67th Annual Meeting of the APS Division of Fluid Dynamics

67th Annual Meeting of the APS Division of Fluid Dynamics San Francisco, California http://www.aps.org/units/dfd/meetings/meeting.cfm?name=DFD14 Sunday, November 23, 2014 8:00AM - 9:57AM Session A1 Non-Newtonian Flows: General — 3000 - Nicholas T. Ouellette, University of Chicago 8:00AM A1.00001 Finite Amplitude Stability Analysis of Polymer Fiber Spinning KARAN GUPTA, PARESH CHOKSHI, Indian Institute of Technology Delhi — The spinning of polymeric fibers suffers from draw resonance instability, manifested by periodic variation of fiber diameter. This occurs when the draw ratio exceeds a certain critical value above which the extensional is unstable. In the present study, weakly nonlinear analysis is performed for polymer fiber spinning to estimate the nature of bifurcation and to construct the finite amplitude branch near critical point. For entangled polymers, we employ the eXtended Pom-Pom model which describes nonlinear rheology of polymer melt. The linear stability analysis provides the critical draw ratio as a function of fluid elasticity represented by Deborah number. In the unstable regime, the nonlinearities saturate the disturbance amplitude to an equilibrium value. Weakly nonlinear analysis is carried out to obtain the equilibrium amplitude along the neutral stability curve. The dynamical equation for the amplitude is the Landau equation with a Landau constant representing nonlinear growth rate. For flows at small De, the Landau constant is found to be negative, indicating supercritical bifurcation. The amplitude branch constructed shows a limit cycle behavior. As the fluid elasticity is increased, initially the equilibrium amplitude is found to decrease and reaches the lowest value when the strain hardening is maximum. Further increase in elasticity, the material undergoes strain softening behavior which leads to increase in amplitude of the oscillations. At very high De, the flow becomes subcritically unstable meaning the flow can become unstable below critical draw ratio. 8:13AM A1.00002 Impact of drag reducing polymers on the onset of instability in a pipe with reverse flow , H.J. SHASHANK, K.R. SREENIVAS, Jawaharlal Nehru Centre for Advanced Scientific Research — The objective of this study is to understand the mechanism by which drag reducing polymer (DRP) additives modify turbulent flow, so as to reduce turbulent drag. Reverse flow in a pipe occurs when the fluid close to the wall moves in an opposite direction to that of the core fluid. Reverse flow is established by using a piston-cylinder mechanism, the programmed motion of which imparts a known impulse to the fluid. When the piston is stopped at the end of the stroke, fluid inertia makes the core of the flow to continue in the same direction. In order to conserve mass, reverse flow is established close to the wall. An inflection point is thus formed, leading to flow instability above a critical Reynolds number 1 . Dye and streak flow visualization experiments are performed to highlight the impact of DRP additives (polyethylene oxide, PEO, dissolved in water). The time of onset of the instability and the wavelength of the observed instability are studied in systems with and without DRP additives. This study will provide further insight into the phenomenon of turbulent polymer drag reduction. 1 Das & Arakeri, J Fluid Mech / Volume 374 / November 1998, pp 251-283 8:26AM A1.00003 New insights into the nature of the asymmetrical flow of shear-thinning polymer solutions in transitional pipe flow , CHAOFAN WEN, ROBERT POOLE, DAVID DENNIS, University of Liverpool — Previous studies of shear-thinning fluids in pipe flow discovered that, although the time-averaged velocity profile was axisymmetric when the flow was laminar or fully turbulent, contrary to expectations it was asymmetric in the laminar-turbulent transition regime. The general consensus of these previous experiments was that the location of the peak velocity remained at a fixed point in space. We present new experimental data which demonstrates that this is in fact not the case. The experiment was performed using an aqueous solution of Xanthan Gum (0.15wt%), a shear-thinning polymer solution. Stereoscopic particle image velocimetry (SPIV) was used to measure the 3C velocity vectors over the entire circular cross-section of the pipe, 220 pipe diameters downstream of the inlet. The exhibition of significant departures from axisymmetry in transitional flows of shear-thinning fluids was observed and in addition it was discovered that the asymmetric flow pattern is not stationary, although the peak velocity does preferentially arise at certain azimuthal locations. The ensemble average of all the SPIV data results in the recovery of the velocity profile measured using laser Doppler velocimetry in previous studies: still asymmetric but to a lesser extent than the instantaneous flow. 8:39AM A1.00004 Microstructure change of shear-bands in concentric cylinder flow of wormlike miceller solutions observed by birefringence profile , MASATOSHI ITO, YUMIKO YOSHITAKE, TSUTOMU TAKAHASHI, Nagaoka University of Technology — The shear-bands formation process of wormlike micellar solutions in start-up shear has been examined by flow-birefringence observation. A concentric cylinder flow cell is used as a platform to generate the start-up shear and a birefringence observation system using the crossed Nichol polarizes with a white light source is built on it together. In this system, the entire flow field along both radial and circumferential direction can be observed. The magnitude of the birefringence is evaluated by the hue profile calculated from the color profile. The orientation angle at each band is estimated by the extinction angle. CTAB/NaSal aqueous solutions were used as a test fluid and it is known that they generate the shear-induced structure (SIS) at a certain shear rate. The birefringence appears homogeneously in the entire area of the flow field at low shear rate. At a certain shear rate, a thin highly oriented band is generated near the driven wall. With increasing the shear rate, this band changes to the SIS state gradually. When the extreme shear-hardening phenomenon appears at start region of high shear rate flow, the birefringence changes to homogeneous. In this case, the stress-optical coefficient keeps a constant that is the almost same value at the low shear. 8:52AM A1.00005 On the rise velocity discontinuity of a deformable bubble in unbounded viscoelastic solutions1 , JOHN TSAMOPOULOS, DIMITRIS FRAGGEDAKIS, YIANNIS DIMAKOPOULOS, Laboratory of Fluid Mechanics and Rheology, Dep. of Chemical Engineering, University of Patras — It is well-documented experimentally, but not well-understood that a bubble steadily rising in a viscoelastic solution exhibits a negative wake and a jump discontinuity in its rise velocity, when its radius exceeds a critical value. In all experiments, the bubble shape forms a cusp in its back side and in some experiments it loses axial symmetry forming a wedge. Some authors have related the velocity jump with the existence of the negative wake or even the wedge formation. We have undertaken a computational study to explore the mechanisms behind these phenomena. To this end, we have used the ePTT model and determined its rheological parameters by fitting it to experiments. Then we developed an FE code (using elliptic grid generation and the SUPG and EVSS methods) and calculated the bubble rise and deformation as its radius increases. This simultaneously affects all parameters: Bond, Archimedes and Deborah numbers. Our predictions reproduce very accurately bubble shapes and the results up to the velocity jump or, in certain cases, beyond it using arc-length continuation. The discontinuity is attributed to a hysteresis loop, but does not require the presence of a wedge in the bubble shape and the negative wake is predicted even before this jump. 1 Supported financially by the General Secretariat of Research & Technology of Greece through the program “Thalis” (Grant titled “COVISCO”) co-funded by the ESF and National resources. 9:05AM A1.00006 Polymer Stress-Gradient Induced Migration in Thin Film Flow Over Topography1 , SOPHIA TSOUKA, YIANNIS DIMAKOPOULOS, JOHN TSAMOPOULOS, Laboratory of Fluid Mechanics and Rheology, Dep. of Chemical Engineering, University of Patras — We consider the 2D, steady film flow of a dilute polymer solution over a periodic topography. We examine how the distribution of polymer in the planarization of topographical features is affected by flow intensity and physical properties. The thermodynamically acceptable, Mavrantzas-Beris two-fluid Hamiltonian model is used for polymer migration. The resulting system of differential equations is solved via the mixed FE method combined with an elliptic grid generation scheme. We present numerical results for polymer concentration, stress, velocity and flux of components as a function of the non-dimensional parameters of the problem (Deborah, Peclet, Reynolds and Capillary numbers, ratio of solvent viscosity to total liquid viscosity and geometric features of the topography). Polymer migration to the free surface is enhanced when the cavity gets steeper and deeper. This increases the spatial extent of the polymer depletion layer and induces strong banding in the stresses away from the substrate wall, especially in low polymer concentration. Macromolecules with longer relaxation times are predicted to migrate towards the free surface more easily, while high surface tension combined with a certain range of Reynolds numbers affects the free surface deformations. 1 Work supported by the General Secretariat of Research & Technology of Greece through the program “Excellence” (Grant No. 1918) in the framework “Education and Lifelong Learning” co-funded by the ESF. 9:18AM A1.00007 Turbulent Fluctuations in Dilute Polymer Solutions , ALEXANDRE DE CHAUMONT QUITRY, NICHOLAS T. OUELLETTE, Yale University — The interaction of complex fluids with turbulent flows presents challenges illustrated in many natural and industrial phenomena. In this study, we report experiment measurements of the modification of turbulence in the presence of long-chain polyacrylamide in water. We use Lagrangian Particle Tracking to study the central region of a Von Karman swirling flow, generated by placing counter-rotating impellers in a cylindrical container. While it has been shown that concentrations as low as 1p.p.m. by weight can affect turbulent fluctuations, it remains theoretically challenging to identify a physical mechanism distinct from an increase in effective viscosity observed at higher concentrations. We attempt to characterize such a mechanism with measurements of spatial and temporal correlations of the velocity and acceleration fields. 9:31AM A1.00008 Lattice Boltzmann simulations of liquid crystal particulate flow in a channel with finite anchoring boundary conditions , RUI ZHANG, TYLER ROBERTS, JUAN DE PABLO, University of Chicago, DEPABLO TEAM — Liquid crystals (LC) posses anisotropic viscoelastic properties, and, as such, LC flow can be incredibly complicated. Here we employ a hybrid lattice Boltzmann method (pioneered by Deniston, Yeomans and Cates) to systematically study the hydrodynamics of nematic liquid crystals (LCs) with and without solid particles. This method evolves the velocity field through lattice Boltzmann and the LC-order parameter via a finite-difference solver of the Beris-Edwards equation. The evolution equation of the boundary points with finite anchoring is obtained through Poisson bracket formulation. Our method has been validated by matching the Ericksen-Leslie theory. We demonstrate two applications in the flow alignment regime. We first investigate a hybrid channel flow in which the top and bottom walls have different anchoring directions. By measuring the apparent shear viscosity in terms of Couette flow, we achieve a viscosity inhomogeneous system which may be applicable to nano particle processing. In the other example, we introduce a homeotropic spherical particle to the channel, and focus on the deformations of the defect ring due to anchorings and flow. The results are then compared to the molecular dynamics simulations of a colloid particle in an LC modeled by a Gay-Berne potential. 9:44AM A1.00009 Brownian Dynamics Simulation of two-dimensional nanosheets under extensional flow , YUEYI XU, Department of Chemical Engineering, Texas Tech University, MICAH GREEN, Department of Chemical Engineering, Texas A&M University — We investigated the morphology change of two-dimensional nanosheets under extensional flow using a coarse-grained model. Nanosheets such as graphene are promising materials for a variety of materials and electronics applications; extensional flow fields are used to cast or process liquid nanosheet dispersions in several processing techniques, including spin coating and compression molding. Process parameters, including bending stiffness and Weissenberg numbers can have a significant impact on the nanosheet morphology and the physical properties of the finished products. We use Brownian Dynamics simulations to study the impact of external flow field on a two-dimensional bead-rod lattice model. Our model was previously demonstrated for steady shear flow. Here we studied the change of morphology of graphene over time and varied the sheet size, bending stiffness and Weissenberg number. Our results showed a flattening behavior that increases with Weissenberg number. Our results also showed significant differences between nanosheets as a function of bending stiffness, with contrasting “plate” and “washrag” results under extension. The intrinsic viscosity first experiences a drop with Weissenberg number followed by a plateau associated with maximum extension. Sunday, November 23, 2014 8:00AM - 9:57AM Session A2 Suspensions: Theory and Modeling — 3002 - Lorenzo Botto, Queen Mary University of London 8:00AM A2.00001 Does Suspension Crowding Screen Hydrodynamic Interactions? , ROSEANNA ZIA, Cornell University, JAMES SWAN, MIT, YU SU, Cornell University — Resistance and mobility functions describe linear couplings between moments of the hydrodynamic traction on a suspended particle and the motion of that or other suspended particles. For two isolated spheres, these functions are well known and have been applied directly in the solution of many important problems for dilute colloidal dispersions. We have devised a new stochastic technique to calculate an analogous set of functions for two spheres immersed in a suspension that are then used to model the near-equilibrium dynamics of concentrated dispersions, including viscoelasticity and long-time diffusion. Of interest is the degree of screening of hydrodynamic interactions by the intervening medium. We find that the mobility is unscreened at the pair level, even in suspensions of high concentration, confirming that hydrodynamic interactions are an essential part of the dynamics of crowded systems and cannot be neglected in favor of simple renormalization schemes. We compare our results for the hydrodynamic interactions between suspended particles to predictions from two-point microrheology. This technique can be used to infer the complex viscosity from long-ranged decay of the pair mobility in viscoelastic materials. Its validity when not in the continuum limit is addressed. 8:13AM A2.00002 Fluctuation, dissipation, and a non-equilibrium “equation of state” via nonlinear microrheology of hydrodynamically interacting colloids , HENRY CHU, ROSEANNA ZIA, Cornell University — In our recently developed non-equilibrium Stokes-Einstein relation for microrheology, we showed that, in the absence of hydrodynamic interactions, the stress in a suspension is given by a balance between fluctuation and dissipation. Here we generalize our theory to develop a simple analytical relation connecting diffusive fluctuation, viscous dissipation and suspension stress in systems of hydrodynamically interacting colloids. In active microrheology, a Brownian probe is driven through a complex medium. The strength of probe forcing compared to the entropic restoring force defines a Peclet number, Pe. In the absence of hydrodynamics, normal stress differences scale as Pe 4 and Pe for weak and strong probe forcing, respectively. But as hydrodynamics become important, interparticle forces give way to lubrication interactions and the normal stresses scale as Pe 2 and Pe δ ln(Pe), where 0.773≤ δ ≤1 as hydrodynamics vary from strong to weak. The new phenomenological theory is shown to agree with standard micromechanical definitions of the stress. A connection is made between the stress and an effective temperature of the medium, prompting the interpretation of the particle stress as the energy density, and the expression for osmotic pressure as a “non-equilibrium equation of state.” 8:26AM A2.00003 Effect of Amphiphiles on the Rheology of Triglyceride Networks , JYOTI SETH, Indian Institute of Technology Bombay — Networks of aggregated crystallites form the structural backbone of many products from the food, cosmetic and pharmaceutical industries. Such materials are generally formulated by cooling a saturated solution to yield the desired solid fraction. Crystal nucleation and growth followed by aggregation leads to formation of a space percolating fractal-network. It is understood that microstructural hierarchy and particle-particle interactions determine material behavior during processing, storage and use. In this talk, rheology of suspensions of triglycerides (TAG, like tristearin) will be explored. TAGs exhibit a rich assortment of polymorphs and form suspensions that are evidently sensitive to surface modifying additives like surfactants and polymers. Here, a theoretical framework will be presented for suspensions containing TAG crystals interacting via pairwise potentials. The work builds on existing models of fractal aggregates to understand microstructure and its correlation with material rheology. Effect of amphiphilic additives is derived through variation of particle-particle interactions. Theoretical predictions for storage modulus will be compared against experimental observations and data from the literature and micro structural predictions against microscopy. Such a theory may serve as a step towards predicting short and long-term behavior of aggregated suspensions formulated via crystallization. 8:39AM A2.00004 Dissipative Particle Dynamics modeling of nanorod-polymer composites , SHAGHAYEGH KHANI, JOAO MAIA, Case Western Reserve University — Recent years have seen a plethora of experimental methods for fabricating nanorodpolymer composites with enhanced physical and mechanical properties. The macroscopic properties of the composites are directly related to the dispersion and organization of the nanoparticles in the matrix. For instance, a significant improvement in the properties of the nanorod-polymer composites is observed upon formation of a percolating network. Thus, controlling the structure of the nanoparticles in the matrix will advance the technology in the field. One way of doing this is by adjusting the chemical interactions which is done through grafting polymer chains on the surface of the rods. Although the enthalpic interactions play the major role in such systems other entropic variables such as the dimension of the rods, density of grafting and etc. may influence the final morphology of the system. The recent developments in the computational techniques have paved the road for further understanding of the controlled assembly of nanorods in polymer matrices. In this study, Dissipative Particle Dynamics (DPD) is employed in order to investigate the effect of enthalpic and entopic variables on the phase behavior of the nanorod-polymer composites. DPD is a coarse-grained mesoscale method which has been found very promising in simulating multi component systems. The interaction parameter between the components of the systems can be mapped onto the Flory-Huggins χ-parameter via well-known Groot-Warren expression. The main goal of this work is to provide a phase diagram that can be used to guide the experiments in designing new materials. 8:52AM A2.00005 Simulation of particle sedimentation in the interface of a stretched capillary bridge , LORENZO BOTTO, School of Engineering and Materials Science, Queen Mary University of London — This talk examines the classical problem of particle sedimentation. In contrast to traditional studies focusing on bulk suspensions, we consider particles entrapped in nearly vertical fluid interfaces and sedimenting owing to gravity or a magnetic field. The interface shape corresponds to that of an axi-symmetric capillary bridge held captive between two parallel circular disks. A transport equation for the particle concentration field has been developed and coupled to the Navier-Stokes equation for the fluid; the resulting system solved numerically in the thin-thread approximation for small capillary and Reynolds numbers. The lower disk is stationary and the upper disk moves with an assigned velocity. The ratio of the settling to stretching velocities is varied. The competition between the sedimentation-induced particle flux and the extensional flow in the neck leads, for intermediate settling velocities, to the formation of a “ring” of high particle concentration; for sufficiently large settling velocities, the particles settle at the bottom of the bridge, potentially modifying the interface shape. These results may help understand the effect of body forces on interfacial tranport, with application to froth flotation processes and the stability of Pickering emulsions. 9:05AM A2.00006 The dynamics of orientable particles in simple shear flow , NAVANEETH KIZHAKKE MARATH, GANESH SUBRAMANIAN, Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore — In simple shear flow, in the Stokes limit, a spheroid rotates indefinitely in any of an infinite single parameter family of periodic orbits, called Jeffery orbits. We have recently used an analytical framework based on spheroidal harmonics to show that weak particle inertia at O(St) and weak fluid inertia at O(Re), St and Re being, respectively, the Stokes number and Reynolds number, lead to an irreversible drift across Jeffery orbits. The spheroid eventually adopts a tumbling or spinning mode, and for an oblate spheroid, the choice of mode depends on its initial orientation. The leading-order inertial corrections leave the time period of rotation (Tp) unchanged. We consider the effects of particle inertia at O(St2 ) and fluid inertia at O(Re3/2 ) on Tp using a reciprocal theorem formulation. It is shown that particle inertia at O(St2 ) results in a decrease in Tp. The fluid inertial contribution is singular in character, arising from the outer region at length scales of O(Re−1/2 ).This allows for a Fourier-space formulation, and the results for moderate- aspect-ratio spheroids show an increase in the Tp. The theoretical predictions are consistent with the results of recent simulations. 9:18AM A2.00007 The concentration instability of a sedimenting suspension of flexible fibers , HARISHANKAR MANIKANTAN, University of California San Diego, LEI LI, SAVERIO SPAGNOLIE, University of Wisconsin-Madison, DAVID SAINTILLAN, University of California San Diego — The stability of a dilute suspension of sedimenting flexible fibers is studied theoretically. Fiber compliance causes individual particles to reorient while sedimenting in a quiescent fluid. We incorporate the rate of reorientation for weakly flexible fibers into a mean-field model to study the stability of a suspension of such fibers to perturbations in concentration. Fiber flexibility is shown to have two opposing effects on suspension stability. First, it establishes a base state that is anisotropic in orientation distribution. We show that such a base state is more prone to a concentration instability than an isotropic distribution, and we illustrate the underlying mechanism. Second, the proclivity of particles to reorient due to flexibility hinders horizontal migration - a key ingredient of the instability mechanism - and suppresses the growth of concentration fluctuations. We analyze this effect by extending our theory to the next order in fiber flexibility, and indeed the growth rate of perturbations is shown to decrease for more compliant fibers. In a Brownian suspension, the dominant effect depends on the relative scales of rotational diffusion and flexibility-induced reorientation. 9:31AM A2.00008 Sedimentation of a flexible fiber in a weak vertical fluid flow , DEWEI QI, Western Michigan University, GUOWEI HE, Institute of Mechanics Chinese Academy of Sciences — Lattice Boltzmann and lattice spring model is used to simulate the later al migration of a flexible fiber in a weak vertical fluid flow. Fiber aspect ratio, rigidity, sediment Reynolds number, and shear Reynolds number are varied. At a low Reynolds number, the fiber migrates to a lower fiber density area. In contrary, at a higher Reynolds number, the fiber migrates to a higher fiber density area. Effect of rigidity on lateral migration is studied. 9:44AM A2.00009 Structure and dynamics of a layer of sedimented microspheres near a horizontal planar wall1 , JERZY BLAWZDZIEWICZ, Texas Tech University, ADAR SONN, HAIM DIAMANT, Tel Aviv University, ELIGIUSZ WAJNRYB, MARIA EKIEL-JEZEWSKA, IPPT PAN, Poland, YAEL ROICHMAN, Tel Aviv University — Structure and dynamics of a sedimented layer of silica microspheres is investigated using computer simulations and confocal-microscopy measurements. The system is characterized by the particle area fraction φs and the dimensionless sedimentation parameter l0 = kB T /(mgd), where kB T is the thermal energy, m is the buoyancy-corrected particle mass, g is the gravitational acceleration, and d is the particle diameter. The range 0 < φs < 0.62 and l0 ≈ 1.6 is explored in our experiments. The near-wall particle distribution exhibits a layered structure, with the second layer developing at φs ≈ 0.4. Particle distribution is well described by a phenomenological model that involves equilibration of a quasi-two dimensional chemical potential. The effective self-diffusivity of the first and second particle layer has been determined. We find that the suspension microstructure is significantly affected by particle polydispersity, whereas the self-diffusivity is only moderately affected. 1 Supported by NSF grant No. CBET 1059745 and National Science Center (Poland) Grant No. 2012/05/B/ST8/03010 Sunday, November 23, 2014 8:00AM - 9:57AM Session A3 Porous Media Flows I: Convection — 3004 - J.M. McDonough, University of Kentucky 8:00AM A3.00001 A theoretical relationship between porosity and permeability , J.M. MCDONOUGH, TINGTING TANG, University of Kentucky — The equations of fluid motion for flow in porous media typically contain the physical parameters porosity and permeability. The first of these is simply a ratio of fluid volume to overall flow-region volume and is easily estimated. Permeability, on the other hand, is more difficult to predict and must usually be calculated using correlations from laboratory experiments for specific porous materials. A well-known example is the Kozeny–Carmen relationship (see, e.g., Carmen, Flow of Gases Through Porous Media, 1956) expressing permeability in terms of porosity for flow in packed beds of solids. In general, there is not a one-to-one permeability-porosity relationship, and this causes difficulties when simulating flows in domains of widely differing porosity. Here we present the derivation of a formula relating these two quantities. We assume validity of using entropy generation rate maximization to set the stable state in non-equilibrium phenomena (Glansdorff and Prigogine, Physica, 1970). This leads to a first-order ordinary differential equation for porosity in terms of permeability which can be solved exactly, resulting in the desired formula for permeability in terms of porosity (as well as strain rates and temperature from the entropy generation formula). 8:13AM A3.00002 Structure and Stability of High Rayleigh-Number Periodic-Orbit Solutions in Porous Medium Convection , BAOLE WEN, GREGORY CHINI, JOHN GIBSON, University of New Hampshire — Direct numerical simulations (DNS) indicate that the instantaneous flow in buoyancy-driven porous medium convection self-organizes into recurring quasi-coherent structures, suggesting that the basic physics can be understood in terms of these “building blocks” and the patterns they form. In this investigation, we use a Newtonhookstep searching algorithm to compute numerically-exact time-periodic (i.e. periodic orbit) solutions to the porous medium convection problem in small laterally-periodic domains at extreme values of the Rayleigh number. Four types of periodic-orbit solutions with different symmetries are presented, and their periods, stability, and heat-transport properties are quantified. Our results confirm that the periodic orbits capture many features of typical quasi-coherent structures observed in DNS of “turbulent” porous medium convection. 8:26AM A3.00003 Buoyant convection from a discrete source in a leaky porous medium1 , MORRIS FLYNN, MARK ROES, Univ. of Alberta, DIOGO BOLSTER, Notre Dame Univ. — The application of turbulent plume theory in describing emptying filling boxes has yielded novel strategies for the natural ventilation of buildings. Making the plume laminar and having it fall through a porous medium yields a new problem of fundamental significance, insights from which may be applied in minimizing the contamination of groundwater by chemicals leached from waste piles. We review the theory for porous media plumes then adapt to the case of an emptying filling box. The long-time solution consists of two ambient layers, each of which has a uniform density. The lower and upper layers are comprised of fluid that is respectively discharged by the plume and advected into the box through the upper opening. Our theory provides an estimate for both the height and thickness of the associated interface in terms of e.g. the source volume and buoyancy fluxes, the outlet area and permeability and the depth-averaged solute dispersion coefficient, which varies with the far-field horizontal flow speed. Complementary laboratory experiments are provided for the case of a line source plume and show very good agreement with model predictions. Our measurements also show that the permeability, kf , of the lower opening decreases with the density of the fluid being discharged. 1 Funding credit: NSERC, CMC, NSF 8:39AM A3.00004 Effect of viscous coupling in multiphase flow in porous media1 , JUAN C. PADRINO, XIA MA, DUAN Z. ZHANG, Los Alamos National Laboratory — Multiphase flow in porous media has traditionally been modeled by the extension of Darcy’s law. This is accomplished by the introduction of the concept of relative permeability, which depends on fluid saturations only. In cases of fluids with significant viscosity difference, additional representation of fluid interactions of viscous nature, not accounted for by Darcy’s law, are needed. In this work we report on new approaches to modeling viscous coupling between phases. Our analysis starts with the ensemble phase averaged momentum equation for multiphase flow. The averaged momentum equation leads to an equation system similar to Darcy’s law, but with additional force terms representing interaction between fluids. These forces arise from the fact that the less viscous fluid pushes the more viscous one to flow through the porous matrix, as shown in our calculation based on the bundle-of-tubes model [Yang et al., 2009, Int. J. Multiphase Flow, 35, 628]. These forces are therefore proportional to the viscosity of the more viscous fluid and the relative velocity between fluids. Based on the formulation developed from the bundle-of-tubes model, we performed numerical simulations of a laboratory experiment of multiphase flows in a porous matrix. Comparisons with the experimental data and other numerical results are presented and discussed. 1 Work partially supported by LDRD project of LANL. 8:52AM A3.00005 Scaling regimes in solute driven convection in porous media with dispersion , MARC HESSE, The University of Texas at Austin, KYUNG WON CHANG, Stanford University — The solute flux during convection in a porous medium with hydrodynamic dispersion is determined by the balance between buoyant driving forces and dissipation by either molecular diffusion or mechanical dispersion. Direct numerical simulations of two-dimensional convection show that the dimensionless flux, given by the Sherwood number, only scales linearly with the −1 −1 Rayleigh number, if both dissipative mechanisms are included, Rah = (Ra−1 . Here Ram is the standard diffusive Rayleigh number and Rad = H/αt m + Rad ) is a dispersive Rayleigh number, where H is the domain height and αt is the transverse dispersivity. The ratio of these Rayleigh numbers ∆ = Ram /Rad determines the contribution of mechanical dispersion to dissipation and identifies diffusion limited (∆ ≪ 1) and dispersion limited (∆ ≫ 1) convection regimes. In both natural and laboratory systems convection is commonly dispersion limited so that the solute flux is determined by Rad and not by Ram as commonly assumed. 9:05AM A3.00006 Boundary layer convection in a radiatively cooled porous medium , JOSEPH HITCHEN, ANDREW WELLS, University of Oxford — In the polar winter, porous sea ice grows by losing heat to the atmosphere through radiative cooling. Sea ice is a reactive, porous medium so cooling causes solidification and creates density gradients in the ice pore space. Previous studies of mushy-layer convection have used highly-conducting boundary conditions with fixed temperatures but we consider the impact of surface radiative cooling using a mixed boundary condition where the heat flux is linked to the evolving boundary temperature. To build initial insight, we consider convective instability in a deep porous layer cooled from above. Using the Biot number to characterise the relative strengths of thermal conduction in the ice and atmospheric heat exchange, we use an energy stability method to determine the critical Rayleigh number, wavenumber and time for convection to occur, driven by density gradients in a transiently growing boundary layer. In the highly conducting limit, we find similar behaviour to previous studies, but a new regime is identified for lower conductivities with a transition region between the two. Calculations suggest that the Biot number for Arctic sea ice may fall in the transitional regime, and therefore the effects of radiative cooling may be important for ice growth. 9:18AM A3.00007 Isotherms Around a Heated Horizontal Cylinder Embedded in a Porous Medium1 , ÁYAX HERNANDO TORRES VICTORIA, Mexican Petroleum Institute, MARIO SANCHEZ ROSAS, FERNANDO ARAGÓN RIVERA, Instituto Politécnico Nacional, México, FAUSTO ALEJANDRO SÁNCHEZ CRUZ, Universidad Autónoma de Nuevo León, México, ABRAHAM MEDINA OVANDO, Instituto Politécnico Nacional, Nuevo León, México — This work presents an experimental study of free and forced convection phenomena that occur in the vicinity of a heated cylinder embedded in a fluid saturated porous medium. The characteristic distribution of the conformed temperature gradients in the porous medium due to pure free convection, and under the action of a continuous and uniform stream were investigated through the use of four different configurations: first by inducing an air stream from below the heated cylinder, second, by placing an air stream on the left hand side of the heat source, third by an air stream acting from the top of the heat source, and fourth by varying the injection angles. The resulting conformation of the buoyant plumes surrounding the heated cylinder when all phenomena reach the steady state were analyzed with an infrared camera. Correspondence is found with the theoretical and numerical solutions proposed by Kurdyumov and Liñán (2000). 1 We wish to thank to the Mexican Petroleum Institute for the unconditional support given to this project. We also thank the Instituto Politécnico Nacional through the SIP Project No. 20141404. 9:31AM A3.00008 Numerical study of thermally stratified flows of a fluid overlying a highly porous material1 , PANAGIOTIS D. ANTONIADIS, MILTIADIS V. PAPALEXANDRIS, Université catholique de Louvain — In this talk we are concerned with thermally stratified flows in domains that contain a macroscopic interface between a highly porous material and a pure-fluid domain. Our study is based on the single-domain approach according to which the same set of governing equations is employed both inside the porous medium and in the pure-fluid domain. Also, the mathematical model that we employ treats the porous skeleton as a rigid solid that is in thermal non-equilibrium with the fluid. First, we present briefly the basic steps of the derivation of the mathematical model. Then, we present and discuss numerical results for both thermally stratified shear flows and natural convection. Our discussion focuses on the role of thermal stratification on the flows of interest and on the effect of thermal non-equilibrium between the solid matrix and the fluid inside the porous medium. 1 This work is supported by the National Fund for Scientific Research (FNRS), Belgium. 9:44AM A3.00009 Phase-Change Modelling in Severe Nuclear Accidents1 , CHRISTOPHER PAIN, DIMITRIOS PAVLIDIS, ZHIHUA XIE, JAMES PERCIVAL, Imperial College London, JEFFERSON GOMES, University of Aberdeen, OMAR MATAR, MOJI MOATAMEDI, Imperial College London, ALI TEHRANI, Office for Nuclear Regulation (UK), ALAN JONES, PAUL SMITH, Imperial College London — This paper describes progress on a consistent approach for multi-phase flow modelling with phase-change. Although, the developed methods are general purpose the applications presented here cover core melt phenomena at the lower vessel head. These include corium pool formation, coolability and solidification. With respect to external cooling, comparison with the LIVE experiments (from Karlsruhe) is undertaken. Preliminary re-flooding simulation results are also presented. These include water injection into porous media (debris bed) and boiling. Numerical simulations follow IRSN’s PEARL experimental programme on quenching/re-flooding. 1 The authors wish to thank Prof. Timothy Haste of IRSN. Dr. D. Pavlidis is funded by EPSRC Consortium “Computational Modelling for Advanced Nuclear Plants,” grant number EP/I003010/1. Sunday, November 23, 2014 8:00AM - 9:57AM Session A4 Bubbles: Biomedical — 3006 - Eric Johnsen, University of Michigan 8:00AM A4.00001 Research of surface modified microbubbles generated by microchannel for selective adsorption , REI MASUDA, TAKUYA ARIYOSHI, Univ of Tokyo, MTSUHISA ICHIYANAGI, Sophia University, SHU TAKAGI, YOICHIRO MATSUMOTO, Univ of Tokyo — Microbubbles have been already used as ultrasound contrast agents to visualize microcirculation system. They are also expected to be used as drag delivery agents. For these bubbles, one of the important requirements is functionality of adsorption to the targeted area. In order to qualify this requirement, it is expected to modify microbubbles with ligand which has ability of specific adsorption to receptor. Biotin as ligand has very high affinity to avidin as receptor, therefore using these materials is supposed to be proper for the first experimental model to satisfy the requirement. In the present study, microbubbles are generated using T-junction type microchannel, because this system has the advantages to control the size and its monodispersity with the wide variety of choice in both liquid phase and gas phase and the capability of surface coating. Polystyrene-dish is confirmed to be coated with avidin. Furthermore, to confirm microbubbles’ selective adsorption, microbubbles generated with liquid containing biotinylated lipids are tried being put on avidin-coated polystyrene-dish. The results will be discussed in the presentation. 8:13AM A4.00002 Deformation of a soft interface by an oscillating microbubble1 , MARC TINGUELY, OMAR MATAR, VALERIA GARBIN, Imperial College London — Acoustically driven oscillating bubbles are used in biomedical applications, for instance to promote pore formation in cell membranes and enhance gene transfection, or to transiently open the blood-brain barrier, which is otherwise impermeable to drugs. However, control over the stresses generated by oscillating bubbles on cells and tissues is still lacking. We use high-speed video microscopy to observe the deformation of a soft interface (agarose gel, a hydrogel that is commonly used as tissue phantom) by the oscillations of a bubble. The mechanical properties of the hydrogel can be tuned to mimic different tissues. The deformation is measured by tracking the displacement of tracer particles embedded in the gel. The results show that the deformation is due to the “push and pull” motion of the bubble against the soft surface. The phase of the deformation varies with the distance to the bubble, which can be explained by the viscoelastic properties of the gel. 1 National Swiss Foundation, and EPSRC Programme Grant EP/K003976/1 8:26AM A4.00003 Shockwave-Gas bubble Interaction in Complex Configurations , FENFANG LI, MANISH ARORA, CLAUS-DIETER OHL, Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University — Shockwave-gas bubble interaction is relevant in biomedical applications such as shock wave lithotripsy and histotripsy where cell rupture needs to be avoided or is advantageous, as well as in the mining industry for microbubble aerated explosive gels. Here we demonstrate an experimental technique to study this interaction in a well-controlled manner utilizing microfluidics and high-speed photography of up to 2 million frames per second. Micron-size gas bubbles are generated with a continuous wave laser beam modulated with a digital hologram, whereas the shockwave and an expanding cavitation bubble are created with a pulsed laser. Gas bubbles are known to generate fast jets when impacted by shockwaves and we observe jets of 125 m/s and more. Complex interactions are reported for geometric arrangements of up to 6 gas bubbles: cascaded and simultaneous collapse of gas bubbles, back reaction of the gas bubbles on the cavitation bubble, and the deflection of jets for neighbouring bubbles. Besides, we find secondary cavitation within the liquid film below the expanding cavitation bubble, which is likely due to trapped gas exposed to low pressures and high shear, i.e. a regime relevant for cavitation in lubricating films. 8:39AM A4.00004 Acoustic Droplet Vaporization in Microchannels , DAVID LI, MARIO FABIILLI, OLIVER KRIPFGANS, J. BRIAN FOWLKES, JOSEPH BULL, University of Michigan — Gas embolotherapy is a proposed cancer therapy where gas bubbles acting as embolic agents are selectively generated near the tumor site to block blood supply, resulting to tumor necrosis. The gas bubbles are generated by using focused ultrasound to selective vaporize intravenously injected microdroplets. In this study, albumin encapsulated dodecafluorocarbon microdroplets were isolated in 25 to 100 micron diameter polydimethylsiloxane microchannels. The droplets were vaporized at 37 ◦ C using a single pulse from a 7.5 MHz single element focused transducer with 8-32 cycles at 2.2 to 5.6 MPa peak negative pressure. The vaporization process was recorded using an ultra-high speed camera attached to an inverted microscope. A theoretical Rayleigh-Plesset like model was derived to describe the both the expansion of small spherical bubbles as well as cylindrical bubbles in a long microchannel. The gas phase was described as an ideal gas and the liquid DDFP and bulk fluid were viscous Newtonian fluids. Additionally, surface tension, viscous losses from the channel, and the phase change process were included in the model. The theoretical model matched very well to experiments with channel diameters or 50 micron or less. This work was supported by NIH grant R01EB006476. 8:52AM A4.00005 Shock-induced bubble collapse in a vessel: Implications for vascular injury in shockwave lithotripsy1 , VEDRAN CORALIC, TIM COLONIUS, Caltech — In shockwave lithotripsy, shocks are repeatedly focused on kidney stones so to break them. The process leads to cavitation in tissue, which leads to hemorrhage. We hypothesize that shock-induced collapse (SIC) of preexisting bubbles is a potential mechanism for vascular injury. We study it numerically with an idealized problem consisting of the three-dimensional SIC of an air bubble immersed in a cylindrical water column embedded in gelatin. The gelatin is a tissue simulant and can be treated as a fluid due to fast time scales and small spatial scales of collapse. We thus model the problem as a compressible multicomponent flow and simulate it with a shock- and interface-capturing numerical method. The method is high-order, conservative and non-oscillatory. Fifth-order WENO is used for spatial reconstruction and an HLLC Riemann solver upwinds the fluxes. A third-order TVD-RK scheme evolves the solution. We evaluate the potential for injury in SIC for a range of pressures, bubble and vessel sizes, and tissue properties. We assess the potential for injury by comparing the finite strains in tissue, obtained by particle tracking, to ultimate strains from experiments. We conclude that SIC may contribute to vascular rupture and discuss the smallest bubble sizes needed for injury. 1 This research was supported by NIH grant no. 2PO1DK043881 and utilized XSEDE, which is supported by NSF grant no. OCI-1053575. 9:05AM A4.00006 Pinched flow fractionation of microbubbles for ultrasound contrast agent enrichment1 , MICHEL VERSLUIS, MAARTEN KOK, TIM SEGERS, Physics of Fluids group, University of Twente — An ultrasound contrast agent (UCA) suspension contains a wide size distribution of encapsulated microbubbles (typically 1-10 µm in diameter) that resonate to the driving ultrasound field by the intrinsic relationship between bubble size and ultrasound frequency. Medical transducers, however, operate in a narrow frequency range, which severely limits the number of bubbles that contribute to the echo signal. Thus, the sensitivity can be improved by narrowing down the size distribution of the bubble suspension. Here, we present a novel, low-cost, lab-on-a-chip method for the sorting of contrast microbubbles by size, based on a microfluidic separation technique known as pinched flow fractionation (PFF). We show by experimental and numerical investigation that the inclusion of particle rotation is essential for an accurate physical description of the sorting behavior of the larger bubbles. Successful sorting of a bubble suspension with a narrow size distribution (3.0± 0.6 µm) has been achieved with a PFF microdevice. This sorting technique can be easily parallelized, and may lead to a significant improvement in the sensitivity of contrast-enhanced medical ultrasound. 1 This work is supported by NanoNextNL, a micro and nanotechnology consortium of the Government of the Netherlands and 130 partners. 9:18AM A4.00007 The effect of relaxation on cavitation dynamics in viscoelastic media1 , LAUREN MANCIA, MATTHEW WARNEZ, ERIC JOHNSEN, University of Michigan — Cavitation plays an important role in diagnostic and therapeutic ultrasound. In certain applications, cavitation bubbles are produced directly in soft tissue, a viscoelastic medium. Although bubble dynamics research in water has received significant attention, the behavior of bubbles in tissue-like media is much less well understood, as the dynamics are strongly affected by the viscoelastic properties of the surroundings, including viscosity, elasticity and relaxation. In the present work, we numerically investigate the role of stress relaxation on spherical bubble dynamics. We simulate bubble dynamics in viscoelastic media with linear and nonlinear relaxation under different types of forcing. Results indicate that the presence of relaxation causes faster growth rates and permits bubble rebound driven purely by residual stresses in the surroundings, a phenomenon not observed in Newtonian media. Differences between nonlinear models become important only following a strong collapse (in which high stresses are generated), thus requiring a robust numerical approach. 1 This work was supported by NSF grant number CBET 1253157 and NIH grant number 1R01HL110990-01A1. 9:31AM A4.00008 Numerical Modeling of 3-D Dynamics of Ultrasound Contrast Agent Microbubbles Using the Boundary Integral Method , MICHAEL CALVISI, Department of Mechanical and Aerospace Engineering, University of Colorado, Colorado Springs, KAWA MANMI, QIANXI WANG, School of Mathematics, University of Birmingham — Ultrasound contrast agents (UCAs) are microbubbles stabilized with a shell typically of lipid, polymer, or protein and are emerging as a unique tool for noninvasive therapies ranging from gene delivery to tumor ablation. The nonspherical dynamics of contrast agents are thought to play an important role in both diagnostic and therapeutic applications, for example, causing the emission of subharmonic frequency components and enhancing the uptake of therapeutic agents across cell membranes and tissue interfaces. A three-dimensional model for nonspherical contrast agent dynamics based on the boundary integral method is presented. The effects of the encapsulating shell are approximated by adapting Hoff’s model for thin-shell, spherical contrast agents to the nonspherical case. A high-quality mesh of the bubble surface is maintained by implementing a hybrid approach of the Lagrangian method and elastic mesh technique. Numerical analyses for the dynamics of UCAs in an infinite liquid and near a rigid wall are performed in parameter regimes of clinical relevance. The results show that the presence of a coating significantly reduces the oscillation amplitude and period, increases the ultrasound pressure amplitude required to incite jetting, and reduces the jet width and velocity. 9:44AM A4.00009 Selective breakup of lipid vesicles under acoustic microstreaming flow1 , ANGELO POMMELLA, VALERIA GARBIN, Imperial College London — The dynamics of lipid vesicles under small deformation in simple shear flow is well characterized: complex behaviors such as tumbling, breathing, and tank-treading are observed depending on the viscosity contrast between inner and outer fluid, vesicle excess area, membrane viscosity, and bending modulus. In contrast, phenomena upon large deformation are still poorly understood, in particular vesicle breakup. Simple shear flow geometries do not allow to reach the large stresses necessary to cause vesicle breakup. We use the acoustic microstreaming flow generated by an oscillating microbubble to study the large deformation and breakup of giant unilamellar vesicles. The deformation is governed by a capillary number based on the membrane elasticity K: Ca = η γ̇a/K where η is the viscosity of the outer fluid, a the vesicle radius, and γ̇ the shear rate. We explore the effect of the mechanical properties of the membrane, and demonstrated selective breakup of vesicles based on the difference in membrane elasticity. The results reveal the influence of membrane mechanical properties in shear-induced vesicle breakup and the possibility to control in a quantitative way the selectivity of the process, with potential applications in biomedical technologies. 1 The authors acknowledge funding from EU/FP7 grant number 618333 Sunday, November 23, 2014 8:00AM - 9:57AM — Session A5 Biofluids: From Idealized Swimming to Boundary Layer Flow 3008 - Alexandra Techet, MIT 8:00AM A5.00001 Effect of longitudinal ridges on the hydrodynamic performance of a leatherback turtle model1 , KYEONGTAE BANG, JOOHA KIM, SANG-IM LEE, HAECHEON CHOI, Seoul Natl Univ — Leatherback sea turtles (Dermochelys coriacea) known as the fastest swimmer and the deepest diver among marine turtles have five longitudinal ridges on their carapace, and these ridges are the most remarkable morphological features distinguished from other marine turtles. To investigate the effect of these ridges on the hydrodynamic performance of the leatherback turtle, we model a carapace with and without ridges using a stuffed leatherback turtle in the National Science Museum, Korea. We measure the drag and lift forces on the ridged model in the ranges of real leatherback turtles’ Reynolds number (Re) and angle of attack (α), and compare them with those of non-ridged model. At α < 6◦ , longitudinal ridges decrease drag on the ridged model by up to 32% compared to non-ridged model. On the other hand, at α > 6◦ , the drag and lift coefficients of the ridged model are higher than those of the non-ridged model, and the lift-to-drag ratio of the ridged model is higher by about 7% than that of the non-ridged model. We also measure the velocity field around both models using a particle image velocimetry and explain the hydrodynamic role of ridges in relation to diving behaviors of leatherback sea turtles. 1 Supported by the NRF program (2011-0028032). 8:13AM A5.00002 Characterization of the Boundary Layer on Full-Scale Bluefin Tuna , BRIAN AMARAL, KIMBERLY CIPOLLA, CHARLES HENOCH, NAVSEA Newport — The physics that enable tuna to cross large expanses of ocean while feeding and avoiding predators is not presently understood, and could involve complex control of turbulent boundary layer transition and drag reduction. Typical swimming speeds of Bluefin tuna are 1-2 m/s, but can be higher during strong accelerations. The goal of this work is to experimentally determine the approximate lateral location at which transition to turbulence occurs on the tuna for various speeds. The question is whether laminar flow or an advanced propulsion mechanism (or both) allows them to swim at high speeds. Uncertainties include the surface roughness of the skin, local favorable and adverse pressure gradients, and discontinuities such as the open mouth or juncture at the fins. Historically, much of the fluid mechanics work in the area of fish locomotion has focused on vortex shedding issues rather than the boundary layer. Here, the focus is obtaining information on the boundary layer characteristics of a rigid tuna model. A full scale model of a Pacific Bluefin tuna was fabricated using a mold made from an actual deceased tuna, preserving the surface features and details of the appendages. The model was instrumented with 32 wall pressure sensors and experiments performed in a tow tank. Results from flow visualization, drag and wall pressure measurements over a range of speeds and varying angles of attack will be presented. 8:26AM A5.00003 Passive appendages aid locomotion through symmetry breaking , SHERVIN BAGHERI, UGIS LACIS, Linné Flow Centre, Dept. Mechanics, KTH, Stockholm, ANDREA MAZZINO, DICCA, University of Genova, Italy, HAMID KELLAY, Universite Bordeaux, France, NICOLAS BROSSE, FREDRIK LUNDELL, Linné Flow Centre, Dept. Mechanics, KTH, Stockholm, FRANCOIS INGREMEAU, Universite Bordeaux, France — Plants and animals use plumes, barbs, tails, feathers, hairs, fins, and other types of appendages to aid locomotion. Despite their enormous variation, passive appendages may contribute to locomotion by exploiting the same physical mechanism. We present a new mechanism that applies to body appendages surrounded by a separated flow, which often develops behind moving bodies larger than a few millimeters. We use theory, experiments, and numerical simulations to show that bodies with protrusions turn and drift by exploiting a symmetry-breaking instability similar to the instability of an inverted pendulum. Our model explains why the straight position of an appendage in flowing fluid is unstable and how it stabilizes either to the left or right of the incoming fluid flow direction. The discovery suggests a new mechanism of locomotion that may be relevant for certain organisms; for example, how plumed seeds may drift without wind and how motile animals may passively reorient themselves. 8:39AM A5.00004 On the Hydrodynamic Function of Sharkskin: A Computational Investigation , AARON BOOMSMA, FOTIS SOTIROPOULOS, University of Minnesota — Denticles (placoid scales) are small structures that cover the epidermis of some sharks. The hydrodynamic function of denticles is unclear. Because they resemble riblets, they have been thought to passively reduce skin-friction–for which there is some experimental evidence. Others have experimentally shown that denticles increase skin-friction and have hypothesized that denticles act as vortex generators to delay separation. To help clarify their function, we use high-resolution large eddy and direct numerical simulations, with an immersed boundary method, to simulate flow patterns past and calculate the drag force on Mako Short Fin denticles. Simulations are carried out for the denticles placed in a canonical turbulent boundary layer as well as in the vicinity of a separation bubble. The computed results elucidate the three-dimensional structure of the flow around denticles and provide insights into the hydrodynamic function of sharkskin. 8:52AM A5.00005 Proprioceptive gait and speed selection in a slender inertial swimmer , MEDERIC ARGENTINA, University Nice Sophia Antipolis, MATTIA GAZZOLA, L. MAHADEVAN, Harvard University — We study the dynamics of a slender inertial swimmer accounting for hydrodynamics, mechanics, muscle activity and sensory feedbacks. Our theory elucidates how elastic properties and proprioception contribute to selecting swimming speed and locomotion gait. Swimmers are shown to take advantage of resonance phenomena to enhance speed and efficiency. Furthermore, we demonstrate how a minimal proprioceptive model, in which the local muscle activation is function of body curvature, is sufficient to exploit hydro-mechanic properties and drive elastic instabilities associated with thrust production. Our results quantitatively agree with live fish experiments and provide a mechanistic basis for the relation U/L ∼ f between the swimmer’s speed U, length L and tail beat frequency f determined empirically by Bainbridge more than half a century ago. 9:05AM A5.00006 The effect of input perturbations on swimming performance , ANDREA M. LEHN, George Washington University, PATRICK J.M. THORNYCROFT, GEORGE V. LAUDER, Harvard University, MEGAN C. LEFTWICH, George Washington University — The influence of flexibility and fluid characteristics on the hydrodynamics of swimming has been investigated for a range of experimental systems. One investigative method is to use reduced-order physical models—pitching and heaving hydrofoils. Typically, a smooth, periodic, input signal is used to control foil motion in experiments that explore fundamental factors (aspect ratio, shape, etc.) in swimming performance. However, the significance of non-smooth input signals in undulating swimmers is non-trivial. Instead of varying external properties, we study the impact of perturbed input motions on swimming performance. A smooth sinusoid is overlaid with high frequency, low amplitude perturbations as the input signal for a heaving panel in a closed loop flow tank. Specifically, 1 cm heave amplitude base sinusoids are added to 0.1 cm heave perturbations with frequencies ranging from 0.5 to 13 Hz. Two thin foils with different stiffness are flapped with the combined input signals in addition to the individual high heave and low heave signals that were added to create the combined inputs. Results demonstrate that perturbations can increase thrust and that adding the perturbed signal to a base frequency alters wake structure. 9:18AM A5.00007 A numerical study of vortex-induced drag of elastic swimmer models , THOMAS ENGELS, M2P2-CNRS, Aix-Marseille University, Marseille, France & Institut für Strömungmechanik und Technische Akustik (ISTA), TU Berlin, Germany, DMITRY KOLOMENSKIY, Department of Mathematics and Statistics, McGill University, Montreal, Canada, KAI SCHNEIDER, M2P2-CNRS & CMI Aix-Marseille University, Marseille, France, JOERN SESTERHENN, Institut für Strömungmechanik und Technische Akustik (ISTA), TU Berlin, Germany — Swimming organisms exploit bending waves to produce propulsive force. The achievable cruising speed, depends on the drag force, which balances the propulsive force. Predicting the cruising velocity at intermediate Reynolds numbers thus requires accurately predicting the drag force. In addition to the friction drag, the vortex induced drag, which may play a significant role, has only recently gained the attention of experimentalists. Based on observations obtained using simplyfied mechanical swimmers, which consist of flexible plates with driven pitching motion, Raspa et al. (PoF 26, 2014), established a basic model to explain the influence of the finite aspect ratio by the formation of trailing longitudinal tip-vortices. Here, these generic swimmers are simulated numerically. We vary the aspect ratio in order to assess the influence of coherent vortices on the drag force. The solid model is based on chordwise flexible foils that undergo large, non-linear deformations. They are actively coupled with a 3D Navier-Stokes solver, based on Fourier transforms and the volume penalization to impose the no-slip boundary conditions. The numerical approach allows to access the entire 3D instantaneous flow field and yields thus new insights into the vortex-induced drag. 9:31AM A5.00008 Strongly Coupled Fluid-Body Dynamics in the Immersed Boundary Projection Method1 , CHENGJIE WANG, JEFF D. ELDREDGE, University of California, Los Angeles — A computational algorithm is developed to simulate dynamically coupled interaction between fluid and rigid bodies. The basic computational framework is built upon a multi-domain immersed boundary method library, whirl, developed in previous work. In this library, the Navier-Stokes equations for incompressible flow are solved on a uniform Cartesian grid by the vorticity-based immersed boundary projection method of Colonius and Taira. A solver for the dynamics of rigid-body systems is also included. The fluid and rigid-body solvers are strongly coupled with an iterative approach based on the block Gauss-Seidel method. Interfacial force, with its intimate connection with the Lagrange multipliers used in the fluid solver, is used as the primary iteration variable. Relaxation, developed from a stability analysis of the iterative scheme, is used to achieve convergence in only 2-4 iterations per time step. Several two- and three-dimensional numerical tests are conducted to validate and demonstrate the method, including flapping of flexible wings, self-excited oscillations of a system of linked plates and three-dimensional propulsion of flexible fluked tail. 1 This work has been supported by AFOSR, under award FA9550-11-1-0098. 9:44AM A5.00009 Vortex-induced drag and the role of aspect ratio in undulatory swimmers1 , RAMIRO GODOY-DIANA, VERONICA RASPA, SOPHIE RAMANANARIVO, BENJAMIN THIRIA, PMMH UMR7636, CNRS, ESPCI ParisTech, U Paris 6, U Paris 7 — During cruising, the thrust produced by a self-propelled swimmer is balanced by a global drag force. For a given object shape, this drag can involve skin friction or form drag, both being well-documented mechanisms. However, for swimmers whose shape is changing in time, the question of drag is not yet clearly established. We address this problem by investigating experimentally the swimming dynamics of undulating thin flexible foils. Measurements of the propulsive performance together with full recording of the elastic wave kinematics are used to discuss the general problem of drag in undulatory swimming. We show that a major part of the total drag comes from the trailing longitudinal vortices that roll-up on the lateral edges of the foils. This result gives a comparative advantage to swimming foils of larger span thus bringing new insight to the role of aspect ratio for undulatory swimmers. Ref: Physics of Fluids, Vol. 26, 041701 (2014). 1 We gratefully acknowledge support by EADS Foundation through project “Fluids and elasticity in biomimetic propulsion.” Sunday, November 23, 2014 8:00AM - 9:57AM Session A6 Biofluids: Active Fluids I — 3010 - Arezoo Ardekani, Purdue University 8:00AM A6.00001 Origination of turbulence in dense suspensions of sperm cells , PETR DENISSENKO, Warwick University, JACKSON KIRKMAN-BROWN, DAVID SMITH, Birmingham University, VASILY KANTSLER, Warwick University — Motile microorganisms with pushing flagella, such as sperm cells, can be directed by “one way” microchannels with ratchet teeth-like wall configuration. We use an array of such micro-channels to gradually concentrate human spermatozoa in a circular arena of 1 mm diameter and 200 micron depth. Velocities of individual cells are measured by particle tracking and velocity of cell-carrying fluid is measured using PIV. At high concentrations, fluid velocities and the velocity fluctuations of individual cells exceeding that of individual swimmers in the dilute regime by an order of magnitude have been measured. Velocity correlations are calculated to study evolution of characteristic length scales as the cell concentration increases. Results are discussed in the context of self-organisation phenomena in active fluids and cooperation of sperm cells. 8:13AM A6.00002 Inverse turbulent cascade in swarming sperm1 , ADAMA CREPPY, OLIVIER PRAUD, XAVIER DRUART, PHILIPPA KOHNKE, FRANCK PLOURABOUE, None, INRA, CNRS, UMR, F-37380 NOUZILLY, FRANCE TEAM2 , UNIVERSITÉ DE TOULOUSE, INPT, UPS, IMFT,UMR 5502, FRANCE TEAM3 — Collective motion of self-sustained swarming flows has recently provided examples of small scale turbulence arising where viscosity effects are dominant. We report the first observation of an universal inverse enstrophy cascade in concentrated swarming sperm consistent with a body of evidence built from various independent measurements. We found a well-defined k−3 power-law decay of velocity field power-spectrum and relative dispersion of small beads consistent with theoretical predictions in two-dimensional turbulence. Concentrated living sperm displays long-range, correlated whirlpool structures the size of which provides turbulence’s integral scale. We propose a consistent explanation for this quasi-two-dimensional turbulence based on self-structured laminated flow forced by steric interaction and alignment, a state of active matter that we call “swarming liquid crystal.” We develop scaling arguments consistent with this interpretation. The implication of multi-scale collective dynamics of sperm’s collective motility for fertility assessment is discussed. 1 This work has been supported by the French Agence Nationale pour la Recherche (ANR) in the frame of the contract MOTIMO (ANR-11-MONU-00901). We thank Pierre Degond, Eric Climent, Laurent Lacaze and Frédéric Moulin for interesting discussions. 2 Physical experiment aspect in MOTIMO contract 3 Biological experiments in MOTIMO contract 8:26AM A6.00003 Bacterial turbulence in motion , ROBERTO RUSCONI, STEVEN SMRIGA, ROMAN STOCKER, Massachusetts Institute of Technology, Cambridge, MA, ELEONORA SECCHI, STEFANO BUZZACCARO, ROBERTO PIAZZA, Politecnico di Milano, Italy — Dense suspensions of motile bacteria exhibit collective dynamics akin to those observed in classic, high Reynolds number turbulence, yet this analogy has remained largely qualitative. Here we present experiments in which a dense suspension of Bacillus subtilis bacteria was flown through narrow microchannels and the velocity statistics of the flowing suspension were accurately quantified with a recently developed velocimetry technique. This revealed a robust intermittency phenomenon, whereby the average velocity profile of the flowing suspension oscillated between a plug-like flow and a parabolic flow. This intermittency is a hallmark of classic turbulence and was associated with the presence of collective structures in the suspension. Furthermore, quantification of the Reynolds stress profile revealed a direct link between the turbulent nature of the suspension and its anomalous viscosity. 8:39AM A6.00004 Diffusion of an ellipsoid in a quasi-2D bacteria suspension , YI PENG, XIANG CHENG, Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN, 55455, USA — Enhanced translational diffusion of spherical particles induced by a suspension of bacteria has been established as a distinct feature for active fluids. Here, instead of spherical tracer particles, we study the diffusion of ellipsoidal particles of various aspect ratios in a free-standing film of bacteria. Using high-speed digital video microscopy, we measured the mean-square displacements and calculated the translational and rotational diffusion coefficients of the elliptical tracer particles. We found that both the translational and rotational diffusion of the particles are dramatically enhanced by the motion of bacteria. The probability distribution functions for linear and angular displacements become non-Gaussian at high bacterial concentrations. Moreover, we also explored the coupling between translational and rotational diffusion induced by the swimming bacteria. 8:52AM A6.00005 Enhanced diffusion self-stimulated by micro-algae in an active, aerotactic bacterial suspension1 , FRANÇOIS PEAUDECERF, RAYMOND E. GOLDSTEIN, University of Cambridge — Suspensions of swimming bacteria form a new class of active fluids that generate complex phenomena. An “active bath” of bacteria for instance produces fluid flows which move passive colloids in a random-like walk, associated with an effective diffusion coefficient higher than for Brownian motion. The value of this enhanced diffusion coefficient depends on the local density of bacteria and their swimming behavior. However, with aerotactic, obligate aerobic bacteria such as B. subtilis, the local oxygen concentration impacts on the distribution of cells and their swimming behavior. We consider the specific case in which non-motile photosynthetic algal cells interacting with a B. subtilis suspension not only play the role of passive colloids, but also produce oxygen under light. We demonstrate that this new kind of active suspension, under heterogeneous illumination, can induce an effective negative phototaxis of the passive algal cells. We explain the origin of this novel phenomenon as the combination of algal oxygen production, diffusion, chemotaxis and motility switching in bacteria resulting in an heterogeneous enhanced diffusion. Finally, we present potential applications for algal cell mixing and sorting, that can inspire new methods for bioengineering. 1 Supported by ERC, Raymond and Beverly Sackler Foundation, and Mines ParisTech 9:05AM A6.00006 Hydrodynamic dispersion of microswimmers in suspension , MATTHIEU MARTIN1 , SALIMA RAFAÏ, PHILIPPE PEYLA, LIPhy - CNRS, Univ. Grenoble — In our laboratory, we study hydrodynamics of suspensions of micro-swimmers. These micro-organisms are unicellular algae Chlamydomonas Rheinhardii which are able to swim by using their flagella. The swimming dynamics of these microswimmers can be seen as a random walk, in absence of any kind of interaction. In addition, these algae have the property of being phototactic, i.e. they swim towards the light. Combining this property with a hydrodynamic flow, we were able to reversibly separate algae from the rest of the fluid. But for sufficiently high volume fraction, these active particles interact with each other. We are now interested in how the coupling of hydrodynamic interactions between swimmers and phototaxis can modify the swimming dynamics at the scale of the suspension. To this aim, we conduct experiments in microfluidic devices to study the dispersion of the micro-organisms in a the liquid phase as a function of the volume fraction. We show that the dispersion of an assembly of puller type microswimmers is quantitatively affected by hydrodynamics interactions. 1 Phd student 9:18AM A6.00007 Photomixing of chlamydomonas rheinhardtii suspensions , JULIEN DERVAUX, LIED, University Paris Diderot, MARINA CAPELLAZZI RESTA, BÉRENGÈRE ABOU, PHILIPPE BRUNET, MSC, University Paris Diderot — Chlamydomonas rheinhardtii is a fast swimming unicellular alga able to bias its swimming direction in gradients of light intensity, an ability know as phototaxis. We have investigated experimentally both the swimming behavior of individual cells and the macroscopic response of shallow suspensions of these micro-organisms in response to a localized light source. At low light intensity, algae exhibit positive phototaxis and accumulate beneath the excitation light. In weakly concentrated thin layers, the balance between phototaxis and cell motility results in steady symmetrical patterns compatible with a purely diffusive model using effective diffusion coefficients extracted from the analysis of individual cell trajectories. However, at higher cell density and layer depth, collective effects induce convective flows around the light source. These flows disturb the cell concentration patterns which spread and may then becomes unstable. Using large passive tracer particles, we have characterized the velocity fields associated with this forced bioconvection and their dependence on the cell density and layer depth. By tuning the light distribution, this mechanism of photo-bioconvection allows a fine control over the local fluid flows, and thus the mixing efficiency, in algal suspensions. 9:31AM A6.00008 Phototactic number-density flux in the localized bioconvection of Euglena gracilis , ERIKA SHOJI, Hiroshima University, NOBUHIKO SUEMATSU, Meiji University, HIRAKU NISHIMORI, AKINORI AWAZU, SHUNSUKE IZUMI, MAKOTO IIMA, Hiroshima University — Euglena gracilis is a unicellular phototactic flagellate; it escapes from light sources if the light intensity is higher than 200W/m2 (negative phototaxis). When the suspension of E.gracilis is illuminated from the bottom by strong light, bioconvection patterns are generated. In the case of E.gracilis, the patterns can be spatially localized. The localization mechanism has not been clarified. We report experimental results related to the localization mechanism. In particular, we experimentally measured the strength of the phototaxis in the lateral direction as well as vertical direction. We prepared a thin container in which the suspension is included, and gave the linearly-changing light intensity. We found the number density gets a peak at a particular light intensity, which never happens if the suspension has the vertical phototaxis only. Further, we succeeded in getting the function representing lateral phototaxis. The relationship between the measured functions and the localized convection cells will be also reported. [1] Localized bioconvection patterns and their initial state dependency in Euglena suspensions in an annular container, E. Shoji, H. Nishimori, A. Awazu, S. Izumi, and M. Iima, J. Phys. Soc. Jpn. 83(2014)04300 9:44AM A6.00009 Suppression of resistance to flow in suspensions of bacteria , HECTOR LOPEZ, FAST, Université Paris Sud, Orsay, France, JÉRÉMIE GACHELIN, PMMH-ESPCI, Paris, France, CARINE DOUARCHE, LPS, Orsay, France, ERIC CLÉMENT, PMMH-ESPCI, Paris, France, HAROLD AURADOU, FAST, Université Paris Sud, Orsay, France — It is usually believed that the influence of small amounts of bacteria on the rheological properties of a fluid is negligible. However, recent theoretical studies predict that the activity results in a decrease of the viscosity at values lower than the suspending fluid viscosity. We present experimental measurements of the viscosity of suspensions of Escherichia coli (volume fractions φ<1%) in a simple Couette flow over a broad range of shear rates. For shear rates larger than 1.5 s−1 , the viscosity is constant and slightly above the viscosity of the suspending fluid. This behavior is similar to the one expected for non-active particles. For lower shear rates the fluid exhibits a non-Newtonian behavior: the viscosity decreases and finally reaches a second Newtonian plateau for shear rates below 0.1 s−1 . For φ <0.6%, the decrease is proportional to the bacteria concentration, as predicted by the theories, suggesting that it is a result of the energy input of each individual microswimmer. For φ >0.6%, we evidence for the first time the existence of a super-lubrication regime where the viscous resistance to shear vanishes. We will demonstrate that this regime holds up over a large window of concentration. Sunday, November 23, 2014 8:00AM - 9:31AM Session A7 Biofluids: Medical Devices 3012 - Sarah Waters, Oxford University — 8:00AM A7.00001 Examination of unsteady flow in a mildly curved vessel with stent-like wall protrusions: A tale of two vessels , CHEKEMA PRINCE, SEAN D. PETERSON, University of Waterloo — New stent designs allow for better conformity to the vessel curvature, maintaining the complex primary and secondary flow patterns present in the native vessel. Despite design improvements, stent induced alterations in local vascular geometry are inevitable and have been associated with stent failure due to in-stent restenosis (ISR). The objective of this study is to elucidate the unsteady flow physics induced by stent implantation, accounting in particular for vessel curvature. The present study focuses on the investigation of unsteady flow through mildly curved vessels with protrusion patterns that emulate current stent designs using computational fluid dynamics (CFD). The modeled geometries include various protrusion frequencies, heights, and widths. Two different arterial velocities waveforms, mimicking the coronary and carotid artery environment, will be considered. A detailed examination of the flow environment induced by the stent presence will be correlated with derived parameters from the flow behavior, such as critical wall shear stress typically associated with ISR development. Specifically, the role of secondary flow in the convective transport of ISR stimuli to the vessel wall will be explored. 8:13AM A7.00002 Unsteady jet in designing innovative drug delivery system1 , CONG WANG, PAUL MAZUR, JULIA COSSE, STEPHANIE RIDER, MORTEZA GHARIB, Caltech — Micro-needle injections, a promising pain-free drug delivery method, is constrained by its limited penetration depth. This deficiency can be overcome by implementing fast unsteady jet that can penetrate sub-dermally. The development of a faster liquid jet would increase the penetration depth and delivery volume of micro-needles. In this preliminary work, the nonlinear transient behavior of an elastic tube balloon in providing fast discharge is analyzed. A physical model that combines the Mooney Rivlin Material model and YoungLapalce’s Law was developed and used to investigate the fast discharging dynamic phenomenon. A proof of concept prototype was constructed to demonstrate the feasibility of a simple thumb-sized delivery system to generate liquid jet with desired speed in the range of 5-10 m/s. 1 This work is supported by ZCUBE Corporation. 8:26AM A7.00003 Secondary flows enhance mixing in a model of vibration-assisted dialysis1 , JOHN PITRE, Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, BRUCE MUELLER, SUSAN LEWIS, Department of Clinical, Social and Administrative Sciences, College of Pharmacy, University of Michigan, Ann Arbor, MI, JOSEPH BULL, Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI — Hemodialysis is an integral part of treatment for patients with end stage renal disease. While hemodialysis has traditionally been described as a diffusion-dominated process, recent in vitro work has shown that vibration of the dialyzer can enhance the clearance of certain solutes during treatment. We hypothesize that the addition of vibration generates secondary flows in the dialysate compartment. These flows, perpendicular to the longitudinal axis of the dialysis fibers, advect solute away from the fiber walls, thus maintaining a larger concentration gradient and enhancing diffusion. Using the finite element method, we simulated the flow of dialysate through a hexagonally-packed array of cylinders and the transport of solute away from the cylinder walls. The addition of vibration was modeled using sinusoidal body forces of various frequencies and amplitudes. Using the variance of the concentration field as a metric, we found that vibration improves mixing according to a power law dependency on frequency. We will discuss the implications of these computational results on our understanding of the in vitro experiments and propose optimal vibration patterns for improving clearance in dialysis treatments. 1 This work was supported by the Michigan Institute for Clinical and Health Research and NIH grant UL1TR000433. 8:39AM A7.00004 Heparin Leakage in Central Venous Catheters by Hemodynamic Transport , MICHAEL BARBOUR, PATRICK MCGAH, University of Washington, KENNETH GOW, Seattle Children’s Hospital and Dept. of Surgery, University of Washington, ALBERTO ALISEDA, University of Washington — Central venous catheters (CVCs), placed in the superior vena cava for hemodialysis, are routinely filled with heparin, an anticoagulant, while not in use to maintain patency and prevent thrombus formation at the catheter tip. However, the heparin-lock procedure places the patient at risk for systemic bleeding incidences, as heparin is known to leak into the blood stream. We propose that the driving mechanism behind heparin leakage is advective-diffusive transport due to the pulsatile blood flow surrounding the catheter tip. This novel hypothesis is based on Planar Laser Induced Fluorescence (PLIF) measurements of heparin transport from a CVC placed inside an in vitro pulsatile flow loop and validated with CFD simulations. The results show an initial, fast (<10s), advection-dominated phase that rapidly depletes the concentration of heparin at the CVC tip, followed by a slow, diffusion-limited phase inside the catheter lumen, where concentration is still high, that is insufficient at replenishing the lost heparin at the tip. These results, which estimate leakage rates consistent with published in vivo data, predict that the concentration of heparin at the catheter tip is effectively zero for the majority of the interdialytic phase, rendering the heparin lock ineffective. 8:52AM A7.00005 Effects of incomplete stent apposition on the changes in hemodynamics inside a curved and calcified coronary artery1 , ERIC POON, ANDREW OOI, University of Melbourne, Australia, PETER BARLIS, UMAIR HAYAT, Northern Health, Australia, STEPHEN MOORE, VLSCI, Australia — Percutaneous coronary intervention (PCI) is the modern gold standard for treatment of coronary artery disease. Stenting (a common PCI procedure) of simple lesion inside a relatively straight segment of coronary artery has proven to be highly successful. However, incomplete stent apposition (ISA) where there is a lack of contact between the stent struts and lumen wall is not uncommon in curved and calcified coronary arteries. Computational fluid dynamics simulations are carried out to study the changes in hemodynamics as a result of ISA inside a curved and calcified coronary artery. For a 3mm coronary artery, we simulate a resting condition at 80 mL/min and a range of hyperemic conditions with coronary flow reserve in between 1 and 2. The heartbeat is fixed at 75 BPM. Five different curvatures of the coronary artery are considered. Negative effects on hemodynamic variables, such as low wall shear stress (<0.5 Pa); high wall shear stress gradient (>5,000 Pa/m) and oscillation shear index (0 ≤ OSI ≤ 0.5), are employed to identify locations with high possibilities of adverse clinical events. This study will lead to better understandings of ISA in curved and calcified coronary arteries and help improve future coronary stent deployment. 1 Supported by the Australian Research Council (LP120100233) and Victorian Life Sciences Computation Initiative (VR0210). 9:05AM A7.00006 Mathematical modelling of flow and transport processes in tissue engineering bioreactors , SARAH WATERS, NATALIE PEARSON, JAMES OLIVER, University of Oxford, REBECCA SHIPLEY, University College London — To artificially engineer tissues numerous biophysical and biochemical processes must be integrated to produce tissues with the desired in vivo properties. Tissue engineering bioreactors are cell culture systems which aim to mimic the in vivo environment. We consider a hollow fibre membrane bioreactor (HFMB), which utilises fluid flow to enhance the delivery of growth factors and nutrients to, and metabolite removal from, the cells, as well as provide appropriate mechanical stimuli to the cells. Biological tissues comprise a wide variety of interacting components, and multiphase models provide a natural framework to investigate such interactions. We present a suite of mathematical models (capturing different experimental setups) which consider the fluid flow, solute transport, and cell yield and distribution within a HFMB. The governing equations are simplified by exploiting the slender geometry of the bioreactor system, so that, e.g., lubrication theory may be used to describe flow in the lumen. We interrogate the models to illustrate typical behaviours of each setup in turn, and highlight the dependence of results on key experimentally controllable parameter values. Once validated, such models can be used to inform and direct future experiments. 9:18AM A7.00007 Numerical Simulations of the Mechanics of Vitrectomy , ETHAN YOUNG, JEFF D. ELDREDGE, JEAN-PIERRE HUBSCHMAN, University of California, Los Angeles — Filling the cavity between the lens and retina in the eye is a clear, gel-like substance known as vitreous humor. The treatment of certain eye abnormalities necessitates the removal of this substance, in a surgical procedure called a vitrectomy, using a device called a vitreous cutter. Understanding the behavior of this viscoelastic biofluid during operations is essential to improving the effectiveness of the procedure. In this work, a three-dimensional computational model of a vitreous cutter is investigated using an immersed boundary method and a viscoelastic constitutive model. The solver uses a fractional-step method to satisfy continuity and traction boundary conditions to simulate the applied suction. The Giesekus constitutive equation is used to model the vitreous, as it captures both elastic and shear-thinning effects. Rheological parameters were obtained from the work of Sharif-Kashani et al. [Retina, 2013]. These simulations were used to quantify both the average and time-varying flow rate through the device during different stages in the cutting cycle. Characteristics of the flow field illustrate how surgical variables like cutting speed, duty cycle, and aspiration pressure affect overall flow rate and suggest targets for improving cutter efficacy. Sunday, November 23, 2014 8:00AM - 9:57AM Session A8 Biofluids: In Suspension — 3001/3003 - Michael Graham, University of Wisconsin-Madison 8:00AM A8.00001 Variation of velocity profile according to blood viscosity in a microfluidic channel1 , EUNSEOP YEOM, Pohang Univ of Sci & Tech, YANG JUN KANG, Chosun Univ, SANG-JOON LEE, Pohang Univ of Sci & Tech — The shearthinning effect of blood flows is known to change blood viscosity. Since blood viscosity and motion of red blood cells (RBCs) are closely related, hemorheological variations have a strong influence on hemodynamic characteristics. Therefore, understanding on the relationship between the hemorheological and hemodynamic properties is importance for getting more detailed information on blood circulation in microvessels. In this study, the blood viscosity and velocity profiles in a microfluidic channel were systematically investigated. Rat blood was delivered in the microfluidic device which can measure blood viscosity by monitoring the flow-switching phenomenon. Velocity profiles of blood flows in the microchannel were measured by using a micro-particle image velocimetry (PIV) technique. Shape of velocity profiles measured at different flow rates was quantified by using a curve-fitting equation. It was observed that the shape of velocity profiles is highly correlated with blood viscosity. The study on the relation between blood viscosity and velocity profile would be helpful to understand the roles of hemorheological and hemodynamic properties in cardiovascular diseases. 1 This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIP) (No. 2008-0061991). 8:13AM A8.00002 Effects of viscosity on endothelial cell damage under acoustic droplet vaporization , ROBINSON SEDA, RAHUL SINGH, DAVID LI, JOHN PITRE, ANDREW PUTNAM, J. BRIAN FOWLKES, JOSEPH BULL, University of Michigan — Acoustic droplet vaporization (ADV) is a process by which stabilized superheated microdroplets are able to undergo phase transition with the aid of focused ultrasound. Gas bubbles resulting from ADV can provide local occlusion of the blood vessels supplying diseased tissue, such as tumors. The ADV process can also induce bioeffects that increase vessel permeability, which is beneficial for localized drug delivery. Previous in vitro studies have demonstrated that vaporization at the endothelial layer will affect cell attachment and viability. Several hypotheses have been proposed to elucidate the mechanism of damage including the generation of normal and shear stresses during bubble expansion. A single 3.5 MHz ultrasound pulse consisting of 8 cycles (∼ 2 .3 µs) and a 6 MPa peak rarefactional pressure was used to induce ADV on endothelial cells in media of different viscosities. Carboxylmethyl cellulose was added to the cell media to increase the viscosity up to 300 cP to and aid in the reduction of stresses during bubble expansion. The likelihood of cell damage was decreased when compared to our control (∼ 1 cP), but it was still present in some cases indicating that the mechanism of damage does not depend entirely on viscous stresses associated with bubble expansion. This work was supported by NIH grant R01EB006476. 8:26AM A8.00003 Micro-PIV of Bubble Splitting in a Bifurcation , SAMANTHA STEPHENSON, DAVID LI, FORIAN HELLMEIER, JOHN PITRE, J. BRIAN FOWLKES, JOSEPH BULL, University of Michigan — Gas embolotherapy is a proposed treatment for cancerous tumors. For this treatment, a liquid droplet solution is injected into the bloodstream and focused ultrasound is used to vaporize droplets upstream of the tumor site, resulting in bubbles that are approximately 125x larger in volume. These bubbles will then occlude the blood vessels, thereby depriving the tumor of nutrients leading to eventual tumor necrosis. However, once the bubbles are formed, they will continue to travel through the bloodstream, through bifurcations that split in to smaller daughter vessels before lodging to occlude flow. Micro-particle imaging velocimetry (PIV) was used to study the flow field surrounding the leading edge of the bubble at the bifurcation point. Consistent symmetric bubble splitting at several different flow rates was achieved. Roll angle of the bifurcation was varied to encourage uneven bubble splitting and reversal. In the absence of the bubble, Poiseuille flow was verified in the parent channel. Results were compared to a boundary elements model developed by Calderon et al. 2010. This research was funded by the NIH grant R01EB006476. 8:39AM A8.00004 Margination and demargination in confined multicomponent suspensions: a parametric study1 , MICHAEL GRAHAM, KUSHAL SINHA, RAFAEL HENRIQUEZ RIVERA, University of Wisconsin-Madison — Blood and other multicomponent suspensions display a segregation behavior in which different components are differentially distributed in the cross-stream direction during flow in a confined geometry such as an arteriole or a microfluidic device. In blood the platelets and leukocytes are strongly segregated to the near wall region and are said to be “marginated.” The effects of particle size, shape and rigidity on segregation behavior in confined simple shear flow of binary suspensions are computationally investigated here. The results show that in a mixture of particles with same shape and different membrane rigidity, the stiffer particles marginate while the flexible particles demarginate, moving toward the center of the channel. In a mixture of particle with same membrane rigidity and different shape, particles with smaller aspect ratio marginate while those with higher aspect ratio demarginate. These results are consistent with theoretical arguments based on wall-induced migration and pair collision dynamics. An analytical solution is presented for a model problem that reveals qualitatively different behavior in various parameter regimes. Finally, effects of viscoelasticity of the suspending phase on margination are examined. 1 This work was supported by the NSF under grants CBET-1132579 and CBET-1436082. 8:52AM A8.00005 Fluid Mechanics of the Red Blood Cell and its Cytoskeleton by an Immersed Boundary Method with Nonuniform Viscosity and Density , THOMAS FAI, Harvard University, CHARLES PESKIN, Courant Inst — The red blood cell cytoskeleton, which is anchored to a lipid bilayer membrane, is an elastic network that helps red cells recover from large deformations as they circulate. Although the cytoskeleton has a convoluted structure, as shown in recent tomographic images, it may be modeled simply as a graph of actin-based junctional complexes (nodes) connected by spectrin polymers (edges). We have developed a discrete cytoskeleton model that incorporates statistical properties of the cytoskeleton, such as the edge length and node degree distributions. A specialized image processing technique is used to gather these distributions directly from tomograms. The network elasticity comes from treating the spectrin polymers as entropic springs, and we show that the spring constant obtained from a well-known model of entropic springs is in reasonable agreement with the experimentally determined shear modulus. By simulating the behavior of red blood cells in shear flow using a variable viscosity and variable density immersed boundary method, we compare this discrete model with its approximately 40,000 nodes to more commonly used continuum ones. 9:05AM A8.00006 Simultaneous measurement of flow over and transmigration through a cultured endothelial cell layer1 , LORI LAMBERT, University of Nebraska - Lincoln, IRAKLIS PIPINOS, TIMOTHY BAXTER, JASON MACTAGGART, DEREK MOORMEIER, KENNETH BAYLES, University of Nebraska Medical Center, TIMOTHY WEI, University of Nebraska - Lincoln — The measurement and analysis of fluid forces on endothelial cells at the cellular and subcellular levels is an essential component of understanding mechanotransduction and atherogenesis. The ultimate goal of this study is to examine and model the transport and transmigration of low-density lipoproteins across a confluent endothelial layer as a function of fluid loading and time. In this study, steady flow over a cultured endothelial cell layer at shear rates up to 20 dynes/cm2 in a 350 µm x 70 µm cross section mircrochannel was measured using µPTV measurements. By using multiple measurement planes parallel to the channel wall, wall shear stress and wall pressure were computed as well as the endothelial cell topography. The study was performed over a period of 18 hours in which the transport and transmigration of fluorescently tagged low-density lipoproteins through a cultured endothelial cell layer were examined as a function of fluid forces, cell topography, and time. 1 The help of Dr. Richard Leighton is gratefully acknowledged. 9:18AM A8.00007 Nanoparticle motion near a blood vessel wall in targeted drug delivery1 , HELENA VITOSHKIN, HSIU-YU YU, DAVID M. ECKMANN, RAVI RADHAKRISHNAN, PORTONOVO S. AYYASWAMY, Univ of Pennsylvania — A computational study of the motion of a spherical nanoparticle close to the bounding wall of a blood vessel in targeted drug delivery is presented. An arbitrary Lagrangian-Eulerian algorithm has been carried out, taking into account both the Brownian and the hydrodynamic effects. Pertinent to targeted drug delivery, we focus on the condition when the particle is in the lubrication layer. The velocity auto-correlation function (VACF) is seen to initially decay faster by a factor of particle radius divided by the fluid gap thickness compared to that in an unbounded medium. Long time decay is found to be algebraic. Focusing on hydrodynamic interaction between the particle and the wall, effects of wall curvature, particle size, and variations in density of the particle are investigated. We also study adhesive interactions of a nanoparticle with an endothelial cell located on the vessel wall by the modeling the nanoparticle tethered by a harmonic spring with varying spring constants. It is shown that the particle velocity is affected by hydrodynamic and harmonic spring forces leading to VACF oscillations which decay algebraically at long times. The results agree with those predicted by earlier theories for particle VACF near a wall. These findings have applications in medication administration and in the colloidal sciences. 1 Supported by NIH grant U01 EB016027. 9:31AM A8.00008 Motion induced between parallel plates with offset centers of radial stretching and shrinking , PATRICK WEIDMAN, University of Colorado — The flow between parallel plates separated by distance h is investigated where the upper and lower plates respectively stretch and shrink at the same rate a and the centers of stretching and shrinking are horizontally separated by distance 2 l. A reduction of the Navier-Stokes equation yields two ordinary differential equations dependent on a Reynolds number R = ah2 /ν. In addition a free parameter γ appears which corresponds to a uniform pressure gradient acting along the line connecting the stretching/shrinking centers. We consider three cases: γ = 0, γ = O(1) and γ = O(R). The flow is described by two functions of the plate-normal coordinate η = z/h: the first f (η) has an analytical solution while the second g(η) must be resolved numerically. The small-R solutions are found and the large-R asymptotic behaviors of the wall shear stresses and the centerline velocities are obtained by matching the viscous boundary layer flows to the interior inviscid motion. 9:44AM A8.00009 Axisymmetric rotational slow viscous flows around asymmetric fused dumbbells , D. PALANIAPPAN, Texas A&M University, Corpus Christi — Symmetric rotational viscous flows involving fused dumbbells are considered in the limit of low-Reynolds numbers. The boundary of the rigid dumbbell is formed by two spherical surfaces of arbitrary radii, a and b respectively, intersecting at a vertex π angle, say n , n an integer. Analytic solutions are obtained for the asymmetric configuration submerged in (i) a rotational flow, and (ii) a rotlet induced flow field. The image system in each case is found in terms of fundamental solutions of the Stokes flow equations. Exact expressions for the torque/couple acting on the dumbbell are computed directly from our singularity solutions. It is found that the radii of the spheres, the center-to-center distance, the vertex angle together with the location of the initial rotlet dictate the flow fields and the torque. Upper and lower bounds for the couple acting on the asymmetric dumbbell are determined as well. Our method is based on the successive reflection theory and avoids the use of complex toroidal and meromorphic functions. The utility of the toroidal frame for the axisymmetric rotational flow in the case of arbitrary vertex angle is also discussed. However, for the rotlet flow, there does not appear to be any technique available other than the one provided here. Sunday, November 23, 2014 8:00AM - 9:44AM Session A9 Biofluids: Ciliary Flows — 3014/3016 - Javier Urzay, Stanford University 8:00AM A9.00001 Asynchronous beating of cilia enhances particle capture rate , YANG DING, Beijing Computational Science Research Center, EVA KANSO, University of Southern California — Many aquatic micro-organisms use beating cilia to generate feeding currents and capture particles in surrounding fluids. One of the capture strategies is to “catch up” with particles when a cilium is beating towards the overall flow direction (effective stroke) and intercept particles on the downstream side of the cilium. Here, we developed a 3D computational model of a cilia band with prescribed motion in a viscous fluid and calculated the trajectories of the particles with different sizes in the fluid. We found an optimal particle diameter that maximizes the capture rate. The flow field and particle motion indicate that the low capture rate of smaller particles is due to the laminar flow in the neighbor of the cilia, whereas larger particles have to move above the cilia tips to get advected downstream which decreases their capture rate. We then analyzed the effect of beating coordination between neighboring cilia on the capture rate. Interestingly, we found that asynchrony of the beating of the cilia can enhance the relative motion between a cilium and the particles near it and hence increase the capture rate. 8:13AM A9.00002 Mixing it up: Corals take an active role in mass transport , VICENTE FERNANDEZ, Massachusetts Insitute of Technology, ORR SHAPIRO, Weizmann Institute of Science, DOUGLAS BRUMLEY, MELISSA GARREN, Massachusetts Insitute of Technology, JEFFREY GUASTO, Tufts University, ESTI KRAMARSKI-WINTER, ASSAF VARDI, Weizmann Institute of Science, ROMAN STOCKER, Massachusetts Insitute of Technology — The growth and health of reef-building corals are limited by corals’ ability to exchange nutrients and oxygen with the surrounding, sometimes quiescent, seawater. Mass transport in coral systems has long been considered to occur passively as a result of molecular diffusion and the ambient fluid flow over the coral. Through a combination of microscale visualization experiments and numerical modeling, we demonstrate instead that motile cilia densely covering the coral surface – previously thought to serve cleaning and feeding purposes– actively stir the coral boundary layer by generating persistent vortices above the coral surface. This active mixing was observed over a variety of corals with differing surface geometries. We have quantified the contribution of ciliary surface vortices to mass transport, finding oxygen flux enhancements of 2 to 3 orders of magnitude under environmentally relevant ambient flow conditions. These results reveal a new, active role of the coral animal in regulating its mass transport by engineering its local hydrodynamic environment, an ability that may have an important role in the evolutionary success of reef corals. 8:26AM A9.00003 How a bacterial pathogen swims in the storm stirred up by its coral host , DOUGLAS BRUMLEY, MELISSA GARREN, VICENTE FERNANDEZ, ROMAN STOCKER, Massachusetts Institute of Technology — One important cause of the worldwide demise of coral reefs is the infection of corals by pathogenic bacteria. These bacteria are always motile, yet how they land on the coral surface remains unclear. In particular, the recently discovered vortical flows produced by the coral with its epidermal cilia create a hostile hydrodynamic environment for motility and the pursuit of chemical cues. We used high-speed imaging coupled with dual-wavelength epifluorescent microscopy to track individual Vibrio coralliilyticus bacteria - known for causing coral disease - in the immediate vicinity of its host, the coral Pocillopora damicornis. By simultaneously determining the fluid velocity and bacterial trajectories, we quantified the ability of the bacteria to target the coral surface. We show that the cilia-driven flows considerably but not entirely disrupt bacterial navigation towards the coral, as a result of (i) the stirring of the chemical cues guiding the cells and (ii) the shear-induced alignment of bacteria within the flow. By enabling the direct visualization of microbial motility in ciliary flows, this system can not only provide insights into coral disease, but also serve as a model system for bacterial disease in other ciliated environments, including the human respiratory system. 8:39AM A9.00004 Cilia beating patterns are not hydrodynamically optimal , HANLIANG GUO, University of Southern California, JANNA NAWROTH, Harvard University, YANG DING, EVA KANSO, University of Southern California — We examine the hydrodynamic performance of two cilia beating patterns reconstructed from experimental data. In their respective natural systems, the two beating patterns correspond to: (A) pumping-specialized cilia, and (B) swimming-specialized cilia. We compare the performance of these two cilia beating patterns as a function of the metachronal coordination in the context of two model systems: the swimming of a ciliated cylinder and the fluid pumping by a ciliated carpet. Three performance measures are used for this comparison: (i) average swimming speed/pumping flow rate; (ii) maximum internal moments generated by the cilia; and (iii) swimming/pumping efficiencies. We found that, in both models, pattern (B) outperforms pattern (A) in almost all three measures, including hydrodynamic efficiency. These results challenge the notion that hydrodynamic efficiency dictates the cilia beating kinematics, and suggest that other biological functions and constraints play a role in explaining the wide variety of cilia beating patterns observed in biological systems. 8:52AM A9.00005 Hydrodynamic interactions of bacteria and particles with ciliated surfaces , JANNA NAWROTH, Harvard University, JOHN DABIRI, Caltech — Cilia are microscopic, hair-like structures on the surface of cells that enable animals to interact with bacteria and fluid boundary layers. Here we present experimental data showing that, in addition to transporting fluids and particles along the surface, the coordinated movement of cilia ensembles generates 3-dimensional, rotational flow fields extending far beyond the length scale of individual cilia. Further, our results suggest that combining such vortices with adhesive stagnation zones creates particle traps that can be tuned to preferentially retaining particles with particular surface properties, and size, on the ciliated surface. 9:05AM A9.00006 The effects of translation and rotation on flagellar synchronization1 , JONATHAN H. TU, MURAT ARCAK, MICHEL MAHARBIZ, Univ of California - Berkeley — Synchrony is often observed in studies of swimming microorganisms. Examples include collective behavior in large populations of microswimmers, metachronal waves passing through arrays of cilia, and flagellar bundling. In this work, we focus on the hydrodynamic interactions that occur between flagella in close proximity. Specifically, we use the method of regularized Stokeslets to numerically investigate the precise mechanisms through which phase synchrony occurs in a pair of side-by-side rigid helices. Because our “end-pinned” model enforces restoring forces at a single end of each helix, we are able to isolate and compare the respective effects of translational and rotational motions. We find that while certain degrees of freedom promote synchrony, others promote anti-synchrony or have little effect. 1 Funded by ONR grant N000141310551 9:18AM A9.00007 Squirmers with swirl: a model for Volvox swimming , TIMOTHY PEDLEY, University of Cambridge, DOUGLAS BRUMLEY, M.I.T., TAKUJI ISHIKAWA, Tohoku University — A Volvox colony takes the form of a perfect sphere that swims because each cell on its surface has a pair of beating flagella. The flagella of the different cells are coordinated, almost certainly hydrodynamically [1], to beat approximately in a meridional plane, with axis of symmetry in the swimming direction, but with a roughly 10 degree azimuthal offset which means that the colonies rotate about their axes as they swim. Experiments on colonies held stationary on a micropipette show that the beating pattern takes the form of a symplectic metachronal wave [1]. Here we extend the Lighthill/Blake axisymmetric, Stokes-flow model of a free-swimming spherical squirmer to include azimuthal swirl. The kinematics of the metachronal wave are used to calculate the coefficients in the eigenfunction expansion and hence calculate the swimming speed and rotation rate (proportional to the square of the beating amplitude); measuring these provides a simple means of assessment of the flagellar beating parameters of individual colonies. Extension of the model to include colony interactions, with each other and a plane boundary, leads to simulations of Volvox “dancing”: the observed bound states of ref [2]. [1] D.R. Brumley et al, Phys. Rev. Lett., 109:268102,2012 [2] K. Drescher et al, Phys. Rev. Lett., 102:168101,2009 9:31AM A9.00008 Buckling and relaxation of an elastic filament in a viscous fluid under compression , MOUMITA DASGUPTA, Clark University, JULIEN CHOPIN, Departamento de Engenharia Civil COPPE/Universidade Federal do Rio de Janeiro, ARSHAD KUDROLLI, Clark University — We discuss an experimental investigation of buckling of an elastic filament in a viscous fluid under compressive loading in which an interplay of elastic and viscous forces are important to the structure observed dynamically. Buckling of an elastic filament in a viscous medium is a common phenomenon in soft matter and biological systems, examples of which include buckling instability during uniflagellated bacteria locomotion and formation of short wavelength curvature of microtubule in surrounding cytoskeleton. The experimental system consists of an elastic PDMS filament with clamped boundary condition immersed in a viscous fluid. One end of the filament is then compressed through a prescribed speed and distance. It buckles with a wavelength which decreases with increasing speed. The amplitude of the buckled mode is observed to decrease from the end which is moved. Over long times, the filament is observed to relax to the fundamental Euler buckling mode. Focusing on the initial buckling, we measure the shapes of filament and the fluid flow, in response to the compression, using PIV and high speed imaging. We thus estimate and discuss the relative viscous and elastic stresses experienced by the filament during the growth of the various modes as a function of compression speed. Sunday, November 23, 2014 8:00AM - 9:57AM — Session A10 Microscale Flows: Electrokinetics and Electrohydrodynamics sachusetts Institute of Technology 3005 - Cullen Buie, Mas- 8:00AM A10.00001 Influence of Film Thickness and Substrate Geometry on the Growth of Taylor Cones in Perfectly Conducting Films1 , THEODORE ALBERTSON, SANDRA TROIAN, California Institute of Technology, MC 128-95, Pasadena, CA 91125 — There are a growing number of space based applications ranging from miniature mass spectrometry devices to small focused ion beam units which rely on nanoscale fluid flow controlled by Maxwell stresses. These new technologies require ever improved understanding of the process by which a liquid stream or droplet is transformed from an initial smooth shape into a steepening cone known as the Taylor cone. While ongoing studies in the literature have elucidated how the cone-jet transition controls the delivery of mass and charge flux, less attention has been paid to the case of microscale films and how frictional effects influence the shape and timescale of evolving conical elongations. Here we describe recent efforts in our group using moving mesh and phase field methods to capture the influence of substrate geometry and film thickness on the formation of transient Taylor cones in perfectly conducting films. The computational model fully couples electrohydrodynamic fluid flow with active pressure and electrical potential fields resulting from the rapidly evolving film shape. We examine the asymptotic behavior of the film deformation process as a function of the electric field strength and substrate curvature, which in experimental systems can be easily tuned. 1 Financial support from a 2014 NASA Space Technology Research Fellowship is gratefully acknowledged. 8:13AM A10.00002 Experimental electrokinetic flow characteristics in cross-shaped microchannels with groove and wavy geometrical inhomegeneities , DIEGO OYARZUN, ALVARO SOCIAS, Universidad de Santiago de Chile, AMADOR GUZMAN, Pontificia Universidad Catolica de Chile — Electrokinetic flow instabilities (EKI) in cross-shaped microchannels occur when a critical local Rayleigh number is reached. To know the range of Rayleigh numbers when either stable or unstable electrokinetic flows happen is very important for suppressing or enhancing the EKI effect on processes such as injection or mixing. One way of enhancing flow mixing is by incorporating geometrical inhomogeneities on the microchannel walls such as groove and/or waves that can disturb or suppress convective and absolute instabilities for highly critical electrokinetic flow regimes. We, first experimentally investigate the flow pattern in a cross-shaped microchannel without groove and waves at the channel walls under an external electric field to determine the Rayleigh number range for stable and unstable flow regimes and the dominants frequencies associated to EKI. Then, we investigate the effect of grooves and waves at the microchannel walls on the electrokinetic flow characteristics for determining the effect of the existence of the geometrical inhomogeneities on the electrokinetic flow patterns for the range of sub-and super-critical Rayleigh numbers. Our primary results for local Rayleigh numbers based on a cross-shaped microchannel with flat walls indicate that at least for subcritical flow regimes there are no unstable, but stable flow regimes. 8:26AM A10.00003 Dynamics of micro-vortices induced by ion concentration polarization in electrodialysis , JOERI DE VALENCA, University of Twente & Wetsus, R.M. WAGTERVELD, Wetsus, ROB LAMMERTINK, PEICHUN AMY TSAI, University of Twente, SOFT MATTER, FLUIDICS AND INTERFACES GROUP, UNIVERSITY OF TWENTE TEAM, WETSUS TEAM — We experimentally investigate the coupled dynamics of global ion transport and local electroconvective flow of an electrolyte solution close to a charge selective membrane under an electric forcing. At small dc electric currents, due to the membrane permselectivity counterions (cations) transport diffusively through the cation exchange membrane (CEM) whereas the passage of co-ions (anions) is inhibited, thereby forming ion concentration polarization or gradients. At large currents, our simultaneous measurements of voltage drop and flow filed reveal several distinct dynamical regimes. Initially, the electrodialysis system exhibits a linear Ohmic electric resistance and then a rate-limiting regime with a voltage jump. Subsequently, electro-osmotic micro-vortices set in and grow linearly both in size and speed with time. After this linearly growing electroconvective regime, the measured voltage drop levels off around a fixed value. The average vortex size and speed saturate as well, however the individual vortices are unsteady and dynamical. Furthermore, the influence of micro-patterned CEM on the couple dynamics will be presented and discussed. 8:39AM A10.00004 Investigation of Material Dependence in Electrothermal Vortex , JIAN WEI KHOR, AVANISH MISHRA, Purdue Univ, XUDONG PAN, Harbin Institute of Technology, STEVEN WERELEY, Purdue Univ — Rapid Electrokinetic Patterning (REP) is an optoelectric method which can be used for manipulation of diverse set of particles with laser. However, requirement of high laser intensity remains a stumbling block to proliferation of this method. In this presentation, we demonstrate that careful selection of the electrode material is critical in producing a cost effective REP chip that requires low laser power. The electrodes in REP provide two aspects of the phenomenon; electrical conduction from the electrical power source that produces the electric field and heat absorption from the laser heating that produces the temperature gradient. Consequently, the physical and thermal properties of the electrodes used in REP are crucial for the formation of the electrothermal vortex, which plays a major role in REP as a manipulation technique. Currently, Indium Tin Oxide (ITO) layer is used as the electrodes. ITO produces the electrothermal vortex needed for REP and also provides a viewing window in to the chip. In this study, possibilities of utilizing other materials to produce equivalent or better REP effects than ITO will be investigated. 8:52AM A10.00005 Highly selective creation of hydrophilic micro-craters on super hydrophobic surface using electrohydrodynamic jet printing1 , JAEHYUN LEE, SANGYEON HWANG, FARIZA DIAN PRASETYO, Sungkyunkwan University, VU DAT NGUYEN, Enjet Incorporation, JUNGWOO HONG, JENNIFER H. SHIN, Korea Advanced Institute Science and Technology, DOYOUNG BYUN, Sungkyunkwan University — Selective surface modification is considered as an alternative to conventional printing techniques in high resolution patterning. Here, we present fabrication of hydrophilic patterns on the super hydrophobic surface, which makes structure on the hydrophilic region. The super hydrophobic surface is able to be chemically changed to hydrophilic with alcohols. As a consecutive process, electrohydrodynamic (EHD) jet printing was utilized to fabricate local hydrophilic craters with 30-200 µm sizes. 3 kinds of target liquids were deposited well on hydrophilic region; PEDOT (poly 3,4 ethylenediocythiophene), polystyrene nano-particles, and salmonella bacteria medium. Additionally, qualitative analysis were presented for modification mechanism and surface properties on super hydrophobic/hydrophilic by analysis of surface energy with contact angle, SEM (scanning electron microscopy) image, and SIMS (secondary ion mass spectroscopy) analysis. This new simple modification method provides possibility to be utilizing in bio-patterning engineering such as cell culturing microchip and lab on a chip. 1 This research was supported by the Basi Science Research Program through the National Research Foundation of Korea (NRF) (Grand number : 2014-023284) 9:05AM A10.00006 Characterization of self-assembled colloidal particle bands in combined electroosmotic and Poiseuille flow1 , NECMETTIN CEVHERI, MINAMI YODA, Georgia Institute of Technology — Periodic and steady electric fields have long been used to manipulate and assemble colloidal particles suspended in conducting fluids, usually aqueous solutions. Most of these studies have, however, focused on suspensions at rest. Recent studies have shown that a combination of steady electric fields and shear flow can be used to manipulate radii a = 245 nm particles in a dilute suspension flowing through ∼ 30 µm deep microchannels. When the electric field is in the opposite direction from the Poiseuille flow (which is essentially simple shear flow near the wall), the particles are first attracted to the wall, then self-assemble into nearly periodic concentrated bands aligned with the flow direction, as the electric field magnitude |E| increases. This talk will discuss the characteristics of these bands, e.g. how their average spacing depends on |E| and the near-wall shear rate γ̇, as well as the dynamics of the particles within the bands, which are moving in the same direction as the flow and appear to be in a disordered liquid (vs. crystalline) state. Bands only form above a threshold value of |E|, and this value depends on parameters such as γ̇, the particle radius a, and the particle zeta-potential ζp . 1 Supported by NSF 9:18AM A10.00007 Repulsive and Attractive Colloidal Particle-Wall Interactions in Poiseuille and Electroosmotic Flow through Microchannels1 , MINAMI YODA, NECMETTIN CEVHERI, Georgia Institute of Technology — Manipulating near-wall polystyrene particles in a dilute suspension flowing through a microchannel is important in microfluidics. Such particles experience wall-normal lift forces in electroosmotic (EO) flows driven by electric fields of magnitude E, and in Poiseuille flows driven by pressure gradients ∆p/L beyond the forces predicted by DLVO theory. Recent evanescent-wave particle tracking studies of combined EO and Poiseuille flow have shown that 245 nm radii particles are repelled from, or attracted to, a fused-silica wall when the EO flow is in the same, or opposite, direction as the Poiseiulle flow, respectively. Estimates of the lift force magnitude F suggest that it scales with the shear rate γ̇ for Poiseuille flow, and not the γ̇ 2 typical of electroviscous lift. Surprisingly, when the force is repulsive, F exceeds the sum of the forces observed for EO flow at the same E and Poiseuille flow at the same ∆p/L and appears to scale as γ̇ 1/2 . Furthermore, in both cases F appears to be proportional to E, suggesting that this lift force is distinct from those observed in “pure” EO and “pure” Poiseuille flows. 1 Supported by NSF 9:31AM A10.00008 Electrohydrodynamic manipulation of particles on drop surfaces1 , EDISON AMAH, KINNARI SHAH, IAN FISCHER, PUSHPENDRA SINGH, NJIT — We have recently shown that particles adsorbed on the surface of a drop can be self-assembled at the poles or the equator of the drop by applying a uniform ac electric field, and that this method can be used to separate on the surface of a drop those particles experiencing positive dielectrophoresis from those experiencing negative dielectrophoresis. In this talk we show that the frequency of the electric field is an important parameter which can be used to modify the intensities of the dielectrophoretic and the hydrodynamic-flow induced forces, and thus control the distribution of self-assembled monolayers. 1 The work was supported by National Science Foundation 9:44AM A10.00009 Effect of electrode geometry on field strength in plastic microfluidic devices and application to cell membrane permeabilization , MARC CHOOLJIAN, UC Berkeley-UCSF Graduate Program in Bioengineering, JACOBO PAREDES, DORIAN LIEPMANN, UC Berkeley — We have developed a method that allows embedding of electrodes in up to 3 walls of a plastic microfluidic channel. Electric field strength and homogeneity of various electrode geometries is analyzed theoretically and experimentally by evaluating the efficiency of on-chip lysis of cells. Electric field-mediated disruption of membranes is an important tool in diagnostics, basic biology, and synthetic biology due to the ability to permeabilize the cell membrane without changing the chemical composition of the buffer. Typically, fields of the required magnitude are applied to the cell by discharging a capacitor through a mixture of cells in a cuvette, resulting in a transient high-voltage pulse. We demonstrate that is possible to substitute a spatially varied DC electric field along a microchannel and to control the timing of the pulses by changing the electrode spacing and the flow rate. Homogeneity of the field with respect to the cross section of the channel is key to achieving critical field strength regardless of the cell’s lateral position in the channel. A comparison of 2D versus 3D electrode geometries on the efficiency of electroporation and on side-effects arising due to the electric field (recirculating flows and hydrolysis) is presented. Sunday, November 23, 2014 8:00AM - 9:57AM Session A11 Rotating Flows I — 3007 - Patrice Meunier, L’Institut de Recherche sur les Phenomenes Hors Equilibre 8:00AM A11.00001 DNS and PIV investigation of nonlinear instability in a precessing cylinder flow1 , HUGH BLACKBURN, THOMAS ALBRECHT, Monash University, RICHARD MANASSEH, Swinburne University of Technology, JUAN LOPEZ, Arizona State University, PATRICE MEUNIER, IRPHE, CNRS — Direct numerical simulation results for flow inside a spinning, precessing cylinder of fluid corresponding to a previous experimental study and its extensions are presented and analysed in relation to experimental results and weakly nonlinear theory based on triad interaction of inviscid Kelvin modes. Simulation outcomes agree well with the experimental results both qualitatively and quantitatively, and additional processing reveals more in-depth support for the weakly nonlinear theory than could be demonstrated in the experiments. Additionally, numerical results provide meridional and azimuthal mean flow data. 1 Supported by Australian Research Council Discovery Program Grant DP130101744 8:13AM A11.00002 Precessional forcing of a mean geostrophic flow in a rotating cylinder , THOMAS ALBRECHT, Monash University, Australia, PATRICE MEUNIER, RPHE, CNRS, and Aix-Marseille Universit ?e, Marseille, France, HUGH BLACKBURN, Monash University, Australia, JUAN LOPEZ, Arizona State University, USA, RICHARD MANASSEH, Swinburne University of Technology, Australia — It has often been observed that inertial waves in rotating flows can interact nonlinearly to create a mean geostrophic motion due to a streaming effect. This mean geostrophic flow has a large effect in rotating flows since it changes the base flow and thereby detunes all the possible resonances. However, in a cylinder, inviscid Kelvin modes (KM) are known theoretically to create no mean geostrophic motion by nonlinear coupling. It was thus assumed that the observed geostrophic flow relies on viscous effects in the Ekman boundary layers together with nonlinear interaction. We present here a simple flow configuration where both the KM and the geostrophic flow can be quantified in order to analyse this mechanism in detail. We have studied the case of a KM forced by precession. This allows to reach a very large amplitude of the KM at the resonance, even for small precession angles. PIV measurements are compared to numerical simulations. The profiles of mean azimuthal velocity are studied in the laminar and in the turbulent case. They seem to be correlated to the profiles of velocity of the forced KM. The amplitude of the geostrophic flow seems to agree with the viscous nonlinear theory which predicts that it scales as the square of the forced KM’s amplitude. 8:26AM A11.00003 Behind the Rotating Flattop: when vortex meets a deformable surface , J.-C. TSAI, Inst. of Physics, Academia Sinica, Taipei, Taiwan, Y.-C. SUN, National Taiwan Normal University, Taipei, Taiwan, K.-H. HUANG, Nationa Taiwan University, Taipei, Taiwan, C.-Y. LAI, Princeton University, C.-Y. TAO, Inst. of Physics, Academia Sinica, Taipei, Taiwan, J.-R. HUANG, National Taiwan Normal University, Taipei, Taiwan — We study experimentally a two-fluid system, driven by a rotating upper boundary, inside a stationary cylinder. For a range of aspect ratios, the interface displays various changes with driving rates, with the most striking being the formation of a plateau. Direct imaging and flow visualization allow us to identify the interplay between the morphology of our two-fluid interface and the vortex loops reported previously in literatures, in a way that we can rigorously define the transition by the switch of topology in the flow structure rather than just the shape of the free surface. Further extensions of the parameter space show a wealth of phenomena involving various instabilities on the interface that call for further understanding. 8:39AM A11.00004 Vibrational Dynamics of Light Body in Rotating Cavity with Liquid1 , NIKOLAI KOZLOV, STANISLAV SUBBOTIN, Laboratory of Vibrational Hydromechanics, PSHPU, Perm — Dynamics of a light body of cylindrical or spherical shape in a rotating cavity (cylindrical or spherical) with liquid is studied. The system is set at rotation, the body occupies a steady position near the cavity axis under the action of centrifugal force. Action of an external periodic force excites inertial oscillations of the body and, as consequence, its differential rotation. The mechanism of the latter is the generation of an average mass force in a viscous boundary layer on the oscillating body surface [Fluid Dyn. 43, 9 (2008); 47, 683 (2012)]. In experiments, two types of external action are used. Rotation of a horizontal cavity in the gravity field leads to circular body oscillations with the frequency of rotation; as a result the body rotates slower than the cavity. External vibration, perpendicular to the rotation axis, leads to a resonant excitation of intensive body oscillations; as a result the body spins in the cavity rotation direction (outrunning rotation), or in the opposite (lagging rotation). The eigenfrequency of rotating system is mainly determined by the ratio of vibration and rotation frequencies n = Ωv /Ωr . Body motion intensity is determined by the dimensionless acceleration Γ = g/Rs Ω2r or Γv = bv Ω2v /Rs Ω2r . 1 The work is supported by Russian Scientific Foundation (project N. 14-11-00476). 8:52AM A11.00005 Flows in rotating cavity excited by oscillating solid core1 , STANISLAV SUBBOTIN, VICTOR KOZLOV, NIKOLAI KOZLOV, Laboratory of Vibrational Hydromechanics, PSHPU — The flow excited by oscillations of free core in a rotating about horizontal axis cavity filled with liquid is experimentally investigated. The core is lighter than the liquid and is located near the rotation axis under the action of centrifugal force. The action of the gravity force field on the rotating system leads to the tidal oscillations of the core. As a result of the pulsating motion in the Stokes boundary layer the average mass force arises, spinning the core relative to the cavity. The phenomenon of the differential rotation was called a “vibrational hydrodynamic top” [Dokl. Phys. 52, 458 (2007)]. The core differential rotation leads to the formation of the flow in the form of a Taylor column. At slow differential rotation the column has the shape of a circular cylinder. Different types of instability manifest themselves: firstly – the excitation of 2D vortex system inside the column; secondly – the excitation of 2D azimuthal waves on the column boundary [Dokl. Phys. 59, 40 (2014)]. The nonlinear interaction of different instability modes resulting in synchronization of phase velocities and wave numbers is found. The stability of the structures of different type is determined by Reynolds number. 1 The work is supported by Russian Scientific Foundation (project N.14-11-00476) 9:05AM A11.00006 Experimental investigation of the rotating flow in a low speed axial compressor , LICHAO JIA, YIDING ZHU, HUIJING YUAN, CUNBIAO LEE1 , Peking Univ — This paper presents detailed experimental data on the flow and turbulence within the boundary layer of an axial compressor rotor blade. The velocity distribution of the entrance of the rotor has also been detected, which will be useful to determine the boundary condition in simulation. The experiments are performed in a low speed wind tunnel at different flow fluxes. During the experiments, high-resolution 2D Particle Image Velocimetry (PIV) measurements are conducted at different axial and radial positions. Phase-locking, boundary detection and virtual particle images methods are used to improve the performance of the PIV. Both the mean and instantaneous internal flow fields of the axial compressor are presented here. The experiments enrich the understanding of the rotating flow phenomenon in the low speed axial compressor. 1 Corresponding author 9:18AM A11.00007 From Newton’s bucket to rotating polygons: experiments on surface instabilities in swirling flows , TOMAS BOHR, BJARNE BACH, MALENE VESTED, ANDERS ANDERSEN, Physics Department, Technical University of Denmark, ERIK LINNARTZ, Physics of Fluids Group, University of Twente — We present an experimental study of “polygons” forming on the free surface of a swirling turbulent water flow in a partially filled cylindrical container, where the rotation of the bottom plate and the cylinder wall is controlled independently. Thus we can move from a rigidly rotating “Newton’s bucket” flow to one where bottom and cylinder walls are rotating oppositely and the surface is turbulent but flat on average. Between those two extremes, we find polygonal states in two distinct bands. Further, we find a “monogon,” a figure with one corner, roughly an eccentric circle rotating in the same sense as the cylinder. We show that the system has a surprising multi-stability and excitability, and that small details can change the stability of polygon states. We investigate accurately the rotation of the plate compared to that of the polygon. Although the the frequency ratios can be close to rational, we do not find phase locking. 9:31AM A11.00008 Three States of Counter–Rotating Turbulent Taylor–Couette Flow , SEDAT TOKGOZ, GERRIT E. ELSINGA, RENE DELFOS, JERRY WESTERWEEL, Delft University of Technology, The Netherlands — In this study we experimentally investigate the change of torque at constant shear Re, and its relation to the coherent flow structures in turbulent Taylor-Couette (TC) flow. Torque measurements at counterrotating turbulent regimes show a change depending on the rotation number. In order to understand the mechanism behind this change we used tomographic PIV and measured the instantaneous 3D flow structures in turbulent TC flow. The instantaneous flow fields are decomposed into large (ILS) and smaller-scale (ISS) motions to study their contributions separately. Three distinctive flow states were found at counterrotating turbulent flow, associated with clear changes in the ILS and ISS structure. Close to only inner cylinder rotation, where well-organised Taylor-vortex-like flow structures are observed, the mean flow is responsible for the torque values. Close to exact-counter rotation, inclined ILS vortices induce velocities in the azimuthal and radial directions, contributing significantly to the torque. Close to only outer cylinder rotation the ILS vortices start to align themselves in the axial direction, resembling co-rotating Taylor column-like structures, which reduces the measured torque. The change of the orientation of the ILS vortices is also confirmed quantitatively. 9:44AM A11.00009 Measurements of small radius ratio turbulent Taylor-Couette flow , ROELAND VAN DER VEEN, SANDER HUISMAN, University of Twente, SEBASTIAN MERBOLD, Brandenburg University of Technology, CHAO SUN, University of Twente, UWE HARLANDER, CHRISTOPH EGBERS, Brandenburg University of Technology, DETLEF LOHSE, University of Twente — In Taylor-Couette flows, the radius ratio (η = ri /ro ) is one of the key parameters of the system. For small η, the asymmetry of the inner and outer boundary layer becomes more important, affecting the general flow structure and boundary layer characteristics. Using high-resolution particle image velocimetry we measure flow profiles, local transport, and statistical properties of the flow for a radius ratio of 0.5 and a Reynolds number of up to 4 · 104 . By measuring flow profiles at varying heights, roll structures are characterized for two different rotation ratios of the inner and outer cylinder. In addition, we systematically vary the rotation ratio and the Reynolds number. These results exemplify how curvature affects flow in strongly turbulent Taylor-Couette Flow. Sunday, November 23, 2014 8:00AM - 9:57AM Session A12 Drops: Heat Transfer and Evaporation I — 3018 - Justin Burton, Emory University 8:00AM A12.00001 Apparent contact angles induced by evaporation into air: interferometric measurements and lubrication-type modeling1 , PIERRE COLINET, YANNIS TSOUMPAS, SAM DEHAECK, ALEXEY RED- NIKOV, Universite Libre de Bruxelles, TIPs - Fluid Physics — For volatile liquids, finite contact angles on solid substrates can occur even in the case of perfect wetting, immobile contact lines and ideally smooth surfaces. This is a fluid-dynamic effect due to evaporation typically intensifying towards a small vicinity of the contact line. In the present talk, we first overview recent theoretical results on the subject, where we focus primarily on the case of diffusion-limited evaporation into air. The model is based upon the so-called de Gennes’ paradigm, incorporating simultaneously the spreading coefficient and the disjoining pressure in the form of an inverse cubic law. Then we carry out comparison with experimental results for the contact angles of evaporating sessile drops of several perfectly-wetting HFE liquids of different volatility recently obtained by Mach-Zehnder interferometry. The scaling-type theoretical prediction for the apparent contact angle is found to be in good agreement with experimental measurements. Another model based upon the Kelvin effect (curvature dependence of the saturation conditions) is also briefly discussed, an important conceptual feature of which being that contact-line singularities (both evaporation- and motion-induced) can be fully regularized, in contrast with the first model. 1 Support from ESA, BELSPO and FRS-FNRS is gratefully acknowledged. 8:13AM A12.00002 Evaporation of a drop on a flat solid substrate with pinned & perfect slip contact line , AMIRHOSSEIN AMINI, G.M. HOMSY, Mechanical Engineering, University of Washington — We study the evolution of the profile of a 2D axisymmetric, incompressible, Newtonian droplet while evaporating on a flat solid substrate. The droplet has an initial circular cross section, the surface tension and the temperature of the solid-liquid interface are constant, and gravity and van der Waals effects are neglected. We deploy the one-sided model1 which, together with the lubrication approximation, results in an evolution equation for the local height of the droplet. The evolution equation is a nonlinear partial differential equation that is 4th order in space and 1st in time and which is solved numerically using the method of lines. The problem is governed by several parameters, the key being the contact line condition and the wall superheat. For the case in which the contact line is pinned, we predict the drop thickness and contact angle as a function of time over a wide range of parameters. Interestingly, we observe a new self-similar regime near the end of the droplet evaporation and derive scaling laws from the numerical solutions. These results are contrasted with those for the case of perfect slip. 1 Burelbach et al. Journal of Fluid Mechanics, 195 463-494 (1988) 8:26AM A12.00003 Local analysis of the contact region of an evaporating sessile drop , S.J.S. MORRIS, Department of Mechanical Engineering, University of California, Berkeley — In experiments by Guéna et al. (2007), a drop of perfectly wetting pure liquid evaporates from a non–heated substrate at a rate controlled by vapour diffusion. The drop spreads until reaching a radius a determined by initial drop volume; the apparent contact line then reverses direction. The apparent contact angle measured at reversal was found experimentally to vary as a−1/6 for a < 1 mm (about); for larger drops θ decreases more strongly. Local analysis (Morris J. Fluid Mech. 739: 308–337. 2014) predicts that θ ∝ a−1/6 ; for the smaller drops obeying the 1/6th rule, predicted values agree with experiment to within 10–30%. Though the behaviour of drops smaller than the capillary length thus appears to be understood, that of larger drops is not. 8:39AM A12.00004 Atmospheric convective transport contribution to evaporative sessile droplets , FLORIAN CARLE, SERGEY SEMENOV, MARC MEDALE, DAVID BRUTIN, Aix-Marseille University - IUSTI UMR 7343 — The scien- tific community struggles with the creation of an accurate quantitative description of sessile droplet evaporation flux rate. The classically used description considers evaporation as a quasi-steady process controlled by the diffusion of vapor into the air, and the whole system is assumed to be isothermal at the ambient temperature. However, when two types of fluids (alcohols and alkanes) are let to evaporate on heated substrates while a side view camera measures their evaporation flux rate, droplets tend to see their evaporation flux rate underestimated by this model mostly due to convection. This experimental study aims to understand how atmospheric convective transport in the vapor phase influences evaporation in order to developed an empirical model that describes with accuracy the evaporation flux rate. The Rayleigh number is used to analyze the contribution of natural convection and an empirical model is developed combining diffusive and convective transport for each type of fluid. The influence of the molecular chain length (and the increasing number of carbon atoms) is also being discussed. 8:52AM A12.00005 Two-phase DNS of evaporating drops with 3D phenomena and contactline dynamics1 , PRASHANT VALLURI, PEDRO J. SÁENZ, KHELLIL SEFIANE, The University of Edinburgh, OMAR K. MATAR, Imperial College London — A novel 3D two-phase model based on the diffuse-interface method is developed to investigate the fully-coupled two-phase dynamics of a sessile drop undergoing evaporation on a heated substrate. General transient advection-diffusion transport equations are implemented to address the conservation of energy and vapour in the gas phase, which also allows the more realistic modelling of interface mass and energy transport based on local conditions. The emphasis of this investigation is on addressing three-dimensional phenomena during evaporation of drops with non-circular contact area. Irregular drops lead to complex interface shapes with intricate contract-angle distributions along the triple line and with a three-dimensional flow which previous axisymmetric approaches cannot show. The versatility of this model also allows the simulation of the more complex case of drops evaporating with a moving contact line. Both constant-angle (CA) and constant-radius (CR) modes of pure evaporation are successfully simulated and validated against experiments. 1 ThermaPOWER project (EU IRSES-PIRSES GA-2011-294905) 9:05AM A12.00006 On the emergence of vortices in irregular evaporating sessile droplets1 , PEDRO J. SÁENZ, DIMITRIOS MAMALIS, KHELLIL SEFIANE, PRASHANT VALLURI, The University of Edinburgh, OMAR K. MATAR, Imperial College London — The spontaneous development of 3D azimuthal vortices parallel to the plane of substrate in an evaporating drop of water with irregular contact area is reported by means of experiments and direct numerical simulations (DNS). In spherical droplets, the non-uniform evaporation flux leads to a 2D axisymmetric flow with fluid being transported along the interface from the contact line (hotter) towards the apex (colder) due to the Marangoni effect. However, infrared recordings of a non-spherical drop show the break of symmetry and the consequent development of a preferential direction for thermocapillary convection. As a result, counter-rotating whirling currents emerge in the drop playing a critical role in regulating the interface thermal motion. This geometry-induced phenomenon is also investigated via simulations with a fully-coupled two-phase model. DNS show good agreement with experiments and reveal the intricate drop dynamics due to this geometry-induced phenomenon. The triggering mechanism is analysed along with the resulting bulk flow. 1 ThermaPOWER project (EU IRSES-PIRSES GA-2011-294905) 9:18AM A12.00007 The effect of vapor diffusion and unsteady heating on the evolution of a sessile droplet on a substrate , MAHNPRIT JUTLEY, VLADIMIR AJAEV, Southern Methodist University — The behavior of a sessile droplet on a heated substrate has been a topic of interest due to its subtle dependencies on the surrounding environmental conditions and its many applications, such as the coating of a solid substrate with another material, the spray cooling of electronics, and DNA microarray technology to name a few. Prediction of the height evolution of a sessile droplet on heated substrate is governed by the unsteady heating of the substrate, the vapor diffusion into the atmosphere above the droplet, and the effects of surface tension, gravity, thermocapillarity, and disjoining pressure. Using lubrication theory and developing coupling relationships between the heat equation in the substrate, height evolution equation of the droplet, and the vapor diffusion equation in the atmosphere, the system of coupled partial differential equations can be derived and solved. Connection of the numerical simulation to experimental studies is discussed. 9:31AM A12.00008 How surfactants influence evaporation-driven flows , ROBERT LIEPELT, ALVARO MARIN, MASSIMILIANO ROSSI, CHRISTIAN J. KÄHLER, Bundeswehr University Munich — Capillary flows appear spontaneously in sessile evaporating drops and give rise to particle accumulation around the contact lines, commonly known as coffee-stain effect (Deegan et al., Nature, 1997). On the other hand, out-of-equilibrium thermal effects may induce Marangoni flows in the droplet’s surface that play an important role in the flow patterns and in the deposits left on the substrate. Some authors have argued that contamination or the presence of surfactants might reduce or eventually totally annul the Marangoni flow (Hu & Larson, J. Phys. Chem. B, 2006). On the contrary, others have shown an enhancement of the reverse surface flow (Sempels et al., Nat. Commun., 2012). In this work, we employ Astigmatic Particle Tracking Velocimetry (APTV) to obtain the 3D3C evaporation-driven flow in both bulk and droplet’s surface, using surfactants of different ionic characters and solubility. Our conclusions lead to a complex scenario in which different surfactants and concentrations yield very different surface-flow patterns, which eventually might influence the colloidal deposition patterns. 9:44AM A12.00009 Free surface profile of evaporative liquids at the vicinity of the contact line , SAMMY HOUSSAINY, PIROUZ KAVEHPOUR, University of California, Los Angeles — Interfacial phenomenon, specifically those associated with evaporation from thin liquid films near the contact line of a liquid drop, play a major role in many current engineering applications which require high local heat fluxes, as evident in heat pipes, grooved evaporators, fuel cells and suction nucleate boiling devices. This study will prove useful in the improvement of such applications. Fluoresces microscopy was used as our main technique of investigating the free surface profiles of evaporative liquids, as it delivers sufficient range and resolution to address the challenge of capturing the microscopic and macroscopic aspects of this phenomenon. Subsequent to our experimental findings, the results are compared with non-volatile liquids for both contact angle and free surface structures. Sunday, November 23, 2014 8:00AM - 9:57AM Session A13 Drops: Levitation and Superhydrophobic Surfaces — 3020 - Konrad Rykaczewski, MIT 8:00AM A13.00001 Surfing a magnetic wave , ELINE DEHANDSCHOEWERCKER, Laboratoire d’Hydrodynamique de l’Ecole Polytechnique (LadHyX), DAVID QUERE, Laboratoire de Physique et Mecanique des Milieux Heterogenes (PMMH), CHRISTOPHE CLANET, Laboratoire d’Hydrodynamique de l’Ecole Polytechnique (LadHyX) — Surfing is a free surface sport in which the athlete rides a wave standing on a board. However, any object plunged into the water or put on its surface is not always captured by an approaching wave, just like the classic example of a fisching float. So, a particle can be captured or not by a wave. Two regimes are defined : surf (captured) and drift (not captured). We focus on the question of the transition between these two regimes. Here we address the question with a magnetic wave. We have developed an experimental setup which allows the control of all relevant physical parameters. Liquid oxygen, which is paramagnetic and undergoes Leidenfrost effect, can be used to ensure magnetic and frictionless particles. A permanent magnet in translatory movement allows us to create a definite magnetic wave. We discuss the motion of oxygen drops deposited on an smooth and horizontal surface above an approaching magnet. First we show the existence of a critical speed below which drops are captured and determine how it depends on the velocity and intensity of the magnetic wave. Then we experimentally investigate the parameters that would affect the movement of drops in each regime. Finally, models have been developed to interpret magnetic drops motion and guarantee an efficient control. 8:13AM A13.00002 Acoustical radiation torque and force for spheres and Bessel beam extinction efficiency , PHILIP L. MARSTON, Washington State University, Pullman, LIKUN ZHANG, University of Texas, Austin — The scattering of optical and acoustical beams is relevant to the levitation and manipulation of drops. Here we examine theoretical developments in the acoustical case. We previously showed how the optical theorem for extinction can be extended to invariant beams. The example of a sphere in a Bessel beam facilitates the direct comparison with a circular disc computed using Babinet’s principle and the Kirchhoff approximation (P. L. Marston, J. Acoust. Soc. Am. 135, 1668-71 (2014)). In related work, by considering traveling or standing wave first-order vortex beams we previously showed that the radiation torque is the ratio of the absorbed power and the radian acoustic frequency (L. Zhang and P. L. Marston, Phys. Rev. E. 84, 065601 (2011)). By modifying the scattering to account for the viscosity of the surrounding fluid in the analysis of the absorbed power, approximations for radiation torque and force are obtained at long wavelengths in special cases (P. L. Marston, Proc. Meetings on Acoustics 19, 045005 (2013)) and these can be compared with results published elsewhere. 8:26AM A13.00003 The performance and operating mechanism of the ultrasonic scrubber1 , J.R. SAYLOR, W. RAN, Clemson University — An ultrasonic standing wave field is commonly used to levitate drops, facilitating drop studies in several ways. In the typical use of such a standing wave field, drops are simply placed at the node of the field and thereby levitated. However, it is also true that any particles or drops located in the general vicinity of the nodes of an ultrasonic standing wave are drawn toward the nodes where they accrue. We have shown that this effect can be used to create an “ultrasonic scrubber”, wherein the combination of a fine water mist and an ultrasonic standing wave field is used to remove particles (e.g. particulate pollutants) from a gas flow directed at the field (Ran, Saylor, & Holt, J. Aerosol Sci., 67, 104-118 (2014)). In this talk details are presented of the operating mechanism responsible for the success of this approach to scrubbing. The results of an experimental study are also presented showing the effect of the gas flow rate and droplet size distribution on the scavenging coefficient for one version of the ultrasonic scrubber. 1 Support from NSF is gratefully acknowledged. 8:39AM A13.00004 Levitation, Herringbones and Propulsion , HELENE DE MALEPRADE, DAN SOTO, CHRISTOPHE CLANET, DAVID QUERE, PMMH, ESPCI / LadHyX, Ecole Polytechnique — Controlling objects motion without contact is a major application issue as it ensures high mobility, low friction and no contamination. Levitation can be induced by blowing air from below through a porous medium, to create a thin air cushion under the object. The airflow is isotropic but if some asymmetry is introduced to rectify it, the levitating object can be controlled and propelled [1]. In our experiments, microscopic textures are engraved on the top of the porous medium, which directs the airflow. The resulting viscous entrainment enables drops, rigid plastic or even glass cards to self-propel [2]. If the micro-textures are displayed on the propelled object, the direction of the motion is reversed, which is found to result from a different mechanism of entrainment. [1] Linke, H. et al, Self-Propelled Leidenfrost Droplets, Phys. Rev. Lett. 96, 2006 [2] Soto, D. and Lagubeau, G. and Clanet, C. and Quere, D. Surfing on a Herringbone, in prep., 2014 8:52AM A13.00005 Free oscillations of a magnetic drop1 , ERIC FALCON, TIMOTHEE JAMIN, YACINE DJAMA, Paris Diderot University, MSC, CNRS, Paris, France — When a flattened drop of liquid is put on a substrate subjected to vertical vibrations, it undergoes a parametric instability above a critical acceleration. An azimuthal pattern occurs around the drop and oscillates at half the forcing frequency: a star-shaped drop is then observed made of several oscillating lobes. Here, we use a drop of ferrofluid, a magnetic liquid that responds to an external magnetic field. We report an experimental study of a ferrofluid drop on a superhydrophobic substrate vertically vibrated in presence of a weak vertical magnetic field of tunable amplitude. We find that the eigenmode frequencies of the drop are shifted by the magnetic field. We show that this shift is due to an interaction between the magnetic field and standing waves on the drop top. 1 This work was supported by ANR Turbulon 12-BS04-0005. 9:05AM A13.00006 Active surfaces: Ferrofluid-impregnated surfaces for active manipulation of droplets , KARIM KHALIL, SEYED REZA MAHMOUDI, Massachusetts Inst of Tech-MIT, NUMAN ABU-DHEIR, KFUPM, KRIPA VARANASI, Massachusetts Inst of Tech-MIT — Droplet manipulation and mobility on non-wetting surfaces is of practical importance for diverse applications ranging from micro-fluidic devices, anti-icing, dropwise condensation, and biomedical devices. The use of active external fields has been explored via electric, acoustic, and vibrational, yet moving highly conductive and viscous fluids remains a challenge. Magnetic fields have been used for droplet manipulation; however, usually, the fluid is functionalized to be magnetic, and requires enormous fields of superconducting magnets when transitioning to diamagnetic materials such as water. Here we present a class of active surfaces by stably impregnating active fluids such as ferrofluids into a textured surface. Droplets on such ferrofluid-impregnated surfaces have extremely low hysteresis and high mobility such that they can be propelled by applying relatively low magnetic fields. Our surface is able to manipulate a variety of materials including diamagnetic, conductive and highly viscous fluids, and additionally solid particles. 9:18AM A13.00007 Visualization of Buoyant Convection in Droplets on Superhydrophobic Surfaces , SUSMITA DASH, ADITYA CHANDRAMOHAN, JUSTIN WEIBEL, SURESH GARIMELLA, Purdue University — We investigate hitherto unreported flow characteristics that are manifested inside a sessile droplet when evaporating on a superhydrophobic surface. Evaporative cooling at the droplet interface establishes a temperature gradient that induces buoyant convection inside the droplet. A single rotating vortex, with a solid body rotation flow pattern, is observed using Particle Image Velocimetry. This flow pattern develops due to the large height-to-diameter aspect ratio of the droplet, which dictates a stable buoyancy-induced convection mode with one rolling vortex. The flow velocity is an order of magnitude higher compared to droplets evaporating on hydrophobic substrates. The high recirculation velocity, combined with the sliding contact line of the droplet, mitigates deposition of particles on the substrate during the evaporation process and enables a single concentrated deposition after complete drying on superhydrophobic substrates. 9:31AM A13.00008 Multifunctional polymer nano-composite based superhydrophobic surface1 , TANMOY MAITRA, ASHISH ASTHANA, Laboratory of Thermodynamics in Emerging Technologies, Department of Mechanical and Process Engineering, ETH Zurich, ROBERT BUCHEL, Particle Technology Laboratory, Department of Mechanical and Process Engineering, ETH Zurich, MANISH K. TIWARI, Mechanical Engineering, University College London, DIMOS POULIKAKOS, Laboratory of Thermodynamics in Emerging Technologies, Department of Mechanical and Process Engineering, ETH Zurich — Superhydrophobic surfaces become desirable in plethora of applications in engineering fields, automobile industry, construction industries to name a few. Typical fabrication of superhydrophobic surface consists of two steps: first is to create rough morphology on the substrate of interest, followed by coating of low energy molecules. However, typical exception of the above fabrication technique would be direct coating of functional polymer nanocomposites on substrate where superhydrophobicity is needed. Also in this case, the use of different nanoparticles in the polymer matrix can be exploited to impart multi-functional properties to the superhydrophobic coatings. Herein, different carbon nanoparticles like graphene nanoplatelets (GNP), carbon nanotubes (CNT) and carbon black (CB) are used in fluropolymer matrix to prepare superhydrophobic coatings. The multi-functional properties of coatings are enhanced by combining two different carbon fillers in the matrix. The aforementioned superhydrophobic coatings have shown high electrical conductivity and excellent droplet meniscus impalement resistance. Simultaneous superhydrophobic and oleophillic character of the above coating is used to separate mineral oil and water through filtration of their mixture. 1 Swiss National Science Foundation (SNF) grant 200021 135479. 9:44AM A13.00009 Is micro-nano texture the only reason for under-water superoleophobicity of fish scale? , NAGA SIVA KUMAR GUNDA, PRASHANT WAGHMARE, SUSHANTA MITRA, Department of Mechanical Engineering, University of Alberta — There is a huge surge in developing liquid repellant surfaces based on the micro/nanostructures that are inherently present in nature, like the one in case of fish scales. Through systematic contact angle measurement of oil drops on fish scales submerged in souring water medium, we have demonstrated that the superhydrophobic/superoleophobic nature of fish scales is attributed to a combination of the mucus layer and the hierarchical structures. The mucus layer on the fish scales produces an unprecedented contact angle close to 180o in contrast to the contact angle of 150o produced in the absence of the mucus layer. We have also identified, through FTIR analysis, that the distinct chemical signatures of mucus accountable for such large contact angles. Sunday, November 23, 2014 8:00AM - 9:57AM Session A14 Drops: General I — 3009/3011 - John Bush, Massachusetts Institute of Technology 8:00AM A14.00001 High Reynolds number droplet de-pinning on textured surfaces: theory and experiments , SUNGYON LEE, BENJAMIN WILCOX, FENG XU, EDWARD WHITE, Texas A&M University — The stability of drops on surfaces subject to forcing by wind and gravity is relevant to heat exchangers, fuel cells, and aircraft icing, and it lacks understanding in a high Reynolds number regime. To experimentally investigate this phenomenon, water drops are placed on the rough aluminum floor of a tiltable wind tunnel and brought to critical conditions for varying drop sizes, inclination angles, and flow speeds. In particular, the evolving 3D droplet shapes under flow are reconstructed based on a laser-speckle interface measurement tool, while the critical flow rates of droplet depinning are also noted. By accounting for the contact angle hysteresis and the pressure build-up in a nearly turbulent boundary layer, the critical depinning flow rate is theoretically predicted and is compared to the experimental results. We also observe and explain the transition of the drop depinning behavior from inertia-dominated to gravity-dominated regimes at non-zero inclination angles. 8:13AM A14.00002 Directional motion of liquid under mechanical vibrations , MAXIME COSTALONGA, PHILIPPE BRUNET, HASSAN PEERHOSSAINI, Université Paris Diderot — When a liquid is submitted to mechanical vibrations, steady flows or motion can be generated by non-linear effects. One example is the steady acoustic streaming one can observe when an acoustic wave propagates in a fluid. At the scale of a droplet, steady motion of the whole amount of liquid can arise from zero-mean periodic forcing. As It has been observed by Brunet et al. (PRL 2007), a drop can climb an inclined surface when submitted to vertical vibrations above a threshold in acceleration. Later, Noblin et al. (PRL 2009) showed the velocity and the direction of motion of a sessile drop submitted to both horizontal and vertical vibrations can be tuned by the phase shift between these two excitations. Here we present an experimental study of the mean motion of a sessile drop under slanted vibrations, focusing on the effects of drop properties, as well as the inclination angle of the axis of vibrations. It is shown that the volume and viscosity strongly affect the drop mean velocity, and can even change the direction of its motion. In the case of a low viscous drop, gravity can become significant and be modulated by the inclination of the axis of vibrations. Contact line dynamic during the drop oscillations is also investigated. 8:26AM A14.00003 A capillary Archimedes’ screw , BAPTISTE DARBOIS TEXIER, STEPHANE DORBOLO, GRASP, Institute of Physics, University of Liege — As used by Egyptians for irrigation and reported by Archimedes, a screw turning inside a hollow pipe can pull out a fluid againt gravity. At a centimetric scale, an analagous system can be found with a drop pending on a rotating spiral which is tilted toward the horizontal. The ascent of the drop to the top of the spiral is considered and a theoretical model based on geometrical considerations is proposed. The climb of the drop is limited by the fluid deposition on the screw at high capillary number and by a centrifugation phenomenon. We find out the range of fluid proprities and spiral characteristics for which an ascending motion of the drop is possible. Finally we discuss the efficiency of such system to extract a fluid from a bath at a centrimetric scale. 8:39AM A14.00004 A laboratory measurement of drop impact on a water surface in the presence of wind1 , XINAN LIU, REN LIU, University of Maryland, College Park — The impact of single water drops on a water surface was studied experimentally in a wind tunnel. Water drops were generated from a needle oriented vertically from the top of the wind tunnel test section. After leaving the needle, the drops move downward due to gravity and downstream due to the effect of the wind, eventually impinging obliquely on the surface of a pool of water on the bottom of the test section. The vertical velocities of drops were about 2.0 m/s and the wind speeds varied from 0 to 6.4 m/s. The drop impacts were recorded simultaneously from the side and above with two high-speed movie cameras with frame rates of 1,000 Hz. Our measurements show that both wind speed and initial drop size dramatically affect the drop impacts and subsequent generation of crowns, secondary drops, stalks and ring waves. In the presence of wind, an asymmetric crown forms after the drop hits the water surface and secondary drops are generated from the fragmentation of the leeward side of the crown rim. This is followed by a stalk formation and ring waves at the location of the water drop impact. It is found that the stalks tilt to leeward and the ring waves in the windward direction are stronger than that in those in the leeward. 1 This work is supported by National Science Foundation, Division of Ocean Sciences 8:52AM A14.00005 Simulation of flows with moving contact lines on curved substrates by immersed boundary methods1 , HANG DING, HAO-RAN LIU, University of Science & Technology of China — We propose an approach to simulate flows with moving contact lines (MCLs) on curved substrates. The approach combines an immersed boundary method with a three-component diffuse-interface model and a characteristic MCL model. The immersed boundary method circumvents the penetration of the gas and the liquid into the solid by convection while the three-component diffuse-interface model can prevent the diffusive fluxes of the gas and liquid from infiltrating into the solid substrate. The characteristic MCL model not only allows for the motion of contact lines, but makes the gas-liquid interface to intersect the solid object at an angle in consistence with the prescribed contact angle, even with tangent variation at the solid surface. We examine the performance of the approach through a variety of numerical experiments: mass conservation and interface shapes at equilibrium were tested through the simulation of drop spreading on a circular cylinder, while the dynamic behavior of MCLs on the curved boundaries was investigated by simulating water entry of and drop impact on a sphere, respectively. At last, we studied the penetration process of a drop into a cluster of circular cylinders. 1 This article was supported by 100 Talents Program of The Chinese Academy of Sciences and the National Natural Science Foundation of China (Grant No. 11172294). 9:05AM A14.00006 Drop motion due to oscillations of an inclined substrate , YI XIA, CHUN-TI CHANG, SUSAN DANIEL, PAUL STEEN, Cornell University — A sessile drop on a stationary inclined substrate remains pinned unless the angle of inclination is greater than some critical value. Alternatively, when shaken at even small angles of inclination, the drop undergoes shape deflections which may lead to drop translation. Translation occurs when large contact angle fluctuations, favored by oscillations at resonance, overcome contact angle hysteresis. In this study, resonance is triggered by substrate-normal oscillations. The drop translation is typically observed to be of constant speed for a given set of parameters. The speed is measured experimentally as a function of resonance mode, driving amplitude and drop volume. This technique of activating the motion of drops having a particular volume can be utilized for applications of droplet selection and transport. 9:18AM A14.00007 Why a falling drop does not in general behave like a rising bubble , RAMA GOVINDARAJAN, TIFR Centre for Interdisciplinary Sciences, Tata Institute of Fundamental Research, Hyderabad, MANOJ TRIPATHI, KIRTI SAHU, Indian Institute of Technology Hyderabad, India — Is a settling drop equivalent to a rising bubble? The answer is known to be in general a no, but we show that when the density of the drop is less than 1.2 times that of the surrounding fluid, an equivalent bubble can be designed for small inertia and large surface tension. Hadamard’s exact solution is shown to be better for this than making the Boussinesq approximation. Scaling relationships and numerical simulations show a bubble-drop equivalence for moderate inertia and surface tension, so long as the density ratio of the drop to its surroundings is close to unity. When this ratio is far from unity, the drop and the bubble are very different. We show that this is due to the tendency for vorticity to be concentrated in the lighter fluid, i.e. within the bubble but outside the drop. As the Galilei and Bond numbers are increased, a bubble displays underdamped shape oscillations, whereas beyond critical values of these numbers, over-damped behaviour resulting in break-up takes place. The different circulation patterns result in thin and cup-like drops but bubbles thick at their base. These shapes are then prone to break-up at the sides and centre, respectively. 9:31AM A14.00008 Self-crumpling elastomers: bending motion induced by a drying stimulus1 , FRANÇOIS BOULOGNE, HOWARD A. STONE, Princeton University, Complex Fluid Group — Capillary forces exerted by a liquid drop can bend elastic slender structures such as fibers or sheets. However, to successfully achieve capillary origami with sheets, it is important to make sure that the adhesion of the elastomer with the surface is low. We report an experimental study of the drying-induced peeling of a bilayer consisting of an elastomeric disk coated with a suspension of nanoparticles. We show that where capillary forces associated with the scale of the droplet can not compete with the adhesion of the elastomer on a surface, nevertheless large tensile stress developed in the coating, which resulted in a moment bending the bilayer. We attribute this stress to the nano-menisci in the pores of the colloidal material and we propose a model that describes successfully the early stage curvature of the bilayer. Thus, we show that the peeling can be conveniently controlled by the particle size and the coating thickness. We believe that such systems can be employed in various situations where delicate surfaces are involved such as in applications with optical and electronic components or in restoration of photographies, painting, wallpaper, fragile collectibles from contamination by dust, pollen, dirt, etc. 1 The research leading to these results received funding from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7/2007-2013) under REA grant agreement 623541. 9:44AM A14.00009 A low-cost, precise piezoelectric droplet generator1 , TANYA LIU, DANIEL M. HARRIS, JOHN W.M. BUSH, Massachusetts Institute of Technology — We present the design for a piezoelectric drop-on-demand generator capable of producing highly repeatable, millimetric droplets. The generator is low-cost, simple to fabricate, and easily reproduced. We demonstrate that droplet diameter can be controlled through variation of the piezoelectric driving waveform parameters. Our experiments demonstrate that if waveform amplitude is fixed, droplet diameter is directly dependent on waveform pulse width, allowing for a range of droplet sizes to be produced for a fixed nozzle diameter. Successful droplet generation occurs only within a finite range of pulse widths; however, outside of this range satellite droplets form or no droplet generation occurs. We also discuss the dependence of droplet size on other system parameters including pressure at the nozzle plane and nozzle diameter. These results make the generator design potentially applicable to a wide range of fluids experiments where repeatable millimetric droplets are required. 1 The authors gratefully acknowledge the financial support of the NSF through Grant CMMI-1333242 and the MIT Undergraduate Research Opportunities Program. Sunday, November 23, 2014 8:00AM - 9:57AM Session A15 Drops: Complex Fluids — 3022/3024 - Gwynn Elfring, University of British Columbia 8:00AM A15.00001 Drop impact experiment as a model experiment to investigate the role of oil-in-water emulsions in controlling the drop size distribution of an agricultural spray , CLARA VERNAY, LAURENCE RAMOS, CHRISTIAN LIGOURE, Laboratoire Charles Coulomb, UMR5221, Université Montpellier 2 et CNRS, France, JEAN-PAUL DOUZALS, Irstea, UMR ITAP, Montpellier, France, RAJESH GOYAL, R&D Novecare, Solvay, Bristol, USA, JEAN-CHRISTOPHE CASTAING, R&D Novecare, Solvay, Aubervilliers, France — Agricultural spraying involves atomizing a liquid stream through a hydraulic nozzle forming a liquid sheet, which is then destabilized into droplets. Solution adjuvants as dilute oil-in-water emulsions are known to influence the spray drop size distribution. To elucidate the mechanisms causing the changes on the drop size distribution, we investigate the influence of emulsions on the destabilization mechanisms of liquid sheets. Model laboratory experiments based on the collision of a liquid drop on a small target are used to produce and visualize liquid sheets. With emulsion, the sheet is destabilized by the nucleation of holes in the sheet that perforate it during its expansion. The physico-chemical parameters of the emulsion, such as the concentration and the emulsion drop size distribution, are varied to rationalize their influence on the destabilization mechanisms. The results obtained with the drop impact experiments are compared to the measurement of the spray drop size distribution. The very good correlation between the number of nucleation events and the volume fraction of small drops in the spray suggests that experiments on liquid sheet are appropriate model experiments to gain an understanding of the physical mechanisms governing the spray drop size distribution. 8:13AM A15.00002 Relaxation-induced coalescence with and without insoluble surfactants , CAROLINA VANNOZZI, Univ of California - Santa Barbara — Rallison and Acrivos in 1978 investigated for the first time numerically the deformation and burst of a viscous drop in extensional flow. This seminal paper led to other important studies regarding the relaxation of viscous drops previously deformed in extensional flows, both for systems with and without surfactants. In this line of research we present the boundary integral simulations of the relaxation process of two viscous drops, previously undergoing a flow-induced head-on collision in an extensional flow, both with and without insoluble surfactants. The clean interface case showed that the acquired drop deformation induces a flow in the thin film between the drops as the interface relaxes back to restore the drop original shape [1]. Under certain circumstances, this flow thins the film, allowing drop coalescence. Surprisingly, this phenomenon, the so-called relaxation-induced coalescence, is possible even for collisions which would not lead to coalescence while the flow is active, and it thus influences the final drop size distribution of blends/emulsions. In the presence of trace amounts of insoluble surfactants relaxation-induced coalescence is still possible, but less likely, depending strongly on the surfactant diffusivity. [1] Vannozzi, J Fluid Mech 2012. 8:26AM A15.00003 Droplet impact on permeable meshes with yield stress fluids , BRENDAN BLACKWELL, RANDY EWOLDT, Univ of Illinois - Urbana — Yield stress fluids can stick and accumulate where they impact. To understand coating of complex topography, we experimentally study the ability of droplets to accumulate on permeable solid meshes (rigid surfaces with small, evenly spaced openings). When inertial stresses are sufficiently high compared to the yield stress, a drop can pass through a mesh, breaking into smaller fluid particles with varying shapes, sizes, and velocities in the process. In contrast, when inertial stresses are sufficiently low compared to the yield stress, a droplet can stick to the mesh as though it were a solid surface. Drop size, impact velocity, mesh geometry, and rheological material properties are varied. Layers of multiple meshes are also examined, demonstrating a range of behaviors and the ability to coat internal aspects of complex topography. Dimensional analysis is performed to characterize material transmittance as a function of the input parameters. 8:39AM A15.00004 Synchronous droplets as a test bed for pulsatory active fluids , GEORGIOS KATSIKIS, MANU PRAKASH, Stanford University — Collective behavior in many-body systems has been studied extensively focusing on a wide range of interacting entities including: flocking animals, sedimenting particles and microfluidic droplets among others. Here, we propose an experimental platform to explore an oscillatory active fluid with synchronous ferrofluid droplets immersed in an immiscible carrier fluid in a Hele-Shaw configuration. The droplets are organized and actuated on a 2-D uniform grid through application of a precessive magnetic field. The state of our system is dependent on three parameters: the grid occupancy with fluid droplets, the grid geometry and the magnetic field. We study the long range orientational order of our system over a range of those parameters by tracking the motion of the droplets and analyzing the PIV data of the carrier fluid flow. Numerical simulations are juxtaposed with experimental data for prediction of the system’s behavior. 8:52AM A15.00005 Complex traffic of drops in 2D microchannels: self-organization, periodicity and reversibility , DANNY RAJ MASILA, RAGHUNATHAN RENGASWAMY, Indian Institute of Technology Madras — In a 2D microchannel, drops slowdown and accelerate in the diverging and converging sections of the channel respectively. Drops entering the microchannel, approach each other when they slow-down in the diverging section. They start to interact hydrodynamically to form different layered structures depending on the spacing between the drops prior to entry into the channel. These patterns break in the converging end of the channel before the drops exit. We devised a multi-agent approach that was able to capture the dynamic pattern formation of drops inside the microchannel. The self-organized dynamic patterns formed are a function of the inlet spacing of the drops. These patterns due to complex drop traffic result in a non-linear outlet spacing between exiting drops. We present a study where we investigate how every spatial event inside the microchannel can result in a temporal signature in the outlet spacing of drops. Understanding the dynamic pattern formation also sheds light on the response of the microfluidic device to flow reversal. We observe that when the layering and breaking patterns of drops are similar the system is reversible. 9:05AM A15.00006 Microscopic reversibility and memory in soft crystals undergoing large deformations , LIAT ROSENFELD, Stanford University, CLAUDIU STAN, SLAC National Accelerator Laboratory, SINDY K.Y. TANG, Stanford University — In this study, we explore the transition from reversible to chaotic behavior in an oscillatory shear flow of water-in-oil emulsions. The emulsion was injected through a microchannel and was forced to rearrange due to a central constriction in the channel. We study the motion of the individual droplets and their neighbors in order to determine their ability to retain their original position after several cycles of oscillations. We have found that the emulsion exhibit behaviors that vary from complete reversibility to complete irreversibility depending on the volume fraction, velocity and strain rate. The reversibility, both in the trajectory and the deformation of every drop, is reproducible even when the drops undergo many rearrangement events over distances of >150 droplet diameters. Moreover, the deformability of the drops and the high volume fraction are crucial conditions for the onset of reversibility. We provide here the first direct visualization and physical analysis of this phenomenon. This work is an important step in describing the flow of concentrated emulsions and suspensions in microchannels and is therefore crucial for understanding the behavior of droplets, bubbles and particles in droplet microfluidic applications. 9:18AM A15.00007 Morphological dynamics of a falling drop in a magnetic field , DAVID S. MARTÍNEZ, PhD. Student, Technical University of Cartagena, MIGUEL ANGEL CABRERIZO-VILCHEZ, University of Granada, ANTONIO VIEDMA, Professor, Technical University of Cartagena, ALIDAD AMIRFAZLI, Professor, Department of Mechanical Engineering — A ferrofluid drop was released and fell through the air; as it travelled through a thin coil a magnetic pulse (4-65 mT for 4 ms) was given. The drop either deformed or split into a multitude of smaller particles. For pulse amplitudes less than 9 mT the drop sequentially deformed to oblate and prolate ellipsoids; the dynamics of drop deformation in this case was modeled using a harmonic damped oscillatory function. Higher magnetic field pulse amplitudes resulted in drop taking the form of a cylindrical ligament; depending on the field strength various numbers of drops were ejected from ends of the ligament. The size of ejected drops decreased with increasing magnetic field strength. Ejected drops travelled with higher velocities as magnetic field strength was increased in a linear fashion. At field strengths of 65mT up to six drops were ejected from the ligament. Ejected drops where all spherical, and the ejection process was over in 22 ms. The cylindrical ligament eventually recovered to a spherical shape due to surface tension forces in 210 ms. Other interesting observations such as momentary interactions without coalescence between consecutive drops due to the wake effect will be discussed. 9:31AM A15.00008 Fractal pattern formation in metallic ink sessile droplets , MILOUD HADJ-ACHOUR, DAVID BRUTIN, Aix-Marseille University - IUSTI UMR 7343 — We report a fingering instability that occurs during the spreading and evaporation of a nanosuspension droplet. The patterns has a fractal structure similar to those reported by N. Shahidzadeh-Bonn and al. (2008) for salt crystallisation, during evaporation of saturated Na2SO4 on a hydrophilic surface. The fingering instability has been widely studied for both Newtonian and non-Newtonian fluids. However, we describe for the first time that a fingering instability is observed for the spreading of a nanosuspension sessile droplet. We demonstrate that in certain cases, the contact line evolves through different spreading regimes according to J. De Coninck et al. (2001) with an enhancement in the evaporation rate due the formation of the fractal patterns. 9:44AM A15.00009 Relative humidity influence on the spreading dynamics of sessile drops of blood , DAVID BRUTIN, WASSIM BOUZEID, Aix-Marseille University - IUSTI UMR 7343 — We studied the effect of relative humidity on the initial stages of spreading dynamics for drops of whole human blood. A range of relative humidities from 8% to 90% was studied. Drops of the same volume were gently deposited on ultra-clean microscope glass substrates. We show that the drop spreading is driven by two distinct regimes. The first is characterized by fast dynamics and competition between viscous forces and capillary forces, whereas the second regime is characterized by competition between viscous dissipation and evaporation and exhibits slower dynamics. At early stages of spreading, the power law r(t) ∼ tn (n = 0.65) was observed regardless of the humidity. At later stages of spreading, the exponent of the power law r(t) ∼ tn (n = 0.19) was found to be higher than that of Tanner’s law because of the effect of humidity and Marangoni stresses. Spreading time and spreading dynamics were found to be related to relative humidity. This is explained by the adhesion of red blood cells to the substrate that is similar to the mechanism observed for nanofluid droplets. The mean velocity of the triple line followed the same behavior as Tanner’s model, where the final wetting radius and the apparent contact angle are functions of relative humidity. Sunday, November 23, 2014 8:00AM - 9:57AM Session A16 Free-Surface Flows I: Impacts — 2000 - Jesse Belden, Naval Underwater System Center 8:00AM A16.00001 Oblique impact of water-skipping elastic spheres , JESSE BELDEN, Naval Undersea Warfare Center, TADD TRUSCOTT, RANDY HURD, Brigham Young University, MICHAEL JANDRON, Naval Undersea Warfare Center, ALLAN BOWER, Brown University — Highly compliant elastic spheres possess remarkable water skipping capabilities. High-speed video reveals that, upon impact with the water, the balls create a cavity and deform significantly. The flattened spheres resemble skipping stones and this augmented geometry results in enhanced lift that causes the ball to launch back into the air. This deformation also excites elastic vibration modes within the sphere. A numerical model reveals that the vibrations are initiated by a stress concentration developed in the early moments of impact. In one mode, an elastic wave propagates around the sphere periphery and may impact the water surface, resulting in an energy loss from the sphere. Thus two timescales govern the success of skipping: the total collision time of impact must be less than the deformation time associated with material vibration. Using a simplified analytical model, we derive the expected scaling of each time in terms of a dimensionless ratio of material shear modulus to fluid inertia forces, G/ρU 2 . Experiments over a range of parameters validate this scaling and result in a regime diagram that distinguishes different types of skipping. We identify critical relations for the material properties and impact conditions to achieve skipping. 8:13AM A16.00002 Performance enhancing water skipping: successive free surface impacts of elastic spheres , RANDY HURD, TADD TRUSCOTT, Brigham Young University, JESSE BELDEN, Naval Undersea Warfare Center — From naval gunners skipping cannonballs to children skipping stones, physicists have long been enamored with the repeated ricochet of objects on the water surface. Elastic spheres, such as the toy Waboba ball, make water skipping more accessible to the masses by expanding the range of impact parameters over which objects can be skipped. For example, it is not difficult to achieve more than twenty skips with such spheres, where skipping a stone twenty times is very difficult. In this talk we discuss the dynamics of water skipping elastic spheres over several successive skips. High-speed video captured using a unique experimental setup reveals how dynamics change with each skip as a result of lost kinetic energy. We place these observations in the context of previous work on single oblique impacts to identify material vibration modes that are excited during ricochet. The material modes excited with each successive impact are seen to decay from high-energy modes to low energy modes until water entry finally occurs. A model for estimating skipping outcome from initial conditions is proposed. 8:26AM A16.00003 Impact with dynamic surface tension , LAURENT DUCHEMIN, NICOLAS VANDENBERGHE, Aix Marseille Universite, CNRS, Centrale Marseille, IRPHE, Marseille, France — We study impacts of a rigid body on a thin elastic sheet floating on a liquid. When struck by a solid object of small size, the elastic sheet deforms and waves propagate in and on the membrane. The impact triggers a longitudinal elastic wave effectively stretching the membrane. The hydro-elastic transverse wave that propagates in the stretched domain is similar to capillary waves on a free surface with an equivalent “surface tension” that results from the stretching of the elastic membrane. Two limiting cases, for which a self-similar solution can be computed, corresponding to short and long times are identified. Surprisingly, our study reveals that the fluid-body system behaves as a regular liquid-gas interface, but with an effective surface tension coefficient that scales linearly with the impact velocity. 8:39AM A16.00004 Resolution of the singularities in the water impact problem: compressibility and viscosity effects , ROUSLAN KRECHETNIKOV, University of Alberta — This work presents an analysis of the flow structure resulting from the flat plate impact on the surface of a compressible viscous liquid at zero deadrise angle. The key goals are to elucidate the effects of compressibility and viscosity and to resolve both near the plate edge, r → 0, and the early times, t → 0, limit singularities in the classical incompressible inviscid pressure-impulse theory. The constructed solution is contrasted to its incompressible flow counterpart, which allows one to identify the characteristic time and spatial scales of each distinct stage of the flow evolution, as defined by different governing physical mechanisms. 8:52AM A16.00005 Water Entry of Deformable Spheres , TATE FANNING, RANDY HURD, Brigham Young University, JESSE BELDEN, Naval Undersea Warfare Center, TADD TRUSCOTT, Brigham Young University — We examine the water entry characteristics and cavity dynamics of highly deformable elastic spheres at high Reynolds numbers (105 ) using high-speed photography and image processing techniques. Upon impact normal to a free surface, these elastic spheres undergo significant deformation. We have observed principal stretches on the order of 1.6 diameters for the most compliant spheres. This initial deformation sets up an oscillatory vibration mode in the sphere that persists throughout its descent through the water column. These oscillations disturb normal cavity formation, resulting in the formation of a periodic, nodular cavity. A comprehensive experimental study allows for prediction of cavity shape, pinch off depth, and time to pinch off. Decreasing sphere stiffness results in decreased pinch off depths, and increased time to pinch off over the range of Reynolds numbers tested. 9:05AM A16.00006 Flipping over: inversion characteristics of a buoyant cylindrical puck during oblique water impact , ZACHARY SMITH, TADD TRUSCOTT, Brigham Young University — The Apollo Command Module had a tendency to flip over upon impact with the ocean surface after returning from space (9/19 times). In an effort to improve upon this idea for potential missions to Saturn’s moon Titan, we present experimental results of a simplified buoyant cylindrical puck impacting the water surface. We examine the dependence of inversion upon vertical and horizontal velocity, center of gravity, and the pitch angle of the puck relative to the free surface. An analytical model is developed which characterizes inversion. High-speed images reveal that the puck does not completely submerge upon impact. Instead, the top of the puck remains above the water surface via a contact line attachment to the cavity. The asymmetric cavity then collapses, applying a moment, which can be sufficient to invert the puck after impact. 9:18AM A16.00007 Investigation of the Entrainment Phenomenon Using a Scaling Approach1 , ARAVIND KISHORE, URMILA GHIA, Univ of Cincinnati — Air entrainment is a commonly observed phenomenon; we see it when filling a glass with water from a faucet, in the frothing of the ocean surface, in white water rapids, etc. The focus of our work is the numerical simulation of the entrainment phenomenon associated with laminar plunging jets. With increasing jet velocity, the interfacial cusp formed between the jet and the liquid pool becomes sharper. At a critical jet velocity, entrainment inception occurs, i.e., the interfacial cusp breaks, the interface ruptures, and air is pulled into the liquid pool. We conduct two-fluid simulations using the Volume-Of-Fluid (VOF) methodology. The large range of length scales in the flow presents a major computational challenge. We postulate an approach based on scaling of the underlying physics and this helps alleviate the constraints that the physics poses on the numerical method. The approach is validated using a simple flow configuration - a cylinder rotating at an interface between two fluids. Our simulations capture the sharpening of the interfacial cusp, and the sudden rupture of the interface. The predicted critical entrainment velocities are within 1% of experimental data, thereby providing confidence in the approach. 1 This work was supported by the UC Simulation Center at the University of Cincinnati 9:31AM A16.00008 The effects of surface tension on the initial development of a free surface adjacent to an accelerated plate , JAMAL UDDIN, DAVID NEEDHAM, University of Birmingham — When a vertical rigid plate is uniformly accelerated from rest into an initially stationary layer of inviscid incompressible fluid, the free surface will undergo a deformation in the locality of the intersection point between the free surface and the plate. This deformation of the free surface will, in the early stages, cause a jet to rise up the plate. An understanding of the local structure of the free surface in the early stages of motion is vital in many situations and has been developed in detail by King & Needham (1994). In this work we consider the effects of introducing weak surface tension, characterized by the inverse Weber number, W, into the problem considered by King & Needham (1994). Our approach is based upon matched asymptotic expansions as W to 0. It is found that four asymptotic regions are needed to describe the problem. The three largest regions have analytical solutions whilst a numerical method based on finite differences is used to solve the time dependent harmonic boundary value problem in the last region. We also present some preliminary comparisons between experiments and theory. 9:44AM A16.00009 On the cusps bordering liquid sheets , JOSE MANUEL GORDILLO, Universidad de Sevilla, HENRI LHUISSIER, Université Paris Diderot, EMMANUEL VILLERMAUX, IRPHE, Aix-Marseille Université, France — The rim at the edge of a steady radially expanding liquid sheet, or bordering a hole expanding in a liquid film, is naturally indented. It presents a collection of cusps at the tip of which the liquid concentrates and is ejected. An experimental description of these cusps for a stationary flat inviscid Savart sheet, formed by the normal impact of a jet with diameter d and velocity u against a solid disk, is given. We identify the stable node-jet structure responsible for the deflection of the incoming flow at the rim and demonstrate how the cusps are the structures that accommodate for both mass and momentum conservation at the sheet edge. Their shape, their number around the sheet, and the residual momentum carried by the ejected liquid are computed. Our model reproduces the experimental observations, correcting the classical interpretation of Savart’s experiments first given by Taylor, who proposed that the radius of the sheet is given by R = d W e/16, with a Weber W e number based on d and u. Indeed, Taylor’s picture disregards that the sheet is not circular, that the liquid is ejected with a non-vanishing remnant radial momentum at the sheet edge and, hence, that the actual radius of the sheet is smaller than R. Sunday, November 23, 2014 8:00AM - 9:57AM Session A17 Nonlinear Dynamics I: Coherent Structures I — 2002 - Steven Brunton, University of Washington 8:00AM A17.00001 A comprehensive investigation of exact coherent states in Newtonian channel flow1 , JAE SUNG PARK, University of Wisconsin-Madison, RASHAD MOARREF, BEVERLEY J. MCKEON, California Institute of Technology, MICHAEL D. GRAHAM, University of Wisconsin-Madison — Exact coherent states have been intensively investigated for better understanding of the transition to turbulence and the complex dynamics in shear flows. Here, we present five families of nonlinear traveling wave solutions in Newtonian channel flow. A Prandtl-von Kármán plot is used to characterize the solutions, in comparison to previously discovered solutions in the same geometry. One solution family shows very intriguing behavior in terms of mean profiles: its upper and lower branches appear to approach the classical Newtonian and viscoelastic turbulent profiles, respectively. On the lower branch of this solution, a spatially subharmonic bifurcation arises, giving rise to period doubling. The solutions are then considered in state space to identify connections to turbulent flow trajectories and paths of an intermittent bursting phenomenon. Lastly, a low-order representation of our exact coherent states is obtained using the resolvent mode decomposition of McKeon & Sharma (JFM, 2010). While lower branch solutions can be approximated by a few resolvent modes, typically one dominant mode, upper branch solutions need a larger number of modes. The dominant features of leading resolvent modes and the dependence of Reynolds number on those modes are further discussed. 1 This work was supported by the Air Force Office of Scientific Research through grants FA9550-11-1-0094 and FA9550-12-1-0469. 8:13AM A17.00002 An infinite hierarchy of guage particle solutions to a regularized Euler equation: numerical methods and beyond1 , HENRY JACOBS, COLIN COTTER, DARRYL HOLM, Imperial College, DAVID MEIER, Brunel University — In this talk we present an infinite hierarchy of exact solutions to a regularized form of Euler’s fluid equations. Each of these solutions is isomorphic to the motion of finitely many guage-theoretic particles, wherein each particle stores internal Lie group structures which correspond to higher-order deformation gradients of the Lagrangian flow map. Collision experiments suggest that two particles at one level in the hierarchy can asymptotically merge into a single particle at a higher-level in the hierarchy. We will display some of these collisions and provide a formal argument to explain this phenomena. These collision events are interpreted as a cascade to smaller scales. 1 European Research Council Advanced Grant 267382 FCCA 8:26AM A17.00003 Exact coherent states in a reduced model of parallel shear flows , CEDRIC BEAUME, Imperial College London, EDGAR KNOBLOCH, UC Berkeley, GREG CHINI, University of New Hampshire, KEITH JULIEN, University of Colorado at Boulder — In plane Couette flow, the lower branch Nagata solution follows simple streamwise dynamics at large Reynolds numbers. A decomposition of this solution into Fourier modes in this direction yields modes whose amplitudes scale with inverse powers of the Reynolds number, with exponents that increase with increasing mode number (Wang et al., Phys. Rev. Lett. 98, 204501 (2007)). In this work, we use this scaling to derive a reduced model for exact coherent structures in general parallel shear flows. The reduced model describes the dynamics of the streamwise-averaged flow and of the fundamental fluctuations and is regularized by retaining higher order viscous terms for the fluctuations. Numerical methods are designed to find good approximates of nontrivial solutions which are then converged using a preconditioned Newton method. This procedure captures both lower branch and upper branch solutions and demonstrates that these branches are connected via a saddle-node bifurcation. 8:39AM A17.00004 Homotopy between plane Couette flow and Pipe flow , MASATO NAGATA, Tianjin University, KENGO DEGUCHI, Imperial College, London — In order to investigate symmetry connections between two canonical shear flows, i.e. plane Couette (PCF) and pipe flow (PF), which are linearly stable for all Reynolds numbers and therefore undergo subcritical transition, we take annular Poiseuille-Couette flow (APCF) as an intermediary Although PCF and PF are very different geometrically, APCF recovers PCF by taking the narrow gap limit, and also PF by taking the limit of vanishing inner cylinder where a homotopy of the basis functions from no-slip to regular conditions at the centre is considered. We show that the double-layered mirror-symmetric solutions in sliding Couette flow (APCF without axial pressure gradient) found by Deguchi & Nagata (2011) can be traced back to the mirror-symmetric solutions in PCF. Also we show that only the double-layered solution successfully reaches the PF limit, reproducing the mirror-symmetric solution in PF classified as M1 by Pringle & Kerswell (2007). 8:52AM A17.00005 Evolution of Power and Structure in an Electroconvective Transition , MARCUS DAUM, ZRINKA GREGURIĆ FERENČEK, JOHN CRESSMAN, School of Physics, Astronomy, and Computational Sciences, George Mason University — Electroconvecting liquid crystals support a wide range of states which are characterized by the system’s ability to create, support, and annihilate structure. Through the creation of electroconvective rolls and defects, the liquid crystal sample is able to absorb and dissipate more energy. By simultaneously acquiring optical and electrical data we are able to accurately compare structure of the sample and injected electrical power. Here we report on spatiotemporal interactions as we abruptly force the sample from an initial state to defect turbulence. By observing the power and the structure of the sample, we have identified qualitatively different transient behaviors based on initial structure within the sample. Using these characterizations, we are able to draw parallels between structural dynamics and large fluctuations in power. 9:05AM A17.00006 First experimental measurement of the Melnikov function , PATRICE MEUNIER, PETER HUCK, EMMANUEL VILLERMAUX, IRPHE, Aix-Marseille Univ., CNRS, Ecole Centrale Marseille — The problem of scalar mixing in a 2D flow has been extensively studied numerically by following Lagrangian tracers or theoretically using the tools of dynamical systems (KAM tori, quasi-periodic orbits, chaotic attractors...). However, in all these modelisations, the diffusion of the scalar is usually neglected for the purposes of the numerical/theoretical tools. We present here an experiment with an exactly 2D flow, which allows to study properly the diffusive and mixing problem at very large Peclet number. To avoid any 3D flow, the fluid is stratified with a linear density gradient using salted water. Moreover, the viscosity of the water is decreased of an order of magnitude by adding 10% ucon oil in the water. The flow under study is created by the co-rotation of two vertical cylinders, leading to a homoclinc point at the center. This base flow is perturbed periodically by a third oscillating cylinder. The dye injected at the center settles on the stable manifold of the homoclinic point. The distance between the stable and the unstable manifold is measured as half the distance between the maximum and the minimum of the dye’s undulation. The results are in good quantitative agreement with the theoretical prediction of the Melnikov function for this flow. 9:18AM A17.00007 Linear analysis of the mean flow of thermosolutal travelling waves , SAM TURTON, Cambridge University, United Kingdom, LAURETTE TUCKERMAN, PMMH-CNRS-ESPCI, France — We carry out a stability analysis on the mean flow extracted from 2D travelling waves in thermosolutal convection over a range of values for separation parameter S, Lewis number Le and Prandtl number Pr. Consistent with similar analyses performed on the mean flow of the cylinder wake, we find that the eigenfrequency provides an accurate measure of the frequency of the travelling waves, in contrast to the frequency obtained by linearizing about the unstable conductive state. The linear growth rates are close to zero just beyond the Hopf bifurcation, and in the case of large Pr, remain so for larger values of the thermal Rayleigh number, implying that the travelling wave mean flow is marginally stable in these regimes. 9:31AM A17.00008 Reynolds Number Effects on Mixing Due to Topological Chaos , SPENCER SMITH, SANGEETA WARRIER1 , Mount Holyoke College — Topological chaos has emerged as a powerful modeling tool to investigate fluid mixing. While this theory can guarantee a lower bound on the stretching rate of certain material lines, it does not indicate what fraction of the fluid actually participates in this minimally mandated mixing. Indeed, the area in which effective mixing takes place depends on physical parameters such as the Reynolds number. To help clarify this dependency, we numerically simulate the effects of a batch stirring device on a 2D incompressible Newtonian fluid in the laminar regime. In particular, we calculate the finite time Lyapunov exponent (FTLE) field for two different stirring protocols, one topologically complex (pseudo Anosov) and one simple (finite order), over a range of viscosities. After extracting appropriate measures indicative of mixing from the FTLE field, we see a clearly defined range of Reynolds numbers for which the relative efficacy of the pseudo Anosov protocol over the finite order protocol justifies the application of topological chaos. The Reynolds number dependance of these mixing measures also reveals several other intriguing phenomena. 1 Undergraduate Student 9:44AM A17.00009 A Jacobian-free Newton-Krylov solver for determination of scaling laws in coherent Rayleigh-Bénard convection1 , DAVID SONDAK, LESLIE SMITH, FABIAN WALEFFE, ANAKEWIT BOONKASAME, University of Wisconsin, Madison — Computational studies of coherent Rayleigh-Bénard convection in a two-dimensional channel with no-slip top and bottom walls are performed in order to determine scaling laws for a range of Rayleigh (Ra) and Prandtl (P r) numbers. Since these coherent states are unstable, a Jacobian-free Newton-GMRES algorithm is developed. This approach allows us to determine scaling of the Nusselt number (N u) with Ra by tracking unstable solutions to the Boussinesq equations. Scaling laws are presented for the primary solution that bifurcates from the conducting state at Ra ∼ 1708, becomes unstable in a Hopf bifurcation at Ra ∼ 5.4 × 104 but have been computed up to Ra ∼ 5 × 106 . We also determine scaling laws for the optimal heat transport up to Ra ∼ 108 . Mechanisms for the observed behavior are discussed including the relationship between the optimal solution and the primary solution as well as the effect of P r. We explore properties of the algorithm and review its potential as a tool in determining scaling laws for thermal convection as well as some areas for improvement. Extensions of this work to three-dimensional Rayleigh-Bénard convection will be discussed. 1 Partial support from NSF-DMS grant 1147523 is gratefully acknowledged. Sunday, November 23, 2014 8:00AM - 9:57AM — Session A18 Vortex Dynamics: Energy Harvesting and Atmospheric Flows 2004 - Dennice Gayme, Johns Hopkins University 8:00AM A18.00001 Optimization of energy harvesting efficiency of an oscillating hydrofoil: Sinusoidal and Non-sinusoidal trajectories , MICHAEL MILLER, BEN STROM, KENNETH BREUER, SHREYAS MANDRE, Brown University — We determine the feasibility of applying optimization algorithms to an oscillating hydrofoil’s motion trajectory to determine maximum efficiency of energy capture. Optimization is performed using the Nelder-Meade downhill simplex method. The objective function is the energy captured measured experimentally in run-time with an oscillating hydrofoil capable of measuring mechanical energy capture in a laboratory flume. For sinusoidal trajectories, optimization is performed over pitch and heave amplitudes as well as frequency; this system is shown to be capable of optimization in run-time. The optimum efficiency of 30% is found for a pitch amplitude of 70◦ , a heave amplitude of 0.8*chord and a dimensionless frequency of 0.13. To treat non-sinusoidal trajectories, we expand them in a truncated Fourier series and consider the coefficients of this series as variables for optimization. The sinusoidal case is simply an extreme case of such a truncated Fourier series, with only one term in the series retained. We present a systematic method for optimization over general non-sinusoidal trajectories by including more and more terms in the Fourier series. 8:13AM A18.00002 Vortex shedding from vertical axis wind turbine blades under linear motion1 , REEVE DUNNE, BEVERLEY MCKEON, California Institute of Technology — A NACA 0018 airfoil was pitched and surged sinusoidally in in a mean free stream flow at Rec = 100, 000 to simulate the flow over vertical axis wind turbine (VAWT) blades. Angle of attack variations between Umax −Umin α = ±30◦ and velocity variation of = .80 at a reduced frequency k = Ωc = .12 result in strong dynamic stall on the blade. Multiple flow Umean 2U ∞ regimes occur during the airfoil motion resulting in vortex shedding over a large range of frequencies. A model of the phase averaged (based on airfoil angle of attack and velocity) flow developed using dynamic mode decomposition highlights the evolution of the leading edge or dynamic stall vortex at the airfoil frequency. Instantaneous results show vortex shedding at frequencies up to 100 times higher than the frequency of the pitch/surge motion and smeared out by the phase averaging process. The implications for forcing on the blade (and associated wind turbine) are described. 1 This research is funded by the Gordon and Betty Moore Foundation through Grant GBMF #2645 to the California Institute of Technology 8:26AM A18.00003 Vortex Identification in the Wake of a Wind Turbine Array , ALEKSANDR ASEYEV, RAUL CAL, Portland State University — A 4 x 3 wind turbine array boundary layer is analyzed through Particle Image Velocimetry data gathered directly forward and aft of the first and last row turbines at the centerline in a wind tunnel. Vortex identification techniques are able to capture vortical structures. Q-criterion, $Delta$-criterion, and $lambda-2$ criterion are evaluated and compared for this flow. Q-criterion and $lambda-2$ criterion provided a clear indication of regions where vortical activity exists while the $Delta$-criterion is not able to capture these regions. Galilean decomposition, Reynolds decomposition, vorticity, and swirling strength were used to further understand the location and behavior of the vortices. The various criterion displayed the high magnitude vortices, resulting from the blade tips and located immediately in areas of high shear. Using Galilean and Reynolds decomposition, swirling motions are shown hugging vortex regions in agreement with the identification criterion. The percentages used in the Galilean decomposition were 20 and 50 percent of a convective velocity of 7 m/s. As the vortices convect downstream, these vortices weaken in magnitude to approximately 25 percent of those present in the near wake. 8:39AM A18.00004 Constructive interference in arrays of energy harvesters in fluid flows1 , VAHID AZADEH RANJBAR, OLEG GOUSHCHA, NIELL ELVIN, YIANNIS ANDREOPOULOS, CUNY-CCNY — In the present work we demonstrate some unique opportunities which exist to increase the power harvested with fluidic piezoelectric generators by almost two orders of magnitude higher than existing methods by exploiting dynamic non-linearities and deploying multi-element arrays in carefully selected positions in a fluid flow field. These ac-coupled generators convert fluid kinetic energy, which otherwise would be wasted, into electrical energy. The available power in a flowing fluid is proportional to the cube of its velocity and if it is properly harvested can be used for continuously powering very small electronic devices or can be rectified and stored for intermittent use. Additional experimental work has shown that non-linear arrays of such energy harvesters can produce high output voltages in a very broadband range of frequencies. In our work, we investigate the effect of geometric parameters such as spatial arrangement and the mutual interference between the elements of a non-linear array on their overall performance and efficiency characteristics. Analytical tools based on the non-linear van der Pol oscillator have been also developed and verified with experimental data. 1 Work supported by National Science Foundation under Grant No. CBET #1033117. 8:52AM A18.00005 Energy Harvesting of a Flapping Airfoil in a Vortical Wake , Z. CHARLIE ZHENG, University of Kansas, ZHENGLUN WEI, Georgia Institute of Technology — We study the response of a two-dimensional flapping airfoil in the wake downstream of an oscillating D-shape cylinder. The airfoil has either heaving or pitching motions. The leading edge vortex (LEV) and trailing edge vortex (TEV) of the airfoil play important roles in energy harvesting. Two major interaction modes between the airfoil and incoming vortices, the suppressing mode and the reinforcing mode, are identified. However, distinctions exist between the heaving and pitching motion in terms of their contributions to the interaction modes and the efficiency of the energy extraction. A potential theory and the related fluid dynamics analysis are developed to analytically demonstrate that the topology of the incoming vortices corresponding to the airfoil is the primary factor that determines the interaction modes. Finally, the trade-off between the input and the output is discussed. It is found that appropriate operational parameters for the heaving motion are preferable in order to preserve acceptable input power for energy harvesters, while appropriate parameters for the pitching motion are essential to achieve decent output power. 9:05AM A18.00006 Buoyancy-Induced Columnar Vortices1 , MARK SIMPSON, ARI GLEZER, Georgia Institute of Technology — Naturally-occurring, buoyancy-driven columnar vortices (“dust devils”) that are driven by an instability of the thermally stratified, ground-heated air layer and are sustained by entrainment of the ground-heated air, occur spontaneously in the natural environment with core diameters of 1-50 m and heights up to one km. These vortices convert low-grade waste heat in the air layer overlying the warm surface into a flow with significant kinetic energy that may be exploited for power generation by coupling the vortex to a vertical-axis turbine. The considerable kinetic energy of the vortex column cannot be explained by buoyancy alone, and the fundamental mechanisms associated with the formation, evolution, and dynamics of an anchored, buoyancy-driven columnar vortex are investigated in a laboratory facility using a heated ground plane and an azimuthal array of flow vanes. The present investigation focuses on the vortex formation, structure, and the dependence of its scaling and strength on the thermal resources and the characteristic scales of the anchoring flow vanes using stereo-PIV with specific emphasis on the production, advection, and tilting of vorticity within the entrained boundary layer. Approaches for the manipulation of these mechanisms for increasing the available kinetic energy and therefore the generated power are also investigated. 1 Supported by ARPA-E 9:18AM A18.00007 Numerical Investigation of Buoyancy-Induced Columnar Vortices1 , NICHOLAS MALAYA, ROY STOGNER, ROBERT MOSER, University of Texas at Austin — Buoyancy driven columnar vortices arise naturally in the atmosphere. A new energy harvesting approach makes use of this phenomenon by creating and anchoring the vortices artificially and extracting energy from them. In this talk, we explore the characteristics of these “solar vortices” through numerical simulation. Computational models of the turning vane system used to generate the solar vortex and the turbine used to extract energy have been developed. The formulation of these models and their validation against available experimental measurements will be discussed, as will the details of the columnar vortex structure and its interaction with the turbine. In addition, the computational models are being used to optimize the turning vane configuration and the turbine characteristics to maximize the power extraction, and to characterize the effects of environmental conditions such as cross winds and topography. Preliminary results from these studies will also be presented. 1 This work supported by the Department of Energy [ARPA-E] under Award Number [DE-FOA-0000670]. 9:31AM A18.00008 On a possible mechanism for the generation of cyclonic vortices regime in a precessing cylindrical container , WALEED MOUHALI, ECE Paris NanoLab, THIERRY LEHNER, Luth Observatoire Paris Meudon, ATER COLLABORATION — We report experimental observations obtained by particle image velocimetry of the behavior of a flow driven by rotation and precession in a cylindrical container. This study is motivated by dynamo effect and geophysics applications. Precessional motion forces inertial waves whose amplitude are predicted by a linear inviscid theory. But, various flow regimes are identified experimentally according to the value of the control parameter ε P the precession rate : the ratio of the precession frequency ΩP to the rotation frequency ΩR (ε = Ω ). When ε is increased from small values, after a linear ΩR regime, we have observed a differential rotation followed by the apparition of four permanent cyclonic vortices as a consequence of instability (eruption of jets from the lateral edges of the cylinder). We propose a mechanism for this instability based on a precedent study : we have proved that the nonlinear mode coupling of two inertial waves of azimuthal wave number m = 0 and m = 1 (mode forced by the precession) in the inviscid regime creates differential rotation also observed experimentally at small ε. The profile of the azimuthal mean velocity and the corresponding axial mean vorticity both show an inflexion point in their radial profile. We show that when the control parameter ε is increased from low values, the forced mode m = 1 can become instable in this induced differential rotation. It could be responsible for the observed instability and for the cyclones formation within the volume after a subsequent Kelvin-Helmholtz type instability. 9:44AM A18.00009 Piezoelectric Energy Harvesters in Isotropic Turbulence1 , AMIR DANESH-YAZDI, OLEG GOUSHCHA, NIELL ELVIN, YIANNIS ANDREOPOULOS, CUNY-CCNY — In the present work, we will report experimental and analytical results related to the extraction of fluidic energy in decaying homogeneous, isotropic turbulence using cantilever beams with attached piezoelectric patches of various materials. Turbulence carries mechanical energy distributed over a range of temporal and spatial scales and the resulting interaction of these scales with the immersed piezoelectric beams creates a strain field in the beam which generates electric charge. Experiments are carried out in large scale wind tunnels in which passive, semi-passive and active turbulence-generating grids are used to excite the piezoelectric cantilever beams at various distances from the grids. We observe that the average power generated in the piezoelectric layer obeys an exponential decay law with respect to the dimensionless distance parameter, as predicted from our theoretical hypothesis. The pertinent parameters that influence the power output of the beams are identified as (1) the dimensionless distance of the beam from the grid with respect to the grid size and (2) the dimensionless length of the beam with respect to the turbulence integral length scale. Furthermore, the efficiencies associated with each step of the energy conversion process in the beams are discussed. 1 Sponsored by NSF Grant: CBET #1033117 Sunday, November 23, 2014 8:00AM - 9:44AM Session A19 Convection and Buoyancy-Driven Flows: Phase Changes — 2006 - Eduardo Ramos, Universidad Nacional Autonoma de Mexico 8:00AM A19.00001 Solidification and natural convection in a Hele-Shaw cell , EDUARDO RAMOS, GUILLERMO RAMIREZ, JONATHAN CISNEROS, GUILLERMO HERNANDEZ-CRUZ, Universidad Nacional Autonoma de Mexico — Water solidification in presence of natural convection has been experimentally observed in a Hele-Shaw cell. Initially, the cell is filled with liquid water, and the upper and lower horizontal walls are kept at -10 ◦ C and 25 ◦ C respectively. The water starts to solidify near the upper wall and a solidification front advances downwards. The unstable temperature gradient triggers natural convection in the liquid filled region of the cell (where thermal and geometrical conditions lead a Rayleigh number larger tan critical). As the solidification front moves and its distance to the lower wall reduces, the Rayleigh number diminishes below the critical value and natural convection stops. The motion of the front and the natural convection motion are recorded with a video camera and measured with image processing and PIV respectively. 8:13AM A19.00002 Rotating ice blocks1 , STEPHANE DORBOLO, NICOLAS ADAMI, Univ de Liege, GRASP TEAM — The motion of ice discs released at the surface of a thermalized bath was investigated. As observed in some rare events in the Nature, the discs start spinning spontaneously. The motor of this motion is the cooling of the water close to the ice disc. As the density of water is maximum at 4◦ C, a downwards flow is generated from the surface of the ice block to the bottom. This flow generates the rotation of the disc. The speed of rotation depends on the mass of the ice disc and on the temperature of the bath. A model has been constructed to study the influence of the temperature of the bath. Finally, ice discs were put on a metallic plate. Again, a spontaneous rotation was observed. 1 FNRS is thanked for financial support 8:26AM A19.00003 The convective dynamics of a suspension of ice crystals1 , DAVID REES JONES, ANDREW WELLS, University of Oxford — The formation of solid crystals from a liquid cooled beneath its freezing temperature occurs in a wide range of environmental and industrial situations, such as in the formation of so-called “frazil ice” in rivers and the polar oceans. Eddies in the fluid flow act to keep the crystals suspended, while the relative buoyancy of the crystals causes them to rise, eventually sedimenting to form a layer of ice, called grease ice in the oceans. Here, we consider the interaction between the fluid dynamics of a suspension of crystals and the thermodynamics of phase change governing the growth and melting of the crystals. The crystals grow when the local temperature lies below the freezing temperature and melt when it lies above. We explore simplified scenarios that illustrate the important features of this multiphase flow and the effect of this “active suspension” on heat transfer. 1 This research is funded by the John Fell Oxford University Press (OUP) Research Fund. 8:39AM A19.00004 Influence of phase changes on Rayleigh-Benard convection1 , MICHAEL ADLER, KELKEN CHANG, RAYMOND SHAW, Michigan Technological University — How does a condensing trace species influence the convection of an inert carrier gas? The question is relevant to cloud formation at typical atmospheric conditions. We simulate Rayleigh-Benard convection in a cylindrical geometry over a range of Rayleigh numbers for which the flow remains laminar. The top and bottom boundaries are fixed at the equilibrium vapor pressure of the condensable species. The buoyancy is influenced by temperature, vapor, and condensate concentrations. The temperature and vapor fields combine via the Clausius-Clapeyron equation to yield a spatially complex supersaturation field; this, in turn, drives condensation. The temperature field is coupled to the condensation process through latent heating. The resulting volume heat source affects the convection and leads to a height dependent Nusselt number. For fixed Rayleigh number, increasing the temperature difference alters the buoyancy and Nusselt number height profiles. An analytical model for the latent heating profile for the purely diffusive case is shown to predict the correct magnitude of heating. This study sets the stage for experiments that will be carried out in the Michigan Tech Pi-Chamber, an aspect ratio 2 chamber with volume 3.14 m3 and controlled temperature and water vapor boundaries. 1 Supported by NSF grant AGS-1039742 8:52AM A19.00005 Natural convection during a phase change of sodium acetate trihydrate , YASUNORI OUCHI, The University of Tokyo, SATOSHI SOMEYA, TETSUO MUNAKATA, National Institute of Advanced Industrial Science and Technology — A latent heat storage system has higher storage capacity than a sensible heat storage system. Sodium acetate trihydrate has large latent heat at the temperature, 58◦ C, suitable for a hot-water supply system. The present study focused on convection in a phase change process to understand the heat transfer from the phase change material(PCM). The convection occurred only in certain conditions of supercooling temperature and PCM concentration. A spicular crystal grew quickly and the thermal convection couldn’t be detected at large supercooling temperature with high concentration of PCM. In the range of 5 ∼ 13◦ C of supercooling temperature, the buoyancy driven convection due to the latent heat of PCM was measured using the PIV. It was also observed that a part of CH3 COONa-3H2 O solution was sucked into the growing spicular crystals to supply CH3 COONa at the condition with small concentration and at 5 ∼ 13◦ C of supercooling temperature. 9:05AM A19.00006 Investigation of Thermal Stratification by Direct Contact Condensation in a Suppression Pool , KOJI OKAMOTO, DAEHUN SONG, NEJDET ERKAN, The University of Tokyo — Thermal stratification in the suppression pool of Boiling Water Reactor were investigated using simple slab-type experimental facilities. The steam direct condensation causes the vibration of bubble interface, resulting in the mixing enhancement at the nozzle. Using the bubble motion model, the Richardson number had been estimated which is a ratio of buoyancy to interface fluctuation momentum. The thermal stratification occurrence had been strong relation to the Richardson number. The flow pattern inside the chamber had also affected by the Richardson number. The flow pattern were measured by PIV. The velocity distribution were compared with the numerical simulation, showing the good agreement. In small Ri, i.e., lower fluctuation condition, the thermal stratification does occur. Thus, the momentum caused by the direct condensation determined the occurrence of the thermal stratification. 9:18AM A19.00007 Growth of mushy layers with temperature modulations1 , GUANG-YU DING, CHAO WU, JIN-QIANG ZHONG, Tongji University, Shanghai, China — Directional solidification of aqueous solutions produces a solid-melt coexisting zone whose growth rate can be predicted by the mushy-layer theory. We present measurements of mushy-layer growth when solidifying aqueous ammonium chloride with the cooling temperature modulated periodically TB (t) = T0 + Acos(ωt). The mush-liquid interface h(t) evolves as the square root of time for a constant TB , but exhibits periodical humps in the present of modulations. The growth rate ḣ(t) is best approximate to ḣ(t) = ḣ0 e−γωt/2π cos(ωt + π + φ(t)), with a decay rate γ = 0.82±0.05 independent on the modulation amplitude A and frequency ω, and a phase-shift φ(t) increasingly lag behind TB as a function of time. We discuss a mushy-layer growth model based on the Neumann solution of the Stefan problem with periodical boundary conditions, and show that the numerical results are in agreement with the experimental observations. 1 Supported by NSFC Grant 11202151. 9:31AM A19.00008 Convective instabilities in a ternary alloy mushy layer1 , DANIEL ANDERSON, George Mason University, PETER GUBA, Comenius University — We investigate a mathematical model of convection, thermal and solutal diffusion in a primary mushy layer during the solidification of a ternary alloy. In particular, we explore the influence of phase-change effects, such as solute rejection, latent heat and background solidification, in a linear stability analysis of a non-convecting base state solution. We identify how different rates of diffusion (e.g. double diffusion) as well as how different rates of solute rejection (double solute rejection) play a role in this system. Novel modes of instability that can be present under statically stable conditions are identified. Parcel arguments are proposed to explain the physical mechanisms that give rise to the instabilities. 1 This work was supported in part by the U.S. National Science Foundation, DMS- 1107848 (D.M.A.) and by the Slovak Scientific Grant Agency, VEGA 1/0711/12 (P.G.). Sunday, November 23, 2014 8:00AM - 9:57AM Session A20 Flow Control: Theory 2008 - Clancy Rowley, Princeton University — 8:00AM A20.00001 Heuristics for Effective Actuator and Sensor Placement in Feedback Flow Control1 , KEVIN CHEN, CLARENCE ROWLEY, Princeton University — Actuator and sensor placement can be just as consequential for the performance of localized feedback flow control as controller design. Yet, effective placement is not well understood, and the use of suboptimal placements is common. We report descriptions and characteristics of effective actuator and sensor placements for optimal flow control. We review optimal placements in the linearized Ginzburg–Landau and Orr–Sommerfeld/Squire models of fluid flow. We then analyze the feedback control of these models by relating physical observations with mathematical tools. Although these tools do not fully predict optimal placements, they do reveal patterns that most or all effective placements share. Most notably, effective actuator–sensor placements provide good authority over unstable modes and transient growth, and avoid large time lags between inputs and outputs. 1 This work was supported by the National Science Foundation’s Graduate Research Fellowship Program and grant CMMI-0932928. 8:13AM A20.00002 Control of streaks induced by free-stream turbulence in incompressible boundary layers: application to a linear model1 , GEORGE PAPADAKIS, LIANG LU, Imperial College London, PIERRE RICCO, University of Sheffield — Active wall-transpiration control of streaks generated within an incompressible boundary layer due to free-stream turbulence is examined. The flow model is based on the linearised unsteady boundary-region (LUBR) equations. The effect of free-stream turbulence appears as explicit forcing of these equations, given by an analytic expression, which is obtained by asymptotic matching with the far field conditions. The presence of the forcing term necessitates the reformulation of the control problem and the re-derivation of its solution. The objective cost function that is minimised comprises the weighted energy of the streak and the actuation. It is shown that the control signal consists of two components, a feed-back part (that depends on the state vector) and a feed-forward part. Explicit equations that provide these two components are derived. The developed method is efficient and has modest memory requirements. Computations with different wavenumbers in the wall normal direction demonstrates the significant effect of forcing for the same initial conditions. The effect of actuation on the perturbation energy and vorticity fields is examined. 1 This work was supported by EPSRC grant EP/I016015/1 8:26AM A20.00003 Approximate Balanced Truncation for Large Unstable Systems , THIBAULT FLINOIS, AIMEE S. MORGANS, PETER J. SCHMID, Imperial College London — A new snapshot-based extension of approximate balanced truncation to unstable systems that does not rely on the computation of global modes is presented. Applying feedback control to fluid flows often allows reaching the desired goal – e.g. drag minimisation, suppression of instabilities – more efficiently than passive and open-loop control, or where these approaches are ineffective. However a low-order approximation of the system’s input-output dynamics is often required in order to make controller design and online implementation tractable. Several system identification and model reduction procedures have been developed to this end: one such method that has received much attention is approximate balanced truncation, or balanced POD. It is applicable to stable systems and has recently been extended to unstable systems. This extension is based on the projection of the system onto its stable subspace, but this procedure can become computationally expensive for large systems. Here we show how balanced POD can be applied to unstable systems in a way that scales well even for very large systems, as it is projection-free and does not require computing global modes. We show that this method can be easily implemented by applying it to model systems and other selected flow configurations. 8:39AM A20.00004 4D-Var identification of DMD Reduced-Order Models1 , GILLES TISSOT, LAURENT CORDIER, BERND R. NOACK, PPRIME Institute — A reduced-order modelling (ROM) strategy is crucial to achieve model-based control in a wide class of flow configurations. In turbulence, ROMs are mostly derived by Galerkin projection of first principles equations onto the proper orthogonal decomposition (POD) modes. POD is widely used since it extracts from a sequence of data an orthonormal basis which captures optimally the flow energy. Unfortunately, energy level is not necessarily the correct criterion in terms of dynamical modelling and deriving a dynamical system based on POD modes leads sometimes to irrelevant models. In this communication, the Dynamic Mode Decomposition (DMD) as recently proposed by Schmid (JFM 2010) is used to determine the DMD modes. A DMD ROM is then derived by Galerkin projection of the Navier-Stokes equations onto a selected set of optimized-DMD modes. Finally, a four-dimensional variational assimilation approach (4D-Var) is employed to identify the coefficients of the DMD ROM. Essentially, 4D-Var combines imperfect observations, a background solution and the underlying dynamical principles governing the system under observation to determine an optimal estimation of the true state of the system. The methodology is illustrated for a DNS cylinder wake flow at Re=200 and PIV measurements at Re=13000. 1 Partially funded by the ANR Chair of Excellence TUCOROM and the Carnot project INTACOO 8:52AM A20.00005 Riccati-based Feedback Stabilization of an Oscillating Vertical Cylinder using a POD Reduced-Order Model1 , LAURENT CORDIER, GILLES TISSOT, BERND NOACK, Institute PPRIME — The aim of this communication is to demonstrate the use of Reduced-Order Model (ROM) based on Proper Orthogonal Decomposition (POD) to stabilize the flow over a circular cylinder in the laminar regime (Reynolds number equal to 60). The control is introduced by vertical oscillations of the cylinder, the objective being to determine by linear control the vertical velocity of the cylinder that stabilizes the flow. Since in Fluid-Structure Interaction, the POD algorithm cannot be applied directly, the fictitious domain method of Glowinski et al. (JMF 1999) is implemented where the solid domain is treated as a fluid undergoing an additional constraint. The POD-ROM is then classically obtained by projecting the Navier-Stokes equations on the first POD modes. The cylinder movement is enforced in the POD-ROM through the introduction of Lagrange multipliers. Finally, a Linear Quadratic Regulator framework is used to determine the optimal control law such that the flow is stabilized. 1 Partially funded by the ANR Chair of Excellence TUCOROM and the Carnot project INTACOO 9:05AM A20.00006 Numerical approximation of spectrum of the linearized Navier-Stokes operator for flow around an infinite cylinder , JONATHAN GUSTAFSSON, SIVAGURU S. SRITHARAN, Naval Postgraduate School — Numerical approximations of the spectrum of the Oseen operator and the linearized Navier-Stokes operator for flow around a cylinder in two dimensions have been studied for Reynolds numbers between 2 and 60. By approximating the eigenfunctions with a spectral method featuring basis functions covering the entire exterior domain, it is possible to obtain a numerical approximation to the continuous spectrum and the isolated eigenvalues (point spectrum). The numerical approximation of the spectra agrees with the previous rigorous results by Babenko (1982). That is a parabolic tongue containing the continuous spectra for the Oseen operator and a parabolic tongue containing the continuous spectrum plus a finite number of isolated eigenvalues for the linearized Navier–Stokes operator. The research feature a novel way of selecting location of collocation points. Future work will include control on the surface of the cylinder and examining its effect on both the unstable eigenvalues and the continuous spectrum. 9:18AM A20.00007 Model-based design of drag-reducing spanwise wall oscillations , MIHAILO JOVANOVIC, ARMIN ZARE, University of Minnesota — We study the model-based design of spanwise wall oscillations for drag reduction in a turbulent channel flow. Our approach selects the optimal period of oscillations by examining the influence of periodic base-flow-modification on Reynolds stresses. These are obtained from the linearized Navier-Stokes equations that are driven by colored-in-time stochastic forcing. Forcing correlations are selected, through a convex optimization procedure, to reproduce the statistical signature resulting from direct numerical simulation of the uncontrolled flow. We show that our analysis reliably predicts the optimal period of wall-oscillations in a simulation-free-manner. This demonstrates the effectiveness of our model-based approach in designing drag reducing wall oscillations and lays ground for utilizing such techniques in other passive or active flow control setups. 9:31AM A20.00008 A framework for studying the effect of compliant surfaces on wall turbulence1 , MITUL LUHAR, California Institute of Technology, ATI SHARMA, University of Southampton, BEVERLEY MCKEON, California In- stitute of Technology — It has long been recognized that compliant surfaces can serve as passive controllers for turbulent flows. However, the lack of a physics-based, computationally cheap theoretical framework that predicts the effect of compliant surfaces on turbulence has restricted progress towards designing performance-improving walls. To address this gap, we extend the resolvent analysis of McKeon & Sharma (2010, J. Fluid Mech.). Under this analysis, the turbulent velocity field is expressed as a linear superposition of propagating modes, identified via a gain-based decomposition of the Navier-Stokes equations. Compliant surfaces, modeled as a complex wall-admittance linking pressure and velocity, affect the gain and structure of these modes. Using a pattern search, we show that walls with unphysical negative damping are required to interact favorably with modes resembling the energetic near-wall cycle, which could explain why previous studies have met with limited success. Our results suggest that positive-damping walls could be effective for modes resembling the so-called very large-scale motions (VLSMs). Since the VLSMs have an organizing influence on smaller-scale turbulence, they may serve as a pathway for compliant walls to affect the entire flow. 1 This work is supported by AFOSR award FA9550-12-1-0469 (PM: Doug Smith) and AFOSR/EOARD award FA9550-14-1-0042 (PM: Gregg Abate). 9:44AM A20.00009 Global eigenfunction based actuation and sensor design for compressible, viscous flows1 , MAHESH NATARAJAN2 , JONATHAN FREUND3 , DANIEL BODONY4 , University of Illinois at Urbana-Champaign — A method is developed to estimate optimal actuator types and locations for controlling compressible, viscous flows using linear feedback. Based on an analysis of the eigensystem of the linearized compressible Navier-Stokes operator for steady baseflow, the forward and adjoint global modes are used to estimate of where the controller should be placed, and what type of controller (mass, momentum, energy, etc.) it should be. The method is demonstrated using direct numerical simulations of a separated boundary layer in a Mach 0.65 diffuser at different Reynolds numbers. The baseflow is taken as the true steady solution or the time-averaged flow. For sufficiently low Reynolds numbers, global stabilization of the flow is achieved; only partial stabilization is achieved at higher Reynolds numbers. 1 Rolls Royce North America and Office of Naval Research student, Department of Aerospace Engineering 3 Professor, Department of Mechanical Science and Engineering and Department of Aerospace Engineering 4 Associate Professor, Department of Aerospace Engineering 2 PhD Sunday, November 23, 2014 8:00AM - 9:57AM Session A21 Instability: Boundary Layers I — 2010 - Carlos Pantano, University of Illinois - Urbana 8:00AM A21.00001 Delaying formation of incipient turbulent spots by tuning compliant panel properties , IGE BORI, Department of Mechanical Engineering, Federal University of Technology, Minna, KHOON SENG YEO, Department of Mechanical Engineering, National University of Singapore — This work attempts to delay formation of incipient turbulent spots farther by tuning compliant panel (CP) properties, in a Blasius boundary layer flow. Direct numerical simulation (DNS) approach was used for the evolution of pulse-initiated disturbance wavepackets over finite-length CP. For over a single CP, modified potential flow theory was revisited in order to choose suitable stiff compliant panel through the estimation of onset velocities for both travelling wave flutter (TWF) and Static divergence (SD) waves. Out of the six cases investigated, case 3 delayed transition farther to the tune of ∼ 3.2% more, while case 6 that is characterized with very soft wall caused the transition to delay farther by almost 7.6% more, when compared with the results from our earlier investigations. In addition, the softness nature of case 6 CP is further confirmed from the spectral analyses results. 8:13AM A21.00002 Laminar-to-Turbulence Transition in Hypersonic Boundary Layers: Role of Dilatational Waves , CUNBIAO LEE, CHUANHONG ZHANG, YIDING ZHU, QING TANG, HUIJING YUAN, JIEZHI WU, SHIYI CHEN, MOHAMED GAD-EL-HAK, Peking Univ — A Mach 6 quiet wind tunnel experiment is carried out to study the turbulent transition in hypersonic boundary layer. It is found that the second instability acoustic mode is the key modulator of the transitional process which experiences a rapid growth and a very fast annihilation due to the effect of bulk viscosity. The second mode interacts strongly with the first vorticity mode to directly promote a fast growth of the latter and lead to immediate transition to turbulence. 8:26AM A21.00003 Receptivity to thermal noise in real airfoil configurations1 , PAOLO LUCHINI, University of Salerno — Thermal noise, the macroscopic manifestation of microscopic particle agitation, is present in fluid flow just as in electron flow in conductors or in other physical transport phenomena. When the flow acts as an amplifier, typically during transition to turbulence, the transition position can be influenced by the amplitude of external disturbances through the so called receptivity of the flow instabilities; internally generated thermal noise represents a thermodynamically enforced lower bound to how much disturbances can be reduced. In a previous paper (Seventh IUTAM Symposium on Laminar-Turbulent Transition, IUTAM Bookseries Volume 18, Springer, 2010, pp 11-18), the present author showed that the maximum transition distance in a Blasius boundary layer corresponds to a Reynolds number little above 6 · 106 and to an N -factor of the order of 13. Results to be exhibited at this conference show that in a real airfoil configuration the maximum transition Reynolds number imposed by thermal noise is even lower than on a flat wall, and not far from the actually observed transition position. It follows that thermal noise might actually have a role in natural transition; and that even a perfectly silenced laboratory environment cannot push the transition position much farther. 1 Work supported by the European Community through the RECEPT grant 8:39AM A21.00004 Linear stability of the flow in a toroidal pipe , PHILIPP SCHLATTER, JACOPO CANTON, RAMIS OERLUE, KTH Mechanics — While hydrodynamic stability and transition to turbulence in straight pipes has been studied extensively, the mechanisms leading to instability in curved pipes are less documented. Here, the first (linear) instability of the flow inside a toroidal pipe is investigated as an initial step in the study of the related laminar-turbulent transition process. In the toroidally bent pipe, the flow is governed by two parameters: the Reynolds number and the curvature of the torus, given as the ratio between the radii of the pipe and of the torus, and is maintained in motion by fixed axial flux. We use classical modal stability analysis, which includes computing nonlinear steady states for each parameter pair, and then studying the stability by solving an eigenproblem of the linearised Navier–Stokes operator. Results show that the flow is indeed modally unstable for all the studied curvatures in the range 0.01–1, with the Reynolds number about 3000. The frequency, wavenumber and mode shapes are strongly dependent on the curvature: The corresponding critical modes are mainly located in the region of the Dean vortices, and represent in general travelling waves. Also time-dependent nonlinear simulations highlight the importance of the linear modes in the transition process in the bent pipe. 8:52AM A21.00005 Investigation of wall temperature and angle of attack effects on boundary layer stability over a blunt cone1 , LIANG XIAN, School of Mathematics and Information Science, Beifang University for Nationalities, LI XINLIANG, LHD, Institute of Mechanics, Chinese Academy of Sciences — A new PSE (parabolized stability equations) method based on the general orthogonal curvilinear system of coordinates is developed. Combined with DNS data, the PSE method is used to investigate the effects of wall temperature and angle of attack (AOA) on stability of the boundary layer over a blunt cone. Results indicate that cooling the surface leads to higher wave number appearing in streamwise for a given frequency disturbance wave. Cooling the surface induces stronger harmonic and 3D disturbances comparing to the adiabatic wall case, which further accelerates the growth of multi disturbance modes in blunt cone boundary layer. Thus finally decreases the transition Reynolds number. Although the non-parallelism is markedly in conical flow, the non-parallel effects on the evolution characteristic of disturbance is not so obviously. So the PSE approach is a useful method in analysis of the hypersonic boundary layer stability over a blunt cone. The combined effects of the wall temperature and the AOA on the transition over the blunt cone are further to be studied in our present work. Many new nonmonotonic changes of transition position versus above variables have been found. 1 This work was supported by the NSFC Projects (11372330, 11072248), the 863 Program (No. 2012AA01A304) and the CAS Program (Nos. KJCX2EW-J01, XXH12503-02-02-04). 9:05AM A21.00006 Velocity streaks in a Blasius boundary layer induced by external streamwise vortices: numerical simulation and linear stability analysis1 , LORENZO SICONOLFI, SIMONE CAMARRI, Università di Pisa, JENS H.M. FRANSSON, Linne Flow Centre, KTH Mechanics — This work investigates numerically the streamwise velocity streaks generated in a Blasius boundary layer (BL) by an array of counter-rotating vortices. The array is positioned outside the BL and generates the streaks by velocity induction. This investigation is motivated by previous studies demonstrating that stable streamwise streaks can lead to a stabilization of the Tollmien-Schlichting (TS) waves and to a subsequent delay of the transition between laminar and turbulent regime. In most of the previous studies streamwise vortices generating the streaks lie inside the BL. Conversely, the conceptual configuration considered here, with vortices outside the BL, has potential advantages due to the lower dissipation rate of the vortices in the streamwise direction. Direct numerical simulations (DNSs) are carried out to study the flow, where the streamwise vortices are introduced in an idealized form. Interesting configurations are identified by DNS and a reference one is selected and investigated in details. Bi-global stability analysis shows benefic effects on the evolution of TS waves and allows the construction of a modified stability curve for the controlled flow. The resulting transition delay is also demonstrated by DNS. 1 PRACE is acknowledged for awarding us access to resource FERMI based in Italy at CINECA 9:18AM A21.00007 Transition to turbulence by interaction of free-stream and discrete mode perturbations1 , RIKHI BOSE, PAUL DURBIN, None — In this work based on DNS, boundary layer streaks have been induced by free-stream turbulence (FST). The FST is numerically generated following the work of Jacobs & Durbin (J. Fluid Mech., 428, 185, 2001). Modal interaction of FST induced streaks and a 2D TS wave have been noted by invoking the FST and TS wave disturbances at the inlet of the computational domain. For higher turbulence intensity of FST (3.5%), the flow undergoes bypass transition to turbulence as in Durbin & Wu (Ann. Rev. Fluid Mech., 39, 107, 2007). When low intensity FST (1%) and TS wave are specified at inlet, transition is triggered via secondary instability. Secondary instability is instigated by interaction of FST induced streaks and TS wave. The pattern of Λ structures observed in these studies is neither of H or K type transition and depends upon inlet frequency spectrum of FST. Frequencies smaller than the TS wave frequency grow and determine the instantaneous pattern of the secondary instability. The span-wise length scale of the Λ structures have approximately half the size seen in Herbert (Ann. Rev. Fluid Mech., 20, 487, 1988). The evolution of the secondary instability is spontaneous rather than forced by the inlet FST or TS wave disturbances. 1 The authors would like to thank National Science Foundation (NSF 1228195) for providing funding to support this research. 9:31AM A21.00008 Inviscid instabilities of non-planar transversely sheared flows governed by the generalized Rayleigh pressure equation1 , MOHAMMED AFSAR, Imperial College, Department of Mathematics, 180 Queen’s Gate, London, UK, ADRIAN SESCU, Mississippi State University, Department of Aerospace Engineering, Starkville, MS 39762, USA — Transition in boundary layer flow over flat/curved surfaces and at moderate to high freestream disturbances or under the influence of various surface roughness elements often involves inviscid secondary instability. This stage in transition can be pictured as being a parametric resonance-type phenomena where a unstable primary flow saturates to a more-or-less steady-state, susceptible to infinitesimal three-dimensional wave-like instability modes that grow much faster than the primary. In decades of research on boundary layers, experimenters have relied upon an inflection point in the wall normal y and/or spanwise directions z of the primary as a pre-cursor to transition. This assertion, based on Rayleigh’s theorem, does not however apply in transversely sheared flows. In this talk, we show that an alternative local criterion for inviscid secondary instability - sharing similarities to the original one-dimensional Rayleigh criterion - exists for a class of non-planar transversely sheared flows at long streamwise wavelength. Our general stability criterion is, remarkably, given by necessity of the surface U (y, z) possessing at least one saddle point in the plane. We analyze this saddle-point criterion numerically show its relevance to secondary instabilities. 1 M.Z.A. would like to anknowledge financial support from Laminar Flow Control (LFC-UK) Research Program at Imperial College London and would like to thank Professor Philip Hall for motivating his interest in this problem. 9:44AM A21.00009 Wavelet-based identification of localized turbulent regions in a transitional boundary layer , JOE YOSHIKAWA, YU NISHIO, SEIICHIRO IZAWA, YU FUKUNISHI, Tohoku Univ — A numerical study in order to develop a method to identify localized turbulent regions in a transitional boundary layer is carried out using a wavelet transformation. Finding the onset of turbulence is quite difficult because it is not easy to distinguish the localized turbulent regions from “non-active” groups of vortices. The base flow with low-speed streaks is generated by placing an array of obstacles. Then a short duration jet is ejected from the wall into the low-speed streak. First, a hairpin vortex appears in the laminar boundary layer which travels downstream growing up. Downstream, localized turbulent regions appear in the boundary layer, where a lot of vortices are entangled with each other. A wavelet analysis is applied to the spatial waveforms of streamwise velocity fluctuations obtained from these two flow fields. It is shown that the hairpin vortex appears as a high amplitude spot in the wavelet spectrum, which is small in both wavenumber-wise and streamwise scales. On the other hand, the isolated turbulent region appears more wide spread in the wavenumber-wise scale. So, using this method, localized turbulent regions can be identified. Sunday, November 23, 2014 8:00AM - 9:57AM Session A22 Instability: Rayleigh-Taylor I — 2012 - Oleg Schilling, Lawrence Livermore National Laboratory 8:00AM A22.00001 The evolution of the Rayleigh-Taylor and Richtmyer-Meshkov instabilities in a finite height domain1 , SNEZHANA I. ABARZHI, Carnegie Mellon University — We apply group theory analysis to systematically study the nonlinear evolution of the Rayleigh-Taylor (RT) and Richtmyer-Meshkov (RM) instabilities in a domain of a finite height. The fluids with similar and contrasting densities are considered in case of two-dimensional RT and RM instabilities that are driven by sustained and impulsive accelerations respectively. The flow is periodic normal to the acceleration direction and has no external sources. For the nonlinear boundary value problem a family of asymptotic solutions is found, and the properties of the family solutions as well as their stability are thoroughly analyzed. For the first time the relation is identified between the family parameter (e.g. the front curvature) and the velocity shear at the front. The growth-rate of shear-driven Kelvin-Helmholtz instability is evaluated. It is shown in the nonlinear RT and RM flows in finite height domain there is an intense motion in a vicinity of the front and there is effectively no motion away from the front. In a finite size the domain the flow is decelerating in comparison to the spatially extended case. The theory outcomes for the numerical modeling of the RT and RM instabilities and for the design of experiments are discussed. 1 The work is supported by the US National Science Foundation. 8:13AM A22.00002 Exploring the effects of a rigid body on the evolution of the Rayleigh Taylor instability , CHRISTOPHER BROWN, STUART B. DALZIEL, University of Cambridge — This talk discusses the effects of a rigid solid boundary impeding the evolution of the Rayleigh-Taylor (RT) instability. Previous experimental studies e.g. those of Linden, Dalziel and Davies Wykes, amongst others, used a solid rigid barrier to separate the two layers which when removed revealed the RT unstable interface. But what happens if the barrier is only partially removed? Initially the interface grows classically, however, this is soon replaced by two circulation cells, one either side of the barrier. The circulation forces fluid from both layers onto the interface at z = 0, resulting in a RT mixing zone superimposed onto the circulation cells. This RT mixing zone grows in a manner similar to that found by Andrews et al. for RT in water tunnels, except here the flow is modified by the end wall. Near to the end wall the two circulation cells are deflected vertically, stretching the mixing zone vertically along the end wall rapidly. Using a combination of ILES simulations and low Atwood number experiments this talk will present a model for a multi-stage mixing process, discussing the effects of the opening size on the density change of each layer, buoyancy driven flux through the opening and mixing efficiency. c British Crown Owned Copyright 2014/AWE. 8:26AM A22.00003 Exploring elastic and plastic regimes of Rayleigh-Taylor instability in solids1 , RINOSH POLAVARAPU, ARINDAM BANERJEE, Lehigh University — The elastic-plastic (EP) transition stage of Rayleigh-Taylor (RT) instability was studied in an accelerated elastic-plastic solid. A novel rotating wheel RT experiment with linear vibratory motion that centrifugally accelerates a test section with two-material interface was utilized. The test section consists of a container filled with air and mayonnaise, a non-Newtonian emulsion, with an initial perturbation between the two materials. Single mode perturbations of various amplitudes and wavelengths were analyzed earlier to find the effects of initial conditions on instability acceleration. Presently, the EP transition process for a stable interface before reaching the instability was verified by accelerating the test section to a magnitude which is slightly less than critical acceleration and imparting linear vibration which alters the radius of circular path and thus varies the magnitude of centrifugal force. The results were compared with various instability and EP transition criteria given by analytical growth models. 1 The authors acknowledge support of the DOE-SSAP (Grant # DE-NA0001975) and DOE-LANL subcontract (Grant #173-667-1) 8:39AM A22.00004 Progress on Multicomponent Reynolds-Averaged Navier–Stokes Model Development and Validation for Rayleigh–Taylor and Reshocked Richtmyer–Meshkov Turbulent Mixing1 , OLEG SCHILLING, Lawrence Livermore National Laboratory — Recent progress on the development and validation of a new K–ǫ multicomponent Reynolds-averaged Navier–Stokes model is discussed. The model includes mixture molecular dissipation and diffusion terms, molecular and turbulent enthalpy diffusion terms, and models for pressure–dilatation and dilatation dissipation. The model has successfully been applied to a set of ten reshocked Richtmyer– Meshkov mixing experiments, and more recently to experiments with larger Mach numbers and various Atwood numbers. An extension of the model to include a modeled density variance transport equation is described. The three-equation model is applied to various Rayleigh–Taylor mixing cases with complex accelerations. The evolution of various turbulence statistics, fields, and turbulent transport equation budgets are compared among these cases to elucidate differences in the turbulence production, dissipation and diffusion mechanisms. It is also shown that the mechanical turbulence timescale is poorly correlated with the molecular mixing timescale determined by the time-evolution of the molecular mixing parameter. 1 This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344. 8:52AM A22.00005 Dynamics of Rayleigh-Taylor driven flows at high Atwood numbers1 , MARK MIKHAEIL, Georgia Institute of Technology, BHANESH AKULA, THOMAS FINN, Texas A&M University, DEVESH RANJAN, Georgia Institute of Technology — For the first time, detailed simultaneous density and velocity turbulent statistics for Rayleigh-Taylor instabilities at Atwood number of 0.75 are measured. A new density probe capable of measuring gas volumetric concentration directly is used in parallel to a three-wire probe to obtain instantaneous density and velocity components simultaneously. Particle Image Velocimetry (PIV) is also implemented to obtain field-wise measurements. The self-similarity behavior of the velocity statistics, corresponding probability density function (PDF) and spectra are presented. Mie-scattering images taken in both stream-wise and span-wise direction at different instability times have illustrated the turbulent structures visible in the instability. 1 This work is graciously supported by DOE-National Nuclear Security Administration grant number DE-NA0001786 9:05AM A22.00006 Rayleigh Taylor Instability with Acceleration Reversals , DENIS ASLANGIL, Lehigh University, ANDREW LAWRIE, University of Bristol, UK, ARINDAM BANERJEE, Lehigh University — Self-similar evolution to turbulence of Rayleigh Taylor Instability (RTI) is studied for various acceleration histories, using high resolution numerical simulations. Incompressible, three-dimensional flow is modelled by MOBILE, a massively parallel solver, here using the Implicit Large Eddy Simulation technique. In the current work, accel-deccel-accel profiles with different reversal times and different deceleration periods are applied to RTI problem, to analyze their effects on self-similar evolution of RTI. Simulations are initialized with two initial conditions having the same initial energy but differing in terms of their mode number range. We will discuss a number of metrics which include low order metrics like mix widths, growth constants, molecular mixing parameter, and higher order turbulence parameters like second and higher order moments, their dissipations, and production-dissipation ratios which will also be useful in validating mix models. 9:18AM A22.00007 Development of a Gas-Driven Implosion Device for the Study of RayleighTaylor Instability in Gelatin Cylinders , ANDREW HIGGINS, JUSTIN HUNEAULT, McGill University — The study of Rayleigh-Taylor (RT) instability growth on the inner surface of imploding cylinders is relevant to a number of inertial confinement and magnetized target fusion schemes. More specifically, the feedthrough of perturbations on the outer surface of the cylinder to its inner surface can be a limiting factor in the convergence and compression provided by implosion schemes. A number of studies have been performed on gas-driven gelatin rings, providing an accessible manner to study the RT instability in a converging geometry. In this study, we present the development of a novel apparatus which uses the single point initiation of a detonable gas mixture that then wraps around a central plate to symmetrically implode gelatin cylinders. The implosion is visualized by high speed camera through a viewing window. The ability to independently vary the initial driving pressure and the internal cavity pressure, as well as the cylinder thickness, the initial perturbation size and mode number allows for the study of a wide range of feedthrough regimes. Of particular interest is the cylinder deceleration phase, where the gas in the cavity begins to decelerate the inner cylinder surface, leading to rapid growth of perturbations on the now RT unstable interface. 9:31AM A22.00008 On the two families of instability waves in rotating stratified media , CHRISTOPHE MILLET, CEA, DAM, DIF, JACQUES VANNESTE, University of Edinburgh, FRANCOIS LOTT, LMD, Ecole Normale Superieure — We reexamine the related problems of baroclinic instability of parallel shear flows, concentrating on the unbounded rotating stratified case. Two families of instability waves, each having a distinct 3D wave pattern and propagation characteristics, have been found. The key feature of one of the families of waves is the spatial transition, at the inertial critical level, from a balanced edge wave near the ground to gravity waves aloft. It is shown that at small Rossby numbers the classical WKB approach would fail to give even a first-order instability wave solution. A global solution based on the method of matched asymptotic expansions is constructed. Matching is carried out in a region where both the quasi-geostrophic and linear approximations hold. Matching the exponentially small terms that arise from the feedback of the inertia-gravity waves on the surface motion can be used to close the potential-temperature dynamics thereby providing a new model of surface dynamics. For large Rossby numbers, another family of instability waves has been found. This family of waves does not appear to have been clearly identified and systematically studied before. The physical mechanisms which give rise to this family of waves are discussed and reported here. 9:44AM A22.00009 The Zombie Instability: Using Numerical Simulation to Design a Laboratory Experiment , MENG WANG, SUYANG PEI, CHUNG-HSIANG JIANG, UC Berkeley, PEDRAM HASSANZADEH, Harvard University, PHILIP MARCUS, UC Berkeley — A new type of finite amplitude-instability has been found in numerical simulations of stratified, rotating, shear flows. The instability occurs via baroclinic critical layers that create linearly unstable vortex layers, which roll-up into vortices. Under the right conditions, those vortices can form a new generation of vortices, resulting in “vortex self-replication” that fills the fluid with vortices. Creating this instability in a laboratory would provide further evidence for the existence of the instability, which we first found in numerical simulations of protoplanetary disks. To design a laboratory experiment we need to know how the flow parameters — shear, rotation and stratification, etc. affect the instability. To build an experiment economically, we also need to know how the finite-amplitude trigger of the instability scales with viscosity and the size of the domain. In this talk, we summarize our findings. We present a map, in terms of the experimentally controllable parameters, that shows where the instability occurs and whether the instability creates a few isolated transient vortices, a few long-lived vortices, or long-lived, self-replicating vortices that fill the entire flow. Sunday, November 23, 2014 8:00AM - 9:57AM Session A23 Geophysical Fluid Dynamics: Stratified Turbulence I — 2001 - C.P. Caulfield, BP Institute and DAMTP, University of Cambridge 8:00AM A23.00001 Multiple instabilities in layered stratified plane Couette flow , C.P. CAULFIELD, BP Institute & DAMTP, University of Cambridge, T.S. EAVES, DAMTP, University of Cambridge — We consider the linear stability and nonlinear evolution of a Boussinesq fluid consisting of three layers with density ρa − ∆ρ/2, ρa and ρa + ∆ρ/2 of equal depth d/3 in a 2D channel where the horizontal boundaries are driven at a constant relative velocity ∆U . Unlike unstratified flow, we demonstrate that for all Rib = g∆ρd/(ρa ∆U )2 > 0, and for sufficiently large Re = ∆U d/(4ν), this flow is linearly unstable to normal mode disturbances of the form first considered by Taylor (1931). These instabilities, associated with a coupling between Doppler-shifted internal waves on the density interfaces, have a growth rate (maximised across wavenumber and Rib ) which is a non-monotonic function of Re. Through 2D simulation, we explore the nonlinear evolution of these primary instabilities at various Re, demonstrating that the primary instabilities grow to finite amplitude as vortices in the intermediate fluid layer before rapidly breaking down, modifying the mean flow to become susceptible to strong and long-lived secondary instabilities of Holmboe (1962) type, associated with vortices now localised in the top and bottom layers. 8:13AM A23.00002 Experimental study of mixing mechanisms in stably stratified TaylorCouette flow , PIERRE AUGIER, COLM-CILLE CAULFIELD, STUART DALZIEL, Univ of Cambridge — We consider experimentally the mechanisms of mixing in stably stratified Taylor-Couette (TC) flow in a TC apparatus for which both cylinders can rotate independently. In the case for which only the inner cylinder rotates, centrifugal instability rapidly splits an initially linear density profile into an array of thin nearly homogeneous layers. Shadowgraph, PIV and density profiles measured by a moving conductivity probe allow us to characterise this process and the resulting flow. In particular, we observe turbulent intrusions of mixed fluid propagating relatively slowly around the tank at the interfaces between the layers, leading to a time-dependent variation in the sharpness and turbulent activity at these interfaces, whose period scales with (but is much larger than) the rotation period. Interestingly, the turbulent intrusions are anti-correlated between adjacent interfaces leading to snake-skin-like patterns in the spatio-temporal diagrams of the density profiles. We also explore how the presence of a density stratification modifies end effects at the top and bottom of the cylinders, in both the presence and absence of primary centrifugal instability. 8:26AM A23.00003 Turbulence and mixing from optimal perturbations to a stratified shear layer , ALEXIS KAMINSKI, DAMTP, University of Cambridge, C.P. CAULFIELD, BPI & DAMTP, University of Cambridge, JOHN TAYLOR, DAMTP, University of Cambridge — The stability and mixing of stratified shear layers is a canonical problem in fluid dynamics with relevance to flows in the ocean and atmosphere. The Miles-Howard theorem states that a necessary condition for normal-mode instability in parallel, inviscid, steady stratified shear flows is that the gradient Richardson number, Rig is less than 1/4 somewhere in the flow. However, substantial transient growth of non-normal modes may be possible at finite times even when Rig > 1/4 everywhere in the flow. We have calculated the “optimal perturbations” associated with maximum perturbation energy gain for a stably-stratified shear layer. These optimal perturbations are then used to initialize direct numerical simulations. For small but finite perturbation amplitudes, the optimal perturbations grow at the predicted linear rate initially, but then experience sufficient transient growth to become nonlinear and susceptible to secondary instabilities, which then break down into turbulence. Remarkably, this occurs even in flows for which Rig > 1/4 everywhere. We will describe the nonlinear evolution of the optimal perturbations and characterize the resulting turbulence and mixing. 8:39AM A23.00004 Energy transfers, mixing efficiency and the internal structure of stratified Rayleigh-Taylor instability , MEGAN DAVIES WYKES, University of Cambridge, ANDREW LAWRIE, University of Bristol, STUART DALZIEL, University of Cambridge — Rayleigh–Taylor instability has been shown in experiments to have a high mixing efficiency (η > 0.75) when it occurs at an interface between two otherwise stably stratified layers. In this presentation, an implicit large eddy simulation (which uses numerical diffusion as a proxy for physical viscous diffusion) is used to model the instability and the resulting turbulent flow. The final state of simulations is shown to have an excellent match with experiments. The simulations allow the tracking of energy in the flow, revealing some interesting behavior with implications for the study of mixing in stratified flows more generally. 8:52AM A23.00005 Subharmonic instability during the off-critical reflection of an internal wave beam , VAMSI KRISHNA CHALAMALLA, SUTANU SARKAR, Univ of California - San Diego — Numerical simulations at laboratory scale are performed to study the reflection of an internal wave beam at a sloping bottom. When the incoming wave Froude number F ri is small, the reflection process can be approximated by linear theory and almost all of the reflected energy is confined to the primary wave frequency. In cases where the incoming wave Froude number F ri is sufficiently high (≈ 0.07 in the present study) and the internal wave angle is close to but greater than the slope angle, the reflected wave undergoes parametric subharmonic instability (PSI) resulting in the formation of two subharmonic waves with frequencies 0.33Ω and 0.67Ω. The energy in the subharmonics is found to be of the same order as that in the primary reflected beam. PSI is not found during critical reflection (α = β) at any incoming wave Froude number. Thus, reflection of internal waves at a near but off-critical slope provides a potential mechanism for mixing through the generation of subharmonic waves with smaller vertical scales that could break down into turbulence. 9:05AM A23.00006 Surface manifestation of internal waves emitted by an evolving stably stratified turbulent shear flow1 , QI ZHOU, PETER DIAMESSIS, Cornell University — Internal waves (IWs) from submerged turbulent sources may manifest themselves at the sea surface by generating coherent and persistent spatial features. Such IWs emitted by the turbulent wake of a towed sphere in a linearly stratified Boussinesq fluid are investigated numerically. The fully nonlinear three-dimensional simulations resolve both the wave-emitting turbulent wake at Reynolds number Re ∈ [5 × 103 , 105 ] and Froude number F r ∈ [4, 16, 64], and the subsurface region where the IWs interact with the sea surface which is modeled by a free-slip rigid lid. As the wake evolves for up to 250 units of buoyancy timescales, IW characteristics such as wavelength and frequency are measured both near the source and at the surface for comparison; the statistics of magnitudes and orientations of IW-induced surface strains are reported. Various IW impacts at the surface, such as local enrichment of surfactant and dispersion of ocean surface tracers, are also discussed. 1 Work funded by ONR grants N00014-08-1-0235 and N00014-13-1-0665 (Dr. R. Joslin) 9:18AM A23.00007 Initially Isotropic Turbulence Subjected to Stabilizing Stratification , STEVE DE BRUYN KOPS, University of Massachusetts Amherst, JAMES RILEY, University of Washington — When turbulence in a stably stratified fluid decays, it often does so without a continuous source of energy. As a result, the turbulence time scale increases relative to the buoyancy time scale so that the Froude number F deceases in time. In wakes, for instance, scaling arguments lead us to expect F ∼ O(1) one buoyancy period after the object has passed, and extensive studies have been carried out to understand how wakes evolve as the buoyancy force becomes increasingly important in time. Even in the unstratified case, though, a turbulent wake is a complicated flow to study. A much simpler configuration is isotropic homogeneous turbulence (IHT). For this study, simulated IHT that exhibits power-law decay is suddenly subjected to stabilizing stratification. The simulations use up to 8192x8192x4096 grid points to resolve the largest and smallest length scales of the flow over a span of at least 10 buoyancy periods. Two Reynolds numbers differing by an order of magnitude are considered, with the lower Reynolds number having a range of turbulence length scales comparable to that in laboratory experiments of stratified turbulent wakes. In this paper, the evolution of the flow as F deceases with time is discussed, as is the effect of the initial Reynolds number. 9:31AM A23.00008 Decaying turbulence as a testing ground for the strongly stratified turbulence theory , ANDREA MAFFIOLI, ANDREA MAFFIOLI, Department of Engineering, University of Cambridge — High-resolution direct numerical simulations of decaying stratified turbulence in parallelepiped-shaped domains are presented. We test various hypotheses and results of the strongly stratified theory, namely that (i) the dissipation scales with horizontal quantities ǫk = Uh3 /Lh , (ii) the potential energy dissipation scales in an analogous way (both of these are hypothesized in the theory and impossible to check in forced simulations), (iii) vertical Froude number is order one F rv = 1 when buoyancy Reynolds 2 much larger than one, (iv) Kolmogorov type scaling for horizontal spectrum of KE (v) Corrsin-Obukhov scaling for horizontal spectrum of number R = Re F rh PE (vi) vertical spectra dependent on stratification N : EK , EP = N 2 kv−3 . Finally, the growth of KE in the horizontally invariant modes with kh = 0, or shear modes, is investigated and their vertical spectra are compared to the vertical spectra of the full kinetic energy. 9:44AM A23.00009 Lagrangian and Eulerian Acceleration Statistics in Turbulent Stratified Shear Flows , FRANK JACOBITZ, University of San Diego, KAI SCHNEIDER, Aix-Marseille University, MARIE FARGE, Ecole Normale Superieure — The Lagrangian and Eulerian acceleration statistics in homogeneous turbulence with shear and stratification are studied using direct numerical simulations. The Richardson number is varied from Ri = 0, corresponding to unstratified shear flow, to Ri = 1, corresponding to strongly stratified shear flow. In addition, the scale dependence of the acceleration statistics is studied using a wavelet-based approach. The probability density functions (pdfs) of both Lagrangian and Eulerian accelerations show a strong and similar influence on the Richardson number and extreme values for Eulerian acceleration are stronger than those observed for the Lagrangian acceleration. Similarly, the Eulerian time-rate of change of fluctuating density is observed to have larger extreme values than that of the Lagrangian time-rate of change. Hence, the time-rate of change of fluctuating density obtained at a fixed location by an Eulerian observer is mainly due to advection of fluctuating density through this location, while the time-rate of change of fluctuating density following a fluid particle is substantially smaller, and due to production and dissipation of fluctuating density. Sunday, November 23, 2014 8:00AM - 9:57AM Session A24 Industrial Applications I — 2003 - Michele Guala, University of Minnesota 8:00AM A24.00001 Study of Hydrokinetic Turbine Arrays with Large Eddy Simulation , DANNY SALE, ALBERTO ALISEDA, University of Washington — Marine renewable energy is advancing towards commercialization, including electrical power generation from ocean, river, and tidal currents. The focus of this work is to develop numerical simulations capable of predicting the power generation potential of hydrokinetic turbine arrays–this includes analysis of unsteady and averaged flow fields, turbulence statistics, and unsteady loadings on turbine rotors and support structures due to interaction with rotor wakes and ambient turbulence. The governing equations of large-eddy-simulation (LES) are solved using a finite-volume method, and the presence of turbine blades are approximated by the actuator-line method in which hydrodynamic forces are projected to the flow field as a body force. The actuator-line approach captures helical wake formation including vortex shedding from individual blades, and the effects of drag and vorticity generation from the rough seabed surface are accounted for by wall-models. This LES framework was used to replicate a previous flume experiment consisting of three hydrokinetic turbines tested under various operating conditions and array layouts. Predictions of the power generation, velocity deficit and turbulence statistics in the wakes are compared between the LES and experimental datasets. 8:13AM A24.00002 Wingtip Devices for Marine Hydrokinetic Turbines , IVAYLO NEDYALKOV, JESSE SHULL, IAN GAGNON, JOHN BRINDLEY, MARTIN WOSNIK, University of New Hampshire — Wingtip devices have become widely used in aircraft and wind turbine applications. There are only a few examples of their usage on Marine Hydrokinetic Turbines (MHK), which have only recently been developed to utility scale. Novel wingtip devices were designed for use specifically in marine applications, to reduce wingtip vortex induced drag and with the additional considerations for suppressing tip vortex cavitation and avoiding significant bio-fouling. A reference foil, a generic wingtip, and new wingtip designs were studied numerically using OpenFOAM, and some of the wingtips (including the reference foil and the generic wingtip) were studied experimentally in the University of New Hampshire High-Speed Water Tunnel. The experimental test bed was designed specifically for this study and can accommodate various wingtips which extend to the center of the tunnel. Lift and drag were measured for different angles of attack and cavitation inception was studied. Additionally, pressure was recorded at 4 locations on each tip. The pressure ports were also used for mass injection studies. 8:26AM A24.00003 Inducer Performance with Varying Inlet Blade Angles with a Stability Control Device , RYAN LUNDGREEN, DANIEL MAYNES, Brigham Young University, KERRY OLIPHANT, Concepts NREC, STEVE GORRELL, Brigham Young University — High suction performance pumps use an inducer as the first stage of the pump to limit the amount of cavitation within the rest of the machine. Suction performance improves significantly when inducers are operated at low flow coefficients. Small blade angles are required at low flow coefficients to maintain a stable operation, however, they are more prone to cavitation blockage and are less robust structurally. It has been shown that a stability control device has a significant stabilizing effect on the flow through an inducer, particularly at low flow coefficients. A local increase in the mass flow rate at the leading edge of the inducer allows the blade to operate at the design flow coefficient regardless of the mass flow rate through the machine. This allows inducers with a greater inlet blade angle, which are less prone to cavitation blockage and can be more structurally robust, to maintain stable operation at low flow coefficients. Numerical simulations were conducted on four different inducers that implemented the stability control device, with inlet blade angles ranging from 7 to 14 degrees. Analysis from the results has led to significant insights into how changes in the inlet blade angle affect the physics and performance of the stability control device. 8:39AM A24.00004 Effects of the geometry of the exit of a tube in an oscillating flow1 , ELIA ECHEVERRÍA, College of Science and Technology, UACM, CARLOS MÁLAGA, Phisics Department, School of Science, UNAM, STEVEN CZITROM, ICMyLUNAM, ARTURO OLVERA, IIMAS-UNAM, CATALINA STERN, Phisics Department, School of Science, UNAM — The problem of optimizing the performance of a wave-driven seawater pump –comprising a resonant duct and an exhaust duct joined by a variable volume air-compression chamber– it is explored by studying oscillating flows at the exit of a tube. It is known that the performance of this pump depends on the geometry of the mouth of its intake tube. An inspection of the integral expression of the Navier-Stokes equation along a central streamline of this flow shows that changing the shape of the tube’s mouth modifies only the inertia and energy losses terms because both depend on the flow field at the chosen streamline. These changes must be such that the integral relation is preserved. Therefore, by measuring the inertial term (known as added mass), the term for losses can be measured indirectly. We developed a method to measure the added mass for oscillating flows in tubes with different mouth shapes and compared these measurements with those obtained for a model of the flow through the pump. Our results suggest a way to find a criterion for choosing the geometry of the mouth of the tubes in order to minimize dissipation and improve efficiency of the pump. 1 This work was supported by funds provided by DGAPA-UNAM (Project PAPITT-IN1188608). 8:52AM A24.00005 A Method for Turbocharging Four-Stroke Single Cylinder Engines1 , MICHAEL BUCHMAN, AMOS WINTER, Massachusetts Institute of Technology — Turbocharging is not conventionally used with single cylinder engines due to the timing mismatch between when the turbo is powered and when it can deliver air to the cylinder. The proposed solution involves a fixed, pressurized volume – which we call an air capacitor – on the intake side of the engine between the turbocharger and intake valves. The capacitor acts as a buffer and would be implemented as a new style of intake manifold with a larger volume than traditional systems. This talk will present the flow analysis used to determine the optimal size for the capacitor, which was found to be four to five times the engine capacity, as well as its anticipated contributions to engine performance. For a capacitor sized for a one-liter engine, the time to reach operating pressure was found to be approximately two seconds, which would be acceptable for slowly accelerating applications and steady state applications. The air density increase that could be achieved, compared to ambient air, was found to vary between fifty percent for adiabatic compression and no heat transfer from the capacitor, to eighty percent for perfect heat transfer. These increases in density are proportional to, to first order, the anticipated power increases that could be realized. 1 This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. 1122374 9:05AM A24.00006 A Novel Pressure Compensating Valve for Low-Cost Drip Irrigation , AMOS WINTER, ALEXANDER WIENS, Massachusetts Inst of Tech-MIT — Nearly one billion people are currently living as subsistence farmers in the developing world. Irrigation could drastically increase quality of life for these individuals by enabling them to grow more and higher value crops. However, current irrigation technologies are too costly for this economic sector, particularly in off-grid applications. The cost of an off-grid irrigation system is primarily driven by the power required to pump the water at a relatively high pressure (¿ 1 bar). We propose a novel pressure compensating drip emitter design which allows these systems to operate at 1/10 the pressure of current products, making them economically viable in developing markets. Our proposed solution is inspired by the resonating nozzle of a deflating balloon. We use a reduced order model to understand the physical principles which drive the cyclic collapse of the balloon nozzle. This knowledge is applied to propose a pressure compensating drip emitter consisting of a simple compliant tube in series with a rigid conical diffuser. A scaling analysis is performed to determine the ideal geometry of the system and the model is applied to demonstrate that the proposed design is capable of pressure compensation in the required operation range. Preliminary experiments are presented. 9:18AM A24.00007 ABSTRACT WITHDRAWN — 9:31AM A24.00008 Airborne Detection and Tracking of Geologic Leakage Sites , JAMEY JACOB, RAKSHIT ALLAMRAJU, ALLAN AXELROD, CALVIN BROWN, GIRISH CHOWDHARY, TAYLOR MITCHELL, Oklahoma State University — Safe storage of CO2 to reduce greenhouse gas emissions without adversely affecting energy use or hindering economic growth requires development of monitoring technology that is capable of validating storage permanence while ensuring the integrity of sequestration operations. Soil gas monitoring has difficulty accurately distinguishing gas flux signals related to leakage from those associated with meteorologically driven changes of soil moisture and temperature. Integrated ground and airborne monitoring systems are being deployed capable of directly detecting CO2 concentration in storage sites. Two complimentary approaches to detecting leaks in the carbon sequestration fields are presented. The first approach focuses on reducing the requisite network communication for fusing individual Gaussian Process (GP) CO2 sensing models into a global GP CO2 model. The GP fusion approach learns how to optimally allocate the static and mobile sensors. The second approach leverages a hierarchical GP-Sigmoidal Gaussian Cox Process for airborne predictive mission planning to optimally reducing the entropy of the global CO2 model. Results from the approaches will be presented. 9:44AM A24.00009 3D Finite Element Formulation of Nonlinear Partial-slip Condition on Curved Geometries , ONKAR SAHNI, FARHAD BEHAFARID, MANE, RPI, LAUREN FOVARGUE, King’s College — For many fluid flow problems, the behavior of the fluid at the physical boundaries doesn’t adhere to the traditional no-slip condition or perfect-slip law and exhibits a partial slip. This partial-slip behavior can be nonlinear. Additionally for real geometries of interest, physical boundaries are composed of arbitrary 3D curved surfaces. In this study we focus on a finite-element formulation that includes 3D nonlinear partial-slip condition on general curved surfaces. Using this formulation, we perform finite-element analysis of flow problems with such a nonlinear boundary condition. We present convergence studies on canonical problems and also include cases with complex curved surfaces. Sunday, November 23, 2014 8:00AM - 9:44AM Session A25 Particle-Laden Turbulence I — 2005 - Lance Collins, Cornell University 8:00AM A25.00001 Bringing Clouds into Our Lab! - The Influence of Turbulence on Early Stage Rain Droplets1 , ALTUG YAVUZ, RUDIE KUNNEN, GERTJAN VAN HEIJST, HERMAN CLERCX, Fluid Dynamics Laboratory, Department of Physics, Eindhoven University of Technology, WORTEX DYNAMICS GROUP TEAM — We are investigating a droplet-laden flow in an air-filled turbulence chamber forced by speaker-driven synthetic jets in many axes. The speakers are running in a random manner; yet they allow us to control and define the statistics of the turbulence. The influence of the turbulence on the behaviour of particles, both individually and collectively, is not well known. Therefore we study the motion of the droplets with tunable size in a turbulent flow, mimicking the early stages of raindrop formation. 3D Particle Tracking Velocimetry (PTV) is chosen as the experimental method to track the droplets and carry out the statistics. Thereby it is possible to study the spatial distribution of the droplets in turbulence using the so-called Radial Distribution Function (RDF), which quantifies the clustering of the droplets under turbulence conditions. Additionally, this technique allows us to measure velocity statistics of the droplets and the influence of the turbulence on droplet trajectories, both individually and collectively. In this contribution, for different turbulence conditions, we will present velocity statistics of the droplets and quantify their clustering using the RDF. 1 FOM 8:13AM A25.00002 On free-stream correction methods for particle-laden flows1 , JEREMY HORWITZ, ALI MANI, Stanford University — We examine the numerical implementation of point-particle drag laws for two-way coupled particle-laden flows. The Stokes drag formula is assumed to be valid for particles smaller than the smallest fluid scales and particle Reynolds numbers less than unity. Numerical implementations however can result in large errors in the computed drag force when the mesh size is comparable to the particle size. We present a quantification of this error and show that its source is rooted in estimation of the “free-stream” velocity. While the Stokes drag formula requires this “free-stream” velocity to be measured away from the particle, current numerical methods use sampling of fluid velocity at the location of the particle. We propose simple extrapolation procedures that estimate the true free-stream velocity in such systems. Investigations on test problems show that the proposed procedures successfully eliminate the steady-state drag error. 1 J. H. Supported by NSF GRF 8:26AM A25.00003 On the mechanism for the clustering of inertial particles in the inertial range of isotropic turbulence , LANCE COLLINS, ANDREW BRAGG, PETER IRELAND, Cornell University — In this talk, we consider the physical mechanism for the clustering of inertial particles in the inertial range of turbulence. By comparisons with DNS data we demonstrate that the mechanism in the theory of Zaichik et.al. (Phys. Fluids. 19:113308, 2007) quantitatively describes the clustering of particles in the inertial range. We then analyze the theory for isotropic turbulence in the limit Reλ → ∞. For arbitrary St (Stokes number), there exists a separation in the inertial range beyond which Str ≪ 1, where Str is the Stokes number based on the eddy turnover timescale at separation r. The inertial-range clustering in this limit can be understood to be due to the preferential sampling of the coarse-grained velocity gradient tensor at that scale. At smaller separations, there may be transitions to Str ∼ 1, where a path history symmetry breaking effect dominates the clustering mechanism, and in some cases Str ≫ 1, which implies ballistic behavior and a flat RDF. The scaling for each of these regimes is derived and compared to DNS, where applicable. Finally, we compare the results with the “sweep-stick” mechanism by Coleman and Vassilicos (Phys. Fluids 21:113301, 2009) and discuss the similarities and differences between the two theories. 8:39AM A25.00004 On the relationship between non-local clustering mechanisms and preferential accumulation , ANDREW BRAGG, LANCE COLLINS, Cornell University — In a recent paper (New J. Phys. 16:055013, 2014) we explained the physical mechanism for the clustering of inertial particles in turbulence contained within the theory of Zaichik et al. (Phys. Fluids. 19:113308, 2007). We showed that for particles with Stokes numbers in the limit St ≪ 1, particles accumulate outside of vortices due to the “centrifuge mechanism” proposed by Maxey. However, for St ≥ O(1), the centrifuge mechanism gives way to a non-local path history symmetry breaking mechanism. Despite the change in the clustering mechanism, the instantaneous particle positions continue to correlate with high-strain, low-vorticity regions of the turbulence. In this talk we show how the non-local mechanism is influenced by, but not dependent upon, the preferential sampling of the fluid velocity gradient tensor along the particles path histories in such a way as to generate a bias for clustering in regions with strong straining motions. Finally, we show how the non-local mechanism still generates clustering, but without preferential accumulation, in the limit where the timescales of the fluid velocity gradient tensor measured along the inertial particle trajectories vanishes (i.e., a white noise velocity field). 8:52AM A25.00005 Dynamics of light particles in turbulence , VARGHESE MATHAI, Univ of Twente, VIVEK PRAKASH, Stanford University, JON BRONS, CHAO SUN, DETLEF LOHSE, Univ of Twente, PHYSICS OF FLUIDS GROUP, UNIV OF TWENTE TEAM — Particle-laden turbulent flows occur widely in nature and industrial applications. The accelerations experienced by these particles can be extreme and intermittent, and are a measure of the forces acting on them. Most of the previous studies have focused on neutrally buoyant and heavy particles in turbulence. In this work, we experimentally study the Lagrangian dynamics of finite-size light particles in a nearly homogeneous and isotropic turbulent channel flow. We explore a range of size ratios and density differences to arrive at the transitional regime when the wake effects start to dominate particle dynamics. Our results suggest that light particle dynamics in turbulence is a strongly two-way coupled problem even for very small density differences with the continuous phase. 9:05AM A25.00006 Secondary flow and particle transport in a square duct , HOORA ABDEHKAKHA, GIANLUCA IACCARINO, Stanford University — Particle transport and deposition play a significant role in various industrial applications. Previous studies have shown that high magnitudes of the vorticity near the corners of a duct can cause higher accumulation of the particles close to the wall. The objective of this study is to investigate the effects of the secondary flows in the transport and deposition of particles in a turbulent square duct flow. In order to enhance our understanding of particle deposition, we performed three-dimensional direct numerical simulation of a square duct in low Mach number turbulent flow using a Lagrangian model for prediction of particle transport and deposition. To have a more comprehensive understanding of the effects of turbulent flow on particle deposition, simulations with different Reynolds numbers and particle Stokes numbers are performed. 9:18AM A25.00007 Turbulent pair dispersion in the presence of gravity1 , KELKEN CHANG, BENEDICT MALEC, RAYMOND SHAW, Michigan Technological University — We present numerical evidence of the alteration in the turbulent pair dispersion of heavy particles with two different Stokes numbers (bidisperse), whose effect on the dispersion is further compounded when a uniform gravitational acceleration is present. Lagrangian particle trajectories for fluid tracers, and bidisperse inertial particles with and without gravity were calculated from a direct numerical simulation of homogeneous, isotropic turbulence. Particle pair dispersion shows a short-time, ballistic (Batchelor) regime and a transition to super-ballistic dispersion that is suggestive of the emergence of Richardson scaling. The commonly used equation of motion for inertial, sedimenting particles and Kolmogorov scaling arguments are shown to capture the essential features of the pair dispersion at very short time and length scales. Between the ballistic and super-ballistic regions, the dispersions of both tracers and monodisperse inertial particles display a sub-ballistic behavior that is strongly suppressed in the bidisperse case. We attribute the suppression of the dispersion to a reduction in the correlation between velocity and acceleration increments, whose behavior we attempt to capture using a stagnation point model. 1 Supported by NSF grant AGS-1039742 9:31AM A25.00008 Collision statistics of inertial particles in two-dimensional homogeneous isotropic turbulence with an inverse cascade , RYO ONISHI, Japan Agency for Marine-Earth Science and Technology, J.C. VASSILICOS, Imperial College London — This study investigates the collision statistics of inertial particles in inverse-cascading 2D homogeneous isotropic turbulence by means of a direct numerical simulation (DNS). A collision kernel model for particles with small Stokes number (St) in 2D flows is proposed based on the model of Saffman & Turner (1956) (ST56 model). The DNS results agree with this 2D version of the ST56 model for St <0.1. It is then confirmed that our DNS results satisfy the 2D version of the spherical formulation of the collision kernel. The fact that the flatness factor stays around 3 in our 2D flow confirms that the present 2D turbulent flow is nearly intermittency-free. Collision statistics for St = 0.1, 0.4 and 0.6, i.e. for St <1, are obtained from the present 2D DNS and compared with those obtained from the three-dimensional (3D) DNS of Onishi et al. (2013). We have observed that the 3D radial distribution function at contact (g(R), the so-called clustering effect) decreases for St = 0.4 and 0.6 with increasing Reynolds number, while the 2D g(R) does not show a significant dependence on Reynolds number. This observation supports the view that the Reynolds-number dependence of g(R) observed in three dimensions is due to internal intermittency of the 3D turbulence. We have further investigated the local St, which is a function of the local flow strain rates, and proposed a plausible mechanism that can explain the Reynolds-number dependence of g(R). Sunday, November 23, 2014 8:00AM - 9:57AM Session A26 Turbulent Boundary Layers I — 2007 - Julio Soria, Monash University 8:00AM A26.00001 A unified theory for wall turbulence via a symmetry approach , ZHEN-SU SHE, XI CHEN, College of Engineering, Peking University, FAZLE HUSSAIN, Department of Mechanical Engineering, Texas Tech University — First principle based prediction of mean flow quantities of wall-bounded turbulent flows (channel, pipe, and turbulent boundary layer - TBL) remains a great challenge from both physics and engineering standpoints. Physically, a non-equilibrium physical principle governing mean properties in turbulent flows is yet unknown. Here, we outline a recently developed symmetry-based approach which derives analytic expressions governing the mean velocity profile (MVP) from an innovative Liegroup analysis. In analogy to the order parameter in Landau’s (1937) mean-field theory, we develop a concept of order functions which are assumed to satisfy a dilation group invariance - representing the effects of the wall on fluctuations - allowing us to construct a set of new invariant solutions of the (unclosed) mean momentum equation (MME). The theory is validated by recent experimental and numerical data, and identifies a universal bulk flow constant 0.45 for all three canonical wall-bounded flows, which asymptotes to the true Karman constant at large Reynolds numbers. The theory equally applies to the quantification of the effects of roughness (She et al. 2012), pressure gradient, compressibility, and buoyancy, and to the study of Reynolds-averaged Navier-Stokes (RANS) models, such as k-ωmodel, with significant improvement of the prediction accuracy. These results affirm that a simple and unified theory of wall-bounded turbulence is viable with appropriate symmetry considerations. 8:13AM A26.00002 Correspondences between self-similar mean dynamics and streamwise velocity behaviors in the inertial region of the turbulent boundary layer , ANG ZHOU, University of New Hampshire, JOSEPH KLEWICKI, University of New Hampshire, University of Melbourne — Self-similar mean dynamics are analytically known to exist over a well-defined inertial domain of turbulent wall-flows [Klewicki 2013, J. Fluid Mech. 718, 596]. Well-resolved streamwise velocity measurements up to δ + = 20,000 are used to investigate three measures of self-similarity in turbulent boundary layers, and compare their behaviors with those determined via analysis of the mean momentum equation. The measures include the Kullback-Leibler divergence (KLD) [Tsuji et al. 2005, Fluid Dyn. Res. 37, 293], the logarithmic decrease of even statistical moments [Meneveau & Marusic 2013, J. Fluid Mech. 719, R1], and the so-called diagnostic plot [Alfredsson & Orlu 2010, Euro. J. Mech. B/Fluids 42, 403]. The present findings indicate that the approximately constant KLD profiles and the approximately logarithmic moment profiles follow the same scaling but reside interior to the bounds of the self-similar inertial domain associated with the mean dynamics. Conversely, the bounds of the self-similar region on the diagnostic plot correspond closely to the theoretically estimated bounds. A self-consistent physical interpretation is briefly discussed. 8:26AM A26.00003 Symmetry-based theory for mean velocities in the flat plate turbulent boundary layer , XI CHEN, College of Engineering, Peking University, FAZLE HUSSAIN, Department of Mechanical Engineering, Texas Tech University, ZHEN-SU SHE, College of Engineering, Peking University — A major difference from channel and pipe flow in zero-pressure-gradient turbulent boundary layer –ZPG-TBL is the streamwise development of the mean velocity components. We report a symmetry-based theory for ZPG-TBL, which yields a complete prediction for both the streamwise and vertical mean velocities, i.e. U(x,y) and V(x,y). A significant result is the identification of a bulk flow constantκb , which achieves a highly accurate description of U above y+ ∼ 150; for a set of DNS data (Schlatter et al. 2010); the relative error is bounded within 0.1%. It is found that κb has a non-trivial streamwise development, and asymptote to 0.45 for large Re’s; the latter is consistent with the true Karman constant recently discovered for channel and pipe flows. The theory assumes a fractional scaling for the total stress, which yields, for the first time, an analytical prediction for V, Reynolds stress profile, friction coefficient and shape factor in ZPG-TBL, in good agreement with both DNS and experimental data. In conclusion, a complete analytical theory is viable for both laminar (i.e. Blasius) and turbulent boundary layers. 8:39AM A26.00004 Statistical structure and scaling behaviors of spanwise vorticity in smoothwall turbulent boundary layers1 , JOSEPH KLEWICKI, University New Hampshire/Melbourne, CALEB MORRILL-WINTER, IVAN MARU- SIC, University of Melbourne — Within the canonical turbulent boundary layer the spanwise component of vorticity, ωz , is the only component that has a non-negligible mean value. For this and other reasons, the motions bearing ωz play a central role in boundary layer dynamics. A compact four element (‘Fossstyle’) hotwire probe was used to acquire well-resolved ωz fluctuation time series over an unprecedented Reynolds number range, 1, 500 ≤ δ + = δuτ /ν ≤ 15, 000. Very good spatial resolution (≤ 9 viscous units) was maintained over the entire δ + range by leveraging the low speed and large scale attributes of the HRNBLWT and FPF wind tunnels at Melbourne and New Hampshire, respectively. The present talk documents the behaviors of the statistical moments and frequency spectra of the ωz fluctuations, and further explores the self-similarity between the mean and rms profiles seen at low Reynolds number. The observed ωz behaviors are discussed relative to mean dynamical structure and the asymptotic properties of the boundary layer vorticity field. 1 The support of the Australian Research Council and the National Science Foundation are gratefully acknowledged. 8:52AM A26.00005 Phase relations of triadic scale interactions in turbulent flows1 , SUBRAHMANYAM DUVVURI, BEVERLEY MCKEON, California Institute of Technology — The quadratic nature of non-linearity in the Navier-Stokes (NS) equations dictates the coupling between scales in a turbulent flow to be of triadic form. An understanding of the triadic coupling affords good insights into the dynamics of turbulence, as demonstrated by Sharma & McKeon (J. Fluid Mech., 2013) through analysis of the NS resolvent operator; a set of three triadically consistent spatio-temporal modes was shown to produce complex structures such as modulating packets of hairpin vortices observed in wall-bounded turbulent flows. Here we interpret Skewness (Sk) of velocity fluctuations and the Amplitude Modulation coefficient (Ram ), proposed by Mathis, Hutchins & Marusic (J. Fluid Mech., 2009), to be a measure of the large- and small-scale phase relationship. Through a simple decomposition of scales, both Sk and Ram are shown to be amplitude weighted (and normalized) measures of phase between scales that have direct triadic coupling. An analytical relationship is established between the two quantities and the result is demonstrated using experimental data from canonical and dynamically forced turbulent boundary layers presented in Duvvuri and McKeon (AIAA 2014-2883). 1 The support of AFOSR (grant no. FA 9550-12-1-0469) and Resnick Institute Graduate Research Fellowship (S.D.) is gratefully acknowledged. 9:05AM A26.00006 Uniform momentum zones in turbulent boundary layers , CHARITHA DE SILVA, IVAN MARUSIC, NICHOLAS HUTCHINS, Univ of Melbourne — We examine the properties of large regions of uniform streamwise momentum in turbulent boundary layers using databases obtained from particle image velocimetry that extend over 2.3 δ (where δ denotes the boundary layer thickness) in the streamwise direction and 1.2 δ in the wall-normal direction. The investigation covers a large range of Reynolds numbers, spanning more than an order of magnitude (Reτ = 103 − 104 ), but with adequate spatial resolution to resolve most structural features. This enables accurate descriptions of the structural evolution of the uniform momentum zones (UMZs) as a function of Reynolds numbers. Our analysis reveals evidence of a hierarchical length scale distribution of structures within turbulent boundary layers, leading to zonal-like organisations. The Reynolds number dependence of these features is also investigated. Interpretation of these results is aided by employing synthetic velocity fields generated by using the attached-eddy model. Comparisons between the model and experimental results show that the widely proposed packet model would lead to a distribution of UMZs that conforms closely to those observed experimentally in this study. 9:18AM A26.00007 A bursting phenomenon in a vortex-gas boundary layer , AARTHI SEKARAN, RODDAM NARASIMHA, Jawaharlal Nehru Centre for Advanced Scientific Research, RAMA GOVINDARAJAN, TIFR Centre for Interdisciplinary Sciences — Bursts are a central phenomenon in turbulent boundary layers as they are an integral part of turbulent energy and stress production. They have consequently been a continuing area of interest since the 1970s following the detailed investigations of Kline et al. (1967). Despite several attempts to understand their dynamics, it has been difficult to arrive at a consensus even on the scaling of the burst frequency. The present investigation simulates the outer part of a plane turbulent boundary layer using the vortex-gas model, in a first step towards understanding the role of the outer layer in boundary layer dynamics. Preliminary results indicate the formation of regions of concentrated vorticity near the wall, at a frequency that is independent of the initial vortex configuration but a function of the mean velocity profile. Further, comparisons with existing experimental data indicate a burst frequency which when scaled on outer variables, is within the range of scatter among different studies. Quadrant occupancy statistics are also related to those in conventional boundary layers. It appears as if a bursting phenomenon of some kind may be a general feature of an inviscid, wall-bounded shear flow, and does not necessitate inclusion of either viscosity or three-dimensionality. 9:31AM A26.00008 Coherent structures in homogeneous shear turbulence compared with those in channels1 , SIWEI DONG, ADRIÁN LOZANO-DURÁN, ATSUSHI SEKIMOTO, JAVIER JIMÉNEZ, Universidad Politécnica de Madrid — Three-dimensional vortex clusters and coherent structures responsible for the momentum transfer (Qs) are studied by DNS in homogeneous shear turbulence (HST) at Reλ = 50, 100 and 250, with emphasis on comparisons with channel turbulence (CH). The anisotropic orientation of those structures only appears for volumes larger than L3c (Lc is the Corrsin scale). Even in that case, their anisotropy is moderate, similar to the detached structures in the CH. Only strictly attached structures in channels are more anisotropic. The Reynolds stress contained in vortex clusters is mainly associated with Q− s, distributed equally between sweeps (Q4) and ejections (Q2), instead of preferentially with the latter, as in the CH. The average fractal dimension of Qs is roughly 2.1 and that of vortex clusters is 1.8. The relative positions of the structures reveal that they form streamwise trains of groups of a Q2 and a Q4, paired side-by-side in the spanwise direction, with vortex clusters in between, as in the CH. 1 Funded by the ERC Multiflow program and CSC 9:44AM A26.00009 Dominant length scale of the “pure” turbulent fluctuations in the outer region of wall turbulence1 , YONG SEOK KWON, JASON MONTY, NICK HUTCHINS, Univ of Melbourne — A new method of decomposing the total velocity in boundary layers, which removes the influence of instantaneous boundary layer thickness variations to the fluctuating velocity component, is proposed. The recent proposition of the quiescent core of turbulent channel flow by Kwon et al. (J. Fluid Mech., vol. 751, 2014, pp. 228–254) permits us to apply the same decomposition to channel flows where the quiescent core is analogous to the free-stream. Using this decomposition, it is observed that the majority of the large-scale streamwise velocity fluctuation within the intermittent region is attributed to the oscillation of the turbulent/non-turbulent interface or the quiescent core. It suggests that the quiescent core and the free-stream play a similar role and the flow nearer to the wall in both flows is more similar than previously thought while the different characteristics of the free-stream and the quiescent core account for the differences in the outer region of two flows. These findings re-affirm the analogy between the quiescent core and the free-stream, which could potentially lead to the unified conceptual model between internal and external flows. 1 This work is financially supported by the Australian Research Council and the Defence Science and Technology Organisation. Sunday, November 23, 2014 8:00AM - 9:57AM Session A27 Stratified Turbulent Flows — 2009 - Elias Balaras, George Washington Univesrity 8:00AM A27.00001 Probing the effect of buoyancy on second-order statistics in stablystratified boundary layers1 , ELIE BOU-ZEID, STIMIT SHAH, Princeton University — Statically-stable turbulent boundary layer flows are particularly challenging due to the potential breakdown of Kolmogorov’s theory and to the emergence of laminar regions, gravity waves, and other complicating flow patterns. To develop a more fundamental understanding of how buoyancy influences turbulence in such flows, direct numerical simulations and large eddy simulations of turbulent boundary layers with rotation are performed. Under the highest stabilities, global intermittency (the almost compete decay of turbulence and then its regeneration) is observed, but could be the result of initial and boundary conditions rather than flow dynamics. Under more moderate stabilities, continuous turbulence is maintained, but it is significantly damped compared to neutral flows. This reduction of the TKE under stable conditions is very well known; however, here we show that it is mainly triggered by reduced mechanical production associated with reduced transport of Reynolds stresses from aloft toward the surface, rather than by direct destruction of TKE by buoyancy. This raises questions about the suitability of some conventional stability parameters, such as the flux Richardson number, in describing the influence of buoyancy in such flows. 1 supported by NSF-PDM under AGS-1026636 8:13AM A27.00002 Energetics of vertical fluid particle dispersion in stably stratified turbulence , JAMES ROTTMAN, SEUNGBUM JO, KEIKO NOMURA, University of California, San Diego — The vertical dispersion of fluid particles in stably stratified turbulence is investigated. We present an analysis framework which describes the associated flow energetics in the Lagrangian frame. The available potential energy (APE) density is a locally defined quantity associated with nonequlibrium displacement. The equilibrium potential energy (EPE) density is defined accordingly and represents the minimum energy required to change the particle equilibrium height. The corresponding evolution equations elucidate the key sequence of processes and clarify previous interpretations of the transport mechanisms. The analysis shows that in the case of stationary flow, the rate of mean square displacement is equal to the rate of mean square equilibrium displacement which is given by the scalar dissipation rate. A dispersion model is developed and compared with previous models. 8:26AM A27.00003 Turbulent transport across an interface between dry and humid air in a stratified environment , DANIELA TORDELLA, LUCA GALLANA, FRANCESCA DE SANTI, SILVIO DI SAVINO, RENZO RICCHIARDONE, MICHELE IOVIENO, Department of Mechanical and Aerospace Engineering, Politecnico di Torino, Italy — The transport of energy and water vapor across a thin layer which separates two decaying isotropic turbulent flows with different kinetic energy and humidity is considered. The interface is placed in a shearless stratified environment in temporal decay. This system reproduces a few aspects of small scale turbulent transport across a dry air/moist air interface in an atmospheric like context. In our incompressible DNS at Reλ = 250, Boussinesq’s approximation is used for momentum and energy transport while the vapor is modeled as a passive scalar (Kumar, Schumacher & Shaw 2014). We investigated different stratification levels with an initial F r between 0.8 and 8 in presence of a kinetic energy ratio equal to 7. As the buoyancy term becomes of the same order of the inertial ones, a spatial redistribution of kinetic energy, dissipation and vapor concentration is observed. This eventually leads to the onset of a well of kinetic energy in the low energy side of the mixing layer which blocks the entrainment of dry air. Results are discussed and compared with laboratory and numerical experiments. A posteriori estimates of the eventual compression/expansion of fluid particles inside the interfacial mixing layer are given (Nance & Durran 1994). 8:39AM A27.00004 Intermittent dynamics in stably stratified plane Couette flows1 , ENRICO DEUSEBIO, JOHN R. TAYLOR, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, COLM-CILLE CAULFIELD, BP Institute & DAMTP, University of Cambridge, RICH R. KERSWELL, School of Mathematics, University of Bristol — Turbulence in a stratified fluid is a fundamental process in the atmosphere and oceans, responsible for mixing density and various tracers and dissipating kinetic and potential energy. Although turbulence is generally suppressed in very statically stable conditions, intermittent bursts of turbulence are still seen when the Reynolds number is sufficiently large. In this work, we study stratified turbulence in plane Couette flow using direct numerical simulations, focusing on the complexity arising from the spatiotemporal intermittency of the flow as the stabilizing stratification increases. Two external dimensionless parameters control the dynamics: the Reynolds number Re and the bulk Richardson number Rib . We trace the boundary between laminar and turbulent states in the Re-Rib plane and discuss the relevant dynamical quantities involved in the relaminarization process. We analyze the structures populating the intermittent regime and the coexistence between laminar and turbulent patches, focusing on similarities and differences between small-Re-small-Rib and large-Re-large-Rib intermittent dynamics. We conclude by discussing the applicability and breakdown of existing stratified turbulence theories, including the Monin-Obukhov self-similarity theory. 1 MUST EPSRC Programme Grant 8:52AM A27.00005 Large-eddy simulations of stratification layer erosion by a jet1 , ALEKSANDR OBABKO, ELIA MERZARI, Argonne National Laboratory, ANANIAS TOMBOULIDES, Aristotle University of Thessaloniki, Greece, SHASHI AITHAL, PAUL FISCHER2 , Argonne National Laboratory — Following Fukushima disaster, the OECD/NEA has chosen the PANDA experiment for 2014 benchmark exercise where predictive capabilities of computational fluid dynamics (CFD) tools are tested for multispecies convection in notorious regime of transition from turbulent to laminar flow and from forced to natural convection. Accurate prediction of these phenomena will beneficial for a range of applications including reactor thermal-hydraulics where it will further our understanding of reactor behavior during accidents and help design safer and more efficient reactors for a carbon-free energy option. In fact, the convection and mixing flow in the containment played an important role in the Fukushima accident as the buoyant hydrogen gas mixed with oxygen and detonated resulting in significant destruction and radioactive pollution. Here we present the three-dimensional large-eddy (LES) simulations of the PANDA experiment with the spectral-element open-source code Nek5000. The results are compared and contrasted for a range of parameters using Boussinesq and low-Mach number approximations. 1 Partially 2 also funded by DOE NE NEAMS Program and used ALCF resources supported by the DOE Office of Science under contract DE-AC02-06CH11357. University of Illinois at Urbana-Champaign 9:05AM A27.00006 Large-eddy simulations of stratified turbulence at the buoyancy scale , SINA KHANI, MICHAEL L. WAITE, University of Waterloo — The effects of buoyancy scale dynamics on large-eddy simulations (LES) of stratified turbulence are presented in this talk. Two common subgrid-scale (SGS) models, the Kraichnan and Smagorinsky models, are considered. It is shown that if the LES filter scale ∆ is small enough compared to the buoyancy scale Lb = 2πurms /N , the horizontal wavenumber energy spectra show an approximately −5/3 slope along with a bump at the buoyancy wavenumber kb = N/urms . Here urms and N are the root-mean-square velocity and the buoyancy frequency, respectively. Our results also suggest that this criterion on ∆ leads to that the dynamics of stratified turbulence, including Kelvin-Helmholtz (KH) instabilities, being better captured in LES. In addition, the minimum necessary ratio of ∆/Lb for resolving KH instabilities depends on the SGS model, with the Smagorinsky model requiring a filter scale three times smaller than the Kraichnan model to adequately capture stratified turbulence. 9:18AM A27.00007 The turbulent, stratified near wake of a sphere at Re = 3700 and Fr = 3 , ANIKESH PAL, University of California San Diego, ANTONIO POSA, ELIAS BALARAS, The George Washington University, SUTANU SARKAR, University of California San Diego — Direct numerical simulation of flow past a sphere in a stratified fluid has been carried out at a sub-critical Reynolds number of 3700 and Froude number of 3. The choice of Re = 3700 allows validation against previous unstratified wake simulation including the recent DNS of Rodriguez et al. (2011). The conservation equations are solved in a cylindrical coordinate system and an immersed boundary method is employed to represent the sphere. The primary focus of this study is to understand buoyancy effects on near wake characteristics. The separated shear layer from the surface of the sphere becomes unstable resulting in transition to turbulence. The recirculation region is found to be affected by buoyancy. The turbulent stratified wake experiences a substantial suppression in the vertical direction in comparison to the corresponding unstratified case. Nevertheless, in the horizontal direction, the turbulent wake expands significantly more than in the unstratified case. Changes in the intensity, spectral content and structure of near-wake fluctuations in the wake are assessed. The momentum and energy transported by the internal gravity waves generated by the turbulent wake are also quantified. 9:31AM A27.00008 Direct simulations of unstratified and stratified turbulent flow past a sphere using a body-conforming grid , KARU CHONGSIRIPINYO, SUTANU SARKAR, Univ of California - San Diego — Direct numerical simulations of unstratified and stratified flows past a sphere are conducted in the sub-critical regime at Re = 3700. The objective is to investigate the flow at and near the body. This study takes advantage of a body-conforming grid which provides better accuracy and boundary layer representation than possible with the immersed boundary method. The body-fitted grid is generated by creating half of a C-type grid using a hyperbolic grid generation method and then rotating it around the wake-cut axis. The incompressible Navier-Stokes equations are solved using a semi-implicit scheme with the Crank-Nicholson method and the low storage RKW3-ADI. The Poisson equation for pressure correction is solved using semi-coarsening multigrid (SMG) from the HYPRE library. The singularity problem at the wake-cut is resolved by rewriting the discretized governing equation in finite-volume formulation and then set all fluxes across the cut to zero. For flux terms at nodes in the proximity of the wake cut, one-sided finite difference is used to avoid crossing the wake cut. Boundary layer separation and vortical structures immediately behind the sphere are examined. Turbulence statistics in the near wake region for unstratified and stratified flows are also compared. 9:44AM A27.00009 High Reynolds Number Near-Field Stratified Wake Measurements behind a Sphere1 , KENNETH KALUMUCK, ALAN BRANDT, KARA SHIPLEY, MICHAEL JOZKOWSKI, Johns Hopkins Univ. Applied Physics Laboratory — To characterize the near-field of a stratified wake at Reynolds numbers, Re ∼ 2 x 105 - 106 , experiments are being conducted in a thermally stratified fresh water lake with large diameter (D ∼ 0.5 m) spheres. The submerged sphere and associated instrumentation are affixed to a frame that is towed through the lake at velocities U ∼ 0.5 - 2 m/s. Measurements of three components of the turbulent fluctuating and mean wake velocities are being made with Acoustic Doppler Velocimeters (ADVs), while density fluctuations (inferred from temperatures) are being made with an array of fast response thermistors. Stratification is such that BV frequencies, N, up to 50 cph (0.09 /s) can be achieved, enabling Froude numbers F=U/ND ≥ 10. Existing stratified near-field wake data for spheres are for Re ∼ 104 and less, while only a very limited set of data under simpler unstratified conditions exists at these large Re, primarily measurements along the sphere (drag, pressure, separation) rather than wake data. Advances in CFD have enabled simulations at these high Reynolds numbers without quantitative data available for validation despite the existence of many natural and man-made systems that operate in these ranges. Here, experimental system design, results of a preliminary data set, and plans for ongoing and future work are presented. 1 This work is sponsored by the ONR Turbulence and Wakes program. Sunday, November 23, 2014 8:00AM - 9:57AM Session A28 Turbulence: Interfaces — 2011 - Carlos B. da Silva, Tecnico Lisboa 8:00AM A28.00001 Experimental investigation of entrainment processes of a turbulent jet , DHIREN MISTRY, University of Cambridge, JAMES R. DAWSON, Norwegian University of Science and Technology — We implemented simultaneous, time-resolved, multi-scale-Particle Image Velocimetry (PIV) and Planar Laser-Induced Fluorescence (PLIF) to study entrainment processes in the far-field of a round, turbulent jet. The experiments were performed using water as the test medium and a passive dye with a Schmidt number Sc ≫ 1 to identify the turbulent/non-turbulent (T/N-T) interface of the jet. The Reynolds number based on the nozzle exit is Re = 25, 300 and is considerably higher than existing studies of entrainment in jets. Independent 2D PIV and PLIF measurements confirmed that the far-field flow characteristics agree well with the classical scaling laws of turbulent jets. We use the auto-correlation of entrainment velocity along the T/N-T interface to show that the interface is dominated by fluid motion of O(λ). We also show that there exists a balance between the mass flux across the interface calculated at the small-scales and the mass flux calculated at larger scales. The interface is more convoluted at smaller scales, which results in a larger interfacial surface area. The mass-flux balance therefore indicates that the entrainment velocity at the interface scales at a rate that is inversely proportional to the surface area. 8:13AM A28.00002 Effect of free stream turbulence on the entrainment characteristics of jets1 , TOMOAKI WATANABE, Nagoya University, 464-8603 Nagoya, Japan, CARLOS B. DA SILVA, LAETA, IDMEC, Instituto Superior Tecnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal, YASUHIKO SAKAI, KOUJI NAGATA, Nagoya University, 464-8603 Nagoya, Japan, NAGOYA UNIVERSITY TEAM, LASEF TEAM — Direct numerical simulations of turbulent planar jets are used to analyze the effects of free stream turbulence on the entrainment characteristics and enstrophy dynamics near the turbulent/turbulent interface (TTI) that separates strong turbulence (inside the jet shear layer) from weaker turbulence outside of the jet. The higher the integral scales and turbulence intensities in the free stream the more effects it has on the jet shear layer, and for strong free stream turbulence the viscous superlayer is absent from the jet edges. 1 Part of this work was supported by JSPS KAKENHI Grant Number 25002531 and MEXT KAKENHI Grant Numbers 25289030, 25289031, 2563005. 8:26AM A28.00003 Flow topology inside the interface layer at the edge of a turbulent jet , RUI JAULINO, RODRIGO TAVEIRA, CARLOS B. DA SILVA, LAETA, IDMEC, Instituto Superior Tecnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal, LASEF TEAM — The invariants of the velocity gradient tensor are analysed in the viscous superlayer and turbulent sublayer within the sharp interface layer that exists at the edges of wakes, jets, mixing layers and boundary layers (turbulent/non-turbulent interface). The invariants display marked differences depending on the characteristics of each one of these layers, and show the imprint of the small scale eddies from the nearby turbulent region. 8:39AM A28.00004 Flow Dynamics Near the Turbulent/Non-Turbulent Interface in Compressible Shear Layers , NAVID S. VAGHEFI, REZA JAHANBAKHSHI, CYRUS K. MADNIA, State University of New York at Buffalo — Direct numerical simulation (DNS) of compressible turbulent shear layers at varying convective Mach numbers are used to assess the flow dynamics in proximity of the turbulent/non-turbulent interface (TNTI) separating the turbulent and the irrotational regions. This interface is identified by using a certain threshold for the vorticity norm. For both incompressible and compressible mixing layers, the TNTI layer thickness is found to be approximately one Taylor length scale. The conditional flow statistics based on the normal distance from the TNTI are compared for different convective Mach numbers. The terms in total kinetic energy, turbulent kinetic energy, and vorticity transport equations are examined in order to determine the effects of compressibility on the transport mechanisms across the TNTI. For all the convective Mach numbers, different terms in these equations are scaled with the Taylor length and velocity scales in interface coordinates. It is observed that for compressible cases, the intense vortical structures (IVS) generate a baroclinic torque as they become close to the TNTI. 8:52AM A28.00005 Deformation of the turbulent/non-turbulent interface by large-scale motions in boundary layers1 , JIN LEE, KAIST, Johns Hopkins University, HYUNG JIN SUNG, KAIST, TAMER A. ZAKI, Johns Hopkins University, Imperial College London — The relationship between large-scale motions (LSMs) and the shape of the turbulent/non-turbulent interface (TNTI) is examined using data from direct numerical simulation (DNS) of turbulent boundary layer (TBL) flow. The Reynolds number based on the momentum thickness and the free-stream velocity ranges from 1180 to 3500. Feature extraction techniques are used to identify cores of the large-scale motions in the perturbation fields. Since turbulence kinetic energy and enstrophy are different inside low- and high-speed LSMs, the wall-normal elevation of TNTI is correlated with the streamwise momentum of these structures. The large-scale crests and troughs of TNTI are matched to the locations of the wall-detached LSMs of low- and high-speed streaks, respectively. In addition, abrupt changes in turbulence statistics across the TNTI reported in previous studies are associated with population trends of the wall-detached LSMs near the TNTI. 1 This work was supported by the Creative Research Initiatives (No. 2014-001493) program of the National Research Foundation of Korea (MSIP). 9:05AM A28.00006 Some Characteristics of Entrainment in a Compressible Turbulent Mixing Layer , REZA JAHANBAKHSHI, NAVID S. VAGHEFI, CYRUS K. MADNIA, State University of New York at Buffalo — The results of direct numerical simulation (DNS) of temporally evolving compressible mixing layer are used to study the entrainment process across the turbulent/non-turbulent interface (TNTI) separating the turbulent and the irrotational regions. This interface is detected by using a certain threshold for the vorticity norm. The compressible form of the conservation equations for mass, momentum, energy, and conserved scalar are solved. The local entrainment velocity is calculated using the enstrophy transport equation. Conditional averages of the terms in this equation across TNTI are examined in order to gain a better understanding of the physical mechanisms contributing to entrainment. The entrainment process in turbulent flows can be associated with two different mechanisms. Nibbling, which is related with small scale motions, and engulfment, which is mostly due to large scale motions. The role of each mechanism is examined. The local entrainment velocity is also decomposed into an inviscid and a viscous part, and the contribution of each part is evaluated. The role of compressibility on the entrainment process is also studied. 9:18AM A28.00007 Local Flow Topology in Compressible Turbulent Shear Layers , CYRUS K. MADNIA, NAVID S. VAGHEFI, State University of New York at Buffalo — The local flow topology is studied using the invariants of the velocity gradient tensor in compressible turbulent mixing layer via direct numerical simulation (DNS). The topological behavior of the flow is analyzed in two different regions: in proximity of the turbulent/non-turbulent interface (TNTI), and inside the turbulent region. The occurrence probability of different flow topologies conditioned by the dilatation level is presented and it is shown that the structures in the locally compressed regions tend to have stable topologies while in locally expanded regions the unstable topologies are prevalent. It is found that the distribution of various flow topologies in regions close to the TNTI differs from inside the turbulent region, and in these regions the most probable topologies are non-focal. At the distances farther than one Taylor microscale from the TNTI, the probability of various topologies is almost constant, and is equal to the values obtained for turbulent region in the mixing layer. 9:31AM A28.00008 On the entrainment dynamics of inergodic, non-stationary flows1 , GIUSEPPE ROSI, DAVID RIVAL, Queen’s University — Entrainment is typically studied through the conditional averaging along the turbulent non-turbulent interface (TNTI) of ergodic flows. However, this method is unsuitable for inergodic, non-stationary flows, as the TNTI is non-similar at different points in space and time. To understand how a TNTI’s mean time dependence effects entrainment, the current study investigates the transport of irrotational fluid into a vortex forming behind an accelerating plate. The plate accelerates to a final velocity within a full-, half- and quarter-chord tow. Phase-averaged, planar, particle tracking velocimetry data is acquired and the forward finite-time Lyapunov exponent and vorticity fields are used to identify the TNTI. The TNTI is then represented by a contour, which is used to approximate the entrainment rate and investigate the transport mechanisms across the TNTI. Early results show that increasing acceleration suppresses vortex growth and entrainment. We hypothesize that shear-layer structure is integral to entrainment by altering the feeding rate of rotational fluid and the TNTI’s convexity. The hypothesis is tested by altering plate-edge geometry and by varying the final chord-based Reynolds number from 5000 to 20 000. 1 Natural Sciences and Engineering Council of Canada 9:44AM A28.00009 Interfacial phenomena in turbulent magnetohydrodynamic channel flows at low magnetic Reynolds number , NAOYA OKAMOTO, YUSUKE OTAKE, TAKASHI ISHIHARA, Nagoya University — Direct numerical simulations (DNS) are performed to examine whether interfacial phenomena can be observed in magnetohydrodynamic (MHD) turbulent channel flows under the influence of imposed magnetic field. The magnetic Reynolds number is assumed to be sufficiently low such that the quasi-static approximation can be applied. For high Hartmann number, the visualization of the vorticity field reveals flow structures consisting of turbulent boundary layers (TBL) near the walls and a quiescent channel core. The statistical analysis of the DNS data shows that physical quantities, such as the spanwise vorticity and streamwise velocity, possess sharp gradients at the edges of the TBL. The features of the sharp gradients are qualitatively similar to those of the turbulent/non-turbulent interfaces which have been observed in hydrodynamic turbulent shear flows, e.g. turbulent boundary layers, wakes and jets. The Joule dissipation rate, which is a characteristic quantity in MHD flow, is shown to have sharp gradients at the edges of the TBL. The average height of the edges is theoretically estimated and the estimation is assessed using the DNS results. Sunday, November 23, 2014 8:00AM - 9:57AM Session A29 Acoustics I: Aeroacoustics — 2014 - Daniel Bodony, University of Illinois at Urbana-Champaign 8:00AM A29.00001 Examining PIV windowing effects on high-speed jet flow physics with POD , MATTHEW BERRY, ANDREW MAGSTADT, ZACHARY BERGER, PATRICK SHEA, Syracuse University, CHRISTOPHER RUSCHER, SIVARAM GOGINENI, Spectral Energies, LLC., MARK GLAUSER, Syracuse University, SYRACUSE UNIVERSITY TEAM, SPECTRAL ENERGIES, LLC. COLLABORATION — The current investigation examines a 2 inch, high-speed, axisymmetric jet with two different PIV setups. Each PIV configuration is simultaneously sampled with far-field pressure. A time-resolved, 10 kHz, PIV system captures a high resolution 1.5 diameter sized window at several downstream locations. A standard, 4 Hz, PIV system utilizes 3 simultaneously captured cameras combined to view a single large interrogation window. Velocity measurements are taken at Mach 0.6, 1.0, and 1.1, along the centerline of the jet in the streamwise (r-z) direction. The low-dimensional modeling technique, proper orthogonal decomposition (POD), is implemented to help resolve the large scale, energetic events, within the flow field. Previous work used these modes to understand how certain flow structures correlated to the far-field acoustics. Due to the different interrogation regions of the PIV systems, windowing effects can yield different results between the setups. We can use this information to determine how windowing effects play a role in the POD convergence rates as well as the velocity to acoustic correlations. 8:13AM A29.00002 Near-field Nonlinear Interactions Leading to Jet Crackle , DAVID BUCHTA, JONATHAN FREUND, University of Illinois at Urbana-Champaign — The noise from high-specific thrust jet exhausts, such as on military jets, is not just particularly intense but also exhibits a peculiar raspy crackling sound. The near acoustic field of this peculiar sound has weak shock-like waves that radiate at a distinct angle with a steepened “footprint” having higher peaks than valleys and thus a positive pressure skewness. We use large-scale direct numerical simulations of free-shear-flow turbulence with Mach numbers ranging from M = 0.9 to 3.5 to study the very near acoustic field and the nonlinear turbulence interactions that lead to this sound. Our simulations reveal that for M ≥ 2.5 sharp, Mach-like waves radiate at a distribution of angles in the near acoustic field. As they propagate, these waves interact nonlinearly. Wave merging increases the length of the correlated waves and decreases the wave density with distance from the turbulence. Locally, the merging waves generate intense pressure fluctuations with elevated pressure skewness, Sk (p′ ) > 0.4, which correlates with the perception of “crackle.” These very-near-field nonlinear interactions may explain the peculiar positive Sk (p′ )—the “footprint” of which has been experimentally observed to propagate approximately linearly to larger distances from the shear layer. 8:26AM A29.00003 Synchronized LES for acoustic near-field analysis of a supersonic jet1 , UNNIKRISHNAN S, DATTA GAITONDE, The Ohio State University, THE OHIO STATE UNIVERSITY TEAM — We develop a novel method using simultaneous, synchronized Large Eddy Simulations (LES) to examine the manner in which the plume of a supersonic jet generates the near acoustic field. Starting from a statistically stationary state, at each time-step, the first LES (Baseline) is used to obtain native perturbations, which are then localized in space, scaled to small values and injected into the second LES (Twin). At any subsequent time, the difference between the two simulations can be processed to discern how disturbances from any particular zone in the jet are modulated and filtered by the non-linear core to form the combined hydrodynamic and acoustic near field and the fully acoustic farfield. Unlike inverse techniques that use correlations between jet turbulence and far-field signals to infer causality, the current forward analysis effectively tags and tracks native perturbations as they are processed by the jet. Results are presented for a Mach 1.3 cold jet. Statistical analysis of the baseline and perturbation boost provides insight into different mechanisms of disturbance propagation, amplification, directivity, generation of intermittent wave-packet like events and the direct and indirect effect of different parts of the jet on the acoustic field. 1 Office of Naval Research 8:39AM A29.00004 Acoustic source analysis of a rectangular supersonic jet1 , JORDAN KREITZMAN, JOSEPH W. NICHOLS, Univ of Minn - Minneapolis — We apply Goldstein’s generalized acoustic analogy to identify acoustic sources in two high-fidelity unstructured large eddy simulation databases of a Mach 1.4 rectangular jet with and without chevrons. Two-point, two-time correlations of the acoustic source terms are evaluated at different positions in the three dimensional flow that develops downstream of the complex nozzle. Two-point statistics are compared to single-point statistics to test the quasi-normality hypothesis and other noise source models for a non-axisymmetric jet. In particular, we assess the predictive capability of a Gaussian model, a fixed-frame model and a modified-distance model. The nozzle geometries used for the simulations exactly match an experimental configuration tested at the NASA Glenn Research Center, allowing for validation in terms of both farfield noise as well as turbulence statistics. 1 We gratefully acknowledge computational resources provided by the Argonne Leadership Computing Facility. 8:52AM A29.00005 Sound amplification by jittering wavepackets in subsonic turbulent jets1 , MENGQI ZHANG, Institut PPRIME - CNRS - Université de Poitiers, AARON TOWNE, California Institute of Technology, PETER JORDAN, Institut PPRIME - CNRS - Université de Poitiers, TIM COLONIUS, California Institute of Technology, GUILLAUME BRÈS, Cascade Technologies, SANJIVA LELE, Stanford University — Recent research confirms that coherent structures in turbulent jets can be understood as hydrodynamic instabilities (wavepackets) of the turbulent mean that amplify and decay as they convect downstream. Linear models used to compute such wavepackets obtain compelling agreement with experiment in terms of both wavepacket structure and phase speed. But the radiated sound can have errors of several orders of magnitude. Data analysis suggests that this is because individual wavepackets evolve, not on the long-time mean of the turbulence, but on a slowly varying mean, which may be described statistically via an ensemble of short-time averages. We use data from a Large Eddy Simulation to explore this idea. The simulation has been carefully validated by an accompanying experiment and found, in particular, to reproduce loud intermittent events observed in the measurements. Slowly varying and short-time-averaged mean flows are extracted from the LES. The Linearised Euler Equations are solved using the slowly varying mean–obtained by low-pass filtering the LES data–as a base flow. The Parabolised Stability and One-Way Euler equations are solved using the short-time ensemble. The solutions comprise jittering wavepackets whose sound radiation is enhanced by several orders of magnitude. 1 This work was supported by the Center for Turbulence Research. 9:05AM A29.00006 Jet noise models using one-way Euler equations , AARON TOWNE, TIM COLONIUS, California Institute of Technology — Experimental and numerical investigations have correlated large-scale coherent structures in turbulent jets with acoustic radiation to downstream angles, where sound is most intense. These structures take the form of wavepackets and can be modeled as linear instability modes of the turbulent mean flow. The parabolized stability equations have been successfully used to estimate the near-field evolution of these wavepackets, but are unable to properly capture the acoustic field. We have recently developed an efficient method for calculating linear instability modes that properly capture both the near-field wavepacket and the associated acoustic field. The linearized Euler equations are modified such that all upstream propagating acoustic modes are removed from the operator. The resulting equations, called one-way Euler equations, can be stably and efficiently solved in the frequency domain as a spatial initial value problem. In this work, we use the one-way Euler equations to model sound generation and propagation in subsonic and supersonic jets. The mean flows are obtained from high resolution large-eddy-simulation (LES) data, and the one-way Euler solutions are validated against direct solution of the linearized Euler equations and compared to the LES data. 9:18AM A29.00007 Correlations Between Large-scale Flow Structures and Acoustic Signatures in an Axisymmetric Jet1 , ANDREW MAGSTADT, MATTHEW BERRY, ZACHARY BERGER, PATRICK SHEA, MARK GLAUSER, Syracuse Univ — In a test campaign studying jet noise, simultaneous far-field acoustic measurements and near-field particle imaging velocimetry (PIV) data were sampled from a supersonic underexpanded axisymmetric jet operating at a Reynolds number of 1.3x10ˆ6. Using overlapping snapshots from three adjacent cameras, separate images of the velocity field were stitched together to form an uninterrupted window. Centered about the axis of the jet, the effective field of view spanned two jet diameters in the cross-stream direction (r) and seven diameters in the streamwise direction (z). This area proved to be sufficiently large to capture important scales of supersonic flow relevant to noise generation. Specifically, Proper Orthogonal Decomposition (POD) has extracted particular energy modes thought to be associated with the large-scale instability wave, shock cells, and turbulent mixing characteristic of supersonic noise. As example, time-dependent modal correlations present evidence linking the existence of shock cells to screech tones. From the data gathered, these experimental and analytical techniques are believed to be valuable tools in isolating energy-based flow structures relevant to noise generation. 1 The authors would like to thank Spectral Energies for their continued support of research at Syracuse University. 9:31AM A29.00008 Simulation and stability analysis of supersonic impinging jet noise with microjet control1 , NATHANIEL HILDEBRAND, JOSEPH W. NICHOLS, Univ of Minn - Minneapolis — A model for an ideally expanded 1.5 Mach turbulent jet impinging on a flat plate using unstructured high-fidelity large eddy simulations (LES) and hydrodynamic stability analysis is presented. Note the LES configuration conforms exactly to experiments performed at the STOVL supersonic jet facility of the Florida Center for Advanced Aero-Propulsion allowing validation against experimental measurements. The LES are repeated for different nozzle-wall separation distances as well as with and without the addition of sixteen microjets positioned uniformly around the nozzle lip. For some nozzle-wall distances, but not all, the microjets result in substantial noise reduction. Observations of substantial noise reduction are associated with a relative absence of large-scale coherent vortices in the jet shear layer. To better understand and predict the effectiveness of microjet noise control, the application of global stability analysis about LES mean fields is used to extract axisymmetric and helical instability modes connected to the complex interplay between the coherent vortices, shocks, and acoustic feedback. 1 We gratefully acknowledge computational resources provided by the Argonne Leadership Computing Facility. 9:44AM A29.00009 A Method for Estimating Far-Field Acoustics Generated by a Turbulent Wall Jet , ADAM NICKELS, LAWRENCE UKEILEY, University of Florida, ROBERT REGER, LOUIS CATTAFESTA, Florida State University — Noise generated via flow interactions within a turbulent wall jet is investigated using a multi-stage estimation method. Two component Particle Image Velocimetry (PIV) and surface pressure measurements are acquired at a Reynolds number based on nozzle height of 25,500. The PIV snapshots and surface pressure measurements were acquired in a synchronous manner to allow for correlation between the quantities. A Proper Orthogonal Decomposition (POD) and Linear Stochastic Estimation (LSE) based technique is used to estimate a set of time-resolved velocity fields at a higher time resolution than was measured. To improve the estimates, the Kalman filter is employed to leverage measurements against a dynamical model established by employing Dynamic Mode Decomposition (DMD) on an independent set of time resolved PIV data. With the estimated time resolved velocity fields and surface pressure measurements, Poisson’s equation for fluctuating pressure is solved to find the pressure acting along the entire surface under the wall jet. The far-field acoustics generated by the wall jet are then estimated by solving Curle’s acoustic analogy, using the integrated surface pressure as the source term. Sunday, November 23, 2014 8:00AM - 9:57AM Session A30 Aerodynamics: Flow Control — 2016 - 8:00AM A30.00001 On the Efficiency of Fluidic Oscillators for Aerofoil Performance Recovery1 , DIMA SARKOROV2 , AVRAHAM SEIFERT3 , Tel Aviv University — The paper describes a recent experiment in which the Suction and Oscillatory blowing (SaOB) actuator was used to control the flow on a thick, turbulent, trailing edge separating aerofoil. The Reynolds number range is 0.5 to 1.5 million. The experiment deals with performance recovery of a thick aerofoil in laminar and turbulent flow conditions. Performance was significantly degraded due to premature boundary layer separation. An array of 12 SaOB actuators was used to effectively restore lift and reduce drag. Overall system efficiency was increased in both turbulent and laminar flow conditions. The AFC outlets, located at 20% chord location were shown to supplement or even being capable of replacing a slat. 1 Support by the EU FP7 is gratefully acknowledged Student 3 Prof. and Head, Meadow Aerolab 2 Graduate 8:13AM A30.00002 Nearfield Flow Topology of a Rounded Wingtip Subject to Circulation Control1 , ADAM EDSTRAND, LOUIS CATTAFESTA, Florida State University — Trailing vortices are an adverse byproduct of lift causing induced drag, accounting for 40% of the total drag on aircraft, and impose a wake hazard on trailing aircraft (Spalart 1998). The metric used to quantify the wake hazard is the average maximum swirl velocity measured in a velocity snapshot. Circulation control uses tangential blowing along a rounded surface, causing the flow to wrap around the surface. This control methodology is extended to a NACA 0012 wingtip by blowing tangentially over a rounded wingtip to control the circulation of the trailing vortex. Stereo particle image velocimetry measurements are acquired along the chord and downstream of the wingtip to characterize the effects of circulation control on vortex formation and evolution. Compared to the baseline case, the vortex core develops along the upper surface of the airfoil further upstream. This upstream development causes more rapid spatial growth of the vortex, resulting with a larger, less intense vortex than the baseline case. However, the circulation, five chords downstream of the leading edge, increases rather than decreases. This increase implies that favorable control of the circulation does not occur. However, there is a 30% reduction in the wake hazard metric due to the increased vortex size. 1 ONR Grant N00014010824 and NSF PIRE Grant OISE-0968313 8:26AM A30.00003 DBD Control of a Turbulent Shear Layer downstream of a Backward Facing Step , JEAN PAUL BONNET, PATRICIA SUJAR-GARRIDO, NICOLAS BENARD, ERIC MOREAU, CNRS University of Poitiers — An open loop control of a turbulent free shear layer downstream of a backward-facing-step at Re 3x104 is performed via a single Dielectric Barrier Discharge (DBD). Several actuation locations are tested, the best result being observed at the hinge of the step. Nanosecond DBD have been tested with no efficiency on the location of the reattachment location. By using AC DBD, a linear evolution of the reattachment is observed. Optimization of frequency, duty cycle and voltage amplitude is performed. An optimal frequency is observed and it is shown that the plasma discharge is able to manipulate the first stages of the formation of the free shear layer and consequently to modify the flow dynamics of the entire flow, with a regularization of the vortex shedding frequency. 8:39AM A30.00004 Closed-loop control of flow separation using instantaneous trajectory patterns1 , ANDREAS SPOHN, VLADIMIR PAREZANOVIĆ, EURIKA KAISER, LAURENT CORDIER, BERND NOACK, Institut PPRIME, Poitiers — A new sensor technique based on visualized instantaneous trajectory patterns is tested to control flow separation. A smooth ramp mounted inside the test section of a water tunnel produces canonical separation conditions. Pulsed hydrogen bubbles furnish instantaneous trajectory patterns of the underlying dynamical system. The evolution of these patterns feeds machine learning algorithms to determine actions that reduce the separated flow region. Compared to periodic forcing the results show even with less actuator action, a major impact on the separated flow. The controlled flow states contain strongly reduced recirculation zones which remain robust even under adverse conditions. Additionally, the visualization of instantaneous trajectory patterns is shown to have some promising options: The Lagrangian coherent structures (LCS) of the controlled dynamical system can be deduced in-time without determination and integration of the instantaneous velocity fields. Additionally, classical procedures to reduce the data dimensionality, as for example the principal component analysis (PCA) and its variants, can be applied directly to the visualizations in order to feed the controller. 1 Funding of the ANR program SepaCoDe and the ANR Chair of Excellence TUCOROM is gratefully acknowledged. 8:52AM A30.00005 The effect of large aspect ratio wing yaw on active separation control1 , PHILIPP TEWES2 , LUTZ TAUBERT3 , ISRAEL WYGNANSKI4 , The University of Arizona — The applicability of the boundary layer independence principle to turbulent boundary layers developing on infinitely yawed wings, suggested that active separation control might be carried out differently to the two presumably independent developing boundary layers. At low incidence or flap deflection the control of the spanwise component of the flow is effective provided the aggregate number of actuators is small. In this case the actuator jets provide jet-curtains that virtually eliminate the spanwise flow component of the flow in their vicinity. At higher incidence or flap deflection, the focus of the active separation control has to shift to the chordwise component that has to overcome a high adverse pressure gradient. The idea was proven experimentally on a flapped wing based on a NACA 0012 airfoil that could be swept back and forward while being suspended from a ceiling of a wind tunnel connected to a six-component balance. The experiments were carried out at Reynolds numbers varying between 300,000 and 500,000. 1 The project was supported in part by a grant from AFOSR Associate 3 Research Assistant Professor 4 Professor 2 Research 9:05AM A30.00006 Experimental measurements on a single sweeping jet1 , DAMIAN HIRSCH, EMILIO GRAFF, MORTEZA GHARIB, California Institute of Technology — “Sweeping jets” proved their effectiveness as Active Flow Control (AFC) actuators in improving the performance of vertical tails of generic and full-scale models. To gain further knowledge about the fundamental flow physics, the jets were investigated experimentally. The influence of a single jet on its surroundings was studied, especially the entrainment region. The results were compared to previous experiments to study the difference between a single isolated jet and multiple jets mounted on a vertical tail. 1 Supported by the Boeing Company. 9:18AM A30.00007 The role of the vorticity field on the increase of drag forces during impulsive deployment of a rectangular flow actuator inside TBL , AMIR ELZAWAWY, Vaughn College of Aeronautics and Technology, YIANNIS ANDREOPOULOS , City College of New York — An experimental Time Resolved PIV data is used to evaluate the significant role of the vorticity field on aerodynamic forces during an impulsive deployment of a 100x100 mm2 flow actuator. In this experiment, the flow actuator is placed inside TBL flow, while it is suddenly deployed, ω = 17 rad/s, against the incoming TBL flow with free stream air velocity of 3.7 m/s. The experiments data has shown a significant increase of drag forces during the impulsive deployment compared with the drag of those stationary actuator cases. In this work, a further analysis is carried out using vorticity moments-based relations of forces of finite bodies exerted by incompressible fluid flows (Wu et al. 2006). These formulations, which shown to be suitable for use with TR-PIV data, are used here to identify the role of the generated vorticity field on the increase of the drag. Only three terms out of the seven terms showed significant contribution to the drag forces enhancement. Two of those were dependent on the vorticity field; the first term represents volume integral of rate of change of the first moment of the vorticity. The second term, which showed a negative contribution to the increase of drag, was the volume integral of the Lamb force. The third term represents the inertia effects at the accelerating boundaries. This identification of the role of each term can provide a basic understanding to the role of the vorticity field and may help in flow actuator design process to obtain enhanced aerodynamics forces with impulsive motion. 9:31AM A30.00008 Fluidic Control of Flexible Structures Embedded in a Turbulent Boundary Layer , ORI FRIEDLAND1 , VICTOR TROSHIN, TAU, AVI SEIFERT, School of mech. Eng., Tel Aviv University — We investigate experimentally the flow around a flexible rectangular thin plate positioned normal to the wind direction and embedded in a thick turbulent boundary layer. The purpose of the study is to reduce the plate oscillations caused by unsteady wind loads. Two methods were tested. First, by mechanical Piezo-electric actuators attached to the plate. Second, by three mass-less Piezo-electric fluidic actuators. The two methods were applied with similar closed-loop control principles: Strain Gauge (SG) sensors captured the plate oscillations and a simple phase-lag and gain was used to attenuate the oscillations. The results show a 20-30% reduction of the plate oscillations by mechanical control and a 30%-40% attenuation of the plate oscillation, compared to the uncontrolled case, using fluidic actuators positioned around the free-end flow separation points. The fluidic control was found to be superior to the mechanical control for the current application and conditions. We Hypothesize flow physics mechanism that link the unsteady pressures created on the plate by actuation to its oscillations. 1 Graduate Student 9:44AM A30.00009 Hybrid flow control of a transport truck side-mirror using AC-DBD plasma actuated guide vane , THEODOROS MICHELIS, MARIOS KOTSONIS, Delft Univ of Tech — A wind-tunnel study is conducted towards hybrid flow control of a full-scale transport truck side-mirror (Re = 4 × 105 ). The mirror is mounted on a structure that models the truck cabin. PIV measurements are performed at a range of velocities from 15 to 25m/s and from leeward to windward angles of −5◦ to +5◦ . A slim guide vane of 6cm chord is employed along the span of the hub of the mirror for redirecting high momentum flow towards the wake region. Separation from the leading edge of the guide vane is reduced or eliminated by means of AC-DBD plasma actuator, operating at voltage of 35kV peak-to-peak and frequency of 200Hz. Time-averaged velocity fields are obtained at the centre of the mirror for three scenarios: a) reference case lacking any control elements; b) guide vane only and c) combination of the guide vane and the AC-DBD. The comparison of cases demonstrates that at 25m/s windward conditions (−5◦ ) the guide vane is capable of increasing momentum (+20%) in the wake of the mirror with additional improvement when plasma actuation is applied (+21%). In contrast, at leeward conditions (+5◦ ), the guide vane reduces momentum (-20%), though with actuation an increase is observed (+5%). Total recovered momentum is 25%. Sunday, November 23, 2014 8:00AM - 9:57AM Session A31 CFD: Particle and Immersed Boundary Methods Carolina — 2018 - Boyce Griffith, University of North 8:00AM A31.00001 Towards numerical consistency and conservation for SPH approximations , NIKOLAUS ADAMS, XIANGYU HU, SERGEJ LITVINOV, Technische Universität München — Typical conservative Smoothed particle hydrodynamics (SPH) approximations introduce two errors: smoothing error is due to smoothing of the gradient by an integration associated with a kernel function; integration error due to approximating of the integration by summation over all particles within the kernel support. The integration error leads to violation of zero-order consistency, i.e., the inability to reproduce a constant field. We show that partition of unity is the condition under which the conservative SPH approximation achieves both consistency and convergence. The condition can be met by relaxing a particle distribution under a constant pressure field and invariant particle volume. The resulting particle distribution is very similar to those observed for liquid molecules. We further show that with two different typical kernel functions the SPH approximation satisfying the partition of unity property is able to achieve very high-order of the integration error for random particle locations. The background pressure used in a weakly compressible SPH simulation implies a self-relaxation mechanism, which explains that convergence with respect to increasing particle numbers could be obtained in SPH simulations, although not predicted by previous numerical analysis. Furthermore, by relating the integration error to the background pressure, we explain why the previously proposed transport-velocity formulation of SPH is able to achieve unprecedented accuracy and stability. 8:13AM A31.00002 Developing a weakly compressible smoothed particle hydrodynamics model for biological flows1 , YAROSLAV VASYLIV, ALEXANDER ALEXEEV, Georgia Institute of Technology — Smoothed Particle Hydrodynamics (SPH) is a meshless particle method originally developed for astrophysics applications in 1977. Over the years, limitations of the original formulations have been addressed by different groups to extend the domain of SPH application. In biologically relevant internal flows, two of the several challenges still facing SPH are 1) treatment of inlet, outlet, and no slip boundary conditions and 2) treatment of second derivatives present in the viscous terms. In this work, we develop a 2D weakly compressible SPH (WCSPH) for simulating viscous internal flows which incorporates some of the recent advancements made by groups in the above two areas. The method is validated against several analytical and experimental benchmark solutions for both steady and unsteady laminar flows. In particular, the 2013 U.S. Food and Drug Administration benchmark test case for medical devices – steady forward flow through a nozzle with a sudden contraction and conical diffuser – is simulated for different Reynolds numbers in the laminar region and results are validated against the published experimental and CFD datasets. 1 Support from the National Science Foundation Graduate Research Fellowship Program (NSF GRFP) is gratefully acknowledged. 8:26AM A31.00003 Momentum exchange in multiphase dispersed flow: a statistical estimator based on multilevel Monte Carlo , MATTEO ICARDI, King Abdullah University of Science and Technology — Eulerian-Eulerian models for multiphase dispersed flow are commonly derived by means of ensemble (or spatial) averaging. They are therefore based on quantities defined over statistical (or spatial) ensemble of particle configurations. However momentum exchange correlations (e.g., drag, lift) are known (and can be defined deterministically) only for the dilute (isolated spheres) and dense (Ergun) limits. Furthermore it is well known that the overall results are often very sensitive to the correlation chosen and to the closure approximations for fluctuations, strongly limiting the predictive capability of the models. In this work the forces acting on random array of spheres and other granular objects have been studied with a novel statistical approach based on multilevel Monte Carlo. Direct Numerical Simulations are used to resolve the flow around the spheres and both the numerical and statistical error are controlled accurately. Mean and fluctuations of the momentum exchange terms can be characterized to derive new correlations for drag and lift in dense poly-dispersed flows that are statistically robust. 8:39AM A31.00004 A Particle-Particle Collision Model for Smoothed Profile Method , FAZLOLAH MOHAGHEGH, JOHN MOUSEL, H.S. UDAYKUMAR, University of Iowa — Smoothed Profile Method (SPM) is a type of continuous forcing approach that adds the particles to the fluid using a forcing. The fluid-structure interaction is through a diffuse interface which avoids sudden transition from solid to fluid. The SPM simulation as a monolithic approach uses an indicator function field in the whole domain based on the distance from each particle’s boundary where the possible particle-particle interaction can occur. A soft sphere potential based on the indicator function field has been defined to add an artificial pressure to the flow pressure in the potential overlapping regions. Thus, a repulsion force is obtained to avoid overlapping. Study of two particles which impulsively start moving in an initially uniform flow shows that the particle in the wake of the other one will have less acceleration leading to frequent collisions. Various Reynolds numbers and initial distances have been chosen to test the robustness of the method. Study of Drafting-Kissing Tumbling of two cylindrical particles shows a deviation from the benchmarks due to lack of rotation modeling. The method is shown to be accurate enough for simulating particle-particle collision and can easily be extended for particle-wall modeling and for non-spherical particles. 8:52AM A31.00005 An Immersed Boundary Method for Rigid Bodies , AMNEET PAL SINGH BHALLA, BAKYTZHAN KALLEMOV, ALEKSANDAR DONEV, Courant Institute of Mathematical Sciences, New York University, BOYCE GRIFFITH, University of North Carolina at Chapel Hill — The traditional immersed boundary (IB) method is a very flexible method for coupling elastic structures to fluid flow. When rigid bodies are modeled using an IB approach, a penalty method is usually employed to approximately enforce the rigidity of the body; this requires small time step sizes and leads to difficult-to-control errors in the solution. We develop a method that exactly enforces a rigidity constraint by solving a linear system coupling a standard semi-implicit discretization of the fluid equations with a rigidity constraint. We develop a preconditioned iterative solver that combines an approximate multigrid solver for the fluid problem with an approximate direct solver for the Schur complement system. We demonstrate the efficiency and study the accuracy of the method on several test problems for both zero and finite Reynolds numbers. 9:05AM A31.00006 An immersed interface vortex particle-mesh solver , YVES MARICHAL1 , PHILIPPE CHATELAIN, GREGOIRE WINCKELMANS, Universite catholique de Louvain — An immersed interface-enabled vortex particle-mesh (VPM) solver is presented for the simulation of 2-D incompressible viscous flows, in the framework of external aerodynamics. Considering the simulation of free vortical flows, such as wakes and jets, vortex particle-mesh methods already provide a valuable alternative to standard CFD methods, thanks to the interesting numerical properties arising from its Lagrangian nature. Yet, accounting for solid bodies remains challenging, despite the extensive research efforts that have been made for several decades. The present immersed interface approach aims at improving the consistency and the accuracy of one very common technique (based on Lighthill’s model) for the enforcement of the no-slip condition at the wall in vortex methods. Targeting a sharp treatment of the wall calls for substantial modifications at all computational levels of the VPM solver. More specifically, the solution of the underlying Poisson equation, the computation of the diffusion term and the particle-mesh interpolation are adapted accordingly and the spatial accuracy is assessed. The immersed interface VPM solver is subsequently validated on the simulation of some challenging impulsively started flows, such as the flow past a cylinder and that past an airfoil. 1 Research Fellow (PhD student) of the F.R.S.-FNRS of Belgium 9:18AM A31.00007 An efficient immersed boundary projection method for flow around moving bodies1 , WEI-XI HUANG, RU-YANG LI, CHUN-MEI XIE, CHUN-XIAO XU, Tsinghua University — An immersed boundary method based on the projection approach is proposed for simulation of flow over moving bodies. In this framework, the momentum forcing added to the incompressible Navier-Stokes equations acts as a Lagrangian multiplier to satisfy the no-slip condition on the immersed boundary, as the role of the pressure on enforcing the divergence-free constraint. The fractional step method with a fully implicit time-advancement scheme is adopted to compute the system, thus eliminating the CFL limitation. Based on the approximate block LU decomposition, velocity-pressure-momentum forcing decoupling is achieved. Moreover, decoupling of the intermediate velocity components and further decoupling of the three directions of the Cartesian coordinates for each velocity component are also performed. As a result, tridiagonal matrix systems for the intermediate velocity, the pressure Poisson equation, and a linear system for the momentum forcing which is one-order lower than the fluid dimensions, are solved, resulting in a significant saving of the computation cost. Both the temporal and spatial accuracies of the proposed method are tested. For validation, several benchmark numerical examples are presented, including flow over a stationary/oscillating circular cylinder and flow around a flapping wing. 1 The work was supported by National Natural Science Foundation of China (Grant number 11322221). 9:31AM A31.00008 A Newton-Krylov method with approximate Jacobian for implicit solution of Navier-Stokes on staggered overset-curvilinear grids with immersed boundaries1 , HAFEZ ASGHARZADEH, IMAN BORAZJANI, University at Buffalo SUNY — Time step-size restrictions and low convergence rates are major bottle necks for implicit solution of the Navier-Stokes in simulations involving complex geometries with moving boundaries. Newton-Krylov method (NKM) is a combination of a Newton-type method for super-linearly convergent solution of nonlinear equations and Krylov subspace methods for solving the Newton correction equations, which can theoretically address both bottle necks. The efficiency of this method vastly depends on the Jacobian forming scheme e.g. automatic differentiation is very expensive and Jacobian-free methods slow down as the mesh is refined. A novel, computationally efficient analytical Jacobian for NKM was developed to solve unsteady incompressible Navier-Stokes momentum equations on staggered curvilinear grids with immersed boundaries. The NKM was validated and verified against Taylor-Green vortex and pulsatile flow in a 90 degree bend and efficiently handles complex geometries such as an intracranial aneurysm with multiple overset grids, pulsatile inlet flow and immersed boundaries. The NKM method is shown to be more efficient than the semi-implicit Runge-Kutta methods and Jabobian-free Newton-Krylov methods. We believe NKM can be applied to many CFD techniques to decrease the computational cost. 1 This work was supported partly by the NIH grant R03EB014860, and the computational resources were partly provided by Center for Computational Research (CCR) at University at Buffalo. 9:44AM A31.00009 An immersed boundary method for imposing solid wall conditions in lattice Boltzmann solvers for single- and multi-component fluid flows , ZHE LI, JULIEN FAVIER, UMBERTO D’ORTONA, Laboratoire M2P2, UMR 7340 CNRS/Aix-Marseille Université, SÉBASTIEN PONCET, Département de génie mécanique, Faculté de génie, Université de Sherbrooke, — In this work, one proposes an immersed boundary-lattice Boltzmann coupled algorithm to solve single- and multi-component fluid flows, in the presence of fixed or moving solid boundaries. The prescribed motion of immersed boundaries is imposed by adding a body force term in the lattice Boltzmann model, which is obtained from the macroscopic fluid velocity definition interpolated at the Lagrangian solid points. Numerical validation test cases show that the proposed numerical solver is second-order accurate. Furthermore, the Shan-Chen’s lattice Boltzmann model is applied for multi-component fluid flows, and a special focus is given to the treatment of different wetting properties of fixed walls. The capability of the new numerical solver is finally evaluated by simulating a cluster of moving cilia in a two-component fluid flow. Sunday, November 23, 2014 8:00AM - 9:57AM Session A32 Particle-Laden Flows: General I — 2020 - Federico Toschi, Technische Universiteit Eindhoven 8:00AM A32.00001 Preferential accumulation and enhanced relative velocity of inertial droplet due to interactions with homogeneous isotropic turbulence , COLIN BATESON, ALBERTO ALISEDA, Univ of Washington — We present results from wind tunnel experiments on the evolution of small inertial (d ≈ 10 − 200 µm) water droplets in homogeneous, isotropic, slowly decaying grid turbulence. High-speed imaging and Particle Tracking Velocimetry are used to calculate relative velocity distributions. We analyze the preferential concentration and enhanced relative velocity of droplets resulting from their inertial interactions with the underlying turbulence. The two-dimensional particle velocity, measured from PTV and long-time tracks along a streamwise plane, are conditionally analyzed with respect to the distance to the nearest particle. We focus on the non-normality of the statistics for the particle-particle separation velocity component to dissect the influence of the inertial interaction with the turbulence on the dynamics of the droplets. We observe a negative bias (in the mean and mode) in the separation velocity of particles for short separations, signaling a tendency of particles to collide more frequently than a random agitation by turbulence would predict. The 2-D Radial Distribution Function is also analyzed and compared to previous 1-D results. 8:13AM A32.00002 Clustering of vertically constrained passive particles in homogeneous and isotropic turbulence1 , MICHEL VAN HINSBERG, Eindhoven University of Technology, MASSIMO DE PIETRO, LUCA BIFERALE, University of Rome Tor Vergata, HERMAN CLERCX, FEDERICO TOSCHI, Eindhoven University of Technology — We analyze the dynamics of small particles confined within a horizontal fluid slab in a three-dimensional (3D) homogenous isotropic turbulent velocity field. Particles can freely move horizontally as fluid tracers but are vertically confined around a given horizontal plane via a simple linear restoring force. The present model may be considered as the simplest description for the dynamics of small aquatic organisms that, due to swimming, active regulation of their buoyancy or other mechanisms, are capable to maintain themselves in a shallow horizontal layer somewhere below the free surface of oceans or lakes. In the model varying the strength of the restoring force can control the thickness of the fluid slab in which the particles can move. Whenever some confinement is present, particle trajectories deviate from fluid tracers and experience an effectively compressible velocity field. We report a quantification of this effective compressibility as well as a quantification of preferential concentration of tracer particles in terms of the correlation dimension. We found that there exists a particular value of the force constant, corresponding to a mean slab depth approximately equal to a few times the Kolmogorov length scale, that maximizes the clustering of the particles. 1 This work is part of the research programmes 11PR2841 and FP112 of the Foundation for Fundamental Research on Matter (FOM), which is part of the Netherlands Organisation for Scientific Research (NWO). The work was partially funded by ERC Grant No 339032. 8:26AM A32.00003 Quantitative prediction of clustering instabilities in gas-solid homogeneous cooling systems , CHRISTINE HRENYA, Univeristy of Colorado, PETER MITRANO, University of Colorado, XIAOQI LI, XIAOLONG YIN, Colorado School of Mines — Dynamic particle clusters are widely documented in gas-solid flow systems, including gasification units for coal or biomass, gravity-driven flow over an array of tubes, pneumatic transport lines, etc. Continuum descriptions based on kinetic theory have been known for over a decade to qualitatively predict the presence of such clustering instabilities. The quantitative ability of such continuum descriptions is relatively unexplored, however, and remains unclear given the low-Knudsen assumption upon which the descriptions are based. In particular, the concentration gradient is relatively large across the boundary between the cluster and the surrounding dilute region, which is counter to the small-gradient assumption inherent in the low-Knudsen-number expansion. In this work, we use direct numerical simulations (DNS) of a gas-solid homogeneous cooling system to determine the critical system size needed for the clustering instability to develop. We then compare the results to the same quantity predicted by a continuum description based on kinetic theory. The agreement is quite good over a wide range of parameters. This finding is reminiscent of molecular fluids, namely the ability of the Navier-Stokes equations to predict well outside the expected range of Knudsen numbers. 8:39AM A32.00004 Inertial Particle Caustics around a vortex , RAVICHANDRAN SIVARAMAKRISHNAN, RAMA GOVINDARAJAN, TIFR Centre for Interdisciplinary Sciences, TIFR Hyderabad — Caustics are formed when particles with different velocities end up in the same location in space. Inertial particle caustics are thought to be responsible for the rapid onset of rainfall in vigorously convecting clouds. It is generally also believed that caustics are effective only at large Stokes numbers. We study inertial particles caustics in the canonical flow of a single vortex. We also show what kinds of vortices can have caustics around them. For point vortices, we show the existence of critical Stokes numbers below which there can be no caustics. We find that the value of the critical Stokes number depends on the initial conditions chosen for the particles. The existence of a critical Stokes number translates to the existence of definite regions around vortices where particles have to start in order to form caustics. We use this fact to perform simulations with many vortices in a periodic box. We use the lagrangian tracking technique of Osiptsov to track the densities associated with the particles. These simulations show that the effect of caustics on particle clustering is more complicated than is generally believed. While larger particle inertia leads to a larger number of very “dense” particles, we find that smaller inertia can lead to higher average densities. 8:52AM A32.00005 Backward transformation of the colored-noise Fokker-Planck equation for shear-induced diffusion processes of non-Brownian particles1 , LAURA LUKASSEN, MARTIN OBERLACK, Chair of Fluid Dynamics / Graduate School of Excellence Computational Engineering, TU Darmstadt — As described in literature, non-Brownian particles in shear flow show a diffusive behavior due to hydrodynamic interactions. This shear-induced diffusion differs from the well-known Brownian diffusion, as there is no separation of time scales. That means that the configuration of non-Brownian particles changes on the same time scale as the hydrodynamic velocity. This fact impedes the derivation of a Fokker-Planck equation describing non-Brownian particles in pure position space. In this context, we derived a new Fokker-Planck approach in coupled position-velocity space to assure the validity of the Markov process assumption which is violated in pure position space formulation (Lukassen, Oberlack, Phys. Rev. E 89, 2014). Here, we present a further validation of our new Fokker-Planck approach that allows us to establish a relation to a modified purely position space Fokker-Planck equation. This backward transformation exhibits additional correction terms when compared to other position space Fokker-Planck equations in that context known from literature. Our extended approach shall enable a better stochastic description of non-Brownian particle flows. 1 The work of L. Lukassen is supported by the “Excellence Initiative” of the German Federal and State Governments and the Graduate School of Computational Engineering at TU Darmstadt. 9:05AM A32.00006 A Stochastic Model for the Relative Motion of High Stokes Number Particles in Isotropic Turbulence , ROHIT DHARIWAL, SARMA RANI, Univ of Alabama - Huntsville, DONALD KOCH, Cornell University — In the current study, a novel analytical closure for the diffusion current in the PDF equation is presented that is applicable to high-inertia particle pairs with Stokes numbers Str ≫ 1. Here Str is a Stokes number based on the time-scale τr of eddies whose size scales with pair separation r. Using this closure, Langevin equations were solved to evolve particle-pair relative velocities and separations in stationary isotropic turbulence. The Langevin equation approach enables the simulation of the full PDF of pair relative motion, instead of only the first few moments of the PDF as is the case in a moments-based approach. Accordingly, PDFs Ω(U |r) and Ω(Ur |r) are computed for various separations r, where the former is the PDF of relative velocity U and the latter is the PDF of the radial component of relative velocity Ur , both conditioned upon the separation r. Consistent with the DNS study of Sundaram & Collins, the Langevin simulations capture the transition of Ω(U |r) from being Gaussian at integral-scale separations to an exponential PDF at Kolmogorov-scale separations. The radial distribution functions (RDFs) computed from these simulations also show reasonable quantitative agreement with those from the DNS of Fevrier et al. 9:18AM A32.00007 Measurements of polystyrene bead trajectories and spatial distributions in a turbulent water flow, square duct using high-speed digital holography1 , RENE VAN HOUT, BORIS RABENCOV, JAVIER ARCA, Technion - Israel Institute of Technology — Near neutrally buoyant, polystyrene beads (583 micrometers) were tracked in a square (50x50mm2 ), closed-loop, turbulent water duct at a bulk flow Reynolds number of 10,602 (friction velocity 0.0208m/s) using single view, inline digital holographic cinematography (at 1 kHz). The volume of interest (50x17.4x17.4mm3 ) was positioned at the bottom part of the channel. The mean bead diameter normalized by inner wall coordinates was d+ =14.2, with Stokes numbers of 8.5. In-house developed algorithms, fine-tuned to tracking single and overlapping beads were developed. Bead in-focus positions were determined by maximum intensity gradient method. Results showed that in agreement with literature publications, ascending beads lagged the mean streamwise water velocity while descending ones had similar velocities. Average streamwise bead velocities and number densities collapsed onto wall-normal-streamwise and spanwise-streamwise planes, indicated preferential segregation of ascending and descending beads up to a height of 100 wall units. Spanwise “lane” separation distances ranged between 150-200 wall units, larger but of the same order as the spanwise extent of coherent near-wall turbulence structures. Duct corners were nearly devoid of beads likely caused by secondary flows. 1 Israel Science Foundation grant 915/10 and COST Actions MP0806 and FP1005 9:31AM A32.00008 Scaling of sediment erosion rates for unsteady, nonequilibrium flows1 , KEN KIGER, KYLE CORFMAN, RAHUL MULINTI, University of Maryland — Traditional approaches to sediment transport are typically based on assumptions of fully developed, equilibrium conditions. Many flows, however, are dominated by the presence of intermittent, coherent large-scale structures, which may not satisfy such assumptions. The current study examines the erosion rates produced by a strongly forced air jet impinging on a mobile bed of fine glass beads. A parametric study is conducted using a range of mean flow and forcing conditions to elucidate the role of the dominant structure on the transport process. The evolution of the bed surface is compared to single-phase PIV measurements. The results show that the use of the time-averaged stress on the bed cannot be used to effectively predict the location or magnitude of erosion. Instead, it is shown that the erosion rate can be related to the location and magnitude of the periodic stress produced by the vortex interacting with the bed. After an initial transient, the erosion for all cases was observed to proceed at a relatively constant rate. The net removal rate was found to correlate closely to the unsteady stress produced by the periodic structures, and predicted the erosion rate for all of the conditions studied when scaled by the integral of the excess wall stress raised to the power 1.2. 1 This work is supported by the AFSOR under grant FA95500810406 9:44AM A32.00009 Reducing thermophoretic deposition in heat exchangers using wavy walled channels1 , ZACHARY MILLS, ALEXANDER ALEXEEV, Georgia Institute of Technology — Using computational simulations, we examined the effect of wavy walled geometries on the fouling of heat exchangers. Our model combines a lattice Boltzmann model for simulating the fluid flow, a finite difference temperature model and a Brownian dynamics model used to model the transport and deposition of aerosol particles. In our previous studies, we investigated how the geometry influences the structure of the flow within the channel. Specifically, we determined the critical pressure gradients at which the flow transitions between different flow regimes for various wave amplitudes and periods. We observed three separate flow regimes including steady flow with and without circulation and unsteady time-periodic flow. We have extended this investigation to examine the effects of these different geometries and flow regimes on heat and mass transport within the channel. In our simulations we investigated particle deposition resulting from convection and thermophoresis. From the results of our investigations we will be able to determine the geometries which reduce the rate of fouling in heat exchangers while increasing heat transport. 1 This work is supported by General Motors Corporation. Sunday, November 23, 2014 8:00AM - 9:44AM Session A33 Atomization, Breakup and Interfacial Flows — 2022 - Luis Bravo, United States Army Research Laboratory 8:00AM A33.00001 Direct Numerical Simulation of Liquid Nozzle Spray with Comparison to Shadowgraphy and X-Ray Computed Tomography Experimental Results , BRET VAN POPPEL, US Military Acad, MARK OWKES, Montana State University, THOMAS NELSON, ZACHARY LEE, TYLER SOWELL, MICHAEL BENSON, US Military Acad, PABLO VASQUEZ GUZMAN, REBECCA FAHRIG, JOHN EATON, Stanford University, MATTHEW KURMAN, CHOL-BUM KWEON, LUIS BRAVO, US Army Research Laboratory — In this work, we present high-fidelity Computational Fluid Dynamics (CFD) results of liquid fuel injection from a pressure-swirl atomizer and compare the simulations to experimental results obtained using both shadowgraphy and phase-averaged X-ray computed tomography (CT) scans. The CFD and experimental results focus on the dense near-nozzle region to identify the dominant mechanisms of breakup during primary atomization. Simulations are performed using the NGA code of Desjardins et al (JCP 227 (2008)) and employ the volume of fluid (VOF) method proposed by Owkes and Desjardins (JCP 270 (2013)), a second order accurate, un-split, conservative, three-dimensional VOF scheme providing second order density fluxes and capable of robust and accurate high density ratio simulations. Qualitative features and quantitative statistics are assessed and compared for the simulation and experimental results, including the onset of atomization, spray cone angle, and drop size and distribution. 8:13AM A33.00002 High Fidelity Simulation of Primary Atomization in Diesel Engine Sprays1 , CHRISTOPHER IVEY, Stanford University, LUIS BRAVO, U.S. Army Research Laboratory, DOKYUN KIM, Cascade Technologies — A high-fidelity numerical simulation of jet breakup and spray formation from a complex diesel fuel injector at ambient conditions has been performed. A full understanding of the primary atomization process in fuel injection of diesel has not been achieved for several reasons including the difficulties accessing the optically dense region. Due to the recent advances in numerical methods and computing resources, high fidelity simulations of atomizing flows are becoming available to provide new insights of the process. In the present study, an unstructured un-split Volume-of-Fluid (VoF) method coupled to a stochastic Lagrangian spray model is employed to simulate the atomization process. A common rail fuel injector is simulated by using a nozzle geometry available through the Engine Combustion Network. The working conditions correspond to a single orifice (90 µm) JP-8 fueled injector operating at an injection pressure of 90 bar, ambient condition at 29 bar, 300K filled with 100% nitrogen with Rel = 16,071, W el = 75,334 setting the spray in the full atomization mode. The experimental dataset from Army Research Lab is used for validation in terms of spray global parameters and local droplet distributions. The quantitative comparison will be presented and discussed. 1 Supported by Oak Ridge Associated Universities and the Army Research Laboratory 8:26AM A33.00003 Numerical simulation of evaporating liquid jet in crossflow , MARIOS SOTERIOU, XIAOYI LI, United Technologies Research Center — Atomization of liquid fuel jets by cross-flowing air is critical to combustor performance. Ability to experimentally probe the fundamentals of this multiscale two phase flows has been hampered by limitations in experimental techniques and the challenges posed by operating conditions. Direct numerical simulation has recently emerged as a promising alternative due to advances in computer hardware and numerical methods. Using this approach, we recently demonstrated the ability to reproduce the physics of atomization of a liquid jet in cross-flow (LJIC) under ambient conditions. In this work we consider this flow in a high temperature environment. The inclusion of evaporation is the major new element. The numerical approach employs the CLSVOF method to capture the liquid-gas interface. Interface evaporation is solved directly with proper treatment of interface conditions and reproduces the relevant species/temperature fields there. A Lagrangian droplet tracking approach is used for the small droplets which are transferred from the Eulerian phase and evaporate using a traditional d2 law model. Other key algorithms of the massively parallelized solver include a ghost fluid method, a multi-grid preconditioned conjugate gradient approach and an adaptive mesh refinement technique. The overall method is verified using canonical problems. Simulations of evaporating LJIC point to the significant effect that evaporation has on the evolution of this flow and elucidate the downstream fuel species patterns. 8:39AM A33.00004 A Computational Study of an Atomizing Liquid Sheet , SURAJ DESHPANDE, MARIO TRUJILLO, University of Wisconsin - Madison — Atomization of a liquid sheet is studied using simulations based on a volume of fluid (VoF) method. Our aim is to evaluate the primary atomization models which are often used in Lagrangian-Eulerian simulations, a prominent spray simulation method. The models assume that growth of sinuous unstable waves on the sheet causes its breakup and use linear theory to predict the wavelength [Dombrowski & Johns 1963; Senecal et al. 1999]. With respect to this, we address two points: (1) applicability of linear theory to instability prediction, and (2) relevance of this prediction to sheet breakup. To this end, a more general linear analysis considering capillary, viscous and boundary layer is performed using Orr-Sommerfeld (OS) theory. Our VoF simulations show that instability mechanism does selectively amplify indistinct noise into discernible interfacial waves, which are very well predicted by OS analysis. These waves, however, do not cause sheet breakup, and this contrasts prior linear theories. The structures which eventually do lead to breakup are shown to be practically independent of viscous and surface tension effects (unlike the linear waves). They scale with sheet thickness, and are ∼ O(100) times larger than predicted by linear theories. 8:52AM A33.00005 Numerical simulation of liquid-layer breakup on a moving wall due to an impinging jet1 , TAEJONG YU, HOJOON MOON, DONGHYUN YOU, POSTECH, DOKYUN KIM, ANDREY OVSYANNIKOV, Center for Turbulence Research — Jet wiping, which is a hydrodynamic method for controlling the liquid film thickness in coating processes, is constrained by a rather violent film instability called splashing. The instability is characterized by the ejection of droplets from the runback flow and results in an explosion of the film. The splashing phenomenon degrades the final coating quality. In the present research, a volume-of-fluid (VOF)-based method, which is developed at Cascade Technologies, is employed to simulate the air-liquid multiphase flow dynamics. The present numerical method is based on an unstructured-grid unsplit geometric VOF scheme and guarantees strict conservation of mass of two-phase flow, The simulation results are compared with experimental measurements such as the liquid-film thickness before and after the jet wiping, wall pressure and shear stress distributions. The trajectories of liquid droplets due to the fluid motion entrained by the gas-jet operation, are also qualitatively compared with experimental visualization. Physical phenomena observed during the liquid-layer breakup due to an impinging jet is characterized in order to develop ideas for controlling the liquid-layer instability and resulting splash generation and propagation. 1 Supported by the Grant NRF-2012R1A1A2003699, the Brain Korea 21+ program, POSCO, and 2014 CTR Summer Program 9:05AM A33.00006 Experimental investigation of two oil dispersion pathways by breaking waves1 , CHENG LI, JOSEPH KATZ, Johns Hopkins University — This experimental study focuses on generation and size distribution of airborne and subsurface oil droplets as breaking surface waves interact with a crude oil slick (MC252 surrogate). Experiments in a specialized wave tank investigate the effects of wave height and wave properties (e.g. spilling vs. plunging), as well as drastically reducing the oil-water interfacial tension by orders of magnitude by introducing dispersant (Coexist 9500-A). This dispersant is applied at varying dispersant-to-oil ratios either by premixing or surface spraying, the latter consistent with typical application. The data include high-speed visualizations of processes affecting the entrainment of subsurface oil and bubbles as well as airborne aerosols. High-speed digital holographic cinematography is employed to track the droplet trajectories, and quantify the droplet size distributions above and below the surface. Introduction of dispersants drastically reduces the size of subsurface droplets to micron and even submicron levels. Ahead of the wave, the 25 µm (our present resolution limit) to 2 mm airborne droplet trajectories are aligned with the wave direction. Behind the wave, these droplets reverse their direction, presumably due to the airflow above the wave. 1 Supported by Gulf of Mexico Research Initiative (GoMRI) 9:18AM A33.00007 Heat transfer and convective structure of evaporating films under pressuremodulated conditions1 , JUAN CARLOS GONZALEZ-PONS2 , JAMES HERMANSON, University of Washington, JEFFREY ALLEN, Michigan Technological University — The interfacial stability, convective structure, and evaporation rate of upward-facing, thin liquid films were studied experimentally. Dichloromethane films approximately 2 mm thick were subjected to impulsive, time-varying superheating. The films resided on a temperature controlled, copper surface in a closed, initially degassed test chamber. Superheating was achieved by modulating the pressure of the saturated pure vapor in the test chamber. The dynamic film thickness was measured at multiple points using ultrasound, and the convective structure information was visualized by schlieren imaging. Two distinct raises in heat transfer rate under unsteady conditions were observed. The first transition appears to be associated with conduction within the film only; the second, to a change in the pattern of convection within the film. Different pressure-modulation cycles were studied to capture one or both of the observed rises in heat transfer. The total film thickness change over multiple cycles, as indicated by ultrasound, allowed determination of the total heat rejected into the evaporating films. Results suggest that there are cycle combinations that lead to an elevation in the average rate of heat transfer compared to films undergoing quasi-steady evaporation. 1 This 2 First work was sponsored by the National Aeronautics and Space Administration under Cooperative Agreement NNX09AL02G. AFD DFD meeting 9:31AM A33.00008 Effects of Soluble Surfactant on Lateral Migration of a Bubble in a Shear Flow1 , METIN MURADOGLU, Koc University, GRETAR TRYGGVASON, The University of Notre Dame — Motivated by the recent experimental study of Takagi et al. (2008), direct numerical simulations are performed to examine effects of soluble surfactant on the lateral migration of a deformable bubble in a pressure-driven channel flow. The interfacial and bulk surfactant concentration evolution equations are solved fully coupled with the incompressible NavierStokes equations. A non-linear equation of state is used to relate interfacial surface tension to surfactant concentration at the interface. A multiscale method is developed to handle the mass exchange between the interface and bulk fluid at high Peclet numbers, using a boundary-layer approximation next to the bubble and a relatively coarse grid for the rest of the flow. It is found that the surfactant induced Marangoni stresses can dominate over the shear-induced lift force and thus alter the behavior of the bubble completely, i.e., the contaminated bubble drifts away from the channel wall and stabilizes at the center of the channel in contrast with the corresponding clean bubble that drifts toward the wall and stabilizes near the wall. 1 The Scientific and Technical Research Council of Turkey (TUBITAK), Grant 112M181 and Turkish Academy of Sciences (TUBA). Sunday, November 23, 2014 8:00AM - 9:44AM Session A34 Sprays and Emissions — 2024 - Fabrizio Bisetti, King Abdullah University of Science and Technology 8:00AM A34.00001 Numerical investigation of spray ignition of a multi-component fuel surrogate , LARA BACKER, KRITHIKA NARAYANASWAMY, PERRINE PEPIOT, Cornell University — Simulating turbulent spray ignition, an important process in engine combustion, is challenging, since it combines the complexity of multi-scale, multiphase turbulent flow modeling with the need for an accurate description of chemical kinetics. In this work, we use direct numerical simulation to investigate the role of the evaporation model on the ignition characteristics of a multi-component fuel surrogate, injected as droplets in a turbulent environment. The fuel is represented as a mixture of several components, each one being representative of a different chemical class. A reduced kinetic scheme for the mixture is extracted from a well-validated detailed chemical mechanism, and integrated into the multiphase turbulent reactive flow solver NGA. Comparisons are made between a single-component evaporation model, in which the evaporating gas has the same composition as the liquid droplet, and a multi-component model, where component segregation does occur. In particular, the corresponding production of radical species, which are characteristic of the ignition of individual fuel components, is thoroughly analyzed. 8:13AM A34.00002 Integral Length Scale Effects on JP-8 Spray Penetration and Ignition at Elevated Pressure and Temperature Conditions , MATTHEW KURMAN, MICHAEL TESS, LUIS BRAVO, CHOL-BUM KWEON, US Army Research Laboratory — The effect of the integral length scale on global spray diagnostics was examined for non-reacting and reacting JP-8 sprays. The scales were set by varying the nominal nozzle diameter from 90 µm, 100 µm, and 147 µm, resulting in the ranges of Re (6.7 x 104 - 9.9 x 104 ) and We (1.3 x 106 - 1.7 x 106 ) setting the spray in the fully atomization mode. A high temperature (900-1000 K) high pressure (60-100 bar) flow through chamber was used to conduct experiments at relevant compression ignition engine operating conditions. Each fuel injector was characterized with an injection analyzer to determine the rate of injection and injected fuel mass. High speed near simultaneous Mie and schlieren images were acquired to determine the liquid and vapor penetration lengths of the non-reacting spray. Ignition delay experiments were conducted by measuring the start of formation of OH radicals. A numerical investigation was also carried out to provide additional insights into the behavior of each spray with the specified conditions. The quantitative results presented will aid in the overall advancement of fuel injector designs and ultimately lead to optimized engines. 8:26AM A34.00003 A spray-flamelet formulation using an effective composition-space variable , BENEDETTA FRANZELLI, AYMERIC VIÉ, MATTHIAS IHME, Center for Turbulence Research, Stanford — The modeling and simulation of spray flames is of primary importance as new combustion systems rely on the use of liquid fuels to feed the combustion process. The description of such flames is commonly performed using a mixture fraction variable that monotonically decreases from the fuel to the oxidizer side. Unfortunately, in the case of spray flames, this mixture-fraction variable is not monotonic as a result of the presence of an evaporation source term in the governing equations. To address this issue, a new composition space variable is defined, which is defined from the arc length along the gas-liquid mixture-fraction space. This monotonic definition enables the complete description of the spray-flame structure in composition space and the formulation of a well-posed spray-flamelet equation. A closure model for the scalar dissipation rate is proposed, and the potential of this effective composition-space variable is demonstrated by comparing simulation results in physical and composition space. 8:39AM A34.00004 Investigation of Mixing and Chemical Reaction Interactions Using RateControlled Constrained-Equilibrium , FATEMEH HADI, Northeastern University, MOHAMMAD JANBOZORGI, University of Southern California, REZA H. SHEIKHI, HAMEED METGHALCHI, Northeastern University — The Rate-Controlled Constrained-Equilibrium (RCCE) method is applied to study the interaction between mixing and chemical reaction in a constant pressure Partially-Stirred Reactor (PaSR). The objective is to understand the influence of mixing on RCCE predictions. The RCCE is a computationally efficient method based on thermodynamics to implement the combustion chemistry. In the RCCE the dynamics of reacting systems is described by a small number of rate-controlling reactions and slowly-varying constraints. The method is applied to study methane combustion via 12 constraints and 133 reaction steps. Simulations are carried out over a wide range of initial temperatures and equivalence ratios. The RCCE predictions are assessed by comparing with those of detailed kinetics model, in which the same kinetics, involving 29 species and 133 reaction steps, is integrated directly. Chemical kinetics and mixing interactions are studied for different residence and mixing time scales. Results show that the RCCE accurately represents the effect of mixing with different mixing strengths. An assessment of numerical performance of the RCCE is also performed. It is shown that the method is effective to reduce the stiffness of the kinetics and thus allows simulations with much lower computation costs. 8:52AM A34.00005 Radiative cooling in a flameholder for NOx reduction , ROBERT BREIDENTHAL, University of Washington, IGOR KRICHTAFOVITCH, DOUG KARKOW, JOSEPH COLANNINO, ClearSign Combustion Corporation — Recent experiments have revealed dramatic reductions in NOx emissions using a ceramic honeycomb as a flameholder. A jet of fuel entrains and mixes air before entering the honeycomb. The honeycomb is positioned at a distance away from the jet nozzle such that the mixed fluid arriving at the upstream edge of the honeycomb is combustible. Combustion occurs within the honeycomb, transferring heat to the ceramic walls, which glow red hot. According to a simple physical model, radiation and thermal conduction transport energy toward the upstream end of the honeycomb, thereby heating the incident cold reactants to maintain combustion. The radiation also transports energy downstream and away from the honeycomb, toward a thermal load. This is an attractive characteristic in boiler applications, for example. Furthermore, the hot combustion products in intimate thermal contact with the walls of the radiating honeycomb are rapidly cooled, consistent with the low NOx emissions. Preliminary experiments with different honeycomb configurations are in accord with this model. 9:05AM A34.00006 Eulerian methods for the description of soot: mathematical modeling and numerical scheme , T.T. NGUYEN, EM2C, CNRS UPR 288 - Fed. de Math., FR CNRS 3487 - ECP, France, A. WICK, Institute for Combustion Technology, RWTH Aachen University, Germany, F. LAURENT, EM2C, CNRS UPR 288 - Fed. de Math., FR CNRS 3487 - ECP, France, R. FOX, Iowa State University, USA, H. PITSCH, Institute for Combustion Technology, RWTH Aachen University, Germany — A development and comparison between numerical methods for soot modeling derived from the population balance equations (PBE) is presented. The soot mechanism includes nucleation, surface growth, oxidation, aggregation and breakage (Mueller et al., Proceed. Combust. Inst., 2009, 2011). For comparison, data from the ethylene premixed flame of Xu et al. (Combust. Flame 108, 1997) over a range of equivalence ratios are used. Two types of methods are introduced. The first is a moment method in which the closure is obtained through a reconstruction of the number density function (NDF). In particular, the NDF can be approximated by a sum of Gamma distribution functions (Yuan et al., J. Aero. Sci. 51, 2012). The second is Eulerian multi-fluid (MF), which is a size discretization method (Laurent et al., Combust. Theory Modelling 5, 2001) considering one or two moments per section. The case of one moment per section is also known as a sectional method. The accuracy of MF methods depends on the number of sections. Eventually, an extension of these two methods considering the surface area as a function of volume is taken into account to describe more precisely the geometry of soot particles. The solutions from these methods are compared with solutions from Monte Carlo method. 9:18AM A34.00007 On the effects of gas-phase species Lewis number in turbulent nonpremixed sooting flames , FABRIZIO BISETTI, ANTONIO ATTILI, King Abdullah University of Science and Technology, MICHAEL MUELLER, Princeton University, HEINZ PITSCH, RWTH Aachen University — Two large DNS of n-heptane/air turbulent nonpremixed combustion are compared to asses the effects of gas-phase species Lewis number on the dynamics of soot formation and growth. A detailed chemical mechanism, which includes PAHs, and a high-order method of moments for soot modeling are employed for the first time in the three-dimensional simulation of turbulent sooting flames. The results obtained employing a complex model (mixture average) for the transport of heat and mass [Attili et al. Comb. Flame, 161, 2014] are compared with those calculated with Le=1 for all gas-phase species, including large soot precursor molecules. It is found that the statistics of temperature and other species governing the heat releasing chemistry are very similar in the two cases as the flow field achieves a fully turbulent state. The dynamics of the soot precursors and soot display quantitative differences between the two cases. Employing the Le=1 approximation, the total mass of soot precursors in the flame decreases by 10 to 20% only, but its field is less inhomogeneous in space and time. Due to the non-linearity of soot growth with respect to the concentration of gas-phase precursors, the domain-averaged rate of soot mass production decreases by a factor of 2. 9:31AM A34.00008 LES of turbulent lifted CH4 /H2 flames using a novel FGM-PDF model1 , S. EBRAHIM ABTAHIZADEH, JEROEN VAN OIJEN, ROB BASTIAANS, PHILIP DE GOEY, Eindhoven University of Technology — This study reports on numerical investigations of preferential diffusion effects on flame stabilization of turbulent lifted flames using LES with a FGM-PDF approach. The experimental test case is the Delft JHC burner to study Mild combustion; a clean combustion concept. In this burner, CH4 based fuel has been enriched from 0 to 25% of H2. Since the main stabilization mechanism of these turbulent flames is autoignition, the developed numerical model should be able to predict this complex event. Furthermore, addition of hydrogen makes modeling even more challenging due to its preferential diffusion effects. These effects are increasingly important since autoignition is typically initiated at very small mixture fractions where molecular diffusion is comparable to turbulence transport (eddy viscosity). In this study, first, a novel numerical model is developed based on the Flamelet Generated Manifolds (FGM) to account for preferential diffusion effects in autoignition. Afterwards, the developed FGM approach is implemented in LES of the H2 enriched turbulent lifted jet flames. Main features of these turbulent lifted flames such as the formation of ignition kernels and stabilization mechanisms are thoroughly analyzed and compared with the measurements of OH chemiluminescence. 1 The authors gratefully acknowledge the financial support of the Dutch Technology Foundation (STW) under project No. 10414. Sunday, November 23, 2014 8:00AM - 9:44AM Session A35 Compressible Flow I: General Turbulence — 2001A - Sharath Girimaji, Texas A&M University 8:00AM A35.00001 p-adaption for compressible flow problems using a goal-based error estimator , DIRK EKELSCHOT, DAVID MOXEY, JOAQUIM PEIRO, SPENCER SHERWIN, Imperial Coll — We present an approach of applying p-adaption to compressible flow problems using a dual-weighted error estimator. This technique has been implemented in the high-order h/p spectral element library Nektar++. The compressible solver uses a high-order discontinuous Galerkin (DG) discretization. This approach is generally considered to be expensive and that is why the introduced p-adaption technique aims for lowering the computational cost while preserving the high-order accuracy and the exponential convergence properties. The numerical fluxes between the elements are discontinuous which allows one to use a different polynomial order in each element. After identifying and localizing the sources of error, the order of approximation of the solution within the element is improved. The solution to the adjoint equations for the compressible Euler equations is used to weigh the local residual of the primal solution. This provides both the error in the target quantity, which is typically the lift or drag coefficient, and an indication on how sensitive the local solution is to the target quantity. The dual-weighted error within each element serves then as a local refinement indicator that drives the p-adaptive algorithm. The performance of this p-adaptive method is demonstrated using a test case of subsonic flow past a 3D wing geometry. 8:13AM A35.00002 Effects of Segmented Slot Blowing at the Leading Edge of a Finite Span Cavity in Supersonic Flow , BENJAMIN GEORGE, LAWRENCE UKEILEY, Univ of Florida - Gainesville, LOUIS CATTAFESTA, KUNIHIKO TAIRA, Florida State University — The effects of finite span on the control of surface pressures within an open cavity in Mach 1.4 flow are studied. Experiments involve a finite span, rectangular cavity with a length to depth ratio of 6 and width to depth ratio of 2 being characterized using unsteady pressure along the floor and Particle Image Velocimetry (PIV). This data is first compared to measurements taken with a full span cavity of the same length to depth ratio to elucidate the flow phenomena caused by the introduction of sidewalls. Thereafter various leading edge slot configurations are employed and their effects compared to previous experiments involving the finite span cavity to gauge the influence of blowing on the three-dimensional flow field. The effectiveness of the slot blowing to suppress the pressure fluctuations are evaluated by examining reductions in both the tonal and broadband levels of the fluctuating surface pressure spectra. PIV data inside the finite span cavity show changes in the mean properties of the flow field when comparing the baseline to the slot blowing cases. Specifically the interaction of blowing with the flow near the cavity sidewalls, the shear layer, and the recirculation region are of interest. 8:26AM A35.00003 On the use of entropy viscosity based high order discontinuous spectral element method for capturing shocks , ARNAB CHAUDHURI, Department of Aerospace Engineering and Engineering Mechanics, San Diego State University, San Diego, CA, 92182, HESAM ABBASSI, Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL, 60607, GUSTAAF B. JACOBS, Department of Aerospace Engineering and Engineering Mechanics, San Diego State University, San Diego, CA, 92182, FARZAD MASHAYEK, Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL, 60607 — A modified entropy viscosity (EV) based high order discontinuous spectral element method (DSEM) has been proposed to deal with compressible flows involving shocks. Particular attention has been made to control undesired artificial dissipation dictated by entropy generation in shock-free shear dominated regions of the flow-field. A shock sensor based EV switch is used for this purpose. Implementation of the proposed method on an existing 3D parallel DSEM solver is successfully verified and validated with numerous benchmark problems. The effectiveness and applicability of stabiliser filters together with EV switch for various steady and unsteady flow configurations involving flow discontinuities are also tested. Results of simulations of compressible subsonic-to-supersonic flows over ramped cavity geometry, moving shocks over wedge involving regular/irregular Mach reflections are presented. 8:39AM A35.00004 DNS and LIA analysis of the shock turbulence interaction , DANIEL LIVESCU, Los Alamos National Laboratory, JAIYOUNG RYU, UC Berkeley — The interaction between isotropic turbulence and a normal shock wave is studied using Direct Numerical Simulations (DNS), with all flow scales (including the shock width) accurately solved, and the Linear Interaction Analysis (LIA). The turbulence quantities from DNS converge to the LIA solutions as the turbulent Mach number, Mt , becomes small, even at low upstream Reynolds numbers. This reconciles a long time open question about the role of LIA and establishes it as a reliable prediction tool for turbulence-shock interaction problems when there is a significant separation between the shock width and turbulence scales and Mt is low, which is encountered in many practical applications. The final LIA formulas are extended to investigate detailed turbulence physics. The extended LIA relations are used to show consistency with the DNS results and study the interaction at high Ms , where the resolution requirements make DNS studies unfeasible. The results show that the shock wave significantly changes the topology of the turbulent structures, with a symmetrization of the third invariant of the velocity gradient tensor and (Ms mediated) of the PDF of the longitudinal velocity derivatives, and an Ms dependent increase in the correlation between strain and rotation. 8:52AM A35.00005 A DNS study of supersonic boundary layer trip induced transition and turbulence1 , IZAAK BEEKMAN, M. PINO MARTIN, University of Maryland — We perform the direct numerical simulation (DNS) of a Mach 7.2, turbulent boundary layer, with a laminar inflow. A two-dimensional, semi-circular, bar-type roughness element is introduced near the inlet to hasten transition to turbulence. We choose this geometry because two-dimensional trip-wire-type devices have been used extensively by the experimental community, but we know of no computational studies to simulate transition behind such a roughness element in supersonic flow. We vary the trip size to investigate how size and trip-imposed length scales affect the transition process and the resulting turbulence. The incoming boundary layer conditions are matched to those of experiments being conducted at Princeton University Gas Dynamics Laboratory, where the free stream Mach number is 7.2. 1 This work is sponsored by the USAF under grant AF/9550-10-1-0535 STW 21 - Revitalization of the Hypersonics Testing and Evaluation Workforce 9:05AM A35.00006 Generation of compressible turbulence using lasers as sources of intense energy1 , AGUSTIN MAQUI, DIEGO DONZIS, Texas A&M University — Intense energy from lasers can be used to photo-dissociate molecules, ejecting fragments with extremely high energy. This energy can be in the form of translation (kinetic energy) as well as rotation and vibration for more complex molecular systems. When lasers are used in a flow, “lines” of concentrated kinetic and internal energy are generated. It is of fundamental as well as practical interest to know whether this source of energy is sufficient to generate turbulence downstream of a supersonic flow. Direct numerical simulations (DNS) are used to study how the flow evolves past the photo-excitation of molecules. Convergence studies are carried out to understand the numerical challenges associated with the strong gradients imposed by the intense energy fluctuations. A comprehensive analysis of single, as well as two point statistics is performed to understand the development towards realistic turbulence. The perturbations introduced are fully characterized to analyze how they determine the flow evolution and if the conditions can be replicated within a wind tunnel. Further results and consequences for particular cases realizable in laboratories will be discussed. 1 The authors gratefully acknowledge the support of AFOSR. 9:18AM A35.00007 Stability of vortical structures of high and low speed mixing layers , MONA KARIMI, SHARATH GIRIMAJI, Department of Aerospace Engineering, Texas A&M Engineering — It is known that mixing layers feature spanwise rollers and long streamwise vortices at low Mach numbers. At high speeds, the spanwise rollers are less evident. In this presentation, we attempt to identify the underlying instability mechanisms that ultimately lead to the growth or stabilization of different structures. Specifically, we examine the effect of Mach number and perturbation orientation on vorticity production. The nature of vortex stretching at different Mach numbers is investigated and the change in its character due to the onset dilatational velocity fluctuations is established. It is shown that the dilatational field suppresses spanwise rollers while leaving the streamwise vorticity intact. 9:31AM A35.00008 Acoustic Radiation from a Mach 14 Turbulent Boundary layer1 , LIAN DUAN, Missouri Univ of Sci & Tech, MEELAN CHOUDHARI, NASA Langley Research Center — Direct numerical simulations (DNS) are used to examine the pressure fluctuations generated by a high-speed turbulent boundary layer with a nominal freestream Mach number of 14 and wall temperature of 0.15 times the recovery temperature. The emphasis is on characterizing the acoustic radiation from the turbulent boundary layer and comparing it with previous simulations at Mach 2.5 and Mach 6 to assess the Mach-number dependence of the freestream pressure fluctuations. In particular, the numerical database is used to provide insights into the pressure disturbance spectrum and amplitude scaling with respect to the freestream Mach number as well as to understand the acoustic source mechanisms at very high Mach numbers. Such information is important for characterizing the freestream disturbance environment in conventional (i.e., noisy) hypersonic wind tunnels. Spectral characteristics of pressure fluctuations at the surface are also investigated. 1 Supported by AFOSR and NASA Langley Research Center. Sunday, November 23, 2014 8:00AM - 9:44AM Session A36 Magnetohydrodynamics — Alcove A - Olev Zikanov, University of Michigan 8:00AM A36.00001 Flow regimes in an electromagnetically forced circular Couette system , JEAN BOISSON, Unité de Mécanique, UMR 8193, ENSTA-ParisTech, Palaiseau, VINCENT PADILLA, FRANÇOIS DAVIAUD, SÉBASTIEN AUMAÎTRE, SPHYNX-SPEC, DSM, CNRS URA 2464, CEA-Saclay, F–91191 Gif-sur-Yvette, UNITÉ DE MÉCANIQUE, UMR 8193, ENSTA-PARISTECH TEAM, SPHYNXSPEC, DSM, CNRS URA 2464, CEA-SACLAY TEAM — We present an experimental study of a liquid metal flow electromagnetically forced in a large aspect ratio coaxial cylindrical geometry with and without a free surface. An azimuthal Lorentz force is applied on the liquid metal gap, through a radial current and an axial magnetic field. Using ultrasonic velocity measurements and direct visualisation of the free surface, we focus on the effect of these two parameters on the flow properties. We show that, depending on the strength of the magnetic field and not only on the applied Lorentz force, dynamical states exist in the two geometries. We first observe a stationary structure at low forcing. Then, two dynamical regimes are exhibited at higher forcing. We characterize them by their different frequencies and speeds. Higher magnetic fields clearly promote the faster regimes. Connections with other magnetohydrodynamics instabilities will be discussed. 8:13AM A36.00002 Numerical simulation of partially ionized gas flows under the influence of electromagnetic fields , KONSTANTINOS PANOURGIAS, University of Patras, JOHN EKATERINARIS, Embry-Riddle Aeronautical University — Partially ionized gases under the influence of electromagnetic fields are described through the coupled system of the compressible Navier-Stokes equations augmented by the equations of species in the mixture (electrons, ions, atoms) and the Maxwell equations. The coupled system is completed with an energy equation for electrons. Stiff source terms encompass the interactions of fluid flow with electromagnetic fields and resulting system of equations is solved numerically. The discontinuous Galerkin finite element method is used for the numerical solution of the above system.For the Maxwell equations, DG method is performed using a divergence free vector basis for the magnetic field in order to preserve zero divergence in the element and retain the global implicit constraint of a divergence free magnetic field vector down to very low levels. In order to avoid severe time step limitations for the Maxwell system, implicit time marching is used with high order implicit Rugne-Kutta methods. The coupled system of the Navier-Stokes and the Maxwell equations is advanced in time simultaneously to avoid wrong wave shapes and propagation speeds that are obtained when the coupling source terms are lagged in time.The method is applied for supersonic plasma flows in strong electromagnetic fields. 8:26AM A36.00003 Inverse cascades and the evolution of decaying magnetohydrodynamic turbulence1 , MORITZ LINKMANN, ARJUN BERERA, University of Edinburgh — Ensemble averaged high resolution direct numerical simulations of inverse cascade are presented, extending on the many single realization numerical studies done up to now. This identifies inverse cascade as a statistical property of magnetohydrodynamic turbulence and thus permits reliable numerical exploration of its dynamics. Our results show that at early times during the decay the properties of the ensemble average are represented by one realization, as the deviations between realizations are small. In contrast, at late times we measure significant deviations between realizations, thus the ensemble average cannot be avoided in this time frame. This is important for measurements of the magnetic energy decay exponent, which has been determined from these ensemble runs to be nE = (0.47 ± 0.03) + (13.9 ± 0.8)/Rλ for initially helical magnetic fields. We show for the first time that even after removing the Lorentz force term in the momentum equation, thus decoupling it from the induction equation, inverse cascade persists. The induction equation is now a linear partial differential equation with an externally imposed velocity field, thus amenable to numerous analysis techniques. A new door has opened for analyzing inverse cascade, with various ideas discussed. 1 This work has made use of the resources provided by the UK supercomputing services HECToR and ARCHER, made available through ECDF. AB acknowledges funding from STFC, and ML is supported by EPSRC. 8:39AM A36.00004 Azimuthal magnetorotational instabilities perturbations1 , YASUHIDE FUKUMOTO, Institute of Mathematics for Industry, Kyushu University, RONG to non-axisymmetric ZOU, Graduate School of Mathematics, Kyushu University — Short-wavelength stability analysis is made of axisymmetric rotating flows of a perfectly conducting fluid subjected to external azimuthal magnetic field, to non-axisymmetric as well as axisymmetric perturbations. The instability caused by the azimuthal magnetic field is referred to as the azimuthal magnetorotational instability (AMRI). We determine the range of unstable angular-velocity distribution and the overall maximum growth rate for the AMRI. Non-axisymmetric perturbations, when coupled to azimuthal magnetic field, widen the instability range of angular-velocity profiles of rotating flows. For strong external field, the maximum growth rate increases, beyond the Oort A-value, without bound in proportion to the strength of the external field. The effect of the electric resistivity is also considered in the limit of very low magnetic Prandtl number. 1 Y. F. was supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (Grant No. 24540407). 8:52AM A36.00005 Spectra and correlations in the solar wind from Voyager 2 around 5 AU , LUCA GALLANA, FEDERICO FRATERNALE, MICHELE IOVIENO, Dipartimento di Ingegneria Meccanica e Aerospaziale, Politecnico di Torino, ENRICO MAGLI, SOPHIE FOSSON, Dipartimento di Elettronica, Politecnico di Torino, MERAV OPHER, University of Boston, Astronomy Department, JOHN RICHARDSON, Kavli institute, MIT, DANIELA TORDELLA, Dipartimento di Ingegneria Meccanica e Aerospaziale, Politecnico di Torino — Solar wind spectra deduced from the data recorded by the Voyager 2 mission during 1979 at about 5 astronomical units from the sun are considered. The data are time series which contain voids that typically become larger and irregularly sparse as the craft moves away from the sun (45% missing data in 1979). By extracting complete subsets and filling gaps with different techniques (polynomial interpolation, Rybicki (AJ 1992) and compressed sensing (e.g. Candes et al. CPAM 2006) reconstruction methods, global DFT for irregularly spaced data) we obtain velocity and magnetic field fluctuations between 10−5 and 10−2 Hz in the MHD inertial range of solar wind. Spectra of all variables show a power law scaling with exponents in between -1.5 and -1.8. PDFs and correlations indicate that the flow has a significant intermittency. The reliability of the reconstruction methods used is analyzed by introducing the same sequence of gaps observed in the Voyager data into a reference dataset extracted from direct numerical simulations of incompressible Navier-Stokes turbulence as well as from synthetic turbulence, and then by comparing the statistics obtained with those of the complete reference dataset. 9:05AM A36.00006 Turbulent convection in a horizontal duct with strong axial magnetic field1 , XUAN ZHANG, OLEG ZIKANOV, University of Michigan - Dearborn — Convection in a horizontal duct with one heated wall is studied computationally. The work is motivated by the concept of a blanket for fusion reactors, according to which liquid metal slowly flows in toroidal ducts aligned with the main component of the magnetic field. We first assume that the magnetic field is sufficiently strong for the flow to be purely two-dimensional and analyze chaotic flow regimes at very high Grashof numbers. Furthermore, three-dimensional perturbations are considered and the relation between the length of the duct and the critical Hartmann number, below which the flow becomes three-dimensional, is determined. 1 Financial support was provided by the US NSF (Grant CBET 1232851). 9:18AM A36.00007 Elevator convection modes in vertical ducts with strong transverse magnetic fields1 , OLEG ZIKANOV, LI LIU, University of Michigan - Dearborn — Instability modes in the form of axially uniform vertical jets, also called “elevator modes,” are known to be solutions of thermal convection problems for vertically unbounded systems. Typically, their relevance to an actual flow state is limited, since they quickly break down to secondary instabilities. We consider a downward flow of a liquid metal in a vertical duct with a heated wall and strong transverse magnetic field and find elevator modes that are likely to be not just relevant, but a dominant feature of the flow. Recent experiments indicate that counterparts of such modes may develop in vertically finite ducts leading to high-amplitude fluctuations of temperature. Potential implications for designs of liquid metal blankets for fusion reactors with poloidal ducts are discussed. 1 Financial support was provided by the US NSF (Grant CBET 1232851). 9:31AM A36.00008 On the dynamic behavior of the flow past a magnetic obstacle , ALBERTO BELTRAN, ROBERTO DOMINGUEZ-LOZOYA, JOEL ROMAN, EDUARDO RAMOS, SERGIO CUEVAS, Universidad Nacional Autonoma de Mexico — We study numerically the duct flow of an electrically conducting incompressible viscous fluid (a liquid metal) past a a localized magnetic field, namely, a magnetic obstacle. We use a quasi-two-dimensional model based on a formulation that includes the induced magnetic field as electromagnetic variable (B-formulation) and analyze the stability of the flow in the parametric space of the Hartmann and Reynolds numbers. We find that even though for a given strength of the localized braking Lorentz force (characterized by the Hartmann number) the flow may become unstable and give rise to a time-periodic wake, when a critical Reynolds number is reached, a further increase in the Reynolds number may result in the flow becoming steady again. Evidently, this behavior is not observed in the flow past a solid obstacle. Experimental observations carried out in a liquid metal (GaInSn) duct flow suggest that this prediction is correct. Sunday, November 23, 2014 10:25AM - 12:30PM — Session B1 Plenary/Awards Session followed by Otto Laporte and Corrsin Lectures Level 1 Hall - Nadine Aubry, Northeastern University 10:25AM B1.00001 Welcome, Presentation of Awards and DFD Fellowships — 11:00AM B1.00002 Fluid Dynamics Prize Otto Laporte Lecture:Turbulence and Aeroacoustics , GENEVIEVE COMTE-BELLOT, Ecole Centrale de Lyon — Some significant advances obtained over the years for two closely related fields, Turbulence and Aeroacoustics, are presented. Particular focus is placed on experimental results and on physical mechanisms. For example, for a 2D channel flow, skewness factors of velocity fluctuations are discussed. The study of isotropic turbulence generated by grids in the “Velvet wind tunnel” of Stanley Corrsin, constitutes a masterpiece. Of particular note are the Eulerian memory times, analysed for all wavenumbers. Concerning hot-wire anemometry, the potential of the new constant voltage technique is presented. Some results obtained with Particule Image Velocimetry are also reported. Two flow control examples are illustrated: lift generation for a circular cylinder, and noise reduction for a high speed jet. Finally, the propagation of acoustic waves through turbulence is considered. Experimental data are here completed by numerical simulations showing the possible occurrence of caustics. 11:45AM B1.00003 Stanley Corrsin Award Lecture: Lagrangian Measurements in Turbulence: From Fundamentals to Applications , EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Self-Organization — In my talk I shall present results from particle tracking experiments in turbulence. After a short review of the history of the field, I shall summarize the most recent technological advances that range form low and high-density particle tracking to direct measurements of the Lagrangian evolution of vorticity. I shall embark on a journey that describes the discoveries made possible by this new technology in the last 15 years. I present results that challenge our understanding of turbulence and show how Lagrangian particle tracking can help us ask questions on turbulent flows that so far were hidden. I shall show how Lagrangian particle tracking may provide important insights into the reversibility of turbulent flows, on vorticity generation, the energy cascade and turbulent mixing. I shall describe the consequences of inertial particle transport on rain formation and end with an outlook on how Lagrangian particle tracking experiments on non-stationary flows in real-world situations may provide high quality data that can support real world engineering problems. I am very thankful for the support by Cornell University, the National Science Foundation, the Research Corporation, the Alfred P. Sloan Foundation, the Kavli Institute for Theoretical Physics, the German Research Foundation, the European Union and the Max Planck Society. I very gratefully acknowledge the excellent partnership with many colleagues in the field of fluid mechanics and turbulence. Sunday, November 23, 2014 1:35PM - 2:10PM — Session C8 Invited Session: Art and Science Duality in Fluid Mechanics 3001/3003 - Mory Gharib, California Institute of Technology 1:35PM C8.00001 Art & Science duality in Fluid Mechanics , JEAN-MARC CHOMAZ, LADHYX, Ecole Polytechnique, Paris — The connections between Art & Science is analysed through examples of my research both in Fluid Mechanics and in Art & Science. Working as a member of the artist group Labofactory and collaborating with more than twenty different artists, I have been exploring for more than twenty-four years a path between art and science that mixes both scientific and artistic imaginations. Formulating questions in science is pure imagination and intuition that does not involve only the sensible side of the brain but the sensitive side, which is able to be non incremental, to understand faster and anticipate. Instead of showing scientific proof or technique, it is possible with Art & Science to directly attempt to share this sensitive side. I will show ten recent installations that involve vortex rings, tornado generators, music propagated in shallow layers, wave tanks used as silent soft drums, boundary layer on a rotating sphere to question climate change, plum ever evolving over a nuclear plan in an water tank, a bubbly fountain in microfluidic... Two installations on the thermohaline circulation staged in a stratified tank and on the generation of earthquake are part of the exhibit “LOST IN FATHOMS” with the artist Anaı̈s Tondeur from 17 October until 29 November 2014 at the GV Art gallery, London. These pieces are like writing poems using fluid mechanics and by doing so re-interrogating our scientific practice and the societal role of science. They symmetrize the relation with the public that involve not only “outreach” but “inreach” or sharing. Sunday, November 23, 2014 1:35PM - 2:10PM — Session C14 Invited Session: Control and Simulation of Thermoacoustic Instabilities 3009/3011 - Parviz Moin, Stanford University 1:35PM C14.00001 Control and simulation of thermoacoustic instabilities1 , THIERRY POINSOT, CERFACS — Combustion instabilities (CI), due to thermoacoustic coupling between acoustic waves and chemical reaction, constitute a major danger for all combustion systems. They can drive the system to unstable states where the whole combustor can oscillate, vibrate, quench or in extreme cases explode or burn. Such phenomena are commonly observed in the final phases of development programs, leading to major difficulties and significant additional costs. One of the most famous examples of combustion instabilities is the F1 engine of the Apollo program which required more than 1000 engine tests to obtain a stable regime satisfying all other constraints (performance, ignition, etc). CIs constitute one of the most challenging problems in fluid mechanics: they combine turbulence, acoustics, chemistry, unsteady two-phase flow in complex geometries. Since combustion instabilities have been identified (more than hundred years ago), the combustion community has followed two paths: (1) improve our understanding of the phenomena controlling stability to build engines which would be “stable by design” and (2) give up on a detailed understanding of mechanisms and add control systems either in open or closed loop devices to inhibit unstable modes. Of course, understanding phenomena driving combustion instabilities to suppress them would be the most satisfying approach but there is no fully reliable theory or numerical method today which can predict whether a combustor will be stable or not before it is fired. This talk will present an overview of combustion instabilities phenomenology before focusing on: (1) active control methods for combustion instabilities and (2) recent methods to predict unstable modes in combustors. These methods are based on recent Large Eddy Simulation codes for compressible reacting flows on HPC systems but we will also describe recent fully analytical methods which provide new insights into unstable modes in annular combustion chambers. 1 Support: European Research Council Advanced Grant INTECOCIS (intecocis.inp-toulouse.fr) Sunday, November 23, 2014 2:15PM - 4:25PM Session D1 Non-Newtonian Flows: Instability and Turbulence — 3000 - Alexander Morozov, University of Edinburgh 2:15PM D1.00001 ABSTRACT WITHDRAWN — 2:28PM D1.00002 Viscoelastic Taylor-Couette instability as an anolog of Magnetorotational instability1 , INNOCENT MUTABAZI, YANG BAI, OLIVIER CRUMEYROLLE, LOMC, UMR 6294, CNRS-University of Le Havre — Our investigation of the viscoelastic instability (VEI) in the corotating Couette-Taylor system is motivated by the prediction of Ogilvie et. al that such an instability is analogous to the MRI (magneto-rotational instability) which is believed to play a key role in the angular momentum transport in accretion disks. This analogy is supported by stretched spring argument developed by Balbus and Hawley which is similar to that of the polymer stretching model in viscoelastic solutions. To our best knowledge, only one experiment by Boldyrev et al. has been reported for the search of the analogy VEI-MRI. We present both theoretical and experimental results obtained in the viscoelastic Couette-Taylor system when both the cylinders are constrained to rotate along the Keplerian and anti-Keplerian lines. The polymer solutions have a constant solution with respect to shear rate and can be described by the Odlroyd-B model. The control parameters are the aspect ratio Γ, the radius ratio η, the Reynolds number Re, the elastic number E = W i/Re and the viscosity ratio S = µp /µ. After linear stability analysis, critical modes are oscillatory and non-axisymmetric. The observed modes are either stationary or oscillatory modes. A state diagram allows for a comparison to MRI 1 Partial EMC3. support from the French National Research Agency (ANR) through the program Investissements d’Avenir (ANR-10 LABX-09-01), LABEX 2:41PM D1.00003 Travelling waves and their stability in elasto-inertial plane Poiseuille flow1 , TOBY SEARLE, ALEXANDER MOROZOV, University of Edinburgh — Purely elastic turbulence occurs in polymer solutions and other viscoelastic fluids when the Reynold’s number is very small (Re < 1) and the Weissenberg number is large. Recent numerical modelling and experimental study has revealed another form of turbulence somewhere between that controlled by inertia and that governed by the elasticity of the fluid. Flows in this elasto-inertial regime support coherent structures that are unlike the usual Newtonian ones, and turbulence sets in at a lower Reynold’s number. It is thought that these structures are similar to those present in purely elastic turbulence. We use 2 dimensional exact solutions in plane Poiseuille flow of an Oldroyd-B fluid to investigate this elasto-inertial regime. First we find viscoelastic travelling wave solutions via numerical continuation from their Newtonian counterparts. We investigate how these solutions are affected by the addition of the polymeric fluid and perform a linear stability analysis in the spanwise direction. We find that these viscoelastic travelling-wave solutions are in fact unstable to 3 dimensional perturbations, and discuss how these instabilities differ from those found in Newtonian turbulence. 1 SUPA, School of Physics and Astronomy, University of Edinburgh, Mayfield Road, Edinburgh, EH9 3JZ, UK 2:54PM D1.00004 The Effect Of Viscosity and Non-Newtonian Rheology On Reaction Enhancement Between Two Initially Distant Scalars1 , FARROKH SHOAEI, JOHN CRIMALDI, University of Colorado, Boulder — The effect of viscosity and non-Newtonian (shear-thinning) rheology on mixing and reaction between two initially distant scalars has been investigated using a two-channel planar laser-induced fluorescence technique (2C-PLIF). The scalars are stirred and mixed in the mildly turbulent (Re=2000) wake of a round cylinder. The scalars are released continuously upstream of the cylinder, with a separation that initially impedes the reaction. The ambient flow is pure water, but the scalar solutions include Xanthan gum to alter their rheology. Results indicate that mixing and reaction rates in the low-Damkohler limit between the two scalars plumes increase as the viscosity of the scalars is increased. The study also shows that the dominant contribution of total reaction derives from the scalar covariance associated with instantaneous flow processes, and depends strongly on viscosity and non-Newtonian rheology of the scalars in the domain. The results have broad implications for biological and ecological mixing processes involving now-Newtonian fluids. 1 This work was supported by the National Science Foundation under Grants No. 0849695 and No. 1205816. 3:07PM D1.00005 Bulk elastic fingering in soft materials , BAUDOUIN SAINTYVES, Harvard University, JOHN BIGGINS, Cambridge University, ZHIYAN WEI, Harvard University, ELISABETH BOUCHAUD, ESPCI ParisTech/CEA, L. MAHADEVAN, Harvard University, HARVARD UNIVERSITY TEAM, EC2M/ESPCI COLLABORATION, CAMBRIDGE UNIVERSITY COLLABORATION — Systematic experiments have been performed in purely elastic polyacrylamide gels in Hele-Shaw cells. We have shown that a bulk fingering instability arises in the highly deformable confined elastomers. A systematic study shows that surface tension is not relevant. This instability is sub-critical, with a clear hysteretic behavior. Our experimental observations have been compared very favorably to theoretical and finite element simulations results. In particular, the instability wavelength and the critical front advance have been shown to be proportional to the distance between the two glass plates constituting the cell. A very important feature is that elasticity doesn’t influence this lengthscale, making this instability very generic. We will also show some new results about an elastic counterpart experiment of the famous Saffman-Taylor experiment, where we push a soft gel in a stiff one. 3:20PM D1.00006 Stability of the boundary layer on a rotating disk for power-law fluids , PAUL GRIFFITHS, School of Mathematics, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK, STEPHEN GARRETT, Department of Engineering, University of Leicester, University Road, Leicester LE1 7RH, UK — The stability of the flow due to a rotating disk is considered for shear-thinning fluids that satisfy the power-law relationship. In this case the basic flow is not an exact solution of the Navier-Stokes equations, however, in the limit of large Reynolds number the flow inside the three-dimensional boundary layer can be determined via a similarity solution. An asymptotic analysis is presented in the limit of large Reynolds number. It is shown that the stationary spiral instabilities observed experimentally in the Newtonian case can be described for shear-thinning fluids by a linear stability analysis. Predictions for the wavenumber and wave angle of the disturbances suggest that shear-thinning fluids may have a stabilizing effect on the flow. The hypotheses of the asymptotic study are confirmed via a numerical investigation. The neutral curves are computed using a sixth-order system of linear stability equations which include the effects of streamline curvature, Coriolis force and the non-Newtonian viscosity model. It is found that the neutral curves have two minima; these are associated with the type I (cross-flow) and type II (streamline curvature) modes. Results indicate that an increase in shear-thinning has a stabilizing effect on both the type I and II modes. 3:33PM D1.00007 Exact coherent states in purely elastic parallel shear flows1 , TOBY SEARLE, ALEXANDER MOROZOV, University of Edinburgh — Parallel shear flows provide a model system for the understanding of the transition to and structure of incompressible Newtonian turbulence. The turbulent attractor is often thought of as structured by a series of exact solutions to the Navier-Stokes equations, where a turbulent flow “pinballs” between these solutions in phase space. The most intuitive mechanism for the appearance of these structures was formulated by F. Waleffe and is known as “the self-sustaining process.” A novel form of turbulence has been discovered in polymeric fluids where the Reynold’s number is low, Re < 1, and the Weissenberg number (characterising the fluid elasticity) is large. Using an analogy with the Newtonian self-sustaining process, we attempt to construct the purely elastic counterpart for plane Couette flow of polymer solutions. By introducing a forcing term to the coupled Navier-Stokes and Oldroyd-B equations, we observe the formation of purely elastic streaks and consider their linear stability. We find that there exists a previously unrecognised purely elastic analogue of the Kelvin-Helmholtz instability that gives rise to the streamwise waviness of Newtonian coherent structures. We discuss how this instability might close the cycle and lead to a sustained purely elastic coherent structure. 1 SUPA, School of Physics and Astronomy, University of Edinburgh, Mayfield Road, Edinburgh, EH9 3JZ, UK 3:46PM D1.00008 An “inverse-Orr” mechanism for spanwise vorticity amplification in viscoelastic fluids , JACOB PAGE, Imperial College London, TAMER ZAKI, Johns Hopkins University, Imperial College London — The linear dynamics of spanwise vorticity fluctuations in homogeneous shear flow of a viscoelastic fluid are examined. A weak Gaussian vortex is superposed onto the mean shear and its time evolution is computed for Oldroyd-B and FENE-P fluids. Unlike the Newtonian case where the vortex is purely advected, the polymeric flow exhibits intriguing behaviors: (i) At high elasticity, the vortex splits into a co-rotating pair which propagate in opposing horizontal directions. (ii) For weaker elasticities, the vortex splitting takes place at early times but the evolution is dominated by an algebraic amplification of vorticity. Both the splitting and amplification are explained using the linear equations for spanwise vorticity and polymer torque for the Oldroyd-B fluid. The splitting results from the ability of the fluid to support vorticity wave propagation along the tensioned mean-flow streamlines. The spanwise vorticity amplification occurs due to a kinematic mechanism that generates polymer torque. This mechanism is an “inverse-Orr” effect where amplification occurs as the disturbance is tilted into the shear. In the case of finite polymer extensibility, similar qualitative features are retained although decay sets in earlier as polymer chains become significantly stretched. 3:59PM D1.00009 POD analysis of viscoelastic flow instabilities , DAVID STEIN, BECCA THOMASES, Univ of California - Davis — Elastic instabilities in low Re viscoelastic flows near extensional points have been identified in experiments and simulations and are thought to be related to elastic turbulence. We study an unsteady two dimensional Stokes Oldroyd-B extensional point flow. Beyond a critical Weissenberg number, the system displays complex time-dependent flow patterns. We examine these quasi-periodic states in detail, and use proper orthogonal decomposition (POD) to extract the dominant oscillatory flow features in an effort to understand the elastic instability and the possible transition to elastic turbulence. 4:12PM D1.00010 Simulation of elastic and elasto-inertial turbulence in straight channel flows , YVES DUBIEF, University of Vermont, VINCENT TERRAPON, SAMIR SID, University of Liege — Elastic turbulence (ET, Nature, 410, 905-908, 2000) is a chaotic flow state generated and sustained by polymer additives at vanishing Reynolds numbers. It is generally accepted that elastic turbulence occurs when the mean flow streamlines are curved. Elasto-inertial turbulence (EIT, PNAS, 220 (26), 10557-10562, 2013) is a similar state of turbulence that happens in inertial flows with mean straight flow streamlines at Reynolds numbers for which the flow is laminar in the absence of polymers. A recent experiment (PRL 110, 174502, 2013) has shown that ET generated by the insertion of cylinders at the inlet of a low Reynolds number channel flow is sustained downstream of the perturbation. This experiment suggests a possible relation between ET and EIT. Our study will first confirm that sustained ET can be triggered in low-Reynolds number channel flows. ET is shown to exist in two- and three-dimensional simulations for Reynolds numbers of the order of 100 or less. Much like the aforementioned experiment, the initial conditions triggering ET cause the flow streamlines to be curved for a short duration at the beginning of the simulation. Our study will then discuss the similarities and differences between ET and EIT. Sunday, November 23, 2014 2:15PM - 4:25PM Session D2 Suspensions: Rheology — 3002 - Itai Cohen, Cornell University 2:15PM D2.00001 Effect of friction on the rheology of dense suspensions , STANY GALLIER, Safran, ELISABETH LEMAIRE, FRANÇOIS PETERS, LAURENT LOBRY, LPMC-CNRS, University of Nice — This work reports three-dimensional numerical simulations of sheared non-Brownian concentrated suspensions using a fictitious domain method. Contacts between particles are modeled using a DEM-like approach (Discrete Element Method), which allows for a more physical description, including roughness and friction. This study emphasizes the effect of friction between particles and its role on rheological properties, especially on normal stress differences. Friction is shown to notably increase viscosity and second normal stress difference |N2 | and decrease |N1 |, in better agreement with experiments. The hydrodynamic and contact contributions to the overall particle stress are particularly investigated and this shows that the effect of friction is mostly due to the additional contact stress since the hydrodynamic stress remains unaffected by friction. Simulation results are also compared with experiments and the agreement is improved when friction is accounted for: this suggests that friction is operative in actual suspensions. 2:28PM D2.00002 Force-induced diffusion in hydrodynamically interacting colloidal dispersions: non-monotonic connections between fluctuation and dissipation , NICHOLAS HOH, ROSEANNA ZIA, Cornell University — Effects of hydrodynamic interactions on equilibrium self-diffusivity are well known; here we explore their influence on the force-induced diffusion of a microrheological probe by tuning the strength of such interactions via an excluded-annulus model. As the probe is driven through the suspension, its force-induced diffusion (microdiffusivity) is determined analytically in the limits of strong and weak forcing, and numerically for all forcing. The total diffusivity comprises that of an isolated probe, the entropic hindrance of the equilibrium microstructure, and non-equilibrium interactions between probe and bath particles. When hydrodynamic interactions are important, three factors contribute to the microdiffusivity: a reduction in probe mobility; Brownian flux due to microstructural deformation; and entropic exclusion (collisions). Long-range hydrodynamic interactions diminish all components of the microdiffusivity; however, lubrication interactions enhance longitudinal encounters. This manifests as a monotonic increase with flow strength, but with a surprising non-monotonic dependence on the strength of hydrodynamic interactions. That is, the role of hydrodynamics in the connection between diffusive fluctuation and viscous dissipation is non-monotonic. 2:41PM D2.00003 Transient, nonlinear rheology of reversible colloidal gels by dynamic simulation , BENJAMIN LANDRUM, WILLIAM RUSSEL, Princeton University, ROSEANNA ZIA, Cornell University — We study the nonlinear rheology of reversible colloidal gels via dynamic simulation as they undergo age- and flow-induced structural evolution, with a view toward understanding and predicting transient behaviors such as multi-step and delayed yield. The gel is formed from 750,000 Brownian spheres interacting via hard-sphere repulsion and O(kT) short-range attraction, where thermal fluctuations are strong enough to allow continued structural rearrangement in the absence of flow. During startup of imposed strain rate, the transition to steady state is characterized by one or more “overshoots” in the stress which suggest initial yield, formation of a stronger gel, and subsequent yield of the new gel. When subjected to step-shear stress, the microstructure undergoes limited creep, followed by viscous flow. This macroscopic “delayed flow” is consistent with previously proposed models of competition between breakage and formation of particle bonds among static load-bearing structures. Our findings suggest, however, that the load-bearing structures evolve, and that the gel’s resistance to delayed failure depends upon this structural evolution and reinforcement. We put forth a micro-mechanical model of stress gradient-driven particle transport that captures this macroscopic behavior. 2:54PM D2.00004 Normal stress differences in suspensions of rigid fibers1 , BRADEN SNOOK, University of Florida; Aix-Marseille Universite, LEVI DAVIDSON, JASON BUTLER, University of Florida, OLIVIER POULIQUEN, ELISABETH GUAZZELLI, Aix-Marseille Universite, UNIVERSITY OF FLORIDA TEAM — Numerical and experimental studies of normal stress differences in suspensions of rigid, non-Brownian fibers were carried out for length (L) to diameter (d) ratios of 11 to 30 at concentrations nL2 d=1.5 to 3, where n is the number density of fibers. The numerical results are in quantitative agreement with the experimental results and allow calculation of the hydrodynamic and contact contributions to the stress in the suspension. The simulations show that the contact contribution to the rheology is dominant in determining the normal stress differences, where the first normal stress difference is positive and approximately twice the magnitude of the second normal stress difference, which is negative. 1 This work was supported by the National Science Foundation (Grant No. 0968313). B. Snook acknowledges support from a Chateaubriand Fellowship provided by the Embassy of France. 3:07PM D2.00005 The effect of geometry on the particle stress in suspensions of rigid particles in simple shear1 , MOHSEN DAGHOOGHI, IMAN BORAZJANI, State Univ of NY - Buffalo — The contribution of particles on the total stress of a suspension is known as particle stress, which consists of three sources: moment of stress on the particle surface, inertial term and Reynolds stress term. The symmetric part of the first term, i.e. stresslet, is considered as the most important term in rheological calculation and contribution of other terms is mainly ignored in low Reynolds regimes. For suspensions of rigid spheres at steady state these terms are negligible comparing to stresslet of the suspension, however this might not be the case for complex particle shapes. Using immersed boundary method, we simulate suspensions of complex shaped particles in simple shear flow to investigate the role of other two terms on the total particle stress and effective viscosity. We validated our results against classical analytical results for the low Reynolds-Stokes problem of suspension of ellipsoidal particles by Jeffery. We studied the effect of volume fraction of suspension and particle shape (aspect ratio) on the rheology of suspensions at Reynolds number range of 0.01 < Re < 10. Our study shows that particle shape has an mportant role on all components of the particle stress, and for Re > 1 the budget of inertial term in the total particle stress is not negligible. 1 This work was supported by the American Chemical Society Doctoral New Investigator grant. The computational resources were partly provided by Center for Computational Research (CCR) at University at Buffalo. 3:20PM D2.00006 Constant stress and pressure rheology: a unified perspective on the arrest of colloidal suspensions , MU WANG, JOHN BRADY, California Institute of Technology — We study the constant pressure and stress rheology of dense hard-sphere colloidal suspensions using a novel Brownian dynamics simulation algorithm. The simulations show an arrested region exhibiting viscosity divergence between the glass and the jamming transitions. Both the suspension shear and normal viscosities near the arrested region display universal power law divergences that depend solely on the volume fraction distance from the corresponding arrest point. We further found that the microscopic particle diffusion correlates with the suspension pressure through a Stokes-Einstein-Sutherland-like relation. With an estimation of the effect of hydrodynamic interactions and a careful analysis of the accessible volume in the experiments, the simulations are in quantitative agreement with the experiments of Boyer et al. [PRL 107, 188301 (2011)] in the non-Brownian limit. The simulations clearly show the fundamental role of the jamming transition in the dense suspension rheology, and illustrate the great care needed when performing and analyzing experiments and simulations near the maximum allowable volume fractions. 3:33PM D2.00007 When hard spheres overlap - generalization of the Rotne-Prager-Yamakawa hydrodynamic tensors1 , ELIGIUSZ WAJNRYB, Institute of Fundamental Technological Research Polish Academy of Sciences, Poland, PAWEL ZUK, University of Warsaw, Physics Department, Poland, KRZYSZTOF MIZERSKI, Institute of Geophysics, Polish Academy of Sciences, Poland, PIOTR SZYMCZAK, University of Warsaw, Physics Department, Poland — The Rotne-Prager-Yamakawa (RPY) approximation is commonly used to model the hydrodynamic interactions between small spherical particles suspended in a viscous fluid at a low Reynolds number. It takes into account long-range contribution to hydrodynamic interactions and yields positive definite diffusion matrix, which is essential for Brownian dynamics modeling. However, when the particles overlap, the RPY tensors lose their positive definiteness, which leads to numerical problems in the Brownian dynamics simulations as well as errors in calculations of the hydrodynamic properties of rigid macromolecules using bead modeling. We extend the RPY approach to the case of overlapping spherical particles of different radii in a consistent way that preserves positive definiteness of diffusion tensors for translational, rotational and dipolar degrees of freedom. Moreover we show how the Rotne–Prager–Yamakawa approximation can be generalized for other geometries and boundary conditions. 1 E.W. acknowledges the support of the Polish National Science Centre (Grant No. 2012/05/B/ST8/03010) 3:46PM D2.00008 Measuring Mechanical Properties by Staring: Using Stress Assessment from Local Structural Anisotropy (SALSA) to Probe Viscosity and Visualize Stress Networks in Colloidal Suspensions , ITAI COHEN, MATTHEW BIERBAUM, JAMES SETHNA, NEIL LIN, Cornell University — Measurement of stress induced thermal fluctuations in materials can be used to determine macroscopic mechanical properties including viscosity in fluids, as well as bulk and shear moduli in solids. When extended to the single particle scale, such measurements also reveal underlying spatially inhomogeneous response mechanisms in systems such as glasses, gels, and polycrystals. Unfortunately, it is not possible to experimentally measure these temporal and spatial stress fluctuations in a colloidal suspension using conventional rheometers. Here however, we show that using fast confocal microscopy it is possible conduct a Stress Assessment from Local Structural Anisotropy (SALSA) to measure such spatio-temporal stress fluctuations. We directly image the microstructure of a nearly hard-sphere suspension using a high-speed confocal microscope and determine particle positions. We compute the structure anisotropy of the suspension and building on the Brady formalism, calculate particle-level stresses. In conjunction with the fluctuation-dissipation theorem, we then determine the bulk viscosity of a colloidal liquid. Furthermore, we show our local measurements allow direct visualization of the complex stress networks in a 3D supercooled liquid under compression. Our method provides an experimental approach that applies to a broad range of processes arising in sheared glasses, compressed gels, and even indented crystals. 3:59PM D2.00009 Paradoxical ratcheting in cornstarch suspensions , TROY SHINBROT, THEO SIU, MATTHEW RUTALA, Rutgers University — Cornstarch suspensions are well known to exhibit strong shear thickening, and we show as a result that they must – and do – climb vertically vibrating rods and plates. This occurs because when the rod moves upward, it shears the suspension against gravity, and so the fluid stiffens, but when the rod moves downward, the suspension moves with gravity, and so the fluid is more compliant. This causes the fluid to be dragged up by the upstroke more than it is dragged down by the downstroke, effectively ratcheting the fluid up the rod every cycle. We show experimentally and computationally that this effect is paradoxically caused by gravity – and so goes away when gravity is removed – and we show that the suspension can be made to balance on the uphill side of an inclined rod in an analog of the inverted “Kapitza pendulum,” closely related to the recent report by Ramachandran & Nosonovsky, Soft Matter 10 (2014) 4633. 4:12PM D2.00010 The role of contact forces in rheology of hard-sphere colloidal suspensions , SAFA JAMALI, ARMAN BOROMAND, JOAO MAIA, Dept. of Macromolecular Science and Engineering, Case Western Reserve University — Dense colloidal suspensions show a wide range of non-Newtonian behavior in response to the flow. While at low and intermediate shear rates the fluid shear-thins, increasing the shear rate above a critical rate gives rise to microstructural changes in the fluid and consequently shear-thickening. It is widely accepted that shear-thickening of a suspension is due to the short-range hydrodynamics (so-called lubrication) forces between colloidal particles which consequently gives rise to formation of hydro clusters that resist against the flow. However, computational efforts based on lubrication theory have not been able to explain discontinuous shear-thickening in suspensions. Recently, some reports have incorporated contact potentials and their dissipative role in colloidal interactions and have successfully reproduced higher viscosity ratios at high shear rates in the shear-thickening regime. We study the effect of contact forces and lubrication potential in rheological behavior of the colloidal suspensions. To do so, we have modified Dissipative Particle Dynamics (DPD) method in order to include the lubrication potentials. Furthermore, different types of contact potentials have been included in our DPD potentials in order to understand the physical nature of contact forces and their effect on the rheology of suspensions. Finally, efforts will be made in order to correlate the microstructural changes and different types of interactions to macroscopic flow behavior of suspensions. Sunday, November 23, 2014 2:15PM - 4:12PM Session D3 Porous Media Flows II — 3004 - Christopher MacMinn, Oxford University 2:15PM D3.00001 Self-similarity in porous convection , ANJA SLIM, Monash University — In geological carbon storage, the carbon dioxide injected into a saline formation is less dense than the resident brine and floats above it. However it is also slightly soluble in brine and progressively dissolves. Brine with dissolved CO2 is slightly denser than “pure” brine and there is the potential for convective overturning. The form of this convection changes as more and more CO2 dissolves, but eventually a surprising regime is reached in which the rate at which CO2 dissolves is constant. In this regime, we find that there is no characteristic length-scale and the horizontally averaged concentration field is self-similar. I will describe features of this regime and develop a system of almost-complete effective equations that describe its evolution. 2:28PM D3.00002 Large poroelastic deformation of a soft material , CHRISTOPHER W. MACMINN, University of Oxford, ERIC R. DUFRESNE, Yale University, JOHN S. WETTLAUFER, Yale University, University of Oxford — Flow through a porous material will drive mechanical deformation when the fluid pressure becomes comparable to the stiffness of the solid skeleton. This has applications ranging from hydraulic fracture for recovery of shale gas, where fluid is injected at high pressure, to the mechanics of biological cells and tissues, where the solid skeleton is very soft. The traditional linear theory of poroelasticity captures this fluid-solid coupling by combining Darcy’s law with linear elasticity. However, linear elasticity is only volume-conservative to first order in the strain, which can become problematic when damage, plasticity, or extreme softness lead to large deformations. Here, we compare the predictions of linear poroelasticity with those of a large-deformation framework in the context of two model problems. We show that errors in volume conservation are compounded and amplified by coupling with the fluid flow, and can become important even when the deformation is small. We also illustrate these results with a laboratory experiment. 2:41PM D3.00003 Poroelastic packing and gravity currents , DUNCAN HEWITT, University of British Columbia, University of Cambridge, JAPINDER NIJJER, University of Toronto, University of Cambridge, JEROME NEUFELD, University of Cambridge — The flow of fluid through a poroelastic medium leads to deformation of the medium. We study flow in deformable media in two different contexts, in both cases undertaking laboratory experiments using small deformable hydrogel spheres. First, we present experimental results of vertical planar flow through a deformable medium driven by an imposed pressure difference. Even this simple system exhibits a complex coupling of flow and deformation. The flow-induced compaction of the medium is non-uniform with depth, and the mass flux measured in a series of experiments for different applied pressure exhibits hysteresis. We compare experimental measurements with the predictions of a simple theoretical model. Second, we consider the gravity-driven flow of fluid injected into a poroelastic medium, where flow-induced deformation leads to uplift of the surface of the medium. In the context of geological CO2 sequestration, uplift of the surface has been observed above CO2 injection sites. We develop a shallow-layer theory which describes both the flow of the injected current and the associated uplift of the surface of the medium. We also compare measurements from laboratory experiments of injection into a poroelastic medium with predictions from the theoretical model. 2:54PM D3.00004 Injection and leakage of fluids in confined porous media , SAM PEGLER, HERBERT HUPPERT, JEROME NEUFELD, Univ of Cambridge — We present a theoretical and experimental study of viscous gravity currents injected into porous media confined vertically by horizontal impermeable boundaries and saturated by a fluid of different density and viscosity. With two-dimensional flow injected at a constant volumetric rate, the pressure gradient introduced by the injection shapes the interface towards a concave similarity solution in which gravity is negligible and the interface grows in proportion to time. Data from a new series of laboratory experiments confirm our theoretical predictions over a range of viscosity ratios. We proceed to consider situations in which the current can leak through a localized “fracture,” at a rate which depends both on the gravitational head of the current below the fracture and the pressure introduced by injection. Confinement constrains the vertical growth of the current and implies a maximum possible rate of leakage. Consequently, two different flow regimes can arise, depending on whether the injection rate exceeds that maximum. If it does, then the proportion of injected fluid retained in the medium can be orders of magnitude greater than has been proposed previously from studies of unconfined aquifers with otherwise identical flow properties. 3:07PM D3.00005 Wettability control on fluid-fluid displacements in patterned microfluidics and porous media , RUBEN JUANES, MATHIAS TROJER, BENZHONG ZHAO, Massachusetts Institute of Technology — While it is well known that the wetting properties are critical in two-phase flows in porous media, the effect of wettability on fluid displacement continues to challenge our microscopic and macroscopic descriptions. Here we study this problem experimentally, starting with the classic experiment of two-phase flow in a capillary tube. We image the shape of the meniscus and measure the associated capillary pressure for a wide range of capillary numbers. We synthesize new observations on the dependence of the dynamic capillary pressure on wetting properties (contact angle) and flow conditions (viscosity contrast and capillary number). We then conduct experiments on a planar microfluidic device patterned with vertical posts. We track the evolution of the fluid-fluid interface and elucidate the impact of wetting on the cooperative nature of fluid displacement during pore invasion events. We use the insights gained from the capillary tube and patterned microfluidics experiments to elucidate the effect of wetting properties on viscous fingering and capillary fingering in a Hele-Shaw cell filled with glass beads, where we observe a contact-angle-dependent stabilizing behavior for the emerging flow instabilities, as the system transitions from drainage to imbibition. 3:20PM D3.00006 The transient spreading flow history and liquid distribution within porous medium: a wettability study , BOJAN MARKICEVIC, PALL Corp. — The interpretation of the capillary pressure at the fluid free interface as fluid potential stipulates that the wetting liquid potential is negative, and flow into porous medium is spontaneous. On the other hand, for a non-wetting flow to take place, an external force needs to be applied. Porous media are heterogeneous materials, which causes the local differences in the liquid potential and irregular liquid free interface irrespective of liquid/solid wettability. However, it is not only the capillary force that causes instabilities; even for a neutral liquid, irregularities are present due to the volumetric factor of pores of different sizes. In the present study, a neutral fluid is used as a referent value, and changes of the free interface shape for gradually increasing wetting and non-wetting interactions are determined numerically. It is shown that the interface instability is higher for the non-wetting liquid and the interface thickness is an asymmetric function of wetting angle with a minimum for the neutral fluid. Clearly, this asymmetry is influenced by pore volume, local flow resistance and capillary pressure, where for the wetting fluid, the fully saturated porous medium emerges earlier compared to its non-wetting counterpart. All three contributions lump into the changes in the flow rate, where it has been shown that by varying the external force onto the system, the interface instability can be reverted and interface stabilizes as the flow rate – and capillary number – increases. This implies that for the media whose wettability is altered, the external pressure needs to be changed in order to maintain the similar liquid distribution. 3:33PM D3.00007 Suppression of the Saffman-Taylor instability through injection of a finite slug of polymer1 , TIMOTHY H. BEESON-JONES, ANDREW W. WOODS, BPI Institute, University of Cambridge, UK — During secondary oil recovery, relatively mobile water can channel through oil owing to the Saffman-Taylor instability. Injection of a finite slug of polymer solution from a central well prior to the water flood suppresses the growth of the instability by reducing the adverse mobility ratio at the leading interface. A linear stability analysis of an axisymmetric base state identifies how perturbations on the leading and trailing interfaces become coupled. It also reveals the dependence of the long-time algebraic growth of each mode on the mobility ratios across the two interfaces. The viscosity of the polymer solution which minimizes the growth rate of the instability is identified, and the impact of different slug sizes on this growth is described. 1 Funded by EPSRC & BP. 3:46PM D3.00008 Direct Numerical Simulation of pore scale flow and reactive transport of CO2 in saline aquifers , MOHAMMAD ALIZADEH NOMELI, AMIR RIAZ, University of Maryland — A long-term geochemical modeling of subsurface CO2 storage is carried out in a single fracture to investigate its impact on CO2 transport and storage capacity. We model the fracture by considering flow of CO2 between finite plates. CO2 is initially dissolved in the brine and then precipitates during the geochemical reactions between H2 O-CO2 and minerals. We study the physics and the critical time of blockage for a fracture to interpret the results. We employ direct numerical simulation tools and algorithms to simulate incompressible flow along with necessary transport equations that capture the kinetics of relevant chemical reactions. The numerical model is based on a finite difference method using a sequential non-iterative approach. It is found that mineral precipitation has an important effect on reservoir porosity and permeability. The fracture ceases to be a fluid channel because of the precipitation of minerals. In addition, using parameter analysis we also determine the effect of various mineral precipitates on porosity of fractures. 3:59PM D3.00009 Fluid-driven fracture of elastic reservoirs followed by viscous backflow , CHING-YAO LAI, ZHONG ZHENG, EMILIE DRESSAIRE, HOWARD STONE, Department of Mechanical and Aerospace Engineering, Princeton University — We developed a laboratory scale experiment to study the physical mechanisms of fluid driven fracture and viscous backflow from elastic reservoirs. When pressurized fluid was injected into a gelatin reservoir, which is elastic but brittle, the fracture grows along an almost continuous plane and forms a fluid-filled disc-like shape. Once the injected fluid is exposed to the atmospheric pressure, the elastic relaxation of the reservoir drives the fluid flows backwards towards the original source. We study the back flow process, e.g. volume recovered as a function of time, as a function of experimental parameters such as injection volume, reservoir elasticity, and fluid viscosity. Scaling arguments are provided to explain the experimental results, which provide insights into the underlying physics of hydraulic fracturing. Sunday, November 23, 2014 2:15PM - 4:25PM Session D4 Bubbles: Acoustics and Cavitation — 3006 - Tadd Truscott, Bringham Young University 2:15PM D4.00001 Experimental study on the manipulation of microbubbles using ultrasound field , SHU TAKAGI, TAICHI OSAKI, TAKASHI AZUMA, The University of Tokyo, MITSUHISA ICHIYANAGI, Sophia University, YOICHIRO MATSUMOTO, The University of Tokyo — Non-contact manipulation techniques of microbubbles are developed by controlling the ultrasound filed. A plane standing wave-type, a ring-type and the focused-type ultrasound are used to manipulate microbubbles. It is shown that Primary Bjarknes force is well-utilized to control the position of microbubble. Microbubble clusters are observed in the actual experiments and they show the complicated behaviors as bubble clusters. These behaviors are discussed through the comparison of the experimental observation and theoretical estimation. It is experimentally shown that the size of bubble clusters gradually increases during the irradiation period of ultrasound. These clusters are captured in the central region of focused ultrasound. These clusters, however, suddenly disappear beyond the certain critical size. This type of phenomena will be discussed in the presentation. 2:28PM D4.00002 Hydrodynamic Forces on Microbubbles under Ultrasound Excitation1 , ALICIA CLARK, ALBERTO ALISEDA, University of Washington — Ultrasound (US) pressure waves exert a force on microbubbles that can be used to steer them in a flow. To control the motion of microbubbles under ultrasonic excitation, the coupling between the volume oscillations induced by the ultrasound pressure and the hydrodynamic forces needs to be well understood. We present experimental results for the motion of small, coated microbubbles, with similar sizes and physico-chemical properties as clinically-available ultrasound contrast agents (UCAs). The size distribution for the bubbles, resulting from the in-house manufacturing process, was characterized by analysis of high magnification microscopic images and determined to be bimodal. More than 99% of the volume is contained in microbubbles less than 10 microns in diameter, the size of a red blood cell. The motion of the microbubbles in a pulsatile flow, at different Reynolds and Womersley numbers, is studied from tracking of high-speed shadowgraphy. The influence of ultrasound forcing, at or near the resonant frequency of the bubbles, on the hydrodynamic forces due to the pulsatile flow is determined from the experimental measurements of the trajectories. Previous evidence of a sign reversal in Saffman lift is the focus of particular attention, as this is frequently the only hydrodynamic force acting in the direction perpendicular to the flow pathlines. Application of the understanding of this physical phenomenon to targeted drug delivery is analyzed in terms of the transport of the microbubbles. 1 NSF GRFP 2:41PM D4.00003 Algal cell disruption using microbubbles to localize ultrasonic energy for biofuel extraction , JOEL KREHBIEL, LANCE SCH, DANIEL KING, JONATHAN FREUND, Univ of Illinois - Urbana — Cell disruption is a critical step in the production of algal-based biofuels, but current mechanical disruption methods require significant energy, typically more than actually available in the cell’s oil. We propose and investigate an ultrasound disruption process using ultrasound contrast agents to localize the delivered energy. Experiments in a flow cell with focused ultrasound show a significant benefit. The degree of disruption increases with increasing peak rarefactional ultrasound pressure for pressures between 1.90 and 3.07 MPa and increasing microbubble concentration up to 12.5 × 107 bubbles/ml. Estimates suggest the energy of this method is less than one fourth of the energy of other industrial mechanical disruption techniques and comparable with theoretical disruption estimates. The increase in efficiency would make this technique viable for bioenergy applications. 2:54PM D4.00004 Theoretical Study on Propagation of Pressure Wave in a Rectangular Duct Containing Spherical Bubbles1 , JUNYA KAWAHARA, KAZUMICHI KOBAYASHI, MASAO WATANABE, Division of Mechanical and Space Engineering, Hokkaido University — Pressure waves propagating in bubbly liquids are affected by the motions of bubbles. Several mathematical models for bubbly liquids have been proposed in order to investigate the acoustic characteristics of bubble cloud. The models are classified into two types: one is the continuum model [e.g., van Wijngaarden, J. Fluid Mech. 33, 465-474 (1968)] and the other is the discrete model [e.g., Fujikawa and Takahira, Acustica 61, 188-199 (1986)]. The continuum model composed of the averaged equations for bubbly liquids treats the macroscopic behavior of bubble cloud. In contrary, the discrete model describes the motion of bubbles individually taking account of bubble/bubble interactions. The aim of our study is to investigate the validity of the continuum model by treating each bubble with the interaction. The present work theoretically investigates the propagation of pressure waves in a rectangular duct containing spherical bubbles with the discrete model [Takahira et al., JSME Int. J. Ser. B 38, 432-439 (1995)]. The results show that the propagation velocities of pressure waves obtained from the present study agree well with those obtained from the continuum models. 1 This work was supported by JSPS KAKENHI Grant Number 261417. 3:07PM D4.00005 Mass transfer effects in linear wave propagation through bubbly liquids1 , DANIEL FUSTER, Institut D’Alembert CNRS-UPMC — In this work we present a model to capture the influence of mass transfer effects on the effective acoustic properties of bubbly liquids. The solution of the conservation equations inside the bubble (e.g. continuity, momentum, energy and species) is coupled to the solution of the conservation equations in the liquid surrounding the bubble using local balances across the interface and a linearized version of the mass transfer flux obtained from the Hertz-Knudsen-Langmuir formula. The model is able to capture the transition from gas bubbles containing a non-soluble gas to gas/vapor bubbles with a given vapor/gas ratio. In addition to the influence of the enthalpy of vaporization, the velocity jump appearing at the interface is shown to have a significant influence in both, the effective phase velocity and the attenuation of the medium near the saturation line. The validity of common assumptions typically used in simplified models and limiting solutions obtained from the current approach are discussed in terms of characteristic non-dimensional numbers. Consistent with previously published data, the influence of mass transfer effect is specially notorious at low frequencies. 1 The author would like to acknowledge the support of Total R&D, and specially to F. Montel and C. Boehm for their fruitful discussions. 3:20PM D4.00006 Velocimetry in both phases of a cavitating flow by fast X-ray imaging , OLIVIER COUTIER-DELGOSHA, ILYASS KHLIFA, SYLVIE FUZIER, LML Laboratory, ALEXANDRE VABRE, CEA, KAMEL FEZZAA, Argonne National Laboratory, HOCEVAR MARKO, Ljubljana University — A promising method to measure velocity fields in a cavitating flow is presented. Dynamics of the liquid phase and of the bubbles are both investigated. The measurements are based on ultra fast X-ray imaging performed at the APS (Advanced Photon Source) of the Argonne National Laboratory. The experimental device consists of a millimetric Venturi test section associated with a transportable hydraulic loop. Various configurations of velocity, pressure, and temperature have been investigated. Radio-opaque particles are used as tracers for the liquid phase, in association with a multi-pixels sensor to record the successive positions of the particles. The use of X-rays instead of light solves the problem of light reflection and dispersion on phase boundaries, since X-rays penetrate a gas/liquid flow in straight lines. Images contain simultaneously the information related to the particles (for PIV analysis in the liquid), to the vapor bubbles (for PIV in the gas). The slip velocity between vapor and liquid is calculated everywhere both velocities can be obtained. 3:33PM D4.00007 Cavitation you can hold in your hand... for a moment , DAVID JESSE DAILY, Electronic Consulting Services, JONATHON PENDLEBURY, Brigham Young University, KENNETH LANGLEY, Hill Air Force Base, TADD TRUSCOTT, Brigham Young University — In a popular party trick a glass bottle is filled with water and firmly struck at the top, breaking the bottle with nothing but bare hands. We present evidence that this trick is caused by cavitation formed by the acceleration of the fluid. Traditional velocity based methods for determining cavity formation do not successfully predict cavitation onset, however, a dimensionless cavitation equation derived from the Navier-Stokes equation predicts cavitation as a function of pressure head and acceleration. Our experiments utilized accelerometers and high-speed photography to observe cavitation with good agreement between experiments and predictions. Elucidating the onset of cavitation based on these simple parameters will help those who attempt this trick appreciate the physical complexity of this phenomenon and improve their bottle breaking skills. 3:46PM D4.00008 Cavitation stuctures formed during the collision of a sphere with an ultraviscous wetted surface , MOHAMMAD MANSOOR, King Abdullah University of Science and Technology, JEREMY MARSTON, Texas Tech University, JAMAL UDDIN, University of Birmingham, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology — We investigate the inception of cavitation and associated structures when a sphere collides with a solid surface covered with a layer of ultra-viscous non-Newtonian liquid with kinematic viscosities, v, of up to 20 million cSt (at nominal low shear). Using a synchronized dual-view high-speed imaging system, we confirm that there is no shear-induced cavitation even in such highly favorable conditions. We show that liquids with high visco-elastic properties can enable the sphere to rebound without any prior contact with the solid wall. A decrease in sphere impact velocity for such non-contact rebound cases results in a systematic delay in cavity inception by depressurization from the time of achieving the minimum gap distance. We find vastly different bubble entrapment characteristics on the sphere surface during entry into the liquid layer for low and high-viscosity liquids. These were found to play an important role in the formation of cavitation structures in non-contact cases. In contrast, when contact occurs, we observe a cylindrical structure attached to the wall having undulations along the cavity interface which were further investigated using high-speed particle image velocimetry (PIV) techniques. 3:59PM D4.00009 Discussion on the Applicability of Rayleigh-Plesset Equation for a Nanoscale bubble using Molecular Dynamics Simulation , SHIN-ICHI TSUDA, Kyushu Univ, KAZUKI OGASAWARA, TAKUMI ITAKURA, Shinshu Univ — Multi-phase flows such as cavitation and boiling have much variety on the scale in time and space compared with single phase flows. It is necessary to recognize the multi-scale structure accurately to construct a sophisticated numerical method for the prediction of various multi-phase flow phenomena. In this point of view, clarification of the valid range of continuum mechanics would be very important. Here, an interesting problem in the case of cavitation is, to what extent Rayleigh-Plesset (R-P) equation, which describes the radius change of a spherical bubble under a pressure given at far from the bubble, can express the behavior of a tiny bubble quantitatively. In this work, we discussed the validity of the application of R-P equation to a nano-scale bubble using Molecular Dynamics (MD) simulation. In the simulation, liquid argon at a decompressed state in a cubic domain was simulated. As a result, a nano-scale bubble was generated after a waiting time, and it rapidly grew to several nanometers, and it reached to an equilibrium state showing a transient behavior. We compared the bubble radius change observed in the MD simulation with the numerical result of R-P equation, and confirmed that R-P equation can well predict the behavior of such tiny bubble. 4:12PM D4.00010 Supersonic microjets induced by hemispherical cavitation bubbles , ROBERTO GONZALEZ-AVILA, CHAOLONG SONG, CLAUS-DIETER OHL, Nanyang Tech Univ — In recent years methods to produce fast microjets have received significant attention due to their potential to be employed in needle-free injection devices that can provide mass inoculation. In this talk we present a novel technique capable of producing jets that can reach up to 400 m/s. The jets are produced by a device that consists only of two electrodes on a plastic substrate and a tapered hole of 13 ∼ 20 µm between them. A short pulse of electric current is applied to the electrodes, then a spark bridges between the electrodes creating a cavitation bubble. Liquid is accelerated through the hole during the expansion and collapse of the bubble producing two separate jets. We found that as the exit velocity of the jet increases the jets become unstable. The second jet exiting the hole, usually faster than the first jet exits as a spray. The effect of viscosity was also studied with silicone oils up to 100 cSt. Finally, we also demonstrate that the jets can penetrate into soft material, thus they have the potential to be used in a needle-free drug-delivery application. Sunday, November 23, 2014 2:15PM - 4:25PM Session D5 Biofluids: Red Blood Cells — 3008 - Michael Plesniak, George Washington University 2:15PM D5.00001 Quantitative imaging of RBC suspensions in bifurcating microchannels , JOSEPH SHERWOOD1 , Imperial College London, DAVID HOLMES, Sphere Fluidics Limited, EFSTATHIOS KALIVIOTIS2 , STAVROULA BALABANI3 , University College London — The local velocity and concentration characteristics of both red blood cells (RBCs) and suspending medium flowing in a bifurcating microchannel were measured simultaneously. An imaging technique involving alternate bright field and laser light illumination was employed to capture both RBC and fluorescent PIV images of human healthy blood, flowing through a sequentially bifurcating 50 micrometer square PDMS microchannel. The acquired images were further processed using PIV algorithms to yield the velocity distribution of RBCs and suspending medium while the brightfield images also provided data on hematocrit distribution and cell-depleted layer. Various flow rates, aggregation states and proportions of flow entering each branch were considered. Asymmetric hematocrit distributions were quantified around the bifurcations and found to be enhanced by aggregation. The data were compared with computational fluid dynamics studies of continuous Newtonian and Non-Newtonian fluids in order to elucidate the impact of the two-phase nature of the flow, particularly RBC aggregation. The work is currently being extended to examine the role of RBC properties on microhemodynamics and the implications for disease. 1 Department of Bioengineering of Mechanical Engineering 3 Department of Mechanical Engineering 2 Department 2:28PM D5.00002 Red blood cell dynamics under high shear rates: in vitro experimental investigations , LUCA LANOTTE, CYRILLE CLAUDET, Laboratoire Charles Coulomb, Université Montpellier 2 (France) - Centre de Biochimie Structurale, CNRS (France), JEAN-MARC FROMENTAL, Laboratoire Charles Coulomb, Université Montpellier 2 (France), MANOUK ABKARIAN, Laboratoire Charles Coulomb, Université Montpellier 2 (France) - Centre de Biochimie Structurale, CNRS (France) — The full understanding of red blood cell (RBC) dynamics is an intriguing challenge that involves transversal branches of science. Despite the potential impact that it could have on medical research and industrial applications, a systematic study of RBCs response under significant shear rates (200<γ̇<3000 s−1 ) is still lacking in scientific literature. In this work, in vitro experiments of microfluidics and rheometric measurements are combined to investigate mechanical properties of highly sheared RBCs. By high-speed microscopy, we investigated RBCs flow through rectangular channels in unconfined conditions. In parallel, RBCs suspensions of different hematocrits have been processed by a cone-plate rheometer and subsequently observed by optical microscopy to ensure reliability to the experimental results. The outcomes of both microfluidics and rheological approaches clearly show the presence of strongly deformed shapes, in addition to the expected elongated ellipsoids. Plausible explanations for formation and stability of these striking highly deformed shapes are here proposed. 2:41PM D5.00003 Influence of red blood cell clustering on phase separation in capillary networks , THOMAS PODGORSKI, CELINE BOUCLY, GWENNOU COUPIER, LIPhy, CNRS-UJF Grenoble — We investigate the flow of red blood cell suspensions in microfluidic bifurcations and capillary networks. At strong degrees of confinement, such as those encountered in the microcirculation, phase separation takes place at bifurcations of the network, leading to strong heterogeneities and fluctuations of the hematocrit (blood cell concentration). We highlight the influence of the mechanical properties of cells : an increase of membrane or cytoplasm rigidity, as can happen in pathologies such as sickle cell disease tends to reduce the phase separation. The influence of the attractive interaction between cells, that leads to clustering (rouleau formation) was also investigated by varying the concentration of macromolecules in the solution (dextran or fibrinogen). We show that hydrodynamic stresses in bifurcations can lead to rupture of clusters at a critical speed which increases with interaction energy. Overall, the clustering phenomenon tends to increase phase separation and hematocrit heterogeneities. 2:54PM D5.00004 Effect of viscoelasticity and RBC migration phenomena in stenotic microvessels , YIANNIS DIMAKOPOULOS, Laboratory of Fluid Mechanics and Rheology, Dep. of Chemical Engineering, University of Patras, ALEXANDROS SYRAKOS, GEORGIOS GEORGIOU, Department of Mathematics, University of Cyprus, JOHN TSAMOPOULOS, Laboratory of Fluid Mechanics and Rheology, Dep. of Chemical Engineering, University of Patras — This study deals with the numerical simulation of the hemodynamics in stenotic microvessels. The blood flow in microvessels differs significantly from that in large arteries and veins, because the Red Blood Cells (RBCs) are comparable in size with the radius of the microvessels and, consequently, local effects such as cell interaction and migration are more pronounced. In terms of complexity of the flow, viscoelasticity along with stress-gradient induced migration effects have a more dominant role, which exceeds the viscous, inertial and transient effects. Recently, a non-homogeneous viscoelastic model has been proposed by Moyers-Gonzalez et al. (2008), which can accurately predict the Fahraeus effects. We developed a numerical algorithm for the time-integration of the set of differential equations that arise from the coupling of momentum, mass, and population balances for RBCs and aggregates with the constitutive laws for both species. The simulations show that a cell-depleted layer develops along the vessel wall with an almost constant thickness. Along this layer, the shear stresses are almost Newtonian because of the plasma, but the normal stresses that are exerted on the wall are high due to the contribution of the individual RBCs and rouleaux. 3:07PM D5.00005 Red Blood Cell Hematocrit Influences Platelet Adhesion Rate in a Microchannel1 , ANDREW SPANN, Stanford University, JAMES CAMPBELL, United States Army Institute of Surgical Research, SEAN FITZGIB- BON, Stanford University, ARMANDO RODRIGUEZ, United States Army Institute of Surgical Research, ERIC SHAQFEH, Stanford University — The creation of a blood clot to stop bleeding involves platelets forming a plug at the site of injury. Red blood cells indirectly play a role in ensuring that the distribution of platelets across the height of the channel is not uniform – the contrast in deformability and size between platelets and red blood cells allows the platelets to preferentially marginate close to the walls. We perform 3D boundary integral simulations of a suspension of platelets and red blood cells in a periodic channel with a model that allows for platelet binding at the walls. The relative rate of platelet activity with varying hematocrit (volume fraction of red blood cells) is compared to experiments in which red blood cells and platelets flow through a channel coated with von Willebrand factor. In the simulations as well as the experiments, a decrease in hematocrit of red blood cells is found to reduce the rate at which platelets adhere to the channel wall in a manner that is both qualitatively and quantitatively similar. We conclude with a discussion of the tumbling and wobbling motions of platelets in 3D leading up to the time at which the platelets bind to the wall. 1 Funded by Stanford Army High Performance Computing Research Center, experiments by US Army Institute of Surgical Research 3:20PM D5.00006 Twisting of Red Blood Cells Entering a Constriction , NANCY ZENG, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science — Most work on the dynamic response of red blood cells (RBCs) to hydrodynamic stress has focused on linear velocity profiles. Relatively little experimental work has examined how individual RBCs respond to pressure driven flow in more complex geometries, such as the flow at the entrance of a capillary. Here, we establish the mechanical behaviors of healthy RBCs undergoing a sudden increase in shear stress at the entrance of a narrow constriction. We pumped RBCs through a constriction in an ex vivo microfluidic device and used high speed video to visualize and track the flow behavior of more than 4,400 RBCs. We show that approximately 85% of RBCs undergo one of four distinct modes of motion: stretching, twisting, tumbling, or rolling. Intriguingly, a plurality of cells (∼30%) exhibited twisting (rotation around the major axis parallel to the flow direction), a mechanical behavior that is not typically observed in linear velocity profiles. We examine the mechanical origin of twisting using, as a limiting case, the equations of motion for rigid ellipsoids, and we demonstrate that the observed rotation is qualitatively consistent with rigid body theory. 3:33PM D5.00007 Intermittency and Synchronized Tumbling and Tank-treading in Red Blood Cell Dynamics in Steady and Oscillatory Shear Flows1 , PROSENJIT BAGCHI, DANIEL CORDASCO, Rutgers University — Red blood cells are known to exhibit a variety of rich and complex dynamics when subjected to a shear flow. Of particular interest is the intermittent behavior that is characterized by coexistence of the tumbling motion, and the tank-treading motion. Several reduced-order theoretical models assuming fixed cell shape emerged that either supported or rejected the possibility of such dynamics, although no full-scale computer simulation of deformable cells has conclusively observed such dynamics. Here we present the first computational evidence of intermittent dynamics of red blood cells in steady and oscillatory shear flows. Our model fully resolves the cell deformation taking in to consideration all essential properties of the cell membrane and internal fluid, and hence, contradicts the notion that intermittency is suppressed in deformable cells. For the intermittent dynamics, we observe sequences of tumbling interrupted by swinging, as well as sequences of swinging interrupted by tumbling. In the synchronized dynamics, the tumbling and membrane rotation occur simultaneously with integer ratio of rotational frequencies. These dynamics are shown to be dependent on the stress-free state of the cytoskeleton, and are explained based on the cell membrane energy landscape. 1 Supported by NSF 3:46PM D5.00008 Effect of Strain Rate on the Mechanical Behavior of Red Blood Cells Entering a Constriction , JORDAN MANCUSO, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — Most work on the effect of hydrodynamic stress on red blood cells (RBCs) has focused on linear velocity profiles. Microfluidic devices have provided a means to examine the mechanical behavior of RBCs undergoing a sudden increase in shear stress at the entrance of a constriction, with previous work primarily focused on a fixed constriction taper angle and corresponding hydrodynamic strain rate. Here we investigate the effect of strain rate on the stretching dynamics exhibited by RBCs as they enter a microfluidic constriction. Systematic variations in the constriction taper angle allow the strain rate to be precisely tuned, and high speed video yields precise measurements of the corresponding RBC deformations. We demonstrate that maximal RBC stretching occurs at an intermediate constriction taper angle, despite the lower magnitude of the strain rate. We interpret the results in terms of the time integral of the elongational strain rate, and we discuss the implications for shear-induced mechanotransduction. 3:59PM D5.00009 Red Blood Cell Dispersion in Morphologically-Inspired Microfluidic Models of Alveolar Capillary Networks , HAGIT STAUBER, RAMI FISHLER, Technion-IIT, DAN WAISMAN, Department of Neonatology Carmel Medical Center Faculty of Medicine - Technion IIT, JOSUE SZNITMAN, Technion-IIT — Microfluidics is frequently used to study blood flow characteristics in microcapillary networks and investigate transport properties of red blood cells (RBC). To date, most of microfluidic studies have not focused on the specific morphology of alveolar capillary networks (ACN), with characteristic length scales of ∼ 5 µm, known to give rise to organ-specific blood flow characteristics. To better understand flow characteristics and dispersion of RBCs in ACNs, we have designed morphologically-inspired microfluidic models of alveolar capillary beds at a real scale. We fabricate lab-on-chip devices featuring confined staggered pillar arrays with diameters of ∼ 10 µm, representative of the dense ACN capillary meshes. Devices are supplied by an external reservoir containing whole blood at various hematocrit levels, to mimic RBC perfusion (Re<0.01) within alveolar capillaries. Whole-field velocity patterns are imaged (PIV) and RBC motion is tracked using particle tracking velocimetry (PTV) from which dispersion coefficients are extracted. Our efforts are aimed at delivering a real-scale quantitative description of the pulmonary ACN microcirculation. 4:12PM D5.00010 Experimental comparison of mammal, fish and bird blood flow in microchannels , KATHRYN FINK, UC Berkeley - UC San Francisco Graduate Program in Bioengineering, KARTHIK PRASAD, DORIAN LIEPMANN, University of California, Berkeley — The non-Newtonian, shear rate dependent behavior of blood in microchannel fluid dynamics has been studied for nearly a century, with a significant focus on the characteristics of human blood. However, for over 200 years biologists have observed significant variation in the size, shape, deformability and aggregation behavior of red blood cells across vertebrate species. With a few exceptions, mammals have denucleated biconcave red blood cells; birds and fish have nucleated ovoid red blood cells with size variations of a full order of magnitude. We present an experimental analysis of flow in long (500-1000 channel diameters) polycarbonate microchannels with a variety of blood samples. Correlation of shear rate and viscosity is compared to existing constitutive equations for human blood to further quantify the importance of red blood cell characteristics. Sunday, November 23, 2014 2:15PM - 4:25PM Session D6 Biofluids: Active Fluids II — 3010 - David Saintillan, University of California, San Diego 2:15PM D6.00001 Swim pressure of active matter , SHO TAKATORI, WEN YAN, JOHN BRADY, California Institute of Technology, CALTECH TEAM — Through their self-motion, all active matter systems generate a unique “swim pressure” that is entirely athermal in origin. This new source for the active stress exists at all scales in both living and nonliving active systems, and also applies to larger organisms where inertia is important (i.e., the Stokes number is not small). Here we explain the origin of the swim stress and develop a simple thermodynamic model to study the self-assembly and phase separation in active soft matter. Our new swim stress perspective can help analyze and exploit a wide class of active soft matter, from swimming bacteria and catalytic nanobots, schools of fish and birds, and molecular motors that activate the cellular cytoskeleton. 2:28PM D6.00002 Hydrodynamic interactions in dilute suspensions of microswimmers , JOAKIM STENHAMMAR, RUPERT NASH, DAVIDE MARENDUZZO, ALEXANDER MOROZOV, University of Edinburgh — We present a numerical method based on a Lattice-Boltzmann algorithm to simulate hydrodynamic interactions between a large number of model swimmers (order 105 ), modelled as extended force dipoles. Similar to previous studies of this problem, both experimental and theoretical, we observe that, depending on the concentration of microswimmers, there exists a transition to large-scale structures, often referred to as bacterial turbulence. We introduce a simple theory to characterize the onset of this transition and compare it to our observations. We will also present results on the influence of the large-scale structures on the enhanced diffusion of tracer particles suspended in a solution of microswimmers. 2:41PM D6.00003 A numerical model of localized convection cells of Euglena suspensions1 , MAKOTO IIMA, ERIKA SHOJI, TAKAYUKI YAMAGUCHI, Hiroshima University — Suspension of Euglena gracilis shows localized convection cells when it is illuminated form below with strong light intensity. Experiments in an annular container shows that there are two elementary localized structures. One consists of a pair of convection cells and a single region where number density of Euglena is high. The other consists a localized traveling wave [1]. Based on the measurements of the flux of number density, we propose a model of bioconvection incorporating lateral phototaxis effect proportional to the light intensity gradient. Using pseudo spectral method, we performed numerical simulation of this model. We succeed in reproducing one of the localized structures, a convection pair with single region of high number density. Also, when the aspect ratio is large, there are a parameter region where the localized structure and conductive state are both stable, which is suggested by experiments [1]. Spatial distribution of the number density implies that the accumulation of microorganism due to the convective flow causes such bistability. [1] Localized bioconvection patterns and their initial state dependency in Euglena suspensions in an annular container, E. Shoji, H. Nishimori, A. Awazu, S. Izumi, and M. Iima, J. Phys. Soc. Jpn. 83(2014)04300 1 CREST(PJ74100011) and KAKENHI(26400396) 2:54PM D6.00004 Effect of hydrodynamic interactions in confined active suspensions , BARATH EZHILAN, DAVID SAINTILLAN, Department of Mechanical and Aerospace Engineering, University of California San Diego — The dynamics of biologically active suspensions in confined geometries is investigated by incorporating accurate boundary conditions within the kinetic theory framework [Saintillan and Shelley, Phys. Fluids. (2008)]. Even in the absence of wall hydrodynamic interactions or imposed flow, swimming microorganisms have a tendency to accumulate at confining boundaries due to self-propulsion. Satisfying a zero wall-normal translational flux condition on the active particle probability distribution function captures this effect. Using a moment-closure approximation, analytical expressions for the equilibrium concentration/polarization profiles are derived in the dilute limit. As particle density increases, we expect particle-particle hydrodynamic interactions to become significant and to destabilize these equilibrium distributions. Using a linear stability analysis and 3D finite volume simulation of the equations for the orientational moments, we study in detail the effect of fluid coupling on the stability properties of the equilibrium states in confined active suspensions. 3:07PM D6.00005 Phase transitions in dense active suspensions , SAM MATTHEW, PALLAB SIMHAMAHAPATRA, SRIKANTH VEDANTAM, MAHESH PANCHAGNULA, Indian Institute of Technology Madras — We study dense suspensions of active particles embedded in a Newtonian fluid medium using discrete element computations. The particles are modeled as soft spheres capable of generating a thrust oriented in the direction of its instantaneous velocity. The embedding fluid provides viscous drag in the Stokes regime. The dynamics of the active suspension are investigated in a square cavity. Simulations of dilute suspensions show classical clustering and collective motion. In dense suspensions, the ratio of the thrust to drag force (denoted by λ) is found to be an important dimensionless parameter governing the system dynamics. Phase transitions in this material are investigated in this parameter space. It was observed that for low values of λ, the material arranges itself an oscillatory modes. At intermediate values of λ, the oscillatory modes transition to a single steady vortex. At higher λ, multiple vortices are observed in the computational domain. At very high λ, diffusive effects dominate and a gas-like phase is observed. All the transitions occur over small changes in λ indicating sharp transitions between the phases. This model system shows multiple phase transitions driven by a single parameter. 3:20PM D6.00006 Density fluctuations and topological structures in collective surface motion of microswimmers , TONG GAO, MICHAEL SHELLEY, Courant Institute of Mathematical Sciences — Active matter that consists of self-propelled particles, such as bacterial suspensions and assays of self-driven biofilaments, can exhibit collective motions with large-scale complex flows and topological defect dynamics. Using a Doi-Onsager kinetic theory, we study suspensions of microswimmers confined to an air/liquid interface, and identify correlations between particle density fluctuations, defect structures, nematic order, and surface flows. When considering a free-standing liquid film where the microswimmers are distributed on the air/liquid interfaces, we capture hydrodynamic coupling of the two active surface, characterized by synchronization of motile disclination defects. We estimate the effective “penetration distance” between the two coupled surfaces through a linear stability analysis. 3:33PM D6.00007 Continuum Level Results from Particle Simulations of Active Suspensions1 , BLAISE DELMOTTE, ERIC CLIMENT, Institut de Mécanique des Fluides de Toulouse - Université de Toulouse, FRANCK PLOURABOUE, Institut de Mécanique des Fluides de Toulouse - CNRS, ERIC KEAVENY, Imperial College — Accurately simulating active suspensions on the lab scale is a technical challenge. It requires considering large numbers of interacting swimmers with well described hydrodynamics in order to obtain representative and reliable statistics of suspension properties. We have developed a computationally scalable model based on an extension of the Force Coupling Method (FCM) to active particles. This tool can handle the many-body hydrodynamic interactions between O(105 ) swimmers while also accounting for finite-size effects, steady or time-dependent strokes, or variable swimmer aspect ratio. Results from our simulations of steady-stroke microswimmer suspensions coincide with those given by continuum models, but, in certain cases, we observe collective dynamics that these models do not predict. We provide robust statistics of resulting distributions and accurately characterize the growth rates of these instabilities. In addition, we explore the effect of the time-dependent stroke on the suspension properties, comparing with those from the steady-stroke simulations. 1 Authors acknowledge the ANR project Motimo for funding and the Calmip computing centre for technical support. 3:46PM D6.00008 Instability and spinodal decomposition of chemically active suspensions , WEN YAN, JOHN BRADY, Caltech — Chemically active particles can self-propel by diffusiophoresis with velocity U = −M ∇c by changing the local solute concentration c via a surface catalytic reaction. Here, M is the particle dffusiophoretic mobility. The particle and fluid motion is such that the convection of solute can be ignored and the concentration field c is governed by Laplace’s equation. We explore the collective dynamics of active particles by both continuum theory and particle-tracking simulation. In simulation the solute concentration field is accurately resolved simultaneously with the particles’ motion by a multipole scattering method allowing the simulation of thousands of active particles. Active suspensions exhibit a Brinkman-like screening of long-range interactions which predicts an instability in the collective dynamics that scales with the volume fraction of active particles to the 1/2 power. For weak phoretic motion (small M ), the instability theory is verified by the simulations. For strong phoretic motion (large M ), the active particles show a spinodal decomposition. Transient fractal structures are identified in 3D, while individual clusters are observed in a particle monolayer. 3:59PM D6.00009 Pair Interaction of Catalytically Active Colloidal Particles , NIMA SHARIFI-MOOD, University of Pennsylvania, SERGEY SHKLYAEV, Institute of Continuous Media Mechanics, UBALDO CÓRDOVA-FIGUEROA, University of Puerto RicoMayagüez — An increasing number of experiments on catalytically-driven (active) colloidal particles have shown that the interaction of chemically active particles is more complicated than usual interaction of two nonreactive (passive) particles. Indeed, each chemically active particle changes the distribution of reactants which, in turn, generates an overall force on other particles. First, we consider a pair of spherically symmetric catalytic particles, which are far from each other, in a colloidal dispersion of reactants and products. In this case there appears a force which can be either attractive or repulsive depending on the stoichiometry factor of the reaction. In fact, the interaction force can be thought of as a force between two charged particles which can bear charges of either the same or opposite signs depending on the stoichiometry factor. Next, we deal with interaction between catalytic and passive (cargo) particles. It is demonstrated that the force on a cargo is exactly the same as the force imposed by a catalytic particle on another one. On the other hand, the force on a catalytic particle imposed by the cargo is much smaller. Within the above-mentioned electrostatic analogy, the cargo particle is equivalent to a particle of vanishing permittivity. 4:12PM D6.00010 Dynamics of micro-swimmers inside a peristaltic pump , ADAM STINCHCOMBE, University of Michigan, CHARLES PESKIN, New York University, ENKELEIDA LUSHI, Brown University — Peristaltic pumping is a form of fluid transport along the length of a tube containing liquid when the tube undergoes a contraction wave. While much is known about the peristalsis of Newtonian liquids, complex ones have received limited attention. There are many examples in nature where motile micro-particles or micro-swimmers (such as bacteria or spermatozoa) are suspended in the fluid inside a peristaltic micro-pump. We present a simulation method that couples the dynamics of many micro-swimmers to each-other, the pump and the fluid flow. The pump and the fluid flow it pushes can affect the swimmer dynamics in interesting ways. Moreover, the presence of the swimmers and their collective motion can affect the net transport and mixing in the pump. The efficiency of mixing abilities of the suspension for a variety of parameters will be discussed. Sunday, November 23, 2014 2:15PM - 4:25PM Session D7 Biofluids: Cardiac Flows 3012 - Iman Borazjani, Buffalo University — 2:15PM D7.00001 Quantification of avian embryonic cardiac outflow hemodynamics through 3D-0D coupling , STEPHANIE LINDSEY, Cornell University; INRIA Paris-Rocquencourt, IRENE VIGNON-CLEMENTEL, INRIA Paris-Rocquencourt, JONATHAN BUTCHER, Cornell University — Outflow malformations account for over 20% of CHDs in the US. While the etiology of these malformations is poorly understood, most can be traced back to perturbations in the patterning of the pharyngeal arch arteries (PAAs), the precursors to the great vessels. Here, we examine the effects of normal and aberrant PAA flow, through the use of two computational models. A 0D electric analog model allows for rapid computation and global tuning of the embryo’s vasculature relative to the arches. A second 3D-0D model replaces the electric analog representation of the arches with a 3D reconstruction, thereby leading to more extensive pressure and flow characterization. We obtain 3D arch artery reconstructions from nano-CT stacks and couple them to 0D outlets. In contrast to standard boundary conditions, such coupling maintains the physiologically desired cranial-caudal flow split in control embryos and predicts how this will change with vessel occlusion. We use flow inputs from Doppler velocity tracings to compute 0D and 3D-0D pulsatile hemodynamic simulations in HH18 (day 3), HH24 (day 4), and HH26 (day 5) geometries. We then calculate flow distributions and wall shear stress maps for control embryos. From here, we modify HH18 geometries to simulate varying levels of PAA occlusion. Pulsatile simulations are run in each geometry and results compared to that of controls. Results serve as a basis for examining flow-mediated growth and adaptation in cardiac outflow morphogenesis. 2:28PM D7.00002 A New Parameter for Cardiac Efficiency Analysis1 , IMAN BORAZJANI, NAVANEETHA KRISHNAN RAJAN, ZEYING SONG, KENNETH HOFFMANN, University at Buffalo SUNY, EILEEN MACMAHON, MAREK BELOHLAVEK, Mayo Clinic Arizona — Detecting and evaluating a heart with suboptimal pumping efficiency is a significant clinical goal. However, the routine parameters such as ejection fraction, quantified with current non-invasive techniques are not predictive of heart disease prognosis. Furthermore, they only represent left-ventricular (LV) ejection function and not the efficiency, which might be affected before apparent changes in the function. We propose a new parameter, called the hemodynamic efficiency (H-efficiency) and defined as the ratio of the useful to total power, for cardiac efficiency analysis. Our results indicate that the change in the shape/motion of the LV will change the pumping efficiency of the LV even if the ejection fraction is kept constant at 55% (normal value), i.e., H-efficiency can be used for suboptimal cardiac performance diagnosis. To apply H-efficiency on a patient-specific basis, we are developing a system that combines echocardiography (echo) and computational fluid dynamics (CFD) to provide the 3D pressure and velocity field to directly calculate the H-efficiency parameter. Because the method is based on clinically used 2D echo, which has faster acquisition time and lower cost relative to other imaging techniques, it can have a significant impact on a large number of patients. 1 This work is partly supported by the American Heart Association. 2:41PM D7.00003 In-vivo characterization of 2D residence time maps in the left ventricle , LORENZO ROSSINI, PABLO MARTINEZ-LEGAZPI, UC San Diego, JAVIER BERMEJO, YOLANDA BENITO, MARTA ALHAMA, RAQUEL YOTTI, CANDELAS PEREZ DEL VILLAR, ANA GONZALEZ-MANSILLA, ALICIA BARRIO, FRANCISCO FERNANDEZ-AVILES, Hospital Gregorio Maranon, Madrid, Spain, SHAWN SHADDEN, UC Berkeley, JUAN CARLOS DEL ALAMO, UC San Diego — Thrombus formation is a multifactorial process involving biology and hemodynamics. Blood stagnation and wall shear stress are linked to thrombus formation. The quantification of residence time of blood in the left ventricle (LV) is relevant for patients affected by ventricular contractility dysfunction. We use a continuum formulation to compute 2D blood residence time (TR ) maps in the LV using in-vivo 2D velocity fields in the apical long axis plane obtained from Doppler-echocardiography images of healthy and dilated hearts. The TR maps are generated integrating in time an advection-diffusion equation of a passive scalar with a time-source term. This equation represents the Eulerian translation of D TR /D t = 1 and is solved numerically with a finite volume method on a Cartesian grid using an immersed boundary for the LV wall. Changing the source term and the boundary conditions allows us to track blood transport (direct and retained flow) in the LV and the topology of early (E) and atrial (A) filling waves. This method has been validated against a Lagrangian Coherent Structures analysis, is computationally inexpensive and observer independent, making it a potential diagnostic tool in clinical settings. 2:54PM D7.00004 Clinical characterization of 2D pressure field in human left ventricles , MARIA BORJA, LORENZO ROSSINI, PABLO MARTINEZ-LEGAZPI, UC San Diego, YOLANDA BENITO, MARTA ALHAMA, RAQUEL YOTTI, CANDELAS PEREZ DEL VILLAR, ANA GONZALEZ-MANSILLA, ALICIA BARRIO, FRANCISCO FERNANDEZ-AVILES, JAVIER BERMEJO, Hospital Gregorio Maranon, Madrid, Spain, ANDREW KHAN, JUAN CARLOS DEL ALAMO, UC San Diego — The evaluation of left ventricle (LV) function in the clinical setting remains a challenge. Pressure gradient is a reliable and reproducible indicator of the LV function. We obtain 2D relative pressure field in the LV using in-vivo measurements obtained by processing Doppler-echocardiography images of healthy and dilated hearts. Exploiting mass conservation, we solve the Poisson pressure equation (PPE) dropping the time derivatives and viscous terms. The flow acceleration appears only in the boundary conditions, making our method weakly sensible to the time resolution of in-vivo acquisitions. To ensure continuity with respect to the discrete operator and grid used, a potential flow correction is applied beforehand, which gives another Poisson equation. The new incompressible velocity field ensures that the compatibility equation for the PPE is satisfied. Both Poisson equations are efficiently solved on a Cartesian grid using a multi-grid method and immersed boundary for the LV wall. The whole process is computationally inexpensive and could play a diagnostic role in the clinical assessment of LV function. 3:07PM D7.00005 Image-based flow modeling in a two-chamber model of the left heart1 , VIJAY VEDULA, JUNG-HEE SEO, KOUROSH SHOELE, RICHARD GEORGE, LAURENT YOUNES, RAJAT MITTAL, Johns Hopkins University — Computational modeling of cardiac flows has been an active topic of discussion over the past decade. Modeling approaches have been consistently improved by inclusion of additional complexities and these continue to provide new insights into the dynamics of blood flow in health and disease. The vast majority of cardiac models have been single-chamber models, which have typically focused on the left or right ventricles, and in these models, the atria are modeled in highly simplistic ways. However, the left atrium acts as a mixing chamber and works with the left ventricle in a highly coordinated fashion to move the blood from the pulmonary veins to the aorta. The flow dynamics associated with this two-chamber interaction is not well understood. In addition, the flow in the left atrium has by itself significant clinical implications and our understanding of this is far less than that of the left ventricle. In the current study, we use 4D CT to create a physiological heart model that is functionally normal and use an experimentally validated sharp-interface immersed boundary flow solver to explore the atrio-ventricular interaction and develop insights into some of the questions addressed above. 1 This research is supported by the U.S. National Science Foundation through NSF Grants IOS-1124804 and IIS-1344772. Computational resources are provided in part through the NSF XSEDE grants TG-CTS100002 and TG-CTS130064. 3:20PM D7.00006 Intrinsic Frequency Method for Noninvasive Diagnosis of Left Ventricular Systolic Dysfunction , NIEMA PAHLEVAN, DEREK RINDERKNECHT, PEYMAN TAVALLALI, DANNY PETRASEK, California Institute of Technology, RAY MATTHEWS, University of Southern California, Keck School of Medicine, MORTEZA GHARIB, California Institute of Technology — We have recently developed a new mathematical method, intrinsic frequency (IF) method, that views the left ventricle-arterial system as a coupled dynamic pumping system which is decoupled upon the closure of the aortic valve. Utilizing this method, given an arterial blood pressure waveform we are able to extract two intrinsic frequencies (ω1 and ω2 ) correlating to systole when the left ventricle (LV) and aorta (vasculature) act as a coupled dynamic pumping system and diastole where the dynamics of the LV is removed. Each of these dynamical pumping states has an inherent frequency of operation (ω1 and ω2 ) which gives information about LV systolic function (ω1 ) as well as arterial dynamics (ω2 ). IF methodology extracts ω1 and ω2 from the pressure wave. This method was applied to invasive aortic pressure waveforms and noninvasively measured carotid pressure waveforms. Our results shows that ω1 is elevated in patients with LV systolic dysfunction (LVSD). However, ω1 remains relatively constant under healthy conditions as age advances. Our results indicate that IF methodology can be used to detect LVSD from a single pressure waveform. One unique advantage of the IF method is only the shape of the waveform is required. Therefore, ω1 can be easily derived from noninvasive measurements and monitored continuously. 3:33PM D7.00007 Comparative study of diastolic filling under varying left ventricular wall stiffness , PRITAM MEKALA, ARVIND SANTHANAKRISHNAN, Oklahoma State University — Pathological remodeling of the human cardiac left ventricle (LV) is observed in hypertensive heart failure as a result of pressure overload. Myocardial stiffening occurs in these patients prior to chronic maladaptive changes, resulting in increased LV wall stiffness. The goal of this study was to investigate the change in intraventricular filling fluid dynamics inside a physical model of the LV as a function of wall stiffness. Three LV models of varying wall stiffness were incorporated into an in vitro flow circuit driven by a programmable piston pump. Windkessel elements were used to tune the inflow and systemic pressure in the model with least stiffness to match healthy conditions. Models with stiffer walls were comparatively tested maintaining circuit compliance, resistance and pump amplitude constant. 2D phase-locked PIV measurements along the central plane showed that with increase in wall stiffness, the peak velocity and cardiac output inside the LV decreased. Further, inflow vortex ring propagation toward the LV apex was reduced with increasing stiffness. The above findings indicate the importance of considering LV wall relaxation characteristics in pathological studies of filling fluid dynamics. 3:46PM D7.00008 Right Heart Vortex Entrainment Volume and Right Ventricular Diastolic Dysfunction , JAMES BROWNING, JEAN HERTZBERG, University of Colorado Boulder, BRETT FENSTER, JOYCE SCHROEDER, National Jewish Health and University of Colorado Denver — Recent advances in cardiac magnetic resonance imaging (CMR) have allowed for the 3-dimensional characterization of blood flow in the right ventricle (RV) and right atrium (RA). In this study, we investigate and quantify differences in the characteristics of coherent rotating flow structures (vortices) in the RA and RV between subjects with right ventricular diastolic dysfunction (RVDD) and normal controls. Fifteen RVDD subjects and 10 age-matched controls underwent same day 3D time resolved CMR and echocardiography. Echocardiography was used to determine RVDD stage as well as pulmonary artery systolic pressure (PASP). CMR data was used for RA and RV vortex quantification and visualization during early and late ventricular diastole. RA and RV vortex entrainment volume is quantified and visualized using the Lambda-2 criterion, and the results are compared between healthy subjects and those with RVDD. The resulting trends are discussed and hypotheses are presented regarding differences in vortex characteristics between healthy and RVDD subjects cohorts. 3:59PM D7.00009 In vivo quantification of intraventricular flow during left ventricular assist device support , VI VU, KIN WONG, Department of Bioengineering, San Diego State University, JUAN DEL ALAMO, PABLO M.L. AGUILO, Department of Mechanical and Aerospace Engineering, University of California San Diego, KAREN MAY-NEWMAN, Department of Bioengineering, San Diego State University, DEPARTMENT OF BIOENGINEERING, SAN DIEGO STATE UNIVERSITY COLLABORATION, DEPARTMENT OF MECHANICAL AND AEROSPACE ENGINEERING, UNIVERSITY OF CALIFORNIA SAN DIEGO COLLABORATION, MECHANICAL ASSIST DEVICE PROGRAM, SHARP MEMORIAL HOSPITAL COLLABORATION — Left ventricular assist devices (LVADs) are mechanical pumps that are surgically connected to the left ventricle (LV) and aorta to increase aortic flow and end-organ perfusion. Clinical studies have demonstrated that LVADs improve patient health and quality of life and significantly reduce the mortality of cardiac failure. However, In the presence of left ventricular assisted devices (LVAD), abnormal flow patterns and stagnation regions are often linked to thrombosis. The aim of our study is to evaluate the flow patterns in the left ventricle of the LVAD-assisted heart, with a focus on alterations in vortex development and blood stasis. To this aim, we applied color Doppler echocardiography to measure 2D, time resolved velocity fields in patients before and after implantation of LVADs. In agreement with our previous in vitro studies (Wong et al, Journal of Biomechanics 47, 2014), LVAD implantation resulted in decreased flow velocities and increased blood residence time near the outflow tract. The variation of residence time changes with LVAD operational speed was characterized for each patient. 4:12PM D7.00010 Bumps and Ridges: Trabeculation Effects in Embryonic Heart Development , NICHOLAS BATTISTA, ANDREA LANE, LAURA MILLER, Univ of NC - Chapel Hill — Trabeculae form in developing zebrafish hearts for Re on the order of 0.1; effects of trabeculae in this flow is not well understood. Dynamic processes, such as vortex formation, are important in the generation of shear at the endothelial surface layer and strains at the epithelial layer, which aid in proper morphology and functionality. In this study, CFD is used to quantify the effects of Re and idealized trabeculae height on the resulting flows. Sunday, November 23, 2014 2:15PM - 4:25PM — Session D9 Focus Session: The Impact of Andy Acrivos on Today’s Fluid Mechanics Science I 3014/3016 - Eric Shaqfeh, Stanford University 2:15PM D9.00001 Field-driven mesoscale phase transition in polarized colloids in microgravity1 , BORIS KHUSID, EZINWA ELELE, New Jersey Institute of Technology, Newark, NJ — An unexpected phase transition in a polarized suspension was reported by Kumar, Khusid, Acrivos, PRL95, 258301, 2005 and Agarwal, Yethiraj, PRL102, 198301, 2009. Following the field application, particles aggregated headto-tail into chains that bridged the interelectrode gap and then formed a cellular pattern, in which large-scale particle-free voids were enclosed by particle-rich thin walls. Surprisingly, the size of particle-free domains scales linearly with the gap thickness but is insensitive to the particle size and the field strength and frequency. Cellular structures were not observed in simulations of equilibrium in a polarized suspension (Richardi, Weis, J Chem Phys 135, 124502, 2011; Almudallal, Saika-Voivod, PRE 84, 011402, 2011). Nonequilibrium simulations (Park, Saintillan, PRE 83, 041409, 2011) showed cellular-like structures but at a particle concentration much higher than in experiments. A requirement for precise matching of densities between particles and a fluid to avoid gravity effects limits terrestrial experiments to negatively polarized particles. We will present data on positively polarized non-buoyancy-matched particles and the development of experiments in the International Space Station needed to evaluate gravity contribution. 1 Supported by NASA’s Physical Science Research Program, NNX13AQ53G. 2:28PM D9.00002 What makes cilia beat? , ASHOK SANGANI, KENNETH FOSTER, Syracuse University — There have been numerous attempts at understanding the mechanism responsible for producing steady beat in cilia that propel eukaryotic cells. The core structure of a cilium, known as the axoneme, consists of nine microtubules doublet surrounding a central pair of microtubules. The dynein motors on the doublets generate active shear forces that are responsible for relative sliding and bending of the cilium. Several theories have been put forward over the last sixty years but none are supported through a careful analysis of the ciliary beating. We have combined the methods of slender body theory and multipole expansions – both developed by Professor Acrivos and his students – to analyze in detail the hydrodynamics of ciliary beating in ten different cases. The analysis is used to infer the internal dynamics of cilia and, in particular, the active forces generated by the dynein motors along the length of cilia. We find that the properties of the axoneme vary along the length of a cilium. In the central region, the active forces generated are primarily dependent on the rate of sliding of the microtubules. This region therefore appears to be optimized to propagate a wave down the length of the cilium. The proximal region near the cell body appears more complex and may be suitable for creating waves. These conclusions from the hydrodynamic analysis are consistent with a recent study that reports different structures of the axoneme in these two regions. The detailed comparison with various theories of axoneme dynamics/ collective behavior of molecular motors show that none of the existing theories are adequate for predicting the correct active moments generated so that the mechanism for ciliary beating still remains unresolved. 2:41PM D9.00003 Transient behavior of liquid drops with a polymerized interface , DOMINIQUE BARTHES-BIESEL, PIERRE-YVES GIRES, ANNE LE GOFF, ERIC LECLERC, ANNE-VIRGINIE SALSAC, BMBI, Universite de Technologie de Compiegne, IFSB TEAM — Capsules consisting of a liquid droplet enclosed by a thin polymerized membrane are commonly encountered in nature or in industry. The mechanical properties of the capsule wall are essential to control particle integrity and release of the internal contents. We have designed a novel method to assess the elastic surface shear modulus Gs of micrometer size capsules. It is based on the comparison between the predictions of a numerical model and the experimental measurement of the steady velocity and deformed profile of a capsule flowing in a square section microfluidic tube. To assess membrane viscosity, we have designed a new set-up, where a capsule exits suddenly from a square channel into a wider rectangular channel. The technique is illustrated for initially spherical capsules with a thin cross-linked albumin (HSA) membrane. From the profile in the square channel, we infer the mean value of Gs. We then follow the transient deformation of the capsule in the rectangular pore. Under the same flow conditions, the experimental relaxation time is about twice the numerical time computed for a capsule with a purely elastic membrane. The HSA membrane has thus some viscosity, probably due to the rearrangement of loose HSA molecules on the inside of the membrane. 2:54PM D9.00004 A generalized Oldroyd model for a suspension of rod-like or disk-like particles , ROBERT DAVIS, RICHARD MARTIN, ALEXANDER ZINCHENKO, University of Colorado Boulder — Early work on emulsion rheology was performed by Frankel and Acrivos [J. Fluid Mech., 44, 65 (1970)] for dilute emulsions of drops with small deformations. Martin, Zinchenko and Davis [J. Rheol. 58, 759, (2014)] developed a more general approach, valid for larger deformations and based on a 5-parameter Oldroyd model with variable coefficients found from fitting the equation to three viscometric and two extensiometric functions in simple shear and hyperbolic flow, respectively, at arbitrary flow intensities. The method was validated with the Frankel-Acrivos small-deformation theory. We have extended the method to ellipsoidal particles subject to Brownian rotations. The viscometric and extensiometric functions were obtained by numerically solving the Fokker-Planck equation for the particle orientation distribution function through expansions into spherical harmonics. The results compare well with the interpolation models of Hinch and Leal [J. Fluid Mech., 76, 187 (1976)] between the limits of weak and strong flows. A benefit of our general approach to constitutive modeling is that it can be applied to concentrated systems (suspensions, emulsions, etc.), while the prior models are limited to dilute systems of non-interacting particles or drops. 3:07PM D9.00005 A coating flow on a rotating vertical disk , EDWARD HINCH, MATTHEW CROWE, DAMTP, Cambridge University, UK — A thin viscous film of liquid on a vertical disk can be stopped from dripping off by slowly (Re ≪ 1) rotating the disk about its horizontal axis. When surface tension is neglected, it is known [Phys Fluids 21 103102] that the thickness of the film is constant around circles whose centres are offset from the centre of the disk. Small nonzero surface tension determines the variation of the thickness on the differing circles, and the above paper found the first 4 terms in an expansion for small gravity. Acrivos [Phys Fluids 22, 05901] has speculated on the maximum strength of gravity before dripping starts. A new integro-differential equation in the non-orthogonal coordinates of the eccentric circles is derived and solved for the distribution of thickness across the circles, to test Acrivos’s speculation. 3:20PM D9.00006 Inertia in Suspension Flows: Bulk Properties and Recirculating Wakes , JEFFREY MORRIS, HAMED HADDADI, Levich Institute and ChE, City College of New York — The influence of suspensions in which the particle scale inertia is non-negligible is considered by examining the inertial effects upon two distinct aspects of suspension flow through numerical simulations using the lattice-Boltzmann method. In one case, we consider the dependence of the bulk flow properties on the particle-scale Reynolds number, Re = ργ̇a2 /µ, where ρ and µ are, respectively the density and viscosity of the suspending fluid, gamma ˙ is the shear rate and a is the radius of a spherical suspended particle. We describe briefly the influence of solid fraction for 0 < φ ≤ 0.35, and Re on the viscosity and normal stresses, showing how the microstructure induced by the flow plays a role in setting these properties. In the second case, we consider the flow of a suspension of φ < 0.1 past a cylinder of radius large relative to the suspended particles at bulk Reynolds numbers yielding a recirculating wake. It is observed experimentally that the resulting wake is largely depleted of particles. The basis for this observation is explored by LB simulation and is found to be due to a two-part mechanism in which particle migrate to a limit cycle and are then displaced by fluctuations from particle interaction. 3:33PM D9.00007 Magnetically Driven Flows of Suspensions of Rods to Deliver Clot-Busting Drugs to Dead-End Arteries , ROGER BONNECAZE, MICHAEL CLEMENTS, The University of Texas at Austin — Suspensions of iron particles in the presence of a magnetic field create flows that could significantly increase the delivery of drugs to dissolve clots in stroke victims. An explanation of this flow rests on the foundation of the seminal works by Prof. Acrivos and his students on effective magnetic permittivity of suspensions of rods, hydrodynamic diffusion of particles, and the flow of suspensions. Intravenous administration of the clot dissolving tissue plasminogen activator (tPA) is the most used therapy for stroke. This therapy is often unsuccessful because the tPA delivery is diffusion-limited and too slow to be effective. Observations show that added iron particles in a rotating magnetic field form rotating rods along the wall of the occluded vessel, creating a convective flow that can carry tPA much faster than diffusion. We present a proposed mechanism for this magnetically driven flow in the form of coupled particle-scale and vessel-scale flow models. At the particle-scale, particles chain up to form rods that rotate, diffuse and translate in the presence of the flow and magnetic fields. Localized vorticity created by the rotating particles drives a macroscopic convective flow in the vessel. Suspension transport equations describe the flow at the vessel-scale. The flow affects the convection and diffusion of the suspension of particles, linking the two scales. The model equations are solved asymptotically and numerically to understand how to create convective flows in dead-end or blocked vessels. 3:46PM D9.00008 Self-diffusiophoresis of catalytically active patchy colloids near a solid boundary , CHARLES MALDARELLI, ALI MOZAFFARI, Levich Institute, NIMA SHARIFI-MOOD, University of Pennsylvania, JOEL KOPLIK, Levich Institute — Active colloidal swimmers designed to move along an envisioned path to ascertain various applications in nanotechnology. In diffusiophoresis, gradients in the solute concentration across the colloid create an imbalance force due to the interactions of the solute with the particle. These forces can also be integrated into a self-propulsion by choosing a reactant as a solute which undergoes a surface reaction only on one face of a colloid. The effect of boundaries in self-diffusiophoresis is not purely to retard the motion, because the boundaries also alter the solutal gradient. We developed an analytical approach to investigate the dynamics of swimming colloid near an infinite planar wall assuming constant flux production and a repulsive interaction between product solute and the colloid. The motion of the colloid was decomposed into translational motions perpendicular and parallel to the wall and a rigid body rotation around the third axis. Our analysis indicates when a patchy colloid approaches the boundary with an inclination angle with respect to the unit normal of the wall, the asymmetric distribution of product around the colloid compels it to rotate and redirects its reaction section towards the wall and thereby the colloid will be moved away from the wall. 3:59PM D9.00009 Lateral migration and diffusion of a mechanical engineer through emulsion of drops induced by Andy’s influence , KAUSIK SARKAR, George Washington University — My initiation to analytical sides of Stokes flow was thorough cyclostyled notes of Andy’s Stanford fluid mechanics notes distributed by Ashok Sangani when he taught a course at Hopkins. Since then, reversibility of Stokes flow and singularity solution remained with me during my research carrier. I will discuss how it and Frankel and Acrivos (1970) paper in JFM influenced my research in drop deformation and emulsion rheology at finite inertia, winning the 2009 Acrivos award by my first PhD student Xiaoyi Li. Finally, I will discuss migration of suspended particles, drops, polymers and biological cells caused by breaking of Stokes reversibility due to deformation and viscoelasticity. Here, we show that the migration is induced by the image stresslet field, as was also indicated by Dave Leighton’s thesis and a paper with Smart [1991, Phys. Fluid A, 3, 21]. We relate the stresslet field to the Interface tensor, and investigate the effects of drop inclination. In contrast to a plausible notion asserted also in the literature, that reduced inclination (increased alignment with flow) decreases migration, it is shown here that reduced inclination increases the stresslet and thereby the migration velocity. 4:12PM D9.00010 Irreversibility in the motion of suspended particles , GERMAN DRAZER, Rutgers, The State University of New Jersey — We discuss the observed irreversibility in the motion of suspended particles in Stokes flow and its effect in a wide range of transport phenomena, from the rheology of non-Brownian suspensions to the separation of colloidal particles in microfluidic devices. The work of Prof. Acrivos in shear-induced diffusion was instrumental to explain the observed irreversible and stochastic behavior in suspensions flows. The same ideas explain the underlying mechanisms in some popular microfluidic devices used for the separation of suspended particles. Sunday, November 23, 2014 2:15PM - 4:12PM Session D10 Microscale Flows: Mixing and Reactions in Droplets — 3005 - Shimon Rubin, Israel Institute of Technology 2:15PM D10.00001 Convection-diffusion driven concentration gradients in nanolitre droplets for microfluidic screening applications , RAVIRAJ THAKUR, School of Mechanical Engineering, Purdue University, AHMED AMIN, Microfluidic Innovations, STEVEN WERELEY, School of Mechanical Engineering, Purdue University — Ability to generate a concentration gradients in emulsified aqueous droplets is a highly desired feature for several lab-on-chip applications. Numerous schemes exists for generating concentration gradients in continuous flow devices such as Y junctions, split-and-recombine techniques, etc. However, varying the sample concentration in emulsified droplets is quite challenging. In this work, we have developed a scheme for generating and controlling concentration gradients in programmable multi-layer PDMS microfluidic chips. Briefly, a high concentration sample is injected into a steady stream of buffer. The buffer with the sample pulse and an immiscible oil phase are flowed through a T-junction in an alternate manner. As the sample pulse advances, the combined effect of diffusion and convection produced dispersion of sample pulse in streamwise direction. This continuous gradient stream is split into discrete droplets at the T-junction. Pulsatile flow condition are maintained using on-chip diaphragm peristaltic pumps. The problem can be thought of an extension of Taylor-Aris dispersion with laminar pulsatile flow in rectangular channels. The concentration profile is found to be dependent upon the frequency of pulsatile flow and thus can be fine-tuned according to application needs. Theoretical framework is established for pump regimes that correlates the diffusion coefficients of the input samples with the resultant concentration profiles. 2:28PM D10.00002 Intra-phase mixing in a bi-component translating drop , THOMAS WARD, Iowa State University — The intra-phase mass transport in a translating spherical drop containing two species will be studied numerically in the zero capillary number limit. The problem is relevant to microfluidic systems where it is common to form two drops of unequal or nearly equal volume in a microfluidic channel where they subsequently merge and then translate. The mixing process in this system is controlled by diffusion due to the small length scales despite the relatively large velocities and low diffusivities. The species conservation equation are discretized using a 4th order finite difference scheme in space with an adaptive explicit Runge-Kutta-Merson scheme to advance in time. With this scheme the solutions conserve mass throughout the numerical integration cycle. Numerical data for Peclet numbers ranging between 1000-10000 will be used to estimate the deviation from the equilibrium concentration as a function of time. Initial species concentration range from ratios of 1:9 to 1:1. 2:41PM D10.00003 Modeling the Dilution of Static Droplet Arrays with Moving Plugs1 , WILLIAM WANG, SIVA VANAPALLI, Texas Tech University — Generation of arrays of immobilized microfluidic droplets with variation in reagent concentration from drop-to-drop is important for a variety of biochemical and screening assays. Recently our laboratory (Sun et al., Lab Chip, 2011) showed that such gradients in chemical concentration can be achieved by coalescing diluting plugs with drops immobilized in a microfluidic parking network. In this study, we investigate the key hydrodynamic mechanisms responsible for generation of concentration gradients in static droplet arrays, with the goal of predicting the dilution profiles observed in experiments. We conduct simulations based on a phenomenological model that includes diffusion, advection due to circulating flow within moving plugs, enhanced material transfer due to coalescence and break-up events, and geometry. Consistent with experiments, we find that the concentration profiles can exhibit segmentation between rows of parked droplets due to coalescence events occurring on alternating sides of the diluting plug. Tail-sweeping of wall material can increase concentrations in the plug tail. Also, coalescence and break-up events can significantly enhance dilution rates and ranges. Our results impact the design of SDAs for creating broad and predictable concentration gradients. 1 NSF CAREER 2:54PM D10.00004 Self-assembly and novel planetary motion of ferrofluid drops in a rotational magnetic field1 , CHING-YAO CHEN, HAO-CHUNG HSUEH, National Chiao Tung University, Taiwan — We experimentally investigate the motion of ferrodrops in a rotating magnetic field. Magnetized and driven by the external field, the ferrodrops are stretched and self-align to form a drop array along the field orientation. An interesting planet-like dual rotation, including local self-spins of individual drops and a global revolution of the drop array, is newly identified. While the drops spin nearly synchronized with the external field, the revolution always lags behind the field and appears a forth and back movement. Prominence of the net revolutionary movement depends on the strength and uniformity of the overall field as well as the number of drops containing in the array. In general, more uniform and stronger rotating field lead to a more prominent global revolution. Phenomenon of such planetary motion can be applied to mix two fluids more effectively than self-spin drops. 1 Supported by ROC NSC 102-2221-E-009-051-MY3 3:07PM D10.00005 Migration of deformable droplets caused by microfluidic inertial effects1 , GUOQING HU, CHUNDONG XUE, XIAODONG CHEN, LNM, Institute of Mechanics, Chinese Academy of Sciences — The inertial effect is an effective way of focusing and sorting droplets suspended in microchannels. Here we conduct numerical simulations and experiments on the droplet motion and deformation in a straight microchannel. In contrast to most existing literature, the present simulations are three-dimensional and full length in the streamwise direction. The migration dynamics and equilibrium positions of the droplets are obtained for different fluid velocities and droplet sizes. Droplets with diameters larger than half of the channel height migrate to the centerline in the height direction and two equilibrium positions are observed between the centerline and the wall in the width direction. In addition to the well-known Segre-Silberberg equilibrium positions, new equilibrium positions closer to the centerline are observed. This finding is validated by preliminary experiments that are designed to introduce droplets at different initial lateral positions. Small droplets also migrate to two equilibrium positions in the quarter of the channel cross section, but with the coordinates between the centerline and the wall. The distributions of the lift forces, angular velocities and the deformation parameters of droplets along the two confinement direction are also investigated in details. 1 Supported by NSFC(#11272321). 3:20PM D10.00006 Micro-droplets lubrication film thickness dynamics , AXEL HUERRE, MMN, UMR CNRS 7083, ESPCI ParisTech, 75005 Paris, France, OLIVIER THEODOLY, LAI, INSERM U600, CNRS UMR 6212, Case 937, 13009 Marseille, France, ISABELLE CANTAT, IPR, UMR CNRS 6251, Universite de Rennes 1, 35000 Rennes, France, ALEXANDER LESHANSKY, Department of Chemical Engineering, TechnionIIT, Haifa, 32000, Israel, MARIE-PIERRE VALIGNAT, LAI, INSERM U600, CNRS UMR 6212, Case 937, 13009 Marseille, France, MARIE-CAROLINE JULLIEN, MMN, UMR CNRS 7083, ESPCI ParisTech, 75005 Paris, France, MMN TEAM, LAI TEAM, IPR TEAM, DEPARTMENT OF CHEMICAL ENGINEERING TEAM — The motion of droplets or bubbles in confined geometries has been extensively studied; showing an intrinsic relationship between the lubrication film thickness and the droplet velocity. When capillary forces dominate, the lubrication film thickness evolves non linearly with the capillary number due to viscous dissipation between meniscus and wall. However, this film may become thin enough that intermolecular forces come into play and affect classical scalings. We report here the first experimental evidence of the disjoining pressure effect on confined droplets by measuring droplet lubrication film thicknesses in a microfluidic Hele-Shaw cell. We find and characterize two distinct dynamical regimes, dominated respectively by capillary and intermolecular forces. In the former case rolling boundary conditions at the interface are evidenced through film thickness dynamics, interface velocity measurement and film thickness profile. 3:33PM D10.00007 Droplet formation and lateral migration via solvent shifting in a microfluidic setup , RAMIN HAJIAN, STEFFEN HARDT, Center of Smart Interfaces, TU Darmstadt — When a non-solvent is added to a solvent/solute mixture and if the solvent and the non-solvent are miscible, a part of the solute transforms to tiny (i.e. micron-/submicron-sized) droplets when the solvent concentration reduces. This phenomenon, resulting from supersaturation, is termed solvent shifting or Ouzo effect. Here we investigate this process in a co-flow microfluidic device. Thanks to the laminar nature of the flow, the mass transfer is mainly diffusive and can be analyzed employing (semi)analytical models. Using the resulting concentration profiles along with the ternary phase diagram (TPD) we analyze droplet formation and their lateral migration in the channel. The ternary system consists of a binary mixture (0.5wt% divinyle benzene (DVB) + 95.5wt% ethanol) and deionized water (non-solvent). Plotting concentration trajectories in the TPD we show that they hit the binodal curve in a region in which droplets of DVB form via nucleation, as opposed to spinodal decomposition. The lateral migration of droplets is partially attributed to the Marangoni effect induced by concentration gradients. However, the main effect governing droplet migration appears to be the phase-separation front (separating the one-phase and two-phase regions) moving toward the center of the channel. 3:46PM D10.00008 Theory of microfluidic step-emulsification , ALEXANDER LESHANSKY, Technion - Israel Institute of Technology, ZHENZHEN LI, SAMUEL METAIS, ESPCI Paris-Tech, LEN PISMEN, Technion - Israel Institute of Technology, PATRICK TABELING, ESPCI Paris-Tech — We present a comprehensive study of the microfluidic step-emulsification process for high-throughput production of monodisperse colloidal droplets. The “microfluidic step emulsifier” combines a shallow microchannel operating with two co-flowing immiscible fluids and an abrupt (step-like) opening to a deep and wide reservoir. Based on Hele-Shaw hydrodynamics, we determine the quasi-static shape of the fluid interface prior to transition to oscillatory step-emulsification at low capillary numbers. The transition threshold obtained from scaling arguments yields an excellent agreement with experimental data. A closed-form expression for the size of the droplets generated in the step-emulsification regime and derived using geometric arguments also shows a very good agreement with the experiment. 3:59PM D10.00009 Numerical simulation of droplet formation regimes and sizes in microfluidic T-junction devices , MEHDI NEKOUEI, SIVA VANAPALLI, Texas Tech University, Department of Chemical Engineering — The T-junction geometry has been widely used for producing monodisperse droplets in microfluidic devices. Droplet formation regimes and sizes are expected to depend on a variety of conditions including flow rates, capillary number, channel geometry and viscosity ratio. Experiments have investigated drop production at a T-junction in a narrow control parameter space and developed analytical models for specific operating regimes. In this study, we take advantage of numerical simulations based on volume-of-fluid method to explore this broad parameter space systematically, and contrast our results with prior experimental data. We find our simulations predict well the regimes of squeezing, dripping and jetting. We also observe that our drop size data is in good agreement with three different experimental reports. Although our results match experimental data, the analytical models do not agree with each other since they are based on specific operating conditions. We use numerical simulations to elucidate the missing components in the physics of drop formation at a T-junction, with an attempt to reconcile existing analytical models. Sunday, November 23, 2014 2:15PM - 4:25PM Session D11 Rotating Flows II — 3007 - Zvi Rusak, Rensselaer Polytechnic Institute 2:15PM D11.00001 Effect of sinusoidal perturbation of the inner cylinder on the stability criteria in a wide gap Taylor-Couette flow , KRISHNASHIS CHATTERJEE, ANNE STAPLES, Virginia Polytechnic Institute and State University — The effects of sinusoidal perturbations of the inner cylinder radius in the axial direction in a Taylor-Couette flow apparatus are studied. The base flow solution in the apparatus is derived and then linear stability analysis is performed using the wide gap approximation. The stability criteria are established based on the critical Taylor numbers which mark the transition from the purely circular base flow to the Taylor Vortex flow regime. The effects of varying the forcing wavelength and modulation amplitude on the stability criteria are investigated. The studies are conducted for different instability wave numbers and inner and outer cylinder rotational velocity combinations. The results are compared with those obtained in the same apparatus using a narrow gap assumption, and with the classical Taylor-Couette case. 2:28PM D11.00002 DNS study on the turbulence statistics of the Taylor-Couette flow in the Reynolds numbers near the torque transition , KOUSUKE OSAWA, YOSHITSUGU NAKA, Tokyo Institute of Technology, NAOYA FUKUSHIMA, The University of Tokyo, MASAYASU SHIMURA, MAMORU TANAHASHI, Tokyo Institute of Technology — The Taylor-Couette flow has been investigated extensively because of its significance in a wide range of engineering applications. In the present study, direct numerical simulations (DNS) have been performed to clarify the characteristics of turbulence statistics of Taylor-Couette flow in Re from 8000 to 20000 where the torque scaling changes according to the Wendt’s empirical formula. Although the flow structures show the existence of the Taylor vortex, the fine scale structures become more pronounced in higher Reynolds numbers. The velocity fluctuations are decomposed into the contribution of Taylor vortex and the remaining turbulent component. A distinct Reynolds number dependence is observed for the turbulence components in the circumferential velocity fluctuation and the Reynolds shear stress while those of the wall normal and the axial velocity fluctuations are insensitive to the Reynolds number change. The budget of the transport equation of the Reynolds stress is evaluated, and the balance of the Reynolds shear stress indicates the Reynolds number dependence in the redistribution and pressure-diffusion terms. This may explain the Reynolds number dependence in the relative contribution of the Taylor vortex and the turbulence components of the Reynolds shear stress. 2:41PM D11.00003 Combined effects of rotation and rib-roughness – a dns study of turbulent channel flow , HELGE I. ANDERSSON, VAGESH D. NARASIMHAMURTHY, Dept. Energy and Process Engineering; Norwegian University of Science and Technology — The combined effects of system rotation and rib-roughness on turbulent channel flow have been investigated by means of direct numerical simulations. Square ribs were placed on both walls in a non-staggered arrangement and the channel was subjected to steady rotation about a spanwise axis for a series of rotation numbers up to Ro = 24. A pressure-loss reduction of about 20 per cent resulted from the imposed rotation at Ro = 6. In spite of the 10 per cent blockage due to the wall-mounted ribs, the flow field exhibited statistical streamwise homogeneity in the core region. The mean velocity varied linearly with a slope such that the mean fluid rotation exactly outweighed the imposed system rotation. The flow field in the vicinity of the ribs was affected differently at the two sides of the rotating channel. The separated flow region behind the ribs on the anti-cyclonic pressure side shrinked with increasing Ro due to the enhanced turbulent mixing caused by the Coriolis force. The original d-type roughness was thus turned into a k-type roughness. 2:54PM D11.00004 Axially localized states in Taylor Couette flows1 , JOSE M. LOPEZ, FRANCISCO MARQUES, Univ Politecnica de Catalunya — We present numerical simulations of the flow in a Taylor Couette system with the inner cylinder rotating and aspect ratio Γ Γ < 0.95, being N the number of Taylor vortices. For these values a complex experimental bifurcation scenario has been reported. The restricted to 0.86 < N transition from wavy vortex flow (WVF) to a very low frequency mode VLF happens via an axisymmetric eigenfunction. The VLF plays an essential role in the dynamics, leading to chaos through a two-tori period-doubling route. This chaotic regime vanishes with further increase in Re and gives rise to a new flow regime ALS characterized by the existence of large jet oscillations localized in some pairs of vortices. The aim of this numerical study is to extend the available information on ALS by means of a detailed exploration of the parameter space in which it occurs. Frequency analysis from time series simultaneously recorded at several points of the domain has been applied to identify the different transitions taking place. The VLF occurs in a wide range of control parameters and its interaction with the axially localized states is crucial is most transitions, either between different ALS or to the chaotic regime. 1 Spanish Ministry of Education and Science grants (with FEDER funds) FIS2013-40880 and BES-2010-041542 3:07PM D11.00005 Precession of a rapidly rotating cylinder flow: traverse through resonance1 , JUAN LOPEZ, Arizona State Univ, FRANCISCO MARQUES, Universitat Politècnica de Catalunya — The flow in a rapidly rotating cylinder that is titled and also rotating around another axis can undergo sudden transitions to turbulence. Experimental observations of this have been associated with triadic resonances. The experimental and theoretical results are well-established in the literature, but there remains a lack of understanding of the physical mechanisms at play in the sudden transition from laminar to turbulent flow with very small variations in the governing parameters. Here, we present direct numerical simulations of a traverse in parameter space through an isolated resonance, and describe in detail the bifurcations involved in the sudden transition. 1 U.S. National Science Foundation grant CBET-1336410 and Spanish Ministry of Education and Science grant (with FEDER funds) FIS2013-40880 3:20PM D11.00006 Instabilities of the sidewall boundary layer in a rapidly rotating split cylinder1 , PALOMA GUTIERREZ-CASTILLO, JUAN LOPEZ, Arizona State Univ — The flow in a rapidly rotating cylinder is studied numerically. The cylinder is split in two with the top rotating slightly faster than the half. The interior basic state is in solid-body rotation with the mean rotation rate. Differential rotation drives boundary layers on the sidewall, and the top and bottom endwalls drive fluid into the sidewall layer. The basic state loses stability to three-dimensional perturbations when both the mean rotation and differential rotation increase. Then, the sidewall boundary layer and the corner flow in the slower half undergo a number of instabilities. These include slow low-azimuthal-wavenumber modes whose frequencies excite inertial waves in the interior as well as fast high-azimuthal-wavenumber modes whose impact is contained in the sidewall boundary layer region. Some of these high azimuthal-wavenumber modes have a complicated behavior with pairs of Gortler vortices present in the bottom corner of the cylinder. The behavior becomes even more complicated with mixed modes with interacting low and high azimuthal wavenumbers, and nonlinear competition due to Eckhaus instabilities and mode interactions. 1 Supported by NSF grant CBET-1336410 3:33PM D11.00007 On the nonlinear stability of the circular Couette flow to viscous axisymmetric perturbations , PUN WONG YAU, SHIXIAO WANG, University of Auckland, ZVI RUSAK, Rensselaer Polytechnic Institute — An axisymmetric viscous nonlinear stability analysis of the circular Couette flow to any finite amplitude perturbation is developed. The analysis is based on investigating the reduced Arnol’d energy-Casimir function Ard , which consists of the sum of the total kinetic energy of the flow E and the Casimir circulation dependent function CS , i.e. Ard = E + CS . In this case, ∆Ard is used as a Lyapunov function, which represents the difference between the reduced Arnol’d function at a later time t and the corresponding base flow value. The requirement for the temporal decay of ∆Ard leads to two novel conditions for the nonlinear stability of this steady flow against axisymmetric viscous perturbations of any finite amplitude. We also establish for the very first time a definite nonlinear stability region in terms of the operational parameters for the circular Couette flow. Once the flow is nonlinearly stable and stays axisymmetric, it always decays asymptotically to a unique steady state defined by the rotating cylinders. The results from this research shed a new fundamental physical insight into a classical flow problem that was studied for many decades. 3:46PM D11.00008 At what spatio-temporal scales can inertial waves be found in rotating turbulence? , PIERRE-PHILIPPE CORTET, ANTOINE CAMPAGNE, Laboratoire FAST, CNRS, Université Paris-Sud, France, BASILE GALLET, Laboratoire SPHYNX, Service de Physique de l’État Condensé, DSM, CEA Saclay, CNRS, 91191 Gif-sur-Yvette, France, FRÉDÉRIC MOISY, Laboratoire FAST, CNRS, Université Paris-Sud, France — We present a spatio-temporal analysis of a statistically stationary rotating turbulence experiments aiming to extract a statistical signature of inertial waves and to determine at what scales and frequencies these waves can be detected. This analysis is performed from two-point correlations of temporal Fourier transform of the velocity fields time series obtained from stereoscopic PIV measurements in the rotating frame. From this data, it is possible to quantify the degree of anisotropy of turbulence due to global rotation both as a function of angular frequency ω and spatial scale normal to the rotation axis r⊥ . This frequency and scale dependent anisotropy is found compatible with the dispersion relation of inertial waves, provided that a weak non-linearity condition is satisfied in terms of a properly defined Rossby number dependant on the spatio-temporal scale (ω,r⊥ ). 3:59PM D11.00009 On the development of lift and drag in a rotating and translating cylinder1 , ANTONIO MARTIN-ALCANTARA, Universidad de Malaga (Spain), ENRIQUE SANMIGUEL-ROJAS, Universidad de Cordoba (Spain), RAMON FERNANDEZFERIA, Universidad de Malaga (Spain) — The two-dimensional flow around a rotating cylinder is investigated numerically using a vorticity forces formulation with the aim of analyzing the flow structures, and their evolutions, that contribute to the lift and drag forces on the cylinder. The Reynolds number, based on the cylinder diameter and steady free-stream speed, considered is Re = 200, while the non-dimensional rotation rate (ratio of the surface speed and free-stream speed) selected were α = 1 and 3. For α = 1 the wake behind the cylinder for the fully developed flow is oscillatory due to vortex shedding, and so are the lift and drag forces. For α = 3 the fully developed flow is steady with constant (high) lift and (low) drag. Each of these cases is considered in two different transient problems, one with angular acceleration of the cylinder and constant speed, and the other one with translating acceleration of the cylinder and constant rotation. Special attention is paid to explaining the mechanisms of vortex shedding suppression for high rotation (when α = 3) and its relation to the mechanisms by which the lift is enhanced and the drag is almost suppressed when the fully developed flow is reached. 1 Supported by the Ministerio de Economia y Competitividad of Spain Grant no. DPI2013-40479-P 4:12PM D11.00010 Robustness of point vortex equilibria in the vicinity of a Kasper Wing , RHODRI NELSON, TAKASHI SAKAJO, Department of Mathematics, Kyoto University — The concept of the Kapser Wing was introduced by Witold Kasper in the early 1970’s. His design proposed to add additional “flaps” or auxiliary aerofoils close to the main aerofoil to control the feeding and shedding of vortices in the vicinity of the wing - the aim being to “trap” a vortex above the main aerofoil thus resulting in an increased lift being experienced. In this study, equilibira consisting of a single point vortex in the presence of an idealised Kapser Wing (modelled as three thin plates) are computed. A background potential flow at an angle attack χ to to the main plate is also present. A range of auxiliary plate configurations is considered and the lift of the system computed. It is seen that the lift experienced by the main plate is “sensitive” to the placement of the auxiliary plates and can be enhanced in comparison to the single plate case (previously considered by Saffman and Sheffield, 1977). The linear stability and non-linear time evolution of the Kasper Wing system is then compared to that of the single plate system. It is seen that the presence of the auxiliary plates, in general, result in a larger range of “useful’ neutrally stable equilibria (according to linear theory) and and can increase the non-linear robustness of the system. Sunday, November 23, 2014 2:15PM - 4:25PM — Session D12 Drops: Bouncing, Impact and Dynamic Surface Interactions I 3018 - Sungyon Lee, Texas A&M University 2:15PM D12.00001 Hydrodynamic quantum analogs1 , JOHN BUSH, MIT — We review recent developments in our understanding of droplets walking on a vibrating fluid bath. Particular attention is given to highlighting the manner in which pilot-wave dynamics gives rise to quantization, and chaotic pilot-wave dynamics to quantum-like statistics. The first links between between pilot-wave dynamics and relativistic effects are explored, along with the relation between this hydrodynamic system and existing realist models of quantum mechanics. Future directions are discussed. 1 The author gratefully acknowledges the support of the NSF through grant CMMI-1333242. 2:28PM D12.00002 Diffraction of walking droplets1 , DANIEL M. HARRIS, Massachusetts Institute of Technology, GIUSEPPE PUCCI, University of Calabria, JOHN W.M. BUSH, Massachusetts Institute of Technology — We present results from our revisitation of the experiment of a walking droplet passing through a single slit, originally investigated by Couder & Fort (PRL, 2006). On each passage, the walker’s trajectory is deviated as a result of the spatial confinement of its guiding wave. We explore the role of the droplet size and the bath’s vibration amplitude on both the dynamics and statistics. We find the behavior to be remarkably sensitive to these control parameters. A complex physical picture emerges. 1 The authors gratefully acknowledge the financial support of the NSF through Grant CMMI-1333242, DMH through the NSF Graduate Research Fellowship Program, and GP through the Programma Operativo Regionale (POR) Calabria - FSE 2007/2013. 2:41PM D12.00003 Hydrodynamic Spin States , ANAND OZA, RODOLFO ROSALES, JOHN BUSH, Massachusetts Institute of Technology — We present the results of a theoretical investigation of droplets walking on a vibrating fluid bath. The droplet’s trajectory is described in terms of an integro-differential equation that incorporates the influence of the propulsive force generated by its monochromatic guiding wave. A stability analysis of orbital solutions shows that walkers may execute stable circular orbits in the absence of an external force. When subjected to rotation, these hydrodynamic spin states exhibit a macroscopic analogue of Zeeman splitting. We conclude by presenting the stability analysis for a pair of orbiting walkers, and compare our results to recent laboratory experiments. 2:54PM D12.00004 Droplets Walking in a 2D Coulomb Potential , LUCAS TAMBASCO, MIT, ANAND OZA, Courant Institute, New York University, JOHN BUSH, MIT — We present the results of a theoretical investigation of a droplet walking on a vibrating fluid bath subject to a central two-dimensional Coulomb force. Using the Hydrogen Atom as motivation, we introduce an attractive Coulombic term to the integrodifferential trajectory equation developed by Oza et al. (JFM, 2014), and analyze the behavior of the droplet’s motion. Linear stability analysis of circular trajectories indicates that stable orbits have quantized radii and can only be achieved for a specific range of vibrational forcing acceleration. In unstable regions, numerical simulations of trajectories show that droplets can either collapse into the center or escape the influence of the Coulomb force. We discuss the value of this Coulombic system as a Hydrodynamic Quantum Analog, and explore its extension to 3 dimensions. 3:07PM D12.00005 Experimental test of temporal reversibility in a memory-driven system , STEPHANE PERRARD, University Paris-Diderot, France, MATTHIEU LABOUSSE, EMMANUEL FORT, ESPCI, France, YVES COUDER, University ParisDiderot, France, UNIVERSITY PARIS-DIDEROT, FRANCE COLLABORATION, ESPCI, FRANCE COLLABORATION — A droplet bouncing sub-harmonically on a vibrated liquid bath can be self-propelled by its interaction with the waves it generates. The resulting “walker” is characterized by the structure of the information loop linking the particle with its pilot wave. The particle can be considered as encoding positional information in the waves it generates. These waves, being sustained for some time, superpose so that the global wave field contains a stored memory of the past trajectory. At its next bounce the drop “reads” this information, which will determine its next move. Is this reading process time reversal? This question is addressed using an experimental trick. A pi phase shift is imposed to the drop, so that the associated wave is reversed. The droplet then “reads” the path-memory phase shifted by pi, so that it comes back on its own previous trajectory. As new emitted waves are pi-shifted, they interfere destructively with the forward path, erasing progressively the recorded memory. The possible application of wave-mediated drop computation will then be discussed. 3:20PM D12.00006 Repeated bouncing of drops on wetting and non-wetting surfaces mediated by a persisting thin air film , JOLET DE RUITER, RUDY LAGRAAUW, DIRK VAN DEN ENDE, FRIEDER MUGELE, MESA+ Institute for Nanotechnology, University of Twente — Liquid drops impinging onto solid surfaces undergo a variety of impact scenarios such as splashing, sticking, and bouncing, depending on impact conditions and substrate properties. Bouncing requires efficient conversion of initial kinetic energy into surface energy and back into kinetic energy. This process is believed to be limited to non-wetting, in particular superhydrophobic surfaces, for which viscous dissipation during drop-substrate contact is minimal. Here, we report a novel bouncing mechanism that applies equally to non-wetting and wetting systems for flat surfaces with contact angles down to 10 degrees. For initial impact speeds up to about 0.5 m/s we demonstrate using dual wavelength interferometry that aqueous and non-aqueous drops remain separated from the substrate by air films of (sub)micrometer thickness at all times throughout a series of up to 16 consecutive bouncing events. We show that the purely dissipative force arising from the viscous squeeze-out of air is responsible for both the momentum transfer and for a substantial part of the residual energy dissipation. 3:33PM D12.00007 Detailed model of bouncing drops on a bounded, vibrated bath , FRANCOIS BLANCHETTE, UC Merced, TRISTAN GILET, Universite de Liege — We present a detailed model of drops bouncing on a bounded vibrated bath. These drops are known to bounce indefinitely and to exhibit complex and varied vertical dynamics depending on the acceleration of the bath. In addition, in a narrow parameter regime, these drops travel horizontally while being guided by the waves they generate. Our model tracks the drop’s vertical radius and position, as well as the eigenmodes of the waves generated via ordinary differential equations only. We accurately capture the vertical dynamics, as well as some of the horizontal dynamics. Our model may be extended to account for interactions with other drops or obstacles, such as slits and corrals. 3:46PM D12.00008 Manipulation of electrically charged drops on a vibrating bath , MARTIN BRANDENBOURGER, STÉPHANE DORBOLO, Université de Liège — The bouncing drop experiment, which allows to store small drops thanks to the vibration of a liquid interface, is sometimes linked to lab-on-a-chip applications. Unfortunately, a lot of these studies focused on the behavior of the bouncing drops instead of their handling. By electrically charging the droplet, we found that an electric field can control the displacement of a droplet stored on a vibrating bath. Even though the charged droplets seems to move with a constant speed at the bath scale, their behavior is shown to be much more complex at the droplet scale. A theoretical model, based on the movement of the droplet during one bounce, has been developed to explain these observations and to understand how to manipulate a droplet without contact with any interfaces. 3:59PM D12.00009 Walking droplets in confined geometries , BORIS FILOUX, OLIVIER MATHIEU, NICOLAS VANDEWALLE, GRASP, Institute of Physics B5a, Sart Tilman, University of Liège, B4000 Liège, Belgium — When gently placing a droplet onto a vertically vibrated bath, coalescence may be avoided : the drop bounces permanently. Upon increasing forcing acceleration, a drop interacts with the wave it generates, and becomes a “walker” with a well defined velocity. In this work, we investigate the confinement of a walker in a mono-dimensional geometry. The system consists of linear submarine channels used as waveguides for a walker. By studying the dynamics of walkers in those channels, we discover some 1D-2D transition. We also propose a model based on an analogy with “Quantum Wires.” Finally, we consider the situation of a walker in a circular submarine channel, and examine the behavior of several walking droplets in this system. We show the quantization of the drop distances, and correlate it to their bouncing modes. 4:12PM D12.00010 Bouncing and coalescence of droplets on falling liquid films1 , ZHIZHAO CHE, AMANDINE DEYGAS, OMAR MATAR, Imperial College London — When a droplet impacts on a falling liquid film, the outcome depends on the fluid properties of the droplet, its speed, and angle of incidence, as well as on the film flow rate and associated flow regimes. In this study, the oblique impact of droplets on a falling liquid film is investigated experimentally. The falling film is created on an inclined substrate and the Reynolds number is varied. Droplets with different sizes and different speeds are used to study the impact process for different Weber and Ohnesorge numbers. Different phenomena of droplet impact are identified and analysed, such as bouncing, partial coalescence, total coalescence, and splashing. An impact regime map is generated, and the effects of droplet impact speed and size, and the film flow rates are studied. The propagation of waves on the liquid film post-impact is analysed. The results show that the flowing film can significantly affect the impact process of droplets, and the latter can alter the propagation of waves on the falling film. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 Sunday, November 23, 2014 2:15PM - 4:12PM Session D13 Drops: Superhydrophobic Surfaces — 3020 - Julie Crockett, Brigham Young University 2:15PM D13.00001 Thermocapillary Dewetting of Superhydrophobic Surfaces , CRISTIAN CLAVIJO, DANIEL MAYNES, JULIE CROCKETT, Brigham Young University, FLUIDS AND THERMAL TRANSPORT LAB TEAM — One of the challenges in preserving the Cassie-Baxter state during liquid flow over micro-structured superhydrophobic surfaces occurs when the force due to pressure in the liquid exceeds that due to the surface tension above the gas cavities thereby forcing liquid in between the micro-structures. This commonly occurs at the impingement point of a jet or droplet where the stagnation pressure is significant. In this work, we present a novel and simple experimental analysis to show that such a wetting state (i.e. Wenzel state) can be reversed for an impinging liquid droplet through a temperature gradient normal to the solid surface. The temperature gradient acts to alter the surface tension along the structures normal to the surface resulting in possible de-wetting. The experiments consisted of 2 mm water and glycerol droplets held at room temperature impinging on heated micro-post superhydrophobic surfaces. The surface temperature was varied between 50 and 90 ◦ C and the height of the micro-posts between 8 and 18 µm. The results show that hotter surface temperatures and taller posts allow for a nearly complete Cassie to Wenzel transition on the order of 1 ms, thus droplets are able to rebound without sticking to the surface. 2:28PM D13.00002 Direct measurement of the interaction forces during the sliding of a water droplet on a pillar-typed superhydrophobic surface , THANH-VINH NGUYEN, HIDETOSHI TAKAHASHI, KIYOSHI MATSUMOTO, ISAO SHIMOYAMA, Univ of Tokyo — We directly measure the interaction forces between a sliding droplet and a single microstructure of a pillar-typed superhydrophobic surface. The measurement is realized using a MEMS (Micro Electro Mechanical Systems)-based two-axis force sensor fabricated under the micropillar. We find that, during the sliding of a water droplet on the micropillar array, the pillar is pulled upward when it meets the advancing and receding edges of the droplet as a result of the normal component of surface tension. The maximum value of the pulling force at the receding edge is larger than that at the advancing edge due to the difference in dynamic contact angles at the receding and advancing edges. Meanwhile, over the inner region of the contact area, the pillar is pushed down by the liquid pressure. Moreover, measured shear force shows that the friction during the droplet sliding is dominated by the adhesion of the droplet at the receding. Finally, we demonstrate that as the volume of the droplet increases, the normal force over the inner region of the contact area decreases while the forces at the receding edges do not change significantly. 2:41PM D13.00003 Superhydrophobic frictions , TIMOTHEE MOUTERDE, PASCAL RAUX, CHRISTOPHE CLANET, DAVID QUERE, (PMMH, ESPCI / LadHyX, Ecole Polytechnique) — We discuss the nature of the friction opposing the motion of drops running on superhydrophobic materials. Despite the high mobility of the drops (which are typically 100 times faster than on usual solids), friction is always found to have a viscous origin. For highly viscous liquids, we recover a Mahadevan-Pomeau regime for which the drop speed is inversely proportional to the viscosity. In the opposite limit (e.g. water), friction depends on the material texture and it is interpreted as resulting either from the development of a viscous boundary layer at the solid/liquid interface, or from the shear inside the subjacent air cushion. In the latter case, air is not only responsible for the existence of superhydrophobic states, but also for the resistance limiting speed of water in these states. 2:54PM D13.00004 Droplet Train Impingement on Superhydrophobic Surfaces , JONATHAN STODDARD, JULIE CROCKETT, DANIEL MAYNES, Brigham Young University — The dynamics of a droplet train impinging on a superhydrophobic (SH) surface is investigated. The surfaces are patterned with either post or ribbed microfeatures and coated with Teflon to render them hydrophobic. The height of the features are nominally 15 microns and the spacing for the post and ribbed surfaces are approximately 16.5 microns and 32 microns respectively. Droplets at the induced frequencies of 600-4600 Hz with sizes varying from 0.7 to 1.5 mm in diameter are forced to impinge on these SH surfaces. When each individual droplet impinges on the surface, a crown forms which spreads out radially until reaching a semi-stable or regularly oscillating crater radius. At this point the water either builds up or breaks up into droplets. In some cases the crown may breakup before reaching the crater, causing splashing. We characterize the occurrence of these dynamics and quantify the crater radius over a Weber number range of 70-1450. In addition we compare the crater radius to the breakup radius of a uniform jet with equivalent momentum and describe the transition dynamics from droplet train impingement to uniform laminar jet impingement on a SH surface. 3:07PM D13.00005 Puddle Jumping , ANDREW WOLLMAN, Portland State University, TREVOR SNYDER, 3D Systems, MARK WEISLOGEL, Portland State University — Rebounding droplets from superhydrophobic surfaces have attracted significant public and scientific attention because they are both enjoyable as well as industrially relevant. Demonstrations of bouncing droplets with volumes between 0.003 and 0.03 ml are common in the literature and limited primarily by gravity. In this presentation we demonstrate large droplet “rebounds” made possible by low-gravity testing in a drop tower. The up to 300 ml drops are best described as puddles that launch in a nearly identical manner to rebounding drops 4 orders of magnitude smaller in volume. A variety of jumping liquid and gas puddles are shown including puddles of highly specified and unusual initial geometry. The large length sales of the capillary fluidic surfaces ∼ O(10 cm) enable 3D printing of all superhydrophobic surface topologies demonstrated. In addition, we demonstrate such puddle jumping as a passive drop-on-demand technique for large low-gravity drop dynamics investigations; such as collisions, rebounds, heat and mass transfer, and containerless possessing. 3:20PM D13.00006 Robust and Drain Resistant Lubricated Omniphobic Fabrics1 , CASSIDEE KIDO2 , VIRAJ DAMLE, XIAODA SUN, AJAY ROOPESH, KYLE DOUDRICK3 , KONRAD RYKACZEWSKI, Arizona State University — The implications of omniphobic fabrics range from stainproof clothing to civilian and military protection from chemical weapons. The challenge comes in developing a product that remains effective in repelling droplets of liquids with a wide range of surface tensions even after being subjected to various stimuli imposed by human use. Omniphobic fabrics can be made by infusing hydrophobic nanoparticle coated fibers with a low surface energy lubricant [1]. These types of lubricant impregnated surfaces can shed large deposited droplets as well as condensed microdroplets of variety of low surface tension liquids [2]. However, here we show that lubricated omniphobic fabrics can easily lose their properties due to degradation of the nanostructure coating or drainage of the lubricant upon contact with a porous surface. We also demonstrate that this issue can be resolved with use of cross-linked polymer coated fibers that are swollen with the lubricant. Use of flexible polymers avoids structure degradation due to fabric deformation, while swelling of the polymer with lubricant minimizes lubricant drainage upon contact maintaining the omniphobic characteristics of the fabric. [1] Shillingford, C. et al. Nanotechnol. 2014, 25, 014019. [2] Rykaczewski, K et al. Sci. Rep 2014, 4. 1 KR acknowledges startup funding from ASU and collaborative effort with Dr. Tim Burgin and James R. Lee from Naval Surface Warfare Center Dahlgren Division. 2 currently at Duke University 3 currently at University of Notre Dame 3:33PM D13.00007 Behavior of severely supercooled water drops impacting on superhydrophobic surfaces1 , TANMOY MAITRA, CARLO ANTONINI, Laboratory of Thermodynamics in Emerging Technologies, Mechanical and Process Engineering Department, ETH Zurich, MANISH K. TIWARI, Department of Mechanical Engineering, University College London, ADRIAN MULARCZYK, ZULKUFLI IMERI, PHILIPPE SCHOCH, DIMOS POULIKAKOS, Laboratory of Thermodynamics in Emerging Technologies, Mechanical and Process Engineering Department, ETH Zurich — Surface icing, commonplace in nature and technology, has broad implications to daily life. To prevent surface icing, superhydrophobic surfaces/coatings with rationally controlled roughness features (both at micro and nano-scale) are considered to be a promising candidate. However, to fabricate/synthesize a high performance icephobic surface or coating, understanding the dynamic interaction between water and the surface during water drop impact in supercooled state is necessary. In this work, we investigate the water/substrate interaction using drop impact experiments down to -17◦ C. It is found that the resulting increased viscous effect of water at low temperature significantly affects all stages of drop dynamics such as maximum spreading, contact time and meniscus penetration into the superhydrophobic texture. Most interestingly, the viscous effect on the meniscus penetration into roughness feature leads to clear change in the velocity threshold for rebounding to sticking transition by 25% of supercooled drops. 1 Swiss National Science Foundation (SNF) grant 200021 135479. 3:46PM D13.00008 Delayed frost formation on hybrid nanostructured surfaces with patterned high wetting contrast , YOUMIN HOU, PENG ZHOU, SHUHUAI YAO, Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Hong Kong — Engineering icephobic surfaces that can retard the frost formation and accumulation are important to vehicles, wind turbines, power lines, and HVAC systems. For condensation frosting, superhydrophobic surfaces promote self-removal of condensed droplets before freezing and consequently delay the frost growth. However, a small thermal fluctuation may lead to a Cassie-to-Wenzel transition, and thus dramatically enhance the frost formation and adhesion. In this work, we investigated the heterogeneous ice nucleation on hybrid nanostructured surfaces with patterned high wetting contrast. By judiciously introducing hydrophilic micro-patches into superhydrophobic nanostructured surface, we demonstrated that such a novel hybrid structure can efficiently defer the ice nucleation as compared to a superhydrophobic surface with nanostructures only. We observed efficient droplet jumping and higher coverage of droplets with diameter smaller than 10 µm, both of which suppress frost formation. The hybrid surface avoids the formation of liquid-bridges for Cassie-to-Wenzel transition, therefore eliminating the ‘bottom-up’ droplet freezing from the cold substrate. These findings provide new insights to improve anti-frosting and anti-icing by using heterogeneous wettability in multiscale structures 3:59PM D13.00009 Bioinspired Antifreeze Secreting Frost-Responsive Pagophobic Coatings1 , XIAODA SUN, VIRAJ DAMLE, KONRAD RYKACZEWSKI, Arizona State University — Prevention of ice and frost accumulation is of interest to transportation, power generation, and agriculture industries. Superhydrophobic and lubricant impregnated pagophobic coatings have been proposed, however, they both fail in frosting conditions [1, 2]. Inspired by functional liquid secretion in natural systems, such as toxin secretion by poison dart frost in response to predator presence, we developed frost-responsive antifreeze secreting pagophobic coatings. These are bi-layered coatings with an inner superhydrophilic “dermis” infused with antifreeze and an outer permeable superhydrophobic “epidermis.” The superhydrophobic epidermis separates the antifreeze from the environment and prevents ice accumulation by repelling impinging water droplets. In frosting conditions, the antifreeze is secreted from the dermis through pores in the epidermis either due to contact with condensed droplets or temporary switch of the epidermis wettability from hydrophobic to hydrophilic caused by surface icing. Here we demonstrate superior performance of this multifunctional coating in simulated frosting, freezing mist/fog, and freezing spray/rain conditions. [1] Varanasi et al., App. Phys. Lett., 97, 2010. [2] Rykaczewski et al., Langmuir, 29, 2013. 1 KR acknowledges startup funding from ASU. Sunday, November 23, 2014 2:15PM - 4:25PM Session D15 Focus Session: Respiratory Bio-Fluid Dynamics I Michigan — 3022/3024 - James Grotberg, University of 2:15PM D15.00001 Impact of the Equation of State in Models for Surfactant Spreading Experiments1 , RACHEL LEVY, Harvey Mudd College — Pulmonary surfactant spreading models often rely on an equation of state relating sur- factant concentration to surface tension. Mathematically, these models have been analyzed with simple functional relationships. However, to model an experiment with a given fluid and surfactant, a physically meaningful equation of state can be derived from experimentally obtained isotherms. We discuss the comparison between model and experiment for NBD-PC lipid (surfactant) spreading on glycerol for an empirically-determined equation of state, and compare those results to simulations with traditionally employed functional forms. In particular we compare the timescales by tracking the leading edge of surfactant, the central fluid height and dynamics of the Marangoni ridge. We consider both outward spreading of a disk-shaped region of surfactant and the hole-closure problem in which a disk-shaped surfactant-free region self-heals. 1 Support from NSF-DMS-FRG 0968154, RCSA-CCS-19788, and HHMI. 2:28PM D15.00002 Large-eddy Simulation of Heat and Water Vapor Transfer in CT-Based Human Airway Models1 , DAN WU, Department of Mechanical and Industrial Engineering, The University of Iowa, MERRYN TAWHAI, Auckland Bioengineering Institute, The University of Auckland, ERIC HOFFMAN, Medicine and Radiology, The University of Iowa, CHING-LONG LIN, Department of Mechanical and Industrial Engineering, The University of Iowa — We propose a novel imaging-based thermodynamic model to study local heat and mass transfers in the human airways. Both 3D and 1D CFD models are developed and validated. Large-eddy simulation (LES) is adopted to solve 3D incompressible Navier-Stokes equations with Boussinesq approximation along with temperature and water vapor transport equations and energy-flux based wall boundary condition. The 1D model provides initial and boundary conditions to the 3D model. The computed tomography (CT) lung images of three healthy subjects with sinusoidal waveforms and minute ventilations of 6, 15 and 30 L/min are considered. Between 1D and 3D models and between subjects, the average temperature and water vapor distributions are similar, but their regional distributions are significantly different. In particular, unlike the 1D model, the heat and water vapor transfers in the 3D model are elevated at the bifurcations during inspiration. Moreover, the correlations of Nusselt number (Nu) and Sherwood number (Sh) with local Reynolds number and airway diameter are proposed. In conclusion, use of the subject-specific lung model is essential for accurate prediction of local thermal impacts on airway epithelium. 1 Supported in part by NIH grants R01-HL094315, U01-HL114494 and S10-RR022421. 2:41PM D15.00003 Development of a 3D to 1D Particle Transport Model to Predict Deposition in the Lungs1 , JESSICA M. OAKES, INRIA Paris-Rocquncourt, UC Berkeley, CELINE GRANDMONT, INRIA Paris-Rocquncourt, UPMC Universite Paris 6, SHAWN C. SHADDEN, UC Berkeley, IRENE E. VIGNON-CLEMENTEL, INRIA Paris-Rocquncourt, UPMC Universite Paris 6 — Aerosolized particles are commonly used for therapeutic drug delivery as they can be delivered to the body systemically or be used to treat lung diseases. Recent advances in computational resources have allowed for sophisticated pulmonary simulations, however it is currently impossible to solve for airflow and particle transport for all length and time scales of the lung. Instead, multi-scale methods must be used. In our recent work, where computational methods were employed to solve for airflow and particle transport in the rat airways (Oakes et al. (2014), Annals of Biomedical Engineering, 42: 899-914), the number of particles to exit downstream of the 3D domain was determined. In this current work, the time-dependent Lagrangian description of particles was used to numerically solve a 1D convection-diffusion model (trumpet model, Taulbee and Yu (1975), Journal of Applied Physiology, 38: 77-85) parameterized specifically for the lung. The expansion of the airway dimensions was determined based on data collected from our aerosol exposure experiments (Oakes et al. (2014), Journal of Applied Physiology, 116: 1561-8). This 3D-1D framework enables us to predict the fate of particles in the whole lung. 1 This work was supported by the Whitaker Foundation at the IIE, a INRIA Associated Team Postdoc Grant, and a UC Presidential Fellowship 2:54PM D15.00004 Oscillatory Flow in the Human Airways from the Mouth through Several Bronchial Generations , ANDREW BANKO, Stanford University, FILIPPO COLETTI, University of Minnesota, CHRIS ELKINS, JOHN EATON, Stanford University — The time-varying flow is studied experimentally in an anatomically accurate model of the human airways from the mouth through the fourth to eighth generation of the bronchi. The airway geometry is obtained from the CT scan of a healthy adult male of normal height and build. The three-component, three-dimensional mean velocity field is obtained throughout the entire model using phase-locked magnetic resonance velocimetry. A pulsatile pump drives a sinusoidal waveform (inhalation and exhalation) with frequency and stroke-length such that the mean trachea Reynolds number at peak inspiration is Re = 4200 and the Womersley number is α = 7. This represents a regime of moderate exertion. Integral parameters are defined to quantify the degree of velocity profile non-uniformity (which correlates with axial dispersion) and secondary flow strength (which correlates with lateral dispersion). It is found that the streamwise momentum flux and secondary flow strength increase and decrease in proportion throughout most of the breathing cycle. On the other hand, the strength of secondary flows during the 10% of the breathing cycle surrounding flow reversal remains approximately half of that at peak inspiration while the streamwise momentum flux goes to zero. The strong and persistent secondary flows have important implications for dispersion of scalar or particulate contaminants in the lungs. 3:07PM D15.00005 Coupling of the interfacial and bulk flow in a knife-edge surface viscometer1 , ADITYA RAGHUNANDAN, CHRISTOPHER TILGER, AMIR HIRSA, Rensselaer Polytechnic Institute, JUAN LOPEZ, Arizona State University — After more than 50 years of investigating how to measure surface (excess) shear viscosity, it remains a controversial issue. The complications stem from the fact that to measure a viscous response the system needs to be flowing, and for a surface film on a liquid substrate this means that the liquid in the bulk will also be flowing. Macroscale measurements, which generally provide greater accuracy, are often made at Reynolds numbers that are large enough for inertia to be non-negligible. However the theoretical models against which the measurements are compared have so far failed to properly account for the coupling. Results will be presented from a numerical study on the coupling between the interfacial and bulk flow. Also, experimental results will be presented for the lung surfactant component DPPC. 1 NASA grant NNX13AQ22G, NSF grants CBET-1064644 and CBET-1064498 3:20PM D15.00006 An exact solution for Stokes flow in an infinite channel with permeable walls1 , GREGORY HERSCHLAG, JIAN-GUO LIU, ANITA LAYTON, Duke University — We derive an exact solution for Stokes flow in an infinite channel with permeable walls. We assume that at the channel walls, the normal component of the fluid velocity is described by Darcy’s law and the tangential component of the fluid velocity is described by the no slip condition. The pressure exterior to the channel is assumed to be constant. We verify that in the limit of small permeability, Poiseuille flow is recovered to leading order, and demonstrate that our exact result agrees with previous approximate results in this limit. By comparing our solution to existing assumptions on inlet profiles in the literature, we find that although the error is small, Poiseuille and Berman flow do not provide correct inlet conditions. 1 DK089066,DMS1263995,DMS0943760,DMS1107444 3:33PM D15.00007 Droplets and modes of respiratory disease transmission , LYDIA BOUROUIBA, Massachusetts Institute of Technology — Direct observation of violent expirations such as sneezes and coughs events reveal that such flows are multiphase turbulent buoyant clouds with suspended droplets of various sizes. The effects of ambient conditions indoors, such as moisture and temperature, coupled with the water content of such clouds are key in shaping the pathogen footprint emitted by potentially sick individuals. Such pathogen footprint can change the patterns of respiratory disease transmission. We discuss how the fluid dynamics of violent expirations can help inform how. 3:46PM D15.00008 Biasing left-right particle distribution via sideways bending of the upper body , JORGE A. BERNATE, ELEANOR LIN, REBECCA FAHRIG, CARLOS MILLA, GIANLUCA IACCARINO, ERIC S.G. SHAQFEH, Stanford University — The ability to target therapeutic aerosols to specific regions of the lungs would result in more effective treatment of localized pulmonary diseases and may also prove beneficial in systemic delivery via the airways. Previous computational and experimental studies have shown that large particles disproportionately enter the left lung. The observed uneven distribution occurs because the trachea bends to the right just before the first bifurcation, causing particles with sufficient inertia to enter the left main bronchus. Via CT imaging, we have shown that it is possible to modify the normal configuration of the trachea by bending sideways. Bending to the right and left results in configurations in which the trachea monotonically and smoothly bends to the first bifurcation. In the left-bent configuration, inertial particles will tend to accumulate towards the right side of the trachea and enter the right main bronchus, and conversely for the right-bent configuration. In this talk, we will present our results of Large-Eddy simulations and particle tracking showing regional deposition and ventilation as a function of the Reynolds and Stokes numbers for realistic models of the upright and bent configurations of an adult human subject. 3:59PM D15.00009 Alveolar flows of the developing lungs:from embryonic to early childhood airways , JANNA TENENBAUM-KATAN, PHILIPP HOFEMEIER, RAMI FISHLER, Technion - Israel Institute of Technology, BARBARA ROTHEN- RUTISHAUSER, Adolphe Merkle Institute, University of Fribourg, JOSUE SZNITMAN, Technion - Israel Institute of Technology — At the onset of life in utero the respiratory system is simply a liquid-filled duct. With our first breath, alveoli are filled with air and become a significant port of entry for airborne particles. As such, alveolar lining is nearly fully functional at birth, though lung development continues during childhood as structural changes increase alveolar surface area to optimize ventilation. We hypothesize that such fluid dynamical changes potentially affect two phenomena occurring within alveoli: (i) flow patterns in airspaces at distinct stages of both in- and ex-utero life and (ii) fate of inhaled particles ex-utero. To investigate these phenomena, we combine experimental and numerical approaches where (i) microfluidic in vitro devices mimic liquid flows across the epithelium of fetal airspaces, and (ii) computational simulations are employed to examine particle transport and deposition in the deep alveolated regions of infants’ lungs. Our approaches capture anatomically-inspired geometries based on morphometrical data, as well as physiological flows, including the convective-diffusive nature of submicron particle transport in alveolar regions.Overall, we investigate respiratory flows in alveolar regions of developing lungs, from early embryonic stages to late childhood 4:12PM D15.00010 An automatic generation of non-uniform mesh for CFD analyses of imagebased multiscale human airway models1 , SHINJIRO MIYAWAKI, University of Iowa, MERRYN H. TAWHAI, University of Auckland, ERIC A. HOFFMAN, CHING-LONG LIN, University of Iowa — The authors have developed a method to automatically generate non-uniform CFD mesh for image-based human airway models. The sizes of generated tetrahedral elements vary in both radial and longitudinal directions to account for boundary layer and multiscale nature of pulmonary airflow. The proposed method takes advantage of our previously developed centerline-based geometry reconstruction method. In order to generate the mesh branch by branch in parallel, we used the open-source programs Gmsh and TetGen for surface and volume meshes, respectively. Both programs can specify element sizes by means of background mesh. The size of an arbitrary element in the domain is a function of wall distance, element size on the wall, and element size at the center of airway lumen. The element sizes on the wall are computed based on local flow rate and airway diameter. The total number of elements in the non-uniform mesh (10 M) was about half of that in the uniform mesh, although the computational time for the non-uniform mesh was about twice longer (170 min). The proposed method generates CFD meshes with fine elements near the wall and smooth variation of element size in longitudinal direction, which are required, e.g., for simulations with high flow rate. 1 NIH grants R01-HL094315, U01-HL114494, and S10-RR022421. Computer time provided by XSEDE. Sunday, November 23, 2014 2:15PM - 4:25PM Session D16 Free-Surface Flows II: Waves — 2000 - Andrew Belmonte, Pennsylvania State University 2:15PM D16.00001 Shallow fluids meets Einstein: an experimental geodesic flow on a curved space , JAY JOHNSON, JEAN-LUC THIFFEAULT, University of Wisconsin - Madison — When a shallow layer of inviscid fluid flows over a smoothlypatterned substrate, the fluid particle trajectories are, to leading order in the layer thickness, geodesics on the two-dimensional curved space of the substrate. We use 3D-printed substrates to show that the pattern made by a jet striking a bumpy surface is described by the geodesic equation. Because the geodesic equation is fourth order, the geodesics are chaotic even for simple substrates. 2:28PM D16.00002 Controlled Pattern Selection of Free Surface Waves , CHUN-TI CHANG, SUSAN DANIEL, PAUL H. STEEN, Cornell University — In this experimental study, we investigate the resonance of surface waves subject to different geometric constraints. For liquid puddles with different footprints and depths, we experimentally probe and compare their dynamics of pattern selection. From the scientific perspective, the comparison relates resonance of sessile drops to Faraday waves. For technological development, the study provides guidelines for applications such as ordered self-assembly of nanoparticles, droplet transport, drop atomization, enhanced mixing, and suspension collection. 2:41PM D16.00003 Capillary Korteweg-de Vries solitons on a levitated cylinder1 , CHI-TUONG PHAM, LIMSI, CNRS & Universite Paris-Sud, STEPHANE PERRARD, CHARLES DUCHENE, Laboratoire MSC, Universite Paris-Diderot, LUC DEIKE, Laboratoire MSC, Universite Paris-Diderot, and Scripps Institution of Oceanography, UCSD — A water cylinder is deposited on a straight channel heated far above boiling temperature so that the water levitates above its own vapor owing to Leidenfrost effect. Our setup allows us to study the one-dimensional propagation of surface waves. We show that the dispersion relation of linear waves follows that of gravity-capillary waves under a dramatically reduced gravity (up to a factor 30), yielding an effective capillary length larger than one centimeter. Nonlinear capillary depression solitary waves propagate without deformation and undergo mutual collisions and reflections at the boundaries of the domain. Their typical width and their amplitude-dependent velocity are in very good agreement with theoretical predictions based on Korteweg-de Vries equation. 1 Supported by ANR-11-BS04-001-01 (FREEFLOW project). 2:54PM D16.00004 Mach-like capillary-gravity wakes , MARC RABAUD, FREDERIC MOISY, Universite Paris-Sud — We determine experimentally the angle α of maximum wave amplitude in the far-field wake behind a vertical surface-piercing cylinder translated at constant velocity U for Bond numbers BoD = D/λc ranging between 0.1 and 4.2, where D is the cylinder diameter and λc the capillary length. In all cases the wake angle is found to follow a Mach-like law at large velocity, α ∼ U −1 , but with different prefactors depending on the value of BoD . For small BoD (large capillary effects), the wake angle approximately follows the law α ≃ cg,min /U , where cg,min is the minimum group velocity of capillary-gravity waves. For larger √ BoD (weak capillary effects), we recover the law α ∼ gD/U found for ship wakes at large velocity. Using the general property of dispersive waves that the characteristic wavelength of the wavepacket emitted by a disturbance is of order of the disturbance size, we propose a simple model that describes the transition between these two Mach-like regimes as the Bond number is varied. This model, complemented by numerical simulations of the surface elevation induced by a moving Gaussian pressure disturbance, is in good agreement with experimental measurements. 3:07PM D16.00005 Chaotic cross-waves in tanks of finite sizes , TATYANA KRASNOPOLSKAYA, EVGENIY PECHUK, VIACHESLAV SPEKTOR, Institute of Hydromechanics National Academy of Sciences of Ukraine — The phenomenon of chaotic cross-waves generation in fluid free surface in two finite size containers is studied. The waves may be excited by harmonic axisymmetric deformations of the inner shell in the volume between two cylinders and in a rectangular tank when one wall is a flap wavemaker. The existence of chaotic attractors was established for the dynamical system presenting cross-waves and forced waves interaction at fluid free-surface in a volume between two cylinders of finite length. In the case of one cross-wave in a rectangular tank no chaotic regimes were found. 3:20PM D16.00006 Dynamic square superlattice of Faraday waves , LYES KAHOUADJI, PMMH-CNRS-ESPCI, France, JALEL CHERGUI, DAMIR JURIC, LIMSI-CNRS, France, SEUNGWON SHIN, Hongik University, Seoul, Korea, LAURETTE TUCKERMAN, PMMHCNRS-ESPCI, France — Faraday waves are computed in a 3D container using BLUE, a code based on a hybrid Front-Tracking/Level-set algorithm for Lagrangian tracking of arbitrarily deformable phase interfaces. A new dynamic superlattice pattern is described which consists of a set of square waves arranged in a two-by-two array. The corners of this array are connected by a bridge whose position oscillates in time between the two diagonals. 3:33PM D16.00007 A theory for stationary polygonal hydraulic jumps , ASLAN KASIMOV, KAUST — When a vertical jet of viscous fluid strikes a horizontal plate, a circular hydraulic jump occurs at some distance from the jet impact point. Under certain conditions, the circular symmetry of the jump breaks and gives rise to stationary or rotating polygonal patterns. We describe experimental observations of the symmetry breaking and propose a model for the structure of the polygonal jumps. 3:46PM D16.00008 Hydraulic jumps in partially closed containers , ANDREW BELMONTE, VISHAL VASAN, Pritchard Labs, Dept of Mathematics, Pennsylvania State University — We present results of experiments investigating the effect of far field boundary conditions on the hydraulic jump in water. The classic hydraulic jump is an axisymmetric flow characterized by a single radial transition of the fluid height, for which the far field depth of the water is a key parameter. With suitable choices of the flow parameters, the circular jump exhibits symmetry breaking, transitioning into polygonal jumps among other possibilities. Here we study the transition between the jumps by suitably controlling the far field condition. This permits the flow to sustain a quasi-steady transition state between circular and polygonal jumps. Further, we investigate the effect of non-axisymmetric boundary conditions on the jump and its stability. 3:59PM D16.00009 Reovering water-wave profiles from bottom pressure measurements1 , VISHAL VASAN, Pennsylvania State Univ, KATIE OLIVERAS, Seattle University, DIANE HENDERSON, Pennsylvania State Univ, BERNARD DECONINCK, University of Washington — Accurate measurements of the surface elevation are essential for understanding flow along coastlines. Often surface elevation is measured indirectly through pressure gauges situated on the bottom bed using linear theory. The full relationship between pressure and surface elevation is however significantly more complicated. In this talk we present a fully nonlinear formula that recovers the surface elevation profile of a traveling water-wave from measurements of the pressure beneath the wave. This is the first analytical investigation to take full nonlinearity into account. From this new relation, we derive a variety of different asymptotic formulas. Surface profile reconstructions from bottom pressure, measured using pressure gauges, are compared to actual heights obtained from surface capacitance gauges. Our comparisons indicate that a new asymptotic reconstruction formula affords significant gains over the traditional approach. Further, it is rapid and easy to implement, requiring only three Fourier transforms. 1 This research was supported by the National Science Foundation under grants NSF-DMS-1008001, NSF-DMS-0708352 and NSF-DMS-1107379. 4:12PM D16.00010 Quantification and prediction of rare events in nonlinear waves , THEMISTOKLIS SAPSIS, WILL COUSINS, MUSTAFA MOHAMAD, MIT — The scope of this work is the quantification and prediction of rare events characterized by extreme intensity, in nonlinear dispersive models that simulate water waves. In particular we are interested for the understanding and the short-term prediction of rogue waves in the ocean and to this end, we consider 1-dimensional nonlinear models of the NLS type. To understand the energy transfers that occur during the development of an extreme event we perform a spatially localized analysis of the energy distribution along different wavenumbers by means of the Gabor transform. A stochastic analysis of the Gabor coefficients reveals i) the low-dimensionality of the intermittent structures, ii) the interplay between non-Gaussian statistical properties and nonlinear energy transfers between modes, as well as iii) the critical scales (or Gabor coefficients) where a critical energy can trigger the formation of an extreme event. The unstable character of these critical localized modes is analysed directly through the system equation and it is shown that it is defined as the result of the system nonlinearity and the wave dissipation (that mimics wave breaking). These unstable modes are randomly triggered through the dispersive “heat bath” of random waves that propagate in the nonlinear medium. Using these properties we formulate low-dimensional functionals of these Gabor coefficients that allow for the prediction of extreme event well before the strongly nonlinear interactions begin to occur. The prediction window is further enhanced by the combination of the developed scheme with traditional filtering schemes. Sunday, November 23, 2014 2:15PM - 4:25PM Session D17 Nonlinear Dynamics II: Coherent Structures II — 2002 - Wenbo Tang, Arizona State University 2:15PM D17.00001 An accurate computation of the flow map gradient , SIAVASH AMELI, SHAWN SHADDEN, UC Berkeley — The flow map gradient (tangent map) is often used in dynamical systems for computation of Lagrangian coherent structures. In more sophisticated methods it is important to recover the complete spectrum of this operator, as well as the eigenvectors. Traditional methods to compute the tangent map using finite differencing often fail in accurately computing these quantities. Due to nonlinear effects of the flow, perturbations of trajectories mapped forward by the tangent map may grow excessively and they collapse on the dominant eigenvector of the map. We describe alternative techniques to overcome these issues. Both continuous or discrete QR factorization and singular value decompositions are used to automatically carry out computation of Lyapunov exponents and directions. Results on sum of Lyapunov exponents for divergent free flow, as well as sensitivity to integration time are compared in contrast to previous methods. 2:28PM D17.00002 Variability of reaction in chaotic flows - an approach based on Lagrangian coherent structures1 , WENBO TANG, CHRISTOPHER LUNA, ADITYA DHUMUNTARAO, Arizona State University — The study of reactive- diffusive systems in the presence of background flows is an important problem of biological, geophysical and engineering interest. The coupling between stirring and reaction brings new complexity, which may lead to strong variability of the outcome of reaction, as compared to homogeneous reaction processes. In this talk, I will discuss several examples of reaction processes, whose variability can be tied to Lagrangian coherent structures (LCS), the deterministic tool developed to address passive scalar transport in chaotic flows. We find that LCS plays different roles in different reaction processes, but the overall strategy to approach such problems has a unified theme. 1 Thanks to: NSF DMS-1212144 2:41PM D17.00003 Effects on finite-time scalar statistics by partitioning metric1 , PHILLIP WALKER, WENBO TANG, Arizona State University — When partitioning a nonlinear aperiodic dynamic system into different regions identified by Lagrangian coherent structures (LCS) there are two approaches, geometric and probabilistic; each offering a handful of different metrics. We consider stochastic scalar dispersion associated with LCS and compare the statistics of the separate flow partitions as identified by several partitioning methods. The differences of the resident time curves between methods indicate the effectiveness of that partitioning method for objectively partitioning the flow into topologically distinct regions. In this talk we explore such correlation between methods and statistics, and effective mixing. 1 Thanks to: NSF DMS-1212144 2:54PM D17.00004 Getting Things Sorted With Lagrangian Coherent Structures , SEVERINE ATIS, THOMAS PEACOCK, Massachussets Institute of Technology; EndLab, 77 Massachusetts Avenue, Cambridge MA 02139, ENVIRONMENTAL DYNAMICS LABORATORY TEAM — The dispersion of a tracer in a fluid flow is influenced by the Lagrangian motion of fluid elements. Even in laminar regimes, the irregular chaotic behavior of a fluid flow can lead to effective stirring that rapidly redistributes a tracer throughout the domain. For flows with arbitrary time-dependence, the modern approach of Lagrangian Coherent Structures (LCSs) provide a method for identifying the key material lines that organize flow transport. When the advected tracer particles possess a finite size and nontrivial shape, however, their dynamics can differ markedly from passive tracers, thus affecting the dispersion phenomena. We present details of numerical simulations and laboratory experiments that investigate the behavior of finite size particles in 2-dimensional chaotic flows. We show that the shape and the size of the particles alter the underlying LCSs, facilitating segregation between tracers of different shape in the same flow field. 3:07PM D17.00005 Lagrangian Coherent Structures are templates for reaction initiation between initially distant scalars1 , KENNETH PRATT, JAMES MEISS, JOHN CRIMALDI, Univ of Colorado - Boulder — Lagrangian Coherent Structures (LCS) are shown to be effective templates for the location of reactions between initially distant scalars in 2D flows. Computations of reactions and finite-time Lyapunov exponent (FTLE) fields demonstrate that reactions are initiated when the scalars come into contact on a common FTLE ridge at a time that depends upon the initial condition. To show robustness of the phenomenon, a hierarchical set of three numerical flows is used: the periodic wake downstream of a stationary cylinder, a chaotic double gyre flow, and a chaotic, aperiodic flow consisting of interacting Taylor vortices. Coalescence of highly concentrated filaments leads to transient reaction rates that are orders of magnitude greater than predicted by the well-mixed state. As a consequence, we show that chaotic flows, known for their ability to efficiently dilute scalars, also have the competing effect of organizing initially distant scalars along the LCS at timescales shorter than that required for dilution. 1 Supported by National Science Foundation Grant No. 1205816 3:20PM D17.00006 Lagrangian coherent structures and the dynamics of inertial particles , SUDHARSAN MADHAVAN, STEVEN BRUNTON, JAMES RILEY, University of Washington — In this work we investigate the dynamics of inertial particles using the finite-time Lyapunov exponent (FTLE). In particular, we analyze preferential concentration of particles with nonzero Stokes number, St, and varying density ratio, R, for the double gyre vector field. We find that heavy particles (aerosols) tend to accumulate strongly onto negative-time (attracting) FTLE ridges of the non-inertial fluid particles, while lighter particles (bubbles) tend to repel from these ridges and accumulate at vortex cores. The transition of the negative-time FTLE ridges from attractors to repellers, based on the value of R, partially explains the preferential concentration of inertial particles. We also investigate the inertial finite-time Lyapunov exponent (iFTLE) based on the trajectories of inertial particles. The iFTLE is used to quantify the effect of St and R on particle stirring, and we present preliminary results establishing a connection between iFTLE and the two-point dispersion. Finally, we analyze the low-pass filtering effect of Stokes number on particle trajectories. 3:33PM D17.00007 Search strategy in a complex and dynamic environment (the Indian Ocean case) , SOPHIE LOIRE, HASSAN ARBABI, PATRICK CLARY, UCSB, STEFAN IVIC, NELIDA CRNJARIC-ZIC, SENKA MACESIC, BOJAN CRNKOVIC, University of Rijeka, IGOR MEZIC, UCSB, UCSB TEAM, RIJEKA TEAM — The disappearance of Malaysia Airlines Flight 370 (MH370) in the early morning hours of 8 March 2014 has exposed the disconcerting lack of efficient methods for identifying where to look and how to look for missing objects in a complex and dynamic environment. The search area for plane debris is a remote part of the Indian Ocean. Searches, of the lawnmower type, have been unsuccessful so far. Lagrangian kinematics of mesoscale features are visible in hypergraph maps of the Indian Ocean surface currents. Without a precise knowledge of the crash site, these maps give an estimate of the time evolution of any initial distribution of plane debris and permits the design of a search strategy. The Dynamic Spectral Multiscale Coverage search algorithm is modified to search a spatial distribution of targets that is evolving with time following the dynamic of ocean surface currents. Trajectories are generated for multiple search agents such that their spatial coverage converges to the target distribution. Central to this DSMC algorithm is a metric for the ergodicity. 3:46PM D17.00008 Untangling tracer trajectories and clarifying coherence in 2D flows using braid theory , MARGAUX FILIPPI, SÉVERINE ATIS, Massachusetts Inst of Tech-MIT, JEAN-LUC THIFFEAULT, MARKO BUDIŠIĆ, University of Wisconsin-Madison, MICHAEL ALLSHOUSE, University of Texas-Austin, THOMAS PEACOCK, Massachusetts Inst of Tech-MIT — Interpreting ocean surface transport is crucial to many areas of oceanography, ranging from marine ecology to pollution management. To better understand surface mixing, we investigate a braid theory method to detect transport barriers bounding coherent structures in two-dimensional fluid flows. Whereas most existing techniques rely on an extensive spatiotemporal knowledge of the flow field, we seek to identify these structures from sparse data sets involving trajectories of a few tracer particles or floats. We present the results of model and laboratory experimental studies to test the robustness and applicability of the braid theory method, and discuss the potential applicability to oceanic data sets. 3:59PM D17.00009 Locating coherent material vortices in three-dimensional unsteady flows , DAVID OETTINGER, DANIEL BLAZEVSKI, GEORGE HALLER1 , ETH Zurich — Recent work has shown that coherent material vortices in two-dimensional unsteady flows are bounded by closed stationary curves of the averaged material strain [1]. These material vortex boundaries are objective (frame-invariant) Lagrangian coherent structures (LCSs) of the elliptic type, which turn out to stretch uniformly under the flow. We extend this approach to three-dimensional unsteady flows to locate toroidal and cylindrical material vortex boundaries as two-dimensional, elliptic LCS surfaces. We provide a detailed numerical procedure building on the approach in [2] and discuss several examples. [1] G. Haller and F.J. Beron-Vera, Coherent Lagrangian vortices: The black holes of turbulence. J. Fluid Mech. 731 (2013) R4 [2] D. Blazevski and G.Haller, Hyperbolic and elliptic transport barriers in three-dimensional unsteady flows, Physica D, 273–274 (2014), 46-62. 1 Full address: Institute for Mechanical Systems, ETH Zurich, Tannenstrasse 3, 8092 Zurich, Switzerland 4:12PM D17.00010 Optimal transport of diffusive scalar from the boundary , PIYUSH GROVER, Mitsubishi Elec Res Lab, YUNFEI SONG, Mitsubishi Elec Res Lab and Lehigh Univ. — Motivated by the problem of microfluidic heat transfer, we consider the optimal control of advection-diffusion in Stokes flows in two dimensional bounded domains. Our aim is to identify the incompressible velocity fields which result in most efficient transport of a diffusive scalar from boundary. We discretize the PDE using a spectral formulation, and derive the optimality conditions for the resulting system of ODEs. We assume that the optimal velocity field can be constructed by a linear combination of the available finite set of basis velocity fields. We compare the results obtained under constraints of fixed energy and fixed enstrophy. We also compare the numerical results with some theoretical predictions and bounds, and discuss the role of chaotic mixing in this process. Sunday, November 23, 2014 2:15PM - 4:25PM — Session D18 Vortex Dynamics: Flow Induced Vibrations and Interactions 2004 - Melissa Green, Syracuse University 2:15PM D18.00001 Flow-induced oscillations of tandem tethered cylinders in a channel flow1 , GARY NAVE, TYLER MICHAEL, Virginia Tech, PAVLOS VLACHOS, Purdue University, MARK STREMLER, Virginia Tech — In single degree-of-freedom (DOF) flow-induced oscillation studies of tandem rigid cylinders, the system most often consists of a front fixed cylinder and a trailing cylinder that is constrained to move perpendicular to the flow. We have conducted experiments in a water channel to investigate the behavior of a single DOF system of cylinders in which the trailing cylinder is constrained to move in a circular arc about the leading cylinder. We will discuss the dynamic response of the trailing cylinder for Reynolds numbers ranging from 10,000 to 20,000 and for inter-cylinder spacings from 3D to 5D, where D is the diameter of the cylinders. The experiments show a multi-frequency response that cannot be classified as a simple harmonic oscillator, as is assumed in typical tandem cylinder models. We compare our results with existing work on transversely constrained cylinders to determine the effect of tethering the cylinders. 1 Work made possible by funding from the Virginia Commonwealth Research Commercialization Fund 2:28PM D18.00002 Effect of mass ratio on fluid induced motions of a circular cylinder with strips1 , ASHWIN VINOD, ARINDAM BANERJEE, Lehigh University — The objective of the current experimental work is to investigate the effects of mass ratio on Fluid Induced Motions, such as vortex induced vibration (VIV) and galloping, of elastically mounted circular cylinders attached with strips to their outer surface. Although the effect of mass ratio on VIV of a smooth circular cylinder is well documented in literature, however, their effects on circular cylinders with strips, capable of inciting galloping oscillations haven’t been investigated and could have potential applications in the domain of vibration based energy harvesters. In the current work, three different mass ratios were tested, out of which, one falls below the critical mass in vortex induced vibration of a circular cylinder. The strips used for the experiments included sandpaper strips of prescribed roughness and smooth strips with no roughness, both of which served as surface protrusion based mechanisms of altering the flow around the cylinder. Interesting variations were observed in the amplitude, frequency response and the power spectrum, depending on the mass ratio of the oscillating system tested. 1 The authors acknowledge support of the Office of Naval Research (Grant # ONR-000141210495 - Dr. Ron Joslin). 2:41PM D18.00003 Wake dynamics of streamwise oscillating cylinders with one and two degrees of freedom , STAVROULA BALABANI1 , NEIL CAGNEY2 , University College London — The effect of a second degree of freedom on the structural response and wake modes of a cylinder undergoing streamwise Vortex-Induced Vibrations (VIV) was studied using DPIV. The 2-DOF oscillating cylinder was found to exhibit similar amplitude response to a cylinder allowed to oscillate only in the streamwise direction, i.e. containing two response branches separated by a low amplitude region, as reported in the literature. The first branch was characterised by negligible transverse motion and the appearance of both alternate (A-II) and symmetric (S-I) vortex-shedding which competed in an unsteady manner. However, this mode competition did not appear to have a significant effect on either the streamwise or transverse motion. The additional DOF was found to simplify the overall dynamics of the system in the second response branch by reducing the number of states that the wake can exhibit: while the 1 -DOF oscillating cylinder was found to exhibit 3 different wake states and hysteresis in the second branch, the 2-DOF one was found to exhibit only one wake mode in the second branch (the SA mode) and the cylinder response was no longer hysteretic. Figure-of-eight motion trajectories were observed throughout the lock in range and the phase angle between the streamwise and transverse motion was found to decrease in a linear manner with reduced velocity, with a sudden jump when the wake changed from the A-II to the SA mode. 1 Department 2 Department of Mechanical Engineering of Earth Sciences 2:54PM D18.00004 The effects of the external cross-flow excitation forces on the vortexinduced-vibrations of an oscillating cylinder , ABOUZAR KABOUDIAN, RAVI CHAITHANYA MYSA, RAJEEV KUMAR JAIMAN, Natl Univ of Singapore — Vortex induced vibrations can significantly affect the effectiveness of structures in aerospace as well as offshore marine industries. The oscillatory nature of the forces resulting from the vortex shedding around bluff bodies can result in undesirable effects such as increased loading, stresses, deflections, vibrations and noise in the structures, and also reduced fatigue life of the structures. To date, most studies concentrate on either the free oscillations or the prescribed motion of the bluff bodies. However, the structures in operation are usually subject to the external oscillatory forces (e.g. due to the platform motions in offshore industries). In this work, we present the effects of the external cross-flow forces on the vortex-induced vibrations of an oscillating cylinder. The effects of the amplitude, as well as the frequency of the external force on the fluid-forces on the oscillating cylinder are carefully studied and presented. Moreover, we present the transition of the response to be dominated by the vortex-induced-vibrations to the range where it is mostly dictated by the external oscillatory forces. Furthermore, we will discuss how the external forces can affect the flow structures around a cylinder. 3:07PM D18.00005 Effect of forced vibration on vortex evolution behind side-by-side cylinders , YINGCHEN YANG, University of Texas at Brownsville, ALIS EKMEKCI, University of Toronto — Vortex patterns in the wake of two side-by-side circular cylinders in stationary state and under forced cross-flow vibration were compared through experimental study. The hydrogen bubble visualization technique was employed for flow visualization. The Reynolds number was fixed at Re = 250 for all the experiments. Two center-to-center pitch ratios were examined: P/D = 3 and 6. For the two cylinders under forced vibration, the vibration frequency was chosen to match with the vortex shedding frequency in stationary state, and the vibration amplitude (A) was fixed at A/D = 0.25. Under forced in-phase vibration, very strong in-phase vortex shedding behind the two cylinders was observed for both P/D = 3 and 6. But the vortices evolve differently in the wake at different P/D. Under forced anti-phase vibration, both in-phase and anti-phase vortex shedding were observed for the two values of P/D. The effect of in-phase and anti-phase vibration on vortex evolution was characterized through comparison with the stationary case. 3:20PM D18.00006 Symmetry breaking in vortex-induced vibration of a rotating cylinder , BANAFSHEH SEYED-AGHAZADEH, YAHYA MODARRES-SADEGHI, University of Massachusetts, Amherst — Vortex-induced vibration (VIV) of a flexiblymounted circular cylinder, free to oscillate in the crossflow direction with imposed rotation around its axis, is studied experimentally. In particular, the influence of asymmetry that is introduced into the system by the forced rotation of the cylinder is considered. The rotation rate, α, defined as the ratio of the surface velocity and free stream velocity, was varied from 0 to 2.6 in small steps. The amplitudes and frequencies of oscillations as well as the flow forces were measured in a Reynolds number range of Re=350-1000. The maximum amplitude of oscillation was found to be limited to values less than a diameter of the cylinder at high rotation rates. Also the lock-in range was found to become narrower at higher rotation rates and finally the oscillation ceased beyond α =2.4. Vortex shedding pattern was found to change from 2S and 2P shedding (two single and two pairs of vortices shed per cycle of oscillation) for a non-rotating cylinder to P shedding (one pair of vortices shed in a cycle of oscillations) for the rotating cylinder. Also, the phase difference between the flow forces and displacement of the cylinder in the crossflow direction was influenced as the rotation rate was increased. At high reduced velocities the phase difference decreased from 180 degree for a non-rotating cylinder to values close to 90 degree for a rotating cylinder. 3:33PM D18.00007 Vortex-induced vibrations under oblique shedding , REMI BOURGUET, IMFT/CNRS, GEORGE KARNIADAKIS, Brown University, MICHAEL TRIANTAFYLLOU, Massachusetts Institute of Technology — A slender flexible body with bluff crosssection placed at normal incidence within a current may be subjected to vortex-induced vibrations (VIV). In practical applications, the structures (e.g. marine risers, towing cables) are often inclined with respect to the direction of the oncoming flow, sometimes at large angles. The vibrations that may appear in such configurations are investigated in the present work on the basis of direct numerical simulation results. We find that a flexible cylinder inclined at 80 degrees exhibits regular large-amplitude vibrations and that the structural responses are excited under the lock-in condition, i.e. synchronization between body oscillation and vortex formation, which is the central mechanism of VIV. We show that the lock-in condition may involve parallel vortex shedding, where the vortex rows are aligned with the body axis, but also oblique vortex shedding patterns. The excited structural wavenumber and the spanwise wavenumber of the obliquely shed vortices coincide; therefore, the flexible structure and the wake are locked both temporally and spatially. In addition, we find that the VIV occurring under oblique shedding may reach very high frequencies compared to the vibrations observed under parallel shedding. 3:46PM D18.00008 Motion Response of 2 DOF Circular Cylinder in Bundle Arrangment , HENDIK HANS, VINH TAN NGUYEN, Institute of High Performance Computing, A*STAR — This study focuses on the motion response of a freely vibrating in streamwise and crossflow circular cylinder in the wake of two leading stationary circular cylinders. Studies on the effects of spatial positioning of the trailing circular cylinder to its amplitude and frequency response are conducted. In order to explain the effects of mass ratio and phase angle on the motion response of the structure, analytical model based on tandem cylinder arrangement are presented. For almost all reduced velocities, the results indicated larger crossflow amplitude of oscillation as the trailing cylinder is aligned to the centerline of one of the leading circular cylinder. Two dominant response frequencies are found on the trailing circular cylinder. Switching between the two response frequencies as the dominant response frequency is found to be strongly related to the natural frequency of the system. Additionally, the mass ratio played a significant role in determining the intermittent domination of the Vortex-Induced Vibrating (VIV) frequency of the structure. For low mass ratio, larger mass ratio is found to increase its amplitude of oscillation. 3:59PM D18.00009 Flow behaviour around tripped circular cylinders , ANTRIX JOSHI, ALIS EKMEKCI, University of Toronto — An experimental study is carried out to investigate if the effects of multiple spanwise tripwires fitted on a circular cylinder in subcritical flow can be explained and predicted using the knowledge accumulated on the influence of only one spanwise tripwire. For this purpose, this study compares the vortex shedding frequency behind circular cylinders fitted with one, two and four spanwise wires. In the two-wire and four-wire fitted cases, the separation between the wires on the cylinder surface is arbitrarily selected as 90◦ . Frequency measurements were done for varying tripwire locations. The Reynolds number was kept at the subcritical value of 10,000 (based upon cylinder diameter). The tripwires were approximately 6% the diameter of cylinder. Results showed that a correlation exists between the effects of single- and multi-wire tripping, and a set of rules can be devised to predict the vortex shedding frequencies around a circular cylinder with complex two-dimensional tripping configurations. 4:12PM D18.00010 Investigation of the Wake Interactions in Tandem Cylinder Arrangements , RAVI CHAITHANYA MYSA, ABOUZAR KABOUDIAN, RAJEEV KUMAR JAIMAN, Natl Univ of Singapore — Vortex-induced vibrations of a single cylinder in a cross-flow are compared with the wake-induced oscillations of the downstream cylinder of a tandem cylinder arrangement in a cross-flow. It is known that the synchronization of frequency of vortex shedding with the natural frequency of the structure leads to large amplitude motions. For larger reduced velocities beyond the lock-in region, the cylinder displacement is abruptly reduced due to the inertia dominated region where the frequency of vortex shedding is larger than the natural frequency of the structure. However, in the case of tandem cylinders, the large amplitudes of the downstream cylinder is found at the reduced velocities greater than that of lock-in region. In this work, we show that the wake from the upstream cylinder interacts with the downstream cylinder which influences the response of the coupled system. Extensive numerical experiments have been performed on a single cylinder and tandem cylinder arrangement in cross-flow. Here, the wake interactions in connection to the forces generated are systematically studied. The upstream cylinder is fixed and the downstream cylinder is free to oscillate in transverse direction. Sunday, November 23, 2014 2:15PM - 4:12PM Session D19 Convection and Buoyancy-Driven Flows: General I sitaet Ilmenau — 2006 - Max Koerner, Technische Univer- 2:15PM D19.00001 Direct numerical simulations of mixed convection in a turbulent channel flow1 , SAMIR SID, VINCENT TERRAPON, University of Liege, YVES DUBIEF, University of Vermont — Wall-bounded turbulence has been extensively studied by the scientific community during the last decades. Much effort has been devoted to identify the role that coherent structures and energy exchanges play in turbulent channel flows. However, in many engineering applications, wall-bounded flows are subjected to additional physical phenomena. For instance, applying a temperature differential to the channel walls leads to a modified turbulent state which results from a balance between buoyancy, inertia and viscosity effects. Although, forced and natural convection have been widely studied separately, the coupling between both and its consequences on turbulence features are still not fully understood. In the present work, direct numerical simulations of a buoyant turbulent channel flow are reported for different values of the Reynolds and Richardson numbers. The energy exchanges between potential and kinetic energy and their impact on coherent structures are investigated. Macroscopic quantities (e.g.: Nusselt number) and statistics are compared with those obtained in forced convection flows. Finally, the influence of the ratio between inertia and buoyancy effects (i.e. Richardson number) is discussed. 1 We acknowledge PRACE for awarding us access to SuperMUC at GSC@LRZ, Germany; JUQUEEN at GCS@Julich, Germany; HERMIT at GCS@HLRS, Germany; FERMI at CINECA, Italy. 2:28PM D19.00002 Experimental Investigation of Large-Scale Flow Structures in Turbulent Mixed Convection1 , MAX KOERNER, CHRISTIAN RESAGK, Institute of Thermodynamics and Fluid Mechanics, Tech Univ Ilmenau, ANDRE THESS, Institute of Engineering Thermodynamics, German Aerospace Center — We report on experimental investigations of the temporal and spatial behavior of large-scale flow structures (LSC) in turbulent mixed convection. Using a reduced scale model room with a passenger cabin based geometry allows a global view on the LSCs, which are mainly responsible for thermal comfort and air quality within rooms. Moreover, the usage of pressurized working gases like air or sulfur hexafluoride (SF6) enables experimental investigations within broad ranges of the Reynolds number Re and Rayleigh number Ra. Thus, it is also possible to achieve realistic values of the dimensionless numbers allowing direct conclusions to be drawn about the LSCs in rooms similar to passenger cabins. The LSCs are determined by measurements of the 2D velocity field using a 2D2C particle image velocimetry system. In order to characterize three-dimensionally evolved flow structures, the measurement plane can be moved throughout the depth of the model room. We found very complex LSCs ranging from two-dimensional to three-dimensional structures and from one-roll systems over simple two-roll ones to chaotic behavior of the flow. The formation the LSCs has a strong dependency on the relation between Re and Ra and they often show distinct coherent oscillations. 1 The authors gratefully acknowledge the DFG (grant no. TH497-32-1) for financial support. 2:41PM D19.00003 A Study of Mixed Convection in a Heated Channel , M.Z. HOSSAIN, JERZY M. FLORYAN, University of Western Ontario — Mixed convection in a channel subject to a spatially periodic heating along one of the walls has been studied. The pattern of the heating is characterized by the wave number α and its intensity is expressed in terms of the Rayleigh number Rap . The primary convection occurring in response to the applied heating has the form of counter-rotating rolls with the wave vector parallel to the wave vector of the heating. The resulting net heat flow between the walls increases proportionally to Rap but the growth saturates when Rap = 0(103 ). The most effective heating pattern corresponds to α ≈1 as this leads to the most intense transverse motion. The primary convection is subject to transition to secondary states with the onset conditions depending on α. Conditions leading to transition between different forms of secondary motions have been determined using the linear stability theory. Three patterns of secondary motion may occur at small Reynolds numbers Re, i.e., the longitudinal rolls, the transverse rolls and the oblique rolls, with the critical conditions varying significantly as a function of α. Increase of α leads to the elimination of the longitudinal rolls and, eventually, elimination of the oblique rolls with the transverse rolls assuming the dominant role. For large α the transition is driven by the Rayleigh-Bénard mechanism while for α =0(1) the spatial parametric resonance dominates. It is shown that the global flow characteristics are identical regardless of whether the heating is applied either at the lower or at the upper walls. 2:54PM D19.00004 Simulations of surfactant laden drops settling in sharp stratifications , DAVID MARTIN, FRANCOIS BLANCHETTE, University of California Merced — To model oil droplets in the oceans, we present simulations of surfactant-laden drops settling in stratifications. Our model uses a thin, axisymmetric interface, treated as a two dimensional front, and we track the local surfactant concentration. By altering the relative surface tension of the drop with the ambient, surfactants impact the flow around the drop as well as the settling speed of the drop. The ambient stratification also affects surface tension, giving rise to complex dynamics. Settling speeds are obtained in the presence of surfactant, and compared to the surfactant free case, and the effects of the stratification are quantified. 3:07PM D19.00005 Effects of Droplet Size on Intrusion of Sub-Surface Oil Spills1 , ERIC ADAMS, GODINE CHAN, DAYANG WANG, Massachusetts Institute of Technology — We explore effects of droplet size on droplet intrusion and transport in sub-surface oil spills. Negatively buoyant glass beads released continuously to a stratified ambient simulate oil droplets in a rising multiphase plume, and distributions of settled beads are used to infer signatures of surfacing oil. Initial tests used quiescent conditions, while ongoing tests simulate currents by towing the source and a bottom sled. Without current, deposited beads have a Gaussian distribution, with variance increasing with decreasing particle size. Distributions agree with a model assuming first order particle loss from an intrusion layer of constant thickness, and empirically determined flow rate. With current, deposited beads display a parabolic distribution similar to that expected from a source in uniform flow; we are currently comparing observed distributions with similar analytical models. Because chemical dispersants have been used to reduce oil droplet size, our study provides one measure of their effectiveness. Results are applied to conditions from the ‘Deep Spill’ field experiment, and the recent Deepwater Horizon oil spill, and are being used to provide “inner boundary conditions” for subsequent far field modeling of these events. 1 This research was made possible by grants from Chevron Energy Technology Co., through the Chevron-MITEI University Partnership Program, and BP/The Gulf of Mexico Research Initiative, GISR. 3:20PM D19.00006 Experiments and models of particle slurries , MATTHEW MOLINARE, University of California, Los Angeles, SARAH BURNETT, University of North Carolina, Chapel Hill, ANDREW LI, University of California, Los Angeles, KATHERINE VARELA, California State University, Long Beach, DIRK PESCHKA, Weierstrass Institute, Mohrenstr, Berlin, JEFFREY WONG, University of California, Los Angeles, ANDREA BERTOZZI, Department of Math, University of California, Los Angeles — We present new experimental and theoretical results for the resuspension of bidisperse particle-laden flows on an inclined plane. In particular, we study the case of two negatively buoyant particle species of similar size and dissimilar densities in a viscous fluid of finite volume. Different regimes of particle separation are observed and studied by adjusting the angle of inclination, total particle concentration, and relative particle volume ratio. In addition to obtaining information about the height profile of shock formations, we measure the advancement and separation of particle and fluid front positions in mono- and bidisperse scenarios. These dynamics are the basis for a quantitative understanding of polydisperse cases, which can be readily applied to industry and catastrophe modeling. 3:33PM D19.00007 Numerical simulation of convective sedimentation using a two-way coupled Eulerian-Lagrangian model , YI-JU CHOU, YUN-CHUAN SHAO, National Taiwan University, YI-CHUN YEH, National Taiwan Normal University — Numerical simulations of convective sedimentation are conducted using an Eulerian-Lagrangian (EL) model. The present EL model is a two-way coupled system that further accounts for added mass and volumetric effects, which makes it more suitable for solid-liquid suspension problems. By comparing with different modeling strategies, the present modeling results reveal the significant deviation from the traditional single-phase modeling results when concentration becomes dense. Moreover, comparison with the point particle representation demonstrates the importance of the volumetric effect in dense suspension problems. Settling of particle clouds in homogeneous and stratified background flow fields is investigated. The results show different transport patterns associated with different conditions of density and particle-induced stratification. The influence of the background stratification on convection of particle clouds is further discussed. 3:46PM D19.00008 Turbulent flow inside a solar concentrator receiver , MANUEL RAMIREZ, EDUARDO RAMOS, Universidad Nacional Autonoma de Mexico — A solar concentrator receiver is a heat exchanger designed to absorb a beam of radiant heat coming from a field of heliostats. Inside the device, a slow forced flow generated bye an external pressure gradient is present, together with a natural convective a turbulent flow produced by the large temperature gradients due to intense heating. We present a model of this device based on the numerical solution of the mass, momentum and energy conservation equations. We consider heating conditions that lead to turbulence convective flow. For this season, a large eddy simulation model is incorporated. The results are potentially useful for the design of solar concentrator receivers. 3:59PM D19.00009 Numerical investigation of high temperature heat pipe incorporated in thermal energy storage systems , MAHBOOBE MAHDAVI, SONGGANG QIU, SAEED TIARI, Sustainable Energy and Energy Efficiency Laboratory, Department of Mechanical Engineering, Temple University, Philadelphia, PA — In the present work, a new type of high temperature heat pipe is investigated which can be incorporated in the thermal energy storage for concentrated solar power systems. A detailed two dimensional axisymmetric numerical procedure is developed to analyze the steady state thermal-hydrodynamic characteristics of the heat pipe. The model accounts for conduction in the wall and wick regions as well as compressible flow in vapor chambers. The geometrical features, working fluid type, wick structure and operational temperature of the heat pipe are adjusted satisfying the heat transport limitations. The proposed numerical model agrees well with available experimental data. The effects of evaporator heat input and vapor core radius on the vapor velocity and pressure fields, vapor and wall temperature distributions as well as heat pipe thermal resistance are studied. The results revealed that the thermal resistance increases with the increase of heat input while decreases with the increase of radius, however, there exists a certain radius which further increase over would not affect the thermal resistance significantly. Sunday, November 23, 2014 2:15PM - 3:46PM Session D20 Acoustics II: General — 2008 - R. Sujith, Indian Institute of Technology, Madras 2:15PM D20.00001 Schlieren imaging of shock waves radiated by a trumpet1 , PABLO L. RENDON, ROBERTO VELASCO-SEGURA, Centro de Ciencias Aplicadas y Desarrollo Tecnologico, Universidad Nacional Autonoma de Mexico, CARLOS ECHEVERRIA, DAVID PORTA, Facultad de Ciencias, Universidad Nacional Autonoma de Mexico, TEO VAZQUEZ, ANTONIO PEREZ-LOPEZ, Centro de Ciencias Aplicadas y Desarrollo Tecnologico, Universidad Nacional Autonoma de Mexico, CATALINA STERN, Facultad de Ciencias, Universidad Nacional Autonoma de Mexico — The flaring bell section of modern trumpets is known to be critical in determining a wide variety of properties associated with the sound radiated by these instruments. We are particularly interested in the shape of the radiated wavefront, which clearly depends on the bell profile. A horn loudspeaker is used to drive high-intensity sound at different frequencies through a B-flat concert trumpet. The sound intensity is high enough to produce shock waves inside the instrument resonator, and the radiated shocks are then visualised using Schlieren imaging. Through these images we are able to study the geometry of the shock waves radiated by the instrument bell, and also to calculate their propagation speed. The results show that propagation outside the bell is very nearly spherical, and that, as expected, the frequency of the driving signal affects the point at which the shock waves separate from the instrument. 1 We acknowledge financial support from PAPIIT IN109214 and PAPIIT IN117712. 2:28PM D20.00002 Quantum Analogies in the Interaction between Acoustic Waves and Bubble Clouds1 , MIGUEL A. PARRALES, Universidad de Sevilla, JAVIER RODRIGUEZ-RODRIGUEZ, Universidad Carlos III de Madrid — Analogies between quantum mechanical and acoustical propagation phenomena have a great interest in academic research due to their ability to shed light on some complex quantum effects, which are impossible to visualize directly in the macroscopic world. In this talk, we describe a number of these analogies concerning the acoustic behavior of bubble clouds. Firstly, we show that the structure of the collective oscillation modes of a spherical bubble cloud resembles that of the atomic orbitals of a hydrogen atom. Secondly, we present an analogy between some perturbation methods used in quantum-electrodynamics and the computation of the acoustic response of the randomly distributed bubble cloud by considering the contribution to the total scattered pressure of the multiple scattering paths that take place inside the clouds. As an application of this analogy, we obtain the scattering cross-section of a diluted cloud, which remarkably mimics the quantum scattering of an neutron wave when passing through an atomic nucleus. Finally, we numerically reproduce the behavior of an electron in a covalent bond between two hydrogen atoms by simulating the acoustic wave propagation through two neighboring spherical bubble assemblages. 1 Funded by the Spanish Ministry of Economy and Competitiveness through grants DPI2011-28356-C03-01 and DPI2011-28356-C03-02. 2:41PM D20.00003 A low-order model for wave propagation in random waveguides , CHRISTOPHE MILLET, MICHAEL BERTIN, DANIEL BOUCHE, CEA, DAM, DIF — In numerical modeling of infrasound propagation in the atmosphere, the wind and temperature profiles are usually obtained as a result of matching atmospheric models to empirical data and thus inevitably involve some random errors. In the present approach, the sound speed profiles are considered as random functions and the wave equation is solved using a reduced-order model, starting from the classical normal mode technique. We focus on the asymptotic behavior of the transmitted waves in the weakly heterogeneous regime (the coupling between the wave and the medium is weak), with a fixed number of propagating modes that can be obtained by rearranging the eigenvalues by decreasing Sobol indices. The most important feature of the stochastic approach lies in the fact that the model order can be computed to satisfy a given statistical accuracy whatever the frequency. The statistics of a transmitted broadband pulse are computed by decomposing the original pulse into a sum of modal pulses that can be described by a front pulse stabilization theory. The method is illustrated on two large-scale infrasound calibration experiments, that were conducted at the Sayarim Military Range, Israel, in 2009 and 2011. 2:54PM D20.00004 On Pressure Wave Simulations in Liquid Metal Neutron Source Targets , JANA R. FETZER, ANDREAS CLASS, Karlsruhe Institute of Technology (KIT) — Sound waves generated by fluid flow at low Mach numbers is associated with separated scales and thus with difficulties to construct efficient numerical methods for their approximation. One method is the Multi Pressure Variables (MPV) approach introduced for aero-acoustic applications [1,2]. The MPV approach is based on a single time scale multiple space scale asymptotic analysis derived for subsonic flow by an asymptotic series expansion in the Mach-number. Distinguished are the flow and acoustic length scales resulting in three pressure contribution, i.e. thermodynamic, acoustic and dynamic pressure which are discretized on numerical meshes of different resolution. We propose to apply MPV to analyse liquid metal cooled spallation targets with a pulsed proton beams. These targets are operating in high power neutron sources for fundamental research. The nearly instantaneous heating of the liquid metal results in volumetric expansion of inertia confined liquid and thus to high pressure waves, which represent a major lifetime limiting thread. Our development accompanies design activities for the META:LIC (MEgawatt TArget: Lead bIsmuth Cooled) target proposed for the European Spallation Source. [1] Munz C.-D. et al. J Computers & Fluids 2003; 32 [2] Klein R. J. Comput Phys 1995; 121 3:07PM D20.00005 Numerical analysis of sound propagation for acoustic lens array in different fluid mediums , KEI FUJISAWA, AKIRA ASADA, University of Tokyo — In this paper, an acoustic sound focusing method using acoustic lens array is investigated numerically. To understand the sound propagation in the acoustic field in water with a lens material of glycerin, compressible Navier-Stokes equation, the mass conservation, energy equation, state equation in cylindrical coordinate system are solved without applying parabolic approximation. The numerical method is based on the finite difference time domain method. The numerical calculation of the sound propagation is carried out in the near field of the acoustic lens array of variable thickness normal to the acoustic beam. The numerical result shows that the sound pressure level along the beam axis increases due to the influence of the acoustic lens array, which indicates the capability of the acoustic lens array to the sound focusing. 3:20PM D20.00006 Global mode and frequency response analysis of low-density jets1 , W. COENEN, Dept. Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, 28911 Leganés, Spain, L. LESSHAFFT, X. GARNAUD, LadHyX, CNRS - École Polytechnique, 91128 Palaiseau, France, A. SEVILLA, Dept. Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, 28911 Leganés, Spain — We present a global stability analysis of a low-density jet, where the wavepacket structures are temporal eigenmodes of the linearized equations of motion in a 2D domain. As a base state we employ a numerical solution of the low-Mach number Navier-Stokes equations. The jet is characterized through the jet-to-ambient density ratio, the Reynolds number, and the momentum thickness of the velocity profile at the jet exit plane. The linear global mode analysis shows that for certain combinations of the control parameters, an isolated eigenmode dominates the eigenvalue spectrum. Its associated growth rate can be used to construct a neutral curve in the parameter space that agrees well with the experimentally observed onset of self-sustained oscillations (Hallberg & Strykowski, JFM, 2006). However, for high values of the Reynolds number, the construction of a neutral curve based on the spectrum loses validity, since for these cases the spectrum is dominated by a continuous branch of eigenvalues, sensitive to changes in domain length and grid refinement. Finally, the flow response to external forcing in a globally stable setting is investigated through the computation of the pseudospectrum, and is found to be dominated by a resonance of the stable eigenmode. 1 Supported by Spanish MINECO under project DPI 2011-28356-C03-02. 3:33PM D20.00007 Understanding the onset of whistling: modelling and analysis , RAMAN SUJITH, VINEETH NAIR, IIT Madras — The onset of whistling in a turbulent pipe flow across an orifice is investigated. An increase in the Reynolds number (Re) shifts the dynamics from a turbulent state to a state dominated by periodic dynamics. This emergence of ordered oscillations is presaged by an intermittent regime with bursts of periodic oscillations appearing in a near-random fashion from the low-amplitude background. The signal at the onset of whistling corresponds to period-2 oscillations as evidenced by an analysis of the amplitude spectrum of the signal and phase space reconstruction. These intermittent bursts have a frequency distinct from the final oscillatory state. A simple acoustic analysis shows that this frequency shift is due to a change in the boundary condition at the orifice exit. The repeating patterns in the dynamics of the intermittent signal are explored through recurrence quantification and the patterns reveal the intermittency to be of type-III; i.e., the type of intermittency that normally precedes a subcritical reverse period-doubling bifurcation. We confirm this subcritical nature of the transition by demonstrating hysteresis, and hence a range of Re for which the system is bi-stable, by decreasing Re after the onset of whistling. Sunday, November 23, 2014 2:15PM - 4:25PM Session D21 Instability: Boundary Layers II — 2010 - Stephen Garrett, University of Leicester 2:15PM D21.00001 The centrifugal instability of the boundary-layer flow over slender rotating cones1 , STEPHEN GARRETT, ZAHIR HUSSAIN, University of Leicester — Existing experimental and theoretical studies are discussed which lead to the clear hypothesis of a hitherto unidentified convective instability mode that dominates within the boundary-layer flow over slender rotating cones. The mode manifests as Görtler-type counter-rotating spiral vortices, indicative of a centrifugal mechanism. Although a formulation consistent with the classic rotating-disk problem has been successful in predicting the stability characteristics over broad cones, it is unable to identify such a centrifugal mode as the half-angle is reduced. An alternative formulation is developed and the governing equations solved using both short-wavelength asymptotic and numerical approaches to independently identify the centrifugal mode. 1 Supported by EPSRC grant EP/G061637/1 2:28PM D21.00002 Rimming flows and pattern formation inside rapidly rotating cylinder1 , DENIS POLEZHAEV, VERONIKA DYAKOVA, VICTOR KOZLOV, Perm State Humanitarian Pedagogical University — The dynamics of fluid and granular medium in a rotating horizontal cylinder is experimentally studied. In a rapidly rotating cylinder liquid and granular medium coat the cylindrical wall under centrifugal force. In the cavity frame gravity field performs rotation and produces oscillatory fluid flow which is responsible for the series of novel effects of pattern formation, namely, axial segregation of heavy particles and pattern formation in the form of sand regular hills extended along the axis of rotation. At least two types of axial segregation are found: a) patterns of spatial period of the same order of magnitude as fluid layer thickness which induced by steady flows generated by inertial waves; b) fine patterns which manifests Gortler - Taylor vortices developing as a consequence of centrifugal instability of viscous boundary layer near the cylindrical wall. Under gravity, intensive fluid shear flow induces partial fluidization of annular layer of granular medium. The oscillatory motion is followed by onset of regular ripples extended along the axis of rotation. 1 The work is supported by Russian Scientific Foundation (project 14-11-00476) 2:41PM D21.00003 Role of Spatio-Temporal Wave-front in causing Flow Transition , SWAGATA BHAUMIK, The Ohio State University, TAPAN SENGUPTA, I. I. T. Kanpur, India — Theoretically, boundary layer transition has been identified to occur via K- and H- or N-type routes (Y. S. Kachanov, Ann. Rev. Fluid Mech., 26, 1994) depending upon the arrangement of Λ-vortices in the transitional zone. While the aligned pattern of these vortices are identified with the first type, a staggered arrangement is attributed to the latter. Subsequently, the H-type breakdown is explained due to triad resonant interaction between a monochromatic spatial 2D TS wave and its two oblique 3D sub-harmonic counterparts. We show via high accuracy DNS of receptivity of the zero-pressure gradient boundary layer that both K- and H-types of transition can be noted for monochromatic deterministic wall-excitation due to growth of spatio-temporal wave-front, which in Bhaumik & Sengupta (Phys. Rev. E, 89, 043018, 2014) has been established as the precursor of flow transition. In addition to the high accuracy dispersion-relation preserving numerical schemes, computations are also carried out over a significantly longer computational domain. The H-type transition is noted for lower frequency of excitation cases, while K-type is seen to occur for higher frequency cases which is in contrast to current theoretical view-point, particularly for H-type transition. 2:54PM D21.00004 Flat-plate boundary-layer receptivity to free-stream vortical disturbances with roughness , RICHARD BOSWORTH, JONATHAN MORRISON, Imperial College — This study focuses on the experimental investigation of the roughness-induced generation of Tollmien-Schlichting (TS) waves in a flat-plate boundary layer, exposed to free-stream vortical disturbances. Experiments are taken in the department’s low-speed, low-turbulence wind tunnel where streamwise and lateral free-stream turbulence intensities are below 0.07%. Repeatable, harmonic, 2D free-stream disturbances are created using a metal ribbon placed upstream of a metal plate, with a leading edge designed specifically for receptivity experiments. The ribbon is forced to vibrate at a frequency conducive to the generation of TS waves within the boundary layer. It is shown that, without roughness present on the plate, the vortical disturbances decay into the boundary layer and that TS waves are not generated. The addition of roughness strips, with heights on the order of the inner deck scaling from Triple Deck Theory, clearly initiate a boundary layer response with the characteristic TS wave profile. This further confirms the theoretical predictions that a scale conversion process is required to generate TS waves from free-stream disturbances in a flat-plate boundary layer. 3:07PM D21.00005 Numerical Study of Boundary Layer Receptivity to Periodic Vortical Disturbances in Freestream , YU NISHIO, SEIICHIRO IZAWA, YU FUKUNISHI, Tohoku Univ — A flow passing a flat plate with an elliptic leading edge whose aspect ratio is 5 is simulated to investigate its receptivity to periodic vortical disturbances in the freestream. The disturbance consists of vortex pairs aligned in the spanwise direction, whose rotational directions are opposite. When they successively collide to the leading edge of the flat plate, streamwise vortices appear in the boundary layer downstream. The tilting and stretching of oncoming vortex columns with their axes normal to the flat plate was believed to be the cause of the generation of the streamwise vortices. However in this study, it is shown that the streamwise vortices in the boundary layer is generated by a combination of the spanwise velocity induced by the parts of vortices outside the boundary layer and the no-slip condition at the wall. 3:20PM D21.00006 Experimental study of crossflow instability on a Mach 6 yawed cone , STUART CRAIG, WILLIAM SARIC, Texas A&M University — Boundary-layer stability and transition represents a key challenge for the designer of hypersonic vehicles, which typically feature highly-swept and conical features inclined to the free stream. The transition process on each of these geometries is typically dominated by the three-dimensional crossflow instability. In order to advance the goal of a physics-based transition prediction method, crossflow experiments were undertaken in the Mach 6 Quiet Tunnel at Texas A&M University. Detailed boundary-layer measurements were performed on a 7-degree cone at a 6-degree angle of incidence using constant-temperature hot-wire anemometry (CTA) to produce boundary-layer contours at constant axial location. These contours illustrate the characteristic streamwise vortex pattern and mean-flow distortion characteristic of crossflow-dominated flows. Additionally, the high frequency response of the CTA system allows for analysis of the spectral content of the flow. These measurements show a high degree of qualitative agreement with analogous studies performed in low-speed flows. 3:33PM D21.00007 Linear stability of steady flow in a precessing sphere – Global and local disturbances , SHIGEO KIDA, Doshisha University — It is known by the linear stability of the steady flow in a precessing sphere that the critical curve behaves as Po=7.9/Re**0.5 for global disturbances at large Re, where Po is the Poincare number and Re is the Reynolds number (Kida 2013). Here we perform the linear stability analysis of the local disturbances localized in the critical regions and the conical shear layer emerging wherefrom, and show that the asymptotic form of the critical curve is Po=21.25/Re**0.8 and that this result agrees with the corresponding labolabory experiment. On the other hand, the critical curve for the global disturbances is not observed in the experiment of a precessing sphere but in the experiment of a slightly elongated spheroid, the minor-to-major axis ratio being 0.9 (Goto et al. 2014). These correspondences between theory and experiment can be understood by noting that conical shear layers exist stably only for a spherical cavity. 3:46PM D21.00008 Numerical study of absolute instability on a rotating disk flow , KEUNSEOB LEE, YU NISHIO, SEIICHIRO IZAWA, YU FUKUNISHI, Tohoku Univ — Numerical simulation is carried out aimed at investigating the absolute instability of the three-dimensional boundary layer flow on a rotating disk. An artificial random disturbance is given at the wall to a laminar flow whose Reynolds number is higher than the critical value for the absolute instability. The disturbance first grows up into spiral vortices aligned regularly in the circumferential direction, and after that, an onset of turbulence takes place. The process is similar to what takes place at the convectively unstable region of much lower Reynolds number. The region of spiral vortices expands not only outward but also toward the center of the disk, and so does the turbulent region. 3:59PM D21.00009 Effect of Wall Suction on Cross-Flow Absolute Instability of a Rotating Disk Boundary Layer , JOANNA HO, THOMAS CORKE, ERIC MATLIS, University of Notre Dame — The effect of uniform suction on the absolute instability of Type I cross-flow modes on a rotating disk is examined. Specifically it investigates if wall suction transforms the absolute instability into a global mode as postulated in the numerical simulations of Davies and Carpenter (2003). The experiment is designed so that a suction parameter of a = W̄0 /(νω)1/2 = 0.2 locates the absolute instability critical Reynolds number, Rca = 650, on the disk. Uniform wall suction is applied from R = 317 to 696. The design for wall suction follows that of Gregory and Walker (1950), where an array of holes through the disk communicate between the measurement side of the disk and the underside of the disk in an enclosure that is maintained at a slight vacuum. The measurement surface is covered by a 20 micron pore size Polyethylene sheet. Temporal disturbances are introduced using the method of Othman and Corke (2006), and the evolution of the resulting wave packets are documented. The present results indicate a rapid transition to turbulence near Rca . 4:12PM D21.00010 Global stability and frequency response of boundary layers developing over shallow cavities , UBAID QADRI, PETER SCHMID, Imperial College, London — In the presence of surface imperfections, the boundary layer developing over an aircraft wing can separate and reattach, leading to a small separation bubble. We study the flow over a shallow rectangular cavity at Reynolds numbers at which the boundary layer is unstable to Tollmien-Schlichting waves. We obtain steady two-dimensional solutions to the incompressible Navier-Stokes equations and study the growth of three-dimensional perturbations on top of these steady base flows. We use the linearized Navier-Stokes operator to identify how the dominant modes of instability vary with the thickness of the upstream boundary layer and with the cavity aspect ratio. We calculate the global frequency response and optimal forcing to map out the influence of the cavity on the growth of TS-waves. Finally, we compare the results with those for boundary layers developing over backward-facing and forward-facing steps. Sunday, November 23, 2014 2:15PM - 4:25PM Session D22 Instability: Rayleigh-Taylor II — 2012 - Jeff Jacobs, University of Arizona 2:15PM D22.00001 2D Rayleigh-Taylor instability: Interfacial arc-length vs. deformation amplitude1 , MARIE-CHARLOTTE RENOULT, University of Le Havre, PIERRE CARLES, Universite Pierre et Marie Curie, SAMEH FERJANI, CHARLES ROSENBLATT, Case Western Reserve Univ — Fluid interface instabilities are usually studied through the time evolution of the amplitude of deformation of the interface. While this approach is convenient, it often fails to fully describe the evolution of a deforming interface, especially when the interface cannot be represented as a single-valued function of a space coordinate. Here, we present experimental data on the Rayleigh-Taylor 2D instability for immiscible fluids having a single-mode sinusoidal initial perturbation, which is obtained through the use of magnetic levitation. We observe that new information can be retrieved by using an alternate metric to the amplitude, viz., the total arc-length of the interface (in 2D), or equivalently its total surface area (in 3D). In particular, we identify a master curve for the evolution of the arc-length over time, following three different regimes and on which all our data points fall. We conjecture that the exploration of such alternate metrics will yield interesting results on a broad range of interface instabilities. 1 Acknowledgments: International Relations UPMC, Partner University Fund, and Fulbright Foundation 2:28PM D22.00002 Interface node behavior due to nonlinearities in a 2D Rayleigh-Taylor instability1 , MARIE-CHARLOTTE RENOULT, University of Le Havre, CHARLES ROSENBLATT, Case Western Reserve Univ, PIERRE CARLES, Universite Pierre et Marie Curie — We report a quantitative study on the symmetry effect of nonlinearities in a typical Rayleigh-Taylor (RT) instability for a single-mode sinusoidal initial perturbation. We use the interface zero-crossings (nodes) to monitor the asymmetrical deformation of the interface due to the growth of nonlinear odd harmonics. A weakly nonlinear model is developed and compared to measurements of node positions in fourteen RT experiments performed using the magnetic levitation technique. Our results suggest that monitoring the nodes’ spatial displacement over time is a powerful technique for detecting the first nonlinear harmonic, and more broadly, exploring the transitional regime between linearity and fully-developed nonlinearity. The nodes approach provides a metric complementary to the deformation amplitude, which is widely used to measure the amplitude effect of nonlinearities in most interface instabilities. 1 Acknowledgments: International Relations UPMC, Partner University Fund, Fulbright Foundation 2:41PM D22.00003 Generalized Cahn-Hilliard Navier-Stokes equations for numerical simulations of multicomponent immiscible flows , ZHAORUI LI, DANIEL LIVESCU, Los Alamos National Laboratory — By using the second-law of thermodynamics and the Onsager reciprocal method for irreversible processes, we have developed a set of physically consistent multicomponent compressible generalized Cahn-Hilliard Navier-Stokes (CGCHNS) equations from basic thermodynamics. The new equations can describe not only flows with pure miscible and pure immiscible materials but also complex flows in which mass diffusion and surface tension or Korteweg stresses effects may coexist. Furthermore, for the first time, the incompressible generalized Cahn-Hilliard Navier-Stokes (IGCHNS) equations are rigorously derived from the incompressible limit of the CGCHNS equations (as the infinite sound speed limit) and applied to the immiscible Rayleigh-Taylor instability problem. Extensive good agreements between numerical results and the linear stability theory (LST) predictions for the Rayleigh-Taylor instability are achieved for a wide range of wavenumber, surface tension, and viscosity values. The late-time results indicate that the IGCHNS equations can naturally capture complex interface topological changes including merging and breaking-up and are free of singularity problems. 2:54PM D22.00004 Combined Rayleigh-Taylor and Kelvin-Helmholtz instabilities on cylindrical interfaces , VADIVUKKARASAN M, MAHESH V PANCHAGNULA, Indian Institute of Technology Madras — Hydrodynamic instabilities that occur on a fluid interface are of interest to a wide range of applications. We study the combined effect of Rayleigh-Taylor (R-T) and Kelvin-Helmholtz (K-H) mechanisms of instability simultaneously attempting to destabilize a cylindrical interface. Linear stability analysis is used to study the process by which relative velocity (characterized by a Weber number) and acceleration (characterized by a Bond number) induced effects distort the interface. We investigate the effect of three dimensional disturbances and study the effect of varying Bo and We. From the dispersion relation obtained in this study, we are able to recover the R-T and K-H mechanism dispersion relations as special cases. From this study, we observe the occurrence of two-dimensional Taylor and flute modes as well as three-dimensional helical modes. A regime chart is presented in the (Bo,We) space to demonstrate the energy budget in the acceleration and shear induced instability mechanisms. In addition, we show that the length scale associated with the distorted interface is minimum in the helical mode. Finally, we show that an optimal Weber number exists above which it is not beneficial to increase relative velocity based kinetic energy. 3:07PM D22.00005 Hydrodynamic Instabilities in Blast-Driven Systems , MARC HENRY DE FRAHAN, ERIC JOHNSEN, Univ of Michigan - Ann Arbor — Mixing from hydrodynamics instabilities such as Richtmyer-Meshkov, Rayleigh-Taylor, and Kelvin-Helmholtz, occurs in a wide range of engineering applications such as inertial confinement fusion, supernova collapse, and scramjet combustion. The success of these applications depends on an accurate understanding of these phenomena. Following previous work investigating hydrodynamic mixing from the interaction of a perturbed interface with a planar blast wave, we model the perturbation growth by analyzing the different acceleration phases of a blast wave: an instantaneous acceleration (a pressure increase) followed by a gradual, time-dependent deceleration (a pressure decrease). Depending on the characteristics of these phases, the instability will be dominated by Richtmyer-Meshkov or Rayleigh-Taylor growth. We use a high-order accurate Discontinuous Galerkin method that prevents pressure errors at interfaces with variable specific heats ratios to simulate these systems and understand the different growth regimes. 3:20PM D22.00006 On the treatment of material interfaces in the presence of finite mass physical diffusion1 , POOYA MOVAHED, ERIC JOHNSEN, University of Michigan, Ann Arbor — In incompressible miscible variable-density flows, density is a function of composition and temperature (but not pressure), and velocity does not remain divergence-free in mixing regions. In numerical simulations of diffuse interfaces, it was previously shown that a specific form of the velocity, based on the density profile, should be prescribed initially, for consistency. In this work, we are interested in extending these ideas to compressible miscible flows, where the density and pressure are coupled through an equation of state. We study the temporal evolution of an isolated material interface in the presence of diffusion processes (mass, momentum and energy). We show that a velocity profile similar to that introduced in the incompressible case should be prescribed initially to avoid generating spurious waves at the interface. A new form of the initial velocity profile is suggested for an isothermal problem in the presence of gravity. The single-mode Richtmyer-Meshkov instability is used to illustrate the importance of this prescribed velocity on large-scale flow dynamics after re-shock. 1 This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1053575. 3:33PM D22.00007 Immiscible experiments on the Rayleigh-Taylor instability using simultaneous particle image velocimetry and planar laser induced fluorescence concentration measurements , MATTHEW MOKLER, JEFFREY JACOBS, University of Arizona — Incompressible Rayleigh-Taylor instability experiments are presented in which two stratified liquids having Atwood number of 0.2 are accelerated in a vertical linear induction motor driven drop tower. A test sled having only vertical freedom of motion contains the experiment tank and visualization equipment. The sled is positioned at the top of the tower within the linear induction motors and accelerated downward causing the initially stable interface to be unstable and allowing the Rayleigh-Taylor instability to develop. Forced and unforced experiments are conducted using an immiscible liquid combination. Forced initial perturbations are produced by vertically oscillating the test sled prior to the start of acceleration. The interface is visualized using a 445nm laser light source that illuminates a fluorescent dye mixed in one of the fluids and aluminum oxide particles dispersed in both fluids. The laser beam is synchronously swept across the fluorescent fluid, at the frame rate of the camera, exposing a single plane of the interface. The resulting images are recorded using a monochromatic high speed video camera. Time dependent velocity and density fields are obtained from the recorded images allowing for 2D full field measurements of turbulent kinetic energy and turbulent mass transport. 3:46PM D22.00008 Numerical study of the single-mode Rayleigh-Taylor instability with nonunity Schmidt number1 , MAXWELL HUTCHINSON, ROBERT ROSNER, Univ of Chicago — Recent experiments[1] and simulations[2,3] of the single mode Rayleigh-Taylor instability question the assumed existence of a bubble terminal velocity regime[4], particularly for low Atwood numbers. We present numerical results using the spectral element method and Boussinesq approximation with purely physical viscosity and diffusivity. The Schmidt number is chosen away from unity and boundary conditions are no-slip in an effort to bring the simulations closer to physically realizable conditions. [1] [2] [3] [4] J. P. Wilkinson and J. W. Jacobs, Phys. Fluids 19, 124102 (2007). P. Ramaprabhu et al., Phys. Fluids 24, 074107 (2012). T. Wei and D. Livescu, Phys. Rev. E 86, 046405 (2012). R. M. Davies and G. Taylor, Proc. R. Soc. A Math. Phys. Eng. Sci. 200, 375 (1950). 1 M. Hutchinson acknolwedges the support of the Department of Energy Computational Science Graduate Fellowship 3:59PM D22.00009 Buoyancy Driven Mixing with Continuous Volumetric Energy Deposition , ADAM J. WACHTOR, FARZANEH F. JEBRAIL, NICHOLAS A. DENNISEN, MALCOLM J. ANDREWS, ROBERT A. GORE, Los Alamos National Laboratory — An experiment involving a miscible fluid pair is presented which transitioned from a Rayleigh-Taylor (RT) stable to RT unstable configuration through continuous volumetric energy deposition (VED) by microwave radiation. Initially a light, low microwave absorbing fluid rested above a heavier, more absorbing fluid. The alignment of the density gradient with gravity made the system stable, and the Atwood number (At) for the initial setup was approximately -0.12. Exposing the fluid pair to microwave radiation preferentially heated the bottom fluid, and caused its density to drop due to thermal expansion. As heating of the bottom fluid continued, the At varied from negative to positive, and after the system passed through the neutral stability point, At = 0, buoyancy driven mixing ensued. Continuous VED caused the At to continue increasing and further drive the mixing process. Successful VED mixing required careful design of the fluid pair used in the experiment. Therefore, fluid selection is discussed, along with challenges and limitations of data collection using the experimental microwave facility. Experimental and model predictions of the neutral stability point, and onset of buoyancy driven mixing, are compared, and differences with classical, constant At RT driven turbulence are discussed. 4:12PM D22.00010 Experiments on the rarefaction wave driven Rayleigh-Taylor instability initiated with a random initial perturbation , ROBERT MORGAN, JEFFREY JACOBS, The University of Arizona — Experiments are presented in which a diffuse interface between two gases is accelerated to become Rayleigh-Taylor unstable. The initially flat interface is generated by the opposing flow of two test gases at matched volumetric flow rates exiting through small holes in the test section. A random, three-dimensional interface perturbation is forced using a loudspeaker. The interface is then accelerated by an expansion wave which is generated by the rupturing of a diaphragm separating the heavy gas from a vacuum tank evacuated to ∼ 0.01atm. The expansion wave generates a large (of order 1000 g), non-constant acceleration acting on the interface causing the Rayleigh-Taylor instability to develop. Planar Mie scattering is employed to visualize the flow using a planar laser sheet generated at the top of the apparatus, which illuminates smoke particles seeded in the heavy gas. The scattered light is then recorded using a CMOS camera operating at 12kHz. The mixing layer width is obtained from an ensemble of experiments and the turbulent growth parameter α is extracted and compared with previous experiments and simulations. Sunday, November 23, 2014 2:15PM - 4:25PM Session D23 Geophysical Fluid Dynamics: Stratified Turbulence II — 2001 - Robert Ecke, Los Alamos National Laboratory 2:15PM D23.00001 A Sweeping based Kinematic Simulation for the Stably Stratified Surface Layer , ADITYA GHATE, SANJIVA LELE, Stanford Univ — A Kinematic Simulation (KS) for a statistically stationary and stably stratified surface layer is proposed. The Fourier coefficients are obtained by numerically solving the linearized NS equations with Boussinesq approximation in spectral space, under the assumption of “rapid” deformation (RDT) due to combined shear and stratification. The linearization of RDT, which is unrealistic for the surface layer, is rectified using Mann’s (JFM, 1994) idea of wavenumber dependent eddy lifetime. The input parameters required by the KS are estimated using either Monin-Obukhov theory, or an appropriate Second Moment Closure. In order to overcome the frozen turbulence hypothesis made in the Mann model, we incorporate inter-scale “sweeping” of eddies following the ideas of Fung, et. al. (JFM, 1992), along with temporal decorrelation associated with the natural eddy time scale. The solenoidal velocity field generated by the KS allows inclusion of a wide range of scales with correct space-time correlations, making it ideal to investigate particle dispersion in a stably stratified environment, and can also serve as inflow for the study of Wind Farm-PBL interactions. The effect of varying Obukhov length will be discussed by analyzing the frozen Eulerian spectra and Lagrangian particle dispersion. 2:28PM D23.00002 An Immersed Boundary Method for the simulation of turbulent stratified flows over rough topography , NARSIMHA RAPAKA, SUTANU SARKAR, Univ of California - San Diego — An Immersed Boundary Method (IBM) is developed to simulate stratified flows over rough topography using a Cartesian grid. The solver is validated in the problem of tidal flow over a laboratory-scale (order of few meters) smoothed triangular ridge. The results including phasing of turbulence, statistics and baroclinic wave flux agree well with the those obtained using in the DNS studies of Rapaka et al. (JFM 2013) and Jalali et al. (JFM 2014) performed with a body-fitted grid. The bottom drag is parameterized using the DNS data and tested for simulations using coarser grids. The beam thickness is increased with the boundary layer being under-resolved but the baroclinic wave flux agree well. Phasing of turbulence is qualitatively similar at both low and high Excursion numbers (Ex). Integrated TKE and dissipation are of the same order as in the DNS. Large Eddy Simulations (LES) are performed on a large scale topography, of the order of few kilometers, with the same Ex and the criticality parameter but significantly larger Reynolds number. The baroclinic response is stronger than the laboratory scale model owing to the larger length of critical slope. The baroclinic flux as well as turbulence energetics and phasing are studied. 2:41PM D23.00003 Mixing structures in stratified shear flow: Dependence on gradient Richardson number , ROBERT ECKE, Los Alamos National Laboratory, PHILIPPE ODIER, Ecole Normale Superieure de Lyon, JUN CHEN, Purdue University — We report experimental measurements of velocity and density fields of wall-bounded stratified shear-flow turbulence as a function of shear velocity and density difference. A turbulent channel flow exits a nozzle onto the underside of a flat transparent plate at fixed velocity 4 < U < 8 cm/s and with a fractional density difference 0.0027 < ∆ρ/ρ < 0.0054 between the inflowing fluid and the quiescent fluid in the tank. Simultaneous velocity and density measurements are obtained using PIV and PLIF, respectively (see P. Odier, J. Chen, and R.E. Ecke, J. Fluid Mech. 794, 498 (2014) for experimental details). For a resultant range of gradient Richardson numbers 0.05 < Rig < 0.5, we compute and compare different measures of turbulent mixing obtained from direct measurement of turbulent dissipation, Reynolds stress, buoyancy flux, and density and velocity gradients. In particular, we obtain mixing lengths, Ozmidov and shear lengths, Thorpe lengths, buoyancy Reynold number, flux Richardson number, diapycnal mixing parameter, and intermittency properties of flows as a function of gradient Richardson number. These quantities characterize the transition from shear dominated flow at low Rig to stratification dominated behavior at larger Rig . 2:54PM D23.00004 Evaluation of the standard k-ǫ closure scheme for modeling stably stratified wall-bounded turbulence1 , AMRAPALLI GARANAIK, FARID KARIMPOUR, SUBHAS VENAYAGAMOORTHY, Colorado State University — Reynolds-averaged Navier-Stokes (RANS) turbulence models are widely used for modeling stratified turbulent flows. The focus of this study is to account for the effect of the buoyancy forces in the two-equation standard k-ǫ closure scheme for modeling stably stratified wall-bounded turbulence. The buoyancy parameter (Cǫ3 ) is analytically revisited and it is found that it can be neglected in the evolution equation of the dissipation rate of the turbulent kinetic energy. Furthermore, we use different propositions for the turbulent Prandtl number (P rt ) to assess their efficacy for modeling stratified wall-bounded flows. Numerical simulations are implemented in a 1-D water column model and the results are compared with data of direct numerical simulation of stably stratified channel flow. 1 Funded by the National Science Foundation and the Office of Naval Research 3:07PM D23.00005 Porous Sphere in Stratified Environments: Entrainment and Diffusion1 , ROBERTO CAMASSA, CLAUDIA FALCON, SHILPA KHATRI, RICHARD MCLAUGHLIN, University of North Carolina, UNC JOINT FLUIDS LAB TEAM — A theoretical, experimental, and numerical study of porous spheres falling in stratified fluids will be presented. The systematic justification of asymptotic regimes resulting in asymptotic models with “heat bath” boundary conditions for salinity are derived in low Reynolds number regimes. Violation of these asymptotic scalings will be discussed in the context of experiments and mathematical modeling. In particular the presence of a salt depletion or enrichment wake left behind by the settling, ab/de-sorbing sphere, and its competition with entrainment, will be presented and highlighted. Experimental results with microporous spheres as well calibrated manufactured drilled spheres will be compared. 1 Supported by : NSF CMG , NSF RTG, ONR 3:20PM D23.00006 LIF measurements of the flow past a sphere descending in a stratified fluid , SHINSAKU AKIYAMA, SHINYA OKINO, HIDESHI HANAZAKI, Kyoto Univ — When a sphere descends in the stratified salt water, a strong upward jet is often generated above the sphere. In this study, the flow is observed by the laser induced fluorescence (LIF) method, assuming the proportionality between the concentrations of salt and fluorescent dye. In particular, the radius of the jet and the thickness of the density boundary layer on the sphere surface are measured. It is found that the radius of the jet is proportional to both F r1/2 (F r: Froude number) and Re−1/2 (Re: Reynolds number), in agreement with the simple dimensional analysis. The density boundary layer on the sphere surface also becomes thinner as F r decreases or Re increases, showing a similar trend. These results are explained by a scenario that the fluid in and near the density boundary layer on the sphere moves up along the sphere surface, changing its density across the isopycnals to finally form a jet above the sphere. 3:33PM D23.00007 The near wake of a towed grid in a stratified fluid1 , XINJIANG XIANG, TRYSTAN MADISON, PRABU SELLAPAN, GEOFFREY SPEDDING, University of Southern California — Though much detailed quantitative information has been assembled to describe the late wakes behind various objects in stably-stratified fluids, much less is known about the early stages when the flow begins to feel the effects of the background density gradient. Here we report on experiments on the early wake of a towed grid, with Re ∈ {2700, 11000}, and F r ∈ {0.6, 9.1}. Internal waves are found for all F r, originating as the flow turns around the obstacle, with wavelength linearly proportional to F r and approximately constant amplitude. The mean centerline stream-wise velocity is strongly affected by the lee waves, and so depends on F r. Strong vertical shear is observed at the wake edge, leading to overturning through Kelvin-Helmholtz instabilities. Stratified turbulence develops up to N t ≈ 10 (except at the lowest F r), with buoyancy Reynolds number independent of F r at higher N t. Developing anisotropy in the horizontal and vertical directions in the early wake is described for both mean and fluctuating quantities. The data and their variation with Re and F r comprise a start towards making a generally available database for detailed comparisons with numerical experiment. 1 Support from ONR N00014-11-1-0553 is gratefully acknowledged. 3:46PM D23.00008 Early wake characteristics of a towed sphere in a stably stratified fluid1 , TRYSTAN MADISON, XINJIANG XIANG, PRABU SELLAPAN, GEOFFREY SPEDDING, University of Southern California — Turbulence in the ocean is intermittent in both space and time, and so the way in which isolated patches of turbulence decay and propagate in the ocean and atmosphere is of practical interest to many. The evolution and decay of bluff body wakes in a linearly density gradient has proven to be useful test case, and though much has been documented on the late wake, much less is known at early times, despite the fact that it is at early times that the non-equilibrium and late wake conditions are determined. Here we present and compare suitable quantitative measures for sphere wakes when 0.1 ≤ N t ≤ 10 for 200 ≤ Re ≤ 2000 and 1 ≤ F r ≤ 16. The comparisons are over both Re and F r and with published results from recent numerical experiments. 1 Support from ONR N00014-11-1-0553 is gratefully acknowledged. 3:59PM D23.00009 Mixing Efficiency in Stratified Turbulent Jets1 , CHUNG-NAN TZOU, ROBERTO CAMASSA, SIAN LEWIS-BEVAN, RICHARD MCLAUGHLIN, NATHAN PERREAU, Unc Joint Fluids Lab, UNC JOINT FLUIDS LAB TEAM — Building upon prior results of the authors establishing rigorously the optimal mixing profile of a turbulent buoyant jet in a special class of stratified environments, a substantial expansion to a broader family of background stratifications is considered both experimentally and analytically with some surprising observations. 1 NSF, ONR 4:12PM D23.00010 Mixing efficiency in shear-driven and convective turbulent stratified flows1 , ALBERTO SCOTTI, BRIAN WHITE, Dept. of Marine Sciences, UNC, Chapel Hill — DNS of steady-state and time-evolving mixing layers are used to calculate the mixing efficiency under different forcing conditions. Two basic mechanisms to sustain turbulence are considered: shear acting against a stably stratified background, and zero-shear, but convectively unstable regions embedded in a stratified fluid. When turbulence is produced by shear the mixing efficiency can be collapsed in terms of the buoyancy Reynolds number for values of the latter less than 20, whether the flow is steady or time evolving. For higher values, no such collapse exists. The efficiency of time-evolving shear-driven mixing layers approaches a constant value during the turbulent phase of about 0.15. In steady-state flows, on the contrary, the mixing efficiency is controlled to leading order by the externally imposed Richardson number. We show that the difference is due to the different way in which buoyancy and momentum are fed into the mixing layer. In overturning mixing layers we observe larger values of the mixing efficiency, approaching 1/2. The results suggests that the practice of adopting a constant mixing efficiency in parameterizing shear-driven episodic mixing events is justified in environmental flows. 1 Work supported by the National Science Foundation Sunday, November 23, 2014 2:15PM - 4:12PM Session D24 Industrial Applications II — 2003 - Amos Winter, Massachusetts Institute of Technology 2:15PM D24.00001 Simulation of an Oscillating Hydrofoil near Boundaries , JENNIFER FRANCK, KA LING WU, Brown University — An oscillating hydrofoil in freestream flow is computationally investigated for the application of a novel hydrokinetic energy device. The hydrofoil is prescribed a rigid body sinusoidal motion in pitch and heave, with the maximum efficiency occurring with a heave amplitude of 0.5 chord lengths, pitch amplitude of 75 degrees, a non-dimensional frequency of 0.15, and a phase difference of 90 degrees between pitch and heave. Simulations are performed using 2D direct numerical simulation with a moving mesh algorithm and compared to particle image velocimetry results of Strom et. al 2014 and previous computational results of Frank and Franck 2013. The high pitch angle during mid-downstroke and upstroke generates large coherent vortices that enhance the power generated from linear translation, yet remain on the wing during stroke reversal to also generate power from the angular motion. In order to assess the hydrofoil’s performance in shallow tidal zones, a lower wall boundary is introduced into the simulation to investigate the ground effect on the unsteady vortex shedding and power generation. 2:28PM D24.00002 Parametric dependence of energy harvesting performance with an oscillating hydrofoil1 , BENJAMIN STROM, DAEGYOUM KIM, SHREYAS MANDRE, KENNETH BREUER, Brown University — We report on experiments on tidal energy conversion from a open channel water flow using an oscillating hydrofoil. The hydrofoil is operated at high angles of attack such that the formation and capture of a leading edge vortex greatly enhances the energy conversion efficiency. A computer-controlled pitch and heave system allows for arbitrary position profiles to be imposed. Force and torque measurements are used to determine the energy harvesting efficiency as a function of Reynolds number, pitch and heave amplitudes, phase shift, the location of the pitching axis, position profile, and the cross sectional shape of the hydrofoil. PIV measurements are used to capture the vortex dynamics and these results are compared to the computational results of Frank and Franck (2013). Efficiency was found to be most sensitive to pitch amplitude, pitching axis and phase shift with relatively little dependence on Reynolds number, heave amplitude, and foil shape. 1 Work supported by DOE-ARPAe 2:41PM D24.00003 The effect of aspect ratio on the performance of an energy harvesting hydrofoil1 , DAEGYOUM KIM, BENJAMIN STROM, YUNXING SU, SHREYAS MANDRE, KENNETH BREUER, Brown University — We investigated the effect of aspect ratio on energy harvesting performance and flow structure of an oscillating hydrofoil. Power measurement and particle image velocimetry were performed in a water flume with a hydrofoil undergoing periodic heaving and pitching motions. Aspect ratio was varied from 2.5 to 4.5, and end plates were also mounted at the hydrofoil tips in order to suppress three-dimensional effects near the tips. For each aspect ratio, energy conversion efficiency was maximum at the same kinematics determined by reduced frequency and pitch amplitude. The efficiency is increased with the aspect ratio, and it is noticeably enhanced with the installation of the end plates. Leading-edge vortex formation and wake dynamics were compared at several spanwise sections among different aspect ratios. Their correlation with the efficiency was also examined. 1 This research was supported by DOE ARPA-E. 2:54PM D24.00004 Real-time Coupled Ensemble Kalman Filter Forecasting & Nonlinear Model Predictive Control Approach for Optimal Power Take-off of a Wave Energy Converter , DANIELE CAVAGLIERI, THOMAS BEWLEY, UC San Diego, MIRKO PREVISIC, Re Vision Consulting LLC — In recent years, there has been a growing interest in renewable energy. Among all the available possibilities, wave energy conversion, due to the huge availability of energy that the ocean could provide, represents nowadays one of the most promising solutions. However, the efficiency of a wave energy converter for ocean wave energy harvesting is still far from making it competitive with more mature fields of renewable energy, such as solar and wind energy. One of the main problems is related to the difficulty to increase the power take-off through the implementation of an active controller without a precise knowledge of the oncoming wavefield. This work represents the first attempt at defining a realistic control framework for optimal power take-off of a wave energy converter where the ocean wavefield is predicted through a nonlinear Ensemble Kalman filter which assimilates data from a wave measurement device, such as a Doppler radar or a measurement buoy. Knowledge of the future wave profile is then leveraged in a nonlinear direct multiple shooting model predictive control framework allowing the online optimization of the energy absorption under motion and machinery constraints of the device. 3:07PM D24.00005 Numerical Simulations and Experimental Measurements of Scale-Model Horizontal Axis Hydrokinetic Turbines (HAHT) Arrays1 , TEYMOUR JAVAHERCHI, NICK STELZENMULLER, University of Washington, JOSEPH SEYDEL, Boeing, ALBERTO ALISEDA, University of Washington — The performance, turbulent wake evolution and interaction of multiple Horizontal Axis Hydrokinetic Turbines (HAHT) is analyzed in a 45:1 scale model setup. We combine experimental measurements with different RANS-based computational simulations that model the turbines with sliding-mesh, rotating reference frame and blame element theory strategies. The influence of array spacing and Tip Speed Ratio on performance and wake velocity structure is investigated in three different array configurations: Two coaxial turbines at different downstream spacing (5d to 14d), Three coaxial turbines with 5d and 7d downstream spacing, and Three turbines with lateral offset (0.5d) and downstream spacing (5d & 7d). Comparison with experimental measurements provides insights into the dynamics of HAHT arrays, and by extension to closely packed HAWT arrays. The experimental validation process also highlights the influence of the closure model used (k-ω SST and k-ǫ) and the flow Reynolds number (Re=40,000 to 100,000) on the computational predictions of devices’ performance and characteristics of the flow field inside the above-mentioned arrays, establishing the strengths and limitations of existing numerical models for use in industrially-relevant settings (computational cost and time). 1 Supported by DOE through the National Northwest Marine Renewable Energy Center (NNMREC) 3:20PM D24.00006 Experimental assessment of blade tip immersion depth from free surface on average power and thrust coefficients of marine current turbine , ETHAN LUST, KAREN FLACK, LUKSA LUZNIK, US Naval Academy — Results from an experimental study on the effects of marine current turbine immersion depth from the free surface are presented. Measurements are performed with a 1/25 scale (diameter D=0.8m) two bladed horizontal axis turbine towed in the large towing tank at the U.S. Naval Academy. Thrust and torque are measured using a dynamometer, mounted in line with the turbine shaft. Shaft rotation speed and blade position are measured using a shaft position indexing system. The tip speed ratio (TSR) is adjusted using a hysteresis brake which is attached to the output shaft. Two optical wave height sensors are used to measure the free surface elevation. The turbine is towed at 1.68 m/s, resulting in a 70% chord based Rec=4 x 105 . An Acoustic Doppler Velocimeter (ADV) is installed one turbine diameter upstream of the turbine rotation plane to characterize the inflow turbulence. Measurements are obtained at four relative blade tip immersion depths of z/D = 0.5, 0.4, 0.3, and 0.2 at a TSR value of 7 to identify the depth where free surface effects impact overall turbine performance. The overall average power and thrust coefficient are presented and compared to previously conducted baseline tests. The influence of wake expansion blockage on the turbine performance due to presence of the free surface at these immersion depths will also be discussed. 3:33PM D24.00007 Large-eddy simulation of turbulent flow past tri-frame configurations of hydrokinetic turbines in an open channel1 , SAURABH CHAWDHARY, St. Anthony Falls Laboratory, Department of Mechanical Engineering, University of Minnesota, XIAOLEI YANG, CRAIG HILL, MICHELE GUALA, FOTIS SOTIROPOULOS, St. Anthony Falls Laboratory, Department of Civil Engineering, University of Minnesota — An effective way to develop arrays of hydrokinetic turbines in streams and tidal sites is to arrange them in tri-frame configurations, where three turbines are mounted together at the apexes of a triangular frame. Turbines mounted on a tri-frame can serve as the building block for rapidly deploying multi-turbine arrays. We employ large-eddy simulation (LES) to understand wake interactions of turbines mounted on tri-frame configurations and develop design guidelines for field deployment. We employ the computational framework of Yang et.al. (2013) to simulate the flow past turbines with the turbines modeled as actuator lines. The computed results are compared with experiments conducted at the Saint Anthony Falls Lab (SAFL) in terms of mean flow and turbulence characteristics. The flow fields are analyzed to elucidate the mechanisms of turbine interactions and wake evolution in tri-frame configurations and to develop design guidelines for maximizing the combined power output while reducing structural loads due to turbulent fluctuations. 1 This work was supported by NSF grant IIP-1318201. The simulations were carried out at the Minnesota Supercomputing Institute. 3:46PM D24.00008 Experimental investigation on axial-flow turbine arrays in erodible and non-erodible channels: Performance, flow-field, and bathymetric interactions1 , CRAIG HILL, FOTIS SOTIROPOULOS, MICHELE GUALA, University of Minnesota — Natural channels ideal for hydrokinetic turbine installations present complex environments containing asymmetric flow, regions of high shear and turbulent eddies that impact turbine performance. To understand the impacts caused by variable topography, baseline conditions in a laboratory flume are compared to turbine performance, flow characteristics, and channel topography measurements from two additional experiments with small-scale and large-scale bathymetric features. Both aligned and staggered multi-turbine configurations were investigated. Small-scale axial-flow rotors attached to miniature DC motors provided measurements of turbine performance and response to i) complex topographic features and ii) flow features induced by upstream turbines. Discussion will focus on optimal streamwise and lateral spacing for axial-flow devices, turbine-topography interactions within arrays and inter-array flow-field measurements. Primary focus will center on results from turbines separated by a streamwise distance of 7dT. Additionally, results indicate possible control strategies for turbines installed in complex natural environments. 1 This work was supported by NSF PFI grant IIP-1318201, CAREER: Geophysical Flow Control (NSF). 3:59PM D24.00009 The dynamic interaction of a marine hydrokinetic turbine with its environment , NITIN KOLEKAR, ARINDAM BANERJEE, Lehigh University — Unlike wind turbines, marine hydrokinetic and tidal turbines operate in a bounded flow environment where flow is constrained between deformable free surface and fixed river/sea bed. The proximity to free surface modifies the wake dynamics behind the turbine. Further, size & shape of this wake is not constant but depends on multiple factors like flow speed, turbine blade geometry, and rotational speed. In addition, the turbulence characteristics of incoming flow also affects the flow field and hence the performance. The current work aims at understanding the dynamic interaction of a hydrokinetic turbine (HkT) with free surface and flow turbulence through experimental investigations. Results will be presented from experimental study carried out in an open channel test facility at Lehigh University with a three bladed, constant chord, zero twist HkT under various operating conditions. Froude number (ratio of characteristic flow velocity to gravitational wave velocity) is used to characterize the effect of free surface proximity on turbine performance. Experimental results will be compared with analytical models based on blade element momentum theory. Characterization of wake meandering and flow around turbine will be performed using a stereo-Particle Image Velocimetry technique. Sunday, November 23, 2014 2:15PM - 4:25PM Session D25 Turbulence: General I — 2005 - Eberhard Bodenschatz, Max Planck Institute for Dynamics and Self- Organization 2:15PM D25.00001 The role of wall confinement on the decay rate of an initially isotropic turbulent field1 , DAVID R. DOWLING, POOYA MOVAHED, ERIC JOHNSEN, University of Michigan, Ann Arbor — The problem of freely decaying isotropic turbulence has been the subject of intensive research during the past few decades due to its importance for modeling purposes. While isotropy and periodic boundary conditions assumptions simplify the analysis, large-scale anisotropy (e.g., caused by rotation, shear, acceleration or walls) is in practice present in most turbulent flows and affects flow dynamics across different scales, as well as the kinetic energy decay. We investigate the role of wall confinement and viscous dissipation on the decay rate of an initially isotropic field for confining volumes of different aspect ratios. We first generate an isotropic velocity field in a cube with periodic boundary conditions. Next, using this field, we change the boundary conditions to no-slip walls on all sides. These walls restrict the initial field to a confined geometry and also provide an additional viscous dissipation mechanism. The problem is considered for confining volumes of different aspect ratios by adjusting the initial field. The change in confining volume introduces an additional length scale to the problem. Direct numerical simulation of the proposed set-up is used to verify the scaling arguments for the decay rate of kinetic energy. 1 This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1053575. 2:28PM D25.00002 Properties of Streamline Segments in Turbulent Channel Flows with Wavy Walls , FABIAN HENNIG, JONAS BOSCHUNG, NORBERT PETERS, RWTH Aachen University — We investigate the turbulent velocity field by means of instantaneous streamlines. The streamlines are partitioned into segments and decompose the velocity field in a non-arbitrary way. We conducted direct numerical simulations of a channel flow with one wavy and one plane wall. The results have been validated against DNS and experimental data from literature. Based on the DNS we investigate the properties of streamlines and streamline segments in detail. 2:41PM D25.00003 The kinematics of the reduced velocity gradient tensor in a fully developed turbulent free shear flow , OLIVER BUXTON, ANDREW WYNN, PAIGE RABEY, Imperial College London — The reduced velocity gradient tensor (VGT) is defined as a 2 × 2 block, from a single interrogation plane, of the full VGT, ∂ui /∂xj . Direct numerical simulation data from the fully developed turbulent region of a nominally two-dimensional mixing layer is used in order to examine the extent to which information on the full VGT can be derived from the reduced VGT. It is shown that the reduced VGT is able to reveal significantly more information about regions of the flow in which strain-rate is dominant over rotation. It is thus possible to use the assumptions of homogeneity and local isotropy to place bounds on the first two statistical moments of the eigenvalues of the reduced strain-rate tensor (symmetric part of the reduced VGT), which in turn relate to the turbulent strain-rates. These bounds are shown to be dependent upon the kurtosis of ∂u1 /∂x1 and another variable defined from the constituents of the reduced VGT. The kurtosis is observed to be minimised on the centreline of the mixing layer and thus tighter bounds are are possible at the centre of the mixing layer than the periphery. Nevertheless, these bounds are observed to hold for the entirety of the mixing layer, despite some significant departures from local isotropy. 2:54PM D25.00004 Slow dynamics at Re = 108 in turbulent Helium flows1 , JAVIER BURGUETE, University of Navarra, PHILIPPE ROCHE, Insitut Néel, CNRS, BERNARD ROUSSET, SBT - CEA Grenoble — The presence of slow dynamics is a recurrent feature of many turbulent flows. This behaviour can be created by instabilities of the mean flow or by other mechanisms [1,2]. In this work we analyze the behavior of a highly turbulent flow (maximum Reynolds number Re = 108 , with a Reynolds based on the Taylor microscale Reλ = 2000). The experimental cell consists on a closed cavity filled with liquid Helium (330 liters) close to the lambda point (between 1.8 and 2.5 K) where two inhomogeneous and strongly turbulent flows collide in a thin region. The cylindrical cavity has a diameter of 78cm and two impellers rotate in opposite directions with rotation frequencies up to 2Hz. The distance between the propellers is 70cm. Different experimental runs have been performed, both in the normal and superfluid phases. We have performed velocity measurements using home-made Pitot tubes. Here we would like to present preliminary results on this configuration. The analysis of the data series reveals that below the injection frequencies there are different dynamical regimes with time scales two orders of magnitude below the injection scale. [1] A. de la Torre, J. Burguete, Phys Rev Lett 99 (2007) 054101. [2] M. Lopez-Caballero, J. Burguete, Phys Rev Lett 110 (2013) 124501. 1 We acknowledge support from the EuHIT network and the SHREK Collaboration 3:07PM D25.00005 Decay Power Law in, High Intensity, Isotropic Turbulent Flow1 , TIMOTHY KOSTER, ALEJANDRO PUGA, JOHN LARUE, Univ of California - Irvine — In the study reported here, isotropy is determined using the measure proposed by George (1992), where isotropy corresponds to those downstream positions where the product of the Taylor Reynolds number and the skewness of the velocity derivative is a constant. Straight forward approach can be used which is based on the observation of Batchelor (1953), that the square of the Talor micorscale is linearly related to downstream distance relative to the virtual origin. The fact that the decay of downstream velocity variance is described by a power law is shown to imply power law behavior for various other parameters such as the dissipation, the integral length scale, the Taylor microscale, the Kolmogorov microscale and the Taylor Reynolds number and that there is an algebraic relationship between the various power law exponents. Results are presented for mean velocities of 6 and 8 m/s for the downstream decay of the parameters listed in the preceding. The corresponding values of the Taylor Reynolds number at the start of the isotropic region are 290 and 400, and the variance decay exponent and virtual origin are found to be respectively -1.707 and -1.298 and -27.95 and -5.757. The exponents in the decay law for the other parameters are found to be within ± 3% of the expected values. 1 University of California Irvine Research Funds 3:20PM D25.00006 Direct and inverse energy cascades in a forced rotating turbulence experiment , ANTOINE CAMPAGNE, Laboratoire FAST, CNRS, Université Paris-Sud, 91405 Orsay, France, BASILE GALLET, Laboratoire SPHYNX, Service de Physique de l’État Condensé, DSM, CEA Saclay, CNRS, 91191 Gif-sur-Yvette, France, FRÉDÉRIC MOISY, PIERRE-PHILIPPE CORTET, Laboratoire FAST, CNRS, Université Paris-Sud, 91405 Orsay, France — Turbulence in a rotating frame provides a remarkable system where 2D and 3D properties may coexist, with a possible tuning between direct and inverse cascades. We present here experimental evidence for a double cascade of kinetic energy in a statistically stationary rotating turbulence experiment. Turbulence is generated by a set of vertical flaps which continuously injects velocity fluctuations towards the center of a rotating water tank. The energy transfers are evaluated from two-point third-order three-component velocity structure functions, which we measure using stereoscopic PIV in the rotating frame. Without global rotation, the energy is transferred from large to small scales, as in classical 3D turbulence. For nonzero rotation rates, the horizontal kinetic energy presents a double cascade: a direct cascade at small horizontal scales and an inverse cascade at large horizontal scales. By contrast, the vertical kinetic energy is always transferred from large to small horizontal scales, a behavior reminiscent of the dynamics of a passive scalar in 2D turbulence. At the largest rotation rate, the flow is nearly 2D and a pure inverse energy cascade is found for the horizontal energy. 3:33PM D25.00007 A tractable representation of the non-linear terms in spectral space applied to isotropic turbulence , LAWRENCE CHEUNG, None, TAMER ZAKI, Johns Hopkins University, Imperial College London — The principal challenge in the analytical treatment of the Navier-Stokes equations in spectral space is the complex nature of the nonlinear triad interactions. In Fourier basis, these interactions are expressed as an infinite convolution sum over all wavenumber pairs. A tractable representation is introduced in terms of a combination matrix which recasts the convolution in a bilinear form. With this new representation, a spectral energy equation is derived, and its bilinear form facilitates the choice of the appropriate canonical basis. The utility of the formulation is demonstrated by considering the problem of homogeneous, isotropic turbulence. By invoking well-established assumptions, for example the presence of an inertial range where the energy decay rate is independent of wavenumber, we derive Kolmogorov’s -5/3 scaling analytically, without any dimensional arguments. The same analytical framework also accurately predicts the spectral characteristics of scalar fluctuations. 3:46PM D25.00008 Pressure Rate of Strain, Pressure Diffusion and Velocity Pressure Gradient Tensor Measurements in a Cavity Shear Layer Flow1 , XIAOFENG LIU, San Diego State University, JOSEPH KATZ, Johns Hopkins University — Pressure related turbulence statistics of a 2D open cavity shear layer flow was investigated experimentally in a water tunnel at a Reynolds number of 40,000. Time-resolved PIV sampled at 4500 fps and a field of view of 25x25 mm was used to simultaneously measure the instantaneous velocity, material acceleration and pressure distributions. The pressure was obtained by spatially integrating the measured material acceleration. Results based on 150,000 measurement samples enable direct estimates of components of the pressure-rate-of-strain, pressure diffusion and velocity-pressure-gradient tensors. The pressure and streamwise velocity correlation changes its sign from negative values far upstream from the downstream corner to positive values near the corner due to the strong adverse pressure gradient imposed by the corner. Moreover, once its sign changes, the pressure-velocity correlation preserves its positive value for the streamwise correlations, and negative value for the spanwise correlations, even after the shear layer propagates beyond the adverse pressure gradient region along both the vertical and horizontal corner walls. The pressure diffusion term is of the same order as the production rate. In the shear layer, the streamwise pressure-rate-of-strain term, R11 , is mostly negative while the perpendicular term, R22 , is positive but with a smaller magnitude, implying turbulent energy redistribution from streamwise to lateral directions. 1 Sponsored by ONR and NSF. 3:59PM D25.00009 Redistribution of kinetic energy by pressure forces in turbulence , ALAIN PUMIR, Ecole Normale Superieure de Lyon, Lyon, France, HAITAO XU, EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Self Organisation, Goettingen, Germany, GREGORY FALKOVICH, Weizmann Institute of Sciences, Rehovot, 76100, Israel, GUIDO BOFFETTA, University of Torino, Torino, I-10125, Torino — In statistically homogeneous turbulent flows, pressure forces provide the main mechanism to redistribute kinetic energy among fluid elements, without net contribution to the overall energy budget. This holds true in both two-dimensional (2D) and three-dimensional (3D) flows, which show fundamentally different physics. As I will demonstrate, pressure forces act on fluid elements very differently in these two cases. Numerical simulations demonstrate that in 3D pressure forces strongly accelerate the fastest fluid elements, an effect which is absent in 2D. In 3D turbulence, these findings suggest a mechanism for a possibly singular build-up of energy, and thus may shed new light on the smoothness problem of the solution of the Navier-Stokes equation in 3D. 4:12PM D25.00010 Atmospheric Scintillations: A Clue for Bird Orientation and Navigation , CHARLES PETTY, ANDREW BOWDEN, ANDRE BENARD, Michigan State University — The index-of-refraction of the troposphere is anisotropic at all scales even if the local turbulent velocity field is statistically homogeneous. This anisotropy is partly due to the coupling between the fluctuating velocity field with the Coriolis field and the Lorentz field. Thus, the redistribution of turbulent kinetic energy and the concomitant anisotropy in the index-of-refraction may provide a practical means for birds (and other animals and insects) to orient and navigate. Consequently, if birds migrate between two points on the Earth by following a great circle path, then local anisotropic scintillation phenomena may provide a means to determine the latitude, the longitude, and the bearing along an orthodromic migration path. Thus, scintillation phenomena may be an important fundamental component in the underlying mechanics that support bird orientation and navigation. Sunday, November 23, 2014 2:15PM - 4:25PM Session D26 Turbulent Boundary Layers II — 2007 - Joseph Klewicki, University of New Hampshire 2:15PM D26.00001 Fluctuating wall shear stress and velocity measurements in a turbulent boundary layer , ROMMEL PABON, LAWRENCE UKEILEY, CASEY BARNARD, MARK SHEPLAK, University of Florida — Knowledge of mean wall shear stress on a surface can shed light on important performance parameters, but the fluctuating shear, even in simple flows, has not been as easily measured, and can be of interest in fundamental boundary layer research. Experiments on a flat plate model were performed to investigate the relationship between the wall shear stress and large scale events in the turbulent boundary layer. A MEMS based differential capacitance shear stress system with 1mm × 1mm floating element which can measure the fluctuating and static components of shear simultaneously, coupled with a hot wire anemometer were used for characterizing the turbulent boundary layer. Velocity profiles and turbulence statistics approaching the wall characterized the two dimensionality of the flat plate, and a trailing edge flap was used to impose a zero pressure gradient. The mean streamwise velocity profile was scaled by the friction velocity using the measured shear stress and independently compared to classical fits. Correlations between the fluctuating shear and measured velocities were used to elucidate the large scale events and to compare with previous fluctuating shear measurements for validation. 2:28PM D26.00002 Floating element measurements of wall-shear stress exerted by highReynolds-number turbulent boundary layers , WOUTIJN J. BAARS, University of Melbourne, KRISHNA M. TALLURU, University of Newcastle, NICK HUTCHINS, IVAN MARUSIC, University of Melbourne — Indirect methods to obtain the wall-shear stress τw , such as the Clauser chart fit, necessitate p inherent assumptions of the boundary layer. Therefore, direct methods are preferred to measure τw and subsequently obtain the friction velocity Uτ = τw /ρ. Floating elements are genuinely small to obtain local wall-shear stress measurements, but cope with low signal-to-noise ratios since the signal scales with the surface area (∝ l2 ), where l is the characteristic length, and the error forces scale with hl; h represents the misalignment of the edges. Therefore, the capacious High Reynolds Number Boundary Layer Wind Tunnel at Melbourne incorporates a large floating element of 3m × 1m over which the changes in boundary layer parameters are negligible, and hence, local measurements of Uτ are made with high accuracy. Smooth-wall results follow the U∞ /Uτ = 1/κ ln (Reθ ) + C trend within ±1% (κ = 0.380 and C = 3.7) for typical test conditions ranging from Reθ = 15, 000 to 45, 000. Moreover, the device is used to measure Uτ corresponding to rough walls, and boundary layers that are perturbed by flush-mounted control devices within the element. 2:41PM D26.00003 Large-scale structures in high Reynolds number turbulent boundary layers , CHRISTIAN J. KÄHLER, NICOLAS BUCHMANN, SVEN SCHARNOWSKI, Bundeswehr University Munich — Large-scale turbulent flow structures in a flat plate turbulent boundary layer with zero pressure gradient were investigated with PIV in a closed-loop transonic wind tunnel at Ma = 0.5 - 0.8, Reτ = 5,100 9,500. The primary aim of the measurements was to simultaneously reach large Reynolds numbers and a low boundary layer thickness to model width ratio. The high Reynolds number is required to generate large scale structures with sufficient amplitude. The low ratio between the boundary layer thickness and the width of the model is necessary to avoid side effects resulting from the side walls and the blockage of the flow. Spatial two-point correlation and conditional correlations are calculated to determine the size and orientation of the large-scale flow structures. The difference between the correlation and conditional correlation analysis indicates that various large-scale structures with typical topologies exist and interact in turbulent boundary layer flows. 2:54PM D26.00004 Spectral scaling in boundary layers and pipes at very high Reynolds numbers1 , MARGIT VALLIKIVI, Princeton University, BHARATH GANAPATHISUBRAMANI, University of Southampton, ALEXANDER SMITS, Princeton University, Monash University — One-dimensional energy spectra in flat plate zero pressure gradient boundary layers and pipe flows are examined over a wide range of Reynolds numbers (2, 600 ≤ Reτ ≤ 72, 500). The peaks associated with the large-scale motions and superstructures in boundary layers behave as they do in pipe flows, with some minor differences. The location of the outer spectral peak (OSP) displays only a weak dependence on Reynolds number, and it occurs at the same wall-normal distance where the variances establish a logarithmic behavior. The outer-scaled wavelength of the OSP appears to decrease with increasing Reynolds number implying that the superstructures represent the inertial range of motions rather than the large scales per se. The location of the OSP appears to mark the start of a plateau that is consistent with a kx−1 slope in the spectrum and the logarithmic variation in the variances. It does not require full similarity between outer and wall-normal scaling on the wavenumber. The extent of kx−1 region depends on the wavelength of the OSP, which appears to emerge as a true inertial scale only at Reynolds numbers typical of atmospheric surface layers. 1 Supported by ONR Grant N00014-13-1-0174 and ERC Grant agreement No. 277472. 3:07PM D26.00005 Hybrid PIV-Particle Tracking Technique Applied to High Reynolds Number Turbulent Boundary Layer Measurements1 , JULIO SORIA, CALLUM ATKINSON, NICOLAS BUCHMANN, Monash University — Zero-pressure gradient turbulent boundary layer (ZPGTBL) experiments in both the LTRAC water tunnel (LTRAC-WT) at Monash University and the University of Melbourne HRNTBL wind tunnel were undertaken up to Reτ = 20, 000. Both experiments represent the flow along the centreline of each test section in the streamwise - wall-normal plane. Two PCO Dimax cameras with 2008 × 2008 pixel arrays were used in conjunction with a high-repetition laser to measure the time-resolved 2C-2D velocity field in the LTRAC-WT. The HRNTBL wind tunnel experiments employed nine PCO pro.4000 cameras each with a 4008 × 2672 pixel array. Two double cavity pulsed Nd:YAG lasers were used to acquire single-exposed PIV images simultaneously on all cameras to yield high spatial resolution 2C-2D velocity fields over a large extend of the ZPGTBL. In both cases, the single-exposed PIV images were analysed using multigrid cross-correlation PIV analysis, which has been enhanced to include a hybrid velocimetry step in which the PIV velocity is used as an estimator to a subsequent particle tracking refinement step. Statistics of the ZPGTBL velocity field as well as the high spatial resolution instantaneous structure and spatio-temporal structure of the ZPGTBL will be presented. 1 The support of the Australian Research Council via Discovery and LIEF grants is gratefully acknowledged. 3:20PM D26.00006 Resolved measurements of the near-wall coherent structures , GERRIT E. ELSINGA, Delft University of Technology, YOSHIFUMI JODAI, Kagawa National College of Technology, Japan — The 3D coherent structures in the near-wall region of a turbulent boundary layer have been measured by time-resolved tomographic PIV. The Reynolds number based on the friction velocity was 814. The measurement volume extended from the wall up to a y+ of 170, and it spanned 680x620 wall units in the streamwise and spanwise direction respectively. The spatial resolution was 16 wall units, which corresponds to 5-6 Kolmogorov length scales. This is considered sufficiently resolved as to infer the nature of the vortical structures, which is still the subject of some debate. Compared to earlier 3D experiments this new data offers much improved spatial resolution within a relatively large flow domain and allows to follow the structures as they develop in time. Visualizations of vortical motions reveal quasi-streamwise vortices near the low speed streaks consistent with some of the proposed models for the near wall region. However, we also find clear evidence of hairpins in this region. Moreover, a new hairpin is observed to develop upstream of one of the pre-existing hairpins creating what may be considered a hairpin packet. This suggests auto-generation mechanisms to be present in the fully turbulent boundary layer. 3:33PM D26.00007 Shear-layer Effect on Wall Pressure Statistics in Turbulent Channel Flow , MASAYUKI SANO, TATSUYA TSUNEYOSHI, Nagoya University, YOSHINOBU YAMAMOTO, University of Yamanashi, YOSHIYUKI TSUJI, Nagoya University — The coherent structures are studied near wall region in turbulent channel flow by means of Direct Numerical Simulation (DNS). We analyze the correlation between coherent structures and wall pressure with positive and negative high amplitude peaks. DNS Reynolds numbers based on the friction velocity and the channel half-width are from 150 to 2000. The probability density functions of pressure indicate that there are different coherent structures associated with positive and negative pressure region. In order to evaluate the influence of the coherent structures on wall pressure, we visualize the conditioned streamwise velocity and second invariant of the deformation tensor conditioned by wall pressure. The second invariant of the deformation tensor is called Q criterion, which represents the intensity of vortices. It is found that the high pressure region is related to shear-layer in the buffer region and the negative pressure region is generated by small-scale vortices. Previous studies reported only the low-Reynolds number case, but the same results are confirmed in relatively high Reynolds number. The shear layer at Reτ =150 and 1000 have the same spanwise scales. The result shows that the spanwise meandering of the large scale structure is normalized by the inner scale. 3:46PM D26.00008 Large scale and small scale interactions of pressure fluctuation in turbulent boundary layers , YOSHIYUKI TSUJI, Nagoya University, YOSHINOBU YAMAMOTO, Yamanashi University — We study the interaction between large and small scale motions from the point of pressure fluctuation. Using the small pressure probe, both the static pressure and wall pressure fluctuations were measured inside the zero-pressure gradient boundary layer at relatively high Reynolds numbers. How the large scales in outside affect the small scales near wall is analyzed by means of statistical method. The correlation between pressure and pressure gradient indicates that the small scales distribute uniformly across the boundary layer. High amplitude positive and negative wall pressure fluctuations are also analyzed and found that they are associated with coherent motions inside the boundary layer. Another interesting aspect is the amplitude modulations of pressure and we would like to comment this topic in the presentation. 3:59PM D26.00009 Statistical comparison of coherent structures in fully developed turbulent pipe flow with and without drag reduction , FRANCESCA SOGARO, ROBERT POOLE, DAVID DENNIS, University of Liverpool — High-speed stereoscopic particle image velocimetry has been performed in fully developed turbulent pipe flow at moderate Reynolds numbers with and without a drag-reducing additive (an aqueous solution of high molecular weight polyacrylamide). Three-dimensional large and very large-scale motions (LSM and VLSM) are extracted from the flow fields by a detection algorithm and the characteristics for each case are statistically compared. The results show that the three-dimensional extent of VLSMs in drag reduced (DR) flow appears to increase significantly compared to their Newtonian counterparts. A statistical increase in azimuthal extent of DR VLSM is observed by means of two-point spatial autocorrelation of the streamwise velocity fluctuation in the radial-azimuthal plane. Furthermore, a remarkable increase in length of these structures is observed by three-dimensional two-point spatial autocorrelation. These results are accompanied by an analysis of the swirling strength in the flow field that shows a significant reduction in strength and number of the vortices for the DR flow. The findings suggest that the damping of the small scales due to polymer addition results in the undisturbed development of longer flow structures. 4:12PM D26.00010 Experimental investigation of turbulent channel flow over a compliant wall using tomographic PIV and Mach-Zehnder interferometry1 , CAO ZHANG, Johns Hopkins University, RINALDO MIORINI, General Electric Global Research Center, JOSEPH KATZ, Johns Hopkins University — The time-resolved 3D flow field and 2D distribution of wall-normal deformation in turbulent channel flow over a compliant surface are simultaneously measured by a combination of tomographic PIV (TPIV) and Mach-Zehnder Interferometry (MZI). The compliant wall is made of PDMS, and the friction Reynolds number is 2.3 × 103 . The mean velocity profile in the log layer is consistent with that of a channel flow over a smooth rigid wall. The flow resolution of the TPIV measurement is enhanced using single-pixel ensemble correlations to resolve the buffer layer. Extensive calibrations of the MZI system show a wall-normal resolution of deformation in the order of 10 nm. The power spectral density of the surface deformation indicates a wide range of the time-scales. The streamwise wavenumber-frequency spectrum displays two main features: (i) An inclined band corresponding to deformations advected with the flow at approximately 80% of the freestream speed, i.e. the velocity in the log layer. Their amplitudes are in the submicron range. (ii) Non advected, low frequency (<500 Hz) events that are larger than the field-of-view, and have much higher amplitudes, up to 100 µm. Ongoing analyses examine the deformation-velocity and deformation-pressure correlations to identify structures that influence the interactions with the compliant wall. 1 Sponsored by ONR. Sunday, November 23, 2014 2:15PM - 4:25PM Session D27 Wall-Bounded Turbulent Flows — 2009 - Xiaohua Wu, The Royal Military College of Canada 2:15PM D27.00001 Direct simulation of flat-plate boundary layer with mild free-stream turbulence , XIAOHUA WU, Royal Military College of Canada, PARVIZ MOIN, Center for Turbulence Research, Stanford University — Spatially evolving direct numerical simulation of the flat-plate boundary layer has been performed. The momentum thickness Reynolds number develops from 80 to 3000 with a free-stream turbulence intensity decaying from 3 percent to 0.8 percent. Predicted skin-friction is in agreement with the Blasius solution prior to breakdown, follows the well-known T3A bypass transition data during transition, and agrees with the Erm and Joubert Melbourne wind-tunnel data after the completion of transition. We introduce the concept of bypass transition in the narrow sense. Streaks, although present, do not appear to be dynamically important during the present bypass transition as they occur downstream of infant turbulent spots. For the turbulent boundary layer, viscous scaling collapses the rate of dissipation profiles in the logarithmic region at different Reynolds numbers. The ratio of Taylor microscale and the Kolmogorov length scale is nearly constant over a large portion of the outer layer. The ratio of large-eddy characteristic length and the boundary layer thickness scales very well with Reynolds number. The turbulent boundary layer is also statistically analyzed using frequency spectra, conditional-sampling, and two-point correlations. Near momentum thickness Reynolds number of 2900, three layers of coherent vortices are observed: the upper and lower layers are distinct hairpin forests of large and small sizes respectively; the middle layer consists of mostly fragmented hairpin elements. 2:28PM D27.00002 Taylor’s hypothesis in turbulent channel flow considered using a transport equation analysis1 , JAMES WALLACE, Univ. of Maryland, CHENHUI GENG, GUOWEI HE, Chinese Acad. of Sci., Inst. of Mech. (LNM), YINSHAN WANG, CHUNXIAO XU, Tsinghua Univ. — A DNS of turbulent channel flow was carried out to examine Taylor’s “frozen turbulence” hypothesis, i.e. the simple time-space transformation that allows (1/U )∂/∂t to approximate streamwise derivatives, ∂/∂x, of velocity fluctuations. These terms in Taylor’s hypothesis appear in the transport equation for instantaneous momentum for this flow. The additional terms, i.e. the additional convective acceleration and the pressure gradient and viscous force terms, act to diminish the validity of Taylor’s hypothesis when they are relatively large compared to the Taylor’s hypothesis terms and are not in balance. A similar analysis also has been applied to the transport equation for instantaneous vorticity. There the additional terms, namely the additional convective rates of change, the stretching/compression/rotation and the viscous diffusion of vorticity terms, similarly act to diminish the validity of Taylor’s hypothesis when they also are relatively large compared to the terms in the hypothesis and are not in balance. Where in the channel flow this diminishment occurs, and to what degree, and which of the non-Taylor’s hypothesis terms in the momentum and vorticity equations contribute most to this diminishment will be presented. 1 Supported by National Natural Sci. Found. and the National Basic Res. Progr. of China and the Burgers Progr. for Fluid Dynamics. 2:41PM D27.00003 Large-scale motions for a high-Reynolds-number turbulent pipe flow at Re τ = 30081 , JUNSUN AHN, KAIST, JAE HWA LEE, UNIST, JIN LEE, HYUNG JIN SUNG, KAIST — Direct numerical simulation (DNS) of turbulent pipe flow at Re τ = 3008 with a very long streamwise domain length (Lx = 30R, R is a pipe radius) was performed to explore the wall scaling laws in the overlap region. The high-Re turbulent pipe flow was found not to follow a log law, but rather to follow a power law. By contrast, the high-Re turbulent channel flow (Re τ ≥ 2006) followed a log law. A mesolayer was observed in both the pipe and channel flows, in agreement with the power law. The retarded log law in the turbulent pipe flow was attributed to the presence of large-scale structures in the outer region of the pipe flow. These large-scale structures were more dominant in the turbulent channel flows than in the turbulent pipe flows. As the Reynolds number increased, the fluids transitioned from a power law to a log law. The development of large-scale structures in the pipe flow was slower than the corresponding development in the channel flow The proportion of large-scale and very-large-scale motions (LSMs and VLSMs) was obtained in comparison with the low-Reynolds-number pipe flow at Re τ = 934. 1 This work was supported by the Creative Research Initiatives (No. 2014-001493) program of the National Research Foundation and was partially supported by KISTI under the Strategic Supercomputing Support Program. 2:54PM D27.00004 Convection of momentum transport events in a turbulent boundary layer , ROELAND DE KAT, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — Momentum transport in turbulent boundary layers increases drag. Understanding how momentum transport events interact and evolve will allow us to find ways to control them. In this study, we determine the convection of momentum transport events from time-resolved particle image velocimetry measurements in a stream-wise wall-normal plane of a turbulent boundary layer at Reτ ≈ 2700. A field-of-view covering approximately 2 × 0.5δ with high spatial, l+ = 20, and temporal resolution, ∆t+ = 0.7, allows us to determine convection velocities of momentum transport events of range of different sizes for wall-normal locations y/δ = 0.02 to 0.47 (y + = 60 to 1260). In the talk, a detailed description of convection of momentum transport events in different quadrants will be presented. 3:07PM D27.00005 Townsend’s similarity hypothesis applies to the intermittent region of a boundary layer1 , GUILLEM BORRELL, JAVIER JIMENEZ, Universidad Politecnica de Madrid — The intermittent region of two boundary layers + with different entrainment rates obtained by direct numerical simulation are compared at δ99 = 1500, one with the natural friction coefficient, and a second where the spreading rate is increased by 70% by a smooth volumetric force. The two flows are compared by thresholding the vorticity magnitude field, using a vorticity isosurface as a reference frame. Three regions can be observed in the conditional analysis. The two that are associated with the turbulent-nonturbulent interface match if u2τ /ν is used as the unit for vorticity, where uτ takes into account the additional friction caused by the forcing. The third one, where the two flows are not comparable, corresponds to the near-wall region where the force is applied. This result suggests that Townsend’s similarity hypothesis is also valid for the intermittent region of the boundary layer. 1 Funded by ERC, PRACE, CICYT and Spanish Ministry of Economy. 3:20PM D27.00006 Effects of external disturbances on turbulent boundary layers , EDA DOGAN, RONALD HANSON, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — The state of a turbulent boundary layer that develops under the influence of different types of freestream turbulence is examined. The freestream turbulence conditions with different length-scale and turbulence intensity are generated using active and passive grids. Downstream of the grid, a flat plate is placed to establish a zero-pressure gradient turbulent boundary layer. The interaction between the freestream and the turbulent boundary layer is investigated using simultaneous measurements of the boundary layer and freestream using single component hot-wire anemometry and multi-camera Particle Image Velocimetry (PIV). Results from the hot-wire measurements of different cases show that the near-wall peak turbulence intensity increases with increasing levels of free stream turbulence indicating the level and extent of penetration by free stream turbulence into the boundary layer. It is also observed that for different level of freestream perturbations to the flow, the momentum loss in the turbulent boundary layer could be similar. The data from these cases will be investigated further using spectral analysis to examine the energetic scales of the flow. The PIV data will be analysed to elucidate the coherent structures associated with these interactions. 3:33PM D27.00007 Reynolds shear stress near its maxima, turbulent bursting process and associated velocity profle in a turbulent boundary layer , NOOR AFZAL1 , Retired Professor, Aligarh Muslim University, Aligarh 201002, Aligarh — The Reynolds shear stress around maxima, turbulent bursting process and associate velocity profile in ZGP turbulent boundary layer is considered in the intermediate layer/mesolayer proposed by Afzal (1982 Ing. Arch 53, 355-277), in addition to inner and outer layers. The intermediate length −1/2 scale δm = δRτ having velocity Um = m Ue with 1/2 ≤ m ≤ 2/3 where Ue is velocity at boundary layer edge. Long & Chen (1981 JFM) intermediate −1/2 layer/ mesolayer scale δm = δRτ with velocity Um the friction velocity uτ , is untenable assumption (Afzal 1984 AIAA J). For channel/pipe flow, Sreenivasan et al (1981989, 1997, 2006a,b) proposed critical layer / mesolayer, cited/adopted work Long and Chen and McKeon, B.J. & Sharma, A. 2010 JFM 658, page 370 stated “retaining the assumption that the critical layer occurs when U (y) = (2/3) UCL (i.e. that the critical layer scales with y + ∼ Rτ +2/3 ),” both untenable assumptions, but ignored citation of papers Afzal 1982 onwards on pipe flow. The present turbulent boundary layer work shows that Reynolds shear maxima, shape factor and turbulent bursting time scale with mesolayer variables and Taylor length/time scale. 1 Residence, Embassy Hotel Rasal Gang Aligarh 202001 UP India 3:46PM D27.00008 Completion of partially known second-order statistics of turbulent flows , ARMIN ZARE, MIHAILO JOVANOVIC, TRYPHON GEORGIOU, University of Minnesota — Second-order statistics of turbulent flows can be obtained either experimentally or via direct numerical simulations. The statistics are relevant in understanding fundamentals of flow physics and for the development of low-complexity turbulence models. For example, such models can be used for control design in order to suppress or promote turbulence. Due to experimental or numerical limitations it is often the case that only partial flow statistics are reliably known. In other words, only certain correlations between a limited number of flow field components are available. Thus, it is of interest to complete the statistical signature of the flow field in a way that is consistent with the known dynamics. Our approach to this inverse problem relies on a model governed by stochastically forced linearized Navier-Stokes equations. In this, the statistics of forcing are unknown and sought to explain available velocity correlations. Identifying suitable stochastic forcing allows us to complete the correlation data of the velocity field. While the system dynamics impose a linear constraint on the admissible correlations, such an inverse problem admits many solutions. We use nuclear norm minimization to obtain correlation structures of low complexity. This complexity translates into dimensionality of filters that can be used to generate the identified forcing statistics. The ability of our approach to reproduce statistical features of a turbulent channel flow is demonstrated using stochastic simulations of the linearized dynamics. 3:59PM D27.00009 Nonlinear interactions and their scaling in the logarithmic region of turbulent channels1 , RASHAD MOARREF, Caltech, ATI S. SHARMA, University of Southampton, JOEL A. TROPP, BEVERLEY J. MCKEON, Caltech — The nonlinear interactions in wall turbulence redistribute the turbulent kinetic energy across different scales and different wall-normal locations. To better understand these interactions in the logarithmic region of turbulent channels, we decompose the velocity into a weighted sum of resolvent modes (McKeon & Sharma, J. Fluid Mech., 2010). The resolvent modes represent the linear amplification mechanisms in the Navier-Stokes equations (NSE) and the weights represent the scaling influence of the nonlinearity. An explicit equation for the unknown weights is obtained by projecting the NSE onto the known resolvent modes (McKeon et al., Phys. Fluids, 2013). The weights of triad modes -the modes that directly interact via the quadratic nonlinearity in the NSE- are coupled via interaction coefficients that depend solely on the resolvent modes. We use the hierarchies of self-similar modes in the logarithmic region (Moarref et al., J. Fluid Mech., 2013) to extend the notion of triad modes to triad hierarchies. It is shown that the interaction coefficients for the triad modes that belong to a triad hierarchy follow an exponential function. These scalings can be used to better understand the interaction of flow structures in the logarithmic region and develop analytical results therein. 1 The support of Air Force Office of Scientific Research under grants FA 9550-09-1-0701 (P.M. Rengasamy Ponnappan) and FA 9550-12-1-0469 (P.M. Doug Smith) is gratefully acknowledged. 4:12PM D27.00010 Time-evolving of very large-scale motions in a turbulent channel flow1 , JINYUL HWANG, JIN LEE, HYUNG JIN SUNG, KAIST, TAMER A. ZAKI, Johns Hopkins University — Direct numerical simulation (DNS) data of a turbulent channel flow at Re τ = 930 was scrutinized to investigate the formation of very large-scale motions (VLSMs) by merging of two large-scale motions (LSMs), aligned in the streamwise direction. We mainly focused on the supportive motions by the near-wall streaks during the merging of the outer LSMs. From visualization of the instantaneous flow fields, several low-speed streaks in the near-wall region were collected in the spanwise direction, when LSMs were concatenated in the outer region. The magnitude of the streamwise velocity fluctuations in the streaks was intensified during the spanwise merging of the near-wall streaks. Conditionally-averaged velocity fields around the merging of the outer LSMs showed that the intensified near-wall motions were induced by the outer LSMs and extended over the near-wall regions. The intense near-wall motions influence the formation of the outer low-speed regions as well as the reduction of the convection velocity of the downstream LSMs. The interaction between the near-wall and the outer motions is the essential origin of the different convection velocities of the upstream and downstream LSMs for the formation process of VLSMs by merging. 1 This work was supported by the Creative Research Initiatives (No. 2014-001493) program of the National Research Foundation of Korea (MSIP) and partially supported by KISTI under the Strategic Supercomputing Support Program. Sunday, November 23, 2014 2:15PM - 4:25PM Session D28 Turbulence: LES Modeling — 2011 - Scott Murman, NASA 2:15PM D28.00001 New subgrid-scale model for large-eddy simulation of turbulent flows , CHANGPING YU, XINLIANG LI, LHD, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, China, ZUOLI XIAO, SHIYI CHEN, State Key Laboratory for Turbulence and Complex Systems, College of Engineering, Peking University, Beijing 100871,China — Based on the theory of energy and helicity cascades, a new subgrid-scale (SGS) stress model is proposed for large eddy simulation of turbulent flows. The new SGS model is a type of eddy-viscosity model, and the eddy-viscosity is proportional to the magnitude of the mean product of the large-scale strain rate tensor and symmetric vorticity gradient tensor. The new SGS model is first tested a priori and a posteriori in homogeneous and isotropic helical turbulence, and the statistical results show that the new model can predict most of the results better than Smagorinsky model and the mixed helical model. Then, we apply the present model to simulate the channel flows, and also our model can support satisfied simulating results of mean velocity, turbulent stress and skin-friction coefficients, etc. The surprising findings is that the new model can describe much more realistic flow structures than DES-SA model and reproduce the skin friction force much more accurately than the Smagorinsky model. The new SGS model is proved to be universal model in large eddy simulation of turbulent flows. 2:28PM D28.00002 A dynamic sub-grid model for variational multiscale methods , SCOTT MURMAN, LASLO DIOSADY, ANIRBAN GARAI, NASA Ames Research Center — The variational multiscale method uses an explicit a priori separation of scales, along with Galerkin projection as the filter operator, to develop an alternative to the classical LES approach.1 A dynamic, parameter-free, multiscale extension of this approach is developed from the variational implementation of Germano’s identity.2 The method is implemented in an entropy-stable Discontinuous-Galerkin spectral-element solver.3 We outline the relevant details of the method using a priori testing, before demonstrating the performance in a posteriori testing on several canonical flows, including homogeneous isotropic turbulence, channel flow, and separated flow over an array of periodic hills. These computed results are compared against DNS, classical LES, and experimental data. 1 Hughes et al. Comput. Visual Sci. Sci. 3, 47 (2000) and Wanderer, J. of Turbulence 6, 7 (2005) 3 Diosady and Murman, AIAA 2014-2784 2 Oberai 2:41PM D28.00003 Autonomic Closure for Large Eddy Simulations , RYAN KING, University of Colorado, Boulder, WERNER DAHM, Arizona State University, PETER HAMLINGTON, University of Colorado, Boulder — Motivated by the application of adjoint techniques for rapidly solving large optimization problems, a fundamentally new autonomic closure is presented that allows an essentially model-free, dynamic subgrid-scale closure for large eddy simulations (LES). The autonomic closure addresses nonlinear, nonlocal, and nonequilibrium turbulence effects and, in its most general form, is based on all possible tensorally-invariant, dimensionally-consistent relations between the local subgrid-stress tensor and resolved scale primitive variables. This introduces a large matrix of spatially and temporally varying coefficients that are optimized using a test filter approach and then applied at the LES filter scale by invoking scale similarity. The autonomic closure avoids the need to specify a model for the subgrid stresses, and instead allows the simulation by itself to determine the best local relation between the subgrid stresses and resolved state variables. A priori tests of this new autonomic closure approach are presented using data from direct numerical simulations of homogeneous isotropic and sheared turbulence, and application of the closure to practical simulations is discussed. 2:54PM D28.00004 A dynamic subfilter-scale stress model for Large eddy simulations , AMIRREZA ROUHI, UGO PIOMELLI, Queen’s University, Canada, BERNARD GEURTS, University of Twente, Netherlands — In conventional large eddy simulation, the filter width is related to the grid size; this decouples the filter width from turbulence physics and results in unwanted dependence of the subfilter model on the grid arrangement. Relating the filter width to the integral length-scale is a potential solution. We proposed an approximation for the integral length-scale, in which a single model parameter was determined based on the global contribution of unresolved (subfilter) scales to the resolved ones denoted as subfilter activity (Piomelli & Geurts, Direct and Large-Eddy Simulation VIII, pp. 15-20, 2011). We have devolped a localized model in which we assign a target value to subfilter activity locally, requiring the model parameter to adapt itself to the local state of the flow. This dynamic modification is coupled with a local formulation for the integral scale. The modified model was applied on channel flow at Reτ up to 2, 000, accelerating boundary layer and backward-facing step flow at high Re with comparable accuracy as the Dynamic Smagorinsky model but with less computational expense. 3:07PM D28.00005 Development of subgrid-scale model using machine learning , MASATAKA GAMAHARA, Department of Applied Information Science, Tohoku University, YUJI HATTORI, Institute of Fluid Science, Tohoku University — Neural network, which automatically finds patterns or rules from big data, is applied to construct an improved sub-grid scale (SGS) model used in LES. SGS stress tensors are obtained by filtering data of direct numerical simulation (DNS) of turbulent channel flow. We use velocity gradient tensors and distance from the wall as inputs of the neural network aiming at improving conventional SGS models which include the Smagorinsky model. The back-propagation method is used in the learning process of the neural network. The results show that the neural network is able to learn SGS stress tensors. High correlation coefficients between SGS stress tensors obtained from DNS data and those estimated by the neural network are obtained. The results do not depend very much on the training data used for learning. Furthermore we investigate dependence on the size of training data, the filter size and the number of neurons. In particular the learning of neural network depends on the filter size. We also obtain high correlation coefficients at all Reynolds numbers tested. In order to find an explicit form of the estimated SGS stress tensors we try to identify the minimum set of independent variables by reducing the number of inputs. Physics behind the obtained model will be also discussed. 3:20PM D28.00006 Slip wall modeling approaches for separated flows and heat transfer , SANJEEB BOSE, Cascade Technologies, BRIAN PIERCE, PARVIZ MOIN, Center for Turbulence Research, Stanford Univ — Resolution of near-wall turbulent structures is computational prohibitive, necessitating the need for wall-modeled large-eddy simulation approaches. Standard wall models are often formulated to represent the wall stress assuming an equilibrium, attached boundary layer. This assumption is invalid in complex flows that include transition to turbulence or boundary layer separation. A dynamic slip wall boundary condition has been recently proposed (Bose & Moin, PoF, 2014) as an alternative for wall-modeled LES, where a slip wall boundary condition is derived from the differentially filtered LES governing equations with no assumption on the state of the local boundary layer. Results will be presented from the application of the dynamic slip wall model to flows with 3D separation (asymmetric stalled diffuser) and from the extension of the model to the prediction of wall heat transfer (turbine blade). The wall modeled LES predicts the primary quantity of interest in these flows: the pressure recovery in the diffuser and the heat transfer coefficient on the turbine blade. 3:33PM D28.00007 A Wall Model for Large-Eddy Simulation of Compressible Channel Flows , BARRETT MCCANN, ANTONINO FERRANTE, University of Washington — We have developed a new wall model for the large-eddy simulation (LES) of compressible channel flows with isothermal walls by extending the incompressible model of Chung and Pullin (J. Fluid Mech. 2009). The wall model computes the local, instantaneous wall shear stress and heat flux, which are then applied as wall boundary conditions, by solving two time-dependent, parameter-free ordinary differential equations (ODEs) at each time step. These ODEs are obtained by integrating the filtered momentum and energy equations in the wall-normal direction from the wall to the first grid point in the log layer. In contrast to so-called “wall resolved” LES, employment of this wall model allows use of relatively coarse computational meshes of fixed size, independent of Reynolds number. The wall model is first validated by comparing our LES results at M = 0.15 and Reτ = 2003 to the DNS results of Hoyas and Jiménez (Phys. Fluids 2006). We present LES results of channel flow simulations at M = 0.15 and M = 0.75, over a three-order of magnitude range of friction Reynolds numbers (Reτ = 2 × 103 , 2 × 104 , 2 × 105 and 2 × 106 ), on a mesh with 256 × 32 × 128 grid points in the streamwise, wall-normal, and spanwise directions. 3:46PM D28.00008 An integral wall model for Large Eddy Simulation (iWMLES) and applications to developing boundary layers over smooth and rough plates1 , XIANG YANG, JASIM SADIQUE, RAJAT MITTAL, CHARLES MENEVEAU, Johns Hopkins University — A new wall model for Large-Eddy-Simulations is proposed. It is based on an integral boundary layer method that assumes a functional form for the local mean velocity profile. The method, iWMLES, evaluates required unsteady and advective terms in the vertically integrated boundary layer equations analytically. The assumed profile contains a viscous or roughness sublayer, and a logarithmic layer with an additional linear term accounting for inertial and pressure gradient effects. The iWMLES method is tested in the context of a finite difference LES code. Test cases include developing turbulent boundary layers on a smooth flat plate at various Reynolds numbers, over flat plates with unresolved roughness, and a sample application to boundary layer flow over a plate that includes resolved roughness elements. The elements are truncated cones acting as idealized barnacle-like roughness elements that often occur in biofouling of marine surfaces. Comparisons with data show that iWMLES provides accurate predictions of near-wall velocity profiles in LES while, similarly to equilibrium wall models, its cost remains independent of Reynolds number and is thus significantly lower compared to standard zonal or hybrid wall models. 1 This work is funded by ONR grant N00014-12-1-0582 (Dr. R. Joslin, program manager). 3:59PM D28.00009 Implicit LES of Flow Over Wall-Mounted Hump , SUSHEEL SEKHAR, NAGI MANSOUR, NASA/Ames Res Ctr, SEKHAR/MANSOUR TEAM — Implicit LES of turbulent flow over wall-mounted hump is conducted to understand the physics of separated flows, and to provide data for RANS modeling and development. A modified version of the FDL3DI code1 that solves the compressible Navier-Stokes equations using high-order compact difference scheme and filter, and the standard recycling/rescaling method for generating a fully developed turbulent boundary layer at the inlet,2 is used. A mean velocity profile with Reθ = 1, 400 is imposed at the inlet. Qualitative assessment shows that the separation bubble is comparable in size with experiment. A detailed analysis, including comparisons of mean velocity profiles with experimental data before separation and after reattachment, is made. Quantitative comparisons of Reynolds stress profiles, as well as budgets of Reynolds stresses and turbulent kinetic energy are also presented. Physics of the flow post-reattachment is the focus of this study. Results from this effort will be used to further set up simulations at a higher Reynolds number (Reθ = 3, 500). 1 D.V. 2 B. Gaitonde & M.R. Visbal, AFRL-VA-WP TR-1998-3060 (1998) Morgan et al., AIAA J., 49 (3), 582-597 (2011) 4:12PM D28.00010 Influence of vortices on turbulence statistics , KOUJIRO ANAYAMA, KATSUNORI YOSHIMATSU, Nagoya University, YUKIO KANEDA, Aichi Institute Technology — We consider the importance or unimportance of the role of vortices at small scales in the determination of the turbulence statistics, on the basis of the method of the so-called “Computational Surgery.” Two fields, true and false fields, are generated. The true field obeys the Navier-Stokes (NS) equations for an incompressible fluid. In the false field, the NS dynamics are artificially modified so that the intense tube-like structures of the vortices are lost. Comparing the two fields, we may get some idea on the role of the vortices. The comparison so far made suggests that the statistics at larger scales are not sensitive to the exact vortex structure at small scales. Sunday, November 23, 2014 2:15PM - 4:25PM Session D29 Experimental Techniques: General Velocimetry — 2014 - Clara Velte, Danmarks Tekniske Uni- versitet 2:15PM D29.00001 2d-LCA - an alternative to x-wires , JAROSLAW PUCZYLOWSKI, MICHAEL HÖLLING, JOACHIM PEINKE, University of Oldenburg — The 2d-Laser Cantilever Anemometer (2d-LCA) is an innovative sensor for two-dimensional velocity measurements in fluids. It uses a micostructured cantilever made of silicon and SU-8 as a sensing element and is capable of performing mesurements with extremly high temporal resolutions up to 150kHz. The size of the cantilever defines its spatial resolution, which is in the order of 150 µm only. Another big feature is a large angular range of 180◦ in total. The 2d-LCA has been developed as an alternative measurement method to x-wires with the motivation to create a sensor that can operate in areas where the use of hot-wire anemometry is difficult. These areas include measurements in liquids and in near-wall or particle-laden flows. Unlike hot-wires, the resolution power of the 2d-LCA does not decrease with increasing flow velocity, making it particularly suitable for measurements in high speed flows. Comparative measurements with the 2d-LCA and hot-wires have been carried out in order to assess the performance of the new anemometer. The data of both measurement techniques were analyzed using the same stochastic methods including a spectral analysis as well as an inspection of increment statistics and structure functions. Furthermore, key parameters, such as mean values of both velocity components, angles of attack and the characteristic length scales were determined from both data sets. The analysis reveals a great agreement between both anemometers and thus confirms the new approach. 2:28PM D29.00002 Development of a nano-scale crossed hot wire to measure velocity in high Reynolds numbers , YUYANG FAN, MARCUS HULTMARK, Princeton University — In very high Reynolds number flows, accurate velocity measurements with conventional hot wires are often limited due to the size and response of the sensors. The Nano-Scale Thermal Anemometry Probes (NSTAPs), previously developed at Princeton, have been used successfully to acquire well-resolved velocity data in multiple high Re facilities. The NSTAP has displayed superior performance compared to conventional hot wires both in the spatial and temporal resolution. However, until now, NSTAPs have been limited to single component measurements. Here, a novel method to combine two inclined NSTAP probes, forming a nano-scale cross-wire, is presented. This enables simultaneous measurements of two fluctuating velocity components with unprecedented spatial and temporal resolution. The two sensing elements of the new x-NSTAP are about one order of magnitude shorter than the conventional cross-wire and are contained within a volume of about 50 × 50 × 50µm. The small sensing volume greatly improves the spatial resolution of high Re measurements. The small thermal mass of the new sensors also improves the frequency response to match that of a single component NSTAP. Measurements with the new x-NSTAP are performed in the Superpipe facility at Princeton and results are presented. 2:41PM D29.00003 An Evaluation of the Jorqenson Equation , JOHN FOSS, ATRA AKANDEH, Michigan State University, DOUGLAS NEAL, LaVision Inc. — Multi-sensor hot-wire probes require some processing algorithm to obtain components of the velocity vector at the measurement location. The Jorgenson equation (1) is used by numerous investigators for this purpose. There exist various algorithms to extract the velocity components from the recorded voltages. The present contribution is not to evaluate such algorithms; rather, it is to evaluate the agreement between the inferred (from (1)) and the known (measured) velocities for a range of pitch (angle α) and yaw (angle β) orientations of the probe body. That is, the ~ at (α, β) - each 9◦ to +/objective is to “give counsel” to those investigators who are considering the use of (1). Calibration data from Neal (2010) provide V ◦ ~ 36 . Since E(α, β) is to represent V (α, β), percentage error magnitudes will be presented. 2 2 n 2 2 E 2 = A + BVef f and Vef f = Un + KT UT + KB UB Jorqenson, F. E. (1971) “Directional Sensitivity of Wire and Fiber Film Probes, An Experimental Study,” DISA Information No. 11 Neal, D.R. (2010) “The Effects of Rotation on the Flow Field Over a Controlled-Diffusion Airfoil”, PhD Michigan State U. (1) 2:54PM D29.00004 Four-sensor Hot-Wire Probes: A Calibration and Data Reduction Strategy , DOUGLAS NEAL, LaVision Inc., JOHN FOSS, Michigan State University — Four-sensor hot-wire probes are capable of simultaneously measuring three components of the velocity vector with a high temporal resolution. Effective use of these probes requires sophisticated calibration and data reduction techniques and a number of different approaches have been published. Lavoie and Pollard (2003) evaluated four of these approaches and found them to vary significantly in terms of complexity, computational costs and accuracy of the results. Lavoie and Pollard showed the work of Wittmer (1998) is the least complicated to implement and has the smallest computational expense. The work of Doebbling (1990) has the best accuracy. A new technique for calibration and data reduction will be presented and compared against the methods of Wittmer (1998) and Doebbling (1990), using the same methodology and evaluation criteria. The results will be shown for a double x-array configuration over the calibration region of +/- 36◦ in pitch and yaw, but these methods are directly applicable to other four-sensor geometries. [1] Lavoie, P., Pollard, A. (2003). “Uncertainty analysis of four-sensor hot-wires and their data-reduction schemes used in the near field of a turbulent jet.” Exp Fluids, 34(3), 358-370. 3:07PM D29.00005 An Electromagnetic Gauge Technique for Measuring Shocked Particle Velocity in Electrically Conductive Samples , DAVID CHENG, AKIO YOSHINAKA, Defence Research and Development Canada Suffield — Electromagnetic velocity (EMV) gauges are a class of film gauges which permit the direct in-situ measurement of shocked material flow velocity. The active sensing element, typically a metallic foil, requires exposure to a known external magnetic field in order to produce motional electromotive force (emf). Due to signal distortion caused by mutual inductance between sample and EMV gauge, this technique is typically limited to shock waves in non-conductive materials. In conductive samples, motional emf generated in the EMV gauge has to be extracted from the measured signal which results from the combined effects of both motional emf and voltage changes from induced currents. An electromagnetic technique is presented which analytically models the dynamics of induced current between a copper disk moving as a rigid body with constant 1D translational velocity toward an EMV gauge, where both disk and gauge are exposed to a uniform external static magnetic field. The disk is modelled as a magnetic dipole loop where its Foucault current is evaluated from the characteristics of the fields, whereas the EMV gauge is modelled as a circuit loop immersed in the field of the magnetic dipole loop, the intensity of which is calculated as a function of space and, implicitly, time. Equations of mutual induction are derived and the current induced in the EMV gauge loop is solved, allowing discrimination of the motional emf. Numerical analysis is provided for the step response of the induced EMV gauge current with respect to the Foucault current in the moving copper sample. 3:20PM D29.00006 Statistical correction on LIFPA measurement1 , WEI ZHAO, University of South Carolina, FANG YANG, Carnegie Mellon University, GUIREN WANG, University of South Carolina — Laser Induced Fluorescence Photobleaching Anemometer (LIFPA) has been applied for velocity fluctuation measurement in micro electrokinetic turbulence in microfluidics. However, due to the intrinsic drawback of LIFPA, i.e. single-point and 1D measurement, LIFPA cannot distinguish velocity components on each directions and should rely on Taylor’s Hypothesis to get spatial series of velocity. Hence, the measurement will have error compared to the actual flow field. Here, the statistical errors of LIFPA measurement, due to 3D flows and Taylor’s Hypothesis, are theoretically estimated. We derived the correction formulas based on the work of Ewing and George (2000) and estimated the correction factor of LIFPA in the direction parallel to laser beam. The influences of directional correction factors on both LIFPA and single-wire Hot-Wire Anemometer (HWA) measurements are also investigated and compared. Later, first derivation variance (FDV) of velocity fluctuation by both Taylor’s Hypothesis and Local Taylor’s Hypothesis (Pinton and Labbe 1994) are compared in microfluidics. It is found the error due to Taylor’s Hypothesis is negligible. And the 3D flow influence on the FDV of velocity fluctuations in LIFPA is smaller than in HWA measurement. 1 The work was supported by NSF under grant no. CAREER CBET-0954977, MRI CBET-1040227, and NSF EPSCoR award, EPS-1317771 respectively. 3:33PM D29.00007 Application of Lorentz force techniques for flow rate measurement , RESCHAD JOHANN EBERT, CHRISTIAN RESAGK, Tech Univ Ilmenau — We report on the progress of the Lorentz force velocimetry (LFV): a contactless non-invasive flow velocity measurement technique. This method has been developed and demonstrated for various applications in our institute and in industry. At applications for weakly conducting fluids such as electrolytes with conductivities in the range of 1 – 10 S/m the challenging bottleneck is the detection of the tiny Lorentz forces in the noisy environment of the test channel. For the force measurement a state-of-the-art electromagnetic force compensation balance is used. Due to this device the mass of the Lorentz force generating magnets is limited. For enabling larger magnet systems and for higher force signals we have developed and tested a buoyancy based weight force compensation method which will be presented here. Additionally, results of LFV measurements at non-symmetric fluid profiles will be shown. By that an evaluation of the feasibility of this measurement principle for disturbed fluid profiles that are relevant for developing the LFV for weakly conducting fluids towards industrial applications can be made. Additionally a prospective setup for using the LFV for molten salt flows will be explained. 3:46PM D29.00008 Deconvolution as a means of correcting turbulence power spectra measured by LDA , PREBEN BUCHHAVE, Intarsia Optics, CLARA VELTE, Technical University of Denmark — Measurement of turbulence power spectra by means of laser Doppler anemometry (LDA) has proven to be a difficult task. Among the problems affecting the shape of the spectrum are noise in the signal and changes in the sample rate caused by unintentional effects in the measuring apparatus or even in the mathematical algorithms used to evaluate the spectrum. We analyze the effect of various causes of bias in the sample rate and show that the effect is a convolution of the true spectrum with various spectral functions. We show that these spectral functions can be measured with the available data from a standard LDA processor and we use this knowledge to correct the measured spectrum by deconvolution. We present results supported by realistic computer generated data using two different spectral estimators, the so-called slotted autocovariance method and the so-called direct method. 3:59PM D29.00009 Dead time effects in turbulence spectra measured by burst-mode LDA , CLARA VELTE, Technical University of Denmark, PREBEN BUCHHAVE, Intarsia Optics, WILLIAM GEORGE, Imperial College London — Dead time effects in laser Doppler measurements have not so far been considered a major problem. We show how dead time occurs in burst-mode laser Doppler anemometry (LDA) when using a so-called burst-mode LDA processor and describe their effects on the measured power spectra. We show how dead time effects may be caused by more than one seed particle being present in the measurement volume at the same time and explain analytically how dead time causes a reduction in the power in the spectrum at low frequencies and an oscillation in the spectrum at the high frequency end. We also present a realistic model for the data sampled from a processor with dead time and use this model to generate turbulence velocity data in a computer. Finally we compare the spectrum computed from realistic values of dead time and sample rate in the computer generated data and compare this spectrum to a measured spectrum in a free turbulent jet with similar parameters. The excellent agreement between the features of these spectra show that our model and explanation of the dead time effect is a valid one. 4:12PM D29.00010 Boundary Wall Shear Measurement with an Automated LDV-Based System , DARIUS MODARRESS, Measurement Science Enterprise, DAVID JEON, California Institute of Technology, PAVEL SVITEK, Measurement Science Enterprise, MORTEZA GHARIB, California Institute of Technology — Wall shear stress is one of the most important measurements in boundary layer flows. Getting wall shear measurements is generally quite difficult due to the need to measure very close to the wall, where poor optical access, particle seeding, and wall effects can bias the results. To simplify that process, a novel system was developed by Measurement Science Enterprise (MSE). The microPro consists of a 12 mm diameter miniLDV attached to a micro-translation stage assembled inside a sealed housing. The microPro automatically locates the wall and measures the mean flow speed profile from a point as close as 50 microns from the window. Accurate estimate of the mean wall shear is obtained from the calculation of the wall velocity gradient obtained from the velocity profile data. We measured wall shear stress on a boundary layer plate mounted in a water tunnel across a range of Reynolds numbers and compared the results against skin friction coefficient models. We also introduced bubbles into the boundary layer to measure the change in wall shear stress with changing void fraction. The measurements show good agreement with established data. This work is supported by the Office of Naval Research (grant ONR- N00014-11-1-0031) and MSE. Sunday, November 23, 2014 2:15PM - 4:25PM — Session D30 Wind Turbines: Atmospheric Forcing and Turbine Models 2016 - Tina Chow, University of California, Berkeley 2:15PM D30.00001 Scanning of wind turbine upwind conditions: numerical algorithm and first applications , MARC CALAF, GERARD CORTINA, Mechanical Engineering, University of Utah, VARUN SHARMA, MARC B. PARLANGE, School of Architecture, Civil and Environmental Engineering, EPFL — Wind turbines still obtain in-situ meteorological information by means of traditional wind vane and cup anemometers installed at the turbine’s nacelle, right behind the blades. This has two important drawbacks: 1-turbine misalignment with the mean wind direction is common and energy losses are experienced; 2-the near-blade monitoring does not provide any time to readjust the profile of the wind turbine to incoming turbulence gusts. A solution is to install wind Lidar devices on the turbine’s nacelle. This technique is currently under development as an alternative to traditional in-situ wind anemometry because it can measure the wind vector at substantial distances upwind. However, at what upwind distance should they interrogate the atmosphere? A new flexible wind turbine algorithm for large eddy simulations of wind farms that allows answering this question, will be presented. The new wind turbine algorithm timely corrects the turbines’ yaw misalignment with the changing wind. The upwind scanning flexibility of the algorithm also allows to track the wind vector and turbulent kinetic energy as they approach the wind turbine’s rotor blades. Results will illustrate the spatiotemporal evolution of the wind vector and the turbulent kinetic energy as the incoming flow approaches the wind turbine under different atmospheric stability conditions. Results will also show that the available atmospheric wind power is larger during daytime periods at the cost of an increased variance. 2:28PM D30.00002 Influence of atmospheric stability on model wind turbine wake interface , AMELIA TAYLOR, VIRGILIO GOMEZ, SANTIAGO NOVOA, SUHAS POL, CARSTEN WESTERGAARD, LUCIANO CASTILLO, Texas Tech Univ — Differences in wind turbine wake deficit recovery for various atmospheric stability conditions (stratification) have been attributed to turbulence intensity levels at different conditions. It is shown that buoyancy differences at the wind turbine wake interface should be considered in addition to varying turbulence intensity to describe the net momentum transport across the wake interface. Mixing, induced by tip and hub vortices or wake swirl, induces these buoyancy differences. The above hypothesis was tested using field measurements of the wake interface for a 1.17 m model turbine installed at 6.25 m hub height. Atmospheric conditions were characterized using a 10 m meteorological tower upstream of the turbine, while a vertical rake of sonic anemometers clustered around the hub height on a downstream tower measured the wake. Data was collected over the course of seven months, during varying stability conditions, and with five different turbine configurations – including a single turbine at three different positions, two turbines in a column, and three turbines in a column. Presented are results showing the behavior of the wake (particularly the wake interface), for unstable, stable, and neutral conditions. We observed that the swirl in the wake causes mixing of the inflow, leading to a constant density profile in the far wake that causes density jumps at the wake interfaces for stratified inflow. 2:41PM D30.00003 Atmospheric Stability Effects on the Structure and Evolution of WindTurbine Wakes , MAHDI ABKAR, FERNANDO PORTÉ-AGEL, Wind Engineering and Renewable Energy Laboratory (WIRE), École polytechnique fédérale de Lausanne (EPFL) — Large-eddy simulation is combined with a turbine model to investigate the effect of atmospheric thermal stability on the structure and evolution of wind-turbine wakes. The simulation results show that atmospheric stability significantly affects the spatial distribution of the mean velocity deficit and turbulence statistics in the wake region. In particular, the enhanced turbulence level associated with positive buoyancy under the convective condition leads to a larger flow entrainment and, thus, a faster wake recovery. It is also shown that atmospheric stability has a significant effect on the wake meandering characteristics downwind of the turbine. In particular, for a given distance downwind of the turbine, wake meandering is larger under the convective condition compared with the neutral and stable ones. In addition, for all the stability cases considered in this study, the vertical wake meandering is lower compared with the lateral one. This is mainly related to the different turbulence levels of the incoming wind in different directions together with the anisotropy imposed by the presence of the ground. The results also confirm that the turbulence levels in all three directions must be considered to describe the impact of atmospheric stability on wind-turbine wakes. 2:54PM D30.00004 Turbulence structures in wind turbine wake: Effects of atmospheric stratification1 , KIRAN BHAGANAGAR, University of Texas San Antonio — Turbulence structure in the wake behind full-scale horizontal-axis WT under the influence of realistic atmospheric turbulent flow conditions has been investigated using actuator-line-model based large-eddy-simulations. Wind turbine simulations have revealed that, in addition to wind shear and ABL turbulence, height-varying wind angle and low-level jets are ABL metrics that influence the structure of turbine wake. Turbulent mixing layer forms downstream of the WT, the strength and size of which decreases with increasing stability. Height dependent wind angle and turbulence are the ABL metrics influencing the lateral wake expansion. Further, ABL metrics strongly impact the evolution of tip and root vortices formed behind the rotor. Two factors play an important role in wake meandering: tip vortex merging due to the mutual inductance form of instability and the corresponding instability of the turbulent mixing layer. 1 NSF CBET Energy for Sustainability 3:07PM D30.00005 Influence of realistic atmospheric forcings on wind turbine wake interactions1 , MITHU DEBNATH, KIRAN BHAGANAGAR, University of Texas, San Antonio — Atmospheric boundary layer structure is dictated by the stratification of the atmosphere; hence stratifications effects are critical in accurate representation of wind turbine wake physics. Large eddy simulation (LES) has been used to resolve atmospheric boundary layer turbulence and the wind turbine (WT) wake turbulence structures. The effect of atmospheric stratification on the evolution of tip and root vortices has been analyzed. For the first time, mutual induction mode of vortex instability leading to vortex merging in the wind turbine wake has been demonstrated under realistic ABL conditions. Vortex merging leads to enhanced Reynolds stresses and increased mixing. Finally, the effect of the turbulent mixing due to the shear layer on power production is analyzed 1 National Science Foundation, CBET Energy for Sustainability 3:20PM D30.00006 Large eddy simulation of unsteady wind farm behavior using advanced actuator disk models1 , MAUD MOENS, MATTHIEU DUPONCHEEL, GREGOIRE WINCKELMANS, PHILIPPE CHATELAIN, Univ Catholique de Louvain — The present project aims at improving the level of fidelity of unsteady wind farm scale simulations through an effort on the representation and the modeling of the rotors. The chosen tool for the simulations is a Fourth Order Finite Difference code, developed at Universite catholique de Louvain; this solver implements Large Eddy Simulation (LES) approaches. The wind turbines are modeled as advanced actuator disks : these disks are coupled with the Blade Element Momentum method (BEM method) and also take into account the turbine dynamics and controller. A special effort is made here to reproduce the specific wake behaviors. Wake decay and expansion are indeed initially governed by vortex instabilities. This is an information that cannot be obtained from the BEM calculations. We thus aim at achieving this by matching the large scales of the actuator disk flow to high fidelity wake simulations produced using a Vortex Particle-Mesh method. It is obtained by adding a controlled excitation at the disk. We apply this tool to the investigation of atmospheric turbulence effects on the power production and on the wake behavior at a wind farm level. A turbulent velocity field is then used as inflow boundary condition for the simulations. 1 We gratefully acknowledge the support of GDF Suez for the fellowship of Mrs Maud Moens 3:33PM D30.00007 Implementation of wind turbine parameterizations in a mesoscale-LES nested model framework , FOTINI CHOW, NIKOLA MARJANOVIC, University of California, Berkeley, JEFFREY MIROCHA, Lawrence Livermore National Laboratory — Wind turbine performance depends on weather conditions, local topography, and wind turbine spacing, among other factors. Atmospheric simulations can be used to predict wind energy production at increasingly higher resolutions. Turbine models placed within such simulations can be used to investigate turbine operation and performance. This work describes the implementation of generalized actuator disk (GAD) and line (GAL) models into the Weather Research and Forecasting (WRF) mesoscale atmospheric model. WRF can be used in a grid nested configuration starting from the mesoscale (∼10 km resolution) and ending with fine scale resolutions (∼1-10 m) suitable for large-eddy simulations (LES). At LES scales it becomes possible to resolve both the thrust and torque forces generated on turbines and imparted to the atmosphere using GAD and GAL models. Both models include real-time yaw and pitch control to respond to changing flow conditions. Here, the GAD and GAL are tested for idealized and real model configurations and compared to data from a wind farm. Comparisons are also made that help determine the importance of turbine blade tilt away from the tower and the inclusion of tower and turbine hub drag effects. 3:46PM D30.00008 A study of two subgrid-scale models and their effects on wake breakdown behind a wind turbine in uniform inflow1 , LUIS MARTINEZ, CHARLES MENEVEAU, Johns Hopkins University — Large Eddy Simulations (LES) of the flow past a single wind turbine with uniform inflow have been performed. A goal of the simulations is to compare two turbulence subgrid-scale models and their effects in predicting the initial breakdown, transition and evolution of the wake behind the turbine. Prior works have often observed negligible sensitivities to subgrid-scale models. The flow is modeled using an in-house LES with pseudo-spectral discretization in horizontal planes and centered finite differencing in the vertical direction. Turbines are represented using the actuator line model. We compare the standard constant-coefficient Smagorinsky subgrid-scale model with the Lagrangian Scale Dependent Dynamic model (LSDM). The LSDM model predicts faster transition to turbulence in the wake, whereas the standard Smagorinsky model predicts significantly delayed transition. The specified Smagorinsky coefficient is larger than the dynamic one on average, increasing diffusion thus delaying transition. A second goal is to compare the resulting near-blade properties such as local aerodynamic forces from the LES with Blade Element Momentum Theory. Results will also be compared with those of the SOWFA package, the wind energy CFD framework from NREL. 1 This work is supported by NSF (IGERT and IIA-1243482) and computations use XSEDE resources, and has benefitted from interactions with Dr. M. Churchfield of NREL. 3:59PM D30.00009 Large Eddy Simulation of wind turbines using the actuator line model and immersed boundary method1 , CHRISTIAN SANTONI, KENNETH CARRASQUILLO-SOLÍS, STEFANO LEONARDI, Univ of Texas, Dallas — Despite the growth of the energy extracted from wind turbines, the flow physics is still not fully understood even under ideal operational conditions. Large Eddy Simulations of the turbulent flow past a wind turbine in a channel have been performed. The numerical setup reproduces the experiment performed in a wind tunnel at the Norwegian University of Science and Technology (NUST). The code is based on a finite difference scheme with a fractional step and Runge-Kutta, which couples the actuator line model (ALM) and the Immersed Boundary Method (IBM). Two simulations were performed, one neglecting the tower and nacelle resulting in the rotating blades only, the other modeling both the rotating blades as well as the tower and nacelle with IBM. Results relative to the simulation with tower and nacelle have a very good agreement with experiments. Profiles of turbulent kinetic energy shows that the effect of the tower and nacelle is not confined to the hub region but extend to the entire rotor. In addition we placed the wind turbine over an undulated topography to understand how it affects the performances and wake of a wind turbine. Comparison with the results obtained for the smooth wall show an interaction between the rough wall and the wake. 1 The numerical simulations were performed on XSEDE TACC under Grant No. CTS070066. The present work is supported by the National Science Foundation (NSF), grant IIA-1243482 (WINDINSPIRE). 4:12PM D30.00010 Comparison of Differences Between Model Wind Turbine Array and Porous Disk Array Boundary Layer Measurements , VASANT VUPPULURI, ELIZABETH CAMP, RAÚL BAYOÁN CAL, Portland State Univ — Wind turbines are often represented in computational studies as actuator disks, also known as porous disks. A wind tunnel study is performed on a 4 × 3 model wind turbine array and equivalent porous disk array using Stereo Particle Image Velocimetry (SPIV) in order to compare the resulting wakes. Measurements are taken both upstream and downstream of the center turbine in the fourth row with SPIV planes both parallel and perpendicular to the rotor disk. The resulting flow fields are used to quantify the cumulative effects of the differences between the rotor and porous disk wakes. Sunday, November 23, 2014 2:15PM - 4:25PM Session D31 CFD: Immersed Boundary Methods — 2018 - Aleksander Donev, New York University 2:15PM D31.00001 Imposing scalar fluxes with the immersed boundary method in WRF , JINGYI BAO, University of California, Berkeley, KATHERINE LUNDQUIST, Lawrence Livermore National Laboratory, FOTINI CHOW, University of California, Berkeley — The Weather Research and Forecasting model (WRF) is being used over increasingly complex terrain at higher grid resolutions. However, when it comes to the situation of complex terrain, resolved terrain slopes can become large, causing numerical errors from grid stretching of the terrain-following coordinates. An immersed boundary method (IBM), a non-conforming grid technique was recently implemented into WRF (Lundquist et al. 2010, 2012), to alleviate numerical errors associated with the extreme distortion of the grid cells. The IBM uses a Cartesian grid with the terrain boundary “immersed” within the grid. The force of the boundaries on the fluid is represented with the addition of a body force term in the momentum equation. In this work, we extend the existing WRF-IBM model to represent the Neumann boundary condition at the surface for potential temperature and moisture in order to simulate realistic atmospheric flows. The Neumann boundary condition for potential temperature and moisture is designed to accommodate different closures, such as the most common Smagorinsky closure. Validation test cases include thermally induced slope flows in an idealized valley with both coupled and uncoupled heat fluxes. In the uncoupled cases, the surface heating is specified as a function of time, and there are no surface or land attributes such as vegetation or soil type. In the coupled cases, the surface fluxes are prescribed by atmospheric parameterizations, which have been modified to recognize the immersed boundary as the terrain surface. These test cases will provide a proof of concept and verify the implementation of the new temperature and moisture boundary conditions. 2:28PM D31.00002 Immersed boundary methods for particles in viscoelastic drilling muds , SREENATH KRISHNAN, ERIC SHAQFEH, GIANLUCA IACCARINO, Stanford Univ — In fracture stimulation of oil and gas wells, polymeric solution with suspended solids (proppants) are pumped to prop open the fracture. The primary aim of our work is to understand the dynamics of such proppants under various flow conditions through numerical computations. The study is concerned with fully resolved simulations, wherein all scales associated with the particle motion and the flow are resolved. The present effort is based on the algorithm proposed by Patankar (CTR Annual Research Briefs 2001:185-196), i.e. the Immersed Boundary (IB) methods, in which the domain grids do not conform to particle geometry and for simplicity are chosen to be Cartesian. Since Cartesian grids cannot efficiently represent a fracture geometry, our focus is on the development of an IB method for viscoelastic flows in unstructured domain grids. This method is implemented in a massively parallel, unstructured finite-volume-based fluid solver developed at Stanford University. The main theme of the presentation will be the description of the algorithm, measures taken to enable efficient parallelization and transfer of information between the underlying fluid grid and the particle mesh. A number of flow simulations will be presented, which validates the accuracy and correctness of the algorithm. 2:41PM D31.00003 Characteristic-based Volume Penalization Method for Arbitrary Mach Flows Around Solid Obstacles1 , NURLYBEK KASIMOV, ERIC BROWN-DYMKOSKI, OLEG VASILYEV, Univ of Colorado - Boulder — A new volume penalization method to enforce immersed boundary conditions in Navier-Stokes and Euler equations is presented. Previously, Brinkman penalization has been used to introduce solid obstacles modeled as porous media. This approach is limited to Dirichlet-type conditions on velocity and temperature, and in inviscid supersonic flows led to wrong shock reflection. It builds upon Brinkman penalization by allowing Neumann and Robin conditions to be applied in a general fashion. Correct boundary conditions are achieved through characteristic propagation into the thin layer inside of the obstacle. Inward pointing characteristics ensure that nonphysical solution inside the obstacle does not propagate out to the fluid region. Dirichlet boundary conditions are enforced similarly to Brinkman method. Penalization parameters are chosen so they act on a much faster timescale than the characteristic timescale of the flow. Main advantage of this method is the systematic means of controlling the error. This approach is general and applicable to a wide variety of flow regimes. The extensions of the methodology to moving obstacles and three dimensional flows are discussed. 1 This work was supported by ONR MURI under grant N00014-11-1-069 2:54PM D31.00004 Effect of gravity on the finite-size particle within spectral resolution , YONGNAM PARK, CHANGHOON LEE, Yonsei University — This study aims at the finest simulation of settling particles by using the immersed boundary method and direct numerical simulation with pseudo-spectral scheme. In many particle-laden simulations with the point-particle approach, due to the heavy particle approximation only Stokes drag force and gravity force are taken into account. On the other hand, most published works using an immersed boundary method considered particles as large as the Talyor micro scale. However, the Stokes number of particle of the Taylor micro scale size is over hundreds such that particles are not directly affected by turbulent flows. Due to the large Stokes number of finite-size particles, the density ratio of finite-size particles is limited to small range. In this study, the size of particles is comparable with the Kolmogorov length scale, and density ratio is larger than 10 in order to compare the results by the point particle simulations. In this simulation range, settling particles attenuate the turbulence because the particles can easily penetrate the vortex core and disturb the evolution of turbulence. Detailed statistics of particle motion will be discussed in the presentation. 3:07PM D31.00005 A velocity-correction projection method based immersed boundary method for incompressible flows1 , SHANGGUI CAI, University of Technology of Compiegne — In the present work we propose a novel direct forcing immersed boundary method based on the velocity-correction projection method of [J.L. Guermond, J. Shen, Velocity-correction projection methods for incompressible flows, SIAM J. Numer. Anal., 41 (1)(2003) 112-134]. The principal idea of immersed boundary method is to correct the velocity in the vicinity of the immersed object by using an artificial force to mimic the presence of the physical boundaries. Therefore, velocity-correction projection method is preferred to its pressure-correction counterpart in the present work. Since the velocity-correct projection method is considered as a dual class of pressure-correction method, the proposed method here can also be interpreted in the way that first the pressure is predicted by treating the viscous term explicitly without the consideration of the immersed boundary, and the solenoidal velocity is used to determine the volume force on the Lagrangian points, then the non-slip boundary condition is enforced by correcting the velocity with the implicit viscous term. To demonstrate the efficiency and accuracy of the proposed method, several numerical simulations are performed and compared with the results in the literature. 1 China Scholarship Council 3:20PM D31.00006 Implementation of Immersed Boundary Method in WENO Scheme to Simulate Blast-Structure Interaction1 , MIN XU, TAO YANG, MINGJUN WEI, New Mexico State University — High-order WENO schemes have been widely used in numerical simulation of shock/blast waves; and immersed boundary method has been gradually accepted as a simple and powerful approach to deal with moving boundaries in computational fluid dynamics. The combination of these two techniques becomes a natural choice in our study of blast-structure interaction. To benchmark our combined approach, we applied it first on classical shockwave problems with exact solutions or well-tested numerical solutions. Then, the algorithm is applied to simulate the interaction between an incoming blast wave and a spring-linked cylinder. Finally, a more complex case, where multiple plates linked by springs are interacting with blast waves and each other, has been investigated. 1 Supported by ARL (AHPCRC) 3:33PM D31.00007 Improvements to Level Set, Immersed Boundary methods for Interface Tracking1 , CHRIS VOGL, RANDY LEVEQUE, University of Washington — It is not uncommon to find oneself solving a moving boundary problem under flow in the context of some application. Of particular interest is when the moving boundary exerts a curvature-dependent force on the liquid. Such a force arises when observing a boundary that is resistant to bending or has surface tension. Numerically speaking, stable numerical computation of the curvature can be difficult as it is often described in terms of high-order derivatives of either marker particle positions or of a level set function. To address this issue, the level set method is modified to track not only the position of the boundary, but the curvature as well. The definition of the signed-distance function that is used to modify the level set method is also used to develop an interpolation-free, closest-point method. These improvements are used to simulate a bending-resistant, inextensible boundary under shear flow to highlight area and volume conservation, as well as stable curvature calculation. 1 Funded by a NSF MSPRF grant 3:46PM D31.00008 A Fluctuating Immersed Boundary Method for Brownian Suspensions of Rigid Particles1 , ALEKSANDAR DONEV, Courant Institute of Mathematical Sciences, New York University — I will describe how to model Brownian suspensions of passive or active particles and rigid bodies using an immersed boundary (IB) approach. I will first discuss minimally-resolved models in which each suspended spherical particle is represented by a single IB marker [F. Balboa Usabiaga and R. Delgado-Buscalioni and B. E. Griffith and A. Donev, Computer Methods in Applied Mechanics and Engineering, 269:139-172, 2014; and S. Delong, F. Balboa Usabiaga, R. Delgado-Buscalioni, B. E. Griffith and A. Donev, J. Chem. Phys., 140, 134110, 2014]. More complex rigid bodies suspensed in fluid can be represented with different degrees of fidelity by enforcing a rigidity constraint for each partially- or fully-resolved body [B. Kallemov, A. Bhalla, A. Donev, and B. Griffith, in preparation]. Thermal fluctuations and thus Brownian motion can be consistently modeled by including a fluctuating (random) stress in the momentum equation, as dictated by fluctuating hydrodynamics. 1 Joint work with multiple collaborators, with special thanks to Boyce Griffith, UNC. 3:59PM D31.00009 The Immersed Interface Method with Triangular Mesh Representation of an Interface1 , SHENG XU, Southern Methodist University — The immersed interface method can be employed to solve an interface problem on a fixed Cartesian grid by incorporating necessary interface-induced Cartesian jump conditions into numerical schemes. In this talk, we present ideas to compute the necessary Cartesian jump conditions from given principal jump conditions using triangular mesh representation of an interface. The triangular mesh representation is simpler and more robust than interface parametrization for a complex or non-smooth interface. We test our ideas by using the computed Cartesian jump conditions in the immersed interface method to solve a Poisson equation subject to an interface with the shape of a sphere, cube, cylinder or cone. Our results demonstrate expected second-order accuracy in the infinity norm. 1 This work is supported by the NSF grant DMS 1320317. 4:12PM D31.00010 Adaptive wavelet-based framework for aeroelastic simulations , RAJ NAIR, OLEG VASILYEV, Univ of Colorado - Boulder — This study presents the novel adaptive wavelet-based framework for modeling fluid-structure interaction. The approach uses the adaptive wavelet collocation method to solve the linear-elastic structural deformation equations inside the solid obstacle and compressible Navier-Stokes equations in the outer fluid region. The method then combines two mathematical approaches: volume penalization for creating a fluid-structure coupling by specifying traction condition on the solid boundary and enforcing the no-slip velocity conditions consistent with the rate of structural deformation on the obstacle boundary and a level-set-method, which dynamically tracks the solid-fluid interface. The method is applied to a two-dimensional aeroelastic flow and preliminary results are discussed. This work serves as the basis for continuing development of a robust adaptive wavelet based fluid-structure interaction model to accurately model the effects of unsteady aerodynamic loads in aeroelastic problems. Sunday, November 23, 2014 2:15PM - 4:25PM Session D32 Particle-Laden Flows: Radiation and Gravity — 2020 - Sivaramakrishnan Balachandar, University of Florida 2:15PM D32.00001 Towards understanding of particle-based solar receivers: impact of preferential concentration on heat transfer statistics , HADI POURANSARI, ALI MANI, Stanford University — This work aims to develop characterization of heating non-uniformities when a particle-laden fluid is heated via a radiative source. We consider a numerical setting with an inflow-outflow configuration in which a premixed turbulent stream is subject to uniform radiation intensity in the heating section. Direct numerical simulation of fluid-particle mixture is developed using finite difference approximation to the low-Mach Navier-Stokes equations and Lagrangian tracking to represent particle transport. The medium is considered optically thin with all radiation absorbed primarily by solid particles and then exchanged conductively to a gaseous carrier phase. In such setting preferential concentration of the particles leads to heating non-uniformities, which can impact system performance. We will present a characterization of mean versus rms temperature fields for a wide range of Stokes numbers. 2:28PM D32.00002 Settling of hot particles through turbulence1 , FILIPPO COLETTI, University of Minnesota, ARI FRANKEL, HADI POURANSARI, ALI MANI, Stanford University — Particle-laden flows in which the dispersed phase is not isothermal with the continuous phase are common in a wealth of natural and industrial setting. In this study we consider the case of inertial particles heated by thermal radiation while settling through a turbulent transparent gas. Particles much smaller than the minimum flow scales are considered. The particle Stokes number (based on the Kolmogorov time scale) and the nominal settling velocity (normalized by the root-mean-square fluid velocity fluctuation) are both of order unity. In the considered dilute and optically thin regime, each particle receives the same heat flux. Numerical simulations are performed in which the two-way coupling between dispersed and continuous phase is taken into account. The momentum and energy equations are solved in a triply periodic domain, resolving all spatial and temporal scales. While falling, the heated particles shed plumes of buoyant gas, modifying the turbulence structure and enhancing velocity fluctuations in the vertical direction. The radiative forcing does not affect preferential concentration (clustering of particles in low vorticity regions), but reduces preferential sweeping (particle sampling regions of downward fluid motion). Overall, the mean settling velocity varies slightly when heating the particles, while its variance is greatly increased. 1 We gratefully acknowledges support from DOE PSAAP II program. 2:41PM D32.00003 Turbulence-radiation interactions in a particle-laden flow1 , ARI FRANKEL, HADI POURANSARI, GIANLUCA IACCARINO, ALI MANI, Stanford University — Turbulent fluctuations in a radiatively participating medium can significantly alter the mean heat transfer characteristics in a manner that current RANS models cannot accurately capture. While turbulence-radiation interaction has been studied extensively in traditional combustion systems, such interactions have not yet been studied in the context of particle-laden flows. This work is motivated by applications in particle-based solar receivers in which external radiation is primarily absorbed by a dispersed phase and conductively exchanged with the carrier fluid. Direct numerical simulations of turbulence with Lagrangian particles subject to a collimated radiation source are performed with a flux-limited diffusion approximation to radiative transfer. The dependence of the turbulence-radiation interaction statistics on the particle Stokes number will be demonstrated. 1 Supported by PSAAP II 2:54PM D32.00004 Influence of thermal radiation on turbulent kinetic energy spectrum in particle-laden flows , JACQUELINE CHEN, HEMANTH KOLLA, Sandia National Laboratories, Livermore, CA, HADI POURANSARI, ALI MANI, Stanford University — We investigate density-weighted spectra of turbulent kinetic energy of the gas phase in particle-laden flows with thermal radiation. Compressible DNS of three-dimensional homogeneous isotropic decaying turbulence laden with point particles reveal that thermal radiation alters the spectrum of the total turbulent kinetic energy by introducing dilatational velocity fluctuations at large wavenumbers, while the spectrum of the divergence-free modes remains unaffected. With increasing time the magnitude of the energy content in the dilatational modes increases while the wavenumber range over which they are active also broadens. Pressure-dilatation correlations, which are source-like terms in the balance equation for density-weighted energy spectrum, confirm the high-wavenumber influence of radiation-induced dilatation. 3:07PM D32.00005 Spreading of non-planar non-axisymmetric gravity and turbidity currents1 , NADIM ZGHEIB, Univ of Florida - Gainesville, THOMAS BONOMETTI, Institut de Mécanique des Fluides de Toulouse, S. BALACHANDAR, Univ of Florida - Gainesville — The dynamics of non-axisymmetric turbidity currents is considered here. The study comprises a series of experiments for which a finite volume of particle-laden solution is released into fresh water. A mixture of water and polystyrene particles of diameter 280 < Dp < 315µm and density ρc = 1007Kg/m3 is initially confined in a hollow cylinder at the center of a large tank filled with fresh water. Cylinders with four different cross-sections are examined: a circle, a plus-shape, a rectangle and a rounded rectangle in which the sharp corners are smoothened. The time evolution of the front is recorded as well the spatial distribution of the thickness of the final deposit via the use of a laser triangulation technique. The dynamics of the front and final deposit are significantly influenced by the initial geometry, displaying substantial azimuthal variation especially for the rectangular case where the current extends farther and deposits more particles along the initial minor axis of the rectangular cross section. Interestingly, this departure from axisymmetry cannot be predicted by current theoretical methods such as the Box Model. Several parameters are varied to assess the dependence on the settling velocity, initial height aspect ratio, local curvature and mixture density. 1 NSF-PIRE OISE-0968313 3:20PM D32.00006 Process of entrainment in particulate gravity currents1 , MRUGESH SHRINGARPURE, University of Florida, JORGE SALINAS, MARIANO CANTERO, Institute Balseiro, Bariloche Atomic Center, San Carlos de Bariloche, Rio Negro, Argentina, S. BALACHANDAR, University of Florida — Various geophysical flows like turbidity currents, river flows, dust storms etc transport huge quantities of dispersed phase over large distances. Typically in such flows a dispersed phase rich layer is swept along with the flow. The amount of dispersed phase carried depends on the dynamics of this layer which are governed by a strong coupling between turbulence and suspended particles. This layer evolves, i.e., grows/shrinks in size, due to entrainment/detrainment of surrounding clear fluid at its interface (where a sharp change from particle rich fluid to surrounding clear fluid occurs). Also in many applications there is entrainment and detainment of particles at the bottom boundary due to settling and resuspension. The entrainment processes that occur here have important consequences. Consistent entrainment means the flow is energetic enough to mix/distribute the dispersed phase layer in the bulk flow. To study these processes, we introduce a layer of suspended particles into a fully turbulent channel flow and capture the entrainment processes in detail. Three parameters - Reynolds number, particle size and Richardson number dictate the entrainment process. Various simulations have been performed that explores this parametric space and identifies various entrainment regimes. 1 We acknowledge support from US NSF through grant OISE 0968313 and OCE 1131016 3:33PM D32.00007 Dynamics of bidensity suspensions in gravity-driven thin film flows , JEFFREY WONG, ANDREA BERTOZZI, Univ of California - Los Angeles — We study bidensity suspensions of a viscous fluid on an incline, where the particles migrate due to a combination of gravity-induced settling and shear induced migration. The physical problem is modeled by a hyperbolic system of conservation laws for the height and particle concentrations. We consider the constant flux problem and show that the system exhibits three-shock solutions corresponding to distinct fronts of particles and liquid traveling at different speeds, as well as singular shock solutions for sufficiently large concentrations, for which the mechanism predicted by the model is similar the single-species case. We also consider initial conditions corresponding to a finite reservoir of fluid, where solutions are rarefaction-shock pairs, and compare to experiments. 3:46PM D32.00008 Periodic oscillations of particles settling under gravity in a viscous fluid , MARIA L. EKIEL-JEZEWSKA, Institute of Fundamental Technological Research, Polish Academy of Sciences, Warsaw — New periodic solutions of three spherical particles settling under gravity in a viscous fluid are found and their relation to chaotic dynamics of a cluster of three randomly distributed particles is shown. An analogue of this solution has not been detected in the point-particle approximation. However, the existence of such an unstable periodic orbit was previously suggested by Janosi et al., Phys.Rev. E,56, 2858 (1997) and was claimed to be responsible for the numerically observed chaotic scattering of three point-particles settling under gravity. The significance of periodic orbits for dynamics of sedimenting particles in other non-regular configurations is also illustrated by other examples. 3:59PM D32.00009 Inertial particle clustering, relative velocity, and collision statistics in the presence of gravity , PETER J. IRELAND, ANDREW D. BRAGG, LANCE R. COLLINS, Cornell University — We use direct numerical simulations to investigate the dynamics of inertial particles in the presence of gravitational forces over a large range of Taylor-scale Reynolds numbers (90 ≤ Rλ ≤ 597). The particle inertial and gravitational forces are parameterized to provide insight into the motion and growth of water droplets in warm, cumulus clouds. We perform a detailed analysis of the effect of the periodic boundary conditions in the simulations and find that extended domain lengths are needed for accurate particle statistics, especially at low Reynolds numbers. While gravity reduces the relative velocities of all particle classes, it has a bifurcated effect on particle clustering, suppressing (enhancing) clustering for weakly (strongly) inertial particles. We provide a physical explanation of these trends by extending the model of Zaichik & Alipchenkov (New J. Phys., 11:103018, 2009) to account for gravitational effects. The particle statistics are strongly anisotropic, and we use spherical harmonic decomposition to quantify this anisotropy. Finally, we compare collision statistics of inertial particles with gravity to those without gravity and suggest practical implications for the onset of precipitation in cumulus clouds. 4:12PM D32.00010 Velocity and acceleration statistics from direct numerical simulations (DNS) of particle-laden homogeneous turbulent shear flow , PARVEZ SUKHESWALLA, ANDREW BRAGG, LANCE COLLINS, Cornell University — We study the effects of imposed mean shear and gravity (acting normal to shear) on the velocity and acceleration statistics of inertial particles in homogeneous turbulent shear flow. Single- and two-particle statistics from high-resolution DNS are analyzed for various particle sizes, shear rates, and settling parameters. We find that mean particle settling speeds are enhanced by turbulence, and are sensitive to changes in shear and gravity. Also, the net drift of particles due to gravity increases their locally averaged velocities in the mean-flow direction. Particle-pair relative velocities are anisotropic, with the radial inward component and the second-order structure function both maximal along the mean-strain contractional axis. Stronger shear shifts this orientation towards the mean flow streamlines, reflecting the larger particle fluctuating velocities in that direction. Gravity and shear together cause particle r.m.s. accelerations to increase with increasing inertia, in contrast to isotropic turbulence, but consistent with past turbulent boundary layer experiments. Analytical predictions for particle mean velocities and accelerations agree well with the DNS data. These results have important implications for the anisotropic collision kernel in shear flow. Sunday, November 23, 2014 2:15PM - 4:25PM — Session D33 Computational Methods and Modeling of Particle Laden Flows 2022 - Marc Massot, Ecole Centrale Paris 2:15PM D33.00001 Kinetic-Based Moment Methods for DNS and LES of particle-laden flows: the Anisotropic Gaussian Closure , SABAT MACOLE, EM2C-Ecole Centrale Paris, AYMERIC VIÉ, Center For Turbulence Research, Stanford, ADAM LARAT, EM2C-Ecole Centrale Paris, FRANCOIS DOISNEAU, EM2C-Ecole Centrale Paris-ONERA, CHRISTOPHE CHALONS, Université de Versailles Saint-Quentin, MARC MASSOT, EM2C-Ecole Centrale Paris — The simulation of particle-laden flows is a challenging topic due to their multiscale character. Lagrangian particle tracking methods are classically used. However, for high performance computing, such approaches deteriorate with the disperse phase inhomogeneities. Moment methods bypass this issue through an Eulerian framework allowing to use the same parallelization paradigm as the gas phase. We present recent developments for DNS and LES based on a Kinetic-Based Moment Method. The moment system is closed by assuming a presumed shape for the NDF. The selected NDF is an Anisotropic Gaussian giving the following properties: 1/ hyperbolicity; 2/ realizability of the moments; 3/maximization of entropy; 4/ H-theorem. The method is evaluated on configurations of increasing complexity that exhibit its potential and drawbacks. This method extends towards LES by means of a full kinetic-based filtering technique instead of filtering the moment equations. Thus realizability conditions are easily derived, and the main properties of the DNS system are preserved. The subgrid terms are closed following the work of Zaichik et al. 2009. The resulting LES strategy is evaluated based on filtered DNS results. 2:28PM D33.00002 A Dual-Scale Approach for Modeling Turbulent Two-Phase Interface Dynamics1 , MARCUS HERRMANN, Arizona State University — Turbulent liquid/gas phase interface dynamics are at the core of many applica- tons. For example, in atomizing flows, the properties of the resulting liquid spray are determined by the interplay of fluid and surface tension forces. The resulting dynamics typically span 4-6 orders of magnitude in length scales, making DNS exceedingly expensive. This motivates the need for modeling approaches based on spatial filtering or ensemble averaging. In this talk, a dual-scale modeling approach is presented to describe turbulent two-phase interface dynamics in a LES-type spatial filtering context. To close the unclosed terms related to the phase interface arising from filtering the Navier-Stokes equation, a resolved realization of the phase interface dynamics is explicitly filtered. This resolved realization is maintained on a high resolution over-set mesh using a Refined Local Surface Grid approach employing an un-split, geometric, bounded, and conservative Volume of Fluid method. The required model for the resolved realization of the interface advection velocity includes the effects of sub-filter surface tension, dissipation, and turbulent eddies. Results of the dual-scale model will be compared to recent DNS by McCaslin & Desjardins of an interface in homogeneous isotropic turbulence. 1 Supported by NSF grant CBET-1054272 and the 2014 CTR Summer Program. 2:41PM D33.00003 A simple stochastic quadrant model for the transport and deposition of particles in turbulent boundary layers , MICHAEL REEKS, University of Newcastle, CHUNYU JIN, IAN POTTS, University of Newcastle UK — We present a simple stochastic quadrant model for calculating the transport and deposition of heavy particles in a fully developed turbulent boundary layer based on the statistics of wall-normal fluid velocity fluctuations obtained from a fully developed channel flow. Individual particles are tracked through the boundary layer via their interactions with a succession of random eddies found in each of the quadrants of the fluid Reynolds shear stress domain in a homogeneous Markov chain process. Deposition rates for a range of heavy particles predicted by the model compare well with benchmark experimental measurements. In addition deposition rates are compared with those obtained continuous random walk (CRW) models including those based on the Langevin equation for the turbulent fluctuations. In addition, various statistics related to the particle near wall behavior are also presented. 2:54PM D33.00004 Immersed Boundary Methods on Non-Uniform Grids for Simulation of a Fully Resolved Bed of Particles in a Near-Wall Turbulent Flow , GEORGES AKIKI, S. BALACHANDAR, University of Florida — This study presents a dynamic distribution of the Lagrangian markers on a sphere when using the immersed boundary method with a non-uniform Eulerian mesh. The points are distributed in accordance with the surrounding Eulerian mesh to keep it optimized as the sphere moves in the channel. Also, a method is proposed to assign weights to the Lagrangian markers, both in the case of uniform and non-uniform distribution of the points. The newly proposed method of weight assignments uses vector spherical harmonics expansion of the weights. The error due to uneven distribution of the Lagrangian points is significantly reduced. These methods are then validated and applied in the simulation of a fully resolved bed of particles in a wall-bounded turbulent flow with periodic boundary conditions along the streamwise and spanwise directions. Results are analyzed for understanding of both effect of wall turbulence on particle motion and interaction, and the back effect of particles on the carrier-phase turbulence. 3:07PM D33.00005 An immersed boundary method for the interaction of turbulence with particles of arbitrary shape , SHIZHAO WANG, MARCOS VANELLA, ELIAS BALARAS, The George Washington University — In this work we present a computational scheme applicable to turbulence/particle interactions, targeting applications involving millions of particles of arbitrary shape. Immersed boundary methods have been frequently applied in simulating such problems, but are usually confined to spherical particles. Extension to rigid/deformable particles of arbitrary shape introduces significant challenges in achieving parallel efficiency. The proposed method is based on the moving least squares immersed boundary approach (Vanella & Balaras, J. Comput. Physics, 228(18), 6617-6628, 2009) on uniform and adaptive block-structured grids. We will present a novel parallelization strategy based on a master/slave model: the processor on which a body/structure resides is designated the master processor, while all the processors that contain at least one block overlapping with the body are designated the slaves. As the particle moves through the fluid, its blocks association and therefore the participating processors change. Effective ways of replicating the mesh metadata on all processors will be discussed. Results for homogeneous turbulence interacting with spherical and ellipsoidal particles and comparisons with experimental results will be given. 3:20PM D33.00006 An improved numerical method for the kernel density functional estimation of disperse flow1 , TIMOTHY SMITH, University of Illinois at Urbana-Champaign, REETESH RANJAN, Georgia Institute of Technology, CARLOS PANTANO, University of Illinois at Urbana-Champaign — We present an improved numerical method to solve the transport equation for the one-point particle density function (pdf), which can be used to model disperse flows. The transport equation, a hyperbolic partial differential equation (PDE) with a source term, is derived from the Lagrangian equations for a dilute particle system by treating position and velocity as state-space variables. The method approximates the pdf by a discrete mixture of kernel density functions (KDFs) with space and time varying parameters and performs a global Rayleigh-Ritz like least-square minimization on the state-space of velocity. Such an approximation leads to a hyperbolic system of PDEs for the KDF parameters that cannot be written completely in conservation form. This system is solved using a numerical method that is path-consistent, according to the theory of non-conservative hyperbolic equations. The resulting formulation is a Roe-like update that utilizes the local eigensystem information of the linearized system of PDEs. We will present the formulation of the base method, its higher-order extension and further regularization to demonstrate that the method can predict statistics of disperse flows in an accurate, consistent and efficient manner. 1 This project was funded by NSF project NSF-DMS 1318161 3:33PM D33.00007 An accurate and efficient Lagrangian sub-grid model for multi-particle dispersion1 , FEDERICO TOSCHI, Eindhoven University of Technology, IRENE MAZZITELLI, University of Rome Tor Vergata, ALESSANDRA S. LANOTTE, CNR - ISAC — Many natural and industrial processes involve the dispersion of particle in turbulent flows. Despite recent theoretical progresses in the understanding of particle dynamics in simple turbulent flows, complex geometries often call for numerical approaches based on eulerian Large Eddy Simulation (LES). One important issue related to the Lagrangian integration of tracers in under-resolved velocity fields is connected to the lack of spatial correlations at unresolved scales. Here we propose a computationally efficient Lagrangian model for the sub-grid velocity of tracers dispersed in statistically homogeneous and isotropic turbulent flows. The model incorporates the multi-scale nature of turbulent temporal and spatial correlations that are essential to correctly reproduce the dynamics of multi-particle dispersion. The new model is able to describe the Lagrangian temporal and spatial correlations in clouds of particles. In particular we show that pairs and tetrads dispersion compare well with results from Direct Numerical Simulations of statistically isotropic and homogeneous 3d turbulence. This model may offer an accurate and efficient way to describe multi-particle dispersion in under resolved turbulent velocity fields such as the one employed in eulerian LES. 1 This work is part of the research programmes FP112 of the Foundation for Fundamental Research on Matter (FOM), which is part of the Netherlands Organisation for Scientific Research (NWO). We acknowledge support from the EU COST Action MP0806. 3:46PM D33.00008 An unstructured overset method for particle-resolved simulation of particle-laden flows1 , WYATT HORNE, KRISHNAN MAHESH, University of Minnesota — Particle-laden flows involve a large range of length scales, ranging from the larger convective length scales down to length scales smaller than particle size. We develop a particle-resolved direct-numerical simulation (PR-DNS) method to enable the accurate study of the physics of particle-laden flow at particle length scales. Unstructured meshes are attached directly to particle surfaces and to the background flow field. The different meshes are allowed to arbitrarily overlap with each other to create a single cohesive solution. A dynamic connectivity procedure is used that cuts solid bodies out of each mesh and establishes interpolation pairs between overlapping meshes. The flow is incompressible, and the numerical method is based on that developed by Mahesh et al. [J. Comput. Phys. (2004) 197:215-240]. The overall discrete conservation properties for mass, momentum and kinetic energy are analyzed. Several cases are presented showing the method’s efficacy for studying particle-laden flow including single particle results and particle-to-particle interaction. 1 This work is supported by the DOE. 3:59PM D33.00009 LES of box turbulence with particles: SGS modeling of the particle acceleration1 , REMI ZAMANSKY, IMFT, MIKHAEL GOROKHOVSKI, Ecole Centrale de Lyon — When the Reynolds number is high, the tur- bulent flow on small length scales is characterized by strong velocity gradients. If such a flow is laden by inertial particles, those gradients, or specifically the turbulent time-scales shorter than the Stokes time, induce fluctuations in the particle motion. In LES, this motivates to simulate the interaction of particle with SGS flow. In our LES of box turbulence with particles, the particle acceleration was decomposed on its resolved and residual parts. The latter was assumed resulting from interactions in the inertial range, and was simulated stochastically along the particle trajectory. It was done by two processes, one for its norm, and another for its direction. Results showed that by introducing the stochastic model for the particle residual acceleration, the particle acceleration statistics from DNS was predicted fairly well. We also proposed the stochastic model for particles bigger than the Kolmogorov size. To this end, the fluctuating drag was derived, and simulated by lognormal process. This model predicted experimental observation: stretched tails in the particle acceleration distribution invariantly to the density and the size of particle. 1 We acknowledge the support from CTR, Stanford University for hosting this work during the 2014 Summer Program. 4:12PM D33.00010 Filter length scale for continuum modeling of subgrid physics , JULIAN SIMEONOV, JOSEPH CALANTONI, Naval Research Laboratory, Stennis Space Center — Modeling the wide range of scales of geophysical processes with direct numerical simulations (DNS) is currently not feasible. It is therefore typical to explicitly resolve only the large energy-containing scales and to parameterize the unresolved small scales. One approach to separate the scales is by means of spatial filters and here we discuss practical considerations regarding the choice of a volume averaging scale L. We use a macroscopically homogeneous scalar field and quantify the smoothness of the filtered field using a noise metric, ν, defined by the standard deviation of the filtered field normalized by the domain-averaged value of the field. For illustration, we consider the continuum modeling of the particle phase in discrete element method (DEM) simulations and the salt fingers in DNS of double-diffusive convection. We find that ν 2 follows an inverse power law dependence on L with an exponent and coefficients proportional to the domain-averaged field value. The empirical power law relation can aid in the development of continuum models from fully resolved simulations while also providing uncertainty estimates of the modeled continuum fields. Sunday, November 23, 2014 2:15PM - 4:25PM Session D34 Turbulent Non-Premixed Flames Toulouse — 2024 - Benedetta Franzelli, Institut National Polytechnique de 2:15PM D34.00001 High-Speed Tomographic PIV Measurements of Strain Rate Properties in Turbulent Partially-Premixed Jet Flames1 , BRUNO CORITON, JONATHAN H. FRANK, Sandia Natl Labs — The effects of combustion on the strain rate field in turbulent jets were studied using 10 kHz tomographic particle image velocimetry (TPIV). Measurements were performed in a series of turbulent partially-premixed jet flames with increasing jet Reynolds numbers and increasing probabilities of localized extinction. Properties of the strain rate were analyzed including the relative ratios of principal strain rates, the preferential alignment of the principal strain rates with vorticity, and the strain rate clustering and intermittency. Comparisons with measurements in turbulent air jets revealed the effects of heat release on the structure and dynamics of the strain rate field. 1 This material is based upon work supported by the US Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences. 2:28PM D34.00002 Simultaneous PIV/OH-PLIF measurements in the wake of a reacting jet in swirling vitiated crossflow , PRATIKASH PANDA, Purdue University, MARIO ROA, University of California, Los Angeles, YASHOWARDHAN WAGH, ROBERT LUCHT, Purdue University — A reacting jet issuing into a swirling, vitiated cross flow was investigated as a means of secondary injection of fuel in a distributed combustion system. Rapid mixing and chemical reaction in the near field of the jet injection is desirable in this application. Current study present time resolved planar measurements within the wake of reactive jets using simultaneous 2D-PIV/OH-PLIF at a repetition rate of 10 kHz. Based on our analysis it is hypothesized that the shear layer and wake field vortices play a significant role is stabilizing a steady reaction front within the near wake region of the jet. The reactive jets were injected through an extended nozzle into the crossflow which is located in the downstream of a low swirl burner (LSB) that produced the swirled, vitiated crossflow. PIV measurements and OH–PLIF based flame visualizations were acquired simultaneously at three measurement planes along the cross- section of the jet. The time resolved measurements provided significant information on the evolution of complex flow structures and highly transient features like, local extinction, re-ignition, vortex-flame interaction prevalent in a turbulent reacting flow. 2:41PM D34.00003 Structure and stabilization of hydrogen-rich transverse jets in a vitiated turbulent flow , SGOURIA LYRA, Sandia National Laboratories, BENJAMIN WILDE, Georgia Institute of Technology, HEMANTH KOLLA, Sandia National Laboratories, JERRY SEITZMAN, TIM LIEUWEN, Georgia Institute of Technology, JACQUELINE CHEN, Sandia National Laboratories — Results are presented from a joint experimental and numerical study of the flow characteristics and flame stabilization of a hydrogen rich jet injected into a turbulent, vitiated crossflow of lean methane combustion products. Simultaneous high-speed stereoscopic PIV and OH PLIF measurements are obtained alongside 3D direct numerical simulations of inert and reacting JICF with detailed H2 /CO chemistry. Under the investigated conditions an autoigniting, burner-attached flame initiates uniformly around the burner edge. Significant asymmetry is observed between the reaction zones located on the windward and leeward sides, due to the substantially different scalar dissipation rates. The unsteady dynamics of the windward shear layer are explored to elucidate the important flow stability implications arising in the reacting JICF. Vorticity spectra extracted from the windward shear layer reveal that the reacting jet is globally unstable and features two high frequency peaks, including a fundamental mode whose Strouhal number of ∼ 0.7 agrees well with previous non-reacting JICF stability studies. Chemical explosive mode analysis shows that the entire windward shear layer, and a large region on the leeward side, are highly explosive prior to ignition and are dominated by non-premixed flame structures after ignition. The predominantly mixing limited nature of the flow after ignition is confirmed by the Takeno flame index, showing that ∼ 70% of the heat release occurs in non-premixed regions. 2:54PM D34.00004 Experimental investigation of the velocity field of a laboratory fire whirl1 , KATHERINE HARTL, Princeton University, PENGFEI WANG, University of Science and Technology of China, ALEXANDER SMITS, Princeton University, Monash Universty — A fire whirl is a swirling diffusion flame that may occur to great destructive effect in urban fires or wildfires. To study fire whirls in the laboratory, we use a burner flame supplied with DME, and induce swirl by entraining air through a split cylinder surrounding the central flame. Stereo Particle Image Velocimetry (PIV) is used to obtain distributions of the three components of velocity inside and outside the fire whirl core. The effects of heat release rate and gap size on whirl height, circulation, and air entrainment are examined, and scaling behavior is discussed. 1 This work was partially supported by the Princeton University Andlinger Center for Energy and the Environment. 3:07PM D34.00005 Models for differential diffusion in turbulent non-premixed combustion1 , HAIFENG WANG, Purdue University — Models for differential diffusion are developed and are incorporated in the flamelet model for turbulent non-premixed combustion. The models are based on the limiting behavior of differential diffusion in turbulent combustion at zero and infinite Reynolds numbers. The effect of differential diffusion in a finite Reynolds number flame is approximated by the blending of the two limits. A turbulent non-premixed CH4/H2/N2 jet flame is adopted as a validation test case. The modeling results are found to be in excellent agreement with the experimental data, including the level of differential diffusion. 1 Acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for support of this research. 3:20PM D34.00006 Effects of differential diffusion on the flame structure of oxygen enhanced turbulent non-premixed jet flames1 , FELIX DIETZSCH, MICHAEL GAUDING, CHRISTIAN HASSE, TU Bergakademie Freiberg — By means of Direct Numerical Simulation we have investigated the influence of differential diffusion for non-premixed oxygen-enhanced turbulent flames. Oxygenenhanced conversion usually yields higher amounts of H2 as compared to conventional air combustion. It is well known that H2 as a very diffusive species leads to differential diffusion effects. In addition to the diffusive transport mixing processes are also often controlled by turbulent transport. Previous investigations of a turbulent CH4/H2 oxygen-enhanced jet flame have shown that in mixture fraction space it is important to distinguish between regions of equal diffusivities and detailed transport. These findings are of particular interest when performing Large-Eddy simulations applying a flamelet approach. Using this approach a LES study was performed of the aforementioned flame considering differential diffusion. Therefore, flamelet equations including differential diffusion via non-unity constant Lewis numbers were solved. However, this study showed that keeping the non-unity Lewis numbers constant, is not sufficient to capture the diffusion phenomena in this particular flame. Direct Numerical Simulations have been conducted in order to investigate how Lewis numbers are affected in mixture fraction space. 1 Computer resources for this project have been provided by the Gauss Centre for Supercomputing/Leibniz Supercomputing Centre under grant: pr83xa. 3:33PM D34.00007 ABSTRACT WITHDRAWN — 3:46PM D34.00008 Validity of the constant non-unity Lewis number assumption in chemically reacting flows , NICHOLAS BURALI, GUILLAUME BLANQUART, Caltech, YUAN XUAN, Penn State — Describing molecular diffusion using constant but non-unity Lewis numbers has been widely used in numerical simulations of chemically reacting flows. These constant Lewis numbers need to be selected carefully, as they should correctly describe the molecular diffusion of different species. However, in practice they are extracted from one-dimensional flame structure calculations. The objective of the current work is to assess the validity of the constant non-unity Lewis number assumption in the description of molecular mixing. Towards this goal, a three-tiered analysis is carried out. First, the sensitivity of key reacting flow characteristics to species Lewis numbers is assessed on both laminar diffusion flames and laminar premixed flames. Second, detailed numerical simulations using the multi-component diffusion model are performed for the same flames, and used as reference data. The validity of different Lewis number extraction criteria is examined by comparing simulation results obtained by using different sets of Lewis numbers to the reference data, and an optimal criterion is proposed. Finally, as a validation, a turbulent flame simulation is performed using Lewis numbers extracted following this optimal criterion, and results are compared to the experimental measurements. 3:59PM D34.00009 Direct Numerical Simulation Study of Thermochemical Nonequilibrium Effect on Mixing and Combustion , ROMAIN FIEVET, STEPHEN VOELKEL, HEESEOK KOO, VENKAT RAMAN, PHILIP VARGHESE, The University of Texas at Austin — Nonequilibrium of internal states of molecules is an important physical phenomenon that could affect flow behavior in supersonic flows. Translational nonequilibrium, where molecular velocities do not conform to the Maxwell distribution could impact dissipation processes in turbulence. Similarly, vibrational and/or rotational nonequilibrium will lead to marked changes in mixing and combustion. In this study, these nonequilibrium effects are explored using direct numerical simulation of a supersonic hydrogen jet issuing into a coflow of air. Nonequilibrium reaction rates derived using detailed computational chemistry methods are used in the flow simulations. It is shown that underpopulation of vibrational states leads to significant change in flame stabilization. Hence, the processing of the incoming air by the bow shocks formed ahead of a scramjet could lead to significant ignition delay. 4:12PM D34.00010 Principal Component Transport in Turbulent Combustion1 , TAREK ECHEKKI, HESSAM MIRGOLBABAEI, North Carolina State University — We present a posteriori validation of the solution of a turbulent combustion problem based on the transport of principal components (PCs). The PCs are derived from a priori principal component analysis (PCA) of the same composition space. This analysis is used to construct and tabulate the PCs’ chemical source terms and diffusion coefficients in terms of the PCs using artificial neural networks (ANN). The a posteriori validation is implemented on a stand-alone one-dimensional turbulence (ODT) simulation of Sandia flame F resulting in a very good reconstruction of the original thermo-chemical scalars profiles at different downstream distances. 1 The work was supported by the National Science Foundation grant DMS- 1217200. Sunday, November 23, 2014 2:15PM - 4:25PM Session D35 Compressible Flow II: Stability and Boundary Layers — 2001A - James Hermanson, University of Washington 2:15PM D35.00001 A Navier-Stokes-Based Approach for Mean Flow Perturbation Analysis1 , SWAGATA BHAUMIK, DATTA GAITONDE, MBU WAINDIM, The Ohio State University, THE OHIO STATE UNIVERSITY TEAM — The manner in which a basic state, obtained from a time-averaged unsteady flowfield, processes perturbations can provide significant insight into the cause and evolution of instabilities. A widely used approach is based on Parabolized Stability Equations (PSE), which limits streamwise mean flow variation and is often applied to 2-D base flows. To avoid some of these issues, we advance a Navier-Stokes-based method, which can address non-trivial three-dimensional fields. The method stems from that employed by Touber and Sandham (Theor. Comput. Fluid. Dyn., 23, 79-107, 2009) to analyze global modes in nominally 2-D shock-wave turbulent-boundary layer interactions (STBLI). We first develop its theoretical underpinnings by examining conditions under which it degenerates to traditional methods. We then illustrate the application by considering perturbations to an entropy layer at Mach 6, a turbulent supersonic jet at Mach 1.3 and STBLI at Mach 2.3. For the entropy layer and jet cases, known linear stability and PSE results are successfully reproduced, while global modes are obtained for STBLI. The results not only validate the proposed technique, but also demonstrate its suitability in analyzing instabilities for any general 3D basic state, including impulse response. 1 Sponsored by AFOSR 2:28PM D35.00002 Parametric sensitivity for frequency response analysis of large-scale flows , MIGUEL FOSAS DE PANDO, Universidad de Cadiz, PETER SCHMID, Imperial College London — When studying the frequency response of globally stable flows, direct and adjoint information from a resolvent analysis has to be computed. These computations involve a sizeable amount of effort, which suggests their reuse to identify sensitivity measures to changes in the governing parameters, base/mean flow fields, boundary conditions or other changes to the underlying linearized operator. We introduce and demonstrate a general technique to determine first-order changes in the frequency response induced by general changes to the governing equations. Examples will include changes to the Reynolds and Mach number for a tonal-noise airfoil problem, sensitivity to heating of a mixing layer past a splitter plate and closeness to global instability for a simplified model equation. 2:41PM D35.00003 An h/p adaptive Discontinuous Galerkin method for Three-Dimensional Compressible Flows , JOHN EKATERINARIS, Embry-Riddle Aeronautical University, KONSTANTINOS PANOURGIAS, University of Patras — High order discontinuous Galerkin (DG) discretizations possess features making them attractive for computations of three-dimensional complex, compressible flows with discontinuities. Development of unified limiting procedures for the DG method that ensure accurate capturing of discontinuities for unstructured meshes, required for simulations in domains with nontrivial geometry, is needed. A TVB limiter is used and applied in the canonical computational space. It results into a unified limiting procedure for DG discretizations with any type of elements. The performance of the unified limiting approach is shown for different types of elements employed in mixed-type meshes and for a number of standard test problems including strong shocks to demonstrate the potential of the method. Furthermore, hierarchical higher-order limiting with the proposed approach is demonstrated. Increased order of expansion and adaptive mesh refinement is introduced in the context of h/p- adaptivity in order to locally enhance resolution of three-dimensional flow simulations that include discontinuities and complex flow features. 2:54PM D35.00004 Meshless Compressible Flow Simulations on Graphical Processor Units (GPUs) , JOHN THOMAS, JACOB ALLDREDGE, Johns Hopkins University — A computationally efficient framework for performing compressible flow simulations over rigid solids is presented. This framework, which is based on a lattice-Boltzmann model, incorporates a volume fraction-based voxelation algorithm to eliminate the explicit meshing process. Moreover, as a framework tuned to run on graphical processers units (GPUs), simulations involving tens-of-billions of grid points can be run on hobby-sized computers in about one day. We validate predictions from this framework using experimental data for flow past wedges, spheres, and airfoils at a variety of Mach numbers. 3:07PM D35.00005 Compression wave structure on droplets under supersonic conditions , ERIC LIN, JAMES HERMANSON, University of Washington — The compression wave structure in the vicinity of droplets deforming in a continuously accelerating, supersonic flow was examined in a draw-down supersonic wind tunnel. This flow configuration allowed droplets to achieve a Mach number of up to 1.7 relative to the surrounding air stream. Neat 2-propanol droplets 100 microns in diameter were generated upstream of the tunnel entrance using a Droplet-On-Demand generator. Schlieren imaging was performed to visualize the deforming droplets and to image the shock wave structure. Theoretical predictions provided a first estimate for bow shock parameters under these flow conditions such as shock thickness, standoff distance, and shock reaction time, suggesting that detached shock waves can be expected to be present for droplets experiencing the locally supersonic conditions in this investigation. The observed shock waves have characteristics broadly consistent with those expected for detached bow shock waves in front of a bluff body. The relative droplet Mach numbers, inferred from the Mach angle suggested by the schlieren images, are consistent with droplet Mach numbers determined previously in this flow configuration by direct imaging. 3:20PM D35.00006 Measurements of Vibrational Non-equilibrium in Supersonic Jet Mixing and Combustion1 , HEATH REISING, TIMOTHY HALLER, NOEL CLEMENS, PHILIP VARGHESE, The University of Texas at Austin — A new experimental facility has been constructed to study the effects of thermal non-equilibrium on supersonic mixing and combustion. The facility consists of a Mach 1.5 turbulent jet issuing into an electrically heated coflow. The degree of non-equilibrium in the jet shear layer is quantified using high spectral resolution time-averaged spontaneous Raman scattering. Since the Raman spectra are time-averaged, they are susceptible to non-linear weighting effects induced by temperature fluctuations. The effect of local turbulent temperature fluctuations on the Raman fitting procedure is quantified by using spectral simulations that use the actual temperature fluctuations present in the flow measured by instantaneous Rayleigh scattering thermometry. It is shown that the temperature fluctuations are not large enough to induce significant errors in the vibrational temperature fitting results. Vibrational non-equilibrium is shown to occur in the jet shear layer, and its magnitude and trend are shown to be similar to recent large-eddy-simulation results. Since CO2 is known to cause faster vibrational relaxation of N2 , a series of experiments were conducted to verify that the non-equilibrium effects could be controlled by CO2 addition. This work is being extended to reacting flows, to assess the impact of non-equilibrium on supersonic shear-layer combustion. 1 This work was supported by the Air Force Office of Scientific Research 3:33PM D35.00007 Space-time measurements in a shock wave/turbulent boundary layer interaction1 , ANNE-MARIE SCHREYER, Aix-Marseille Universite, CNRS, IUSTI UMR 7343, Marseille / Centre National d’Etudes Spatiales CNES, DLA, Paris, PIERRE DUPONT, Aix-Marseille Universite, CNRS, IUSTI UMR 7343, Marseille, France — We study a reflected shock interaction with separation at Mach 2, contributing to a better understanding of rocket engine nozzle flows. The flow field contains a wide range of characteristic frequencies between O(100)Hz for the oscillation of the reflected shock and O(100)kHz for the turbulent microscales. To explain the origin and interdependence of the physical phenomena in the interaction, we need access to the spatio-temporal links. We thus require a measurement technique allowing the resolution of the entire frequency range while also providing sufficient spatial resolution and a large field of view. Our newly developed Dual-PIV system satisfies these requirements. First measurements with this system in an interaction flow field were performed in the continuous hypo-turbulent wind-tunnel at IUSTI at a momentum thickness Reynolds number of Reθ = 5024 and a deflection angle of θ = 8.75◦ . We present a detailed characterization of the flow field including turbulence measurements. From measurements at a range of temporal delays, we determined autocorrelations at crucial points in the flow field (incoming boundary layer, mixing layer, relaxation zone). From these, spatio-temporal information like the integral scales and the convection velocity are deduced. 1 This work received financial support by the CNES within the research program ATAC and also the ANR within the program DECOMOS. This support is gratefully acknowledged. 3:46PM D35.00008 ABSTRACT WITHDRAWN — 3:59PM D35.00009 Shock wave Boundary layer interaction in supersonic flow over a forwardfacing step , JAYAPRAKASH NARAYAN M., RAGHURAMAN GOVARDHAN, Indian Institute of Science — Shock wave boundary layer interactions (SWBLI) are known to result in low-frequency large-scale shock oscillations, the origin of which has been a subject of debate. Motivated by this debate, we study in the present work, the SWBLI in supersonic flow over a Forward-Facing Step (FFS) at a Mach number of 2.5. The FFS configuration, which consists of a 90 degree step of height h, may be thought of as an extreme case of the compression ramp geometry, with the main geometrical parameter here being (h/δ) (δ is the boundary layer thickness). This configuration is less studied and has some inherent advantages for experimentally studying SWBLI as the size of the separation bubble is large. In the present experimental study, we use high-speed schlieren and PIV measurements to help understand the features of SWBLI in the forward-facing step case. PIV measurements show a clear time-averaged separation bubble ahead of the step, with very large variations of the separation bubble in time. From instantaneous PIV velocity fields, a number of features are extracted including size of the separation bubble and the shock location, to comment on their variations in time, and to determine correlation coefficients. 4:12PM D35.00010 Preliminary LES of Hypersonic Shock/Turbulent Boundary Layer Interaction , CLARA HELM, PINO MARTIN, University of Maryland, College Park — Preliminary results from the Large Eddy Simulation (LES) of two hypersonic Shock/Turbulent Boundary Layer Interactions (STBLIs) are presented. First it is demonstrated with the simulation of a Mach 3 interaction over a 24o compression ramp that the LES method used is capable of resolving the relevant features of the complex dynamics present in separated STBLIs. Features such as the separation low-frequency dynamics, turbulence magnification, shear layer dynamics, and wall and skin friction distributions are validated against the Direct Numerical Simulation (DNS) data of the same Mach 3 flow. The LES is then validated for the computation of hypersonic conditions by simulating an attached Mach 7 STBLI generated by an 8o compression ramp and comparing results to DNS data of the same flow conditions. Lastly, initial results from the LES of a Mach 7 separated interaction over a 33o compression ramp at experimentally achievable conditions will be presented and discussed. This work is supported by the Air Force Office of Scientific Research under grant AF/9550-10-1-0164. Sunday, November 23, 2014 2:15PM - 3:59PM Session D36 Rarefied Flows — Alcove A - Venkatraman Ayaswamy 2:15PM D36.00001 Plane Poiseuille flow of a highly rarefied gas between the two walls of Maxwell-type boundaries with different accommodation coefficients: Effect of a weak external force , TOSHIYUKI DOI, Department of Applied Mathematics and Physics, Tottori University — Plane Poiseuille flow of a highly rarefied gas between the two walls of Maxwell-type boundaries with different accommodation coefficients is studied based on kinetic theory when the gas is subject to a weak external force perpendicular to the walls. The flow behavior is studied numerically based on the spatially one-dimensional Boltzmann equation for a hard-sphere gas derived by the asymptotic analysis for a slow variation in the longitudinal direction. Due to the effect of a weak external force, there is an appreciable difference in the mass-flow rate between the flows in which the two walls are interchanged when the mean free path is sufficiently large. If both of the accommodation coefficients are close to unity, the mass-flow rate is reduced due to the effect of the external force. In contrast, if the accommodation coefficient of one wall is considerably smaller than unity, the mass-flow rate of the gas can be enhanced when this wall is placed on the side to which the external force points. 2:28PM D36.00002 Frequency-Domain DSMC Method for Oscillatory Gas Flows , DANIEL LADIGES, JOHN SADER, The University of Melbourne — Gas flows generated by resonating nanoscale devices inherently occur in the non-continuum, low Mach number regime. Numerical simulation of such flows presents a tremendous challenge, which has motivated the development of several direct simulation Monte Carlo (DSMC) methods for low Mach number flows. We present a frequency-domain DSMC method for oscillatory low Mach number gas flows, based on the linearized Boltzmann equation. This circumvents the need for temporal simulations, providing direct access to both amplitude and phase information using a pseudo-steady algorithm. The proposed method is demonstrated with several examples, and good agreement is found with both existing time-domain DSMC methods and accurate numerical solutions of the Boltzmann-BGK equation. Analysis of these simulations, using a rigorous statistical approach, shows that this frequency-domain method provides a significant improvement in computational speed compared to existing time-domain DSMC methods. 2:41PM D36.00003 Relaxation rates in the Maxwellian collision model and its variable hard sphere surrogate , ROBERT RUBINSTEIN, None — The variable hard sphere and related models have proven to be accurate and computationally convenient replacements for the inverse power law model of classical kinetic theory in DSMC calculations. We provide theoretical support for this success by comparing the relaxation rates in the linearized Boltzmann equation for the Maxwellian model with those of its variable hard sphere surrogate. We demonstrate that the linearized collision operators for these two models agree closely under well defined and broadly applicable conditions and show some implications of this agreement for time dependent solutions of the linearized Boltzmann equation. 2:54PM D36.00004 Hybrid DSMC-LBM scheme for pressure-driven flows , GIANLUCA DI STASO, FEDERICO TOSCHI, HERMAN J.H. CLERCX, Eindhoven Univ of Tech — Lattice Boltzmann Method (LBM) is a standard numerical methodology that, in principle, can be used to simulate flows ranging from hydrodynamic to rarefied gas as it is based on a discretisation of the Boltzmann equation. However at increasing rarefaction the number of needed speeds may become very large making the scheme prohibitively expensive under very rarefied conditions. Another classical technique, also based on the discretisation of the Boltzmann equation, and effective under rarified conditions is the Direct Simulation Monte Carlo (DSMC). We present results on the development of a hybrid scheme able to combine both numerical techniques to efficiently and accurately study flows with varying rarefaction. Details of the implementation and validation against few representative flows will be presented. 3:07PM D36.00005 A continuum breakdown parameter based on the characteristic function of the molecular velocity distribution , ARGHAVAN ALAMATSAZ, AYYASWAMY VENKATTRAMAN, Univ of California - Merced — Rarefied flows characterized by Knudsen numbers (Kn) greater than 0.1 are frequently encountered in several applications including low-pressure, high speed and microscale flows and require computationally expensive molecular approaches such as direct simulation Monte Carlo (DSMC) to accurately capture the physical phenomena unique to these flows. However, most of these flows also contain regions where traditional inexpensive continuum techniques such as the Navier-Stokes (NS) equations are sufficiently accurate making a hybrid NS-DSMC approach attractive and optimal. Such a hybrid method typically requires a robust continuum breakdown parameter (CBP) to determine regions where each method should be applied. Historically, hybrid methods have used CBPs based on the macroscopic properties which are lower order moments of the molecular velocity distribution function (VDF) and their gradients which can have significant inaccuracies. In this work, we propose a novel CBP that utilizes all moments of the VDF by computing the characteristic function with limited computational overhead. We also compare the performance of this CBP using standard benchmark problems including structure of a normal shock wave and Fourier-Couette flow for various Kn from continuum to free-molecular. 3:20PM D36.00006 An evaluation of collision models in the Method of Moments for rarefied gas problems1 , DAVID EMERSON, XIAO-JUN GU, STFC Daresbury Laboratory — The Method of Moments offers an attractive approach for solving gaseous transport problems that are beyond the limit of validity of the Navier-Stokes-Fourier equations. Recent work has demonstrated the capability of the regularized 13 and 26 moment equations for solving problems when the Knudsen number, Kn (where Kn is the ratio of the mean free path of a gas to a typical length scale of interest), is in the range 0.1 and 1.0–the so-called transition regime. In comparison to numerical solutions of the Boltzmann equation, the Method of Moments has captured both qualitatively, and quantitatively, results of classical test problems in kinetic theory, e.g. velocity slip in Kramers’ problem, temperature jump in Knudsen layers, the Knudsen minimum etc. However, most of these results have been obtained for Maxwell molecules, where molecules repel each other according to an inverse fifth-power rule. Recent work has incorporated more traditional collision models such as BGK, S-model, and ES-BGK, the latter being important for thermal problems where the Prandtl number can vary. We are currently investigating the impact of these collision models on fundamental low-speed problems of particular interest to micro-scale flows that will be discussed and evaluated in the presentation. 1 Engineering and Physical Sciences Research Council under Grant EP/I011927/1 and CCP12. 3:33PM D36.00007 Generalized slip-flow theory and its related Knudsen-layer analysis for a slightly rarefied gas I1 , MASANARI HATTORI, Department of Mechanical Engineering and Science, Kyoto University, Kyoto 615-8540, Japan, SHIGERU TAKATA, Department of Aeronautics and Astronautics & Advanced Research Institute of Fluid Science and Engineering, Kyoto University, Kyoto 615-8540, Japan — A systematic asymptotic analysis of the Boltzmann equation shows that the overall behavior of a gas can be described by fluid-dynamic-type equations with the appropriate slip/jump boundary condition when the Knudsen number is small [the generalized slip-flow theory (Sone, Molecular Gas Dynamics, 2007)]. Near the boundary, a non-fluid-dynamic correction (the Knudsen-layer correction) to the overall solution is required. Although the generalized slip-flow theory has been established up to the second order of the Knudsen number expansion, the data of those corrections have been completed only for the BGK model. Completing the corresponding data for the original Boltzmann equation has been demanded. In the present work, partial results of completing the data for a hard-sphere gas under the diffuse reflection condition are reported. 1 The present work is supported by JSPS KAKENHI Grant Numbers 23360083 and 13J01011. 3:46PM D36.00008 Dual-Wavelength Interferometry Plasma Electron Density Measurements , BRIAN NEISWANDER, ERIC MATLIS, THOMAS CORKE, University of Notre Dame — Plasma is an optically controllable medium with potential for improving high-speed adaptive optics technology, particularly in aero-optical wavefront-control. The index of refraction of a plasma depends on the electron density and gas density. These two parameters are highly coupled and must be uniquely determined in order to assess the effectiveness of plasma as a high-speed adaptive optic medium. Presented here are time-resolved experimental measurements of plasma electron density and gas density for a low-pressure cylindrical dielectric barrier discharge (DBD). Optical measurements were obtained using a dual-wavelength Michelson interferometer system featuring visible (0.633 µm) and infrared (3.39 µm) HeNe lasers. Along with results, a method used to increase the accuracy of the measurement system by incorporating a piezoelectric actuated scanning mirror and phase-demodulation analysis will be discussed. Sunday, November 23, 2014 4:45PM - 6:03PM Session E1 Non-Newtonian Flows: Applications — 3000 - Anubhav Tripathi, MIT 4:45PM E1.00001 Connecting the Rheological Behavior of Clathrate Hydrate Slurries to Flow Performance , MICHELA GERI, MIT, RAMA VENKATESAN, Chevron ETC, GARETH MCKINLEY, MIT, MIT TEAM, CHEVRON ETC TEAM — Clathrate hydrates represent a major flow assurance issue for deep water drilling operations. To develop a proper constitutive model, an extensive set of rheological measurements has been performed on a model hydrate forming emulsion. Upon hydrate formation a sharp increase in the fluid viscosity is observed (by a factor of 100 to 1000). Steady shear measurements show that the hydrate slurry has a shear thinning behavior as well as a yield stress on the order of 1-10 Pa which increases with aging of the fluid. Thixotropy becomes evident as a hysteretic behavior in the flow curve, even when no rheological aging has occurred. Creep tests also reveal that the fluid microstructure accumulates back stress. Oscillatory measurements show that in the linear viscoelastic region hydrate slurries develop viscoelastic gel-like behavior with the elastic modulus exceeding the viscous modulus. These characteristics guide the development of an elastoviscoplastic constitutive model that can capture the salient dynamic features in simple unidirectional flows (e.g. steady or transient Poiseuille) such as apparent wall slip, plug flow or excessive pressure drop in start-up flow. 4:58PM E1.00002 The role of extensional viscosity in frog tongue projection , ALEXIS NOEL, Georgia Inst of Tech, CAROLINE WAGNER, GARETH MCKINLEY, Massachusetts Inst of Tech, JOE MENDELSON, DAVID HU, Georgia Inst of Tech — Frogs and other amphibians capture insects through high-speed tongue projection, some achieving tongue accelerations of over fifty times gravity. In this experimental study, we investigate how a frog’s sticky saliva enables high-speed prey capture. At the Atlanta zoo, we used high-speed video to film the trajectory of frog tongues during prey capture. We have also designed and built a portable extensional rheometer; by following the capillary-driven thinning in the diameter of a thread of saliva we characterize the relaxation time and extensional viscosity and so infer the adhesive force between the frog tongue and prey. 5:11PM E1.00003 Creep and fracture of a model yoghurt1 , SEBASTIEN MANNEVILLE, MATHIEU LEOCMACH, CHRISTOPHE PERGE, Laboratoire de Physique - Ecole Normale Superieure de Lyon, THIBAUT DIVOUX, Centre de Recherche Paul Pascal - CNRS — Biomaterials such as protein or polysaccharide gels are known to behave qualitatively as soft solids and to rupture under an external load. Combining optical and ultrasonic imaging to shear rheology we show that the failure scenario of a model yoghurt, namely a casein gel, is reminiscent of brittle solids: after a primary creep regime characterized by a macroscopically homogeneous deformation and a power-law behavior which exponent is fully accounted for by linear viscoelasticity, fractures nucleate and grow logarithmically perpendicularly to shear, up to the sudden rupture of the gel. A single equation accounting for those two successive processes nicely captures the full rheological response. The failure time follows a decreasing power-law with the applied shear stress, similar to the Basquin law of fatigue for solids. These results are in excellent agreement with recent fiber-bundle models that include damage accumulation on elastic fibers and exemplify protein gels as model, brittle-like soft solids. 1 Work funded by the European Research Council under grant agreement No. 258803 5:24PM E1.00004 Predicting Pressure Profiles of Cement Columns in Oil Wells Using a Thixotropic Model , RAFAEL OLIVEIRA, FLÁVIO MARCHESINI, Halliburton — It is important to the oil and gas industry to provide proper well-bore isolation from the surrounding porous formations. This can be aided by predicting and preventing formation fluid invasion after primary cementing an oil well. In that regard, this work investigates the downhole pressure profile of a cement column placed in the annular space between the casing and the formation. The developed model takes into account the influence of (i) fluid loss to the geological formation, (ii) thixotropy and structure development during gelation, and (iii) compressibility and shrinkage of the cement slurry. This is a one-dimensional model where shear rates are estimated by the downhole velocity of the cement slurry and the annular distance. The thixotropic model recently proposed by de Souza Mendes and Thompson (Rheologica Acta, 2013) is used to calculate shear stresses, which are then plugged into the momentum equation. This equation is coupled with an equation for pressure evolution derived from mass balance and compressibility considerations. The model is under validation against large-scale cementing experiments, and application to current oil field data show promising results. 5:37PM E1.00005 Phase Behavior of Dilute Carbon Black Suspensions and Carbon Black Stabilized Emulsions , MICHAEL GODFRIN, Center for Biomedical Engineering, School of Engineering, Brown University, AYUSH TIWARI, Department of Civil Engineering, Thapar University, ARIJIT BOSE, Department of Chemical Engineering, University of Rhode Island, ANUBHAV TRIPATHI, Center for Biomedical Engineering, School of Engineering, Brown University — We use para-amino benzoic acid terminated carbon black (CB) as a tunable model particulate material to study the effect of inter-particle interactions on phase behavior and steady shear stresses in suspensions and particle-stabilized emulsions. We modulate inter-particle interactions by adding NaCl to the suspension, thus salting surface carboxylate groups. Surprisingly, yield stress behavior emerged at a volume fraction of CB as low as φCB = 0.008, and gel behavior was observed at φCB >0.05, well below the percolation threshold for non-interacting particles. The yield stress was found to grow rapidly with carbon black concentration suggesting that salt-induced hydrophobicity leads to strong inter-particle interactions and the formation of a network at low particle concentrations. The yield stresses of CB-stabilized emulsions also grows rapidly with carbon black concentrations, implying that inter-droplet interactions can be induced through the tuning of carbon black concentration in emulsion systems. Emulsions stabilized by ionic surfactants show no inter-droplet interactions. In contrast, oil droplets in the CB-stabilized emulsion move collectively or are immobilized because of an interconnected CB network in the aqueous phase. 5:50PM E1.00006 Capillary thinning and breakup of saliva threads and rheological aging of mucin solutions , CAROLINE WAGNER, LYDIA BOUROUIBA, GARETH MCKINLEY, Massachusetts Institute of Technology — The elasticity of saliva, which is essential for many of its primary functions such as lubrication, arises largely as a result of the presence of MUC5B mucins. These are large glycoproteins composed of numerous repeated polymeric subunits forming a weakly crosslinked network. It has been noted for nearly a century that once removed from the mouth, saliva quickly loses its elasticity, which can be quantified by a decrease in its capillary breakup time. We model saliva as a dilute finitely extensible nonlinear elastic (FENE-P) fluid with polymer chains composed of dispersed Hookean dumbbells of maximum extensibility b related to the number of MUC5B subunits. We show that under conditions of simple elongational flow, an analytic prediction of the time evolution of the radius and the filament breakup time can be derived. Furthermore, our model shows that decreasing the maximum extensibility b leads to a decrease in the breakup time, which suggests that the aging process of saliva outside the mouth involves a shortening of the MUC5B mucin chains into smaller groupings. Finally, we compare the analytic breakup times from the model with experimental results obtained using a capillary breakup extensional rheometer and human whole saliva. Sunday, November 23, 2014 4:45PM - 6:03PM Session E2 Suspensions: Confined Flows — 3002 - Roseanna Zia, Cornell University 4:45PM E2.00001 Diffusion and rheology in a suspension of hydrodynamically interacting colloids enclosed by a spherical cavity1 , CHRISTIAN APONTE-RIVERA, ROSEANNA ZIA, Cornell University — We study diffusion and rheology of a suspension of hydrodynamically interacting colloidal spheres enclosed by a spherical cavity, utilizing the Stokesian Dynamics framework to account for long-range many-body and pairwise lubrication interactions between the particles and between particle and enclosure. Previous studies of 1D- and 2D-confined suspensions have revealed that boundaries exert a pronounced qualitative influence on microstructure, dynamics, and rheology. While studies of the motion of a point particle in a cavity have been reported, the neglect of finite size sacrifices significant qualitative information, resulting in an incorrect coupling between torque and velocity, among others. We have derived new hydrodynamic mobility functions for finite-size particles confined by a spherical boundary that faithfully capture the physics of the boundary and its influence on particle dynamics. We obtain the full grand-mobility matrix and, from these, the position-dependent short-time self-diffusivity for an isolated particle and the dynamics of a hydrodynamically interacting pair suspended in the cavity. Both of these are studied over a range of particle-to-cavity size ratios. 1 This material is based upon work supported by the NSF GRFP under Grant No. DGE-0707428. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF. 4:58PM E2.00002 Effective viscosity of 2D suspensions - Confinement effects , PHILIPPE PEYLA, STEPHANE PRIEM, DOYEUX VINCENT, University Joseph Fourier, ALEXANDER FARUTIN, University Joseph Fourier - CNRS, MOURAD ISMAIL, University Joseph Fourier — We study the rheology of a sheared 2D suspension of non-Brownian disks in presence of walls. Although, it is of course possible today with modern computers and powerful algorithms to perform direct numerical simulations that fully account for multiparticle 3D interactions, the analysis of the simple case of a 2D suspension, provides valuable insights and helps to understand 3D results. For instance, we examine the role of particle-wall and particleparticle interactions in determining the rheology of confined sheared suspensions. In addition we evaluate the intrinsic viscosity as well as the contribution of hydrodynamic interactions to the dissipation as a function of a wide range of confinements. Thanks to the direct visualisation of the whole 2D Stokes flow, we are able to give a clear interpretation about the rheology of semi-dilute confined suspensions. 5:11PM E2.00003 Immersed collision of a sphere with a textured wall: from sticking to bouncing dynamics , THIBAULT CHASTEL, ANNE MONGRUEL, Physique et Mécanique des Milieux Hétérogène (UMR 7636 - CNRS - ESPCI - Université Pierre et Marie Curie - Université Paris-Diderot) — We investigate experimentally the dynamics of a sphere immersed in a viscous fluid and impacting a wall decorated with square micro-pillars. High frequency laser interferometry is used for measuring small displacements of the sphere with spatial resolution of 200 nm. For creeping flow, the classical lubrication force on the sphere is regularized by an effective slip length that can be correlated to the texture geometry. For Reynolds number of the order of 1 to 10, the sphere can either stick to or bounce off the wall. We show how the micro-textures affect the critical Stokes number for bouncing transition. A simple model using the slip length is presented to describe the near-wall dynamics of the sphere. 5:24PM E2.00004 Growth of clogs in microchannels , EMILIE DRESSAIRE, NYU Polytechnic School of Engineering, ALBAN SAURET, SVI laboratory and Princeton University , EMMANUEL VILLERMAUX, Aix Marseille Université, IRPHE, Marseille, France, HOWARD A. STONE, Princeton University — Porous membranes are used to detect and remove contaminants suspended in a fluid phase, e.g. to filter water. A typical filtration membrane allows the fluid to pass through but traps contaminants. Once a clog is formed in a pore, incoming particles aggregate upstream. This aggregate grows over time, which leads to a dramatic reduction of the flow rate. We consider a model that predicts the growth of the colloidal aggregate formed upon clogging of a microchannel. We present an analytical description to capture the time-evolution of the volume of the aggregate. We then focus on multiple parallel channels to model membrane filtration. In this situation, the growth dynamics of the aggregates are intrinsically coupled. The results of this modeling are compared with experimental data. Our work illustrates the critical influence of clogging events on the flow rate of porous membranes used in practical applications. 5:37PM E2.00005 Simulation study of suspension plugs in unsteady microchannel flows , AMANDA HOWARD, MARTIN MAXEY, FRANCIS CUI, ANUBHAV TRIPATHI, Brown University — The analysis or processing of particles in suspension may often involve a sample of finite length that is moving in a microchannel flow. Using numerical simulations, we examine the development of such a suspension plug of non-Brownian particles in unsteady, low Reynolds number shear flows in a microchannel. We focus on the early development in an oscillating Poiseuille flow, the distortion of the plug and the degree to which the motion is reversible relating this to prior work on oscillating suspension flows. For an initial particle volume fraction of 30% in the plug, the forward and then reversed flow leads to minimal net forward motion of the plug front at the centerline even after several oscillations. However a forward migration is seen near the walls. This net flux of particles is balanced by a flux of particles towards the wall within the plug. The exact response depends on the strain amplitude of the oscillation, the particle volume faction and other parameters of the flow. We are also able to examine the shear-driven particle fluxes at the tail of the plug. Both regions illustrate the effect of strong inhomogeneities in particle concentration on transport. We will relate the results to our recent experimental observations. 5:50PM E2.00006 Transport of a viscoelastic particle suspension in tortuous geometries , ALEXANDER BARBATI, GARETH MCKINLEY, Massachusetts Inst of Tech-MIT — Particle transport in viscoelastic fluids is of paramount importance in a variety of physical and industrial processes. We consider the transport of rigid particles through varied and microscale tortuous sections to model larger-scale particle transport, as commonly occurs in many reservoir stimulation processes. Beginning with the development of dynamic and geometric similarity parameters, we construct a rigid microfluidic device to probe the effects of fluid elasticity, fluid inertia, particle size, and particle volume fraction on particle transport. We characterize the microchannel flows with a combination of particle image velocimetry of embedded tracer particles and direct observation of particle accumulation and occlusion within the device. These on-chip experiments are accompanied by off-chip measurements of fluid rheology, and numerical computations of the flow field. Sunday, November 23, 2014 4:45PM - 6:03PM Session E3 Porous Media Flows III: Mixing and Transport — 3004 - Petr Denissenko, University of Warwick 4:45PM E3.00001 The Impact of Miscible Viscous Fingering on Mixing , JANE CHUI, PIETRO DE ANNA, RUBEN JUANES, MIT — Viscous fingering is a hydrodynamic instability that occurs when a less viscous fluid displaces a more viscous one. Instead of progressing as a uniform front, the less viscous fluid forms fingers that vary in size and shape to create complex patterns. The interface created from these patterns affects mixing between the two fluids, and therefore understanding how these patterns evolve in time is of critical importance in applications such as enhanced oil recovery and microfluidics. In this work, we focus on experimentally quantifying the impact of miscible viscous fingering on mixing. We use a radial Hele-Shaw cell as an analog of radial flows in porous media, and the local concentration field is measured temporally and spatially with the use of a fluorescein tracer. We first observe two distinct growth regimes in the evolution of the diffuse invading front: an initial regime of rapid growth due to the viscous fingering instability, and a latter regime of growth equivalent to a stable uniform displacement. We propose a scaling framework that predicts the time of transition between these two regimes, and subsequently the total length of the invading front. This framework will help to accurately determine the interface available for mixing when viscous fingering is observed. 4:58PM E3.00002 Statistics of admixture distribution in flows through rigid foams1 , PETR DENISSENKO, PIERRE LE FUR, JOZEF VLASKAMP, MARK WILLIAMS, Warwick University, XIAOLEI FAN, Manchester University, ALEXEI LAPKIN, Cambridge University — Diffusion and dispersion of admixture in flows through rigid foams need to be accounted for when modelling catalytic reactions on the foam surface. We study diffusion of admixture and scaling exponents of admixture concentration both experimentally and by numerical simulations. A liquid admixture was continuously released from a point source at the upstream boundary of a block of rigid SiC foam. Foam thickness was varied from 20 to 80 average pore sizes. A flow with Re of up to 300 based on the pore size was imposed by a progressive cavity pump. The distribution of the tracer at the exit from the foam was measured using LIF and the concentration moments have been calculated. Numerical simulation of the flow in laminar regime has been performed within OpenFoam for the Re from 1 to 100. Geometry of the sample was acquired by Micro Computed Tomography scanning of the actual foam sample. A steady-state SIMPLE method was used to solve the incompressible steady flow in the volume of 20x20x40 average pore sizes. Diffusion and dispersion of passive scalar has been studied by following individual streamlines. Results are interpreted in terms of mixing, heat transfer, and selectivity of catalytic reactions at the foam surface. 1 This Research is supported by European Commission under the 7th Framework Program 5:11PM E3.00003 Mixing-Scale Dependent Dispersion For Transport in Heterogeneous Flows , MARCO DENTZ, Spanish National Research Council (IDAEA-CSIC), Barcelona, Spain, FELIPE P.J. DE BARROS, University of Southern California (USC), Los Angeles, CA, USA — Dispersion quantifies the impact of microscale velocity fluctuations on the effective movement of particles and the evolution of scalar distributions in heterogeneous flows. It depends on the interaction between the velocity fluctuation scales and the scale on which the scalar is homogenized. The mixing, or coarse grained scale is the characteristic length below which the scalar is well mixed. It evolves in time as a result of dispersion and deformation of material fluid elements in the heterogeneous flow. We propose to use the mixing scale as a natural coarse graining scale for dispersion in heterogeneous flows. Using a stochastic modeling approach, we derive explicit expressions for the mixing-scale dependent dispersion coefficients and their variance. The fundamental mechanisms of local dispersion and compression of material fluid elements on evolving velocity scales determine the evolution of mixing-scale dependent dispersion and its self-averaging behavior. 5:24PM E3.00004 Direct Numerical Simulation of turbulent flow in a porous, face centered cubic cell1 , XIAOLIANG HE, SOURABH APTE, BRIAN WOOD, Oregon State University — DNS of flow through a 3D, periodic, face centered cubic (FCC) unit cell geometry at Re = 300, 550, and 950 based on diameter is performed. This low porosity arrangement of spheres is characterized by rapid flow expansions and contractions, and thus features an early onset to turbulence. The simulations are performed using a fictitious domain approach [Apte et al, J. Comp. Physics 2009], which uses non-body conforming Cartesian grids, with resolution up to D/∆ = 250 (3543 cells total). The results are used to investigate the structure of turbulence in the Eulerian and Lagrangian frames, the distribution and budget of turbulent kinetic energy, and the characteristics of the energy spectrum in complex packed beds and porous media. The porescale flow physics, which are important to properties such as bulk mixing performance and permeability, are investigated. Specifically, the data generated is being used to understand the important turbulence characteristics in low porosity packed beds of relevance for heat tranfer applications in chemical/nuclear reactors. 1 Funding: NSF project number 1336983. 5:37PM E3.00005 Modeling of nanoparticle transport and deposition in a porous medium: Effects of pore surface heterogeneity1 , NGOC PHAM, The University of Oklahoma, DIMITRIOS PAPAVASSILIOU, The University of Oklahoma & NSF — Pore surface charge heterogeneity has been found to affect particle retention in flow through porous media. In this study, retention of nanoparticles under different surface blocking conditions is numerically investigated. Micro-CT scanning is used to reconstruct the 3D geometry of sandstone and image-based analysis is used to characterize the pore space and the mineral composition of the rock. Flow of water through the sample is simulated with the lattice Boltzmann method. The motion of nanoparticles is modeled by injection of particles moving under convection and molecular diffusion and recording their trajectories in time [1,2]. When interacting with the pore surface, particles can be retained onto the surface with a particular deposition rate. As deposited particles hinder the retention of other particles by blocking occupied sites, the deposition is considered to be a second order process. Particle breakthrough under different modeled and real distributions of surface heterogeneity as a function of various surface blocking conditions is investigated. The effect is stronger when parts of the surface are much more favorable for deposition than others. [1] Voronov, R.S., VanGordon, S., Sikavitsas, V.I., Papavassiliou, D.V., Int. J. Num. Metods in Fluids, 67, 501, 2011 [2] Pham, N., Swatske, D.E., Harwell, J.H., Shiau, B.-J., Papavassiliou, D.V., Int. J. Heat & Mass Transf., 72, 319, 2014 1 Acknowledgements: Advanced Energy Consortium (AEC BEG08-022) & XSEDE (CTS090017). 5:50PM E3.00006 Advective-diffusive transport in microflows , PATRICK ANDERSON, MICHEL SPEETJENS, OLEKSANDR GORODETSKYI, Eindhoven University of Technology — Advective-diffusive transport in microflows is studied by means of the diffusive mapping method, a recent extension of the mapping method by Gorodetskyi et al. (Phys. Fluids 24, 2012) that includes molecular diffusion. This greatly expands the application area of the mapping technique and makes the powerful concepts of eigenmode decomposition and spectral analysis of scalar transport accessible to an important class of flows: inline micromixers with diffusion. The staggered herringbone micro-mixer is adopted as a prototypical three-dimensional micro mixer. Simulations with the diffusive mapping method are in close agreement with experimental observations in literature and expose a strong impact of diffusion on the transport. Diffusion enables crossing of Lagrangian transport barriers and thus smoothens concentration gradients and accelerates homogenization. Spectral analysis of the mapping matrix reveals this already occurs on a modal level in that individual eigenmodes progressively smoothen and spread out across transport barriers with stronger diffusion. Concurrently, the corresponding eigenvalues diminish and thus fundamentally alter the mixing process by invariably causing homogenization, irrespective of the Lagrangian flow structure. Sunday, November 23, 2014 4:45PM - 6:03PM Session E4 Bubbles: Rupture 3006 - James Bird, Boston University — 4:45PM E4.00001 The living times of bubbles at the interface1 , BENJAMIN CAMERON, LYDIA BOUROUIBA, Massachusetts Institute of Technology, NICOLAS VANDENBERGHE, EMMANUEL VILLERMAUX, IRPHE, Aix-Marseille Université — The lifetime of a water bubble at the surface of a pool prior to its burst remains an open question. It is known that the death of a bubble is initiated by the nucleation of a hole in its shell. However, the mechanisms governing the occurrence of such nucleation sites and prescribing the lifetime of bubbles remain unclear. Combining original visualizations, quantitative measurements of bubbles lifetimes and simple theoretical ideas, we report direct observations of the onset of the bursting process and rationalize the link between the rich interfacial events leading to the hole nucleation on the shell and the resulting robust bubble lifetimes distributions. These play a critical role in shaping the final size distribution of the droplets emitted. We will underline the consequences of the process in the sensible world, like air-sea water vapor exchanges. Bubbles bursting at the surface of water sources also allow for high levels of contamination and long-term exposure to a range of respiratory human pathogens and irritants indoors. Indeed, the droplets created by such bursts can contribute to the transfer of pathogens to the air, followed by their dispersal, thus bridging this subtle problem with unexpected new areas in health. 1 Thanks to the financial support of the MISTI-FRANCE MIT program. 4:58PM E4.00002 Jet drops from microbubble rupture , YINGXIAN YU, CASEY BARTLETT, JAMES BIRD, Boston University — When a bubble bursts at an interface, the surface energy creates an upward jet that can break into smaller droplets. Extensive research has demonstrated that the size of the droplets depends on the size of the initial bubbles. Yet this research has almost entirely been conducted for bubbles that are larger than 100 microns. As the bubbles approach 100 microns, the linear relation seems to deviate, although there have not been systematic experiments in this regime – mainly because these smaller bubbles and the even smaller droplets that they create have been difficult to visualize in the past. Here we directly measure the jetting phenomena for bubbles that are smaller than 100 micron using a combination of microfluidics and high-speed photography, and we relate our results to theory. Lab Name: Interfacial Fluidic Dynamics Laboratory Faculty Mentor’s Name: James C. Bird 5:11PM E4.00003 On the Physics of Fizziness: How Bubble Bursting Controls Droplets Ejection , THOMAS SEON, ELISABETH GHABACHE, ARNAUD ANTKOWIAK, CHRISTOPHE JOSSERAND, CNRS & UPMC - Institut d’Alembert — Either in a champagne glass or at the oceanic scales, the tiny bubbles rising at the surface burst in ejecting myriads of droplets. Focusing on the bubble bursting jet, prelude for these aerosols, we propose a simple scaling for the jet velocity, we unravel experimentally the intricate roles of bubble shape, capillary waves and liquid properties, and we demonstrate that droplets ejection can be tuned by changing the liquid properties. In particular, as capillary waves are shown to always evolve into a self-similar collapsing cavity, faster and smaller droplets can be produced by sheltering this collapse from remnant ripples using damping action of viscosity. These results pave the road to the characterization and control of the bursting bubble aerosols. Applications to champagne aroma diffusion will be discussed. 5:24PM E4.00004 On the Physics of Fizziness: How liquid properties control bursting bubble aerosol production? , ELISABETH GHABACHE, ARNAUD ANTKOWIAK, CHRISTOPHE JOSSERAND, THOMAS SEON, CNRS & UPMC Institut d’Alembert — Either in a champagne glass or at the oceanic scales, the tiny capillary bubbles rising at the surface burst in ejecting myriads of droplets. Focusing on the ejected droplets produced by a single bubble, we investigate experimentally how liquid properties and bubble size affect their characteristics: number, ejection velocities, sizes and ejection heights. These results allow us to finely tune the bursting bubble aerosol production. In the context of champagne industry, aerosols play a major role by spreading wine aroma above the glass. We demonstrate that this champagne fizz can be enhanced by selecting the wine viscosity and the bubble size, thanks to specially designed glass. 5:37PM E4.00005 Bursting the Taylor cone bubble , ZHAO PAN, TADD TRUSCOTT, Brigham Young University — A soap bubble fixed on a surface and placed in an electric field will take on the shape of a cone rather than constant curvature (dome) when the electrical field is not present. The phenomenon was introduced by J. Zeleny (1917) and studied extensively by C.T. Wilson & G.I. Taylor(1925). We revisit the Taylor cone problem by studying the deformation and bursting of soap bubbles in a point charge electric field. A single bubble takes on the shape of a cone in the electric field and a high-speed camera equipped with a micro-lens is used to observe the unsteady dynamics at the tip. Rupture occurs as a very small piece of the tip is torn away from the bubble toward the point charge. Based on experiments, a theoretical model is developed that predicts when rupture should occur. This study may help in the design of foam-removal techniques in engineering and provide a better understanding of an electrified air-liquid interface. 5:50PM E4.00006 Thin cylindrical sheets of air , SIGURDUR THORODDSEN, King Abdullah University of Science and Technology, DANIEL BEILHARZ, AXEL GUYON, Ecole Polytechnique, ER QIANG LI, MARIE-JEAN THORAVAL, King Abdullah University of Science and Technology — Drops impacting at low velocities onto a pool surface can stretch out thin hemispheric sheets of air. These air sheets can remain intact until they reach submicron thicknesses, whereby they rupture to form myriad of microbubbles. By impacting a higher-viscosity drop onto a lower-viscosity pool, we have explored new geometries of such air films. In this way we are able to maintain stable air-layers which can wrap around the entire drop to form anti-bubbles, i.e. spherical air layers bounded by inner and outer liquid masses. Furthermore, for the most viscous drops they enter the pool trailing a viscous thread from the pinch-off from the nozzle. The air sheet can also wrap around these treads and remain stable over extended time to form a cylindrical air sheet. We study the parameter regime where these structures appear and their subsequent breakup. The stability of these air cylinders is inconsistent with inviscid stability theory, suggesting stabilization by lubrication forces within the submicron air layer. Sunday, November 23, 2014 4:45PM - 6:03PM Session E6 Biofluids: Biofilms and Microenvironments — 3010 - Jiang Sheng 4:45PM E6.00001 Low Reynolds Number Biofilm Streamers Form as Highly Viscous Liquid Jets , ALOKE KUMAR, Department of Mechanical Engineering, University of Alberta, Edmonton, Alberta, Canada T6G 2G8, MAHTAB HASSANPOURFARD, Department of Chemical and Materials Engineering, University of Alberta, Edmonton, Alberta, Canada T6G 2G8, SIDDHARTHA DAS, Department of Mechanical Engineering, University of Maryland, College Park, MD-20742, USA — There are recent experimental investigations that suggest that in presence of low Reynolds number (Re ≪ 1) transport, preformed bacterial biofilms may deform into filamentous structures termed as streamers. Streamer formation time-scales usually far exceed reported rheological relaxation time scales for biofilms. Here we propose a theory that hypothesizes that streamers form due to the viscous response of the viscoelastic biofilms. The theoretical model is based on a stability analysis and can accurately explain hitherto unresolved issues, such as extremely large time needed for appearance of streamers and exponential growth of streamer dimensions after it has formed. We also provide results from our own initial experiments that indicate towards the validity of this “liquid-state” hypothesis. 4:58PM E6.00002 Bacterial adhesion and biofilm formation over a substrate with micro printed oily patches , MARYAM JALALI, JIAN SHENG, Texas Tech University — Over the past few years, there has been a significant focus on the processes involved in biodegradation of crude oil. In prior studies, using soft lithography and surface functionalization, we have fabricated solid substrates with micro-scale chemical patterns, and applied them to studying the bacteria-surface interactions as well as the formation of biofilm over these micro-patterned surfaces. A strong correlation between biofilm morphology and substrate patterns was found. In our current work we investigate the bacterial adhesion and biofilm formation of hydrocarbon degrading bacteria on micro printed oily surfaces with different micro-scale textures. The oily patterns were formed by contact printing of crude oil on a glass substrate with PDMS stamps. The oil patterned surface is additionally combined with a microfluidics as its bottom substrate. This unique lab-on-a-chip device allows us to investigate the complex interactions microscopically and over a long time. Additionally, it allows us to conduct experiments to elucidate the dynamic interactions such as swimming, dispersion, attachment, detachment, and adsorption between bacteria and micro printed oily surfaces under flow conditions in-situ. The growth rates and morphology of bacterial colony and biofilm are also studied and reported. 5:11PM E6.00003 Application of micro-PIV to the study of staphylococci bacteria biofilm dynamics , ERICA SHERMAN, University of Nebraska - Lincoln, DEREK MOORMEIER, KENNETH BAYLES, University of Nebraska Medical Center, TIMOTHY WEI, University of Nebraska - Lincoln — Staphylococci bacteria are recognized as the most frequent cause of biofilm-associated infections. A localized staph infection has the potential to enter the bloodstream and lead to serious infections such as endocarditis, pneumonia, or toxic shock syndrome. Changes in flow conditions, such as shear stress, can lead to stable biofilm growth or the dispersion of portions of the biofilm downstream. Exploration of biofilm physiology indicates a link between production of a specific enzyme called nuclease and biofilm architecture -; however the physical impact of this enzyme in directing the location and behavior of biofilm growth remains unclear. This talk investigates the link between sites of nuclease production and the development of biofilm tower structures using the application of micro-PIV and fluorescently labeled bacterial cells producing nuclease. Staphylococcus aureus bacteria were cultured in a BioFlux1000 square microchannel of a 65 by 65 um cross section, and subjected to a steady shear rate of 0.6 dynes. Micro-PIV and nuclease production measurements were taken to quantify the flow over a biofilm tower structure prior and during development. Data were recorded around the structure at a series of two dimensional planes, which when stacked vertically show a two dimensional flow field as a function of tower height. 5:24PM E6.00004 Effect of low Reynolds number flow on the quorum sensing behavior of sessile bacteria , FRANCOIS INGREMEAU, KEVIN KIM MINYOUNG, BONNIE BASSLER, HOWARD STONE, Princeton University, MECHANICAL AND AEROSPACE ENGINEERING, COMPLEX FLUIDS GROUP TEAM, MOLECULAR BIOLOGY LAB TEAM — Sessile and planktonic bacteria can be sensitive to the bacteria cell density around them through a chemical mediated communication called quorum sensing. When the quorum sensing molecules reach a certain value, the metabolism of the bacteria changes. Quorum sensing is usually studied in static conditions or in well mixed environments. However, bacteria biofilms can form in porous media or in the circulatory system of an infected body: quorum sensing in such flowing environment at low Reynolds number is not well studied. Using microfluidic devices, we observe how the flow of a pure media affects quorum sensing of bacteria attached to the wall. The biofilm formation is quantified by measuring the optical density in brightfield microscopy and the quorum sensing gene expression is observed through the fluorescence of a green fluorescent protein, which is a reporter for one of the quorum sensing genes. We measured without flow the amount of Staphylococcus aureus biofilm when the quorum sensing gene expression starts. In contrast, when the media is flowing in the microchannel, the quorum sensing expression is delayed. This effect can be understood and modelled by considering the diffusion of the quorum sensing molecules in the biofilm and their convection by the flowing media. 5:37PM E6.00005 Mixture Theory Study of Role of Growth Factor Gradients on Breast Cancer Chemotaxis , SREYASHI CHAKRABORTY, MARY SCHUFF, Mechanical Engineering, Purdue University, USA, ELIZABETH VOIGT, Mechanical Engineering, Virginia Tech University, USA, ERIC NAUMAN, Mechanical Engineering, Purdue University, USA, MARISSA RYLANDER, Mechanical Engineering, Virginia Tech University, USA, PAVLOS VLACHOS, Mechanical Engineering, Purdue University, USA — The transport of chemotactic agents is strongly influenced by variation in interstitial flows in different types of tissue. The mixture theory model of the fluid and solute transport in the microvasculature of tissues accounts for transport in the vessel lumen, vessel wall and the interstitial space separately. In the present study we use this model to develop a three dimensional geometry of the tumor microenvironment platform incorporating a physiological concentration of growth factor protein through blood flow in an extracellular collagen matrix. We quantify chemotaxis in response to solute gradients of varying magnitude formed by diffusion of proteins into the surrounding collagen. The numerical analysis delineates the dependence of hydraulic permeability coefficient on solute concentration. The preliminary results show the existence of a linear concentration gradient in the central plane between the micro-channels and a strong nonlinear gradient at the remaining parts of the system. 5:50PM E6.00006 Microbial transport through porous media: The importance of the microscale , PIETRO DE ANNA, YUTAKA YAWATA, ROMAN STOCKER, RUBEN JUANES, Massachusetts Inst of Tech-MIT — Bacteria play a key role in a plethora of subsurface processes, from geothermal energy, to enhanced oil recovery, to bioremediation. These large-scale consequences arise from microscale interactions within the highly heterogeneous subsurface environment. In particular, flow generates strong chemical gradients at the pore-scale and we hypothesized that, by actively responding to these microscale gradients, bacteria significantly change their transport properties at the macro-scale. We tested this hypothesis using video microscopy of Bacillus subtilis in microfluidic replica of porous media. We found that the bacteria’s motility and chemotaxis resulted in a two-fold increase in their ability to spread in the pore volumes, as a result of active migration out of micro-pockets of stagnant fluid. These findings illustrate that microscale flow heterogeneity has strong implications for the transport of biota through the subsurface, and thus likely for the biogeochemical processes they mediate. Sunday, November 23, 2014 4:45PM - 6:03PM Session E7 Fluids Education I — 3012 - Aline Cotel, University of Michigan 4:45PM E7.00001 Lost in Fathoms1 , ANAı̈S TONDEUR, JEAN-MARC CHOMAZ, LadHyx, CNRS-École Polytechnique, Palaiseau, France — In 2012, at the very point where two continents collided, the island of Nuuk disappeared without trace. At the same time, in Brisbane, the 34th International Geological Congress advanced a new era–the Anthropocene: an age where mankind has become a global telluric force. Was the disappearance of Nuuk island a one-off or a direct consequence of the emergence of the Anthropocene? This project was developed during a year of research as an artistin-residence at LadHyX and has evolved from an expedition of the emergent part of the Mid-Atlantic ridge and the region of deep oceanic water dive. This talk will present Lost In fathoms a narratives composed of installations, drawings and photographs by the means of which we investigate the causes involved in the disappearance of Nuuk island. It challenges the perception of oceanic and geologic time scales and human’s impact on the environment. This project is exhibited from October 16th to November 29th 2014, at GV Art Gallery in London, a contemporary art gallery devoted to art and science shows. 1 Acknowledgment: GV Art Gallery London, Chaire DDX École Polytechnique, LaSIPS Université Paris-Saclay 4:58PM E7.00002 Measuring Visual Expertise in Fluid Dynamics1 , JEAN HERTZBERG, TIM CURRAN, KATHERINE GOODMAN, University of Colorado, Boulder — What role does visual expertise play in the learning of abstract physics? In surveys for the Flow Visualization course at the University of Colorado, Boulder, students often commented positively about a new awareness of fluid dynamics in everyday life. Could this new awareness, termed visual expertise, be measured in some way? Working with research psychologists at CU Boulder, who had already been working in this area on projects such as face recognition, a study was developed. This study had subjects with no prior fluid dynamics expertise classify flow images as either turbulent or laminar. The first group was given error-driven learning ; that is, they had to guess the correct category for each image, were given feedback as to whether they had guessed correctly, and after a period of training, were tested on both the training images and a set of similar but new images. A second group was given simple instruction for the training images; that is, they were shown the image along with the name of the correct category, before being tested on both training images and new images. Preliminary results of the pilot study are presented, along with next steps. 1 This work is supported by NSF grant number 1240294. 5:11PM E7.00003 A tale of two slinkies: learning about scientific models in a student-driven classroom , PUNIT GANDHI, University of California, Berkeley, CALVIN BERGGREN, Texas Lutheran University, JESSE LIVEZEY, RYAN OLF, University of California, Berkeley — We describe a set of conceptual activities and hands-on experiments based around understanding the dynamics of a slinky that is hung vertically and released from rest. The motion, or lack thereof, of the bottom of the slinky after the top is dropped sparks students’ curiosity by challenging their expectations and provides context for learning about scientific model building. This curriculum helps students learn about the model building process by giving them an opportunity to enlist their collective intellectual and creative resources to develop and explore two different physical models of the falling slinky system. By engaging with two complementary models, students not only have the opportunity to understand an intriguing phenomenon from multiple perspectives, but also learn deeper lessons about the nature of scientific understanding, the role of physical models, and the experience of doing science. The activities we present were part of a curriculum developed for a week-long summer program for incoming freshmen as a part of the Compass Project at UC Berkeley, but could easily be implemented in a wide range of classrooms at the high school or introductory college level. 5:24PM E7.00004 Transparent 2-D converging-diverging nozzle for gas dynamics instruction , DELL OLMSTEAD, JON VIGIL, GREG NARANJO, C. RANDALL TRUMAN, University of New Mexico — A nozzle lab was created to combine qualitative and quantitative instruction of supersonic converging-diverging nozzle operation and design. The lab uses readily-available compressed nitrogen flowing through a 6.5mm square throat to produce exit Mach numbers up to 2.9. Several nozzles of different area ratio with transparent sidewalls can be quickly interchanged. Measured thrust, plenum pressure, plenum temperature, and exit pressure are displayed real-time and may be overlaid with data from other nozzle contours. Additionally, a Schlieren imaging system was used to observe shockwaves inside the nozzle and near its exit as plenum pressure was increased. Deviation between design and measured variables is around 3%. The correlation of Schlieren images of the exhaust with data from several different nozzles operated over the same total pressure range helps students understand not only how converging-diverging nozzles operate, but why they are used in some, but not all, propulsion applications. 5:37PM E7.00005 Bird Flight as a Model for a Course in Unsteady Aerodynamics , JAMEY JACOB, JONATHAN MITCHELL, MICHAEL PUOPOLO, Oklahoma State University — Traditional unsteady aerodynamics courses at the graduate level focus on theoretical formulations of oscillating airfoil behavior. Aerodynamics students with a vision for understanding bird-flight and small unmanned aircraft dynamics desire to move beyond traditional flow models towards new and creative ways of appreciating the motion of agile flight systems. High-speed videos are used to record kinematics of bird flight, particularly barred owls and red-shouldered hawks during perching maneuvers, and compared with model aircraft performing similar maneuvers. Development of a perching glider and associated control laws to model the dynamics are used as a class project. Observations are used to determine what different species and sizes of birds share in their methods to approach a perch under similar conditions. Using fundamental flight dynamics, simplified models capable of predicting position, attitude, and velocity of the flier are developed and compared with the observations. By comparing the measured data from the videos and predicted and measured motions from the glider models, it is hoped that the students gain a better understanding of the complexity of unsteady aerodynamics and aeronautics and an appreciation for the beauty of avian flight. 5:50PM E7.00006 Towards a global virtual community of female engineering students and professionals1 , ALINE COTEL, SARA RIMER, SAHITHYA REDDIVARI, Univ of Michigan - Ann Arbor — ct- The need for strategies to empower Liberian women is exemplified in the recent study carried out by ActionAid International, which examined the state of Liberian undergraduate women in urban areas. The results show that these women often face sexual intimidation by faculty and instructors, women are often excluded from student organizations, there exists a lack of institutional support for female organizations at the universities, and that the women do not feel safe in the university due to low security standards. The situation is even direr for the female engineering students with less than 5% of the engineering student population being women, therefore they are quite isolated in their engineering studies with minimal role models and professional support as they persist. We have planned a leadership camp for female Liberian engineering undergraduate women. The ultimate goal is to empower the Liberian women engineers with the skills, support and inspiration necessary to becoming successful engineering professionals. The leadership camp is planned and facilitated collaboratively by the members of the University of Michigan Society of Women Engineers (UM-SWE) student chapter and the Liberia Society of Women Engineers (L-SWE) student organization. The 2 week-long leadership camp has a workshop-based format with two themes: (i) academic and professional skills, and (ii) student organization development. 1 Funded by UM CRLT, IRWG, STEM Africa Sunday, November 23, 2014 4:45PM - 6:03PM — Session E9 Focus Session: The Impact of Andy Acrivos on Today’s Fluid Mechanics Science II 3014/3016 - Gary Leal, University of California, Santa Barbara 4:45PM E9.00001 Interfacial effects on droplet electrohydrodynamics: particle vortices, patchy membranes, and vesicle drums1 , PETIA VLAHOVSKA, Brown University — The analytical work by Acrivos group on drop dynamics in linear flows and rheology of dilute emulsions (papers with Frankel and Barthes-Biesel) have provided solid basis for more than 40 years of research on drops and capsules. These classical papers have inspired my research on drops with “complex” interfaces - surfactant-laden and particle-covered drops, and vesicles (drops encapsulated with lipid bilayer membranes). I will present some of our recent experimental observations on these systems in uniform DC and AC electric fields, where the coupling of the electric-field-induced flow and complex mechanics of the interface drives peculiar (and yet to be explained) behaviors: drum-like and asymmetric dumbbell shapes of vesicles; domains formation and motion in multicomponent membranes; particle assembly in dynamic vortices; drop kayaking. Possible implications of our findings to the design of patchy particles and electrorheology of emulsions will be discussed. 1 Supported by NSF-CBET 1132614. 4:58PM E9.00002 Binding and Unbinding of Vesicles and Capsules in Axisymmetric Flow , L. GARY LEAL, MARTIN KEH, University of California, Santa Barbara — Prof. Andreas Acrivos pioneered the use of scaling and asymptotic analysis, as well as the use of boundary integral methods, by chemical engineers in fluid flow and transport problems. These are skills that have been used by many of his former students in their own research. Here we consider the title problem using a combination of boundary-integral based numerical methods and scaling analysis to study the dynamics and mechanisms of adhesion and de-adhesion of vesicles at a solid boundary in the presence of flow. The adhesion process is dominated by drainage of the thin film down to a point where non-hydrodynamic attractive forces cause adherence. The unbinding process is dominated by peeling, though the final force to pull a vesicle from a solid surface is larger than expected due to lubrication effects. 5:11PM E9.00003 Discontinuous shear thickening and steady-state multiplicity in a granular suspension , MORTON DENN, Benjamin Levich Institute, City College of New York, HENRI DE CAGNY, ZHONGCHENG PAN, DANIEL BONN, Van der Waals-Zeeman Institute, Institute of Physics, University of Amsterdam, RYOHEI SETO, ROMAIN MARI, JEFFREY MORRIS, Benjamin Levich Institute, City College of New York — A concentrated suspension of neutrally buoyant non-Brownian spheres sheared between concentric cylinders with an inner radius-to-gap ratio of 0.037 undergoes discontinuous shear thickening under shear rate control but passes through an S-shaped viscosity curve with multiple states under stress control. This behavior is well described by simulation results that incorporate particle-particle frictional forces into the hydrodynamic description and lead to an analytical interpolation relation between low friction and high friction states that predicts an S-shaped viscosity curve. 5:24PM E9.00004 Force v. force-free motion in colloids , JOHN BRADY, California Institute of Technology — Consider a neutrally buoyant particle suspended in a fluid. Since the particle and fluid densities are the same there is no force on the particle and no motion. Now add a very large number of other particles to the fluid. These other particles are more dense than the fluid and so will settle due to gravity. Will the first particle move? In which direction and how fast? And viewed in a frame moving with the first particle what is the velocity disturbance caused by this force-free particle? These questions arise in the context of phoretic motion in colloidal dispersions. 5:37PM E9.00005 Global symmetry relations in linear and viscoplastic mobility problems , KEN KAMRIN, MIT, JOE GODDARD1 , UC San Diego — The mobility tensor of a textured surface is a homogenized effective boundary condition that describes the effective slip of a fluid adjacent to the surface in terms of an applied shear traction far above the surface. In the Newtonian fluid case, perturbation analysis yields a mobility tensor formula, which suggests that regardless of the surface texture (i.e. nonuniform hydrophobicity distribution and/or height fluctuations) the mobility tensor is always symmetric. This conjecture is verified using a Lorentz reciprocity argument. It motivates the question of whether such symmetries would arise for nonlinear constitutive relations and boundary conditions, where the mobility tensor is not a constant but a function of the applied stress. We show that in the case of a strongly dissipative nonlinear constitutive relation — one whose strain-rate relates to the stress solely through a scalar Edelen potential — and strongly dissipative surface boundary conditions — one whose hydrophobic character is described by a potential relating slip to traction — the mobility function of the surface also maintains tensorial symmetry. By extension, the same variational arguments can be applied in problems such as the permeability tensor for viscoplastic flow through porous media, and we find that similar symmetries arise. These findings could be used to simplify the characterization of viscoplastic drag in various anisotropic media. 1 (Joe Goddard is a former graduate student of Acrivos) 5:50PM E9.00006 Effective reaction rate for porous surfaces under strong shear: Beyond Damkohler , ERIC S.G. SHAQFEH, PREYAS SHAH, Stanford Univ — Traditonally, surface reactive porous media are modeled via an effective reac- tion/mass transfer rate based on the original ansatz of Damkohler, i.e, reaction limited transport at the microscale in the absence of flow. We are interested in modeling the microscale mass transfer to porous surfaces occuring in leaky tumor vasculature, where the Damkohler number can be O(1) and the Peclet number may be large. We model it as a uniform bath of a species in unbound shear flow over a wall with first order reactive circular patches (pores). We analyze the flux through a single pore using both analytic and boundary element simulations and observe the formation of a 3-D depletion region (wake) downstream of the pore. Wake sharing between adjacent pores in a multibody setting such as 2 pores aligned in the shear direction leads to a smaller flux per pore. Obtaining this interaction length scale and using the renormalized periodic Green’s function, we study the flux through a periodic and disordered distribution of pores. This flux appears as the reaction rate in an effective boundary condition, valid up to non-dilute pore area fractions, and applicable at a wall-normal effective slip distance. It replaces the details of the surface and can be used directly in large scale physics simulations. Sunday, November 23, 2014 4:45PM - 6:03PM Session E10 Microscale Flows: Formation and Dynamics of Drops — 3005 - Sindy Tang, Stanford University 4:45PM E10.00001 Confined microbubbles at high capillary numbers1 , MARTIN SAUZADE, THOMAS CUBAUD, Stony Brook University — We experimentally investigate the flow behavior of bubbles in highly viscous silicone oils within various microgeometries. A square focusing section is used to examine the bubble generation process at large capillary numbers. We notably vary the continuous phase viscosity from 1 to 10,000 cS and study the dynamics of interfacial cusps during bubble pinch-off. The resulting segmented flows are then scrutinized in straight microchannels for both dissolving and non-dissolving bubbles. Finally, we examine the motion of bubbles in periodically constricted microchannels over a wide range of flow conditions. Our findings highlight the possibility to control and exploit the interplay between capillary and mass transfer phenomena with highly viscous fluids in microsystems. 1 This work is supported by NSF (CBET- 1150389) 4:58PM E10.00002 Flow regime and dynamic critical pressure of droplet entering confined microchannel , ZHIFENG ZHANG, Mechanical Engineering, Washington State University, 14204 NE Salmon Creek Ave, Vancouver, Washington 98686, USA, JIE XU, Department of Mechanical & Industrial Engineering, University of Illinois at Chicago, Chicago, Illinois 60607, XIAOLIN CHEN, Mechanical Engineering, Washington State University, 14204 NE Salmon Creek Ave, Vancouver, Washington 98686, USA, CAE LAB TEAM, XU GROUP: MICROFLUIDICS LAB TEAM — Droplet entering a microchannel contraction is a phenomenon widely seen in many applications, such as two phase separation, droplet microfluidics, size-based cancer cell detection, and micro combustors. Better understanding of the droplet flow regime as well as the droplet deformation pattern is crucial to device design in aforementioned applications. In present numerical study, we explore the transient behavior of droplet being squeezed into a micro-contraction and we report different flow regimes observed according to the blockage status of the droplet in the entrance of the contraction. We then quantify the relation between droplet deformation, back pressure and the flow velocity in the channel. In the end, we explore the effects of channel geometry by comparing the results in three different shapes of channels (circular square and rectangular). 5:11PM E10.00003 Picoliter Droplets of Controlled Composition for SERS Studies , CHRYSAFIS ANDREOU, MARTIN MOSKOVITS, CARL MEINHART, UCSB — Generation of picoliter scale droplets of specified composition is a valuable tool for chemical analysis and synthesis, with many bioanalytical lab-on-chip applications. Here we present a microfluidic droplet generator that creates water-in-oil droplets of picoliter volume, by merging two laminar aqueous streams immediately before the droplet-generating junction. We use this device to investigate the dependence of surface enhanced Raman spectroscopic signal on the availability of silver nanoparticles, and analyte. By controlling the composition of the generated droplets, we can limit the availability of plasmonic nanoparticles and analyte molecules with unprecedented accuracy. Experiments, as well as numerical simulations, are used to investigate the plasmonic enhancement stemming from small numbers of silver nanoparticles confined within the droplets. 5:24PM E10.00004 Exploiting droplet formation in microfluidic devices to create functional particles1 , EMILIA NOWAK, MARK SIMMONS, University of Birmingham — Microfluidic devices offer excellent capabilities for the formation of microstructured particles which have functional attributes e.g. in controlled delivery of pharmaceuticals, enhanced nutrition and flavours in food. In this work, a microfluidic device is employed to form microstructured particles in two steps: (i) by formation of single/double emulsions and (ii) solidification of the droplet by either gelation or solvent evaporation. Both may impart non-Newtonian properties to the component phases. The influence of phase flow rates (capillary number), surfactant type/concentration and the rheology of the component phases upon the particle formation and hydrodynamic behaviour are described. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 5:37PM E10.00005 Dynamics of jet breakup induced by perturbation , HO CHEUNG SHUM, JINGMEI LI, SZE YI MAK, University of Hong Kong — We study the breakup of jet to form droplets, as induced by controlled perturbation, in a microchannel. Controlled mechanical perturbation is introduced to the tubing through which the jet phase is injected into the device, which is monitored under high-speed optical imaging. We measure the frequency of droplet formation and the sizes of the droplets as the frequency and amplitude of the perturbation is varied. Droplets can be induced to form at the perturbation frequency only above a critical frequency and amplitude. In this manner, the droplet size can be precisely controlled. The amplitude needed to induce breakup decreases as the interfacial tension of the system is lowered. Moreover, by selectively varying the wettability of the inner wall of the channel, double emulsion droplets can be generated in one step by applying large-amplitude perturbation of the jet phase. Our work demonstrates the potential of using controlled perturbation to generate droplets with tunable size and shapes, with implications on new designs of liquid dispensing nozzles. 5:50PM E10.00006 Dynamics of soft interfaces in droplet-based microfluidics , QUENTIN BROSSEAU, Brown University, INGMAR POLENZ, Max Planck Institute for dynamics and Self-Organization, JEAN-CHRISTOPHE BARET, CRPP Unniversité de Bordeaux — The numerous applications of microfluidic emulsions (i.e compartmentalisation) rely on both the short and long term stabilization of droplet interfaces. This is achieved mainly via the addition of surfactant or the formation of a rigid capsule. Therefore in order to predict and control the stability of the emulsions, a precise control of the interfaces modified in this way is required. In this talk, I present a microfluidic method for the characterisation of the dynamic properties of complex soft interfaces. Monodisperse droplets with controlled interfaces are produced at a flow focusing junction and deformed in sudden planar expansions. The deformation of the droplets is then analysed at different interface ages, allowing us to follow the dynamics of two processes occurring at the interface. First, the evolution of the interfacial tension, from which we extract the surfactant adsorption kinetics for small time and length scales. Secondly, the in-situ kinetics of a interfacial polymerization reaction permitting us to determine the mechanical properties of the resulting polymer shell and encapsulated droplet. Sunday, November 23, 2014 4:45PM - 6:03PM Session E11 General Fluid Dynamics I — 3007 - Jerry Westerweel, Delft University of Technology 4:45PM E11.00001 ABSTRACT WITHDRAWN — 4:58PM E11.00002 Mechanics of Flow over Wrinkled Surfaces , SHABNAM RAAYAI, GARETH MCKINLEY, Massachusetts Inst of Tech-MIT, MARY BOYCE, Columbia University — The surfaces of many plants and animals are covered with a variety of microtextures such as ribs or 3D tubules which can control surface-mediated properties such as skin friction. Inspired by the drag reducing ability of these natural structures, passive drag reduction methods such as microfabricated riblet surfaces have been developed. Microgroove textures on the surface of objects such as hulls and wings which are aligned in the free-stream direction have been shown to reduce drag by 4-8% in flows with zero or mild pressure gradients [1]. We introduce sinusoidal wrinkles as model ribbed surfaces with drag reducing capabilities. Surface wrinkling arises spontaneously as the result of mismatched deformation of a thin stiff coating bound to a thick soft substrate and can be designed over a wide range of length scales. Using numerical modeling we show that wrinkled surfaces can substantially reduce the skin friction coefficient in high Reynolds number laminar flow. We show that this reduction is a result of purely viscous mechanisms through a geometry-mediated increase in the thickness of the boundary layer and retardation of the flow in the interstitial grooves of the textured surfaces. [1] D. Bechert et al., Experiments in Fluids 28 (2000) 5:11PM E11.00003 Effective Slip and Drag Reduction in Transitional Flow over Superhydrophobic Surfaces , MARGARET HECK, The University of Oklahoma, DIMITRIOS PAPAVASSILIOU, The University of Oklahoma & National Science Foundation — Superhydrophobic surfaces (SHS) have recently attracted attention as a passive technique for reducing drag in both laminar and turbulent flows [1,2]. These surfaces result in a reduced contact area between a liquid and a solid by trapping air between roughness elements [1,3]. The liquid then glides over the pockets of air between the roughness elements, exhibiting hydrodynamic slip [1]. Most studies considering systems involving SHS focus on laminar or low Reynolds number (Re) turbulent flow. Turbulence closure models may be useful when considering large Re flows and complicated surface topologies of the SHS. This study explores the behavior of the effective slip length and the drag reduction for Newtonian fluids flowing over SHS in the transitional flow regime using computations and both laminar and traditional turbulence models. For Re in the range between 1,000 and 5,000, slip length is observed to differ from that for purely laminar or fully developed turbulent flow. The values obtained using popular models for turbulence, such as the standard k-e and the standard k-w models, are compared, and a possible explanation for the observed variance is explored using predictive models, such as that proposed by Fukagata et al. [4]. [1] Rothstein, JP, Annu. Rev. Fluid Mech., 42, 89, 2010; [2] Voronov, R., Papavassiliou, D.V., and L.L. Lee, Ind. Eng. Chem. Res., 47(8), 2455, 2008; [3] Heck, M.L., and D.V. Papavassiliou, Chem. Eng. Com., 200, 919-934 (2013); [4] Fukagata, K., N. Kasagi, P. Koumoutsakos,. Phys. Fluids, 18, 051703 (2006). 5:24PM E11.00004 Direct numerical simulation of flow past superhydrophobic surfaces , PAOLO LUCHINI, Universita‘ di Salerno - DIIN, ALESSANDRO BOTTARO, Universita‘ di Genova - DICCA — Superhydrophobic surfaces trap a discontinuous air layer through their texture which, in addition to changing the apparent contact angle of water drops, also changes the friction coefficient of a continuous water flow. Locally this effect can be represented through a slip coefficient (e.g. Lauga & Stone, J. Fluid Mech. 489, 55–77, 2003), or equivalently through an effective displacement of the wall by a distance (different for each different velocity component) comparable to the spacing of the texture. For this reason they are being considered for drag reduction in turbulent flow, more sensitive to this displacement than laminar flow for its intrisic small features. Since the upper limit on texture size imposed by the destruction of the surface-tension-bound air layer eventually constrains the reduction available, to quantify the effect accurately is essential. In its simplest representation, the superhydrophobic surface may be assumed to be flat and composed of alternating patches of no-slip and free-slip wall. Here direct numerical simulations will be presented of turbulent flow past such a surface, and their results compared with those produced by the corresponding effective wall displacement. 5:37PM E11.00005 Properties of the Mean Momentum Balance in Polymer Drag Reduced Channel Flow , CHRISTOPHER WHITE, University of New Hampshire, YVES DUBIEF, University of Vermont, JOSEPH KLEWICKI, University of New Hampshire and University of Melbourne — The redistribution of mean momentum and the underlying mechanisms of the redistribution process in polymer drag reduced channel flow are investigated by employing a mean momentum equation based analysis. The work is motivated by recent studies that showed (contrary to long-held views) that polymers modify the von Karman coefficient, κ, at low drag reduction, and at some relatively high drag reduction eradicate the inertially dominated logarithmic region. Since κ is a manifestation of the underlying dynamical behaviors of wall-bounded flow, understanding how polymers modify κ is inherently important to understanding the dynamics of polymer drag reduced flow, and, consequently, the phenomenon of polymer drag reduction. The goal of the present study is to explore and quantify these effects within the framework of a mean momentum based analysis. 5:50PM E11.00006 Why bigger may in fact be better... in the context of table tennis , TADD TRUSCOTT, ZHAO PAN, Brigham Young University, JESSE BELDEN, Naval Undersea Warfare Center — We submit that table tennis is too fast. Because of the high ball velocities relative to the small table size, players are required to act extremely quickly, often exceeding the limits of human reaction time. Additionally, the Magnus effect resulting from large rotation rates introduces dramatically curved paths and causes rapid direction changes after striking the table or paddle, which effectively reduces reaction time further. Moreover, watching a professional game is often uninteresting and even tiring because the ball is moving too quickly to follow with the naked eye and the action of the players is too subtle to resolve from a distance. These facts isolate table tennis from our quantitatively defined “fun game club,” and make it less widely appealing than sports like baseball and soccer. Over the past 100 years, the rules of table tennis have changed several times in an effort to make the game more attractive to players and spectators alike, but the game continues to lose popularity. Here, we experimentally quantify the historic landmark equipment changes of table tennis from a fluid dynamics perspective. Based on theory and observation, we suggest a larger diameter ball for table tennis to make the game more appealing to both spectators and amateur players. Sunday, November 23, 2014 4:45PM - 6:03PM Session E12 Drops: Sessile and Static Surface Interactions — 3018 - Pirouz Kavehpour, University of California, Los Angeles 4:45PM E12.00001 Optimization of Aggregation Kinetics of SERS-active Nanoparticles in Evaporating Sessile Droplets , MEYSAM BARMI, CHRYSAFIS ANDREOU, MEHRAN HOONEJANI, MARTIN MOSKOVITS, CARL MEINHART, University of California, Santa Barbara — We studied evaporating sessile droplets containing silver nanoparticles as a platform for chemical detection using Surface-Enhanced Raman Spectroscopy (SERS), yielding molecule-specific Raman spectra. Controlling the degree of aggregation of nanoparticles is the key to the signal enhancement. Therefore, the aggregation kinetics of nanoparticles in droplets was investigated both experimentally and numerically to determine the evaporation and aggregation parameters leading to optimal detection. The signal depends on the degree of aggregation, which is affected by the initial concentration of nanoparticles (cNP), the dimerization rate (k), and the droplet evaporation time (τ evap). We introduced the aggregation parameter Γa ≡ τ evap /τ a, which is the ratio of the evaporation to the aggregation time scales. We found the different aggregation regimes based on aggregation parameter and stirring level within the droplet. For a well-stirred droplet, the optimal condition for SERS detection was found to be Γa,opt = k cNP τ evap ≈ 0.3. The intensity of the SERS signal is near maximal in a wide range of aggregation parameters between 0.05 and 1.25 defining the time window during which trace analytes can be measured. 4:58PM E12.00002 Design of a Condensation-Based Contact Angle Goniometer1 , AJAY ROOPESH, VIRAJ DAMLE, KONRAD RYKACZEWSKI, Arizona State University — Condensation of low surface tension fluids such as refrigerants, natural gas, and carbon dioxide is important to a variety of industrial processes. Condensation of these fluids often occurs at elevated pressures and/or cryogenic temperatures, making measurement of their wetting properties using standard approaches challenging. It was recently demonstrated that these properties are critical in designing omniphobic surfaces for low surface tension fluid condensation rate enhancement [1]. To this end, we have developed an alternative goniometer design capable of contact angle measurement at wide pressure and temperature range. In this design, droplets are not dispensed through a pipette but generated through localized condensation on a tip of a preferentially cooled small metal wire encapsulated within a thick thermal insulator layer. Here we present a computational and an experimental study of the relation between the condensation-based goniometer geometry, subcooling, and droplet generation rate. We also compare water contact angle measurements using standard and condensation-based goniometer. [1] Rykaczewski et al., Sci. Rep., 4, 2014. 1 KR acknowledges startup funding from ASU. 5:11PM E12.00003 The bifurcation diagram of drops in a sphere/ plane geometry: influence of contact angle hysteresis , RIËLLE DE RUITER, MATHIJS VAN GORCUM, MESA+ Institute for Nanotechnology, University of Twente, CIRO SEMPREBON, Max-Planck-Institute for Dynamics and Self-Organization, MICHÈL DUITS, MESA+ Institute for Nanotechnology, University of Twente, MARTIN BRINKMANN, Max-Planck-Institute for Dynamics and Self-Organization / Saarland University, FRIEDER MUGELE, MESA+ Institute for Nanotechnology, University of Twente — We study liquid drops that are present in a generic geometry, namely the gap in between a sphere and a plane. For the ideal system without contact angle hysteresis, the drop position is solely dependent on the contact angle, drop volume, and sphere/ plane separation distance. Performing a geometric analysis and Surface Evolver calculations, a continuous and fully reversible transition between axisymmetric non-spherical shapes and non-axisymmetric spherical shapes is predicted. We also study these transitions experimentally, varying the contact angle using electrowetting. Then, pinning forces drastically alter the pitchfork bifurcation as the unstable branch gets stabilized, and introduce a history-dependence in the system. As a consequence, the outward movement of drops following pinning can be either continuous or discontinuous, depending on the minimum contact angle that is attained. 5:24PM E12.00004 A new approach to stability and oscillations of constrained drops and capillary bridges , DAVID FABRE, VERONIQUE CHIREUX, FREDERIC RISSO, PHILIPPE TORDJEMAN, IMFT, University of Toulouse — Static equilibria of liquid inclusions under the effect of gravity and capillarity is a large class of situations which encompasses drops hanging from a ceiling or from a capillary, sessile drops, liquid bridges, etc... In such equilibria the surface shape is governed by the Yong-Laplace equation, which is usually solved in a local way using a “shooting” method. We introduce a new method which solves the Laplace-Young in a global way, using an iterative deformation of the shape towards the equilibrium shape. The method is easy to implement and versatile, and allows to prescribe constraints such as the volume of liquid, the angle of attachment, etc... We subsequently consider the issue of stability and oscillations of such configurations. Using finite elements and considering small-amplitude displacements of the surface with respect to the static configuration previously computed, we introduce a global stability approach which allows to predict the stability limits, the oscillation frequencies and the eigenmode shapes for quite general geometries. The approach will be illustrated and compared with experiments in two situations, namely a drop attached to a capilary and a liquid bridge resulting from the coalescence of two facing millimetric drops. 5:37PM E12.00005 Hysteresis in the surface topography of laterally confined fluids1 , JUAN GOMBA, Univ. Nacional del Centro- CIFICEN - Argentina, CARLOS PERAZZO, University Favaloro - Argentina, JONATAN RAÚL MAC INTYRE, Univ. Nacional del Centro- CIFICEN - Argentina — Possible steady states of a fluid within a vessel of finite width are theoretically studied. The fluid is under the action of surface tension, gravity and molecular interaction with the substrate. At steady state, the liquid surface can assume only two possible forms: a film of uniform thickness or a droplet surrounded by a uniform film. Depending on the volume of the liquid and the width of the containing vessel, the system will adopt one of the two steady states. However, there is a range of parameters for which both states can be reached either. In this case the system can present hysteresis, ie the final state may depend on its previous history. 1 Authors thank to CONICET and ANPCYT 5:50PM E12.00006 Asymptotic Expansion of the Axisymmetric Linear-Elastic Shell Equations with Application to Determining the Elastic Moduli of Ultra-Thin Shells , MARTIN NEMER, CARLTON BROOKS, Sandia National Laboratories — An asymptotic expansion of the axisymmetric linear-elastic shell equations is presented in the limit that h/r → 0 where h is the shell thickness and r is the characteristic radius of curvature. This solution is obtained using a WKB expansion, which doesn’t rely on the shell being close to spherical, allows for turning points in curvature, and can be extended to include higher-order terms. The obtained solution is used to analyze experimental results for obtaining elastic moduli of ultra-thin shells of metal oxides on molten-metal drops. Sunday, November 23, 2014 4:45PM - 5:50PM Session E13 Drops: Evaporation and Condensation — 3020 - Mandre Shreyas, Brown University 4:45PM E13.00001 Dynamics of evaporative colloidal patterning1 , L. MAHADEVAN, C. NADIR KAPLAN, Harvard University, NING WU, Colorado School of Mines, SHREYAS MANDRE, Brown University, JOANNA AIZENBERG, Harvard University — Evaporating suspensions of colloidal particles lead to the formation of a variety of patterns, ranging from rings left behind a coffee drop to periodic bands or uniform solid films deposited on a substrate suspended vertically in a container of the colloidal solution. To characterize the transition between different types of patterns, we develop minimal models of the liquid meniscus deformation due to the evaporation and colloidal deposition. A complementary multiphase model allows us to investigate the detailed dynamics of patterning in a drying solvent. This approach couples the inhomogeneous evaporation at the evolving liquid-air interface to the dynamics inside the suspension, i.e. the liquid flow, local variations of the particle concentration, and the propagation of the deposition front where the solute forms a wet, incompressible porous medium at high concentrations. The results of our theory are in good agreement with direct observations. 1 This research was supported by the Air Force Office of Scientific Research (AFOSR) under Award FA9550-09-1-0669-DOD35CAP and the Kavli Institute for Bionano Science and Technology at Harvard University. 4:58PM E13.00002 Leidenfrost drops and micro-particles: organization and evaporation , LAURENT MAQUET, Université de Liège, PIERRE COLINET, Université Libre de Bruxelles, STÉPHANE DORBOLO, Université de Liège — We investigate the behavior of hydrophilic microparticles dropped into Leidenfrost drops. These particles appears to go through the drop until they reach the bottom surface of the drop where they are dewetted. Due to the evaporation of the drop, the surface of the drop decreases. Thus, the particles that are trapped at the surface of the drop due to the dewetting begin to cover more and more the drop. At a point, they even cover the whole surface of the drop. The superficial density of the particles at the surface is ∼ 0.8 and the fraction of the beads that stay trapped at the surface until the cover is complete is always larger than 0.7. We measured evaporation rates and compared the case of drops with and without particles. These evaporation rates are always decreased by the presence of the particles. This is due to the dewetting. Indeed, the effective surface of evaporation is decreased by the presence of particles at the surface. Thus, knowing how the evaporation is affected by the presence of the particles, we can measure contact angles at the lower surface of these levitating drops. 5:11PM E13.00003 A phase-field point-particle model for particle-laden interfaces , CHUAN GU, LORENZO BOTTO, School of Engineering and Materials Science, Queen Mary University of London — The irreversible attachment of solid particles to fluid interfaces is exploited in a variety of applications, such as froth flotation and Pickering emulsions. Critical in these applications is to predict particle transport in and near the interface, and the two-way coupling between the particles and the interface. While it is now possible to carry out particle-resolved simulations of these systems, simulating relatively large systems with many particles remains challenging. We present validation studies and preliminary results for a hybrid Eulerian-Lagrangian simulation method, in which the dynamics of the interface is fully-resolved by a phase-field approach, while the particles are treated in the “point-particle” approximation. With this method, which represents a compromise between the competing needs of resolving particle and interface scale phenomena, we are able to simulate the adsorption of a large number of particles in the interface of drops, and particle-interface interactions during the spinodal coarsening of a multiphase system. While this method models the adsorption phenomenon efficiently and with reasonable accuracy, it still requires understanding subtle issues related to the modelling of hydrodynamic and capillary forces for particles in contact with interface. 5:24PM E13.00004 Condensation Dynamics on Mimicked Metal Matrix Hydrophobic Nanoparticle-Composites1 , VIRAJ DAMLE, XIAODA SUN, KONRAD RYKACZEWSKI, Arizona State University — Use of hydrophobic surfaces promotes condensation in the dropwise mode, which is significantly more efficient than the common filmwise mode. However, limited longevity of hydrophobic surface modifiers has prevented their wide spread use in industry. Recently, metal matrix composites (MMCs) having microscale hydrophobic heterogeneities dispersed in hydrophilic metal matrix have been proposed as durable and self-healing alternative to hydrophobic surface coatings interacting with deposited water droplets [1]. While dispersion of hydrophobic microparticles in MMC is likely to lead to surface flooding during condensation, the effect of dispersion of hydrophobic nanoparticles (HNPs) with size comparable to water nuclei critical radii and spacing is not obvious. To this end, we fabricated highly ordered arrays of Teflon nanospheres on silicon substrates that mimic the top surface of the MMCs with dispersed HNPs. We used light and electron microscopy to observe breath figures resulting from condensation on these surfaces at varied degrees of subcooling. Here, we discuss the relation between the droplet size distribution, Teflon nanosphere diameter and spacing, and condensation mode. [1] Nosonovsky, M. et al. Langmuir 27 (2011). 1 KR acknowledges startup funding from ASU. 5:37PM E13.00005 Parameterization of the scavenging coefficient for particle scavenging by drops , STEVEN FREDERICKS, J.R. SAYLOR, Clemson University — The removal of particles by drops occurs in many environmentally relevant scenarios such as particle fallout from rain, as well as in many industrial applications such as sprays for dust control in mines. In applications like these the ability of a drop to scavenge a particle is quantified by the scavenging coefficient, E, which is the fraction of particles removed. Though the physics controlling particle scavenging by drops suggests that E is controlled by several dimensionless groups, E is typically correlated to just the Stokes number. A survey of published experimental data shows significant scatter in plots of E versus the Stokes number, occasionally exceeding three orders of magnitude. There is also a large discrepancy between the published theories for E. A parameterization study was conducted to ascertain if and how inclusion of other dimensionless groups could better collapse the extant data for E and the results of that study are presented in this talk. Brief mention will also be made of recent experiments by the authors where E was measured for a liquid drop suspended in an ultrasonic standing wave field, where the drop diameter and gas velocity can be independently varied unlike the more typical experiments where these quantities are coupled. Sunday, November 23, 2014 4:45PM - 6:03PM Session E15 Focus Session: Respiratory Bio-Fluid Dynamics II — 3022/3024 - Josue Sznitman, Technion - Israel Institute of Technology 4:45PM E15.00001 Correlations of Flow Structure and Particle Deposition with Structural Alterations in Severe Asthmatic Lungs1 , SANGHUN CHOI, SHINJIRO MIYAWAKI, JIWOONG CHOI, ERIC A. HOFFMAN, Univ of Iowa, SALLY WENZEL, Univ of Pittsburgh, CHING-LONG LIN, Univ of Iowa — Severe asthmatics are characterized by alterations of bifurcation angle, hydraulic diameter, circularity of the airways, and local shift of air-volume functional change. The characteristics altered against healthy human subjects can affect flow structure and particle deposition. A large-eddy-simulation (LES) model for transitional and turbulent flows is utilized to study flow characteristics and particle deposition with representative healthy and severe asthmatic lungs. For the subject-specific boundary condition, local air-volume changes are derived with two computed tomography images at inspiration and expiration. Particle transport simulations are performed on LES-predicted flow fields. In severe asthmatics, the elevated air-volume changes of apical lung regions affect the increased particle distribution toward upper lobes, especially for small particles. The constricted airways are significantly correlated with high wall shear stress, leading to the increased pressure drop and particle deposition. The structural alterations of bifurcation angle, circularity and hydraulic diameter in severe asthmatics are associated with the increase of particle deposition, wall shear stress and wall thickness. 1 NIH grants: U01-HL114494, R01-HL094315 and S10-RR022421. Computer time: XSEDE. 4:58PM E15.00002 Perception of Better Nasal Patency Correlates with Increased Mucosal Cooling after Surgery for Nasal Obstruction , GUILHERME GARCIA, CORBIN SULLIVAN, Medical College of Wisconsin, DENNIS FRANK-ITO, Duke University, JULIA KIMBELL, University of North Carolina at Chapel Hill, JOHN RHEE, Medical College of Wisconsin — Nasal airway obstruction (NAO) is a common health problem with 340,000 patients undergoing surgery annually in the United States. Traditionally, otolaryngologists have focused on airspace cross-sectional areas and nasal resistance to airflow as objective measures of nasal patency, but neither of these variables correlated consistently with patients’ symptoms. Given that the sensation of nasal airflow is also associated with mucosal cooling (i.e., heat loss) during inspiration, we investigated the correlation between the sensation of nasal obstruction and mucosal cooling in 10 patients before and after NAO surgery. Three-dimensional models of the nasal anatomy were created based on pre- and post-surgery computed tomography scans. Computational fluid dynamics (CFD) simulations were conducted to quantify nasal resistance and mucosal cooling. Patient-reported symptoms were measured by a visual analog scale and the Nasal Obstruction Symptom Evaluation (NOSE), a disease-specific quality of life questionnaire. Our results revealed that the subjective sensation of nasal obstruction correlated with both nasal resistance and heat loss, but the strongest correlation was between the NOSE score and the nasal surface area where heat flux exceeds 50 W/m2. In conclusion, a significant post-operative increase in mucosal cooling correlates well with patients’ perception of better nasal patency after NAO surgery. 5:11PM E15.00003 4DCT-based assessment of regional airflow distribution in healthy human lungs during tidal breathing1 , JIWOONG CHOI, NARIMAN JAHANI, SANGHUN CHOI, ERIC HOFFMAN, CHING-LONG LIN, The University of Iowa — Nonlinear dynamics of regional airflow distribution in healthy human lungs are studied with four-dimensional computed tomography (4DCT) quantitative imaging of four subjects. During the scanning session, subjects continuously breathed with tidal volumes controlled by the dual piston system. For each subject, 10 instantaneous volumetric image data sets (5 inspiratory and 5 expiratory phases) were reconstructed. A mass-preserving image registration was then applied to pairs of these image data to construct a breathing lung model. Regional distributions of local flow rate fractions are computed from time-varying local air volumes. The 4DCT registration-based method provides the link between local and global air volumes of the lung, allowing derivation of time-varying regional flow rates during the tidal breathing for computational fluid dynamics analysis. The local flow rate fraction remains greater in the lower lobes than in the upper lobes, being qualitatively consistent with those derived from three static CT (3SCT) images (Yin et al. JCP 2013). However, unlike 3SCT, the 4DCT data exhibit lung hysteresis between inspiration and expiration, providing more sensitive measures of regional ventilation and lung mechanics. 1 NIH grants U01-HL114494, R01-HL094315 and S10-RR022421. 5:24PM E15.00004 Surfactant Delivery into the Lung , JAMES GROTBERG, University of Michigan, MARCEL FILOCHE, Ecole Polytechnique — We have developed a multiscale, compartmentalized model of surfactant and liquid delivery into the lung. Assuming liquid plug propagation, the airway compartment accounts for the plug’s volume deposition (coating) on the airway wall, while the bifurcation compartment accounts for plug splitting from the parent airway to the two daughter airways. Generally the split is unequal due to gravity and geometry effects. Both the deposition ratio RD (deposition volume/airway volume), and the splitting ratio, RS , of the daughters volumes are solved independently from one another. Then they are used in a 3D airway network geometry to achieve the distribution of delivery into the lung. The airway geometry is selected for neonatal as well as adult applications, and can be advanced from symmetric, to stochastically asymmetric, to personalized. RD depends primarily on the capillary number, Ca, while RS depends on Ca, the Reynolds number, Re, the Bond number, Bo, the dose volume, VD , and the branch angles. The model predicts the distribution of coating on the airway walls and the remaining plug volume delivered to the alveolar region at the end of the tree. Using this model, we are able to simulate and test various delivery protocols, in order to optimize delivery and improve the respiratory function. 5:37PM E15.00005 Nonlinear analysis of the influence of surfactant on the stability of a liquid bilayer inside a tube , YUANYUAN SONG, DAVID HALPERN, University of Alabama, JAMES GROTBERG, University of Michigan — The lung’s airways are coated internally with a liquid bilayer consisting of a serous layer immediately coating the airway wall and a more viscous mucus layer which is exposed to the gas core. A surface tension instability at the interfaces may lead to the formation of liquid plugs that block the passage of air. This is known as airway closure. Here we consider this thin liquid bilayer coating within a compliant tube in the presence of insoluble surfactant at the mucus-gas interface. Surfactant can reduce the surface tension and induce a stress gradient, both of which are stabilizing. Lubrication theory is used to derive a system of nonlinear evolution equations for the thickness of the layers, the location of the tube wall, and the surfactant concentration. The effects of various parameters, the thickness of the bilayer to the tube radius, the layer thicknesses ratio, the surface tension ratio, and the viscosity ratio between the two layers, and wall compliance parameters, are investigated numerically. For a single layer in a rigid tube, surfactant can increase the closure time by approximately a factor of five. However, for a bilayer, the presence of surfactant slows down the closure time by a significantly larger factor, twenty times or more dependent on system parameters. 5:50PM E15.00006 Spatial organization of cilia tufts governs airways mucus transport: Application to severe asthma1 , MUSTAPHA KAMEL KHELLOUFI, Aix Marseille University, DELPHINE GRAS, Med Bio Med, PASCAL CHANEZ, Aix Marseille University, ANNIE VIALLAT, CNRS — We study the coupling between both density and spatial repartition of beating cilia tufts, and the coordinated transport of mucus in an in-vitro epithelial model. We use a fully differentiated model epithelium in air liquid interface (ALI) obtained from endo-bronchial biopsies from healthy subjects and patients with asthma. The asthma phenotype is known to persist in the model. Mucus transport is characterized by the trajectories and velocities of microscopic beads incorporated in the mucus layer. When the beating cilia tufts density is higher than dc =11/100x100 µm2 a spherical spiral coordinated mucus transport is observed over the whole ALI chamber (radius=6mm). Below dc , local mucus coordinated transport is observed on small circular domains on the epithelium surface. We reveal that the radii of these domains scale with the beating cilia tufts density with a power 3.7. Surprisingly, this power law is independent on cilia beat frequency, concentration and rheological properties of mucus for healthy subject and patient with asthma. The rotating or linear mucus transport is related to dispersion of the cilia tufts on the epithelium surface. We show that impaired mucus transport observed in severe asthma model epithelia is due to a drastic lack and dysfunction of cilia tufts. 1 The author acknowledges the support of the French Agence Nationale de la Recherche (ANR) under reference ANR-13-BSV5-0015-01. Sunday, November 23, 2014 4:45PM - 6:03PM Session E16 Free-Surface Flows III: Turbulence and Wakes — 2000 - James Duncan, University of Maryland 4:45PM E16.00001 Turbulent hydraulic jumps: characterization of macro and micro bubble generation1 , MILAD MORTAZAVI, ALI MANI, Center for Turbulence Research, Stanford — Bubble generation is a ubiquitous two-phase flow phenomenon occurring constantly in nature and in a wide variety of industrial processes. This work considers a hydraulic jump as a canonical setting to investigate bubble generation by nonlinear breaking waves. We have performed direct numerical simulation of a turbulent hydraulic jump with inlet Froude number of 2.0 and Reynolds number of 11000. We show remarkable similarities in the bubble size distribution and its evolutions between this statistically stationary wave and reported results for transient breaking waves. Additionally, in the hydraulic jump large bubbles are observed to be generated in patch-like structures with a distinct frequency which matches a dominant frequency in the velocity spectrum. It is speculated that this frequency is associated with the roller frequency generated at the toe of the jump. Curvature and relative velocity at the liquid-liquid impact points are investigated as an attempt to model microbubble generation by making connections to the Mesler entrainment mechanism. 1 Supported by ONR 4:58PM E16.00002 Accuracy of the 2D+t Approximation for a Laminar Wake in Surface Waves1 , LAURA PAULEY, CHRIS MATHIOT, Penn State University — Wakes in the ocean can be produced by a stationary object in a current or by a moving object in stationary water. When viewed in a reference frame moving with the object, the wake can persist thousands of object diameters downstream. Due to the extensive domain, an unsteady two-dimensional (2D+t) computation is often used to sweep downstream through the wake development. The 2D+t computation approximates the development of the wake at a fixed location as an object moves past but applies cyclical boundary conditions in the streamwise direction. A Parabolized Navier-Stokes (PNS) method has the same numerical efficiency as the 2D+t method but includes additional streamwise gradient terms found in the three-dimensional governing equations. The present paper investigates the accuracy of the 2D+t approximation for a laminar wake interacting with a surface wave described by the Stokes drift velocity distribution. When the laminar wake and Stokes drift are in the same direction, a secondary recirculating flow develops in the cross-span plane. The 2D+t results are compared with results from 3D Navier-Stokes computations and results from PNS computations to identify criteria at which the 2D+t method will yield accurate results. 1 Second author supported by the Penn State College of Engineering Research Initiative 5:11PM E16.00003 A Laboratory Study of Rain-Induced Underwater Turbulence Using Particle Image Velocimetry1 , R. LIU, X. LIU, J.H. DUNCAN, University of Maryland — The characteristics of rain-induced turbulence under a free surface are studied experimentally with Paticle Image Velocimetry (PIV) techniques in a 1.22-m-by-1.22-m water pool with a water depth of 0.3 m. A rain generator consisting of an open-surface water tank with an array of 22-gauge hypodermic needles attached to the tank bottom is mounted above the water pool. The tank is connected to a 2D translation stage to provide a small-radius horizontal circular motion to the needles, thus avoiding repeated drop impacts at the same location under each needle. The drop diameter is 2.6 mm and the height of the rain generator above the water surface of the pool is varied from 1 m to 2.5 m to provide different impact velocities. Both the flow field of a single drop impact and the turbulent layer under the free surface during rain simulations were measured with PIV. It was found that the drop penetration, the thickness of the turbulent layer under the free surface and the RMS velocity fluctuation are strongly correlated to the impact velocities of raindrops. The influence of this turbulence on the height of rebounding jet stalks from drop impacts is discussed. 1 The support of the National Science Foundation, Division of Ocean Sciences, is gratefully acknowledged. 5:24PM E16.00004 Vorticity-based correction for modelling of free-surface wave interacting with turbulent current , WEI ZHANG, Research Fellow — This paper describes a new vorticity-based correction model for studying the interaction between free-surface wave and turbulent current. To track free-surface movements, the volume of fluid (VOF) method is employed. The momentum equations are rewritten to avoid the numerically generated vorticity effects along the air-water interface. Simultaneously unsteady RANS equations are used, while standard k-epsilon model is adapted with modification to the production term by introducing the vorticity to limit the production of turbulent kinematic energy at free surface. To validate the numerical model used here, standalone wave and current cases are studied to ensure the accuracy of each component of the numerical model. The model is then used to simulate the interaction between the second-order stokes wave and turbulent current for both wave following and countering in a setting of shallow water wave flume. The results are compared with experimental measurement available in the literature. 5:37PM E16.00005 Water Surface Ripples Generated by the Turbulent Boundary Layer of a Surface-Piercing Moving Wall1 , N. WASHUTA, N. MASNADI, J.H. DUNCAN, University of Maryland — Free surface ripples created by subsurface turbulence along a surface-piercing moving wall are studied experimentally. In this experiment, a meter-wide stainless steel belt travels horizontally in a loop around two rollers with vertically oriented axes, which are separated by 7.5 meters. One of the two 7.5-m-long belt sections between the rollers is in contact with the water in a large open-surface water tank and the water level is adjusted so that the top of the belt pierces the water free surface. The belt is launched from rest with a 3g acceleration in order to quickly reach a steady state velocity. This belt motion creates a temporally evolving boundary layer analogous to the spatially evolving boundary layer created along the side of a ship hull moving at the belt velocity, with a length equivalent to the length of belt that has passed the measurement region. The water surface ripples generated by the subsurface turbulence are measured in a plane normal to the belt using a cinematic LIF technique. It is found that the overall RMS surface fluctuations increase linearly with belt speed and that the spatial distributions of the fluctuations show a sharp increase near the wall. 1 The support of the Office of Naval Research is gratefully acknowledged. 5:50PM E16.00006 Surface Ripples Generated in a Couette Flow with a Free Surface1 , N. MASNADI, N. WASHUTA, J.H. DUNCAN, University of Maryland — Free surface ripples created by subsurface turbulence in the gap between a vertical surface-piercing moving wall and a parallel fixed wall are studied experimentally. The moving wall is created with the aide of a meter-wide stainless steel belt that travels horizontally in a loop around two rollers with vertically oriented axes, which are separated by 7.5 meters. One of the two 7.5-m-long belt sections between the rollers is in contact with the water in a large open-surface water tank and forms the moving wall. The fixed wall is an acrylic plate located 4 cm from the belt surface. The water surface ripples are measured in a plane normal to the belt using a cinematic LIF technique. Measurements are done at a location about 100 gap widths downstream of the leading edge of the fixed plate in order to have a fully developed flow condition. It is found that the overall RMS surface fluctuations increase linearly with belt speed. The frequency-domain spectra of the surface height fluctuation and its temporal derivative are computed at locations across the gap width and are used to explore the physics of the free surface motions. 1 The support of the Office of Naval Research is gratefully acknowledged. Sunday, November 23, 2014 4:45PM - 5:50PM Session E17 Astrophysical Fluid Dynamics — 2002 - Joseph Barranco, San Francisco State University 4:45PM E17.00001 Fluid Instabilities inside Astrophysical Explosions1 , KE-JUNG CHEN, STAN WOOSLEY, UC Santa Cruz, ALEXANDER HEGER, Monash U, ANN ALMGREN, WEIQUN ZHENG, LBNL — We present our results from the simulations of fluid instabilities inside supernovae with a new radiation-hydrodynamic code, CASTRO. Massive stars are ten times more massive than Sun. Observational and theoretical studies suggest that these massive stars tend to end their lives with energetic explosions, so-called supernovae. Many fluid instabilities occur during the supernova explosions. The fluid instabilities can be driven by hydrodynamics, nuclear burning, or radiation. In this talk, we discuss about the possible physics of fluid instabilities found in our simulations and how the resulting mixing affects the observational signatures of supernovae. 1 This work was supported by the DOE HEP Program under contract DE-SC0010676; the National Science Foundation (AST 0909129) and the NASA Theory Program (NNX14AH34G) 4:58PM E17.00002 End-effects versus stratification in quasi-Keplerian Taylor–Couette flow , COLIN LECLERCQ, RICH KERSWELL, University of Bristol — There has been much controversy in the past decade over the impact of end-wall boundary conditions on transition to turbulence in centrifugally stable Taylor–Couette experiments, e.g. Paoletti & Lathrop, PRL (2011); Ji et al., Nature (2006); Balbus, Nature (2011). In this configuration, the meridional flow driven by the axial boundaries is no longer confined to their vicinity, potentially leading to turbulence through a classical supercritical route at high rotation rates (Avila et al., POF (2008); Avila, PRL (2012)). But the question of subcritical transition in the limit of infinite cylinders remains of fundamental importance to the theory of weakly ionised accretion disks, as it may help to understand the inferred existence of turbulence there. We investigate theoretically the effect of stratification on azimuthally symmetric quasi-Keplerian base flows in the finite-length Taylor–Couette system. The challenge is to find a practical way to suppress the meridional flow, while not triggering the stratorotational instability. Different strategies will be discussed, including layered density profiles obtained with a stratifying agent of variable concentration and linear density profiles caused by a temperature difference between the top and bottom boundaries. 5:11PM E17.00003 The Fluid Dynamics of Saturn’s “String of Pearls” , CHRISTOPHER GEBHART, PHILIP MARCUS, UC Berkeley — A long-lived feature in Saturn’s northern hemisphere is a row of 20 - 30 discrete dark clouds at latitude 33 ◦ that extends approximately 1/3 of the way around the planet. It was the named the “String of Pearls” (SoP). It was suggested by others that these clouds are associated with a row of cyclonic vortices. However, generally a row of vortices with the same sign is unstable, and the vortices merge. Using a version of Correlation Image Velocimetry (CIV) that we developed to extract velocities from satellite images of clouds (Advection Corrected CIV), we have created a velocity field map of the SoP. From that map, we believe that the SoP lies on a strong, wavy westward jet stream that is sandwiched between a row of anticyclones on its northern side and a row of cyclones on its southern side, i.e., between a Karman Vortex Sheet (KVS). We previously showed that a KVS that sandwiches a jet stream is stable using 2D, quasigeostrophic simulations. Here, we present preliminary results on the stability and dynamics of a KVS using numerical simulations of the fully 3D, anelastic equations. We compare our simulations with the observations of the SoP. 5:24PM E17.00004 Zombie Vortices: The Dead Zones of Protoplanetary Disks are Not Dead , CHUNG-HSIANG JIANG, PHILIP MARCUS, SUYANG PEI, UC Berkeley, JOE BARRANCO, SFSU, PEDRAM HASSANZADEH, Harvard, DANIEL LECOANET, UC Berkeley — Numerical simulations, using both the anelastic and fully compressible equations of motion, show that the “dead zones” of protoplanetary disks (PPDs) around forming stars are unstable and filled with vortex-dominated turbulence with Mach and Rossby numbers of order 0.2 – 0.3. The dead zones are regions in which the temperature is too cool for the gas to ionize and be destabilized by instabilities associated with the magnetic field. The “dead zones” were thought, by most authors, to be stable to all purely-hydrodynamic instabilities because the flow has an angular momentum that increases with increasing radius in a PPD and is therefore stable by Rayleigh’s theorem. However, that theorem in not applicable to stratified flows, such as those in a PPD. We summarize our simulations with emphasis on the finite-amplitude trigger of the instability and show that when the trigger is Kolmogorov noise, the Mach number of the noise that is needed to create instability is proportional to Re−1/2 , where Re is the Reynolds number of the initial noise. 5:37PM E17.00005 Zombie Vortices: Angular Momentum Transport and Planetesimal Formation1 , JOSEPH BARRANCO, San Francisco State University, PHILIP MARCUS, SUYANG PEI, CHUNG-HSIANG JIANG, University of Cal- ifornia, Berkeley, PEDRAM HASSANZADEH, Harvard University, DANIEL LECOANET, University of California, Berkeley — Zombie vortices may fill the dead zones of protoplanetary disks, where they may play important roles in star and planet formation. We will investigate this new, purely hydrodynamic instability and explore the conditions necessary to resurrect the dead zone and fill it with large amplitude vortices that may transport angular momentum and allow mass to accrete onto the protostar. One unresolved issue is whether angular momentum transport is mediated via asymmetries in the vortices, vortex-vortex interactions, or acoustic waves launched by the vortices. Vortices may also play a crucial role in the formation of planetesimals, the building blocks of planets. It is still an open question how grains grow to kilometer-size. We will investigate the interactions of dust with vortices generated via our new hydrodynamic instability, and bridge the gap between micron-sized grains and kilometer-sized planetesimals. 1 Supported by NSF AST-1010052 Sunday, November 23, 2014 4:45PM - 6:03PM Session E18 Superfluids — 2004 - William Irvine, University of Chicago 4:45PM E18.00001 Approach and separation of quantum vortices with balanced cores , ROBERT M. KERR1 , University of Warwick, C. RORAI, NORDITA, Stockholm, J. SKIPPER, University of Warwick, K.R. SREENIVASAN, New York University — Using two innovations, smooth but different, scaling laws for the reconnection of pairs of initially orthogonal and anti-parallel quantum vortices are obtained using the three-dimensional Gross-Pitaevskii equations. For the anti-parallel case, the scaling laws just before and after reconnection obey the dimensional δ ∼ |t − tr |1/2 prediction with temporal symmetry about the reconnection time tr and physical space symmetry about xr , the mid-point between the vortices, with extensions forming the edges of an equilateral pyramid. For all of the orthogonal cases, before reconnection δin ∼ (t − tr )1/3 and after reconnection δout ∼ (tr − t)2/3 , which are respectively slower and faster than the dimensional prediction. In these cases, the reconnection takes place in a plane defined by the directions of the curvature and vorticity. 1 Robert.Kerr@warwick.ac.uk 4:58PM E18.00002 The Effect of Quantized Vortices on Particle Motion in Superfluid Helium , AKIRA HIRANO, DAIKI KATO, RYOUHEI OHTAKA, Nagoya University, AKIFUMI IWAMOTO, National Institute of Fusion Science, TAKAHIRO ITO, YOSHIYUKI TSUJI, Nagoya University, NAGOYA UNIVERSITY TEAM, NATIONAL INSTITUTE OF FUSION SCIENCE COLLABORATION — Superfluid 4 He (HeII) exits as liquid phase below 2.17K and indicates peculiar flow structure such as no viscosity and super heat conduction. Therefore, HeII is used as refrigerant in superconducting magnet. HeII property is well understood by so called two-fluid model that is composed of superfluid and normalfluid component. Quantized vortices are generated in superfluid component when the heat flux is larger than the critical value in a thermal counter flow. In this study, we use solid hydrogen particles as tracer and visualize tracer particle motions. The particles are forced by Stokes drag with the normalfluid and trapped by the quantized vortex with superfluid. The particle motions differ depending on the interaction between particle and quantized vortex. In order to analysis the particle trajectory, we adopt Pparticle Tracking Velocimetry. We identified two distinct types of particle trajectories moving straightly with normal fluid and moving irregularly with superfluid and apparently trapped by quantized vortex. They are compared with previous studies. The distribution of the vertical velocity component of particle motion was bimodal, which are consistent with theoretical values. We discuss in detail how the particle moves trapped by quantized vortices. 5:11PM E18.00003 Three Dimensional Quantized Vortex Dynamics in Superfluid Helium , DAVID MEICHLE, PETER MEGSON, DANIEL LATHROP, University of Maryland — Vorticity is constrained to line-like topological defects in quantum superfluids, such as liquid Helium below the Lambda transition temperature of 2.17 Kelvin. A tangle of vortices exists in a dissipative dynamical state called quantum turbulence, which has quantitative features distinct from classical turbulence. To study the vortex dynamics, we have invented a novel method to disperse fluorescent nanoparticles directly into the superfluid which become trapped on the vortex cores. Using a newly constructed multi-camera stereographic microscope, we present new data showing vortex reconnections and Kelvin waves with fully three-dimensional particle trajectories. These events are of scientific interest as they play a key role in the dissipation of quantum turbulence. 5:24PM E18.00004 Knots and Coils in Superfluid Vortices1 , DUSTIN KLECKNER, University of Chicago, James Franck Institute, DAVIDE PROMENT, University of East Anglia, MARTIN SCHEELER, WILLIAM T.M. IRVINE, University of Chicago, James Franck Institute — Recent work has demonstrated that linked and knotted vortices will spontaneously unknot or untie in both classical fluids and superfluids. This effect would appear to jeopardize any notion of conservation of fluid topology (helicity), but this need not be the case: vortices can transfer their knottedness to helical coils, preserving some measure of the original topology. By simulating superfluid vortices in the Gross-Pitaevskii equation, we find a geometric mechanism for efficiently transferring helicity in exactly this manner. Remarkably, the same transfer of topology to geometry also appears in viscous fluid vortices, suggesting it is a generic feature of non-ideal fluids. 1 This work was supported by the NSF MRSEC shared facilities at the University of Chicago (DMR-0820054) and an NSF CAREER award (DMR1351506). W.T.M.I. further acknowledges support from the A.P. Sloan Foundation and the Packard Foundation. 5:37PM E18.00005 Quantum analogues of classical wakes in Bose-Einstein condensates , GEORGE STAGG, NICK PARKER, CARLO BARENGHI, Newcastle University — We show that an elliptical obstacle moving through a Bose-Einstein condensate generates wakes of quantum vortices which resemble those of classical viscous flow past a cylinder or sphere. Initial steady symmetric wakes, similar to those observed in classical flow at low Reynolds number, lose their symmetry and form clusters of like-signed vortices, in analogy to the classical Bénard–von Kármán vortex street. The key ingredient to produce classical-like wakes is that vortices are generated at a sufficiently high rate that they undergo strong interactions with their neighbours (rather than being swept away). The role of ellipticity is to facilitate the interaction of the vortices and to reduce the critical velocity for vortex nucleation. Our findings, demonstrated numerically in both two and three dimensions, confirm the intuition that a sufficiently large number of quanta of circulation reproduce classical physics. The effects which we describe (dependence of the critical velocity and cluster size on the obstacle’s size, velocity and ellipticity) are also relevant to the motion of objects (such as vibrating wires, grids and forks) in superfluid helium, as the obstacle’s ellipticity plays a role which is analogous to rough boundaries. 5:50PM E18.00006 Acoustics of the Lambda Transition in Superfluid Helium , PETER MEGSON, DAVID MEICHLE, DANIEL LATHROP, Univ of Maryland-College Park — Liquid Helium undergoes a phase transition and becomes a quantum superfluid when cooled below the Lambda transition temperature of 2.17 Kelvin. The superfluid, which is a partial Bose Einstein Condensate, exhibits unique macroscopic properties such as flow without viscosity and ballistic temperature propagation. We have recorded striking audio-frequency sounds using a micro electromechanical microphone (MEMS) present as the Helium goes through the Lambda transition. Characterization of this sound, as well as its relevance to theories of the Lambda transition will be presented. Sunday, November 23, 2014 4:45PM - 6:03PM Session E19 Particle-Laden Flows: General II — 2006 - Martin Maxey, Brown University 4:45PM E19.00001 Deformation regimes for immersed single particle collisions , ANGEL RUIZ-ANGULO, Centro de Ciencias de la Atmosfera, Universidad Nacional Autonoma de Mexico, MELANY HUNT, California Institute of Technology — This work presents experimental measurements of the approach and rebound of single particles colliding with a “deformable” surface in a viscous liquid. The complex interaction between the fluid and the solid phases is coupled through the dynamics of the flow as well as the deformation process. A simple pendulum experiment was used to produced single controlled collisions; steel particles were used to impact different aluminum alloy samples in different aqueous mixtures of glycerol and water as a viscous fluid. For the combination of materials proposed, the elastic limit is reached at relatively low velocities. The deformations produced by the collision were analyzed using an optical profilometer. The measurements showed that the size of the indentations is independent of the fluid media. It was found that the size of the indentations was the same for collisions in air than for the rest of the collisions using various viscous fluids. 4:58PM E19.00002 Particle-laden thin film flow with surface tension , LI WANG, ALIKI MAVROMOUSTAKI , ANDREA BERTOZZI, Department of Math, University of California, Los Angeles — We derive a dynamic model which describes the evolution of a thin film, laden with negatively buoyant particles on an incline including the surface tension effect. The original model (Murisic et. al [J. Fluid Mech. 2013]) that only considers the leading order effects such as gravity and shear induced migration can produce singular shock when the particle concentration is above a critical value. Our model builds on a similar equilibrium theory and results in a 2 × 2 system of conservation laws augmented with forth order diffusion. Such diffusion is both a stand-alone regularization and a modification of the original flux, thus posing challenges for both the design of numerical schemes and analysis. We present the model and a proposed numerical method that produces solutions in which the singularity is suppressed, leading to more physical solutions. 5:11PM E19.00003 Dynamics of particle migration in a channel flow of viscoelastic fluids1 , GAOJIN LI, Purdue University, GARETH MCKINLEY, Massachusetts Institute of Technology, AREZOO ARDEKANI, Purdue University — Understanding the dynamics of particle transport in channel flows is important for many problems related to industrial, environmental and biological applications. Cross streamline migration of particles due to inertial and/or viscoelastic effects has been studied and utilized for particle focusing, particle separation and fluid mixing in microfluidic devices. Most previous studies on viscoelastic-induced particle focusing are limited to low Reynolds number flows and some of the mechanisms leading to particle migration remain unclear. In this work, we numerically study the interio-elastic migration of particles in a microfluidic channel flow driven by a constant pressure gradient. Simulations cover the following range of parameters: Reynolds numbers 4 ≤ Re ≤ 100, Weissenberg numbers 0 ≤ Wi ≤ 2, for weakly viscoelastic fluids with elasticity numbers 0 ≤ El= Wi/Re ≤ 0.2. Both viscoelasticity and shear-thinning effects are considered. The competition between inertia and viscoelasticity leads to different equilibrium particle positions between the channel centerline and the wall. The equilibrium position moves towards the centerline at higher El for a given Reynolds number due to the dominance of the cross-streamline viscoelastic force compared to the inertial lift. Shear-thinning effects increase the effective shear rate, and consequently, the dominance of the inertial lift drives the particle towards the wall. 1 A. M. Ardekani and Gaojin Li acknowledge support from NSF Grant No. CBET-1150348. 5:24PM E19.00004 Dispersion of a suspension plug in oscillatory pressure-driven flow , FRANCIS CUI, AMANDA HOWARD, MARTIN MAXEY, ANUBHAV TRIPATHI, Brown University — We investigate the dispersion of suspension plugs in a microcapillary as they are sheared in periodic pressure-driven flows. To study this novel configuration, a new experimental method was implemented to observe the shear-induced evolution of semi-infinite suspension plugs consisting of non-colloidal spherical particles (90-µm mean diameter) at dilute and high concentrations for various values of applied strain. In this cyclic shearing flow, irreversible particle migration arises from numerous unpredictable hydrodynamic interactions between particles and walls. Although the periodic velocity profiles do not lead to any significant increase in plug length, significant streamwise particle migration was observed near the walls of the capillary, becoming more pronounced with increasing strain amplitude γ0 . This experimental outcome agrees with the results of numerical simulation, which produces analogous behavior for a suspension sheared between parallel walls. Calculating dimensionless particle diffusivities Dz for various γ0 allows us to determine a cutoff point demarcating regimes of reversibility and irreversibility. 5:37PM E19.00005 Shock propagation over a deformable particle , THOMAS JACKSON, PRASHANTH SRIDHARAN, University of Florida, JU ZHANG, Florida Institute of Technology, SIVA BALACHANDAR, University of Florida — The interaction of strong shock waves with a deformable particle is an important fundamental problem in applications of multiphase flow; e.g., volcanic blasts, shock past a bubble, or explosives loaded with particles. In these applications the shock strength is greater than the yield strength of the particles, and as a result the particles will move and deform. We consider the impedance and shock-speed ratios, which define the nature of the deformation, for a variety of materials. Understanding the dynamic behavior of isolated particles at the microscale is important for developing point-particle models at the macroscale. Numerical results will be presented using the axisymmetric assumption to reduce computational costs. For a variety of shock strengths, we plot as a function of time a number of quantities, such as maximum particle temperature and pressure, mass integrated temperature and pressure, particle position. We also show results for non-spherical particles to determine the effect of particle shape. Here, we consider an ellipsoid align along or normal to the flow direction. Finally, preliminary results using a fully 3-D code will be presented. 5:50PM E19.00006 Shock-particle cloud interaction: Isolated unsteadiness contributions from shock and vortical structures , ZAHRA HOSSEINZADEH-NIK, JONATHAN D. REGELE, Iowa State University — The interaction between shock waves and particles in a multiphase shock tube is an efficient way to study dense compressible particle-laden flows. However, it is difficult to study the interaction between the two phases at the particle scale. Recent numerical simulations [Regele et al., Int. J. Multiphase Flow 61, 1-13 (2014)] show that after a shock wave impacts a particle cloud strong unsteady effects exist inside and in the wake immediately behind the cloud. This unsteadiness is attributed to the fluctuation associated with vortical structures and reverberating compression wave radiation. It is still unclear how the unsteady flow behavior is partitioned between vortical structures and reverberating finite disturbances. In this work numerical simulations are performed that attempt to replicate the unsteady wake behavior for the same mean flow conditions that are observed after a shock wave passage without using a shock wave to initialize the flow. The results are volume-averaged to compare the unsteady velocity magnitude with that of the previous results to determine the contribution from vortical unsteadiness. Sunday, November 23, 2014 4:45PM - 6:03PM Session E20 Flow Control: General — 2008 - Mihailo Jovanovic, University of New Mexico 4:45PM E20.00001 Three-dimensional simulations of flow around cylinders with fairing to suppress VIV , FANGFANG XIE, Massachusetts Institute of Technology, YUE YU, Lehigh University, YIANNIS CONSTANTINIDES, Chevron Energy Technology Company, GEORGE KARNIADAKIS, Brown University — Three-dimensional simulation of stationary and moving cylinders with free-to-rotate fairings are conducted at Reynolds number 100 ≤ Re ≤ 10,000. Fairings are nearly-neutrally buoyant devices, which are fitted along the axis of long circular risers to suppress vortex-induced vibrations (VIV) and reduce the drag force. The effect of gap between fairings along the cylinder axis on the hydrodynamic forces (Cd , Cl ) and the translational and rotational motion of fairings (xrms , yrms, θrms ) are investigated. With the increase of Re, the drag coefficient Cd of fairing decreases. Compared to the plain cylinder case, fairings without gap can reduce Cd by 15% while the fairing with gap can reduce Cd by almost 50%. The lift force (Cl ) and angular momentum of fairing (Mf h ) for different gaps are also decreased. Correspondingly, the vibration (yrms ) and rotation (θrms ) amplitudes of fairing are also reduced. We also investigate the change in flow structure induced by the fairing gaps. A pair of stream-wise vortices are generated in the gap region, which extracts energy from the flow in the cross-flow direction hence causing decrease of the lift force. As Re increases, pressure recovery in the wake of the fairing is observed, which is the main reason for the substantial decrease of drag force. 4:58PM E20.00002 Receptivity of the Boundary Layer to Vibrations of the Wing Surface1 , TOMASS BERNOTS, ANATOLY RUBAN, DAVID PRYCE, Imperial College London, LAMINAR FLOW CONTROL UK GROUP TEAM — In this work we study generation of Tollmien-Schlichting (T-S) waves in the boundary layer due to elastic vibrations of the wing surface. The flow is investigated based on the asymptotic analysis of the Navier-Stokes equations at large values of the Reynolds number. It is assumed that in the spectrum of the wing vibrations there is a harmonic which comes in resonance with the T-S wave on the lower branch of the stability curve. It was found that the vibrations of the wing surface produce pressure perturbations in the flow outside the boundary layer which can be calculated with the help of the piston theory. As the pressure perturbations penetrate into the boundary layer, a Stokes layer forms on the wing surface which appears to be influenced significantly by the compressibility of the flow, and is incapable of producing the T-S waves. The situation changes when the Stokes layer encounters an roughness; near which the flow is described using the triple-deck theory. The solution of the triple-deck problem can be found in an analytic form. Our main concern is with the flow behaviour downstream of the roughness and, in particular, with the amplitude of the generated Tollmien-Schlichting waves. 1 This research was performed in the Laminar Flow Control Centre (LFC-UK) at Imperial College London. The centre is supported by EPSRC, Airbus UK and EADS Innovation Works. 5:11PM E20.00003 Relaminarisation of fully turbulent flow in pipes , JAKOB KUEHNEN, BJOERN HOF, IST Austria — Drag reduction still remains one of the most alluring applications of turbulence control. We will show that flattening the streamwise velocity profile in pipes can force turbulent flow to decay and become laminar. Two different experimental control schemes are presented: one with a local modification of the flow profile by means of a stationary obstacle and one with a moving wall, where a part of the pipe is shifted in the streamwise direction. Both control schemes act on the flow such that the streamwise velocity profile becomes more flat and turbulence gradually grows faint and disappears. Since, in a smooth straight pipe, the flow remains laminar from that position a reduction in skin friction by a factor of 5 can be accomplished. We will present measurements with high-speed particle image velocimetry, measurements of the pressure drop and videos of the development of the flow during relaminarisation. The guiding fundamental principle behind our approach to control the velocity profile will be explained and discussed. 5:24PM E20.00004 Model-based and adaptive laminar-flow control via dielectric-barrierdischarge plasma actuators: an experimental comparison , NICOLÒ FABBIANE, Linnè FLOW Centre, KTH Mechanics, BERNHARD SIMON, SVEN GRUNDMANN, Center of Smart Interfaces, TU Darmstadt, SHERVIN BAGHERI, DAN S. HENNINGSON, Linnè FLOW Centre, KTH Mechanics — This work compares two of the mostly investigated reactive-control techniques in delaying the laminar-to-turbulence transition in boundarylayer (BL) flows: a Linear Quadratic Gaussian (LQG) regulator and a Filtered-X Least Mean Squares (FXLMS) algorithm. The two compensators are compared on damping 2D TS-waves excited via both single-frequency and white-noise disturbances in a zero-pressure-gradient BL flow. Surface hot-wire sensors are used to detect the incoming waves and measure the effectiveness of the control action that is provided by a dielectric-barrier-discharge plasma actuator positioned between the two sensors. Based on DNS of the experimental set-up a linear reduced order model is built using the Eigensystem Realization Algorithm and used for the LQG design. The two control techniques show comparable performances when tested at their design condition. However, when tested off-design the LQG compensator shows a stronger sensitivity to model variations. If the free-stream velocity is changed, the LQG regulator estimates a wrong phase information of the incoming disturbance resulting in a less effective control action. The FXLMS compensator, instead, is capable to adapt to the new condition and prescribe the correct phase information with no significant performance loss 5:37PM E20.00005 Estimation and feedback control of vortex shedding , SIMON ILLINGWORTH, University of Melbourne — We consider estimator-based control of the cylinder wake in low-Reynolds-number simulations. There are two parts to the study. In the first part, we show that feedback control with a single sensor becomes increasingly difficult as Reynolds number increases. This is because of a larger region of absolute instability. The convective nature of the flow means that the single sensor is unaware of what is happening downstream of it, making feedback control very difficult. This motivates the second part of the study, where we consider estimator-based feedback control. Keeping with a single sensor measurement, we investigate how well one can estimate the entire flow field using only this single sensor. To do so we use a Kalman filter, and excellent results are seen. We then combine this Kalman filter with suitable feedback control laws to suppress vortex shedding. This control strategy still uses only a single sensor but, crucially, the control actions are based on (the estimate of) what is happening in the full domain. This control strategy achieves much better suppression in the far wake when compared with the single-sensor controller without estimator. 5:50PM E20.00006 Controlling the onset of turbulence by spatially-localized feedback controllers , NEIL DHINGRA, MIHAILO JOVANOVIC, University of Minnesota — We develop a model-based technique for a design of spatially-localized controllers which map sections of a flow field to actuator activity at the channel’s walls. These controllers minimize variance amplification from flow disturbances to the energy of velocity fluctuations. Traditional approaches, based on linear quadratic regulator theory, result in control laws which require that each actuator observes the entire flow field. In contrast, our approach augments the objective function (that penalizes the variance amplification) with a term that promotes sparsity in the feedback gains. By increasing the emphasis on sparsity, we obtain localized family of spatially-invariant distributed controllers. We demonstrate computational efficiency of the developed algorithm and illustrate its utility in controlling the onset of turbulence. Sunday, November 23, 2014 4:45PM - 6:03PM Session E21 Instability: Vortex Flows — 2010 - Paul Fontana, Seattle University 4:45PM E21.00001 Two-dimensional linear instability of a Lamb-Oseen counter-rotating vortex dipole , REMI JUGIER, LAURENT JOLY, JEROME FONTANE, Universite de Toulouse, ISAE, DAEP, PIERRE BRANCHER, Universite de Toulouse, UPSIMFT — The present study investigates the stability of a family of quasi-steady two-dimensional vortex dipoles resulting from the adaptation to the mutual deformation and viscous diffusion of counter-rotating Lamb-Oseen vortices of equal size and circulation. For sufficiently large Reynolds numbers, the internal structure is set by the dipole aspect ratio a/b (radius versus separation distance), giving rise to a family of smooth quasi-steady solutions. These base flow solutions consist of self-propagating dipoles with a vorticity trail leaking from the dipole during the self-adaptation process and from diffusion across the dipole Kelvin oval. The present work is a generalization of the recent study by Brion et al. (Phys. Fluids 2014), who found that a specific kind of vortex dipole, the Lamb-Chaplygin dipole, was unstable with respect to two-dimensionnal perturbations. We show how the growth rate and spatial structure of the unstable modes vary with aspect ratio a/b and Reynolds number. The growth rates are lower than for the Lamb-Chaplygin dipole and decreases when the aspect ratio is lowered. It is advocated here that these new two-dimensional modes of small aspect-ratio smooth dipoles are good candidates for steady actuation in aircrafts conditions. 4:58PM E21.00002 Experimental and theoretical analysis of vortex breakdown in the wake of the 25◦ Ahmed body , CYRIL JERMANN, PHILIPPE MELIGA, M2P2 (CNRS - Centrale Marseille), GREGORY PUJALS, PSA Peugeot Citroen, FRANCOIS GALLAIRE, LFMI (EPFL - Lausanne), ERIC SERRE, M2P2 (CNRS - Centrale Marseille) — We study experimentally and theoretically the wake of the 25o Ahmed body, considered a suitable test-case to reproduce the two counter-rotating longitudinal vortices widely encountered in automotive aerodynamics. The three-dimensional experimental mean flow is reconstructed at high Reynolds number (Re = 2.8 × 106 ) from a series of cross-flow time-averaged planes acquired with a moving automated Stereo-PIV system. We observe a sharp decay of the axial velocity and vorticity in the near-wake, 0.5 times the projected length of the slanted surface downstream the square back, where the streamwise vortices is subjected to a strong adverse pressure gradient and the turbulent kinetic energy exhibits a peak in the vortex core. A stability analysis of the experimental velocity shows that the flow undergoes vortex breakdown roughly at the same position, through a transition from supercritical (x < 0.5) to subcritical (x > 0.5) conditions and the accumulation of upstream propagating axisymmetric waves. 5:11PM E21.00003 The linear stability analysis of swirling flows in a finite-length pipe , RUI GONG, SHIXIAO WANG, Auckland University, ZVI RUSAK, Rensselaer Polytechnic Institute — The linear stability of an inviscid, axisymmetric and rotating columnar flow in a finite length pipe against general type of perturbations is studied. The perturbation mode is subject to a set of boundary conditions that may reflect the physical situation of the swirling flow in a finite-length pipe. This type of stability problem was first studied by Wang and Rusak (1996). The current study generalized their original analysis (valid for axisymmetric perturbations) to a stability analysis of general type of perturbations. The underlying physical mechanism of the unstable mode is examined and discussed in terms of the energy transfer mechanism between the perturbations and the base flow. 5:24PM E21.00004 Onset of unsteady flow in wavy walled channels at low Reynolds number1 , TAPAN SHAH, ZACHARY MILLS, ALEXANDER ALEXEEV, Georgia Institute of Technology — Using computational modeling, we examine the development of an unsteady laminar flow of a Newtonian fluid in a channel with sinusoidal walls, driven by a constant pressure gradient. The lattice Boltzmann method was used as our computational model. Our simulations revealed two types of unsteady flows occurring in sinusoidal channels. When the amplitude of the wavy walls is relatively small, vortices forming in the channel furrows are shed downstream. For larger wall wave amplitudes, vortices remain inside the furrows and exhibit periodic oscillations and topological changes. Our simulations establish the optimum wall amplitude and period leading to an unsteady flow at the minimum pressure gradient. The results are important for designing laminar heat/mass exchangers utilizing unsteady flows for enhancing transport processes. 1 This work is supported by General Motors Corporation. 5:37PM E21.00005 Numerical Study of Electrolytic Flow Instabilities Driven by an Azimuthal Lorentz Force in a Cylindrical Geometry1 , JAMES PÉREZ-BARRERA, JOSÉ ENRIQUE PÉREZ-ESPINOZA, ALEJANDRO ORTÍZ, SERGIO CUEVAS, EDUARDO RAMOS, Universidad Nacional Autónoma de México — We present numerical simulations of the flow produced by an azimuthal Lorentz force in an electromagnetic stirrer. The stirrer consists of a cylindrical cavity with two copper concentric cylindrical electrodes, filled with an electrolytic solution. Underneath the cavity, a permanent magnet creates an almost uniform magnetic field, perpendicular to the circular section of the stirrer. An electric potential difference between the electrodes produces a radial D.C. current that passes through the fluid and interacts with the axial magnetic field, generating an azimuthal Lorentz force that drives the fluid. Experiments have shown the appearance of a flow instability that gives rise to a varying number of anticyclonic vortices for given values of the current intensity and fluid layer thickness. The MHD governing equations are expressed in terms of the velocity, pressure and electric potential. Numerical simulations are carried out using a hybrid Finite volume-Fourier method to ensure periodicity in the azimuthal direction. Numerical results show the formation of different modes of perturbation in the velocity field, which give rise to a varying number of traveling vortical structures. 1 Work supported by CONACYT, Mexico under Project 131399. JPB acknowledges a grant from CONACYT. 5:50PM E21.00006 Anomalous vortex shedding and wake profiles in quasi-two-dimensional flows , PAUL W. FONTANA, DOMINIC A. DAMS, Seattle University — Vortex shedding by circular cylinders in a vertical soap film channel exhibits anomalously low shedding frequencies compared with observations in conventional systems. Furthermore, the Strouhal number (St = f D/U∞ , where f is the shedding frequency, D the cylinder diameter, and U∞ the upstream flow speed) is not uniquely determined by the Reynolds number (Re = DU∞ /ν, where ν is the kinematic viscosity). We have previously argued that Ekman friction is a likely cause [Bull. Amer. Phys. Soc. 57(17), R10.7]. Other possibilities include gravity, which in this system acts as a forcing mechanism not typically present during vortex shedding measurements, surface tension effects, or variable-viscosity effects due to variations in film thickness. Theory to predict the shedding frequency is lacking and so it is unclear if or how each of these mechanisms should affect it, but understanding the anomaly may elucidate the shedding process. We present two-dimensional profiles of velocity, viscosity, and surface friction measured in the wake of the cylinder under several sets of flow parameters and discuss their implications for the various candidates. The results do not support variable viscosity as a cause. Sunday, November 23, 2014 4:45PM - 6:03PM Session E22 Instability: Wakes — 2012 - Francois Gallaire, Ecole Polytechnique Federale de Lausanne 4:45PM E22.00001 Flow around a rectangular forebody with modified inlet conditions , RENZO TRIP, JENS H.M. FRANSSON, KTH Mechanics — The near wake behind a rectangular forebody with a smooth leading edge and a blunt trailing edge is investigated, whereby the boundary layer over the forebody is modified by means of wall suction. The laminar boundary layer subject to wall suction yields the asymptotic suction boundary layer (ASBL), whereas an initially turbulent boundary layer will start to relaminarize for high enough wall suction. Wall suction therefore provides the possibility to perform a comprehensive parametrical study. The wake characteristics, such as the base pressure and the shedding frequency, are related to the boundary layer thickness and shape. Time resolved, with respect to the vortex shedding frequency, planar Particle Image Velocimetry (PIV) measurements are performed to gain fundamental knowledge on the role of the topology of the recirculation region in this respect. The mean flow fields do also allow for a global stability and sensitivity analysis on the vortex shedding instability. 4:58PM E22.00002 On the destabilizing influence of surface tension in planar wakes , EYAL HEIFETZ, Tel Aviv University, LUCA BIANCOFIORE, Imperial College London, FRANÇOIS GALLAIRE, EPFL — A counterintuitive destabilizing effect of the surface tension in planar immiscible wakes was observed by means of a linear global analysis (Tammisola et al., PoF, 2011) and Direct Numerical Simulations (Biancofiore et al., FDR, 2014), respectively. This destabilization can be interpreted by the presence of two different temporal unstable modes found when analyzing the local stability of an extracted velocity profile from the base flow. We approximate the wake velocity profile through a piecewise broken-line profile. We then explain the presence of these two temporal unstable modes using the Rossby wave (RW) perspective, which associates to each vorticity discontinuity an individual RW. The introduction of a finite amount of surface tension at the interface creates two capillary waves (CW) which travel with the same relative velocity but in opposite directions. The interaction of these four waves originates in two temporal unstable modes for both sinuous and varicose symmetries. Furthermore, we have captured the spatio temporal evolution of the interacting four-waves system by means of an impulse response analysis. The spreading of the wavepacket is significantly influenced by the coupling of the Rossby waves with the capillary waves, and is seen to favor absolute instability. 5:11PM E22.00003 Prediction of the hub vortex instability within wind turbine wakes and effects of the incoming wind and turbine aerodynamic characteristics , GIACOMO VALERIO IUNGO, University of Texas at Dallas, FRANCESCO VIOLA, Ecole Polytechnique Fédérale de Lausanne (EPFL), SIMONE CAMARRI, University of Pisa, FERNANDO PORTÉ-AGEL, FRANCOIS GALLAIRE, Ecole Polytechnique Fédérale de Lausanne (EPFL) — Instability of the hub vortex, which is a vorticity structure present in wind turbine near-wake and mainly oriented along the streamwise direction, is predicted from wake velocity measurements. In this work, stability analysis is performed on wind tunnel velocity measurements acquired in the wake produced from a wind turbine model immersed in a uniform flow. Turbulence effects on wake dynamics are taken into account by modeling the Reynolds stresses through eddy-viscosity models, which are calibrated on the wind tunnel data. This formulation leads to the identification of one dominant mode associated with the hub vortex instability, which is characterized by a counter-winding single-helix mode. Moreover, this analysis also predicts accurately the frequency of the hub vortex instability observed experimentally. The hub vortex instability is also investigated by considering incoming wind fields with different turbulence characteristics, different turbine aerodynamic designs and operational regimes, which affect the morphology of the wake vorticity structures and their dynamics. The ultimate goal of this work consists in providing useful information for predicting wind turbine wake dynamics and their effects on downstream wake recovery, thus to maximize wind power harvesting. 5:24PM E22.00004 Stability of the laminar wake behind spinning axisymmetric bluff bodies: sensitivity and control1 , JOSE IGNACIO JIMENEZ-GONZALEZ, CARLOS MARTINEZ-BAZAN, Universidad de Jaen, WILFRIED COENEN, CARLOS MANGLANO, ALEJANDRO SEVILLA, Universidad Carlos III de Madrid — We carry out direct and adjoint global stability analyses of the laminar wake behind several spinning axisymmetric bluff bodies, i.e. sphere, hemisphere, bullet-shaped bodies of ellipsoidal nose and spherical nose respectively; for moderate Reynolds numbers (Re≤450) and values of the spin parameter (Ω ≤1), defined as the ratio between the azimuthal velocity at the outer body surface and the free-stream velocity. Both the axisymmetric base flow computations and the assembling of the eigenvalue problems are tackled by means of the finite element solver FreeFEM++, computing finally the eigenmodes with an Arnoldi algorithm in Matlab. We show that spin acts as a stabilization mechanism for the wake behind bodies with a cylindrical trailing part, while it destabilizes the wake of the other geometries. The computation of the adjoint modes and the identification of the wavemaker allow us to discuss the nature of the different unstable modes found and understand the differences in the stabilizing or destabilizing effect of rotation due to the base flow modifications. The controllability of the unstable regimes by means of base bleed is also addressed. 1 Supported by the Spanish MINECO, Junta de Andalucı́a and EU Funds under projects DPI2011-28356-C03-03 and P11-TEP7495. 5:37PM E22.00005 Wake transition and vortex street interaction in flows generated by traveling localized Lorentz forces in a shallow electrolyte layer1 , JOEL ROMAN, SERGIO CUEVAS, Universidad Nacional Autonoma de Mexico — We present an experimental and numerical study of the vortex street produced by a traveling localized Lorentz force, namely a magnetic obstacle, in a thin layer of electrolyte. The Lorentz force is generated by the interaction a localized magnetic field created by a small permanent magnet which travels with a uniform velocity underneath a rectangular container and a uniform D.C. current applied transversally to the motion of the magnet. We find that by increasing the Reynolds number (based on the velocity of the magnet) the wake generated by the magnetic obstacle presents a transition from the Bénard-von Kármán (BvK) wake to the reversed BvK wake. In addition, we analyze the flow past a pair magnetic obstacles side-by-side in a thin layer of electrolyte by varying the separation between the magnets and the intensity of the applied current. The attention is focused in the interference of the wakes created by the magnetic obstacles. Numerical simulations based on a quasi-two dimensional numerical model present a satisfactory agreement with experimental results. 1 Work supported by CONACYT, Mexico under Project 131399. J. Roman acknowledges a grant from CONACYT. 5:50PM E22.00006 Mean flow stability wave models for coherent structures in open shear flows: experimental assessment of potentials and limitations1 , KILIAN OBERLEITHNER, LOTHAR RUKES, OLIVER PASCHEREIT, Technical University Berlin, JULIO SORIA, Monash University, Melbourne — We report on a number of experimental and theoretical investigations of shear flow instabilities in jet flows. In these studies, linear stability analysis is employed to the time-averaged flow taken from experiments, contrasting the “classic” stability approach that is based on a stationary base flow. The eigenmodes of the time-averaged flow are considered as models for the nonlinearly saturated state of the instability waves. The accuracy of these models is validated through a detailed comparison with experiments. In this talk we outline the potential and limitation of these flow models for convectively and globally unstable jet flows. 1 The first author was supported by a fellowship within the Postdoc-Program of the German Academic Exchange Service (DAAD). The support of the Australian Research Council (ARC) and the German Research Foundation (DFG) is greatfully acknowledged. Sunday, November 23, 2014 4:45PM - 5:50PM — Session E23 Geophysical Fluid Dynamics: Stratified Turbulence III 2001 - Hieu Pham, Univesrity of California at San Diego 4:45PM E23.00001 Low-Level Jets, Coherent Structures and Turbulence in a Stably Stratified Atmospheric Boundary Layer , IMAN GOHARI, MASOUD JALALI B., PhD Student, SCOTT BADEN, SUTANU SARKAR, professor — Accurate numerical modeling of stably stratified Atmospheric Boundary Layers (SABL) is known to be a pacing item for progress in numerical weather prediction, atmospheric dispersion and other related applications such as harnessing wind energy. The stabilizing effect of buoyancy not only dampens the strength of turbulence relative to near-neutral and convectively unstable cases but also qualitatively affects turbulence structure by changing anisotropy in velocity, anisotropy in length scale and spatio-temporal intermittency. Low-Level Jets (LLJs) form at O(100) m heights in the rotation-influenced SABL. It is thought that the LLJ leads to two turbulent layers, wherein the turbulence in the top layer originates from shear production and that in the lower one is mainly controlled by surface conditions. Properties of the SABL during the evolution of LLJ are not well understood. Therefore, Large Eddy Simulations of a rotation-influenced SABL with different surface cooling rates have been performed. Fluctuations (coherent structures, turbulence, and internal gravity waves) during LLJ evolution are quantified through second-order moments, high-order derivative statistics, spectra, length scales and eduction of coherent structures. 4:58PM E23.00002 On Parameterizing Turbulence in the Stably Stratified Atmospheric Boundary Layer1 , JORDAN M. WILSON, SUBHAS K. VENAYAGAMOORTHY, Colorado State University — Parameterizing turbulent mix- ing in the stably stratified atmospheric boundary layer remains an active area of research connecting available field measurements with appropriate model parameters. The research presented studies the pertinent mixing lengths for shear- and buoyancy-dominated (or weakly stable and very stable) regimes in the stable atmospheric boundary layer (SABL). Incorporating shear and buoyancy effects, two length scales can be constructed, LkS = k1/2 /S and LkN = k1/2 /N , respectively. Extending the conceptual framework of Mater & Venayagamoorthy (2014)2 , LkS and LkN are shown to be accurate representations of large-scale motions from which relevant model parameters are developed using observations from three field campaigns. An a priori analysis of large-eddy simulation (LES) data evaluates the efficacy of parameterizations applied to the vertical structure of the SABL. The results of this study provide a thorough evaluation of the pertinent mixing lengths in stably stratified turbulence through applications to atmospheric observations and numerical models for the boundary layer extendable to larger-scale weather prediction or global circulation models. 1 S.K.V. 2 Phys. gratefully acknowledges the support of the National Science Foundation under Grant No. OCE-1151838 Fluids, 26(3), 036601. 5:11PM E23.00003 Stratified Turbulence Measurements in Complex Terrain Using Hot-film Probes and a Collocated Sonic Anemometer1 , C. HOCUT, U.S. Army Research Laboratory, E. KIT, Tel Aviv University, D. LIBERZON, Technion - Israel Institute of Technology, H.J.S. FERNANDO, University of Notre Dame, MATERHORN TEAM — In the fall of 2012 and spring 2013, the Mountain Terrain Atmospheric Modeling and Observations Program (MATERHORN) conducted extensive field experiments at the Granite Mountain Atmospheric Science Testbed (GMAST), US Army Dugway Proving Grounds (DPG), Utah. This provided a unique opportunity to deploy tower mounted three-dimensional hot-film combo probes, consisting of sonic anemometers collocated with hot-film anemometers able to respond to the wind direction. The combo probes follow mean winds using a feedback control loop and use a Neural Network to calibrate the hot-films in-situ. Once calibrated, these probes can handle a vast range of background flow conditions and scales from mesoscale flow down to the Kolmogorov scale. Of particular interest are the observed variation in velocity spectra during the evenings. Sometimes the velocity spectra shows the turbulence is Kolmogorov and is isotropic at small scales while in other spectra there is evidence of turbulence production at finer scales. An explanation on different spectral shapes will be presented as well as the relevant length/time scales of the production events. 1 Funded by ONR grant N00014-11-1-0709 5:24PM E23.00004 Turbulence and Mixing Near a Sloping Boundary of a Lake1 , CHRIS REHMANN, ZHIMIN LI, HUI HU, Iowa State University — Fluxes in stratified water bodies such as lakes and oceans are often controlled by turbulence and mixing at sloping boundaries, and determining how the mixed fluid moves from the boundary to the interior is important for estimating basin-wide transport of heat and other scalars. A field experiment in one lake showed that fluid mixed at the boundary can be transported by intrusions that form as the mixed fluid collapses while an experiment in another lake suggested that the transport is caused by advection and dispersion by internal waves. Further work on this problem involves two parallel approaches. An analytical model, based on rapid distortion theory, is used to determine the effect of straining by vertical mode-2 waves on the turbulence and to compute the efficiency of the mixing. This approach is complemented with laboratory measurements of velocity and scalar fields with molecular tagging velocimetry. These measurements allow the scalar fluxes to be quantified as a function of the ratio of the wave frequency and the critical wave frequency. 1 We acknowledge support from the U.S. National Science Foundation. 5:37PM E23.00005 Turbulence and dissipation in a computational model of Luzon Strait , MASOUD JALALI, SUTANU SARKAR, University of California, San Diego — Generation sites for topographic internal gravity waves can also be sites of intense turbulence. Bottom-intensified flow at critical slopes leads to convective instability and turbulent overturns [Gayen & Sarkar (2011)]. A steep ridge with small excursion number, Ex, but large super criticality can lead to nonlinear features according to observations [Klymak et al. (2008)] and numerical simulations [Legg & Klymak (2008)]. The present work uses high resolution 3-D LES to simulate flow over a model with multiscale topography patterned after a cross-section of Luzon Strait, a double-ridge generation site which was the subject of the recent IWISE experiment. A 1:100 scaling of topography was employed and environmental parameters were chosen to match the slope criticality and F r number in the field. Several turbulent zones were identified including breaking lee waves, critical slope boundary layer, downslope jets, internal wave beams, and vortical valley flows. The multiscale model topography has subridges where a local Ex may be defined. Wave breaking and turbulence at these subridges can be understood if the local value of Ex is employed when using the Ex-based regimes identified by Jalali et al. (2014) in their DNS of oscillating flow over a single triangular obstacle. Sunday, November 23, 2014 4:45PM - 6:03PM Session E24 Wind Turbines: Wind Farms — 2003 - Johan Meyers, Katholieke University, Leuven 4:45PM E24.00001 Coupled wake boundary layer model of windfarms1 , RICHARD STEVENS, DENNICE GAYME, CHARLES MENEVEAU, Johns Hopkins University — We present a coupled wake boundary layer (CWBL) model that describes the distribution of the power output in a windfarm. The model couples the traditional, industry-standard wake expansion/superposition approach with a top-down model for the overall windfarm boundary layer structure. Wake models capture the effect of turbine positioning, while the top-down approach represents the interaction between the windturbine wakes and the atmospheric boundary layer. Each portion of the CWBL model requires specification of a parameter that is unknown a-priori. The wake model requires the wake expansion rate, whereas the top-down model requires the effective spanwise turbine spacing within which the model’s momentum balance is relevant. The wake expansion rate is obtained by matching the mean velocity at the turbine from both approaches, while the effective spanwise turbine spacing is determined from the wake model. Coupling of the constitutive components of the CWBL model is achieved by iterating these parameters until convergence is reached. We show that the CWBL model predictions compare more favorably with large eddy simulation results than those made with either the wake or top-down model in isolation and that the model can be applied successfully to the Horns Rev and Nysted windfarms. 1 The ‘Fellowships for Young Energy Scientists’ (YES!) of the Foundation for Fundamental Research on Matter supported by NWO, and NSF grant #1243482 4:58PM E24.00002 Temporal characteristics of POD modes from wind farm LES1 , CLAIRE VERHULST, CHARLES MENEVEAU, Johns Hopkins University — Large eddy simulations of a fully developed wind farm in the turbulent atmospheric boundary layer have been analyzed using 3D Proper Orthogonal Decomposition (POD). In this study we consider the temporal variations of the POD modes and their relationship to unsteadiness in the wind turbine power production. We find that the streamwise-constant counter-rotating roller modes vary on time-scales much longer that the mean advection time from turbine to turbine. The structure of these roller modes and their long-time variations are consistent with meandering of high- and low-speed streaks in the turbulent flow within the wind farm. Another class of POD modes—one with significant streamwise-variation—is found to correspond to advection of velocity perturbations in the streamwise direction. Temporal variations of the shear-type modes are found to strongly correlate with power production of the wind farm as a whole. Overall, the long-time power production is well captured by reconstructions using fewer than 50 POD modes (< 1% of the total), but variations faster than the inter-turbine advection time are only captured by higher-order, less energetic modes. 1 This work was supported by NSF grant 1243482 (the WINDINSPIRE project). 5:11PM E24.00003 A geometry-based approach for optimizing wind turbine layout , NIRANJAN GHAISAS, CRISTINA ARCHER, University of Delaware — Layout studies are critical in designing large wind farms, since wake effects can lead to significant reductions in wind power generation. Optimizing wind farm layout using computational fluid dynamics is practically unfeasible today because of the high computational requirements of the numerical simulations. Simple statistical models, based on geometric quantities associated with the wind farm layout, are therefore attractive because they are less demanding computationally. Results of large-eddy simulations of the Lillgrund wind farm are used here to develop such geometry-based models. Several geometric quantities (e.g., blockage ratio, or the fraction of the swept-area of a wind turbine which is blocked by upstream turbines) are found to correlate very well (> 0.95) with the power generated by the turbines. These models are particularly accurate at predicting the farm-averaged power and are therefore used here to study layout effects in large wind farms. Several layout parameters are considered, such as angle between rows and columns, turbine spacing, staggering of alternate rows, and wind direction. This study demonstrates the utility of simple, inexpensive, and reasonably accurate geometric models to identify general principles governing optimal wind farm layout. 5:24PM E24.00004 Optimal control of energy extraction in LES of large wind farms1 , JOHAN MEYERS, JAY GOIT, WIM MUNTERS, KU Leuven. Mechanical Engineering, Celestijnenlaan 300A, B3001 Leuven, Belgium — We investigate the use of optimal control combined with Large-Eddy Simulations (LES) of wind-farm boundary layer interaction for the increase of total energy extraction in very large “infinite” wind farms and in finite farms. We consider the individual wind turbines as flow actuators, whose energy extraction can be dynamically regulated in time so as to optimally influence the turbulent flow field, maximizing the wind farm power. For the simulation of wind-farm boundary layers we use large-eddy simulations in combination with an actuator-disk representation of wind turbines. Simulations are performed in our in-house pseudo-spectral code SP-Wind. For the optimal control study, we consider the dynamic control of turbine-thrust coefficients in the actuator-disk model. They represent the effect of turbine blades that can actively pitch in time, changing the lift- and drag coefficients of the turbine blades. In a first infinite wind-farm case, we find that farm power is increases by approximately 16% over one hour of operation. This comes at the cost of a deceleration of the outer layer of the boundary layer. A detailed analysis of energy balances is presented, and a comparison is made between infinite and finite farm cases, for which boundary layer entrainment plays an import role. 1 The authors acknowledge support from the European Research Council (FP7-Ideas, grant no. 306471). Simulations were performed on the computing infrastructure of the VSC Flemish Supercomputer Center, funded by the Hercules Foundation and the Flemish Govern 5:37PM E24.00005 An ideal limit for the performance of a large, fully-developed wind farm , P. LUZZATTO-FEGIZ, C.P. CAULFIELD, University of Cambridge — Wind turbines are often deployed in arrays of hundreds of units, where interactions lead to drastic losses in power output. Remarkably, while the theoretical “Betz” maximum has long been established for the output of a single turbine, no corresponding theory appears to exist for a generic, large-scale energy extraction system, although models exist for specific turbine designs and layouts. Recent work with vertical-axis turbines indicates that large performance gains may be achievable (Dabiri 2011), making the search for a theoretical upper bound even more compelling. We develop a model for an array of energy-extraction devices of arbitrary design and layout, first focusing on the fully-developed regime. When tailoring the model to reflect current designs, the predicted power output is in good agreement with field measurements. Furthermore, by considering a suitable ideal limit, we establish an upper bound on the performance of a large wind farm. This is found to be several times larger than the output of existing arrays, thus supporting the notion that performance improvements may be possible. Finally, we extend our model to include spatially developing flows, as well as to account for the effect of atmospheric stability, finding good agreement with laboratory and field data. 5:50PM E24.00006 Effect of the tip speed ratio in the power production of aligned wind turbines1 , KENNETH CARRASQUILLO, CHRISTIAN SANTONI, MARIO ROTEA, YAOYU LI, STEFANO LEONARDI, The University of Texas at Dallas — The increased demand for wind energy had led to a constant increase in the size of wind turbines and subsequently of the wind farms. A drawback of using large arrays of wind turbines is the decrease in efficiency due to the wake interference. For example, the second row of turbines extracts about 15% less power than the first row. Previous studies indicated that the power production of the entire wind farm is not maximized if the turbines work at their optimum tip speed ratio (TSR). In fact, reducing the TSR on the upwind turbines with respect to an optimum value, the momentum deficit decreases and the downwind turbines power production increases. Although the power production on the upwind turbines decreases, the power production of the entire wind plant may increase. Large Eddy Simulations of the turbulent flow over three NREL5MW aligned turbines have been performed. The most downwind turbine is kept at maximum power production with TSR=7.5, while the TSR of the other two turbines is varied. The effect of the TSR on power production and its fluctuations will be discussed. The UTDWF code is used to perform the simulations, which is based on a finite difference scheme with the Line Actuator to model the turbine blades and the Immersed Boundary Method for the tower and nacelle. 1 The numerical simulations were performed on XSEDE TACC under Grant No. CTS070066. This work was supported by the NSF, grant IIA-1243482 (WINDINSPIRE). Sunday, November 23, 2014 4:45PM - 6:03PM Session E25 Turbulence: Multiphase Flow — 2005 - Igor Bolotnov, North Carolina State University 4:45PM E25.00001 Drag Modification by Micro-bubbles in Taylor-Couette Turbulence: A Numerical Approach , VAMSI SPANDAN, RODOLFO OSTILLA-MONICO, ROBERTO VERZICCO, DETLEF LOHSE, Physics of Fluids, University of Twente — We simulate two phase Taylor-Couette (flow between two co-axial independently rotating cylinders) using the Euler-Lagrange approach in which bubbles are treated as point particles with effective forces such as drag, lift and added mass acting on them. The outer cylinder is stationary, while the inner cylinder is rotated to reach a Reynolds number Re ∼ 104 with almost 105 bubbles dispersed into the carrier phase.Two-way coupling is implemented between the dispersed phase and the carrier phase allowing us to study the effect of these point like bubbles on the overall structure of the flow. The two-way coupling is implemented through a unique forcing scheme where the back reaction from a single bubble is spread out over a finite computational volume rather than a finite number of nodes as previously done in literature, which ensures grid independent results. We observe that the bubbles are responsible for disrupting the coherent vortical structures in the carrier flow ultimately resulting in drag modification. In addition we also study the spatial distribution and effect of neutrally buoyant particles dispersed into the flow. 4:58PM E25.00002 Direct numerical simulation of turbulent core-annular flow in a vertical pipe1 , KIYOUNG KIM, HAECHEON CHOI, Seoul Natl Univ — The core-annular flow has been considered as a useful tool to effectively transport highly viscous oil by having lower viscous fluid such as water near the pipe surface. There have been several studies to investigate turbulent core-annular flows but most of them have been conducted experimentally. We solve the three-dimensional Navier-Stokes equations in a cylindrical coordinate and use the level-set method for interface tracking between two fluids (oil and water). A few different flow parameters such as the superficial velocity of fluids and mean pressure gradient are considered in a vertical pipe. The results show that the oil core region is nearly a plug flow and the water region experiences high shear rates, which generate turbulence structures different from those of single phase flow. The interface wave suppresses the near-wall coherent structures but produces complex fluid motions caused by its interaction with the wall. The phenomenon of maximum drag reduction and the effect of water turbulence on total drag will be discussed at the presentation. 1 We gratefully acknowledge financial support from the NRF programs(No. 2012M2A8A4055647), Mest, Korea 5:11PM E25.00003 Numerical study of bubble generation in a turbulent two-phase Couette flow1 , ANDREY OVSYANNIKOV, ALI MANI, PARVIZ MOIN, Stanford University, DOKYUN KIM, Cascade Technologies Inc. — The objective of this work is to develop an understanding bubble generation mechanism due to interactions between free surfaces and turbulent boundary layers as commonly seen near ship walls. To this end, we have focused on a canonical problem that involves Couette flow between two vertical parallel walls with an air-water interface in between. We have considered flow at Reynolds number of 8000 and Froude number of 3.6, both based on half domain dimension and water properties. Our calculations resolve both Kolmogorov lengths and the Hinze scale. Additionally, a conservative VOF method coupled to a subgrid Lagrangian breakup model is used to represent the ligament breakup phenomena and their resulting bubbles and drops. We will present results from these calculations revealing bubble formation rates, bubble size distribution, and effects of bubbles on modulation of turbulence 1 Supported by ONR 5:24PM E25.00004 Single deformable bubble interaction with turbulence in uniform and shear flows1 , JINYONG FENG, IGOR BOLOTNOV, North Carolina State University — Combined direct numerical simulation (DNS) and interface tracking method (ITM) approach is utilized to study the effect of bubble deformability on the bubble-induced turbulence. Set of simulations is performed with 5mm diameter bubble in laminar and turbulent flows. Uniform shear and constant mean velocity profiles are used to perform evaluation of bubble-induced turbulence in various cases. The simulation capabilities allow estimating the turbulent kinetic energy before and after the bubble thus providing the information about bubble’s influence on the liquid turbulence. The effect of bubble deformability is studied by separately changing the surface tension parameter. The bubble is controlled in one location of the domain using external forces. The force evolution is managed by proportional-integral-derivative (PID) controller. The steady-state values of the lateral and stream-wise forces result in the lift and drag force estimates on the bubble. DNS approach allows for comprehensive, well-defined studies of bubble-induced turbulence and interfacial forces by separately varying bubble’s deformability, relative velocity, level of turbulence and local shear. This work presents new opportunities for the development of multiphase computational fluid dynamics closure laws. 1 The presented work is supported by the National Science Foundation under Grant No. 1333993. 5:37PM E25.00005 The evaporation of dense sprays as a mixing process , ALOIS DE RIVAS, EMMANUEL VILLERMAUX, Aix-Marseille Université, France — A dense spray of micron-sized droplets (water or ethanol) is formed in air by a pneumatic atomizer in a closed chamber, and is then conveyed through a nozzle in ambient air, forming a plume whose extension depends on the relative humidity of the diluting medium. We focus on the dry ambient medium, and large plume Reynolds number limit. Standard shear instabilities develop at the plume edge, forming the stretched lamellar structures familiar with passive scalars, except that these vanish in a finite time, because individual droplets evaporate at their border. Experiments also demonstrate that the lifetime of an individual droplet embedded in a lamellae is much larger than expected from the usual d-square law for an isolated droplet. By analogy with the way mixing
times are understood from the convection-diffusion equation for passive scalars, we show that the lifetime of a lamellae where φ is a parameter which incorporates the thermodynamic and diffusional properties of the vapor in the diluting stretched at a rate γ is tv = γ1 ln 1+φ φ phase. The droplets field thus behaves as a -non conserved- passive scalar. 5:50PM E25.00006 Suppression of self-organized structure coarsening in homogenous isotropic turbulence , YOUHEI TAKAGI, Osaka University — Self-organized structure by spinodal decomposition is often seen in quenched binary mixture. Complex network structure is formed through coarsening process of self-organized structure when the phase separation due to spinodal decomposition proceeds. The phase separation governed by the Cahn-Hilliard equation have been well investigated for stationary fluid in previous studies, however, the turbulent effect on the formation of structures was not fully discussed. In this study, we carried out a numerical simulation for homogenous isotropic turbulence with phase separation, the relation between turbulent vortex formation and self-organized structure coarsening. The governing equations are incompressible Navier-Stokes equation considering phase separation force and Cahn-Hilliard equation with the chemical potential based on the Landau-Ginzburg free energy. From the identification and visualization of turbulent structures, it was found that the local entrainment of small eddy structure suppressed the coarsening process of self-organized structure. The energy used in phase separation was related to the initial process of vortex sheet-tube transition in turbulent flow, and the energy cascade from large turbulent structure to small eddy was different from that without phase separation. Sunday, November 23, 2014 4:45PM - 6:03PM Session E26 Turbulent Boundary Layers III — 2007 - Ivan Marusic, University of Melbourne 4:45PM E26.00001 Prediction of Frictional Drag over Rough Walls using Surface Statistics1 , KAREN FLACK, MICHAEL SCHULTZ, United States Naval Academy — Although the frictional drag of rough-wall-bounded flows has been studied extensively, several practical questions remain largely unresolved. First, the relationship between the shape of the roughness function in transitionally-rough regime and the surface topography which gives rise to it are not well understood. Second, it is not completely clear which textural parameters best describe a rough surface in a hydraulic sense. Furthermore, the range of roughness wavelengths that influence the skin-friction is not well established. The focus of the present work is to attempt to address these questions with a systematic study of the skin-friction of fifteen rough surfaces that were generated by grit blasting. The hydrodynamic tests were carried out over a large Reynolds number range. Five surfaces were prepared by grit blasting with a single scale blast media. These underwent hydrodynamic testing and were subsequently blasted with secondary and tertiary scale media in order to investigate the role that the incorporation of additional roughness length scales plays in determining the shape of the roughness function and the resulting hydraulic length scale. The presentation will focus on the appropriate statistical scales for prediction of the roughness function. Spatial filtering prior to the calculation of surface statistics will also be discussed. 1 Work supported by the Office of Naval Research 4:58PM E26.00002 Effects of Sudden Change in Surface Roughness on Turbulent Boundary Layers , RONALD HANSON, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — In almost all practical applications the Reynolds number of the turbulent boundary layer is high and the separation between the inner and outer layer scales become appreciable. Interaction between these scales has profound consequences for the control of turbulent wall flows. In this experimental study we consider the response of the turbulent boundary layer occurring over a surface which transitions from a rough to smooth boundary condition. The transition in surface condition leads to the formation of an internal layer. Above this layer the flow is characteristic of the upstream condition. Within the internal layer the near-wall turbulence establishes itself under the influence of the outer region that remains from the incoming rough-wall. We examine the interaction between the newly established near-wall region and the outer region that persists from upstream conditions. Single and two-component hot-wire measurements are performed simultaneously; the single wire is used to improve the spatial resolution of the measurement of u and to acquire near-wall data. Wide-field Particle Image Velocimetry measurements capture the entire development of the boundary layer over the smooth wall downstream of the roughness. 5:11PM E26.00003 Reorganisation of the large-scale structures in turbulent boundary layers using highly ordered and directional surface roughness , - KEVIN, BAGUS NUGROHO, University of Melbourne, GOKUL PATHIKONDA, JULIO BARROS, University of Illinois, KENNETH CHRISTENSEN, University of Notre Dame, JASON MONTY, NICHOLAS HUTCHINS, University of Melbourne, UOM - UIUC RIBLETS STUDY COLLABORATION1 — The potential of riblet-type surface roughness with converging-diverging (herring-bone type) arrangements to reorganise the large-scale coherent structures that populate the logarithmic region of turbulent boundary layers is investigated at moderate Reynolds number. The ability of this transitionally rough surface to generate large-scale counter rotating roll-modes suggests that a preferential arrangement of the naturally occurring large-scale structures may have been introduced. Prior analysis of the pre-multiplied energy spectra of streamwise velocity fluctuation indicates an increase (or decrease) in the large-scale streamwise turbulence energy over the converging region (or diverging) of the riblets. In this study we examine this possible spanwise redistribution of the coherent structures using instantaneous planar Particle Image Velocimetry (PIV) in the wall-parallel plane (within the logarithmic region) as well as cross-plane Stereoscopic PIV. The characteristics of the large-scale structure over the converging-diverging surface are compared with those of the corresponding smooth-wall case, revealing pronounced modification of the size, strength and alignment of these features over the directional surface. 1 Collaboration between University of Melbourne and University of Illinois on converging-diverging riblets study 5:24PM E26.00004 Wind Tunnel Simulation of the Atmospheric Boundary Layer1 , TRISTEN HOHMAN, TYLER VAN BUREN, Princeton Univ, ALEXANDER SMITS, Princeton Univ, Monash Univ, LUIGI MARTINELLI, Princeton Univ — We aim to generate an artificially thickened boundary layer in the wind tunnel with properties similar to the neutral atmospheric boundary layer (ABL). We implement a variant of Counihan’s technique which uses a combination of a castellated barrier, elliptical vortex generators, and floor roughness elements to create a thick boundary layer in a relatively short wind tunnel. We demonstrate an improved spanwise uniformity than in Counihan’s original design by using a tighter vortex generator spacing with a smaller wedge angle to keep frontal area approximately constant. This is achieved while keeping the turbulence intensity and power spectral density unchanged. It was found possible to generate a boundary layer at Reθ ∼ 106 , displaying logarithmic mean velocity behavior, a constant stress region, and turbulence intensities that compare favorably with full scale ABL measurements and laboratory rough-wall boundary layers. In addition, the longitudinal power spectral density agrees well with von Kármán’s model spectrum and the integral length scale agrees well with data from ABL measurements. 1 Supported by Princeton University Grand Challenges 5:37PM E26.00005 Statistical structure of spanwise vorticity in high Reynolds number roughwall turbulent boundary layers1 , CALEB MORRILL-WINTER, University of Melbourne, JOSEPH KLEWICKI, University of Melbourne, University of New Hampshire, IVAN MARUSIC, University of Melbourne — A defining characteristic of boundary layers is the presence of vorticity. Within the 2-D turbulent boundary layer the only component of vorticity to have a non-negligible mean value is the spanwise component, ωz . In the present experiments, a compact four element (“Foss-style”) hotwire probe was used to acquire well-resolved ωz fluctuations over the range, 3, 000 ≤ δ + = δuτ /ν ≤ 20, 000 for 36 grit sandpaper roughness. Over the entire Reynolds number range good spatial resolution was maintained by utilizing the low speed, large scale attributes of the HRNBLWT at the University of Melbourne. The present talk addresses the statistical structure of ωz above a rough wall including comparisons with its smooth wall counterpart. The observed low Reynolds number smooth wall self-similarity between the mean and the rms profiles of ωz is clarified for the rough-wall case. The rough wall ωz behavior is described in a context consistent with the mean momentum equation. 1 The support of the Australian Research Council is gratefully acknowledged. 5:50PM E26.00006 Cylinder array height effects on evolution of tracked vortex packets within a turbulent boundary layer , YAN MING TAN, ELLEN LONGMIRE, University of Minnesota — A zero pressure gradient turbulent boundary layer with Reτ = 2480 was perturbed by a spanwise array of cylinders. When a narrowly spaced array extended to the top of the log region, perturbed packets appeared to reorganize via a top-down mechanism, suggesting that packet organization can originate from above. We test this hypothesis by extending the array height to the edge of the boundary layer to completely disrupt the packet organization. On the other hand, previous measurements showed that the downstream packet organization was reinforced by an array spacing matching the dominant spanwise spacing of unperturbed packets. A shorter array with reduced blockage was tested to see whether the same effect is achievable. To compare the flow organization in the different cases, fixed and flying PIV measurements were obtained in streamwise-spanwise planes at multiple wall normal locations. The flying PIV system allows tracking and quantification of packet evolution through the array and over a distance of 7δ downstream. Sunday, November 23, 2014 4:45PM - 6:03PM Session E27 Turbulence: Polymers/Drag Reduction — 2009 - Yves Dubief, University of Vermont 4:45PM E27.00001 Turbulence scalings in pipe flows exhibiting polymer-induced drag reduction , IVAN ZADRAZIL, CHRISTOS MARKIDES, Imperial College London — Non-intrusive laser based diagnostics technique, namely Particle Image Velocimetry, was used to in detail characterise polymer induced drag reduction in a turbulent pipe flow. The effect of polymer additives was investigated in a pneumatically-driven flow facility featuring a horizontal pipe test section of inner diameter 25.3 mm and length 8 m. Three high molecular weight polymers (2, 4 and 8 MDa) at concentrations of 5 – 250 wppm were used at Reynolds numbers from 35000 to 210000. The PIV derived results show that the level of drag reduction scales with different normalised turbulence parameters, e.g. streamwise and spanwise velocity fluctuations, vorticity or Reynolds stresses. These scalings are dependent of the distance from the wall, however, are independent of the Reynolds numbers range investigated. 4:58PM E27.00002 Tampering with the turbulent energy cascade with polymer additives1 , PEDRO VALENTE, CARLOS DA SILVA, IST - U. Lisbon, FERNANDO PINHO, FEUP - U. Porto — We show that the strong depletion of the viscous dissipation in homogeneous viscoelastic turbulence reported by previous authors does not necessarily imply a depletion of the turbulent energy cascade. However, for large polymer relaxation times there is an onset of a polymer-induced kinetic energy cascade which competes with the non-linear energy cascade leading to its depletion. Remarkably, the total energy cascade flux from both cascade mechanisms remains approximately the same fraction of the kinetic energy over the turnover time as the non-linear energy cascade flux in Newtonian turbulence. 1 The authors acknowledge the funding from COMPETE, FEDER and FCT (grant PTDC/EME-MFE/113589/2009) 5:11PM E27.00003 On the connection between Maximum Drag Reduction and Newtonian fluid flow , RICHARD WHALLEY, University of Liverpool, JAE-SUNG PARK, ANUBHAV KUSHWAHA, University of Wisconsin-Madison, DAVID DENNIS, University of Liverpool, MICHAEL GRAHAM, University of Wisconsin-Madison, ROBERT POOLE, University of Liverpool — To date, the most successful turbulence control technique is the dissolution of certain rheology-modifying additives in liquid flows, which results in a universal maximum drag reduction (MDR) asymptote. The MDR asymptote is a well-known phenomenon in the turbulent flow of complex fluids; yet recent direct numerical simulations of Newtonian fluid flow have identified time intervals showing key features of MDR. These intervals have been termed “hibernating turbulence” and are a weak turbulence state which is characterised by low wall-shear stress and weak vortical flow structures. Here, in this experimental investigation, we monitor the instantaneous wall-shear stress in a fully-developed turbulent channel flow of a Newtonian fluid with a hot-film probe whilst simultaneously measuring the streamwise velocity at various distances above the wall with laser Doppler velocimetry. We show, by conditionally sampling the streamwise velocity during low wall-shear stress events, that the MDR velocity profile is approached in an additive-free, Newtonian fluid flow. This result corroborates recent numerical investigations, which suggest that the MDR asymptote in polymer solutions is closely connected to weak, transient Newtonian flow structures. 5:24PM E27.00004 Influence of large-scale motions on the effectiveness of active drag-reduction control in wall turbulence1 , BING-QING DENG, CHUN-XIAO XU, WEI-XI HUANG, GUI-XIANG CUI, Tsinghua University — To investigate the influence of the large-scale motions on the effectiveness of the active drag-reduction control, direct numerical simulations are performed in turbulent channel flows with opposition control at Reτ =180 and 1000. Consistent with the results of Chang et al. (Phys. Fluids, vol. 14, pp. 4069–4080, 2002), the drag reduction rate of the opposition control decays with Reynolds number. In the outer layer, by influencing the mean shear, the control imposed on the wall can reduce the Reynolds stress at the same rate as the drag reduction, while the distribution of the energy at different scales is little different than the uncontrolled case. In the near-wall region at Reτ =1000, suppression of the near-wall structures under the large-scale high-speed streaks by the control are much weaker than those under the region of the large-scale low-speed streaks, which leads to the falloff of the effectiveness of the control in suppressing the near-wall turbulence at high Reynolds numbers. By further analyzing the drag reduction rates, it is found that the effectiveness of the control is mainly determined by the suppression degree of the near-wall motions which is influenced by the large-scale motions. 1 The work was supported by National Natural Science Foundation of China (Grant numbers 11132005, 11322221). 5:37PM E27.00005 Multi-scale study on process of contravariant and covariant polymer elongation and drag reduction in viscoelastic turbulence , KIYOSI HORIUTI, SHU SUZUKI, Dept. Mechano-Aerospace Engineering, Tokyo Institute of Technology, Japan — We study the elongation process of polymers released in the Newtonian homogeneous isotropic turbulence by connecting a mesoscopic description of ensemble of elastic dumbbells using Brownian dynamics (BDS) to the macroscopic description for the fluid using DNS. The dumbbells are allowed to be advected non-affinely with the macroscopically-imposed deformation. More drastic drag reduction is achieved when non-affinity is maximum than in the complete affine case. In the former case, the dumbbell is convected as a covariant vector, and in the latter as a contravariant vector. We derive the exact solution for the governing equation of the motion of dumbbells. The maximum stretching of dumbbell is achieved when the dumbbell aligns in the direction of vorticity in the contravariant case, and when the dumbbell directs outward perpendicularly on the vortex sheet in the covariant case. Alignment in the BDS-DNS data agrees with the theoretical results. In the mixture of contravariant and covariant dumbbells, the covariant dumbbells are transversely aligned with the contravariant dumbbells. Compared with the cases without mixture, stretching of covariant dumbbell is enhanced, while that of contravariant dumbbell is reduced. Application of this phenomenon is discussed. 5:50PM E27.00006 Breakup of colloidal aggregates in turbulent channel flow1 , ALFREDO SOLDATI, CRISTIAN MARCHIOLI, University of Udine — Breakup of small aggregates in turbulence is of high relevance to industrial applications (from processing of colloids and nanomaterials to flocculation) and environmental processes (marine snow formation). In spite of their importance, breakup phenomena are poorly understood from a fundamental viewpoint and a basic description of breakup dynamics is still lacking. In this work we examine the complex role of turbulence and the way it generates fluctuating hydrodynamic stresses to which an aggregate is exposed. We use pseudo-spectral DNS and Lagrangian tracking to determine the breakup rate of sub-Kolmogorov colloidal massless aggregates in non-homogeneous anisotropic turbulence, considering both instantaneous and ductile breakup. Instantaneous breakup occurs when the stress generated by the surrounding fluid exceeds the critical value required to break that aggregate: σ > σcr . Ductile breakup is consequence of a non-instantaneous process activated when σ > σcr and occurs when the energy dissipated by the surrounding R fluid, E = ǫ(τ |σ > σcr )dτ with ǫ the fluid kinetic energy dissipation rate and τ time, exceeds the critical breakup value. Effects on breakup rates due to aggregate inertia will also be discussed. 1 Financial support from grant LR14-2010 “Sviluppo di filtri catalitici e antiparticolato ad alta efficienza per una sostenibile mobilita’ compatibile con Euro 6” and from COST Action FP1005 “Fiber suspension flow modelling” is gratefully acknowledged. Sunday, November 23, 2014 4:45PM - 6:03PM Session E28 Compressible Turbulence 2011 - Carlo Scalo, Purdue University — 4:45PM E28.00001 Effects of shock structure on temperature field in compressible turbulence , QIONGLIN NI, SHIYI CHEN, College of Engineering, Peking University, Beijing 100871, China — Effects of shock structure on temperature in compressible turbulence were investigated. The small-scale shocklets and large-scale shock waves were appeared in the flows driven by solenoidal and compressive forcings, i.e. SFT & CFT, respectively. In SFT the temperature had Kolmogorov spectrum and ramp-cliff structures, while in CFT it obeyed Burgers spectrum and was dominated by large-scale rarefaction and compression. The power-law exponents for the p.d.f. of large negative dilatation were -2.5 in SFT and -3.5 in CFT, approximately corresponded to model results. The isentropic approximation of thermodynamic variables showed that in SFT, the isentropic derivation was reinforced when turbulent Mach number increased. At similar turbulent Mach number, the variables in CFT exhibited more anisentropic. It showed that the transport of temperature was increased by the small-scale viscous dissipation and the large-scale pressure-dilatation. The distribution of positive and negative components of pressure-dilatation confirmed the mechanism of negligible pressure-dilatation at small scales. Further, the positive skewness of p.d.f.s of pressure-dilatation implied that the conversion from kinetic to internal energy by compression was more intense than the opposite process by rarefaction. 4:58PM E28.00002 Compressible turbulent channel flow with impedance boundary conditions , CARLO SCALO, Stanford University, JULIEN BODART, Universite de Toulouse, ISAE, SANJIVA LELE, Stanford University — We have performed large-eddy simulations of compressible turbulent channel flow at one bulk Reynolds number, Reb = 6900, for bulk Mach numbers Mb = 0.05, 0.2, 0.5, with linear acoustic impedance boundary conditions (IBCs). The IBCs are formulated in the time domain following Fung and Ju (2004) and coupled with a Navier-Stokes solver. The impedance model adopted is a three-parameter Helmholtz oscillator with resonant frequency tuned to the outer layer eddies. The IBC’s resistance, R, has been varied in the range, R = 0.01, 0.10, 1.00. Tuned IBCs result in a noticeable drag increase for sufficiently high Mb and/or low R, exceeding 300% for Mb = 0.5 and R = 0.01, and thus represents a promising passive control technique for delaying boundary layer separation and/or enhancing wall heat transfer. Alterations to the turbulent flow structure are confined to the first 15% of the boundary layer thickness where the classical buffer-layer coherent vortical structures are replaced by an array of Kelvin-Helmholtz-like rollers. The non-zero asymptotic value of the Reynolds shear stress gradient at the wall results in the disappearance of the viscous sublayer and very early departure of the mean velocity profiles from the law of the wall. 5:11PM E28.00003 Numerical investigation of non-equilibrium effects in hypersonic turbulent boundary layers , PILBUM KIM, JOHN KIM, XIAOLIN ZHONG, JEFF ELDREDGE, University of California, Los Angeles — Direct numerical simulations of a spatially developing hypersonic boundary layer have been conducted in order to investigate thermal and chemical non-equilibrium effects in a hypersonic turbulent boundary layer. Two different flows, pure oxygen and pure nitrogen flows with specific total enthalpy, h0,O2 = 9.5017 M J/kg and h0,N2 = 19.1116 M J/kg, respectively, have been considered. The boundary edge conditions were obtained from a separate calculation of a flow over a blunt wedge at free-stream Mach numbers M∞,O2 = 15 and M∞,N2 = 20. The inflow conditions were obtained from a simulation of a turbulent boundary layer of a perfect gas. Non-equilibrium effects on turbulence statistics and near-wall turbulence structures were examined by comparing with those obtained in a simulation of the same boundary layer with a perfect-gas assumption. 5:24PM E28.00004 Reynolds Stress Anisotropy Effects on the Modeling of Production in Compressible Homogeneous Turbulent Shear Flow , JOHN PANICKACHERIL, GREGORY BLAISDELL, Purdue University — Direct numerical simulations of compressible homogeneous shear flow are performed for a range of gradient Mach numbers and time scale ratios. A pseudospectral Fourier collocation method is used to perform the simulations. Compressibility effects associated with larger gradient Mach numbers result in increased anisotropy of the Reynolds stresses and a reduced growth rate of turbulent kinetic energy. However, the time scale ratio also affects Reynolds stress anisotropy. The DNS results seem to indicate that the model coefficient Cµ in the production term of the turbulent kinetic energy equation is mainly dependent on the time scale ratio for low gradient Mach numbers and collapses to a single function of gradient Mach number for higher gradient Mach numbers. A model for Cµ is formulated based on this behavior in the DNS. Computations using this model are then compared to the DNS results. 5:37PM E28.00005 Fluctuations of thermodynamic variables in compressible isotropic turbulence , DIEGO DONZIS, SHRIRAM JAGANNATHAN, Texas A&M Univ — A distinguishing feature of compressible turbulence is the appearance of fluctuations of thermodynamic variables. While their importance is well-known in understanding these flows, some of their basic characteristics such as the Reynolds and Mach number dependence are not well understood. We use a large database of Direct Numerical Simulation of stationary compressible isotropic turbulence on up to 20483 grids at Taylor Reynolds numbers up to 450 and a range of Mach numbers (Mt ≈ 0.1 − 0.6) to examine statistical properties of thermodynamic variables. Our focus is on the PDFs and moments of pressure, density and temperature. While results at low Mt are consistent with incompressible results, qualitative changes are observed at higher Mt with a transition around Mt ∼ 0.3. For example, the PDF of pressure changes from negatively to positively skewed as Mt increases. Similar changes are observed for temperature and density. We suggest that large fluctuations of thermodynamic variables will be log-normal at high Mt . We also find that, relative to incompressible turbulence, the correlation between enstrophy and low-pressure regions is weakened at high Mt which can be explained by the dominance of the so-called dilatational pressure. 5:50PM E28.00006 The effect of thermal non-equilibrium in decaying turbulence using direct numerical simulations , SUALEH KHURSHID, DIEGO DONZIS, Texas A&M University — We study effects of thermal non-equilibrium (TNE), in particular vibrational non-equilibrium, in decaying turbulence using Direct Numerical Simulation (DNS). The exchange mechanism between molecular vibrational and translational energy modes is introduced using the well-known Landau-Teller approximation. A change in the fundamental cascade is observed with dissipation (ǫ) increasing significantly relative to cases without TNE at time scales of O(τv ) where τv is the characteristic relaxation time of vibrational energy. This is also found to depend on the initial degree of TNE (∆Ev0 ). The relative contributions of energy transfer through classical energy cascade and transfer v0 through TNE exchanges can be represented by a new non-dimensional parameter S = ∆E . For example, S can be used to understand DNS data, in particular τv ǫ to distinguish different regimes in the interaction. S is also useful to characterize the time at which dissipation peaks as well as its peak value. Results are compared satisfactorily with experimental evidence available. Turbulence is also observed to decelerate the transfer from vibrational to translational mode in flows with initial vibrationally hot states. Sunday, November 23, 2014 4:45PM - 6:03PM Session E29 Experimental Techniques: Multiphase Flow — 2014 - Charles A. Petty, Michigan State University 4:45PM E29.00001 Three-Dimensional Feature Extraction from Multiphase Flows , BARRY SCHARFMAN, ALEXANDRA TECHET, Massachusetts Inst of Tech-MIT — Light field imaging (LFI) and synthetic aperture (SA) refocusing techniques have been combined in an emerging method to resolve three-dimensional (3D) flow fields over time. Image volumes of a scene are captured using an array of multiple cameras. SA refocusing yields a stack of post-processed images at different focal depths, each with a narrow depth of field. Although this technique has previously been used to reconstruct flow features that are small relative to the field of view, blur artifacts are more clearly visible when this method is applied to relatively larger features. The presence of these artifacts prevents 3D scene reconstruction. To eliminate the artifacts, circles are detected in the raw camera images, and their rims are converted to white pixels, while the rest of each raw image is made black. This simplifies 3D feature detection in the stack of refocused images and allows the scene to be reconstructed in 3D. Simulations and experiments using the aforementioned modified SA method show that it is possible to extract the center coordinates in 3D and radii of spheres found in a scene being recorded with simple back illumination. This technique has been applied to various types of multiphase flows, including bubble flow fields in air and sneezes. 4:58PM E29.00002 Phase-locked measurements of gas-liquid horizontal flows1 , IVAN ZADRAZIL, OMAR MATAR, CHRISTOS MARKIDES, Imperial College London — A flow of gas and liquid in a horizontal pipe can be described in terms of various flow regimes, e.g. wavy stratified, annular or slug flow. These flow regimes appear at characteristic gas and liquid Reynolds numbers and feature unique wave phenomena. Wavy stratified flow is populated by low amplitude waves whereas annular flow contains high amplitude and long lived waves, so called disturbance waves, that play a key role in a liquid entrainment into the gas phase (droplets). In a slug flow regime, liquid-continuous regions travel at high speeds through a pipe separated by regions of stratified flow. We use a refractive index matched dynamic shadowgraphy technique using a high-speed camera mounted on a moving robotic linear rail to track the formation and development of features characteristic for the aforementioned flow regimes. We show that the wave dynamics become progressively more complex with increasing liquid and gas Reynolds numbers. Based on the shadowgraphy measurements we present, over a range of conditions: (i) phenomenological observations of the formation, and (ii) statistical data on the downstream velocity distribution of different classes of waves. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 5:11PM E29.00003 Hard X-ray nanotomography of colloidal suspensions1 , YESEUL KIM, SU JIN LIM, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University, JUN LIM, Beamline Division, Pohang Light Source, BYUNG MOOK WEON, School of Advanced Materials Science and Engineering, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University — Colloidal suspensions are complex fluids that include colloidal nanoparticles or microparticles suspended in a liquid medium. In-situ characterizations of colloidal suspensions are necessary in many topics: for instance, wetting properties for colloidal particles on a fluid-fluid interface are essential but hard to be directly taken with conventional imaging techniques. Here we show that hard X-ray nanotomography clearly visualizes individual colloidal particles inside fluids in three dimensions (3D). In particularly, we demonstrate 3D images for colloidal particles adsorbed on water-oil emulsions: contact angle and configuration of colloids could be measured. We believe that hard X-ray nanotomography would be a powerful tool to identify the nature of colloidal particles inside or on fluids. 1 This work (NRF-2013R1A22A04008115) was supported by Mid-career Researcher Program through NRF grant funded by the MEST. 5:24PM E29.00004 High-speed imaging of the transient ice accretion process on a NACA 0012 airfoil1 , RYE WALDMAN, HUI HU, Iowa State University — Ice accretion on aircraft wings poses a performance and safety threat as aircraft encounter supercooled droplets suspended in the cloud layer. The details of the ice accretion depend on the atmospheric conditions and the fight parameters. We present the measurement results of the experiments conducted in the Iowa State icing wind tunnel on a NACA 0012 airfoil to study the transient ice accretion process under varying icing conditions. The icing process on the wing consists of a complex interaction of water deposition, surface water transport, and freezing. The aerodynamics affects the water deposition, the heat and mass transport, and ice accumulation; meanwhile, the accumulating ice also affects the aerodynamics. High-speed video of the unsteady icing accretion process was acquired under controlled environmental conditions to quantitatively measure the transient water run back, rivulet formation, and accumulated ice growth, and the experiments show how varying the environmental conditions modifies the ice accretion process. 1 Funding support from the Iowa Energy Center with Grant No. 14-008-OG and National Science Foundation (NSF) with Grant No. CBET- 1064196 and CBET- 1438099 is gratefully acknowledged. 5:37PM E29.00005 An Experimental Investigation on the Wind-Driven Water Film Flows over Rough Arrays by using a Digital Image Projection (DIP) Technique1 , KAI ZHANG, ALRIC ROTHMAYER, HUI HU, Iowa State University, IOWA STATE UNIVERSITY TEAM — In the present study, an experimental investigation was conducted to quantify the transient behavior of the wind-driven surface water film flows over a rouged surface in order to examine the water mass trapped effect due to the presence of roughness arrays pertinent to aircraft icing phenomena. A novel digital image projection (DIP) technique was developed and applied to achieve time-resolved measurements of the thickness distributions of the unsteady surface water film flows over the roughness arrays, in comparison with those over a flat plate as the comparison baselines. The measurement results reveal clearly that, at relatively low wind speed, the roughness arrays would perform as a dam to block the wind-driven water film flow at the front side of the roughness arrays. For the cases with relatively high wind speeds, the trapped water mass was found to stagnate mainly at the backside of the roughness arrays. The time-averaged mass trapping ratio was found to be very sensitive to the wind speed, but less sensitive to the flow rate of the surface water film flows over the test plate. 1 Funding support from National Aeronautical and Space Administration (NASA) with Grant No. NNX12AC21A and National Science Foundation (NSF) with Grant No. CBET-1435590 is gratefully acknowledged. 5:50PM E29.00006 Velocity and thickness measurement of a thin-liquid film via a single-tip optical fiber probe micro-fabricated by femtosecond pulse laser , YUSUKE IKEDA, Faculty of Engineering, Shizuoka University, YUKI MIZUSHIMA, Graduate school of Science and Technology, Shizuoka Univerisity, TAKAYUKI SAITO, Research Institute of Green Science and Technology, Shizuoka University — Optical fiber probing is a simple and compact measurement system for a gas-liquid two phase flow. This probe detects a gas-liquid interface responsively. We have developed a new measurement technique for a thin-liquid-film that utilizes a single-tip optical fiber probe (Fs-TOP) micro-fabricated through femtosecond laser pulses. The Fs-TOP is installed horizontally along the channel base, and vertically traversed to the other side of the base. The signal from the Fs-TOP is sufficiently understood by using the originally developed 3D ray-tracing-numerical simulation. The maximum liquid film thickness was measured as well as the average liquid film thickness. Based on the simulation, it was found out that when a fraction of liquid phase is 52%, the installed position of the Fs-TOP is equal to the average liquid film thickness. We measured velocity of the liquid film by the Fs-TOP and visualization. The results accorded with each other, and implied that the Fs-TOP measurement was superior to the visualization, against a high-velocity flow. Sunday, November 23, 2014 4:45PM - 6:03PM Session E30 Aerodynamics: General — 2016 - 4:45PM E30.00001 Apparent Mass Nonlinearity for Paired Oscillating Plates , KENNETH GRANLUND, MICHAEL OL, US Air Force Research Lab — The classical potential-flow problem of a plate oscillating sinusoidally at small amplitude, in a direction normal to its plane, has a well-known analytical solution of a fluid “mass,” multiplied by plate acceleration, being equal to the force on the plate. This so-called apparent-mass is analytically equal to that of a cylinder of fluid, with diameter equal to plate chord. The force is directly proportional to frequency squared. Here we consider experimentally a generalization, where two coplanar plates of equal chord are placed at some lateral distance apart. For spacing of ∼ 0.5 chord and larger between the two plates, the analytical solution for a single plate can simply be doubled. Zero spacing means a plate of twice the chord and therefore a heuristic cylinder of fluid of twice the cross-sectional area. This limit is approached for plate spacing <0.5c. For a spacing of 0.1-0.2c, the force due to apparent mass was found to increase with frequency, when normalized by frequency squared; this is a nonlinearity and a departure from the classical theory. Flow visualization in a water-tank suggests that such departure can be imputed to vortex shedding from the plates’ edges inside the inter-plate gap. 4:58PM E30.00002 Numerical study on the aerodynamics of a golf ball and its comparison with a smooth sphere , JING LI, MAKOTO TSUBOKURA, Faculty of Engineering, Hokkaido University, MASAYA TSUNODA, Sumitomo Rubber Industries Ltd. — The present study has numerically investigated the flow over a golf ball and a smooth sphere by conducting large-eddy simulation (LES) using hundreds of millions of unstructured elements. Simulations were conducted at various Reynolds numbers ranging from the subcritical to the supercritical regimes. Special attention was paid to the phenomenon of drag crisis as well as the effect of surface roughness on the drag crisis. The simulation result shows that the surface roughness introduced by the dimples of the golf ball causes a local instability of the flow around the ball and subsequently leads to a momentum transfer in the near-wall region inside the dimples. The flow with high momentum in the near-wall region travels further downstream, which consequently results in the drag crisis occurring at a relatively lower Reynolds number compared with that of the smooth sphere. Moreover, the Magnus effect resulting from the rotating motion of a sphere was also one of the main concerns in this study. The simulation result shows that lift forces are imposed on both the rotating smooth sphere and rotating golf ball. For most cases the lift force points to the positive direction, however, the negative lift force appears also under certain conditions. 5:11PM E30.00003 Characterisation of turbulence downstream of a linear compressor cascade1 , LUCA DI MARE, Imperial College London, THOMAS JELLY, IVOR DAY, University of Cambridge — Characterisation of turbulence in turbomachinery remains one of the most complex tasks in fluid mechanics. In addition, current closure models required for Reynolds-averaged Navier-Stokes computations do not accurately represent the action of turbulent forces against the mean flow. Therefore, the statistical properties of turbulence in turbomachinery are of significant interest. In the current work, single- and two-point hot-wire measurements have been acquired downstream of a linear compressor cascade in order to examine the properties of large-scale turbulent structures and to assess how they affect turbulent momentum and energy transfer in compressor passages. The cascade has seven controlled diffusion which are representative of high-pressure stator blades found in turbofan engines. Blade chord, thickness and camber are 0.1515 m, 9.3% and 42 degrees, respectively. Measurements were acquired at a chord Reynolds number of 6.92 × 105 . Single-point statistics highlight differences in turbulence structure when comparing mid-span and end-wall regions. Evaluation of two-point correlations and their corresponding spectra reveal the length-scales of the energy-bearing eddies in the cascade. Ultimately, these measurements can be used to calibrate future computational models. 1 The authors gratefully acknowledge Rolls-Royce plc for funding this work and granting permission for its publication. 5:24PM E30.00004 Investigation of Drag Coefficient for Rigid Ballute-like Shapes1 , MARIA-ISABEL CARNASCIALI, ANTHONY MASTROMARINO, Univ of New Haven — One common method of decelerating an object during atmospheric entry, descent, and landing is the use of parachutes. Another deceleration technology is the ballute – a combination of balloon and parachute. A CFD study was conducted using commercially available software to investigate the flow-field and the coefficient of drag for various rigid ballute-like shapes at varying Reynolds numbers. The impact of size and placement of the burble-fence as well as number, size, and shape of inlets was considered. Recent experimental measurements conducted during NASA’s Low-Density Supersonic Decelerator program revealed a much higher coefficient of drag (Cd ) for ballutes than previously encountered. Using atmospheric drag to slow down and land reduces the need for heavy fuel and rocket engines and thus, high values of drag are desired. 1 Funding for this work, in part, provided by the CT Space Grant Consortium. 5:37PM E30.00005 Application of smoothed particle hydrodynamics method in aerodynamics , MIGUEL CORTINA, University of Texas at San Antonio — Smoothed Particle Hydrodynamics (SPH) is a meshless Lagrangian method in which the domain is represented by particles. Each particle is assigned properties such as mass, pressure, density, temperature, and velocity. These properties are then evaluated at the particle positions using a smoothing kernel that integrates over the values of the surrounding particles. In the present study the SPH method is first used to obtain numerical solutions for fluid flows over a cylinder and then we are going to apply the same principle over an airfoil obstacle. 5:50PM E30.00006 Sailing effect on high performance bicycle wheels , FLAVIO NOCA, hepia - University of Applied Sciences - Switzerland, JEAN-PIERRE MERCAT, BRIEUC CRETOUX, FRANCOIS-XAVIER HUAT, MAVIC - France — Recently, MAVIC and hepia (University of Applied Sciences in Switzerland) developed the most aerodynamic bicycle wheel on the market. The key feature of this wheel is its ability to sail in a cross-wind, just like a sailboat. The phenomenon relies on features on the tire itself. While it was thought in the past that wheel/tire smoothness was the key to good performance, our team discovered that adequately designed patterns on the tire allowed cross-winds to remain attached around the front wheel. The flowfield is very similar to that of an airfoil at incidence, and thrust forces (in the direction of travel) can even be generated. Experiments are being conducted in a wind tunnel and in a towing tank in order to examine the aerodynamic influence of patterned structures on the leading edge of airfoils and wheels at intermediate Reynolds numbers. Sunday, November 23, 2014 4:45PM - 6:03PM Session E31 CFD: Higher-Order Methods — 2018 - Timothy Colonius, CalTech University 4:45PM E31.00001 Discontinuous Galerkin Methods and High-Speed Turbulent Flows , MUHAMMED ATAK, University of Stuttgart, JOHAN LARSSON, University of Maryland, CLAUS-DIETER MUNZ, University of Stuttgart — Discontinuous Galerkin methods gain increasing importance within the CFD community as they combine arbitrary high order of accuracy in complex geometries with parallel efficiency. Particularly the discontinuous Galerkin spectral element method (DGSEM) is a promising candidate for both the direct numerical simulation (DNS) and large eddy simulation (LES) of turbulent flows due to its excellent scaling attributes. In this talk, we present a DNS of a compressible turbulent boundary layer along a flat plate at a free-stream Mach number of M=2.67 and assess the computational efficiency of the DGSEM at performing high-fidelity simulations of both transitional and turbulent boundary layers. We compare the accuracy of the results as well as the computational performance to results using a high order finite difference method. 4:58PM E31.00002 CoreSVM: a generalized high-order spectral volume method bearing Conservative Order RElease1 , RAPHAEL LAMOUROUX, JEREMIE GRESSIER, LAURENT JOLY, GILLES GRONDIN, None — The spectral volume method (SVM) introduced by Wang in 2002 is based on a compact polynomial reconstruction where the interpolation’s degree is driven by the partition of the spectral volumes. We propose a generalization of the SVM which releases the polynomial degree from this constraint and more importantly that allows to resort to any polynomial order inferior to the regular stencil order without changing the original spectral volume partition. Using one-dimensional advection and Burgers equation, we prove that the proposed extended method exhibits versatile high-order convergence together with conservativity properties. This new method is thus named the CoreSVM for Conservative Order-REleased SVM and we therefore explore its potential towards the numerical simulation of stiff problems. It is stressed that CoreSVM is indeed particularly suited to handle discontinuities, as the order-reduction serves to damp the numerical oscillations due to Runge’s phenomenon. To ensure computational stability, local p-coarsening is used to obtain the highest adequate polynomial degree. It is advocated finally that, since the CoreSVM sets the polynomial order adaptation free from any stencil changes, these features do not come at the expense of any extra remeshing or data adaptation cost. 1 Part of this research was funded by the French DGA. 5:11PM E31.00003 Stable, high-order SBP-SAT finite difference operators to enable accurate simulation of compressible turbulent flows on curvilinear grids, with application to predicting turbulent jet noise , JAESEUNG BYUN, DANIEL BODONY, CARLOS PANTANO, University of Illinois at Urbana-Champaign — Improved order-of-accuracy discretizations often require careful consideration of their numerical stability. We report on new high-order finite difference schemes using Summation-By-Parts (SBP) operators along with the Simultaneous-Approximation-Terms (SAT) boundary condition treatment for first and second-order spatial derivatives with variable coefficients. In particular, we present a highly accurate operator for SBP-SAT-based approximations of second-order derivatives with variable coefficients for Dirichlet and Neumann boundary conditions. These terms are responsible for approximating the physical dissipation of kinetic and thermal energy in a simulation, and contain grid metrics when the grid is curvilinear. Analysis using the Laplace transform method shows that strong stability is ensured with Dirichlet boundary conditions while weaker stability is obtained for Neumann boundary conditions. Furthermore, the benefits of the scheme is shown in the direct numerical simulation (DNS) of a Mach 1.5 compressible turbulent supersonic jet using curvilinear grids and skew-symmetric discretization. Particularly, we show that the improved methods allow minimization of the numerical filter often employed in these simulations and we discuss the qualities of the simulation. 5:24PM E31.00004 Positivity-preserving and entropy-bounded Discontinuous Galkerin method for conservation laws , YU LV1 , MATTHIAS IHME, Stanford University — Although Discontinuous Galerkin (DG) methods have gained considerable success for application to advection-dominated flows, the robustness and the treatment of geometric singularities and flow-field discontinuities remain open problems. In this talk, a DG-method is formulated that is positivity-preserving and entropy-bounded to guarantee algorithmic stability and conservation. After demonstrating the efficacy in one- and two-dimensional tests, this formulation is generalized to unstructured and curvilinear meshes. Details on the algorithmic implementation are presented, and applications to complex geometries in three dimensions are discussed. 1 Yu Lv was graduated from Zhejiang University, China for his undergraduate. At 2011, he got his Master’s degree for aero, University of Michigan. Now he is working at Stanford as a Ph.D research assistant. 5:37PM E31.00005 A New Approach for Imposing Artificial Viscosity for Explicit Discontinuous Galerkin Scheme , YEE CHEE SEE, YU LV, MATTHIAS IHME, Stanford University — The development of high-order numerical methods for unstructured meshes has been a significant area of research, and the discontinuous Galerkin (DG) method has found considerable interest. However, the DG-method exhibits robustness issues in application to flows with discontinuities and shocks. To address this issue, an artificial viscosity method was proposed by Persson et al. for steady flows. Its extension to time-dependent flows introduces substantial time-step restrictions. By addressing this issue, a novel method, which is based on an entropy formulation, is proposed. The resulting scheme doesn’t impose restrictions on the CFL-constraint. Following a description of the formulation and the evaluation of the stability, this newly developed artificial viscosity scheme is demonstrated in application to different test cases. 5:50PM E31.00006 Towards A Fast High-Order Method for Unsteady Incompressible NavierStokes Equations using FR/CPR1 , CHRISTOPHER COX, CHUNLEI LIANG, MICHAEL PLESNIAK, George Washington Univ — A high-order compact spectral difference method for solving the 2D incompressible Navier-Stokes equations on unstructured grids is currently being developed. This method employs the gGA correction of Huynh, and falls under the class of methods now refered to as Flux Reconstruction/Correction Procedure via Reconstruction. This method and the artificial compressibility method are integrated along with a dual time-integration scheme to model unsteady incompressible viscous flows. A lower-upper symmetric Gauss-Seidel scheme and a backward Euler scheme are used to efficiently march the solution in pseudo time and physical time, respectively. We demonstrate order of accuracy with steady Taylor-Couette flow at Re=10. We further validate the solver with steady flow past a NACA0012 airfoil at zero angle of attack at Re=1850 and unsteady flow past a circle at Re=100. The implicit time-integration scheme for the pseudo time derivative term is proved efficient and effective for the classical artificial compressibility treatment to achieve the divergence-free condition of the continuity equation. 1 We greatly acknowledge financial support from The George Washington University under the Presidential Merit Fellowship Sunday, November 23, 2014 4:45PM - 6:03PM Session E32 Particle-Laden Flows: Particle-Resolved Simulations — 2020 - Andrea Prosperetti, Johns Hopkins University 4:45PM E32.00001 Direct numerical simulation of gravity-driven avalanches immersed in a viscous fluid , THOMAS BONOMETTI, EDOUARD IZARD, LAURENT LACAZE, IMFT, Universite de Toulouse, CNRS/INPT/UPS, OTE TEAM — This work deals with direct numerical simulations of sediment transport at the scale of O(103 ) grains. A soft-sphere discrete element method is coupled to an immersed boundary method in order to compute the flow around moving and colliding grains in an incompressible Newtonian fluid. A lubrication force is added for representing fluid-particles interaction near contact. The numerical method is shown to adequately reproduce the effective coefficient of restitution measured in experiments of the normal and oblique rebound of a grain on a wall. An analytical model is proposed and highlights the importance of the grain roughness and Stokes number on the rebound phenomenon. Three-dimensional configurations of gravity-driven dense granular flows in a fluid, namely the granular avalanche on an inclined plane and the collapse of a granular column, are performed. The granular flow regimes (viscous, inertial and dry) observed in experiments are identified as a function of the grain-to-fluid density ratio and the Stokes number. In particular, the simulations provide insights on the grain and fluid velocity profiles and force balance in each regime. In the second case, results agree well with experiments and the pore pressure feedback is observed for the first time in direct numerical simulations. 4:58PM E32.00002 Turbulent channel flow laden with finite-size neutrally-buoyant particles1 , FRANCESCO PICANO, KTH Mechanics, WIM-PAUL BREUGEM, Aero & Hydrodynamics Dep., TU-Delft, LUCA BRANDT, KTH Mechanics — Dense suspensions are widely encountered in many applications and in environmental flows. While their rheological features in laminar flows have been longly studied, much less is known on their behavior in turbulent/inertial regime. The present works aims to fill this gap by investigating the turbulent channel flow of a Newtonian fluid laden with rigid neutrally-buoyant spheres at relatively high volume fractions. An Immersed Boundary Method has been used to account for the phase interaction performing Direct Numerical Simulation in the range of volume fractions Φ = 0 − 0.2 and a typical particle radius of 10 wall units. The results show that the mean velocity profiles are significantly altered by the presence of a solid phase with a decrease of the von Karman constant in the log-law. The overall drag is found to monotonically increase with the volume fraction. At the highest volume fraction here investigated, Φ = 0.2, the velocity fluctuation intensities and the Reynolds shear stress are found to decrease. The analysis of the mean momentum balance shows that the particle-induced stresses govern the dynamics in the dense cases and are responsible of the the overall drag increase since the turbulent shear stress is reduced with respect the unladen case. 1 This work was supported by the European Research Council Grant No. ERC-2013-CoG-616186, TRITOS. 5:11PM E32.00003 Lattice Boltzmann simulation of particle inertial focusing in micro channels , YU CHEN, MORAN WANG, Department of Engineering Mechanics, School of Aerospace, Tsinghua University — We perform three dimensional lattice Boltzmann simulations to study particle inertial focusing in micro channels. Interpolation based curved boundary condition is employed to accurately treat the non-slip boundary condition of the particle surface. Force evaluation is via the corrected momentum exchange method recently proposed by our group, which ensures Galilean invariance and smooth force transition as the particle move across lattice nodes. Our results show good agreement with experiments, four equilibrium positions were found in square channel and two were found in rectangle channel. The two stage focusing is observed in our simulations which is also reported by others. For curving channels, additional force from dean flow further reduces equilibrium positions. A large portion of the curving channel needs to be simulated, as periodic boundary condition may not be valid here. By utilizing the parallel computing advantage of LBM, we perform large scale simulations of inertial focusing in curving channels. Detailed flow information and precisely monitored particle motion may provide valuable insight to understanding the mechanism of inertial focusing in micro channels and inspire developing of new designs. 5:24PM E32.00004 Fully resolved simulations of particle sedimentation1 , ADAM SIERAKOWSKI, YAYUN WANG, ANDREA PROSPERETTI, Johns Hopkins University — Progress in computational capabilities – and specifically in the realm of massively parallel architectures – render possible the simulation of fully resolved fluid-particle systems. This development will drastically improve physical understanding and modelling of these systems when the particle size is not negligible and their concentration appreciable. Using a newly developed GPU-centric implementation of the Physalis method for the solution of the incompressible Navier-Stokes equations in the presence of finite-sized spheres, we carry out fully resolved simulations of more than one thousand sedimenting spheres. We discuss the results of these simulations focusing on statistical aspects such as particle velocity fluctuations, particle pair distribution function, microstructure, and others. 1 Supported by NSF grant CBET 1335965 5:37PM E32.00005 Fully resolved simulation of the settling motion of a finite-sized spherical particle in a cellular flow field1 , JUNGWOO KIM, Seoul National University of Science and Technology — For particle-laden flows related to particle transport and dispersion, a knowledge of particle settling velocity is one of the important subjects. In that respect, for last several decades, many numerical studies with point particle approaches have been done. However, existing analytical expressions and empirical correlations used in point particle approaches are made based on many assumptions including the fact that the particle size is much smaller than the typical length scale of a given flow field. So, the settling velocity of a finite-sized particle in turbulent flows remains an unresolved issue. Therefore, we perform fully resolved simulations of the settling motion of a finite-sized spherical particle in a cellular flow field. The cellular flow field considered has been regarded as one of the good model problems for the study of the particle settling. One of the important parameters is the ratio of the particle diameter (d) and the cell size in the cellular flow (L). In this study, the change of the particle settling velocity is examined in the range of 0.01≤d/L≤0.1. In addition, the instantaneous drag and lift force components are compared with existing expressions for the corresponding force on the particle. Those results would show the validity and limitation of the present point particle approach in understanding the settling motion of a spherical particle in turbulent flows. 1 This work is supported by the NRF programs (2012M2A8A4055647) of the MSIP, Korea. 5:50PM E32.00006 Particle Resolved DNS of Turbulent Oscillatory Flow Over a Layer of Fixed Particles1 , CHAITANYA GHODKE, Oregon State Univ, JAVIER URZAY, Center for Turbulence Research, Stanford University, SOURABH APTE, Oregon State Univ — Particle resolved direct numerical simulations are performed using fictitious domain approach (Apte et al., JCP 2009) to investigate oscillatory turbulent flow over a layer of fixed particles representative of a sediment layer in coastal environments. Five particle Reynolds numbers in the range, ReD = 660 − 4240 are studied and results are compared against available experimental data (Keiller & Sleath, JFM 1976). Flow is characterized in terms of coherent vortex structures, Reynolds stress variation, turbulent cross-correlations and PDF distributions. The nature of the unsteady hydrodynamic forces on particles and their correlation to sweep and burst events is reported. The net lift coefficient remains positive over the cycle and is well correlated with phase averaged near-bed velocity. Maximum in the lift coefficient occurs when the strength of the horseshoe vortices is maximum. At this phase the lift fluctuations are correlated negatively with pressure and positively with velocity fluctuations in the region above the particle bed. Preliminary analysis shows non-Gaussian distribution for velocity fluctuation and follows 4th order Gram-Charlier. These detailed findings could eventually be useful in improving the existing criterion for predicting sediment incipient motion. 1 Supported by NSF project # 1133363 as well as Center for Turbulence Research Stanford University Summer Program 2014. Sunday, November 23, 2014 4:45PM - 6:03PM Session E33 Cavitation and Ventilation — 2022 - David Dowling, University of Michigan 4:45PM E33.00001 Wall-induced path variation of a large deformable rising bubble1 , HYUNGMIN PARK, HYEONJU JEONG, Seoul National University — In the present study, we experimentally investigate the wall-induced path variation of a large deformable bubble (Re ∼ O(103 )) rising near a vertical wall in quiescent water. To change the wall effect, we consider different wall materials (acrylic, PTFE and sponge) and vary the initial distance between the bubble and the wall. Depending on the conditions, various motions like a periodic bouncing, sliding, migrating away, and non-periodic oscillation without collisions are captured. Analysis on the energy balance shows that, contrary to a low-Re bubble, the surface deformation plays a great role in bubble’s rising behaviour. Especially, across the bubble-wall collision, the excessive surface energy compensates the deficit of kinetic energy, which enables the bubble to maintain a constant bouncing kinematics, despite the wall effect. The wall effect, appearing as a energy loss, decreases as the distance to the wall increases. Compared to the no-slip surface, the hydrophobic surface enhances or reduces the wall effect with the wall distance, whereas the porous surface reduces the energy loss due to the wall. The dependence of near-wall bubble motion on a wall configuration may give us an idea about how to predict or model the near-wall gas void-fraction. 1 Supported by the NRF programs (NRF-2012M2A8A4055647, NRF-2013R1A1A1008373) of Korean government. 4:58PM E33.00002 Slugs in a large diameter column with air and high viscosity silicone oil1 , ABBAS HASAN, BARRY AZZOPARDI, University of Nottingham — Very little information is known about the behaviour of high viscous liquids (> 100 Pa.s) in two phase slug flows. Experiments were carried out to study the behaviour of silicone oil (300 Pa.s) in gas bubble column using electrical capacitance tomography technique. The main aim of this paper is to study the characteristics and parameters of gas-liquid slug flows through large scale experiments with realistic liquids in a large diameter pipe (240 mm). These include; mean void fraction, Taylor bubble velocity, lengths of liquid slugs and Taylor bubbles, liquid film and fraction flowing down past the Taylor bubble. It was found that the gas mainly travels as large bubbles with ellipsoidal shape which occupy a significant portion of the pipe cross section with tiny bubbles in the liquid. In addition, the top surface of the gas-liquid column experiences a periodic oscillation (rising and falling) as the large bubbles rise to the top surface and burst. The results presented in this work have been compared with previous studies to show the effects of the viscosity and the pipe diameter on the behaviour of large bubbles in gas-liquid two phase slug flows. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 5:11PM E33.00003 Ventilation Inception and Washout, Scaling, and Effects on Hydrodynamic Performance of a Surface Piercing Strut1 , CASEY HARWOOD, YIN LU YOUNG, STEVEN CECCIO, University of Michigan — High-lift devices that operate at or near a fluid free surface (such as surface-piercing or shallowly-submerged propellers and hydrofoils) are prone to a multiphase flow phenomenon called ventilation, wherein non-condensable gas is entrained in the low-pressure flow, forming a cavity around the body and dramatically altering the global hydrodynamic forces. Experiments are being conducted at the University of Michigan’s towing tank using a canonical surface-piercing strut to investigate atmospheric ventilation. The goals of the work are (i) to gain an understanding of the dominant physics in fully wetted, partially ventilated, and fully ventilated flow regimes, (ii) to quantify the effects of governing dimensionless parameters on the transition between flow regimes, and (iii) to develop scaling relations for the transition between flow regimes. Using theoretical arguments and flow visualization techniques, new criteria are developed for classifying flow regimes and transition mechanisms. Unsteady transition mechanisms are described and mapped as functions of the governing non-dimensional parameters. A theoretical scaling relationship is developed for ventilation washout, which is shown to adequately capture the experimentally-observed washout boundary. 1 This material is based upon work supported by the National Science Foundation Graduate Student Research Fellowship under Grant No. DGE 1256260. Support also comes from the Naval Engineering Education Center (Award No. N65540-10-C-003). 5:24PM E33.00004 Shock wave induced shedding of cavitation clouds , HARISH GANESH, SIMO MAKIHARJU, STEVEN CECCIO, University of Michigan — Mechanisms responsible for periodic shedding of vapor clouds from partial cavities forming on a wedge are explored using time resolved X-ray densitometry. Time resolved 2-D void fraction flow field measurements of such partial cavities are obtained to identify the mechanisms of transition from closed partial cavities to open cavities exhibiting periodic shedding of vapor clouds or large gas pockets break off. From the void fraction field measurements, presence of an advancing bubbly shock front responsible for periodic shedding is identified as a primary cause of shedding. The void fraction measurements also reveal the presence of a reentrant flow at the cavity closure that produces intermittent shedding of smaller scale cavities at slightly higher cavitation numbers than periodically shedding cavities. A discussion on the observed occurrence and properties of the bubbly shock wave, and its role in causing periodic shedding is presented based on the one-dimensional model of shock propagation in bubbly mixtures. The observed cavity shape and its dependence with cavitation number is also compared with analytically predicted cavity shape using free streamline theory. 5:37PM E33.00005 Effect of Non-Condensable Gas on Cavity Dynamics and Sheet to Cloud Transition1 , SIMO MAKIHARJU, HARISH GANESH, STEVEN CECCIO, University of Michigan — Partial cavitation occurs in numerous industrial and naval applications. Cavities on lifting surfaces, in cryogenic rocket motors or in fuel injectors can damage equipment and in general be detrimental to the system performance, especially as partial cavities can undergo auto-oscillation causing large pressure pulsations, unsteady loading of machinery and generate significant noise. In the current experiments incipient, intermittent cloud shedding and fully shedding cavities forming in the separated flow region downstream of a wedge were investigated. The Reynolds number based on hydraulic diameter was of the order of one million. Gas was injected directly into the cavitation region downstream of the wedge’s apex or into the recirculating region such that with the same amount of injected gas less ended up in the shear layer. The cavity dynamics were studied with and without gas injection. The hypothesis to be tested were that i) relatively miniscule amounts of gas introduced into the shear layer at the cavity interface can reduce vapor production and ii) gas introduced into the separated region can dampen the auto oscillations. The authors also examined whether the presence of gas can switch the shedding mechanism from one dominated by condensation shock to one dominantly by re-entrant jet. 1 The work was supported by ONR grant number N00014-11-1-0449. 5:50PM E33.00006 Large eddy simulation of cavitating flows1 , ASWIN GNANASKANDAN, KRISHNAN MAHESH, University of Minnesota — Large eddy simulation on unstructured grids is used to study hydrodynamic cavitation. The multiphase medium is represented using a homogeneous equilibrium model that assumes thermal equilibrium between the liquid and the vapor phase. Surface tension effects are ignored and the governing equations are the compressible Navier Stokes equations for the liquid/vapor mixture along with a transport equation for the vapor mass fraction. A characteristic-based filtering scheme is developed to handle shocks and material discontinuities in non-ideal gases and mixtures. A TVD filter is applied as a corrector step in a predictor-corrector approach with the predictor scheme being non-dissipative and symmetric. The method is validated for canonical one dimensional flows and leading edge cavitation over a hydrofoil, and applied to study sheet to cloud cavitation over a wedge. 1 This work is supported by the Office of Naval Research Sunday, November 23, 2014 4:45PM - 6:03PM Session E34 CFD: Algorithms for Complex Flows — 2024 - David Trebotich, Lawrence Berkeley National Laboratory 4:45PM E34.00001 Large Eddy Simulation of a Fully Developed Turbulent Channel Flow by using Local Mesh Refinement , MEHTAP CEVHERI, THORSTEN STOESSER, Cardiff University — Application of mesh refinement to incompressible flows have been investigated over few decades. However, large eddy simulation for incompressible turbulent flows by using local mesh refinement (LMR) is a difficult task due to the lack of continuity at the coarse-fine interfaces. In order to solve this problem, ghost cell values for locally refined regions, which are located at the interface between coarse and fine grid domains, are calculated by using moving least square (MLS) algorithm and a numerical simulation of fully developed turbulent channel flow is performed to validate the accuracy of this implementation. The developed code has been coupled to the multigrid method to solve the unsteady, incompressible Navier-Stokes problem on a Cartesian grid with staggered variable arrangement. Wall-Adapting Local Eddy Viscosity (WALE) and one equation subgrid scale models have been used to simulate the unresolved turbulent fluctuations. The results of the developed code are compared with DNS data. 4:58PM E34.00002 Large-eddy simulation of wind turbine wake interactions on locally refined Cartesian grids1 , DIONYSIOS ANGELIDIS, FOTIS SOTIROPOULOS, St. Anthony Falls Laboratory, Department of Civil Engineering, 2 Third Avenue SE, Minneapolis, MN 55414, USA — Performing high-fidelity numerical simulations of turbulent flow in wind farms remains a challenging issue mainly because of the large computational resources required to accurately simulate the turbine wakes and turbine/turbine interactions. The discretization of the governing equations on structured grids for mesoscale calculations may not be the most efficient approach for resolving the large disparity of spatial scales. A 3D Cartesian grid refinement method enabling the efficient coupling of the Actuator Line Model (ALM) with locally refined unstructured Cartesian grids adapted to accurately resolve tip vortices and multi-turbine interactions, is presented. Second order schemes are employed for the discretization of the incompressible Navier-Stokes equations in a hybrid staggered/non-staggered formulation coupled with a fractional step method that ensures the satisfaction of local mass conservation to machine zero. The current approach enables multi-resolution LES of turbulent flow in multi-turbine wind farms. The numerical simulations are in good agreement with experimental measurements and are able to resolve the rich dynamics of turbine wakes on grids containing only a small fraction of the grid nodes that would be required in simulations without local mesh refinement. 1 This material is based upon work supported by the Department of Energy under Award Number DE-EE0005482 and the National Science Foundation under Award number NSF PFI:BIC 1318201. 5:11PM E34.00003 A low-cost RK time advancing strategy for energy-preserving turbulent simulations , FRANCESCO CAPUANO, GENNARO COPPOLA, LUIGI DE LUCA, University of Naples Federico II, GUILLAUME BALARAC, LEGI University of Grenoble — Energy-conserving numerical methods are widely employed in direct and large eddy simulation of turbulent flows. Semi-discrete conservation of energy is usually obtained by adopting the so-called skew-symmetric splitting of the non-linear term, defined as a suitable average of the divergence and advective forms. Although generally allowing global conservation of kinetic energy by convection, it has the drawback of being roughly twice as expensive as standard divergence or advective forms alone. A novel time-advancement strategy that retains the conservation properties of skew-symmetric-based schemes at a reduced computational cost has been developed in the framework of explicit Runge-Kutta schemes. It is found that optimal energy-conservation can be achieved by properly constructed Runge-Kutta methods in which only divergence and advective forms for the convective term are adopted. The new schemes can be considerably faster than skew-symmetric-based techniques. A general framework for the construction of optimized Runge-Kutta coefficients is developed, which has proven to be able to produce new methods with a specified order of accuracy on both solution and energy. The effectiveness of the method is demonstrated by numerical simulation of homogeneous isotropic turbulence. 5:24PM E34.00004 Multirate time-stepping least squares shadowing method for unsteady turbulent flow1 , HYUNJI JANE BAE, PARVIZ MOIN, Stanford — The recently developed least squares shadowing (LSS) method reformulates unsteady turbulent flow simulations to be well-conditioned time domain boundary value problems. The reformulation can enable scalable parallel-in-time simulation of turbulent flows (Wang et al. Phys. Fluid [2013]). A LSS method with multirate time-stepping was implemented to avoid the necessity of taking small global time-steps (restricted by the largest value of the Courant number on the grid) and therefore result in a more efficient algorithm. We will present the results of the multirate time-stepping LSS compared to a single rate time-stepping LSS and discuss the computational savings. 1 Hyunji Jane Bae acknowledges support from the Stanford Graduate Fellowship 5:37PM E34.00005 Adaptation of a Multi-Block Structured Solver for Effective Use in a Hybrid CPU/GPU Massively Parallel Environment , DAVID GUTZWILLER, Numeca USA, MATHIEU GONTIER1 , Numeca International, ALAIN DEMEULENAERE2 , Numeca USA — Multi-Block structured solvers hold many advantages over their unstructured counterparts, such as a smaller memory footprint and efficient serial performance. Historically, multi-block structured solvers have not been easily adapted for use in a High Performance Computing (HPC) environment, and the recent trend towards hybrid GPU/CPU architectures has further complicated the situation. This paper will elaborate on developments and innovations applied to the NUMECA FINE/Turbo solver that have allowed near-linear scalability with real-world problems on over 250 hybrid GPU/GPU cluster nodes. Discussion will focus on the implementation of virtual partitioning and load balancing algorithms using a novel meta-block concept. This implementation is transparent to the user, allowing all pre- and post-processing steps to be performed using a simple, unpartitioned grid topology. Additional discussion will elaborate on developments that have improved parallel performance, including fully parallel I/O with the ADIOS API and the GPU porting of the computationally heavy CPUBooster convergence acceleration module. 1 Head of HPC and Release Management, Numeca International Director, Numeca USA 2 Managing 5:50PM E34.00006 High Resolution DNS of Turbulent Flows using an Adaptive, Finite Volume Method1 , DAVID TREBOTICH, Lawrence Berkeley National Laboratory — We present a new computational capability for high resolution simulation of incompressible viscous flows. Our approach is based on cut cell methods where an irregular geometry such as a bluff body is intersected with a rectangular Cartesian grid resulting in cut cells near the boundary. In the cut cells we use a conservative discretization based on a discrete form of the divergence theorem to approximate fluxes for elliptic and hyperbolic terms in the Navier-Stokes equations. Away from the boundary the method reduces to a finite difference method. The algorithm is implemented in the Chombo software framework which supports adaptive mesh refinement and massively parallel computations. The code is scalable to 200,000+ processor cores on DOE supercomputers, resulting in DNS studies at unprecedented scale and resolution. For flow past a cylinder in transition (Re=300) we observe a number of secondary structures in the far wake in 2D where the wake is over 120 cylinder diameters in length. These are compared with the more regularized wake structures in 3D at the same scale. For flow past a sphere (Re=600) we resolve an arrowhead structure in the velocity in the near wake. The effectiveness of AMR is further highlighted in a simulation of turbulent flow (Re=6000) in the contraction of an oil well blowout preventer. 1 This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, Applied Mathematics program under contract number DE-AC02-05-CH11231 Sunday, November 23, 2014 4:45PM - 5:50PM Session E35 Compressible Reacting Flows and Rocket Combustion of Texas at Austin — 2001A - Venkat Raman, University 4:45PM E35.00001 Assessment of chemistry models for compressible reacting flows , SIMON LAPOINTE, GUILLAUME BLANQUART, California Institute of Technology — Recent technological advances in propulsion and power devices and renewed interest in the development of next generation supersonic and hypersonic vehicles have increased the need for detailed understanding of turbulence-combustion interactions in compressible reacting flows. In numerical simulations of such flows, accurate modeling of the fuel chemistry is a critical component of capturing the relevant physics. Various chemical models are currently being used in reacting flow simulations. However, the differences between these models and their impacts on the fluid dynamics in the context of compressible flows are not well understood. In the present work, a numerical code is developed to solve the fully coupled compressible conservation equations for reacting flows. The finite volume code is based on the theoretical and numerical framework developed by Oefelein (Prog. Aero. Sci. 42 (2006) 2-37) and employs an all-Mach-number formulation with dual time-stepping and preconditioning. The numerical approach is tested on turbulent premixed flames at high Karlovitz numbers. Different chemical models of varying complexity and computational cost are used and their effects are compared. 4:58PM E35.00002 Simulation of Supersonic Reactive Flow in Ramped Cavity Combustor with Fuel Injector , ZIA GHIASI, JONATHAN KOMPERDA, DONGRU LI, FARZAD MASHAYEK, University of Illinois at Chicago, COMPUTATIONAL MULTIPHASE TRANSPORT LABORATORY TEAM — Numerical simulation of supersonic reactive flows is emerging as an essential stage toward efficient design and development of scramjets. The flow inside the combustion chamber of scramjet is a prime example of multi-scale and multi-physics flow and is often accompanied by concurrent presence of shock waves and turbulence. Developing a robust numerical method for such simulations leads to numerous challenges due to the presence of complex geometries, shocks, turbulence and chemical reaction, which require massively parallel computation. In the present work we use the Discontinuous Spectral Element Method (DSEM) for high-fidelity simulation of reactive, supersonic and turbulent flows. The code features an entropy-based artificial viscosity method for capturing shock waves and uses implicit Large Eddy Simulation (LES) method for turbulence modeling. The turbulence-combustion interaction is captured using the Filtered Mass Density Function (FMDF) method. A supersonic reactive flow within a ramped cavity flame holder with round fuel injectors is simulated for hydrogen/air reaction, and the physics of the flow is studied. 5:11PM E35.00003 Numerical Simulations of a Reacting Sonic Jet in a Supersonic Cross-flow , NITESH ATTAL, PRAVEEN RAMAPRABHU, University of North Carolina at Charlotte — Interaction of a jet with a background cross-flow is a situation common to many engineering systems, including combustors in SCRAMJETS, gas turbines etc. Such an interaction enhances fuel-air mixing through the distortion of coherent structures into counter-rotating vortex pairs that are tilted, stretched and then sundered by the velocity gradient in the cross-flow, eventually leading to turbulent mixing. The ignition process and flame characteristics depend sensitively on the extent and efficiency of this turbulent mixing process. We describe results from detailed 3D numerical simulations of a sonic circular jet of diameter (D=0.5 cm) issuing a mixture of H2 (Fuel) diluted with 50% N2 at 300K into a turbulent, Mach 2 cross-flow of air at 1200K. The simulations were performed in a computational domain of 20x16x16 jet diameters using the compressible flow code FLASH [1], with modifications [2] to handle detailed (H2 -O2 ) chemistry and temperature-dependent material properties. We discuss the role of shock driven mixing, ignition and flame anchoring on the combustion efficiency of the system. [1] B. Fryxell et al., Astrophys. J., Suppl. Ser. 131,273(2000) [2] N. Attal et al., Comput. Fluids(under review) 5:24PM E35.00004 Simultaneous high-speed schlieren and OH chemiluminescence imaging in a hybrid rocket combustor at elevated pressures , VICTOR MILLER, ELIZABETH T. JENS, FLORA S. MECHENTEL, BRIAN J. CANTWELL, Stanford University, STANFORD PROPULSION AND SPACE EXPLORATION GROUP TEAM — In this work, we present observations of the overall features and dynamics of flow and combustion in a slab-type hybrid rocket combustor. Tests were conducted in the recently upgraded Stanford Combustion Visualization Facility, a hybrid rocket combustor test platform capable of generating constant mass-flux flows of oxygen. High-speed (3 kHz) schlieren and OH chemiluminescence imaging were used to visualize the flow. We present imaging results for the combustion of two different fuel grains, a classic, low regression rate polymethyl methacrylate (PMMA), and a high regression rate paraffin, and all tests were conducted in gaseous oxygen. Each fuel grain was tested at multiple free-stream pressures at constant oxidizer mass flux (40 kg/m2 s). The resulting image sequences suggest that aspects of the dynamics and scaling of the system depend strongly on both pressure and type of fuel. 5:37PM E35.00005 Experimental characterization of solid propellants combustion by digital holography , JUN CHEN, MICHAEL POWELL, JIAN GAO, IBRAHIM GUNDUZ, Purdue University, DANIEL GUILDENBECHER, Sandia National Laboratories, STEVE SON, Purdue University — Aluminum and other additions are widely used in solid propellants to improve performance. In this study, we apply digital holography as a three-dimensional diagnostic tool to characterize the burning of composite solid propellants with addition of different composite particles. Structures around the burning surfaces and reaction zones are identified, whereas the drop morphologies and their size/velocity distributions are quantified. The nano-second exposure of this imaging technique enables time-freezing measurements of the highly dynamic combustion process. The results are compared with discoveries from high-speed imaging. This technique is also applied to study the combustions of solid propellants under high-pressure environment. Sunday, November 23, 2014 4:45PM - 6:03PM Session E36 Instability: Viscoelastic Effects — Alcove A - Paul Durbin, Iowa State University 4:45PM E36.00001 Absolute Instability of a Variable Visccosity Jet , VINOD SRINIVASAN, Indian Inst of Science — The linear stability of an incompressible jet issuing into an ambient of higher viscosity is examined. Motivated by experimental results, only axisymmetric disturbances are considered. It is shown that for a constant-density jet with a prescribed velocity profile and Reynolds number, there exists a critical viscosity ratio between the jet centerline and far-field values, at which the jet transitions from convective to absolute instability. The axial disturbance corresponding to absolute instability is similar to the “column” mode found in the absolute instability of low-density jets. The boundary between absolute and convective instability is tracked as a function of viscosity ratio, Reynolds number, jet shear layer thickness, and density ratio. Shadowgraph and schlieren visualization is performed for a hot water jets issuing into a cold medium, over an experimental parameter range suggested by linear theory. A sudden increase in the jet spreading angle is interpreted as the onset of a global mode; hot film anemometry measurements corroborate the hypothesis. Global modes are observed for sufficiently large viscosity ratios. The onset and disappearance of global modes qualitatively matches the predictions of linear theory. 4:58PM E36.00002 Instability of Newtonian and Viscoelastic Submerged Jets , BAVAND KESHAVARZ, GARETH MCKINLEY, MIT, MIT, DEPT. OF MECH. ENG. TEAM — We study the behavior of high speed submerged liquid jets using a novel experimental setup equipped with strobe imaging flow visualization. Visualizations for Newtonian liquids at high Reynolds numbers (Re ∼ 150) show that for large wavenumbers a varicose mode dominate and the nonlinear growth of instability leads to the appearance of axisymmetric bags that roll up and encapsulate the central jet. At lower wave-numbers the varicose mode initially starts to grow close to the nozzle but is overwhelmed by the sinuous mode as the jet moves downstream. Due to the difference in the wave speeds for these two different modes of instability, the varicose waves slowly pile up into continuously growing sinuous waves leading to some unique concertina or chevron like morphologies. Tests with different viscoelastic model solutions show that by increasing the fluid elasticity the disturbance growth of both modes can be substantially inhibited due to streamline tension. These observations are the first experimental validation of theoretical predictions obtained from an elastic Rayleigh equation [1]. [1] J. M. Rallison and E. J. Hinch, “Instability of a high-speed submerged elastic jet,” J. Fluid Mech., vol. 288, pp. 311-324, 1995. 5:11PM E36.00003 Interfacial instability of wormlike micellar solutions sheared in a TaylorCouette cell , HADI MOHAMMADIGOUSHKI, SUSAN J. MULLER, Chemical Engineering, UC Berkeley — We report experiments on wormlike micellar solutions sheared in a custom-made Taylor-Couette (TC) cell. The computer controlled TC cell allows us to rotate both cylinders independently. Wormlike micellar solutions containing water, CTAB, and NaNo3 with different compositions are highly elastic and exhibit shear banding. We visualized the flow field in the θ-z as well as r-z planes, using multiple cameras. When subject to low shear rates, the flow is stable and azimuthal, but becomes unstable above a certain threshold shear rate. This shear rate coincides with the onset of shear banding. Visualizing the θ-z plane shows that this instability is characterized by stationary bands equally spaced in the z direction. Increasing the shear rate results to larger wave lengths. Above a critical shear rate, experiments reveal a chaotic behavior reminiscent of elastic turbulence. We also studied the effect of ramp speed on the onset of instability and report an acceleration below which the critical Weissenberg number for onset of instability is unaffected. Moreover, visualizations in the r-z direction reveals that the interface between the two bands undulates with shear bands evolving towards the outer cylinder regardless of which cylinder is rotating. 5:24PM E36.00004 Non-Modal Stability Analysis of High Strain-Rate Plastic Shear Flow , ALI NASSIRI, GREGORY CHINI, BRAD KINSEY, University of New Hampshire — High-speed oblique impact of two metal plates results in the development of an intense shear region at their interface, which leads to interfacial profile distortion and interatomic bonding. If the relative velocity is sufficient, a wavy pattern with a well-defined wavelength and amplitude is observed. The wavy structure has similarities to shear instabilities observed in fluid dynamics and predicted by hydrodynamic stability theories. However, since the impact is a short-time transient dynamical phenomenon, non-modal stability analysis presumably is more relevant than conventional eigenvalue analysis. Here, a non-modal shear flow stability analysis of a perfectly plastic material is performed to investigate the transient growth of disturbances and to assess if a connection exists with the corresponding predictions obtained from modal analysis. 5:37PM E36.00005 Instabilities around a rotating ellipsoid in a stratified fluid1 , BENJAMIN MIQUEL, PATRICE MEUNIER, STEPHANE LE DIZES, Aix Marseille Universite, CNRS, Centrale Marseille, IRPHE UMR 7342, 13384 Marseille, France — Geosismic observations have revealed the stacking of horizontal layers of water with different densities in the ocean, particularly above and beneath lens-shaped eddies. We present a simplified model together with an experimental setup to reproduce and identify the mechanism responsible for this layering phenomenon: we consider the stably stratified flow around a rotating, solid ellipsoid. Experimentally, a flat oblate rotating ellipsoid reproduces faithfully the boundary condition of an oceanic eddy, whereas the case of a rotating sphere provides an analytically tractable base flow, suitable for a numerical linear analysis. Two instabilities are witnessed experimentally and numerically. The first instability is the classical, inviscid, strato-inertial instability that tends to develop at the equator of the ellipsoid independently of the value of the Schmidt number. The second instability is localised in the vicinity of the poles and appears only if the Schmidt number differs from one. Hence, this instability is reminiscent of the double-diffusive McIntyre instability, a valuable candidate to explain layering in oceanic eddies. 1 Funded by ANR OLA 5:50PM E36.00006 Analysis of the onset of elastic instabilities in a homogenous stagnation point flow using dilute polymer solutions , FILIPE CRUZ, SIMON HAWARD, MANUEL ALVES, CEFT, Departamento de Engenharia Quı́mica, Faculdade de Engenharia da Universidade do Porto, Porto 4200-465, Portugal, GARETH MCKINLEY, Hatsopoulos Microfluidics Laboratory, MIT, Cambridge, MA 02139, United States of America — We compare numerical and experimental results for viscoelastic flows in the optimized cross-slot extensional rheometer - OSCER (Haward et al., Phys Rev Lett 109:128301, 2012) up to the onset of elastically-driven flow instabilities. Model polymer solutions with almost constant shear viscosity are used in the experiments, and the FENE-CR constitutive model is used in the 2D numerical simulations together with an in-house finite-volume viscoelastic flow solver. We match the model parameters to the rheology of the fluids used in the experiments, and the simulations are conducted for a wide range of flow rates, ranging from Newtonian-like flow at low Weissenberg numbers (Wi) up to the onset of time-dependent elastic instabilities at high Wi. We test the applicability of a dimensionless stability criterion (McKinley et al., J Non-Newt Fluid Mech 67:19-47, 1996) for predicting the onset of flow instability for both the experimental and computational data sets, using a spatially-resolved procedure to locally compute the stability criterion in the vicinity of the stagnation point. By evaluating this dimensionless criterion on a pointwise basis we are able to clearly distinguish the OSCER flow geometry from the archetypal cross-slot geometry. 6:15PM - 6:15PM — Session F1 Poster Session (6:15 - 7:00 PM) Level 2 Lobby - F1.00001 AERODYNAMICS AND ACOUSTICS — F1.00002 Vortex shedding by matched asymptotic vortex method , XINJUN GUO, SHREYAS MANDRE, Brown University — An extension of the Kutta condition, using matched asymptotic expansion applied to the Navier-Stokes equations, is presented for flow past a smooth body at high Reynolds number. The goal is to study the influence of unsteady fluid dynamical effects like leading edge vortex, unsteady boundary layer separation, etc. In order to capture accurately the location and strength of vortex shedding, the simplified Navier-Stokes equations in the form of boundary layer approximation are solved in the thin inner region close to the solid body. In the outer region far from the structure, the vortex methods are applied, which significantly reduces the computational cost compared to CFD in the whole domain. With this method, the flow past an airfoil with two degrees of freedom, pitching and heaving, is investigated. F1.00003 A reduced model for vortex shedding from a body using matched asymptotics1 , SHREYAS MANDRE, XINJUN GUO, PONNULAKSHMI V. K., Brown University — Flow around a solid body at high Reynolds number is often computed efficiently using inviscid vortex methods, if the distribution of vorticity shed from the surface of the body can be predicted accurately. The only method currently available for predicting the shed vorticity is by the application of the Kutta condition, which applies to slender wings at the leading and trailing edges. Therefore, benefit from the high Reynolds number approximation is limited to situations where the Kutta condition is applicable. We present a method based on matched asymptotic analysis to compute the strength and distribution of vorticity shed from rigid bodies of smooth but otherwise arbitrary shape executing arbitrary motion in a uniform far-field flow. The method decomposes he flow domain in an inviscid outer region and a thin viscous boundary layer near the solid body. The flow is approximated by inviscid vorticity dynamics in the outer region and Prandtl’s boundary layer theory in the boundary layer. The treatment of the boundary layer dynamics may be considered analogous to the Kutta condition, which yields an approximation to the shed vorticity. An approximately 100-fold increase in computational speed may be achieved using this method compared to direct numerical simulations. 1 AFOSR Award # FA9550-12-1-0099 F1.00004 Kinematics and Flow Evolution of a Flexible Wing in Stall Flutter1 , JOHN FARNSWORTH, University of Colorado - Boulder, JAMES AKKALA, JAMES BUCHHOLZ, University of Iowa, THOMAS MCLAUGHLIN, United States Air Force Academy — Large amplitude stall flutter limit cycle oscillations were observed on an aspect ratio six finite span NACA0018 flexible wing model at a free stream velocity of 23 m/s and an initial angle of attack of six degrees. The wing motion was characterized by periodic oscillations of predominately a torsional mode at a reduced frequency of k = 0.1. The kinematics were quantified via stereoscopic tracking of the wing surface with high speed camera imaging and direct linear transformation. Simultaneously acquired accelerometer measurements were used to track the wing motion and trigger the collection of two-dimensional particle image velocimetry field measurements to the phase angle of the periodic motion. Aerodynamically, the flutter motion is driven by the development and shedding of a dynamic stall vortex system, the evolution of which is characterized and discussed. 1 This work was supported by the AFOSR Flow Interactions and Control Portfolio monitored by Dr. Douglas Smith and the AFOSR/ASEE Summer Faculty Fellowship Program (JA and JB). F1.00005 Vorticity Transport on a Flexible Wing in Stall Flutter1 , JAMES AKKALA, JAMES BUCHHOLZ, University of Iowa, JOHN FARNSWORTH, University of Colorado - Boulder, THOMAS MCLAUGHLIN, United States Air Force Academy — The circulation budget within dynamic stall vortices was investigated on a flexible NACA 0018 wing model of aspect ratio 6 undergoing stall flutter. The wing had an initial angle of attack of 6 degrees, Reynolds number of 1.5 × 105 and large-amplitude, primarily torsional, limit cycle oscillations were observed at a reduced frequency of k = πf c/U = 0.1. Phase-locked stereo PIV measurements were obtained at multiple chordwise planes around the 62.5% and 75% spanwise locations to characterize the flow field within thin volumetric regions over the suction surface. Transient surface pressure measurements were used to estimate boundary vorticity flux. Recent analyses on plunging and rotating wings indicates that the magnitude of the pressure-gradient-driven boundary flux of secondary vorticity is a significant fraction of the magnitude of the convective flux from the separated leading-edge shear layer (Wojcik and Buchholz, J. Fluid Mech. 2014; Buchholz et al. AIAA Paper 2014-0071), suggesting that the secondary vorticity plays a significant role in regulating the strength of the primary vortex. This phenomenon is examined in the present case, and the physical mechanisms governing the growth and evolution of the dynamic stall vortices are explored. 1 This work was supported by the Air Force Office of Scientific Research through the Flow Interactions and Control Program monitored by Dr. Douglas Smith, and through the 2014 AFOSR/ASEE Summer Faculty Fellowship Program (JA and JB). F1.00006 ISGV Self-rectifying Turbine Design For Thermoacoustic Application , SHERVIN SAMMAK, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA, MARYAM ASGHARY, KAVEH GHORBANIAN, Department of Aerospace Engineering, Sharif University of Technology, Tehran 11115-8639, Iran — Thermoacoustic engines produce the acoustic power from wasted heat and then electricity can be generated from acoustic power. Utilizing self-rectifying turbine after a thermoacoustic engine allows for deploying standard generators with high enough rotational speed that remarkably reduce abrasion, size and cost and significantly increase efficiency and controllability in comparison with linear alternators. In this paper, by evaluating all different type of self-rectifying turbine, impulse turbine with self-piched controlled (ISGV) is chosen as the most appropriate type for this application. This kind of turbine is designed in detail for a popular engine, thermoacoustic stirling heat engine (TASHE). In order to validate the design, a full scale size of designed turbine is modeled in ANSYS CFX. As a result, optimum power and efficiency gained based on numerical data. F1.00007 BIOLOGICAL FLUID DYNAMICS — F1.00008 Modeling of Shear-Induced Red Blood Cell Migration for Guiding Microfluidic Device Design , EDEN DURANT, ADAM HIGGINS, KENDRA SHARP, Oregon State Univeristy — Through refinement and extension of a two-phase flow model previously reported for modeling blood in cylindrical flows (Gidaspow, 2009), we have developed a computational model for blood flow in complex microfluidic. Treating plasma as a Newtonian fluid and the Red Blood Cells (RBCs) as a granular phase, whose local concentrations are determined statistically, we have captured the migration of RBCs and concomitant formation of a cell free plasma layer at the channel walls. This model provides us with a threedimensional distribution of RBCs and the development of the stead-state flow profile, and enables us to study the influence of complex microfluidic geometries, including flow obstacles and varying channel dimensions, on the rate and extent of RBC margination. Simulations on 50 and 100 micron square channels match observed trends including decreasing RBC margination rate in larger channels, increasing RBC margination rate with higher hematocrit, and decreasing cell free layer width with increasing hematocrit. This predictive capability will allow microfluidic devices to be tailored and optimized for specific biomedical applications such as separation of blood constituents.. F1.00009 Hydrodynamics of Sessile Choanoflagellates1 , GREG BUSTAMANTE, HOA NGUYEN, Trinity University — Choanoflagellates are unicellular organisms whose intriguing morphology includes a set of collars/microvilli emanating from the cell body, surrounding the beating flagellum. Certain types of choanoflagellates are sessile, i.e., they can attach themselves to a substrate via a pedicel which extends from the cell body. We investigate the interactions of the flagellum - microvilli - pedicel system in the feeding behavior of sessile choanoflagellates using the method of images for regularized Stokeslets. The results of the fluid-particle motions and streamlines explain their effective capture of bacteria in the fluid. 1 Murchison Undergraduate Research Grant F1.00010 Determining Suction Feeding Efficiency in the Bowfin fish (Amia) using Particle Image Velocimery and Computaional Fluid Dynamics , YENNY RUA, KARIM KHARBOUCH, Fairfield Univ, CHRISTOPHER SANFORD, Hofstra University, SHANON RECKINGER, Fairfield Univ — Suction feeding is the most common form of prey capture in aquatic vertebrates. During the early evolution of fishes there was a major change in shape of the mouth, from a wedge shaped mouth opening in more primitive fishes to a more circular and planar mouth. This change in shape resulted from increased mobility of a key upper jaw bone, the maxilla. It has been suggested that this change in shape dramatically increased suction feeding efficiency. This study examines the hydrodynamic effects of these two mouth shapes in the same animal, the bowfin fish (Amia calva). 2D Particle Image Velocimetry (PIV) is used to analyze suction feeding events. Post-processing algorithms have been developed to determine the flow rate of water into the mouth of the fish; the area of fluid, the velocity of fluid and the volume of fluid affected by the fish; the velocity of the fluid at the mouth, as well as the velocity of the fluid as a function of the distance from the mouth, finally the force exerted on the fluid by the fish is also determined. Lastly, a numerical model has been developed for comparison using a non-uniform mesh, which adapts dynamically in space and time to the fish feeding event. The realistic geometry of the fish’s head is modeled in CAD. F1.00011 Bacterial accumulation mediated by flow compression-expansion , GASTÓN L. MIÑO, MIT, ERNESTO ALTSHULER, Physics Faculty-IMRE, University of Havana, Cuba, ANKE LINDNER, PMMH-ESPCI, France, ROMAN STOCKER, MIT, CARLOS A. CONDAT, ADOLFO J. BANCHIO, VERONICA I. MARCONI, IFEG-CONICET and FaMAF-Universidad Nacional de Cordoba, Argentina, ERIC CLÉMENT, PMMH-ESPCI, France — Swimming bacteria can be concentrated using a suitable microfluidic device. The combination of flow rate and surface shape can have significant impact on the microorganisms’ behavior. In those processes rheotaxis, accumulation caused by ratchets and even attachment to surfaces leading biofilm formation can be observed. Under these conditions, the transport of the active suspension is deeply modified, and differs significantly from passive suspensions. In this work, we present experimental evidence of Escherichia coli suspension flowing in a straight channel with a funnel-like constriction in the middle. This constriction is characterized by the aperture (wf ) and its angle (Θf ). We explore how the modification of wf and Θf affects the accumulation of bacteria in the channel. Concentrations of bacteria passing the constriction were observed for all the cases. However, the range of the flow rate that produced such accumulation varied with the geometry. In order to obtain a better understanding of this phenomenon, experiments are compared with a simple phenomenological model. F1.00012 On the Evolution of Pulsatile Flow Subject to a Transverse Impulse Body Force , GIUSEPPE DI LABBIO, Concordia University, ZAHRA KESHAVARZ-MOTAMED, MIT, LYES KADEM, Concordia University — In the event of an unexpected abrupt traffic stop or car accident, automotive passengers will experience an abrupt body deceleration. This may lead to tearing or dissection of the aortic wall known as Blunt Traumatic Aortic Rupture (BTAR). BTAR is the second leading cause of death in automotive accidents and, although quite frequent, the mechanisms leading to BTAR are still not clearly identified, particularly the contribution of the flow field. As such, this work is intended to provide a fundamental framework for the investigation of the flow contribution to BTAR. In this fundamental study, pulsatile flow in a three-dimensional, straight pipe of circular cross-section is subjected to a unidirectional, transverse, impulse body force applied on a strictly bounded volume of fluid. These models were simulated using the Computational Fluid Dynamics (CFD) software FLUENT. The evolution of fluid field characteristics was investigated during and after the application of the force. The application of the force significantly modified the flow field. The force induces a transverse pressure gradient causing the development of secondary flow structures that dissipate the energy added by the acceleration. Once the force ceases to act, these structures are carried downstream and gradually dissipate their excess energy. F1.00013 Simulation of flow in the microcirculation using a hybrid Lattice-Boltzman and Finite Element algorithm , ANDRES GONZALEZ-MANCERA, DIEGO GONZALEZ CARDENAS, Universidad de los Andes — Flow in the microcirculation is highly dependent on the mechanical properties of the cells suspended in the plasma. Red blood cells have to deform in order to pass through the smaller sections in the microcirculation. Certain deceases change the mechanical properties of red blood cells affecting its ability to deform and the rheological behaviour of blood. We developed a hybrid algorithm based on the Lattice-Boltzmann and Finite Element methods to simulate blood flow in small capillaries. Plasma was modeled as a Newtonian fluid and the red blood cells’ membrane as a hyperelastic solid. The fluid-structure interaction was handled using the immersed boundary method. We simulated the flow of plasma with suspended red blood cells through cylindrical capillaries and measured the pressure drop as a function of the membrane’s rigidity. We also simulated the flow through capillaries with a restriction and identify critical properties for which the suspended particles are unable to flow. The algorithm output was verified by reproducing certain common features of flow int he microcirculation such as the Fahraeus-Lindqvist effect. F1.00014 Predicting Weight Support Based on Wake Measurements of a Flying Bird in Still Air , ERIC GUTIERREZ, Aeronautics and Astronautics, Stanford University, DAVID LENTINK, Mechanical Engineering, Stanford University — The wake development of a freely flying Pacific Parrotlet (Forpus coelestis) was examined in still air. The bird was trained to fly from perch to perch through the laser sheet while wearing custom-made laser safety goggles. This enabled a detailed study of the evolution of the vortices shed in its wake using high-speed particle image velocimetry at 1000Hz in the plane transverse to the flight path. The measurement started when the bird was approximately 0.25 wingbeats in front of the laser sheet and stopped after it traveled 3.5 wingbeats beyond the laser sheet. The instantaneous lift force that supports body weight was calculated based on the velocity field, using both the Kuttta-Joukowski and the actuator disk quasi-steady model. During the first few flaps, both models predict an instantaneous lift that is reasonably close to the weight of the bird. Several flaps away from the laser sheet, however, the models predict that the lift steadily declines to about 50% of the weight of the bird. In contrast to earlier reports for bat wakes in wind tunnels, these findings for bird wakes in still air suggest that the predictive strength of quasi-steady force calculations depends on the distance between the animal and the laser sheet. F1.00015 In vivo measurement of aerodynamic weight support in freely flying birds , DAVID LENTINK1 , ANDREAS HASELSTEINER, RIVERS INGERSOLL, Department of Mechanical Engineering, Stanford University — Birds dynamically change the shape of their wing during the stroke to support their body weight aerodynamically. The wing is partially folded during the upstroke, which suggests that the upstroke of birds might not actively contribute to aerodynamic force production. This hypothesis is supported by the significant mass difference between the large pectoralis muscle that powers the down-stroke and the much smaller supracoracoideus that drives the upstroke. Previous works used indirect or incomplete techniques to measure the total force generated by bird wings ranging from muscle force, airflow, wing surface pressure, to detailed kinematics measurements coupled with bird mass-distribution models to derive net force through second derivatives. We have validated a new method that measures aerodynamic force in vivo time-resolved directly in freely flying birds which can resolve this question. The validation of the method, using independent force measurements on a quadcopter with pulsating thrust, show the aerodynamic force and impulse are measured within 2% accuracy and time-resolved. We demonstrate results for quad-copters and birds of similar weight and size. The method is scalable and can be applied to both engineered and natural flyers across taxa. 1 The first author invented the method, the second and third authors validated the method and present results for quadcopters and birds. F1.00016 Roll Dynamics in a Free Flying Dragonfly , JAMES MELFI, Cornell University, ANTHONY LEONARDO, Janelia Farms Research Campus, Z. JANE WANG, Cornell University — Dragonflies are capable of executing fast turning maneuvers. A typical free-flight maneuver includes rotations in all three degrees of freedom; yaw, pitch, and roll. This makes it difficult to identify the key changes to wing kinematics responsible for controlling each degree of freedom. Therefore we focus on a single motion; roll about the body longitudinal axis in a combined experimental and computational study. To induce rolling, a dragonfly is released from a magnetic tether while inverted. Both wing and body kinematics are recorded using multiple high speed cameras. The kinematics are replayed in a computer simulation of the flight, with forces and torques based on quasi-steady aerodynamics. By examining the effect of each kinematic change individually, we determine the key changes a dragonfly uses to both instigate, maintain, and end a rolling motion. F1.00017 BUBBLES, INTERFACES AND MULTIPHASE FLOWS — F1.00018 Analytical solution for linear long water waves, propagating in divergent channels , AGUSTÍN MORA, Univ Nacl Autonoma de Mexico, ERIC BAUTISTA, OSCAR BAUTISTA, JOSÉ HERNÁNDEZ, Instituto Politécnico Nacional — In this work, we obtain an analytical solution for the deformation of linear long water waves, propagating in a divergent channel of slowly varying crosssection, whose width of the channel obey a power-law distribution. By using an order of magnitude analysis and proposing characteristics lengths, the nonlinear governing equations of shallow water waves are simplified and written in dimensionless form. We derive a dimensionless wave equation that predicts the surface elevation, which is solved by using a novel technique which has the purpose on seeking the appropriate Bernoulli equation of the Boundary Value Problem studied. The analytical solution thus obtained is a function of two dimensionless parameters: a kinematical parameter, κ2 , which is the ratio of the wavelength to the channel length and one geometrical parameter Γ, which is the ratio of the width of regionR1 to the width of region R3 . The present analytical solution, covers a wide range of linear long water waves, spreading in long divergent channels with different geometrical transitions. For values of the parameter Γ > 1 the channels proposed in the present work, represent an efficient attenuator of linear long water waves. In addition, the application of the present mathematical model is not limited to thestudied cases in the present work, because the methodology used here can be extended to more complex channel transitions and also the formula presented in this work, can be easly implemented in order to validated numerical solution of long water waves. F1.00019 Large-eddy simulation of a solid-particles suspension in a turbulent boundary layer1 , MUSTAFA RAHMAN, RAVI SAMTANEY, KAUST — We decribe a framework for the large-eddy simulation of solid particles suspended and transported within an incompressible turbulent boundary layer. The underlying approach to simulate the solid-particle laden flow is Eulerian-Eulerian in which the particles are characterized by statistical descriptors. For the fluid phase, the large-eddy simulation (LES) of incompressible turbulent boundary layer employs stretched spiral vortex subgrid-scale model and a virtual wall model similar to the work of Inoue & Pullin (J. Fluid Mech. 2011). Furthermore, a recycling method to generate turbulent inflow is implemented. For the particle phase, the direct quadrature method of moments (DQMOM) is chosen in which the weights and abscissas of the quadrature approximation are tracked directly rather than the moments themselves. The numerical method in this framework is based on a fractional-step method with an energy-conservative fourth-order finite difference scheme on a staggered mesh. It is proposed to utilize this framework to examine transport of sand in desert sandstorms. 1 Supported by KAUST OCRF funded CRG project on simulation of sandstorms. F1.00020 Dynamics of elastic dumbbells sedimenting in a viscous fluid: oscillations and hydrodynamic repulsion , MAREK BUKOWICKI, MARTA GRUCA, MARIA EKIEL-JEŹEWSKA, Institute of Fundamental Technological Research, Polish Academy of Sciences — We consider a symmetric system of two elastic fibers, modeled as elastic dumbbells, sedimenting in a vertical plane in a viscous fluid at low Reynolds number regime. We focus on the problem how the elasticity (which breaks time-reversal symmetry of the motion) affects the system’s dynamics. The point-particle model is used. We observe oscillating, but non-periodic motion of the elastic particles. Independently of the value of the spring constant, the hydrodynamic repulsion appears between the dumbbells. The trajectory shift is slower when the spring constant k tends to 0 or to ∞. In these limiting cases we recover the periodic dynamics reported in the literature. For a given finite but non-zero spring constant we observe existence of a universal time-dependent trajectory to which the system converges. F1.00021 Contact line dynamics on chemically heterogeneous surfaces , MARTIN BRINKMANN, DANIEL HERDE, Max Planck Institute for Dynamics and Self-Organization, Göttingen, Germany, TAK S. CHAN, Saarland University, Saarbrücken, Germany, STEPHAN HERMINGHAUS, Max Planck Institute for Dynamics and Self-Organization, Göttingen — Modeling the dynamics of liquid interfaces in contact to heterogeneous solids involves a multitude of different length scales. A description of the unsteady motion of the three phase contact line on substrates with spatially varying wettability remains a challenging task, even for highly viscous liquids. To investigate this problem in the framework of continuum mechanics, we first consider the motion of an effectively two-dimensional viscous droplet on a plane substrate using a boundary element method to numerically solve the Steady Stokes equation. A dynamic bistability is observed on a smooth, sinusoidal wettability pattern if the magnitude of the slip length is comparable to the height of the droplet and, at the same time, the extension of the droplet is close to a multiple of the wavelength of the pattern. Employing a linear response formalism we study the complementary case of a liquid interface forced to move over the substrate at a fixed average velocity. The stick slip motion of the contact line amounts to an additional viscous dissipation close to the contact line and, hence, causes the macroscopic dynamic contact angle to increase further in the presence of spatially heterogeneous surface energies. F1.00022 Lateral migration of a spherical particle in square channel flows1 , MASAKO SUGIHARA-SEKI, NAOTO NAKAGAWA, ATSUSHI KASE, RYOKO OTOMO, Kansai University, MASATO MAKINO, Yamagata University, TOMOAKI ITANO, Kansai University — Particles suspended in the Poiseuille flow through circular cylindrical tubes have been known to migrate perpendicular to the flow direction due to the inertial effect and to be focused toward an equilibrium radial position. Recently, the distributions of neutrally buoyant spherical particles were measured in square channel flows and it was reported that there are eight equilibrium positions of the particles in the channel cross-section, located near the centers of the channel faces and near the channel corners (Miura et al. JFM 749, 320-330, 2014). The present study is aimed to simulate numerically the motion of a spherical particle suspended in square channel flows for the channel Reynolds number (Re) up to 1000. The computation of lateral forces exerted on the particle indicated the presence of equilibrium positions at the center of the channel faces and at the channel corners. The channel corner equilibrium positions were found to be unstable for low Re, whereas the channel face equilibrium positions are always stable. As Re increases, the channel corner equilibrium positions are shifted toward the channel corner, while the channel face equilibrium positions are shifted toward the channel center. These results account for the experimental measurements. 1 JSPS KAKENHI Grant Number 25630057 F1.00023 Solid intruders falling into foams , ANDRÉS PÉREZ, SALOMÓN ÁLVAREZ, FLORENCIO SÁNCHEZ, IGNACIO CARVAJAL, IPN Esime Zacatenco, ABRAHAM MEDINA, IPN Esime Azcapotzalco — We have made experiments where we follow the trajectory of solid spherical intruders falling, due to the gravity action, in vertical rectangular acrylic-box, brimful of foam. Through this method, we can measure the mean bubble size and the effective resistance of a given foam. We found that effective resistances are very different among the cases when the foams were made in boxes open to atmosphere or in closed boxes. F1.00024 Hydrodynamic interactions between many-particles falling under gravity in a viscous fluid: analysis of periodic and quasi-periodic motions , MARTA GRUCA, Institute of Fundamental Technological Research Polish Academy of Sciences, DIVISION OF COMPLEX FLUIDS TEAM — We investigate dynamics of many particles settling under gravity in a viscous fluid within the Stokes flow regime. We consider several families of regular initial configurations of a large number of point-particles which lead to periodic and quasi-periodic motions of the particles. We vary the relative distance between particles and observe how does it affect the dynamics. We observe the oscillations under some out-of-phase rearrangements of the particles. We also see a large influence of initial conditions on the system stability. By perturbating the regular configurations we obtain the dynamics corresponding to the dynamics of drop of suspension. We also explore the dynamics of such system in porous media where analogous quasi-periodic motions have been found. F1.00025 Diffuse-interface modeling of liquid-vapor coexistence in equilibrium drops using smoothed particle hydrodynamics1 , JAIME KLAPP, Cinvestav-Abacus, LEONARDO DI G SIGALOTTI, UAM-A, JORGE TROCONIS, ELOY SIRA, FRANKLIN PENA, IVIC-Venezuela, ININ-IVIC TEAM, CINVESTAV-UAM-A TEAM — We study numerically liquid-vapor phase separation in two-dimensional, nonisothermal, van der Waals (vdW) liquid drops using the method of Smoothed Particle Hydrodynamics (SPH). In contrast to previous SPH simulations of drop formation, our approach is fully adaptive and follows the diffuse interface model for a single-component fluid, where a reversible, capillary (Korteweg) force is added to the equations of motion to model the rapid but smooth transition of physical quantities through the interface separating the bulk phases. Surface tension arises naturally from the cohesive part of the vdW equation of state and the capillary forces. The drop models all start from a square-shaped liquid and spinodal decomposition is investigated for a range of initial densities and temperatures. The simulations predict the formation of stable, subcritical liquid drops with a vapor atmosphere, with the densities and temperatures of coexisting liquid and vapor in the vdW phase diagram closely matching the binodal curve. We find that the values of surface tension, as determined from the Young-Laplace equation, are in good agreement with the results of independent numerical simulations and experimental data. The models also predict the increase of the vapor pressure with temperature and the fitting to the numerical data reproduces very well the Clausius-Clapeyron relation, thus allowing for the calculation of the vaporization pressure for this vdW fluid. 1 Cinvestav-Abacus F1.00026 Compressible bubbly shock problem: Revisited , ASHA CHIGURUPATI, SANJIVA LELE, Stanford University — The problem of shock waves in bubbly flows has been studied in great depth and is a part of an extensive body of literature. Most of this literature assumes the liquid to be incompressible. In this study, we look at the problem of shock waves in bubbly flows where liquid is treated as compressible. A simple 1-D flow problem is considered to study the effect of liquid compressibility on shock speed. The results obtained show higher values of shock speeds for incompressible case when compared to compressible case. This difference is negligible for higher void fractions (of the order of 10ˆ-1) but grows immensely as you decrease the void fractions further by several orders of magnitude. Results pertaining to the structure and propagation of shock waves, particularly in this range of void fractions, will be presented. Future investigations will focus on studying the accompanying bubble-bubble interactions and looking at transient solutions of the problem. F1.00027 Lattice Boltzmann simulation of self-driven bubble transport in a micro-channel with a virtual check valve1 , ROU CHEN, Indiana University-Purdue University Indianapolis, Indianpolis, IN, USA, WEI DIAO, YONGGUANG CHENG, School of Resources and Environmental Science, Wuhan University, Hubei, China, LIKUN ZHU, HUIDAN (WHITNEY) YU, Indiana University-Purdue University Indianapolis, Indianpolis, IN, USA — An innovative self-circulation, self-regulation mechanism has recently been proposed to experimentally generate gaseous species from liquid reactants with little or zero parasitic power consumption. When a bubble grows at a location close to a virtual check valve, expansion of the left meniscus of the bubble is hindered due to its capability to provide a higher capillary pressure than the right meniscus does. We perform numerical simulation of bubble transport in a channel with a virtual check valve using lattice Boltzmann method to provide benchmarks for the experiments. A stable discretized lattice Boltzmann equation is employed to simulate incompressible bubble-liquid flows with density ratio above 1000. Polynomial wall free energy boundary condition is introduced and examined for static cases with a bubble sitting on solid surfaces for a triple contact among bubble, liquid, and solid surface. In this work, we focus on the effects of channel ratio between with and without check valve on the dynamics of bubble-driven liquid circulation. 1 This work is supported by NSF Collabrotive Research (1264739) F1.00028 Dynamics of skirting droplets , CALEB AKERS, JACOB HALE, DePauw University — It has been observed that non-coalescence between a droplet and pool of like fluid can be prolonged or inhibited by sustained relative motion between the two fluids. In this study, we quantitatively describe the motion of freely moving droplets that skirt across the surface of a still pool of like fluid. Droplets of different sizes and small Weber number were directed horizontally onto the pool surface. After stabilization of the droplet shape after impact, the droplets smoothly moved across the surface, slowing until coalescence. Using high-speed imaging, we recorded the droplet’s trajectory from a top-down view as well as side views both slightly above and below the fluid surface. The droplets’ speed is observed to decrease exponentially, with the smaller droplets slowing down at a greater rate. Droplets infused with neutral density micro beads showed that the droplet rolls along the surface of the pool. A qualitative model of this motion is presented. F1.00029 A boundary element method for particle and droplet electrohydrodynamics in the Quincke regime , DEBASISH DAS, DAVID SAINTILLAN, None — Quincke electrorotation is the spontaneous rotation of dielectric particles suspended in a dielectric liquid of higher conductivity when placed in a sufficiently strong electric field. This phenomenon of Quincke rotation has interesting implications for the rheology of these suspensions, whose effective viscosity can be controlled and reduced by application of an external field. While spherical harmonics can be used to solve the governing equations for a spherical particle, they cannot be used to study the dynamics of particles of more complex shapes or deformable particles or droplets. Here, we develop a novel boundary element formulation to model the dynamics of a dielectric particle under Quincke rotation based on the Taylor-Melcher leaky dielectric model, and compare the numerical results to theoretical predictions. We then employ this boundary element method to analyze the dynamics of a two-dimensional drop under Quincke rotation, where we allow the drop to deform under the electric field. Extensions to three-dimensions and to the electrohydrodynamic interactions of multiple droplets are also discussed. F1.00030 Pick up and remove particles by water droplet using dissipative particle dynamics , CHUANJIN LAN, SOUVIK PAL, University of California Merced, ZHEN LI, Brown University, YANBAO MA, University of California Merced — Particle removal is a crucial concern for many engineering processes, such as, glass cleaning and substrate cleaning, where the removal of nanoparticles is a great challenge. In order to clean the surface without causing any mechanical damage to it, we use water droplets to pick up and remove the nanoparticles. Dissipative particle dynamics simulation is used to model the interaction between the water droplet and nanoparticles, as well as the solid substrate surface. The hydrophilic nanoparticles are successfully cleaned up by water droplet, and the detailed motion of these particles together with droplet is also captured. The results show that the water droplet can be used as an efficient tool for removal of nanoparticles from a surface. F1.00031 A study of water droplet between an AFM tip and a substrate using dissipative particle dynamics , SOUVIK PAL, CHUANJIN LAN, University of California, Merced, ZHEN LI, Brown University, E. DANIEL HIRLEMAN, YANBAO MA, University of California, Merced — Formation of a water droplet between a sharp AFM tip and a substrate due to capillary condensation affects the tip-substrate interaction. As a consequence, AFM measurements lose precision and often produce incorrect sample topology. Understanding the physics of liquid bridges is also important in the field of Dip-pen nanolithography (DPN). Significant research is being carried out to understand the mechanics of the formation of the liquid bridge and its dependence of surface properties, ambient conditions etc. The in-between length scale, i.e., mesoscale (∼ 100 nm) associated with this phenomenon presents a steep challenge for experimental measurements. In addition, molecular dynamics (MD) can be computationally prohibitive to model the entire system, especially over microseconds to seconds. Theoretical analysis using Young Laplace equation has so far provided some qualitative insights only. We study this system using Dissipative Particle Dynamics (DPD) which is a simulation technique suitable for describing mesoscopic hydrodynamic behavior of fluids. In this work, we carry out simulations to improve understanding of the process of formation of the meniscus, the mechanics of manipulation and control of its shape, and better estimation of capillary forces. The knowledge gained through our study will help in correcting the AFM measurements affected by capillary condensation. Moreover, it will improve understanding of more accurate droplet manipulation in DPN. F1.00032 CONTROL AND STABILITY — F1.00033 Effect of initial amplitude on the interfacial and bulk dynamics in RichtmyerMeshkov instability under conditions of high energy density1 , ZACHARY DELL, Carnegie Mellon University, ROBERT STELLINGWERF, Stellingwerf Consulting, SNEZHANA ABARZHI, Carnegie Mellon University — We systematically study the effect of the initial amplitude on the interfacial and bulk dynamics of the Richtmyer-Meshkov instability (RM) induced by strong shocks. The shock propagates from the light to the heavy fluid. The fluid densities are contrast. The fluid interface is initially perturbed with a cosine wave perturbation. Its amplitude is varied from 0% to 100% of the initial perturbation wavelength. A broad range of shock strengths and density ratios is considered. Smoothed particle hydrodynamics code is employed to ensure shock capturing and interface tracking. Detailed diagnostics of the flow scalar and vector fields is performed. Whenever possible the simulation results are compared with existing theoretical analyses achieving good agreement. The focus question of our study is how the energy deposited by the shock is partitioned between the interfacial and volumetric components. We analyze the dependence of the initial growth-rate of RMI, the velocity away from the interface, and the transmitted shock velocity as functions of the initial amplitude. Particularly, we found that for a Mach number 5 and an Atwood number 0.8, the initial growth rate is highest and the interfacial energy is the largest when the initial amplitude is about a quarter of the wavelength. 1 The work is supported by the US National Science Foundation. F1.00034 Flow Instability Past A Rounded Cylinder1 , DOOHYUN PARK, KYUNG-SOO YANG, Inha University, Korea — Numerical simulation of flow past a rounded cylinder has been performed to study the effects of rounding corners of an angulated cylinder on the primary (2D) and the secondary (3D) instabilities associated with the corresponding flow configuration. We consider the rounded cylinders ranging from a square cylinder of height D to a circular cylinder of diameter D by rounding the four corners of a square cylinder with a quarter circle of fixed radius (r). An immersed boundary method was adopted for implementation of the cylinder cross-sections in a Cartesian grid system. The key parameters are Reynolds number (Re) and corner radius of curvature (r). Firstly, the characteristics of the primary instability such as critical Reynolds number (Rec ), force coefficients, and Strouhal number for vortex shedding are reported against r. It was found that Rec is maximum at r/D =0.25, meaning that this flow is more stable than the two extreme cases of the square and circular cylinders. Furthermore, there are the optimal values of r/D for force coefficients, which vary with Re. Secondly, we studied the onset of 3D instabilities by using Floquet stability analysis. It turned out that the criticalities of 3D instability modes are significantly affected by r. 1 This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2012R1A2A2A01013019). F1.00035 Resolvent mode identification in a turbulent boundary layer1 , KEVIN ROSENBERG, BEVERLEY MCKEON, California Institute of Technology — The resolvent analysis developed by McKeon and Sharma (J. Fluid Mechanics, 2010) has demonstrated a connection between the most amplified disturbances in wavenumber/frequency space and observed structures in wall turbulence. Three simultaneous hotwire measurements are made across a zero-pressure gradient turbulent boundary layer to identify the resolvent modes associated with these structures. A resolvent mode is designated by a streamwise wavenumber, a spanwise number, and a temporal frequency (k, n, ω respectively) and physically represents a travelling wave. The three wires are aligned in the wall normal direction and spaced in the streamwise and spanwise directions. The signals are filtered at the frequency corresponding to the resolvent mode of interest and ensemble averaged over a single period; the resulting phase differences between wires and their respective separation distances allows for the calculation of the spatial wavenumbers. The eventual goal is to sense these modes in real time as this will provide an important first step towards the development of closed-loop control schemes, specifically within the context of the resolvent framework. 1 The support of the Air Force Office of Scientific Research under grant # FA 9550-12-1-0469 (P.M. Doug Smith) is gratefully acknowledged. F1.00036 Feedback control for oscillation damping in cavity flow , MOHAMED-YAZID RIZI, SATIE, ENS Cachan, 61 Avenue du Président Wilson, F-94230 Cachan Cedex, France, LUC PASTUR, LIMSI-CNRS, BP 133, F-91403 Orsay Cedex, France, MOHAMED ABBASTURKI, HISHAM ABOU-KANDIL, SATIE, ENS Cachan, 61 Avenue du Président Wilson, F-94230 Cachan Cedex, France, YANN FRAINGNEAU, LIMSI-CNRS, BP 133, F-91403 Orsay Cedex, France, SATIE, ENS CACHAN TEAM, LIMSI-CNRS TEAM — It is well known that the cavity flow provides a benchmark configuration to understand the self-sustained oscillations of the impingement shear layer that constitutes in numerous application the first source of acoustic noise. This study is focused on the design of a closed-loop control strategy to suppress the cavity flow self-sustained oscillations. We show that a simple time-delayed feedback control law kills the limit cycle and stabilizes the steady base flow. This control law happens to be easy to implement experimentally and rather robust to changes in flow conditions. To establish a linear dynamical model representing the cavity flow, a closed-loop identification technique (Eigensystem Realization Algorithm - ERA) is used. As expected, results indicate that the oscillation frequencies of the cavity are mainly due to the unstable modes of the linear dynamics. An H2 optimal controller is designed by exploiting the identified linear dynamical model. Our H2 controller stabilizes the cavity oscillations and is robust to parameters variations. F1.00037 Propagations of fluctuations and flow separation on an unsteadily loaded airfoil , ANDREW TENNEY, JACQUES LEWALLE, Syracuse Univ — We analyze pressure data from 18 taps located along the surface of a DU-96-W180 airfoil in bothand steady flow conditions. The conditions were set to mimic the flow conditions experienced by a wind turbine blade under unsteady loading to test and to quantify the effects of several flow control schemes. Here we are interested in the propagation of fluctuations along the pressure and suction sides, particularly in relation to the fluctuating separation point. An unsteady phase of the incoming fluctuations is defined using Morlet wavelets, and phase-conditioned crosscorrelations are calculated. Using wavelet-based pattern recognition, individual events in the pressure data are identified with several different algorithms utilizing both the original time series pressure signals and their corresponding scalograms. The data analyzed in this study was collected by G. Wang in the Skytop anechoic chamber at Syracuse University in the spring of 2013; the work of Zhe Bai on this data is also acknowledged. F1.00038 On a possible mechanism for the generation of cyclonic vortices regime in a precessing cylindrical container , WALEED MOUHALI, ECE PAris, THIERRY LEHNER, Luth Observatoire de Meudon, ATER TEAM — We report experimental observations obtained by particle image velocimetry of the behavior of a flow driven by rotation and precession of a cylindrical container. Various flow regimes are identified according to the value of the control parameter ε (also called the Poincaré number) which is the ratio of the precession frequency ΩP to the rotation frequency ΩR . In particular, when ε is increased from small values, we have observed an induced differential rotation followed by the apparition of permanent cyclonic vortices. In particular, when ε is increased from small values, after a linear regime, we have observed a differential rotation followed by the apparition of four permanent cyclonic vortices as a consequence of instability (eruption of jets from the lateral edges of the cylinder). We propose a mechanism for the explanation of this generation. based on the differential rotation created by nonlinear mode coupling of two inertial waves of azimuthal wave numbers m = 0 and m = 1 (mode forced by the precession) in the inviscid case. The profile of the azimuthal mean velocity and the corresponding axial mean vorticity both show an inflexion point in their radial profile leading to a possible localized instability. F1.00039 BUOYANCY AND THERMAL CONVECTION — F1.00040 Computational investigation of negatively buoyant jets , LEANDRE BERARD, MEHDI RAESSI, Univ of Mass - Dartmouth — We present computational results on the stability of negatively buoyant jets at various Richardson numbers and injection angles. The results show a critical Richardson number that is the boundary between stable and unstable behavior. The critical Richardson number is seen to vary with injection angle. The computational results also reveal the mechanisms leading to instability and shedding of oil ring structures. The computational tool is a 3D GPU-accelerated MPI-parallel two-phase flow solver. The governing equations are solved using the two-step project method in the finite volume context. The fluid interfaces are tracked using the volume-of-fluid method. The pressure Poisson solver is accelerated using GPUs, which provides an average acceleration factor of 5 in 3D parallel simulations. F1.00041 On the spiral roll state in thermal convection in spherical shell , TOMOAKI ITANO, TAKAHIRO NINOMIYA, KEITO KONNO, MASAKO SUGIHARA-SEKI, Faculty of Eng. Sci., Kansai Univ. — It is found that the “giant” spiral roll state in thermal convection in non-rotating spherical shell with a finite Pr reported by Zhang and others (Phys.Rev.E, 2002) exists with a subtle modification under conservation of invariance of C2 symmetry even at a relatively thicker spherical shell. By means of a detail numerical bifurcation analysis with aid of direct numerical simulation for the time-development of the system, it is elucidated that this state orginates, in the parameter space, at a higher Rayleigh number but a lower azimuthal wavenumber than the set of parameters where the state previously found by Zhang exists. F1.00042 Thermal convection in a rotating horizontal annulus1 , ALEXEY VJATKIN, ALEVTINA IVANOVA, VICTOR KOZLOV, Laboratory of vibrational hydromechanics, PSHPU — Thermal convection of viscous fluid in a coaxial horizontal gap rotating around its own axis is investigated experimentally. The temperature of inner boundary is higher than that of the outer one. The threshold of mean convection excitation is studied. It is found that, despite the stabilizing effect of the centrifugal force of inertia, the convection in the layer occurs in a threshold way at lowering the rotation velocity and is excited by thermovibrational mechanism. In viscous liquids crisis of heat transfer is associated with the appearance of vortices extended along the azimuth (three-dimensional structures), and the longitudinal two-dimensional rolls appear on the background of them. In the experiments with low-viscosity fluids the opposite sequence of convective processes development is observed. With the advent of convective structures their slow azimuthal drift relative to the cavity is registered. It is shown that the drift is associated with the azimuthal steady motion of the fluid, which is generated in Stokes layers near the boundaries. The increase of viscosity results in growth of wavelength of the longitudinal rolls and significant reduction of the velocity the drift of vortex system. Experimental results agree with theoretical predictions. 1 The work was done in the frame of PSHPU Strategic Development Program (project 029-F) and supported by RFBR (Grant 13-01-00675a) F1.00043 Comparison of CFD simulation of night purge ventilation to full-scale building measurements , ASHA CHIGURUPATI, CATHERINE GORLE, GIANLUCA IACCARINO, Stanford University — Efforts to improve the understanding of air motion in and around buildings can lead to more efficient natural ventilation systems, thereby significantly reducing a building’s heating and cooling demands. CFD simulations enable solving the details of the flow and convective heat transfer in buildings and have the potential to predict the performance of natural ventilation with a high degree of accuracy. Understanding the actual predictive capability of CFD simulations is however complicated by the complexity of the geometry and physics involved, and the uncertainty and variability in the boundary conditions. In the present study we model the night flush process in the Y2E2 building on Stanford University’s campus and compare the results to measurements in the full-scale, operational building. We model half of the building, which consists of three floors with office spaces and two atriums. We solve the RANS equations using ANSYS/Fluent and k-e RNG theory turbulence closure model for the duration of one night flush and will present a comparison of the CFD results to measurements of the temperature on each floor in both atriums. Future investigations will focus on the potential of reducing the discrepancy between observed and predicted values by varying uncertain model parameters and boundary conditions. F1.00044 TURBULENCE — F1.00045 Identification of Coherent Structures in Premixed Reacting Flows , EILEEN HAFFNER, MELISSA GREEN, Syracuse University, ELAINE ORAN, University of Maryland, SYRACUSE UNIVERSITY TEAM, UNIVERSITY OF MARYLAND TEAM — Many studies have been conducted on the best ways to quantitatively characterize the turbulence-flame interaction in reacting flows. It has been observed that increased turbulence intensity both wrinkles and broadens the flame front throughout the preheat zone and reaction zone. A Lagrangian coherent structures analysis is used to identify the individual coherent turbulent structures as the maximizing ridges of the Finite-Time Lyapunov exponent scalar field (FTLE). This method provides different information than Eulerian criteria which have predominantly been used in previous reacting flow studies. Preliminary results show that LCS ridges exhibit a clear qualitative correlation to the contour of the fuel mass-fraction of the flame. A quantitative characterization of how the LCS results correlate to observed flame geometries will allow for a better understanding of how these structures affect the flame brush, and could lead to improved efficiency in particular engines. F1.00046 Large-eddy simulations of the restricted nonlinear system1 , JOEL BRETHEIM, DENNICE GAYME, CHARLES MENEVEAU, Johns Hopkins University — Wall-bounded shear flows often exhibit elongated flow structures with streamwise coherence (e.g. rolls/streaks), prompting the exploration of a streamwise-constant modeling framework to investigate wall-turbulence. Simulations of a streamwise-constant (2D/3C) model have been shown to produce the roll/streak structures and accurately reproduce the blunted turbulent mean velocity profile in plane Couette flow. The related restricted nonlinear (RNL) model captures these same features but also exhibits self-sustaining turbulent behavior. Direct numerical simulation (DNS) of the RNL system results in similar statistics for a number of flow quantities and a flow field that is consistent with DNS of the Navier-Stokes equations. Aiming to develop reduced-order models of wall-bounded turbulence at very high Reynolds numbers in which viscous near-wall dynamics cannot be resolved, this work presents the development of an RNL formulation of the filtered Navier-Stokes equations solved for in large-eddy simulations (LES). The proposed LES-RNL system is a computationally affordable reduced-order modeling tool that is of interest for studying the underlying dynamics of high-Reynolds wall-turbulence and for engineering applications where the flow field is dominated by streamwise-coherent motions. 1 This work is supported by NSF (IGERT, SEP-1230788 and IIA-1243482) F1.00047 Wall-Resolved Large-Eddy Simulation of Turbulent Flow Past a NACA0012 Airfoil1 , WEI GAO, WEI ZHANG, RAVI SAMTANEY, KAUST — Large-eddy simulation (LES) of turbulent flow past a NACA0012 airfoil is performed at angle of attack (AoA) 3o and Rec = 2.3 × 104 . The filtered incompressible Navier-Stokes equations are spatially discretized using an energy conservative fourth-order scheme developed by Morinishi et al. (J. of Comput. Phys., 1998), and the subgrid-scale (SGS) tensor is modeled by the stretched-vortex SGS model developed by Pullin and co-workers (Phys. of Fluids, 2000, J. of Fluid Mech., 2009). An extension of the original stretched-vortex SGS model is utilized to resolve the streak-like structures in the near-wall flow regions. The mean velocity and turbulence intensity profiles on airfoil surface and in wake are validated against experimental data reported in Dong-Ha Kim et al. (AIAA, 2009). To further verify our LES capacity, some high-order turbulence quantities are also compared with the DNS results produced by our in-house DNS code. The effect of grid-refinement on the wall-resolved LES approach is also discussed. 1 Supported by KAUST OCRF funded CRG project on simulation of turbulent flows over bluff bodies and airfoils. F1.00048 Spectral analysis of structure functions and their scaling exponents in forced isotropic turbulence1 , MORITZ LINKMANN, University of Edinburgh, W. DAVID MCCOMB, Retired, SAMUEL YOFFE, University of Strathclyde, ARJUN BERERA, University of Edinburgh — The pseudospectral method, in conjunction with a new technique for obtaining scaling exponents ζn from the structure functions Sn (r), is presented as an alternative to the extended self-similarity (ESS) method and the use of generalized structure functions. We propose plotting the ratio |Sn (r)/S3 (r)| against the separation r in accordance with a standard technique for analysing experimental data. This method differs from the ESS technique, which plots the generalized structure functions Gn (r) against G3 (r), where G3 (r) ∼ r. Using our method for the particular case of S2 (r) we obtain the new result that the exponent ζ2 decreases as the Taylor-Reynolds number increases, with ζ2 → 0.679 ± 0.013 as Rλ → ∞. This supports the idea of finite-viscosity corrections to the K41 prediction for S2 , and is the opposite of the result obtained by ESS. The pseudospectral method permits the forcing to be taken into account exactly through the calculation of the energy input in real space from the work spectrum of the stirring forces. The combination of the viscous and the forcing corrections as calculated by the pseudospectral method is shown to account for the deviation of S3 from Kolmogorov’s “four-fifths”-law at all scales. 1 This work has made use of the resources provided by the UK supercomputing service HECToR, made available through the Edinburgh Compute and Data Facility (ECDF). A. B. is supported by STFC, S. R. Y. and M. F. L. are funded by EPSRC. F1.00049 Investigation of electric charge on inertial particle dynamics in turbulence1 , JIANG LU, RAYMOND SHAW, Michigan Technological University — The behavior of electrically charged, inertial particles in homogeneous, isotropic turbulence is investigated. Both like-charged and oppositely-charged particle interactions are considered. Direct numerical simulations (DNS) of turbulence in a periodic box using the pseudospectral numerical method are performed, with Lagrangian tracking of the particles. We study effects of mutual electrostatic repulsion and attraction on the particle dynamics, as quantified by the radial distribution function (RDF) and the radial relative velocity. For the like-charged particle case, the Coulomb force leads to a short range repulsion behavior and an RDF reminiscent of that for a dilute gas. For the oppositely-charged particle case, the Coulomb force increases the RDF beyond that already occurring for neutral inertial particles. For both cases, the relative velocities are calculated as a function of particle separation distance and show distinct deviations from the expected scaling within the dissipation range. 1 This research was supported by NASA grant NNX113AF90G F1.00050 Statistical Modeling of Epistemic Uncertainty in RANS Turbulence Models , IMAN RAHBARI, VAHID ESFAHANIAN, University of Tehran — RANS turbulence models are widely used in industrial applications thanks to their low computational costs. However, they introduce model-form uncertainty originating from eddy-viscosity hypothesis, assumptions behind transport equations of turbulent properties, free parameters in the models, and wall functions. In contrast, DNS provides detailed and accurate results but in high computational costs making it unaffordable in industrial uses. Therefore, quantification of structural uncertainty in RANS models using DNS data could help engineers to make better decisions from the results of turbulence models. In this study, a new and efficient method for statistical modeling of uncertainties in RANS models is presented, in which deviation of predicted Reynolds stress tensor from results of DNS data is modeled through a Gaussian Random Field. A new covariance kernel is proposed based on eigendecomposition of a sample kernel, hyperparameters are found by minimization of negative log likelihood employing Particle Swarm Optimization algorithm. Thereafter, the random field is sampled using Karhunen-Loeve expansion followed by solving RANS equations to obtain the quantity of interest for each sample as uncertainty propagation. In the present study, fully developed channel flow as well as flow in a converging-diverging channel are considered as test cases. F1.00051 Turbulent Flow Over a Low-Camber Pitching Arc Wing , MAJID MOLKI, Southern Illinois University Edwardsville — Aerodynamics of pitching airfoils and wings are of great importance to the design of air vehicles. This investigation presents the effect of camber on flow field and force coefficient for a pitching circular-arc airfoil. The wing considered in this study is a cambered plate of zero thickness which executes a linear pitch ramp, hold and return of 45 ◦ amplitude. The momentum equation is solved on a mesh that is attached to the wing and executes a pitching motion with the wing about a pivot point located at 0.25-chord or 0.50-chord distance from the leading edge. Turbulence is modeled by the k − ω SST model. Using the open-source software OpenFOAM, the conservation equations are solved on a dynamic mesh and the flow is resolved all the way to the wall (y + ≈ 1). The computations are performed for Re = 40,000 with the reduced pitch rate equal to K = cθ̇/2U∞ = 0.2. The results are presented for three wings, namely, a flat plate (zero camber) and wings of 4% and 10% camber. It is found that the flow has complex features such as leading-edge vortex, near-wake vortex pairs, clockwise and counter-clockwise vortices, and trailing-edge vortex. While vortices are formed over the flat plate, they are formed both over and under the cambered wing. F1.00052 Ahmad Reza Estakhr Number, (Fluid Dynamics) , AHMAD REZA ESTAKHR, Physics Research Center — The Estakhr Number is a dimensionless number defined as: En = λ η where λ denotes mean free path and η denotes Kolmogorov length scale. The Mach p γπ where the M a denotes Mach number, γ denotes the ratio of specific heats and is dimension and Estakhr Numbers are therefore related by: En = M a 2 less. At high Reynolds number the Knudsen, Estakhr and Reynolds Numbers are therefore related by: En = Kn Re where the Kn denotes Knudsen number and Re denotes Reynolds number. F1.00053 NUMERICAL SIMULATIONS — F1.00054 Shock interaction behind a pair of cylindrical obstacles , HENG LIU, RAOUL MAZUMDAR, VERONICA ELIASSON, Univ of Southern California — The body of work focuses on two-dimensional numerical simulations of shock interaction with a pair of cylindrical obstacles, varying the obstacle separation and incident shock strength. With the shock waves propagating parallel to the center-line between the two cylindrical obstacles, the shock strengths simulated vary from a Mach of 1.4 to a Mach of 2.4, against a wide range of obstacle separation distance to their diameters. These cases are simulated via a software package called Overture, which is used to solve the inviscid Euler equations of gas dynamics on overlapping grids with adaptive mesh refinement. The goal of these cases is to find a so-called “safe” region for obstacle spacing and varying shock Mach numbers, such that the pressure in the “safe” region is reduced downstream of the obstacles. The benefits apply to both building and armor design for the purpose of shock wave mitigation to keep humans and equipment safe. The results obtained from the simulations confirm that the length of the “safe” region and the degree of shock wave attenuation depend on the ratio of obstacle separation distance to obstacle diameter. The influence of various Mach number is also discussed. F1.00055 High-Order Ghost-Fluid Method for Compressible Flow in Complex Geometry1 , MOHAMAD AL MAROUF, RAVI SAMTANEY, KAUST — We present a high-order embedded boundary method for numerical solutions of the Compressible Navier Stokes (CNS) equations in arbitrary domains. A high-order ghost fluid method based on the PDEs multidimensional extrapolation approach of Aslam (J. Comput. Phys. 2003) is utilized to extrapolate the solution across the fluid-solid interface to impose boundary conditions. A fourth order accurate numerical time integration for the CNS is achieved by fourth order Runge-Kutta scheme, and a fourth order conservative finite volume scheme by McCorquodale & Colella (Comm. in App. Math. & Comput. Sci. 2011) is used to evaluate the fluxes. Resolution at the embedded boundary and high gradient regions is accomplished by applying block-structured adaptive mesh refinement. A number of numerical examples with different Reynolds number for a low Mach number flow over an airfoil and circular cylinder will be presented. 1 Supported by OCRF-CRG grant at KAUST. F1.00056 Momentum equations of Newtonian fluids fully in the Eulerian perspective , QIFENG LV, SIJING WANG, MORAN WANG, Tsinghua University — The Navier-Stokes equations are used to describe the flow of Newtonian fluids in the Eulerian perspective. However, we find the right-hand sides of the Navier-Stokes equations were derived not from the Eulerian perspective but rather from the Lagrangian perspective, although this makes the Navier-Stokes equations simple and also valid in the laminar flow. In fact, the Lagrangian Cauchy strain rates were used in the derivation of the Navier-Stokes equations. Thus, here we derive the Cauchy strain rates from the Eulerian perspective. We then find the difference between the Eulerian and the Lagrangian Cauchy strain rates cannot be neglected when in turbulent flows or compressible fluid flows. Thereby, On the basis of the Eulerian Cauchy strain rates, we derive a set of momentum equations for the flow of Newtonian fluids fully in the Eulerian perspective. F1.00057 MICRO AND NANO FLUID DYNAMICS — F1.00058 Dimensionless numbers describing the onset of flow transitions in flow-focusing and T-junction microfluidic devices , SIVA VANAPALLI, JIHAY KIM, MEHDI NEKOUEI, Texas Tech University, Department of Chemical Engineering — T-junction and flow focusing geometries are the most commonly used drop generators in microfluidic devices. Several studies have documented the different behaviors of dispersed phase in these devices including break-up modes such as squeezing, dripping, and jetting and a non-break-up mode involving co-flowing laminar streams, called parallel stream. However, the control parameters that govern the transitions between these behaviors are not fully known. Using a combination of experiments and numerical simulations, we find that the onset of the dispersed phase transitions can be described by two dimensionless numbers – Weber number based on outer phase and Reynolds number based on the inertia of the inner phase and viscous stress of the outer phase. The flow transition from drop regime to jetting occurs at Weo ∼ O(1), and the flow transition from drop regime to parallel stream occurs at Re* ∼ 1. This scaling of flow transition was not affected by the change in the viscosity ratio, concentration of surfactant, the height of the channel, and the wettability of the device. Thus, our studies suggest that these two dimensionless numbers capture the onset of flow transitions in microfluidic drop generators. F1.00059 Study on three-dimensional printing using electrohydrodynamic inkjet by analysis of mass flow rate1 , HAN SEO KO, SOO-HONG LEE, PIL-HO LEE, SANG WON LEE, Sungkyunkwan University — An electrohydrodynamic (EHD) jet can produce much smaller droplets than nozzle sizes even for highly viscous liquid. Micro scale patterns are produced by a direct patterning of the EHD inkjet printing technique to obtain lamination layers. A cone-jet mode shows good performance for line and surface printings. A prediction method for a flow rate was proposed by performing experiments and deriving an equation. The calculation was carried out by dividing the electric field and the fluid regions. Dielectric liquids were used as the working fluid, whose flow rate was measured at the applied voltage of 1.5kV to 2.5kV. The measured flow rate was affected by viscosity, surface tension, and density as fluid properties, and dielectric constant and electric conductivity as properties of electric fields for the voltage. Then, parameters of the printing were investigated by printed line width and thickness at various conditions. As a result, the applied static pressure had more effect on the line printing although the line width was affected by the stage velocity. The significant role of the parameters was confirmed to produce scaffolds using the three-dimensional EHD printing. 1 This work supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (No. S-2011-0023457). F1.00060 Direct patterning using aerodynamically assisted electrohydrodynamic jet printing1 , SANGYEON HWANG, BAEKHOON SEONG, WONYOUNG LEE, DOYOUNG BYUN, Sungkyunkwan Univ — Electrical force and aerodynamic force are considered to be preferred sources for generating a liquid jet to emit the target fluid on a tiny scale. The former is known as an electrohydrodynamic (EHD) jet, while the latter is called flow focusing. Here, we report the effect of a combined energy source on the micro scale jet and patterns and investigate the scaling law of pattern width according to the ratio of two energy sources. In a conventional EHD jet, after a short length of straight section the charged viscous jet turns into complex shape which occurs difficulty in patterning fine lines. A coaxially driven gas stream smoothed the asymmetric jet lengthening the straight section of the jet. The jet could be issued constantly within the range that did not exceed the stable region in the parametric space. Under such stable conditions, the jet became narrow as compared to the one from the normal EHD jet. Hence, the patterns formed at a high gas pressure were noticeably smaller than the others, demonstrating the controllability of jet thickness. Various liquids had been used as the target fluids to investigate the effect of liquid properties. 1 This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) (Grand number : 2014-023284 F1.00061 Capillary penetration of a liquid between two tilted plates making a small angle , ABEL LOPEZ-VILLA, Esime Azcapotzalco IPN, FRANCISCO HIGUERA, Etsi Aeronauticos UPM, AYDET JARA, SERGIO DE SANTIAGO, ABRAHAM MEDINA, Esime Azcapotzalco IPN — The penetration of a wetting liquid in the narrow gap between two tilted plates making a small angle is analyzed in the framework of the lubrication approximation. At the beginning of the process, the liquid rises independently at different distances from the line of intersection of the plates except in a small region around this line where the effect of the gravity is negligible. The maximum height of the liquid initially increases as a function of time and the angle of inclination. At later times, the motion of the liquid is confined to a thin layer around the line of intersection whose height increases as a power of time. The evolution of the liquid surface is computed numerically and compared with the results of a simple experiment. F1.00062 Dynamics of hemiwicking , JUNGCHUL KIM, ILDOO KIM, HO-YOUNG KIM, Seoul Natl Univ — We study the wicking dynamics of a liquid on a surface decorated with micropillars, which has been long known to be governed by the Washburn-like dynamics. However, rough substrates cannot be described by a single geometric parameter like a tube with a constant radius. So far, different forms of scaling laws for liquid propagation distance were suggested by different researchers, but most of the laws are valid for the specific experimental conditions (e.g. pillar aspect ratio) employed in each work. Here we propose a novel scaling law for the wetting speed as a function of the width, gap, and height of the pillars, and the physicochemical liquid properties, which is valid for considerably wide parameter space. Also, we discuss the maximum pillar spacing up to which the current assumption of densely spaced pillars is valid. F1.00063 Numerical study of dynamic behavior of contact line approaching a micro-scale particle , YUSUKE MIYAZAKI, Dept. of Mechanical Engineering, Tokyo University of Science, TAKAHIRO TSUKAHARA, ICHIRO UENO, Research Institute for Science & Technology, Tokyo University of Science — The behavior of contact line (CL) the boundary line of solid-liquid-gas interface is one of the important topics regarding the dynamic wetting. Many experimental and theoretical approaches have been performed about static and axisymmetric systems: e.g., Ally et al. (Langmuir 2010 vol. 26, 11797) measured the capillary force on a micro-scale particle attached to a liquid surface and they compared with their physical model. However, there are few numerical simulations of the dynamic and asymmetric systems Focusing on the CL passing micro-scale solid particles, we simulated solid-liquid-gas flows. Gas-liquid interface is captured by a VOF method and the surface tension model is the CSF model. Solid-fluid interaction is treated by an immersed boundary method. We studied the broken-dam problem with a fixed sphere in either macro or micro scale. Our results of the macro scale agree reasonably with the experimental result. In the micro scale, where the domain is of 2.0 × 2.0 × 2.0 µm3 and the sphere diameter is 0.5 µm, we tested two types of sphere surface: hydrophobic and hydrophilic solids. We demonstrated that, as the liquid touches the hydrophilic sphere, the velocity of CL is higher than the hydrophobic case. F1.00064 A Two-Stage Microfluidic Device for the Isolation and Capture of Circulating Tumor Cells1 , ANDREW COOK, Fisk University, SAYALI BELSARE, Birla Institute of Technology and Science Pilani, TODD GIORGIO, Vanderbilt University, RICHARD MU, Fisk University — Analysis of circulating tumor cells (CTCs) can be critical for studying how tumors grow and metastasize, in addition to personalizing treatment for cancer patients. CTCs are rare events in blood, making it difficult to remove CTCs from the blood stream. Two microfluidic devices have been developed to separate CTCs from blood. The first is a double spiral device that focuses cells into streams, the positions of which are determined by cell diameter. The second device uses ligand-coated magnetic nanoparticles that selectively attach to CTCs. The nanoparticles then pull CTCs out of solution using a magnetic field. These two devices will be combined into a single 2-stage microfluidic device that will capture CTCs more efficiently than either device on its own. The first stage depletes the number of blood cells in the sample by size-based separation. The second stage will magnetically remove CTCs from solution for study and culturing. Thus far, size-based separation has been achieved. Research will also focus on understanding the equations that govern fluid dynamics and magnetic fields in order to determine how the manipulation of microfluidic parameters, such as dimensions and flow rate, will affect integration and optimization of the 2-stage device. 1 NSF-CREST: Center for Physics and Chemistry of Materials. HRD-0420516; Department of Defense, Peer Reviewed Medical Research Program Award W81XWH-13-1-0397 F1.00065 Non-linear dynamics of viscous bilayers subjected to an electric field: 3D phase field simulations1 , CHRISTOS DRITSELIS, GEORGE KARAPETSAS, VASILIS BONTOZOGLOU, University of Thessaly — The scope of this work is to investigate the non-linear dynamics of the electro-hydrodynamic instability of a bilayer of immiscible liquids. We consider the case of two viscous films which is separated from the top electrode by air. We assume that the liquids are perfect dielectrics and consider the case of both flat and patterned electrodes. We develop a computational model using the diffuse interface method and carry out 3D numerical simulations fully accounting for the flow and electric field in all phases. We perform a parametric study and investigate the influence of the electric properties of fluids, applied voltage and various geometrical characteristics of the mask. 1 The authors acknowledge the support by the General Secretariat of Research and Technology of Greece under the action “Supporting Postdoc- toral Researchers” (grant number PE8/906), co-funded by the European Social Fund and National Resources. F1.00066 Self-assembly of nanoparticles in evaporating particle-laden emulsion drops , MIN PACK, XIN YANG, YING SUN, Drexel University — In this study, we demonstrate the scalable fabrication of nanostructures (e.g., nanomesh and nanoring arrays) via inkjet printing of oil-in-water emulsion drops containing nanoparticles in water. Nanoscale oil drops dispersed in water are used here as templates for assembly of nanoparticles on a substrate. The effect of oil vapor pressure on particle deposition morphologies is studied by using a variety of oils. For oil drops with a lower vapor pressure, non-uniform evaporation rate along the air-water interface drives dispersed oil drops to move and accumulate near the air/water/substrate contact line. These oil drops remain on the substrate while water is evaporating enabling nanoparticles to self-assemble into nanomeshes. While keeping the same oil concentration, oil drops with a higher vapor pressure completely evaporates near the contact line before water dries out, leading to nanoparticle deposition of coffee-ring structures. If nanoparticles are confined inside the dispersed oil drops, nanoring arrays are formed as the emulsion evaporates. The characteristics of the nanomeshes and nanorings are controlled by tuning the size and concentration of oil drops and nanoparticles, substrate wettability, surfactant concentration, and vapor pressure of oil. F1.00067 FIB design for nano-fluidic applications1 , REMY FULCRAND, ALESSANDRO SIRIA, ANNE-LAURE BIANCE, LYDERIC BOCQUET, Intitut Lumière Matière, Université Lyon1 - CNRS, UMR5306, 69622 Villeurbanne cedex, France, LIQUIDES ET INTERFACES TEAM — In this paper we briefly review the techniques available to researchers in the nano-fluidic domain to fabricate nano-pores and nano-channels. In this context the focused ion beam (FIB) technique will be introduced as a useful and versatile tool for nano-fluidics. We illustrate it with two specific examples involving nano-pores as building blocks for nano-fluidics. Nano-pores, either biological, solid-state, or ultra thin pierced grapheme, are powerful tools which are central to many applications, from sensing of biological molecules to desalination and fabrication of ion selective membranes. 1 The authors acknowledge financial support from ERC-AG project MICROMEGAS F1.00068 Study of surface charge density on solid/liquid interfaces by modulating the electrical double layer1 , HYUK KYU PAK, JONG KYUN MOON, Ulsan National Institute of Science and Technology, Korea — A solid surface in contact with water or aqueous solution usually carries specific electric charges. These surface charges attract counter ions from the liquid side. Since the geometry of opposite charge distribution parallel to the solid/liquid interface is similar to that of a capacitor, it is called an electrical double layer capacitor (EDLC). Therefore, there is an electrical potential difference across an EDLC in equilibrium. When a liquid bridge is formed between two conducting plates, the system behaves as two serially connected EDLCs. In this work, we propose a new method for investigating the surface charge density on solid/liquid interfaces. By mechanically modulating the electrical double layers and simultaneously applying a DC bias voltage across the plates, an AC electric current can be generated. By measuring the voltage difference between the plates as a function of bias voltage, we can study the surface charge density on solid/liquid interfaces. Our experimental results agree very well with the simple equivalent circuit model proposed here. Furthermore, using this method, one can determine the polarity of the adsorbed state on the solid surface depending on the material used. 1 This work was supported by Center for Soft and Living Matter through IBS program in Korea. F1.00069 Investigation and visualization of flow through porous media at the pore scale , SOPHIE ROMAN1 , CYPRIEN SOULAINE, ANTHONY KOVSCEK, Stanford Univ, DEPARTMENT OF ENERGY RESOURCES ENGINEERING TEAM — In this work, a micro-Particle Image Velocimetry (micro-PIV) system is used to quantitatively investigate the dynamics of fluid displacement in simplified porous media. The porous media under study are 2D etched micromodels containing a flow pattern either composed of circular grains homogeneously distributed or made of a sandstone replica pattern. The fluid is seeded with microparticles which are used to estimate the velocity field with PIV algorithms. The exact pore-scale velocity profiles are obtained in the case of a fully saturated porous medium with a typical pore size of 5-40µm. The experimental velocity measurements are compared with 2D direct numerical simulations of the flow through the two different geometries under consideration. We have shown that the micro-PIV measurements have produced results in very good agreement with the numerical simulations for single-phase flows. Therefore, this experimental technique can be used with confidence to investigate flow properties in porous media. In particular this technique can be powerful for the study of immiscible two-phase flow in porous media in a wide range of parameters, for which numerical tools are still in development and need reliable data to be validated. 1 Department of Energy Resources Engineering, Stanford University, Stanford, CA 94305, USA F1.00070 COMPLEX FLOWS AND COMPLEX FLUIDS F1.00071 ABSTRACT WITHDRAWN — — F1.00072 Enhancement of Sublimation of Single Graphene Layer by Interacting with Gas Molecules in Rarefied Environment1 , RAMKI MURUGESAN, JAE HYUN PARK, Gyeongsang National University — Graphene has excellent mechanical properties. One of them is the resistance to high temperature environment. Since the sublimation temperature of graphene is over 4500 K, it has been used for diverse high temperature applications in order to protect the system. In this study, using extensive molecular dynamics simulations, we show that the sublimation of graphene could be enhanced (occurs at the lower temperature) by interacting with the gas molecules. With increase in temperature, the bonds in graphene becomes so sensitive to interact with the incoming gas molecules. When the temperature is low, the graphene is stable to the impingement of gas molecules: The light H2 gases are stick to the graphene surface and remains being attached while the heavy CO2 and H2 O are bounced back from the surface. However, at high temperature H2 gases are absorbed on the graphene and destroy the C−C bonds by forming C−H bonds. The local breakage of bond at the impingement spot spreads the entire graphene soon, causing a complete sublimation. Even though the heavy CO2 and H2 O molecules also break the C−C bonds at high temperature,but their impingement effect is localized and the breakage does not propagate over the entire surface. 1 This research was supported by Agency for Defence Development (ADD) F1.00073 Characteristics of contaminant deposition onto a cylindrical body surrounded by porous clothing1 , MINKI CHO, JINWON LEE, Pohang Univ of Sci & Tech, HYUNSUK JUNG, HAEWAN LEE, Agency for Defense Development, POHANG UNIV OF SCI & TECH TEAM, AGENCY FOR DEFENSE DEVELOPMENT TEAM — In order to characterize the deposition pattern of air-borne contaminants on a human body protected by a garment, the air flow through the clothing and in the air gap between the clothing and the skin was numerically solved, and the deposition of the suspended contaminants on the skin was obtained over a wide variety of conditions-wind speed, human motion and clothing conditions. The penetrating air flow was sensitive to the pressure inside the air gap, for which a simple model was successfully formulated. Also the profile of the non-uniform deposition velocity or the Sherwood number could be well modeled based on the developing concentration boundary layer inside the air gap. The boundary layer thickness grew vary rapidly, nearly proportional to the square of the distance from the front stagnation point, which is much different from any other boundary layer studied in many engineering fields before. A rather universal function for the distribution of deposition speed over a cylindrical body was obtained, which remained valid for a very wide range of conditions. The characteristics for non-uniform and/or periodic external wind due to human motion were also analyzed. 1 This study is supported by Agency for Defense Development. F1.00074 Stability of viscoelastic wakes , LUCA BIANCOFIORE, Imperial College London, LUCA BRANDT, KTH, TAMER ZAKI, Imperial College London — Theoretical and computational studies of synthetic wakes have explained the dynamics of several industrial and technological flows, for example mixing in fuel injection and papermaking, and the flow behind bluff bodies. Despite the industrial importance of complex non-Newtonian flow, previous work has focused on Newtonian fluids. Nonlinear simulations of viscoelastic, spatially-developing wakes are performed in order to analyze the influence of polymer additives on the behavior of the flow. Viscoelasticity is modeled using the FENE-P closure. A canonical wake profile (Monkewitz, Phys. Fluids, 88) is prescribed as an inflow condition, and the downstream evolution is computed using the full Navier-Stokes equations for a range of Reynolds and Weissenberg numbers. The simulations demonstrate that the influence of the polymer can be stabilizing or destabilizing, depending on the inlet velocity profile. Smooth profiles are stabilized by elasticity while sharp profiles are destabilized. The disturbance kinetic energy budget is examined in order to explain the difference in behavior and in particular the influence of the polymeric stresses on flow stability. F1.00075 Mass flow rate of granular material in silos with lateral exit holes , ABRAHAM MEDINA, ESIME Azcapotzalco IPN, ARMANDO SERRANO, FLORENCIO SANCHEZ, ESIME Zacatenco IPN — In this work we have analyzed experimentally the mass flow rate, m’, of the lateral outflow of cohesionless granular material through circular orifices of diameter D and rectangular and triangular slots of hydraulic diameter DH made in vertical walls of bins. Experiments were made in order to determine also the influence of the wall thickness of the bin, w. Geometrical and physical arguments, are given to get a general correlation for m’ embracing both quantities, D (DH ) and w. The angle of repose is also an important factor characterizing these flows. F1.00076 A continuum approach to study reversible shear thickening fluid behavior , HUA-YI HSU, Department of Mechanical Engineering, National Taipei University of Technology — In this work we explore if it is possible to reproduce reversible shear thickening behavior by using continuum equations. A critical shear stress indicates the transition of reversible shear thickening. To the end it is noted that reversible shear thickening fluid behavior is affected by (i) Hydrodynamic force (ii) Brownian motion, and (iii) Electrostatic force. To incorporate the features, we simulate shearing flow between two walls in the presence of external potential source term. The shear- stress-versus-viscosity trend is similar to the experimental results. At low shear stress range, the viscosity decreases as the stress increases. After reaching the critical shear stress, the viscosity increases as the stress increases. An understanding of the overall force balance and the critical shear stress emerges from the governing equations. F1.00077 Shock wave mitigation using Newtonian and non-Newtonian fluids , XINGTIAN TAO, BRENDAN COLVERT, VERONICA ELIASSON, Univ of Southern California — The effectiveness of a wall of liquid as a blast mitigation device is examined using a shock tube and a custom-designed and -built shock test chamber. High-speed schlieren photography and high-frequency pressure sensors allow measurement during the relevant shock interaction time periods of the liquid-gas interface. The characteristic quantities that reflect these effects include reflected-to-incident shock strength ratio, transmitted-to-incident shock strength ratio, transmitted and reflected impulse, and peak pressure reduction. In particular, the effects of viscous properties of the fluid are considered when using non-Newtonian dilatant and pseudoplastic fluids. Experiments have been performed with both Newtonian and non-Newtonian fluids. The impact of a shock waves on Non-newtonian fluids is compared to that of Newtonian fluids. Experiments show that non-Newtonian fluids have very strong reflection properties, acting like solid walls under the impact of a shock wave. Further work is to be performed to compare quantitatively the properties of Newtonian vs. non-Newtonian fluids. F1.00078 EXPERIMENTAL FLUID MECHANICS — F1.00079 Lift Force Acting on Bodies in Viscous Liquid Under Vibration1 , VITALIY SCHIPITSYN, ALEVTINA IVANOVA, OLGA VLASOVA, VICTOR KOZLOV, Laboratory of vibrational hydromechanics, PSHPU — The averaged lift force acting on a rigid body located near the wall of the cavity with a viscous liquid under high-frequency oscillations of various types is studied experimentally and theoretically. The experiments are conducted with cylindrical and rectangular solids. Amplitude and frequency of vibration, viscosity and density of fluid, specific solid size, its density and shape vary. Lift force was measured by the dynamic hanging of the body in the gravity when the body oscillates without touching the cavity walls. The vibrations generate a repulsive force, holding a heavy body above the bottom of the cavity, and the light one at some distance from the ceiling. Lift force changes qualitatively in case of combined translational and rotational oscillations of the cavity containing fluid and solid; it is much greater than at the translational vibrations and appears throughout the entire volume of the liquid. The work contains a theoretical description of the mechanism of lift force generation and the comparison of the experimental and theoretical results. The agreement of the results is found in the limit of high dimensionless frequencies. The considered effects could be interesting for vibrational control of solid inclusions in viscous liquids. 1 Work was done in the framework of the Program of strategic development of PSHPU (project 030-F) and supported by Ministry of Education of Perm Region (project C26/625) and grant 4022.2014.1 (Leading Scientific School) F1.00080 Experiments and models of particle slurries , KATHERINE VARELA, California State University, Long Beach, SARAH BURNETT, University of North Carolina, Chapel Hill, ANDREW LI, MATTHEW MOLINARE, University of California, Los Angeles, DIRK PESCHKA, Weierstrass Institute, Mohrenstr, Berlin, Germany, JEFFREY WONG, ANDREA BERTOZZI, University of California, Los Angeles — We present new experimental and theoretical results for the resuspension of bidisperse particle-laden flows on an inclined plane. In particular, we study the case of two negatively buoyant particle species of similar size and dissimilar densities in a viscous fluid of finite volume. Different regimes of particle separation are observed and studied by adjusting the angle of inclination, total particle concentration, and relative particle volume ratio. In addition to obtaining information about the height profile of shock formations, we measure the advancement and separation of particle and fluid front positions in mono- and bidisperse scenarios. These dynamics are the basis for a quantitative understanding of polydisperse cases, which can be readily applied to industry and catastrophe modeling. F1.00081 Experimental Study of Spanwise Wake Compression of a Trapezoidal Pitching Panel1 , JUSTIN KING, ZACHARY BERGER, MELISSA GREEN, Syracuse University — Stereoscopic particle image velocimetry was used to characterize the highly three-dimensional flow created by a rigid, trapezoidal pitching panel used to model an idealized fish caudal fin. Previous work has demonstrated that spanwise compression of the wake occurs until the wake ultimately breaks down as it convects in the streamwise direction. However, quantitative verification of the spanwise velocity relevant to the structure of this compression was not evaluated in the prior work. Experiments were conducted over a range of Strouhal numbers from 0.17 to 0.56 at three locations along the spanwise extent of the wake. Ongoing stereo PIV measurements confirm spanwise flow in the wake toward the midspan, which agrees with the previously-observed linear spanwise compression as the wake moved downstream. 1 This work was supported by the Office of Naval Research under ONR Award No. N00014-14-1-0418. F1.00082 Study of interfaces in an Axisymmetric Supersonic Jet using Background Oriented Schlieren (BOS)1 , CARLOS ECHEVERRÍA, DAVID PORTA, ALEJANDRO AGUAYO, HIROKI CARDOSO, CATALINA STERN, UNAM — We have used several techniques to study a small axisymmetric supersonic jet: Rayleigh scattering, Schlieren Toepler and PIV. Each technique gives different kind of information. In this paper, a BOS set-up is used to study the structure of the shock pattern. A shadowgraph of a dot matrix is obtained with and without a flow. The displacement field of the dots is related to changes in the index of refraction, which can be related, through the Gladstone-Dale equation, to changes in density. Previous results with this technique were not conclusive because of the relative size of the dots compared to the diameter of the nozzle. Measurements have been taken for three different exit speeds. 1 We acknowledge support from UNAM through DGAPA PAPIIT IN117712 and the Graduate Program in Mechanical Engineering F1.00083 Calibration of a Background Oriented Schlieren (BOS) Set-up1 , DAVID PORTA, CARLOS ECHEVERRı́A, HIROKI CARDOSO, ALEJANDRO AGUAYO, CATALINA STERN, UNAM — We use two materials with different known indexes of refraction to calibrate a Background Oriented Schlieren (BOS) experimental set-up, and to validate the Lorenz-Lorentz equation. BOS is used in our experiments to determine local changes of density in the shock pattern of an axisymmetric supersonic air jet. It is important to validate, in particular, the Gladstone Dale approximation (index of refraction close to one) in our experimental conditions and determine the uncertainty of our density measurements. In some cases, the index of refraction of the material is well known, but in others the density is measured and related to the displacement field. 1 We acknowledge support from UNAM through DGAPA PAPIIT IN117712 and the Graduate Program in Mechanical Engineering F1.00084 Rise of aqueous foam in vertical pipes , VALERIANO ÁLVAREZ, ANDRÉS PEREZ, IGNACIO CARVAJAL, FLORENCIO SÁNCHEZ, ESIME Zacatenco, ABRAHAM MEDINA, ESIME Azcapotzalco — In this work we made many experiments on aqueous foams formation, the generation is done through a constant flow of gas from the bottom of the tube, by means of a capillary, we observed two cases, the first case was with the top tube open to the atmosphere and the second case is the same tube now covered. We measure vertical profiles of the foam rise velocities. These velocities are affected by properties of the foam solution and the interaction with the material tube as the surfactant type, angle contact, the type of foam generator, and different diameter of the pipe. 6:15PM - 6:15PM — Session F2 Student Poster Session (Student Poster Competition) - 6:15 - 7:00 PM - Level 3 Lobby F2.00001 GENERAL FLUID DYNAMICS — F2.00002 Reynolds Number Effects on Mixing Due to Topological Chaos , SANGEETA WARRIER1 , SPENCER SMITH2 , Mount Holyoke College — Topological chaos has emerged as a powerful modeling tool to investigate fluid mixing. While this theory can guarantee a lower bound on the stretching rate of certain material lines, it does not indicate what fraction of the fluid actually participates in this minimally mandated mixing. Indeed, the area in which effective mixing takes place depends on physical parameters such as the Reynolds number. To help clarify this dependency, we numerically simulate the effects of a batch stirring device on a 2D incompressible Newtonian fluid in the laminar regime. In particular, we calculate the finite time Lyapunov exponent (FTLE) field for two different stirring protocols, one topologically complex (pseudo Anosov) and one simple (finite order), over a range of viscosities. After extracting appropriate measures indicative of mixing from the FTLE field, we see a clearly defined range of Reynolds numbers for which the relative efficacy of the pseudo Anosov protocol over the finite order protocol justifies the application of topological chaos. The Reynolds number dependance of these mixing measures also reveals several other intriguing phenomena. 1 Undergraduate 2 Supervising Student Professor F2.00003 Integral Method and Eigenspace Decomposition for RANS Turbulent Mixing Flows , WADE SPURLOCK, Stanford University, ERIC PARISH, University of Michigan, DANIEL ISRAEL, Los Alamos National Laboratory, LANL COMPUTATIONAL PHYSICS SUMMER WORKSHOP COLLABORATION — Integral methods can be used to obtain low-order dynamical systems which approximate the solution of RANS models. Such techniques are beneficial for calibrating coefficients and verifying a model for flows away from self-similarity. We apply an integral method approach to turbulent mixing flows and compare analytic results to full-field RANS simulations in xRage, a code developed at Los Alamos National Laboratory. Flows that exhibit late time self-similarity are considered. Eigenspace decomposition is then used to identify dominant solution characteristics and visualize higher dimensional closure models. Results are shown for the temporal shear and Rayleigh-Taylor layers. F2.00004 On the stability of homogeneous three-dimensional turbulent flows , ANANDA MISHRA, SHARATH GIRIMAJI, Texas A&M University — Flows experiencing spatially uniform deformation rate, appellated as homogeneous flows, are the most elementary cases to exhibit hydrodynamic instabilities. While the stability characteristics of homogeneous flows subject to planar strain and rotation are well-established, those of three-dimensional flows are not. We address the stability characteristics of general incompressible flows undergoing three-dimensional streamline convergence, divergence and swirl. Two of the salient findings are:(i) flow stability is completely contingent upon the third invariant of the background velocity gradient tensor - flows with a positive third invariant are stable while those with a negative value are unstable; and, (ii) with the sole exception of two-dimensional elliptic flows, inertial effects are destabilizing and pressure effects are stabilizing. F2.00005 2D CFD Analysis of an Airfoil with Active Continuous Trailing Edge Flap1 , DYLAN JAKSICH, JINWEI SHEN, University of Alabama — Efficient and quieter helicopter rotors can be achieved through on-blade control devices, such as active Continuous Trailing-Edge Flaps driven by embedded piezoelectric material. This project aims to develop a CFD simulation tool to predict the aerodynamic characteristics of an airfoil with CTEF using open source code: OpenFOAM. Airfoil meshes used by OpenFOAM are obtained with MATLAB scripts. Once created it is possible to rotate the airfoil to various angles of attack. When the airfoil is properly set up various OpenFOAM properties, such as kinematic viscosity and flow velocity, are altered to achieve the desired testing conditions. Upon completion of a simulation, the program gives the lift, drag, and moment coefficients as well as the pressure and velocity around the airfoil. The simulation is then repeated across multiple angles of attack to give full lift and drag curves. The results are then compared to previous test data and other CFD predictions. This research will lead to further work involving quasi-steady 2D simulations incorporating NASTRAN to model aeroelastic deformation and eventually to 3D aeroelastic simulations. 1 NSF ECE Grant #1358991 supported the first author as an REU student. F2.00006 Experimental Investigation of Passive Shock Wave Mitigation using Obstacle Arrangements , MONICA NGUYEN, QIAN WAN, VERONICA ELIASSON, University of Southern California AME Department — With its vast range in applications, especially in the defense industry, shock wave mitigation is an ongoing research area of interest to the shock dynamics community. Passive shock wave mitigation methods range from forcing the shock wave to abruptly change its direction to introducing barriers or obstacles of various shapes and materials in the path of the shock wave. Obstacles provide attenuation through complicated shock wave interactions and reflections. In this work, we have performed shock tube experiments to investigate shock wave mitigation due to solid obstacles placed along the curve of a logarithmic spiral. Different shapes (cylindrical and square) of obstacles with different materials (solid and foam) have been used. High-speed schlieren optics and background-oriented schlieren techniques have been used together with pressure measurements to quantify the effects of mitigation. Results have also been compared to numerical simulations and show good agreement. F2.00007 A Shock-Driven Mechanism for Standing-Wave Patterns in Vertically Oscillated Grains1 , ALEX GILMAN, JON BOUGIE, Loyola University Chicago — We develop a simplified model for standing-wave pattern formation in vertically oscillated granular layers based on an instability in shocks found in these layers. When layers of particles are oscillated at accelerational amplitudes greater than that of gravity, the layers leave the plate, and shocks are created upon re-established contact with the plate. Additionally, standing-wave patterns form when the accelerational amplitude exceeds a critical value. For a given layer depth and accelerational amplitude, varying driving frequency alters the shock strength as well as pattern wavelength; increasing layer depth produces stronger shocks and longer pattern wavelengths for a given frequency. We demonstrate relationships between properties associated with shocks and properties associated with standing wave patterns, and present a simple mechanism by which a non-uniform shock front drives standing-wave configurations. We justify this mechanism using mathematical relationships derived from a continuum granular model. We then compare these mathematical relationships to full numerical solutions of continuum equations to Navier-Stokes order for uniform, frictionless, inelastic spheres. 1 This research is supported by the Loyola Undergraduate Research Opportunities Program. F2.00008 Computational Analysis of Wake Field Flow between Multiple Identical Spheres , WESLEY BRAND, MORTON GREENSLIT, ZACH KLASSEN, JAY HASTINGS, WILLIAM MATSON1 , University of Minnesota, Morris — It is well understood both that objects moving through a fluid perturb the motion of nearby objects in the same fluid and that some configurations of objects moving through a fluid have little inter-object perturbation, such as a flock of birds flying in a V-formation. However, there is presently no known method for predicting what configurations of objects will be stable while moving through a fluid. Previous work has failed to find such stable configurations because of the computational complexity of finding individual solutions. In this research, the motions of two spheres in water were simulated and combinations of those simulations were used to extrapolate the motions of multiple spheres and to find configurations where the lateral forces on each sphere were negligible and the vertical forces on each sphere were equivalent. Two and three sphere arrangements were simulated in COMSOL Multiphysics and Mathematica was used both to demonstrate that combinations of two sphere cases are identical to three sphere cases and to identify stable configurations of three or more spheres. This new approach is expected to simplify optimization of aerodynamic configurations and applications such as naval and aerospace architecture and racecar driving. 1 Advisor F2.00009 Vortex breakdown of a co-flowing swirling jet with density difference under normal and inverted gravity fields , SHUNSUKE TSUTSUMI, Ritsumeikan University, ADZLAN AHMAD, University of Malaysia Sarawak, HIROSHI GOTODA, Ritsumeikan University — We study vortex breakdown (VB) of a co-flowing swirling jet with a density difference under normal and inverted gravity fields. The density difference is created by issuing CO2 from an inner tube into ambient air. The formation region of unstable VB for a CO2 jet is larger under inverted gravity than under normal gravity. The trends of the changes in the stagnation point height are given particular attention while investigating the stable breakdown region. A physical model derived by considering the momentum balance in the flow is adopted to reasonably interpret the decrease in the stagnation point height of stable VB under inverted gravity with increasing inner swirl number of the inner jet, or the increase in the stagnation point height with increasing bulk flow velocity of the outer jet, for a swirling jet with a density difference. F2.00010 FILM AND INTERFACES — F2.00011 Surfactant Spreading on Thin Viscous Fluid Films1 , CAITLYN BONILLA, NATHANIEL LESLIE, JEANETTE LIU, DINA SINCLAIR, RACHEL LEVY, Harvey Mudd College — We examine the spreading of insoluble lipids on a viscous Newtonian thin fluid film. This spreading can be modeled as two coupled nonlinear fourth-order partial differential equations, though inconsistencies between the timescale of experiments and simulations have been reported in recent research. In simulations, we replace traditional models for the equation of state relating surfactant concentration to surface tension with an empirical equation of state. Isotherms collected via a Langmuir-Pockels scale provide data for the equation of state. We compare the timescale of simulation results to measurements of the fluorescently tagged lipid (NBD-PC) spreading as well as the height profile, captured with laser profilometry. 1 Research supported by NSF-DMS-FRG 9068154, RCSA-CCS-19788, HHMI F2.00012 Using falling soap film to visualize flow in a wavy channel , VONTRAVIS MONTS, OSAZUWA EDOKPOLO, ZACHARY MILLS, ALEXANDER ALEXEEV, Georgia Institute of Technology — The disturbances created by the walls of a sinusoidal shaped channel lead to the development of unsteady, time periodic flow. This periodic flow is the result of vortex shedding occurring along the crests of the channel walls. We used a falling soap film to investigate the influence of the channel geometry on the flow. In falling soap films, variations in the thickness of the film correspond to streamlines in the flow. These thickness variations are made visible by reflecting monochromatic light off the film. This allows for soap films to be an accurate, but inexpensive method of visualizing two dimensional flows. In our experiments we used a gravity driven soap film flowing through a wavy channel of several periods and used a high speed camera to record the resulting flow. The collected footage was then analyzed to collect data on the flow. From this data we were able to characterize the dependence of the size of the vortices and their shedding frequency on the amplitude and period of the sinusoidal channel walls as well as the Reynolds number of the flow. F2.00013 Wakes and flow-induced oscillations of tandem cylinders in a flowing soap film1 , WENCHAO YANG, JOSAM WATERMAN, MARK STREMLER, Virginia Tech — We investigate the wake dynamics and flow-induced oscillations of a tandem two-cylinder system aligned vertically in a flowing soap film. The cylinders interact with the soap film as circular disks. The upstream cylinder is fixed in place, while the downstream cylinder is free to oscillate as a pendulum that is driven by interactions with the wake of the upstream cylinder. The soap film is a convenient system for investigating quasi-2D dynamics and considering how they compare with the typical 3D system. Wake structures are visualized by the film’s interference fringes; both these and the cylinder locations are recorded with a high-speed camera system. The force response of the downstream cylinder is measured with a microcantilever laser-mirror sensor system. Varying the distance between the cylinders reveals multiple modes of behavior, including variations in the force response and the regularity of the oscillations. 1 Work made possible by funding from the Virginia Commonwealth Research Commercialization Fund F2.00014 Electrohydrodynamics of a surfactant-covered drop1 , ANDREW OBERLANDER, Brown University, MALIKA OURIEMI, IFPEN, France, PETIA VLAHOVSKA, Brown University — We present an experimental study of the behavior of a drop covered with insoluble surfactant in a uniform DC electric field. Steady drop shapes, drop evolution upon application of the field, and drop relaxation after the field is turned off are observed for a polybutadiene (PB) drop suspended in silicon oil (PDMS). The surfactant is generated at the drop interface by reaction between end-functionalized PB and PDMS. The experimental data is compared with the theory of Nganguia et al (2013) for the steady shapes, and a new model developed by us which accounts for polarization relaxation. The latter effect turns to be significant for our system of very low conductivity fluids, for which the Maxwell-Wagner time is of the order of tens of seconds. We will discuss the complex interplay of shape deformation, surfactant redistribution, and interfacial charging in droplet electrohydrodynamics. Our results are important for understanding electrorheology of emulsions commonly found in the petroleum industry. 1 supported by NSF-CBET -1132614 F2.00015 Enhancing capillary rise on a rough surface1 , MELISSA CHOW, JASON WEXLER, IAN JACOBI, HOWARD STONE, Princeton University — Liquid-infused surfaces have been proposed as a robust alternative to traditional air-cushioned superhydrophobic surfaces. However, if these surfaces are held vertically the lubricating oil can drain from the surface, and cause the surface to lose its novel properties. To examine this failure mode, we measure the drainage from a surface with model roughness that is scaled-up to allow for detailed measurements. We confirm that the bulk fluid drains from the surface until it reaches the level of the capillary rise height, although the detailed dynamics vary even in simple surface geometries. We then test different substrate architectures to explore how the roughness can be designed to retain greater amounts of oil. 1 Supported under MRSEC NSF DMR 0819860 (PI: Prof. N. Phuan Ong) REU Site Grant: NSF DMR-1156422 (PI: Prof. Mikko Haataja), PREM CSUN Prime # NSF 1205734 and ONR MURI Grants N00014-12-1-0875 and N00014-12-1- 0962 (Program Manager Dr. Ki-Han Kim). F2.00016 Stereo Refractive Imaging of Breaking Free-Surface Waves in the Surf Zone , TRACY MANDEL, JOEL WEITZMAN, JEFFREY KOSEFF, Stanford University, ENVIRONMENTAL FLUID MECHANICS LABORATORY TEAM — Ocean waves drive the evolution of coastlines across the globe. Wave breaking suspends sediments, while wave run-up, run-down, and the undertow transport this sediment across the shore. Complex bathymetric features and natural biotic communities can influence all of these dynamics, and provide protection against erosion and flooding. However, our knowledge of the exact mechanisms by which this occurs, and how they can be modeled and parameterized, is limited. We have conducted a series of controlled laboratory experiments with the goal of elucidating these details. These have focused on quantifying the spatially-varying characteristics of breaking waves and developing more accurate techniques for measuring and predicting wave setup, setdown, and run-up. Using dynamic refraction stereo imaging, data on free-surface slope and height can be obtained over an entire plane. Wave evolution is thus obtained with high spatial precision. These surface features are compared with measures of instantaneous turbulence and mean currents within the water column. We then use this newly-developed ability to resolve three-dimensional surface features over a canopy of seagrass mimics, in order to validate theoretical formulations of wave-vegetation interactions in the surf zone. F2.00017 Mechanism of the lift force acting on a levitating drop over a moving surface1 , MASAFUMI SAITO, YOSHIYUKI TAGAWA, MASAHARU KAMEDA, Tokyo Univ of Agri & Tech — The purpose of this study is to understand the levitation mechanism of a drop over a moving surface. In our experiment we softly deposit a silicon-oil drop onto the inner wall of a rotating hollow cylinder. With sufficiently large velocity of the wall, the drop steadily levitates. The drop reaches a stable angular position in the cylinder, where the drag and lift balance the weight of the drop. The lift force, which is vital for the levitation, is generated inside a thin air film existing between the drop and the wall. Here three-dimensional shape of the air film plays a crucial role for the magnitude of the lift force. Note that, although the shapes of some levitating drops had been reported, the lift estimated from the shape had not been validated. Using interferometric technique, we measure the three-dimensional shape of the air film under the drop. We then calculate the lift by applying the lubrication theory. This lift is compared with that estimated from the angular position. Both lifts show a fair agreement. In addition, we investigate the shapes of the air film under drops with various sizes, viscosities and wall velocities. We discuss effects of these parameters on the shape and the lift. 1 JSPS KAKENHI Grant Number 26709007 F2.00018 BIO AND BIO-INSPIRED — F2.00019 The recreation of a unique shrimp’s mechanically induced cavitation bubble1 , RYAN MILLER, CHRISTOPHER DOUGHERTY, VERONICA ELIASSON, GAURI KHANOLKAR, University of Southern California — The Alpheus heterochaelis, appropriately nicknamed the “pistol shrimp,” possesses an oversized claw that creates a cavitation bubble upon rapid closure. The implosion of this bubble results in a shock wave that can stun or even kill the shrimp’s prey (Versluis et al., 2000). Additionally, the implosion is so violent that sonoluminescence may occur. This light implies extreme temperatures, which have been recorded to reach as high as 10,000 K (Roach, 2001). By developing an analogous mechanism to the oversized claw, the goal of this experiment is to verify that cavitation can be produced similar to that of the pistol shrimp in nature as well as to analyze the resulting shock wave and sonoluminescence. High-speed schlieren imaging was used to observe the shock dynamics. Furthermore, results on cavitation collapse and light emission will be presented. 1 USC Provost Undergraduate Research Fellowship/Rose Hills Undergraduate Research Fellowship F2.00020 Swimming Vorticella convallaria in various confined geometries1 , LUZ SOTELO, University of Texas-Pan American, DONGHEE LEE, University of Nebraska-Lincoln, SUNGHWAN JUNG, Virginia Polytechnic Institute and State University, SANGJIN RYU, University of Nebraska-Lincoln — Vorticella convallaria is a stalked ciliate observed in the sessile form (trophont) or swimming form (telotroch). Trophonts are mainly composed of an inverted bell-shaped cell body generating vortical feeding currents, and a slender stalk attaching the cell body to a substrate. If the surrounding environment is no longer suitable, the trophont transforms into a telotroch by elongating its cell body into a cylindrical shape, resorbing its oral cilia and producing an aboral cilia wreath. After a series of contractions, the telotroch will completely detach from the stalk and swim away to find a better location. While sessile Vorticella has been widely studied because of its stalk contraction and usefulness in waste treatment, Vorticella’s swimming has not yet been characterized. The purpose of this study is to describe V. convallaria’s swimming modes, both in its trophont and telotroch forms, in different confined geometries. Using video microscopy, we observed Vorticellae swimming in semi-infinite field, in Hele-Shaw configurations, and in capillary tubes. Based on measured swimming displacement and velocity, we investigated how V. convallaria’s mobility was affected by the geometry constrictions. 1 We acknolwedge support from the First Award grant of Nebraska EPSCoR. F2.00021 Bio-inspired propulsor using internally powered flexible fins , PETER YEH, ALPER ERTURK, ALEXANDER ALEXEEV, Georgia Institute of Technology — Using experiments and three dimensional numerical simulations, we study the underwater locomotion of internally powered flexible plates. The flexible plate is composed of Macro-Fiber Composite (MFC) piezoelectric laminates. A sinusoidally varying voltage is applied to the MFCs, causing bending and generating thrust similar to a flapping fin in carangiform motion. In our fully coupled FSI simulations, we model the swimmer as a rectangular elastic plate actuated by a sinusoidal internal moment. The steady state swimming velocity and thrust are measured experimentally and compared to our numerical simulations. Our results can be used to design underwater self-propelling vehicles driven by internally powered flexible fins. F2.00022 The Effects of Including Piezoelectric Film as Part of a Wing Surface1 , CHARLOTTE SAPPO, Smith College — Micro air vehicles (MAVs) are size- and weight-restricted, unmanned, flying vehicles that often exploit biology for inspiration. Membrane wings, one commonly employed biological adaptation, improves aerodynamic efficiency. These efficiency gains are due to the passive deformations and vibrations of the membrane. Piezoelectric films have the potential to further utilize these vibrations through the conversion of this motion into measureable electrical energy. In this investigation, an amplifier circuit was designed to measure the charge generated by a flexible polyvinylidene fluoride (PVDF) film adhered to a rectangular wing frame (aspect ratio of 2). The trailing edge was unattached and free to vibrate. The circuit consisted of two charge amplifiers, to convert the high impedance charge of the piezoelectric film into an output voltage, and an instrumentation amplifier, to reject common-mode noise. Amplifying and filtering the output signal appropriately, through the use of the feedback capacitance and resistance, was discovered to be of the utmost importance for this endeavor. Results from shaker and wind tunnels tests are presented. 1 NSF ECE grant 1358991 supported the first author as a REU student F2.00023 Effects of Fluid Shear Stress on Cancer Stem Cell Viability1 , BRITTNEY SUNDAY, Case Western Reserve University, URSULA TRIANTAFILLU, RIA DOMIER, YONGHYUN KIM, University of Alabama — Cancer stem cells (CSCs), which are believed to be the source of tumor formation, are exposed to fluid shear stress as a result of blood flow within the blood vessels. It was theorized that CSCs would be less susceptible to cell death than non-CSCs after both types of cell were exposed to a fluid shear stress, and that higher levels of fluid shear stress would result in lower levels of cell viability for both cell types. To test this hypothesis, U87 glioblastoma cells were cultured adherently (containing smaller populations of CSCs) and spherically (containing larger populations of CSCs). They were exposed to fluid shear stress in a simulated blood flow through a 125-micrometer diameter polyetheretherketone (PEEK) tubing using a syringe pump. After exposure, cell viability data was collected using a BioRad TC20 Automated Cell Counter. Each cell type was tested at three physiological shear stress values: 5, 20, and 60 dynes per centimeter squared. In general, it was found that the CSC-enriched U87 sphere cells had higher cell viability than the CSC-depleted U87 adherent cancer cells. Interestingly, it was also observed that the cell viability was not negatively affected by the higher fluid shear stress values in the tested range. In future follow-up studies, higher shear stresses will be tested. Furthermore, CSCs from different tumor origins (e.g. breast tumor, prostate tumor) will be tested to determine cell-specific shear sensitivity. 1 National Science Foundation grant #1358991 supported the first author as an REU student. F2.00024 Design of a millifluidic device for thermophoretic analysis of biomolecules , RYAN PHELPS, TYLER SHARBY, Department of Biology and Marine Biology, Roger Williams University, JENNIFER KREFT PEARCE, Department of Physics, Roger Williams University — Thermophoresis is the migration of particles due to a temperature gradient, which is enhanced in small channels due to the high temperature gradients that can be achieved. Thermophoresis can be used to analyze biomolecules such as proteins and DNA. It can also be used to study the absorption of small molecules to lipid membranes. For this experiment a millifluidic device is used. The channel in which the sample is injected is 500 microns wide. The temperature gradient is produced by hot and cold water baths. This device is a low cost alternative to the commercially available systems for thermophoresisbased analysis of biological molecules. F2.00025 Lattice-Boltzmann-based simulations of membrane protein dynamics , TYLER SHARBY, RYAN PHELPS, MICHAEL ANTONELLI, Department of Biology and Marine Biology, Roger Williams University, JENNIFER KREFT PEARCE, Department of Physics, Roger Williams University — The cell membrane is a complex structure composed of a phospholipid bilayer and embedded proteins. Recent work has shown that regions of different mobility exist in the membrane due to a variety of factors and that protein motion can be significantly subdiffusive due to the presence of stationary obstacles. We present work that shows that the combination of stationary obstacles and regions of different mobility can lead to aggregation of proteins in certain regions of the cell membrane. The concentration of stationary proteins is below the percolation threshold. The mechanism of this process is hydrodynamically-mediated interactions of diffusing proteins with themselves, as in hydrodynamic memory, and with obstacles. F2.00026 Microfluidic mixing using a heterogeneous array of actuated synthetic cilia1 , MATTHEW BALLARD, PUJA DE, ALEXANDER ALEXEEV, Georgia Institute of Technology — We use three-dimensional numerical simulations to examine mixing of an initially segregated viscous fluid solution in a microchannel containing a heterogeneous array of actuated synthetic cilia. We model the cilia as elastic filaments attached to the channel walls and actuated by an external periodic force. Fluid flow is modelled using a lattice Boltzmann model treating concentration as a scalar, coupled with a lattice spring model simulating the elastic cilia. To investigate the effects of the oscillating heterogeneous cilia on microfluidic mixing of fluid solutions of different diffusivity, we consider the effects of cilia relative size, elasticity, spacing, and oscillation pattern. We demonstrate that arrays of heterogeneous cilia can provide enhanced mixing over that achievable with a homogeneous array of identical cilia. Thus, our findings further the understanding of how heterogeneous arrays of active bio-mimetic structures can be used to enhance mixing in microfluidic devices. 1 This work is supported by a grant from the National Science Foundation. F2.00027 THERMAL AND COMBUSTION — F2.00028 Dynamic behavior of thermoacoustic combustion oscillations in a lean premixed gasturbine model combustor with and without active control , RYOSUKE TSUJIMOTO, SHOHEI DOMEN, YUTA OKUNO, YOSHITAKE NAKAGAKI, HIROSHI GOTODA, Ritsumeikan University — We experimentally study the dynamic behavior of thermoacoustic combustion oscillations in a laboratory-scale lean premixed gas-turbine model combustor with and without active control. We adopt the delayed feedback control method based on the concept of chaos control to suppress thermoacoustic combustion oscillations. The unstable periodic orbits in the attractor of uncontrolled thermoacoustic combustion oscillations are led to the desired orbits with a small diameter of the attractor when the perturbation is switched on, resulting in the notable suppression of thermoacoustic combustion oscillations. Color-recurrence plots (Gotoda et al., Physical Review E, 89, 022910 (2014)) are used for characterizing the complexity of the combustion state with and without delayed feedback control. F2.00029 Experimental Configuration Effects on ICE Tumble Flow Evaluation1 , BRYAN SANTANA, Universidad del Turabo, PAULIUS PUZINAUSKAS, University of Alabama — The generation of ICE (Internal Combustion Engine) in-cylinder charge motions, such as swirl and tumble, have shown positive effects on reducing fuel consumption and exhaust emission levels at partial engine loads. Tumble flow is commonly measured utilizing a steady-flow rig and two-dimensional PIV (Particle Image Velocimetry) systems, among others. In order to optimize the tumble flow, it is important to retrieve accurate measurements. The tumble flow values could be affected by variations in the geometry and/or design of the steady-flow rig utilized during flow tests. In this research, a four-valve per cylinder head was tested on a steady flow bench, varying several aspects of the configuration to evaluate how they influence bulk momentum as well as PIV measurements. The configuration variations included symmetrical, asymmetrical and runner-fed configurations throughout testing. Volumetric flow rate and tumble strength flow measurements were retrieved at the selected L/D ratios. Additionally, several PIV seeding particles were characterized for size and shape. Corresponding PIV flow measurements using each type of seeding were made to evaluate how the particles influence the results. 1 NSF ECE Grant #1358991 supported Bryan Santana Rivera as an REU student. F2.00030 Laser Diagnostic Methods to Characterize Soot Evolution in Diesel-relevant Fuels1 , STEVEN OVERHEIM, BRIAN FISHER, University of Alabama — Soot particles are a harmful byproduct of diesel combustion and can be detrimental to the environment and our health. The purpose of this research is to gain a better understanding of how the soot formation, growth, and oxidation are directly related to the chemical structure of the fuel in a diffusion flame. Such understanding is expected to help with soot reduction methods in the future. A new method to analyze soot concentrations was developed combining previous successful methods of experimentation. The new method employs combined elastic scattering and extinction to characterize soot formation, growth, and oxidation throughout the flame. These concentrations are quantifiable through the use of a 532-nm Nd:YAG laser and carefully calibrated photodetectors as optical measuring tools. This study focused on the doping of the diffusion flame with toluene, which has an aromatic molecular structure. The diffusion flame is doped with a low concentration of toluene, 1000 ppm, in its fuel stream and compared to a methane-fueled base flame. By comparing the doped flame to the methane/oxygen base flame, the higher level of active soot formation in the doped flame was clearly observed. Future work on the project will entail further data analysis to convert measured signals into quantitative soot size and concentration information. 1 NSF ECE Grant #1358991 supported the first author as an REU student. F2.00031 On the Possibility of Condensation during Supercritical Fuel Injection1 , LU QIU, ROLF REITZ, University of Wisconsin-Madison — Supercritical fuel injection into a nitrogen environment was simulated using Peng-Robinson equation of state. The real gas simulation was found to match the experimental injectant density much better than the ideal gas simulation, emphasizing the importance of applying realistic equation of state model. Possible fuel condensation processes were also investigated by considering the stability of the single phase by utilizing fundamental thermodynamics principles. Several conclusions from the experiments are also seen from the simulations. First, though both the injection and chamber pressures are above the critical pressure of the injectant, condensation can become possible as long as their temperature difference is large enough, and when this occurs, the fluid is able to enter the two-phase region. Condensation is found to be enhanced when the chamber temperature is further reduced, indicating that the fluid is in a state further away from the phase border. In addition, the newly formed condensed phase is found to exist only in the jet boundary where there are strong interactions between the “hot” injectant and the “cold” nitrogen. Finally, it was concluded that the local strong heat and mass exchange sent the mixture into the two-phase region by crossing the dew point line with the commencement of condensation. 1 The research work was sponsored by Department of Energy and Sandia National Laboratories through the Advanced Engine Combustion Program (MOU 04-S-383). F2.00032 Flow Diagnostics of Swirl Stabilized Combustion with and without Porous Inert Media1 , CAROLINA VEGA RECALDE, Oklahoma State University, AJAY AGRAWAL COLLABORATION, JOSEPH MEADOWS COLLABORATION, ZACHARY SMITH COLLABORATION, JOHN KORNEGAY COLLABORATION — Due to regulations, the industry and the scientific community have become interested in combustion noise and thermo-acoustic instabilities, especially those produced under lean-premixed (LPM) conditions. Instabilities are self-excited and arise when energy from combustion is added to the system faster than energy is dissipated by heat transfer. Given that porous inert media (PIM) has been shown to mitigate combustion noise and thermo-acoustic instabilities in lean direct injection (LDI) using kerosene fuel, the present study examined the flow fields produced with and without PIM. By using time-resolved particle image velocimetry (TR-PIV) and proper orthogonal decomposition (POD) techniques, the non-reacting and reacting flow fields will be studied to determine the underlying mechanisms. The purpose of this experiment is to gain more understanding of how this PIM material works. Since PIM has been shown to reduce these instabilities, modifications to combustors can be made to make them more efficient and safe for the environment. 1 NSF ECE Grant #1358991 and NASA Award No. NNX13AN14A are gratefully acknowledged. F2.00033 Modeling the Optimal Heat Transfer Fluidization Velocity in Gas-Fluidized Beds1 , THOMAS PREDEY, JON BOUGIE, ALEKSANDR GOLTSIKER, Loyola University Chicago — Fluidized beds are vital to a wide range of industrial applications and are useful for studying two-phase flow. However, modeling the optimal heat transfer fluidization velocity (OHTFV) in such beds has remained difficult. Previous investigations have commonly taken one of two approaches. One such approach attempts to find a general scaling formula for homogeneous fluidized beds by taking a harmonic average between the terminal and minimum fluidization velocities. Modern approaches using computer simulations and a wide range of parameters are more commonly used in industry today, but are generally concerned with specific applications. We propose a third approach, taking into account the inhomogeneity of the fluidized bed system while limiting the input parameters to gas velocity and particle size. We use this approach to find a general formula for OHTFV that accounts for the collective behavior of the particles rather than focusing on each individual particle in the bed. We then compare this model to previous experimental results. 1 This research is supported by the Loyola Undergraduate Research Opportunities Program. F2.00034 Flux Variability from Turbulence and Bulk Velocity Variations in Relativistic Hydrodynamic Jets1 , MAXWELL POLLACK, DAVID PAULS, PAUL WIITA, The College of New Jersey — We simulated relativistic hydrodynamic jets using the Athena MHD code incorporating special relativity (Beckwith & Stone 2011). We compared the long-timescale variations produced by changes in the bulk velocity within the jet, amplified by Doppler boosting, to the short-timescale variations caused by turbulence in the flow. The flux variability due to changes in bulk velocity was calculated along a band spanning the width of the jet at a fixed distance down its stream, positioned just behind a reconfinement shock. We computed the relativistic turbulence variability by summing the results from our relativistic turbulence code over multiple zones; this required incorporating time delays. Power Spectral Densities were then computed for both turbulent and bulk velocity flux variations, and compared. For reasonable jet widths of ∼40 light-years, we found turbulent fluctuations on timescales of days to years and bulk-velocity variations contributing on longer timescales. We found that the slopes of the turbulent and bulk PSDs were usually between −1.5 and −2.2, in accord with observations of Active Galactic Nuclei. 1 Supported by the Mentored Undergraduate Summer Experience at TCNJ F2.00035 Modeling Relativistic Jets Using the Athena Hydrodynamics Code1 , DAVID PAULS, MAXWELL POLLACK, PAUL WIITA, The College of New Jersey — We used the Athena hydrodynamics code (Beckwith & Stone 2011) to model early-stage two-dimensional relativistic jets as approximations to the growth of radio-loud active galactic nuclei. We analyzed variability of the radio emission by calculating fluxes from a vertical strip of zones behind a standing shock, as discussed in the accompanying poster. We found the advance speed of the jet bow shock for various input jet velocities and jet-to-ambient density ratios. Faster jets and higher jet densities produce faster shock advances. We investigated the effects of parameters such as the Courant-Friedrichs-Lewy number, the input jet velocity, and the density ratio on the stability of the simulated jet, finding that numerical instabilities grow rapidly when the CFL number is above 0.1. We found that greater jet input velocities and higher density ratios lengthen the time the jet remains stable. We also examined the effects of the boundary conditions, the CFL number, the input jet velocity, the grid resolution, and the density ratio on the premature termination of Athena code. We found that a grid of 1200 by 1000 zones allows the code to run with minimal errors, while still maintaining an adequate resolution. 1 This work is supported by the Mentored Undergraduate Summer Experience program at TCNJ. F2.00036 Discrete Element Simulation of Density Induced Segregation in Binary Granular Mixtures1 , ANNETTE VOLK, Univ of Cincinnati, URMILA GHIA, University of Cincinnati — We study density induced segregation of binary granular mixtures under vertical vibration using the open source discrete element method (DEM) code LIGGGHTS. Published experiments of vertically vibrated binary mixtures, varying in density ratio and observed under differing intensities of vertical vibration, are simulated and the final segregation state is quantitatively compared. Simulation results compare well with experimental data when the density ratio between the binary particles is relatively small but the comparison slowly deteriorates as the density ratio increases. A sensitivity study is performed for the coefficient of restitution since this quantity is absent from the published experimental data but has been shown to affect the amount of segregation. In industrial applications, mixing/segregation time is vital to processing, and hence, the relationship between the time to reach final segregation state and the density ratio of the binary particles is also investigated. Finally, in an effort to increase computational efficiency while maintaining accuracy, the effect of domain size on both time to reach final segregation state and amount of segregation in the final state is assessed. 1 This material is based on work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. 2013170606 Monday, November 24, 2014 8:00AM - 10:10AM Session G1 Non-Newtonian Flows: Rheology — 3000 - Juan C. Del Alamo, University of California, San Diego 8:00AM G1.00001 Quantitative Rheological Model Selection , JONATHAN FREUND, RANDY EWOLDT, University of Illinois at Urbana-Champaign — The more parameters in a rheological the better it will reproduce available data, though this does not mean that it is necessarily a better justified model. Good fits are only part of model selection. We employ a Bayesian inference approach that quantifies model suitability by balancing closeness to data against both the number of model parameters and their a priori uncertainty. The penalty depends upon prior-to-calibration expectation of the viable range of values that model parameters might take, which we discuss as an essential aspect of the selection criterion. Models that are physically grounded are usually accompanied by tighter physical constraints on their respective parameters. The analysis reflects a basic principle: models grounded in physics can be expected to enjoy greater generality and perform better away from where they are calibrated. In contrast, purely empirical models can provide comparable fits, but the model selection framework penalizes their a priori uncertainty. We demonstrate the approach by selecting the best-justified number of modes in a Multi-mode Maxwell description of PVA-Borax. We also quantify relative merits of the Maxwell model relative to powerlaw fits and purely empirical fits for PVA-Borax, a viscoelastic liquid, and gluten. 8:13AM G1.00002 Improved johnson segalman model for couette flow , NARIMAN ASHRAFI, None — An improved nonlinear viscoelastic model is proposed and examined for a flow between parallel plates. The model takes into account the interrelations of velocity gradients and stress components through introduction of appropriate coefficients. For a typical viscoelastic material, the coefficients are evaluated and incorporated within the model to simulate the flow of nonlinear Couette flow. In special cases, of the proposed model, typical upper convected Maxwell model and Johnson-Segalman fluid can be recovered from the proposed model further verifying the formulation. The proposed form of constitutive equation almost completely models the physical behavior of a wide range of nonlinear materials, yet it is computationally appropriate as well. The model also allows for the velocity and stress components to be represented by truncated series functions to be used for numerical purposes. 8:26AM G1.00003 2-Point Particle Tracking Microrheology of Directional Viscoelastic Gels , MANUEL GOMEZ-GONZALEZ, JUAN C. DEL ALAMO, University of California, San Diego — By applying Particle Tracking Microrheology we can measure the stiffness of the cell cytoplasm, using a spherical microparticle as a probe. PTM relies on the assumption of isotropy, but this hypothesis breaks for highly oriented materials. In order to apply PTM to them, we have calculated the drag force of a particle embedded in a directional viscoelastic gel, modeled as a directional viscoelastic network frictionally coupled to a viscous isotropic fluid. The directional network is modeled with the Leslie-Ericksen equations and the fluid with the Stokes equation. The motion of particles embedded in such a directional gel is dependent on up to three viscoelasticity coefficients, but only two can be calculated from tracking a single probing particle. We have calculated the first order perturbation that the motion of one probe induces on a distant particle, as a function of the three viscoelasticity coefficients. By correlating the motion of two distant particles we can measure such a perturbation and obtain three independent equations that univocally determine the three viscoelasticity coefficients. We show the accuracy of the Directional 2-Point PTM by applying it to a control numerical experiment, and finally we apply it to an essential biological sample such as nematic F-actin. 8:39AM G1.00004 Immersed Particle Dynamics in Fluctuating Fluids with Memory , CHRISTEL HOHENEGGER, University of Utah, SCOTT MCKINLEY, University of Florida — Multibead passive microrheology characterizes bulk fluid properties of viscoelastic liquids by connecting statistically measurable quantities (e.g. mean-square displacement, auto-correlation to mechanical fluid properties (loss and storage modulus). Understanding how these material properties relate to biological quantities (e.g. exit time, first passage time through a layer) is of crucial importance for many pharmaceutical and industrial applications. To correctly model the correlations due to the fluid’s memory, it is necessary to include a thermally fluctuating stress in the Stokes equations (Landau and Lifschitz 1958). We present such a model for an immersed particle passively advected by a fluctuating Maxwellian fluid. We describe the resulting stochastic partial differential equations for the underlying non-Markovian, stationary fluid velocity process and we present a covariance based numerical method for generating particle paths. Finally, we apply standard experimental one and two-point microrheology protocol to recover bulk loss and storage modulus and quantify the resulting errors. Our approach can be applied to a Stokes fluid with memory created by a large suspension of active swimmers or to the diffusion of a particle in a crowded environment. 8:52AM G1.00005 Membrane Elastegrities: A New Model Viscoelastic Structure , ELEFTHERIOS PAVLIDES, JENNIFER PEARCE, Roger Williams University — We propose a new class of structures, membrane elastegrities, a network of rigid and elastic members maintaining shape through elastic forces, named by analogy to tensegrities that maintain shape through tension alone. Numerous researchers have proposed tensegrities as models to biological structure. Elastegrities expand tensegrity properties primarily by suggesting a mechanism for containing and pumping non-Newtonian fluids in living organisms. The chiral icosahedral elastegrity compared to the 6-strut tensegrity have identical symmetry, negative Poisson Ratio, and the reaction force to external forces is distributed throughout the elastic members causing reversible deformation. They also have important differences: a) elastic hinge connections enable containment and pumping of fluids versus nodal connections, b) simple assembly by folding a flat shape-memory material versus assembly requiring scaffolding, c) hinge connections limit freedom of movement resulting in isometric forces as members rotate cooperatively contracting versus large freedom of movement with unpredictable deformation. d) The chiral icosahedral elastegrity can contain liquid and requires increased force for equal displacement as it rotates towards a zero volume octahedron suggesting a mechanism for a non-Newtonian pump. 9:05AM G1.00006 Rubber and gel origami: visco- and poro-elastic behavior of folded structures , ARTHUR EVANS, NAKUL BENDE, JUNHEE NA, RYAN HAYWARD, CHRISTIAN SANTANGELO, University of Massachusetts, Amherst — The Japanese art of origami is rapidly becoming a platform for material design, as researchers develop systematic methods to exploit the purely geometric rules that allow paper to folded without stretching. Since any thin sheet couples mechanics strongly to geometry, origami provides a natural template for generating length-scale independent structures from a variety of different materials. In this talk I discuss some of the implications of using polymeric sheets and shells over many length scales to create folded materials with tunable shapes and properties. These implications include visco-elastic snap-through transitions and poro-elastically driven micro origami. In each case, mechanical response, dynamics, and reversible folding is tuned through a combination of geometry and constitutive properties, demonstrating the efficacy of using origami principles for designing functional materials. 9:18AM G1.00007 Particle migration in two-phase, viscoelastic flows1 , NICK JAENSSON, MARTIEN HULSEN, PATRICK ANDERSON, Eindhoven Univ of Tech — Particles suspended in creeping, viscoelastic flows can migrate across stream lines due to gradients in normal stresses. This phenomenon has been investigated both numerically and experimentally. However, particle migration in the presence of fluid-fluid interfaces is hardly studied. We present results of simulations in 2D and 3D of rigid spherical particles in two-phase flows, where either one or both of the fluids are viscoelastic. The fluid-fluid interface is assumed to be diffuse and is described using Cahn-Hilliard theory. The particle boundary is assumed to be sharp and is described by a boundary-fitted, moving mesh. The governing equations are solved using the finite element method. We show that differences in normal stresses between the two fluids can induce a migration of the particle towards the interface in a shear flow. Depending on the magnitude of the surface tension and the properties of the fluids, particle migration can be halted due to the induced Laplace pressure, the particle can be adsorbed at the interface, or the particle can cross the interface into the other fluid. 1 Dutch Polymer Institute (DPI), P.O. Box 902, 5600 AX Eindhoven, The Netherlands 9:31AM G1.00008 Experimental study of the solid-liquid interface in a yield-stress fluid flow upstream of a step , LI-HUA LUU, PHILIPPE PIERRE, CHAMBON GUILLAUME, IRSTEA — We present an experimental study where a yield-stress fluid is implemented to carefully examine the interface between a liquid-like unyielded region and a solid-like yielded region. The studied hydrodynamics consists of a rectangular pipe-flow disturbed by the presence of a step. Upstream of the step, a solid-liquid interface between a dead zone and a flow zone appears. This configuration can both model geophysical erosion phenomenon in debris flows or find applications for industrial extrusion processes. We aim to investigate the dominant physical mechanism underlying the formation of the static domain, by combining the rheological characterization of the yield-stress fluid with local measurements of the related hydrodynamic parameters. In this work, we use a model fluid, namely polymer micro-gel Carbopol, that exhibits a Hershel-Bulkley viscoplastic rheology. Exploiting the fluid transparency, the flow is monitored by Particle Image Velocimetry thanks to internal visualization technique. In particular, we demonstrate that the flow above the dead zone roughly behaves as a plug flow whose velocity profile can successfully be described by a Poiseuille equation including a Hershel-Bulkley rheology (PHB theory), with exception of a thin zone at the close vicinity of the static domain. The border inside the flow zone above which the so-called PHB flow starts, is found to be the same regardless of the flow rate and to move with a constant velocity that increases with the flow rate. We interpret this feature as a slip frontier. 9:44AM G1.00009 The Propagation of the Gravity Current of Viscoplastic Fluid1 , YE LIU, UBC — We are studying the spreading of the viscoplastic fluid of Bingham type over a horizontal plane, using both mathematical derivation and numerical experiments. We are interested in its final shape and whether theory and numerics correspond well. There are two theories for comparison: lubrication theory from asymptotics, and slipline theory from plasticity. The numerical method we are using is based on the volume-of-fluid method, with both regularization and Augmented Lagrangian for the constitutive law of Bingham type fluid. 1 UBC IRSN 9:57AM G1.00010 Effects of confinement & surface roughness in electrorheological flows , AHMED HELAL, Department of Mechanical Engineering, Massachusetts Institute of Technology, MARIA J. TELLERIA, Pneubotics, JULIE WANG, Department of Mechanical Engineering, Massachusetts Institute of Technology, MARC STRAUSS, MIKE MURPHY, Boston Dynamics, GARETH MCKINLEY, A.E. HOSOI, Department of Mechanical Engineering, Massachusetts Institute of Technology — Electrorheological (ER) fluids are dielectric suspensions that exhibit a fast, reversible change in rheological properties with the application of an external electric field. Upon the application of the electric field, the material develops a field-dependent yield stress that is typically modeled using a Bingham plastic model. ER fluids are promising for designing small, cheap and rapidly actuated hydraulic devices such as rapidly-switchable valves, where fluid flowing in a microchannel can be arrested by applying an external electric field. In the lubrication limit, for a Bingham plastic fluid, the maximum pressure the channel can hold, before yielding, is a function of the field-dependent yield stress, the length of the channel and the electrode gap. In practice, the finite width of the channel and the surface roughness of the electrodes could affect the maximum yield pressure but a quantitative understanding of these effects is currently lacking. In this study, we experimentally investigate the effects of the channel aspect ratio (width/height) and the effects of electrode roughness on the performance of ER valves. Based on this quantitative analysis, we formulate new performance metrics for ER valves as well as design rules for ER valves that will help guide and optimize future designs. Monday, November 24, 2014 8:00AM - 9:57AM Session G2 Suspensions: General — 3002 - Elisabeth Guazzelli, Ecole d’ingenieurs universitaire - Polytech marseille 8:00AM G2.00001 Geometrically-protected reversibility in hydrodynamic Loschmidt-echo experiments , RAPHAEL JEANNERET, JOOST WEIJS, DENIS BARTOLO, ENS Lyon — We demonstrate an archetypal Loschmidt-echo experiment where thousands of droplets interact in a reversible fashion via a viscous fluid. Firstly, we show that, unlike equilibrium systems, periodically driven microfluidic emulsions self-organize and geometrically protect their macroscopic reversibility. This self-organization is not merely dynamical, it has a clear structural signature akin to the one found in a mixture of molecular liquids. Secondly, we evidence that above a maximal shaking amplitude both structural order and reversibility are lost simultaneously in the form of a 1st order non-equilibrium phase transition. Thirdly, we account for this discontinuous transition, in term of a memory-loss process. 8:13AM G2.00002 Colloidal deposition and aggregation in the presence of charged collectors , BEHNAM SADRI, ARVIND RAJENDRAN, University of Alberta, SUBIR BHATTACHARJEE, Water planet engineering, COLLOIDS AND COMPLEX FLUID LABORATORY TEAM1 — The transport of colloidal particles in porous media is of great importance in sub-surface environments. These colloidal particles facilitate transport of contaminants, low-soluble compounds and metals in groundwater. Here, we have studied transport dynamics of colloids inside porous medium using a combination of column experiments and batch studies. Polystyrene latex beads (100nm), as colloidal agents, and soda lime glass beads, as porous medium, are employed in this work. On the one hand, batch experiments are undertaken to better understand concurrent aggregation and deposition of particles. On the other hand, column experiments are performed to understand the flow induced deposition of colloidal particles in the interstitial voids. Effect of collector surface preparation, pH, colloidal suspension concentration and collector beads mass is studied. Chemical release and shear field are revealed as two significant factors lying behind the coagulation of colloidal particles. These findings help us to better distinguish mechanisms responsible for the transport of colloids inside porous medium. 1 We are collaborators. Behnam Sadri is master of science student while two other professor are supervising his research work. 8:26AM G2.00003 The role of short-ranged and long-ranged hydrodynamic interactions on aggregation of colloidal particle in colloid-polymer mixtures , ARMAN BOROMAND, SAFA JAMALI, JOAO MAIA, Case Western Reserve University — Colloidal Gels i.e. disordered arrested systems has been studied extensively during the past decades both experimentally and computationally. Despite their widespread applications in various industries e.g. cosmetic, food, their physical principals are still far beyond being understood. The interplay between different types of interactions e.g. quantum scale, short-ranged, and long-ranged turned dynamics and thermodynamics of the colloidal systems to one the most intriguing areas in Physics. Many authors have implemented different simulation techniques such as molecular dynamics (MD) and Brownian dynamics (BD) to capture better picture during phase separation in colloidal system with short-ranged attractive force e.g. colloid-polymer mixtures. However, BD neglects multi-body hydrodynamic interactions (HI) and MD is limited considering the time and length scale of gel formation and long-time dynamics. In this presentation we used Core-modified dissipative particle dynamics (CM-DPD) with modified depletion potential, as a coarse-grain model, to address the gel formation process in short ranged-attractive colloidal systems. Due to the possibility to study short- and long-ranged HI separately in this method we studied the effect of each of those interactions on the final morphology and report on one of the controversial question in this field. In the second part of the presentation, we include colloidal-polymer interactions to extend/modify the Asakura-Oosawa potential model to semi-dilute region of polymer solution. 8:39AM G2.00004 Flow-structure interaction of falling cones in unbounded flow media , DAN TROOLIN, LAI WING, TSI, YAQING JIN, A.H. HAMED, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, CARLO ZUNIGA ZAMALLOA, University of Illinois at Urbana-Champaign, LEONARDO P. CHAMORRO, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — The kinematics of falling objects in a fluid media at rest are dominated by the vorticity dynamics generated in the vicinity of the object. At a critical Reynolds number, the large-scale vortical structures shed by the body lose their axisymmetric character leading to unsteady lift and, consequently, body rotation. The dynamics of these motions depend on the body shape and can range from simple oscillatory motions to chaotic behavior. In this study, 2D and 3D Particle Image Velocimetry (PIV) are used to characterize the turbulence in the vicinity of cones of various shapes (aspect ratios) falling in a fluid media at complete rest. Translations and rotations experienced by the cones are tracked with a miniature and highly sensitive 3-axis accelerometer and 3-axis gyroscope inserted in the object. Coupling between vortex dynamics and body motions is characterized at various Reynolds numbers and cone shapes. 8:52AM G2.00005 Effects of Hydrodynamic Interaction in Aerosol Particle Settling: Mesoscopic Particle-level Full Dynamics Simulations1 , SHUIQING LI, MENGMENG YANG, Tsinghua University, JEFFREY MARSHALL, University of Vermont — A new mesoscopic particle-level approach is developed for the full dynamics simulation (FDS) of the settling of systems of aerosol micro-particles. The approach efficiently combines an adhesive discrete-element method for particle motions and an Oseen dynamics method for hydrodynamic interactions. Compared to conventional Stokeslet and Oseenlet simulations, the FDS not only accounts for the cloud-scale fluid inertia effect and the particle inertia effect, but also overcomes the singularity problem using a soft-sphere model of adhesive contact. The effect of hydrodynamic interactions is investigated based on FDS results. The particle inertia is found to reduce the mobility of particle clouds and to elongate the cloud on vertical direction. Meanwhile, the fluid inertia decreases the settling velocity by weakening the hydrodynamic interaction and tends to flatten the cloud, leading to breakup. Expressions for the settling velocity of particle cloud are proposed with consideration of fluid inertia effect and the cloud shape. Finally, the transformation in settling behavior from a finite particle cloud to an unbounded uniform suspension is explained. 1 This work has been funded by the National Natural Science Funds of China (No. 50976058), and by the National Key Basic Research and Development Program (2013CB228506). 9:05AM G2.00006 Effect of roughness in periodically sheared clouds of particles , PHONG PHAM, University of Florida - Gainesville, BLOEN METZGER, IUSTI-CNRS UMR 7343, Aix-Marseille University, France., JASON BUTLER, University of Florida Gainesville, IUSTI-CNRS UMR 7343, AIX-MARSEILLE UNIVERSITY, FRANCE. COLLABORATION, UNIVERSITY OF FLORIDA - GAINESVILLE COLLABORATION — We investigate experimentally the evolution of small clouds of non-Brownian particles submitted to a periodic shear under low Reynolds number conditions. The particle motion is irreversible during the first cycle. Beyond that, the particle motion is reversible. We find that the amount of irreversibility increases as the particle roughness is increased. An accurate prediction of the particles’ trajectories is obtained with a minimal model including normal lubrication and a frictionless contact force. These experiments provide evidence that, in viscous flows, contacts between particles occur and strongly influence the particle dynamics. 9:18AM G2.00007 Mixing at low Reynolds number by shearing suspensions , MATHIEU SOUZY, BLOEN METZGER, CHERIFA ABID, IUSTI, EMMANUEL VILLERMAUX, IRPHE, XIAOLONG YIN, Colorado School of Mines — Sheared suspensions provide a unique system where mixing spontaneously occurs even under low Reynolds numbers conditions. Under flow, particles within the fluid experience frequent collisions with one another, and are thus deviated from their laminar streamlines. Particles can be thought of as many “stirrers” inducing disturbances in the fluid phase, which produce an efficient mixing. Using index matching and laser induced fluorescence, we investigate experimentally the evolution of the concentration profiles of a layer of dye initially applied on the outer wall of a cylindrical Couette cell, in a sheared suspension of neutrally buoyant, non-Brownian particles. Close to the walls, although the particle-translational-diffusive motion is frustrated, particle rotation significantly enhances the rate of mass transfer, which is found to propagate across the gap super-diffusively. The fine-scale mixing properties of this disordered flow are investigated as well. The stretching laws of isolated scalar blobs are measured and used to infer the probability density function of the concentration in the medium. 9:31AM G2.00008 Dynamics of shear-induced migration of spherical particles in pipe flow , ELISABETH GUAZZELLI, Aix Marseille University, CNRS, IUSTI UMR 7343, BRADEN SNOOK, Aix Marseille University, CNRS, IUSTI UMR 7343 and Department of Chemical Engineering, University of Florida, JASON BUTLER, Department of Chemical Engineering, University of Florida, AIX-MARSEILLE UNIVERSITY, CNRS, IUSTI UMR 7343 TEAM, DEPARTMENT OF CHEMICAL ENGINEERING, UNIVERSITY OF FLORIDA TEAM — We study the largeoscillation flow of a concentrated suspension in a pipe. Particle volume fraction and particle velocity are examined through refractive index matching techniques. The particles are seen to migrate toward the center of the pipe, i.e. from the region of high to low shear-rate. The dynamics of the shear-induced migration process is analyzed and in particular compared to the prediction of the suspension balance model using realistic rheological laws. 9:44AM G2.00009 Inertial migration of spherical particles in square channel flows , KAZUMA MIURA, Kansai University Graduate School, TOMOAKI ITANO, MASAKO SUGIHARA-SEKI, Kansai University — It has been known that particles suspended in the laminar pipe flow migrate laterally toward a certain radial position due to the inertial effect. In this research, we investigated experimentally the inertial migration of neutrally buoyant spherical particles in square channel flows in the range of Reynolds numbers (Re) from 100 to 1200. The measurement of the particle positions at several cross-sections revealed that there are eight equilibrium positions of the particles in the cross-section, four of them located near the centers of the channel faces and the other four located near the channel corners. The corner equilibrium positions were found to exist only for Re larger than about 260. It was also shown that an increase in Re shifts the channel face equilibrium positions toward the channel center, whereas it shifts the corner equilibrium positions toward the channel corner. As the observation sites become downstream, the particles are more focused near the equilibrium positions. The distribution of the particles measured in a short distance from the channel inlet indicated that the lateral forces exerted on the particles located near the centers of the channel faces would be larger compared to the particles at the other positions in the cross-sections. Monday, November 24, 2014 8:00AM - 10:10AM Session G3 Porous Media Flows IV: General — 3004 - Shima Parsa, Harvard University 8:00AM G3.00001 Application of micro-PIV technique to study multiphase flow of water and liquid CO2 in 2D porous media , F. KAZEMIFAR, G. BLOIS, D.C. KYRITSIS, Univ. of Illinois, K.T. CHRISTENSEN, Univ. of Notre Dame — We study the multiphase flow of water and liquid/supercritical CO2 in 2D porous micromodels, with the goal of developing a more complete understanding of pore-scale flow dynamics for the scenario of geological sequestration of carbon dioxide. Fluorescent microscopy and the micro-PIV technique are employed to simultaneously visualize both phases and obtain the velocity field in the aqueous phase. This technique provides a powerful tool for studying such flow systems and the results give valuable insight into flow processes at the pore scale. The fluid-fluid interface curvature from the images can be used to estimate the local capillary pressure. The velocity measurements illustrate active and passive flow pathways and circulation regions near the fluid-fluid interfaces induced by shear. Thin water films observed on the solid surfaces confirm the hydrophilic nature of the micromodels. The velocity of the said films is measured by particle tracking. 8:13AM G3.00002 ABSTRACT WITHDRAWN — 8:26AM G3.00003 Experimental characterization of the deviation from Darcy flow at low Reynolds numbers through elastic porous matrices1 , SID BECKER, BEN MUNRO, University of Canterbury; Mechanical Engineering — The subject of this study concerns viscous flow through an elastic porous matrial for which the solid matrix is capable of experiencing deformation under the influence of the flow field. The inherent challenges associated with developing experimental testing of flow in deformable porous media are largely related to the fabrication of a deformable matrix. In this study a method of media fabrication is presented that uses an indirect solid free form fabrication process combining 3D Printing with an infused Polydimethylsiloxane elastomer. This allows for the precise control of the matrix parameters: elasticity and pore geometry. The conjugate flow-media behavior is then observed in an experimental test rig which captures the global flow behavior, the local matrix deformation, and the onset of the deviation from Darcy flow at low Re. The experimental data is presented such that the results can be used for numerical validation. Dimensionless combinations of parameters are considered in the prediction of the point of deviation from Darcy flow at low Re and confirmed from the experimental data. 1 Supported by the Marsden Fund Council from Government funding, Administered by the Royal Society of New Zealand 8:39AM G3.00004 Experimental Analysis of Entrance Effects in Low Reynolds Flow in Porous Media1 , BEN MUNRO, SID BECKER, University of Canterbury Department of Mechanical Engineering — The topic of this research concerns the experimentally observed influences of the developmental effects in a rigid porous media. A test rig has been constructed that accurately measures the pressure drop across the media and the corresponding average bulk flow velocity. The porous media has been developed using a 3D printer so that the pore geometries are uniform throughout the media. The fluid is a mixture of glycerol and water for which the viscosity is varied. Measurements of the global pressure drop versus bulk flow rate have been made over a range of Re in which the overall length of the porous media (in the direction of flow) has been varied. Because all tests have been conducted at low Re (and thus within the Darcy regime) comparisons of experimentally determined permeability between the overall media lengths provide insight into the non linear component of pressure drop that occur within the developing region. 1 Supported by the Marsden Fund Council from Government funding, Administered by the Royal Society of New Zealand 8:52AM G3.00005 Rapid capillary filling of high aspect ratio helically-supported channels in microgravity1 , MAVERICK TERRAZAS, DAVID THIESSEN, Washington State University — Arrays of capillary channels supported by helical wires may be useful as passive phase separators in microgravity. In particular, we are interested in liquid-filled channels connected by manifolds to a pressure reservoir maintained below ambient pressure to collect and transport droplets from a two-phase flow back to the low-pressure reservoir. This drop capture requires that the entire array along with the manifolds be filled with liquid by connecting them to a pressure reservoir. The priming of an array of 6-mm diameter channels in low-gravity aircraft flights has been demonstrated and shown to be stable in the presence of an exterior, transverse two-phase flow with droplet diameters ranging up to several centimeters. The priming of one or two 6-mm diameter channels with springs of large pitch has been demonstrated in drop-tower experiments as well. 1 Supported by NASA. 9:05AM G3.00006 Evaporation in dense suspension droplets1 , JIN YOUNG KIM, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University, BYUNG MOOK WEON, School of Advanced Materials Science and Engineering, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University — When a drop on a solid surface dries, a variety of drying dynamics emerge eventually. Here we show how colloidal particles affect drying dynamics in colloidal suspensions. By comparing drying dynamics of pure and colloidal fluids using confocal microscopy and mass balance, we demonstrate that the drying dynamics of colloidal fluids strongly depend on the colloid size and the initial concentration. The role of colloidal particles is complicated in the drying processes and related to the hydrodynamics for the porous medium. This work would offer clues for the dynamic nature of colloidal fluids and help to understand the drying-mediated processes such as spreading, painting, coating, and evapotranspiration. 1 This work (NRF-2013R1A22A04008115) was supported by Mid-career Researcher Program through NRF grant funded by the MEST. 9:18AM G3.00007 Drying by bubble nucleation of plant-inspired nanoscale porous media , OLIVIER VINCENT, ALEXANDRE SZENICER, JULES GUIOTH, ERIK HUBER, DAVID SESSOMS, ABRAHAM STROOCK, Cornell University — Drying from porous media is a very common phenomenon, with examples of increasing importance such as drying of soils and plants during drought, or drying of rocks subsequent to underground gas flow. Understanding and predicting drying in these examples is particularly challenging due to the large range of lengthscales that coexist in the porous medium, which can span from nanometers to meters. Inspired by the structures of the water conducting tissues that can be found in trees, we built artificial porous structures with two well separated lenghtscales: voids or channels at the micrometer scale that are interconnected by pores only a few nanometers wide. This presentation will explore the dynamics of drying in these model structures and show that drying occurs by bubble nucleation (cavitation) inside the medium rather than by the receding of liquid-vapor interfaces from the edges. We will explore the consequences of that unusual drying mode on the drying front propagation, with different regimes that can be obtained by varying the sizes and shapes of the tailored features in the nanoporous medium. 9:31AM G3.00008 Experimental study of the flow over random porous media , REZA GHEISARI, PARISA MIRBOD, Department of Mechanical and Aeronautical Engineering, Clarkson University, Potsdam, New York, 13699 — Flow over porous media has significant applications in biological systems, and industrial processes. The main focus of the majority of works in this area has been on the formulation of appropriate conditions at the interface separating the pure fluid from the porous medium flow. Furthermore, recent experimental measurements have been limited to explore the flow over superhydrophic surfaces as well as homogenous patterns. Previous studies show that the drag force due to sliding friction can be dramatically reduced if the elastic restoring force of the solid phase is small compared to the lift force generated by transiently trapped air inside the porous material. In this study, particle image velocimetry was used to observe slip velocities, shear stress, and drag reductions over a random soft porous media. Results illustrate the significant effect of these patterns on the streamlines, which can potentially affect drag force. 9:44AM G3.00009 From Red Cells to Soft Porous Lubrication , QIANHONG WU, THOMAS GACKA, Villanova University, RUNGUN NATHAN, Penn State Berks, ROBERT CRAWFORD, Villanova University, VUCBMSS TEAM — Biological scientists have wondered, since the motion of red cells was first observed in capillaries, how the highly flexible red cell can move with so little friction in tightly fitting microvessels without being damaged or undergoing hemolysis. Theoretical studies (Feng and Weinbaum, 2000, JFM; Wu et al., 2004, PRL) attributed this frictionless motion to the dramatically enhanced hydrodynamic lifting force generated inside the soft, porous, endothelial surface layer (ESL) covering the inner surfaces of our capillaries, as a red blood cell glides over it. Herein we report the first experimental examination of this concept. The results conclusively demonstrate that significant fraction of the overall lifting force generated in a soft porous layer as a planing surface glides over it, is contributed by the pore fluid pressure, and thus frictional loss is reduced significantly. Moreover, the experimental predictions showed excellent agreement with the experimental data. This finding has the potential of dramatically changing existing lubrication approaches, and can result in substantial savings in energy consumption and thus reduction in greenhouse gas emissions. 9:57AM G3.00010 Effects of polymer retention on dynamics of single phase flow , SHIMA PARSA, DAVID WEITZ, Harvard University — We study the effect of adsorption of polymer solution on dynamics of a single phase flow in a model porous medium. We use confocal microscopy to fully visualize the flow of fluid in 3D micromodel of porous media. Polymer flooding is known to be an effective method for enhanced oil recovery. However, the physical mechanism is not clearly understood. We study the effect of polymer retention on the dynamics of single phase flow using particle image velocimetery. The distribution of velocities in the medium changes greatly after flow of high concentrations of polymer through the medium. Comparing the magnitude of velocities before and after the polymer flow, we observe reduction of accessible pores to the fluid at similar injection rates. Independent measurement of the permeability of the medium confirms the decrease in the porosity. Measurements of the retention of polymer in porous media shows a weak dependence on the hydrodynamic radius of the polymer. In these experiments, the viscoelastic behavior of the polymer is isolated from velocity measurements. Monday, November 24, 2014 8:00AM - 9:44AM Session G4 Bubbles: Surfactants and Foams — 3006 - Emmanuelle Rio, Laboratoire de Physique des Solides 8:00AM G4.00001 Aging of clean foams1 , BYUNG MOOK WEON, School of Advanced Materials Science and Engineering, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University, PETER S. STEWART, School of Mathematics and Statistics, University of Glasgow — Aging is an inevitable process in living systems. Here we show how clean foams age with time through sequential coalescence events: in particular, foam aging resembles biological aging. We measure population dynamics of bubbles in clean foams through numerical simulations with a bubble network model. We demonstrate that death rates of individual bubbles increase exponentially with time, independent on initial conditions, which is consistent with the Gompertz mortality law as usually found in biological aging. This consistency suggests that clean foams as far-from-equilibrium dissipative systems are useful to explore biological aging. 1 This work (NRF-2013R1A22A04008115) was supported by Mid-career Researcher Program through NRF grant funded by the MEST. 8:13AM G4.00002 The role of surface elasticity in liquid film formation1 , LORENE CHAMPOUGNY, Univ of Paris - Sud 11 CNRS, BENOIT SCHEID, Univ. Libre de Bruxelles, FREDERIC RESTAGNO, EMMANUELLE RIO, Univ of Paris - Sud 11 CNRS, LABORATOIRE DE PHYSIQUE DES SOLIDES TEAM, TIPS - FLUID PHYSICS UNIT TEAM — The formation of thin liquid films, either free standing (soap films) or deposited on a solid substrate (coated films), is of utmost importance for many applications, ranging from the control of foam stability to surface functionalization. In this work, the behavior of thin liquid films during their generation from a surfactant solution is investigated through comparison between a hydrodynamic model including surface elasticity and experiments. “Twin” models are proposed to describe the coating of films onto a solid plate (Landau-Levich-Derjaguin configuration) as well as soap film pulling (Frankel configuration) in a single framework. Experimental data are successfully fitted using the models, surface elasticity being the only adjustable parameter. For a given surfactant solution, the analyses of soap and coated films both yield the same value for the effective surface elasticity, showing that it is an intrinsic parameter of a surfactant solution. Conversely, we demonstrate that Frankel- or Landau-Levich-like experiments can be used in practice as surface rheometers to determine the numerical value of the (effective) surface elasticity of a solution, especially for values lower than those measurable by classical devices. 1 L.C. was supported by ANR F2F. B.S. thanks the F.R.S.-FNRS for funding as well as the IAP-MicroMAST project. 8:26AM G4.00003 Rupture of vertical soap films , EMMANUELLE RIO, Laboratoire de Physique des Solides, Université Paris Sud, CNRS, Orsay — Soap films are ephemeral and fragile objects. They tend to thin under gravity, which gives rise to the fascinating variations of colors at their interfaces but leads systematically to rupture. Even a child can create, manipulate and admire soap films and bubbles. Nevertheless, the reason why it suddenly bursts remains a mystery although the soap chosen to stabilize the film as well as the humidity of the air seem very important. One difficulty to study the rupture of vertical soap films is to control the initial solution. To avoid this problem we choose to study the rupture during the generation of the film at a controlled velocity. We have built an experiment, in which we measure the maximum length of the film together with its lifetime. The generation of the film is due to the presence of a gradient of surface concentration of surfactants at the liquid/air interface. This leads to a Marangoni force directed toward the top of the film. The film is expected to burst only when its weight is not balanced anymore by this force. We will show that this leads to the surprising result that the thicker films have shorter lifetimes than the thinner ones. It is thus the ability of the interface to sustain a surface concentration gradient of surfactants which controls its stability. 8:39AM G4.00004 Wall slip of foams close to the jamming transition , S. COHEN-ADDAD1 , Univ Pierre et Marie Curie, M. LE MERRER2 , Univ Lyon, R. LESPIAT, Saint-Gobain, R. HOHLER3 , Univ Pierre et Marie Curie — Aqueous foams are dense packings of gas bubbles in a surfactant solution. They exhibit unique rheological properties [1]. When they flow along a solid smooth wall, they slip and experience viscous drag. This feature is crucial in many applications involving flow through microfluidic channels, pipes or spreading on surfaces. We focus on foams in the vicinity of the jamming transition where the bubbles are quasi spherical with small contact films at the wall and thick liquid channels between bubbles. What are the mechanisms of friction at play at the scale of the films, the channels and the bubbles that are at the origin of the macroscopic friction law? To address this question, we measure the velocity of a bubble monolayer or a wet 3D foam as it creeps along an immersed inclined plane, as a function of the inclination angle, bubble size and confinement. Two regimes of friction are evidenced: In addition to a previously reported non-linear Bretherton-like drag, we present the first direct evidence for a linear Stokes-like drag. We show that the key parameter governing the transition between the regimes is set by the Bond number for the monolayer or the confinement pressure for the foam. [1] S. Cohen-Addad, R. Hohler, O. Pitois, Annu. Rev. Fluid Mech. (2013), 45, 241. 1 Institut des NanoSciences de Paris, sylvie.cohen-addad@insp.upmc.fr Lumiere Matiere 3 Institut des NanoSciences de Paris 2 Institut 8:52AM G4.00005 A numerical model to simulate foams during devolatilization of polymers , IRFAN KHAN, RAVINDRA DIXIT, Dow Chemical Co — Customers often demand that the polymers sold in the market have low levels of volatile organic compounds (VOC). Some of the processes for making polymers involve the removal of volatiles to the levels of parts per million (devolatilization). During this step the volatiles are phase separated out of the polymer through a combination of heating and applying lower pressure, creating foam with the pure polymer in liquid phase and the volatiles in the gas phase. The efficiency of the devolatilization process depends on predicting the onset of solvent phase change in the polymer and volatiles mixture accurately based on the processing conditions. However due to the complex relationship between the polymer properties and the processing conditions this is not trivial. In this work, a bubble scale model is coupled with a bulk scale transport model to simulate the processing conditions of polymer devolatilization. The bubble scale model simulates the nucleation and bubble growth based on the classical nucleation theory and the popular “influence volume approach.” As such it provides the information of bubble size distribution and number density inside the polymer at any given time and position. This information is used to predict the bulk properties of the polymer and its behavior under the applied processing conditions. Initial results of this modeling approach will be presented. 9:05AM G4.00006 Towards Modeling Local Foam Drainage Using the Arbitrary Lagrangian Eulerian Method , ANDREW BRANDON, RAMAGOPAL ANANTH, Naval Research Laboratory — Liquid drainage in foams is a multi-scale, multi- dimensional phenomena that is tied directly to how well a foam performs. For example, the amount of metal within a metal foam after it solidifies affects the strength of the foam and the amount of liquid within an aqueous fire fighting foam determines how effective it is at extinguishing a fire. Liquid drainage is driven by gravity and is governed by the liquid’s density and viscosity as well as the surface tension at the liquid gas interface. There are numerous one dimensional, single phase models that approximate liquid drainage by employing a global description but there are no multidimensional models that use a local description. In this presentation, I will describe an ongoing effort to develop a two dimensional, multiphase, Arbitrary Lagrangian Eulerian model for the study of local liquid drainage in foams. I will present an improved algorithm for the solution of the incompressible fluid equations in the Arbitrary Lagrangian Eulerian method, the novel method used for moving the domain in time, and results from this model development effort. 9:18AM G4.00007 Blast wave mitigation by liquid foam1 , MARTIN MONLOUBOU, BENJAMIN DOLLET, ARNAUD SAINT-JALMES, ISABELLE CANTAT, Institut de Physique de Rennes, SOFT MATTER TEAM — Due to their high apparent viscosity, liquid foams are good systems to absorb energy. This property is for instance used in the military domain to mitigate blast waves or explosions [Britan, 2009; Del Prete, 2013]. However, the underlying dissipation mechanisms are still not well understood. We address this issue by resolving in space and time a shock wave impacting a foam sample. We use a shock tube to send a shock wave on a foam with controlled liquid fraction, bubble size and physico-chemistry. The impacting shock creates an expanding cavity in the foam and propagates through the whole sample. The dynamics is recorded with a high speed camera and pressure signals are simultaneously measured. We show the influence of the bubble size and of the shock amplitude on the velocity and on the attenuation of the pressure signal, and on the foam destruction rate. [1] Britan et al., Colloids and Surfaces A, 344:48-55, 2009. [2] Del Prete et al., Shock Waves, 23:39-53, 2013. 1 This work is supported by the DGA. 9:31AM G4.00008 Pinch-off Scaling Law of Soap Bubbles , JOHN DAVIDSON, SANGJIN RYU, University of NebraskaLincoln — Three common interfacial phenomena that occur daily are liquid drops in gas, gas bubbles in liquid and thin-film bubbles. One aspect that has been studied for these phenomena is the formation or pinch-off of the drop/bubble from the liquid/gas threads. In contrast to the formation of liquid drops in gas and gas bubbles in liquid, thin-film bubble pinch-off has not been well documented. Having thin-film interfaces may alter the pinch-off process due to the limiting factor of the film thickness. We observed the pinch-off of one common thin-film bubble, soap bubbles, in order to characterize its pinch-off behavior. We achieved this by constructing an experimental model replicating the process of a human producing soap bubbles. Using high-speed videography and image processing, we determined that the minimal neck radius scaled with the time left till pinch-off, and that the scaling law exponent was 2/3, similar to that of liquid drops in gas. Monday, November 24, 2014 8:00AM - 10:10AM Session G5 Biofluids: Flapping Wings — 3008 - David Rival, University of Calgary 8:00AM G5.00001 The Hovering Hat , YANGYANG HUANG, University of Southern California, MONIKA NITSCHE, University of New Mexico, EVA KANSO, University of Southern California — Birds and insects often flap their wings to hover and fly. Interestingly, recent experiments have shown that non-flapping rigid objects are also capable of stably hovering in an oscillatory background flow, given they possess particular geometric asymmetry in the vertical direction. The up-down asymmetry creates downwash vortex shedding and thus generates lift against gravity. Here, we use a two-dimensional vortex sheet model to study the motion of an inverted V-shaped object, moving passively in an oscillatory flow. We found that, depending on the fluid oscillation frequency, the hat descends, hovers or ascends. We also found an optimal opening angle of the hat for hovering that requires minimal energy from the background flow. We conclude by showing the passive stability of the hat and the role of the hydrodynamic forces and moments in preventing it from tipping over. 8:13AM G5.00002 Thrust generated by a flapping flexible plate , FLORINE PARAZ, CHRISTOPHE ELOY, LIONEL SCHOUVEILER, Aix Marseille Université, CNRS, Centrale Marseille, IRPHE, Marseille, France — In order to a gain better insight into the physics of swimming with a flexible caudal fin, we have performed experiments with a rectangular elastic plate immersed in a water flow. The plate leading edge is forced into harmonic motion, while its trailing edge responds passively to this actuation. A resonance has been evidenced experimentally, pointing out a strong coupling between the natural frequencies of the structure and the forcing frequencies. In this experiment, the forcing amplitude plays a non-trivial role, emphasizing the role of non-linearities in this problem. To better understand the origin of these non-linearities, a weakly non-linear model has been developed. We assumed a quasi two-dimensional plate of zero thickness immersed in a potential flow and subject to a resistive drag-like force. The plate deflection has then been decomposed into a forcing heaving mode and natural flexural modes. This modeling approach allowed us to predict the response to the heave forcing as a function of its amplitude and frequency. The frequencies of the resonances, as well as the deflection enveloppes, are well captured by this model. The performance of the system, measured through the generated thrust, is also well predicted by this model. 8:26AM G5.00003 Effect of pitching history on the flow topology for freely pitching wings , SWATHI KRISHNA, KAREN MULLENERS, Leibniz Universität Hannover — Insect flapping flight represents an interesting aerodynamic problem because of the characteristic unsteadiness and the low Reynolds number of the airflow. The time dependent wing kinematics play a prominent role in the generation and evolution of the leading edge vortex (LEV). The changes in unsteady flow attributes due to differences in the time dependent wing kinematics pose an interesting case for study. Variations in the temporal evolution of the wing’s angle of attack lead to changes in the size, position and strength of the LEV. Time-resolved planar particle image velocimetry was conducted on a freely pitching wing to gain insight into how the pitching history affects the vortex dynamics. To quantify the observed trends, a topological analysis of the instantaneous flow around the pitching wing is conducted. The isolated singular points (nodes, saddles, and foci) of a instantaneous vector field are analysed. Based on the type and distribution of the critical points, a better understanding of the emergence and spatio-temporal development of the prominent vortical structures is obtained. Additionally, proper orthogonal decomposition is carried out to study the influence of temporal changes in pitch on the dynamic behaviour of the vortical structures. 8:39AM G5.00004 Model-Based Optimization for Flapping Foil Actuation , JACOB IZRAELEVITZ, MICHAEL TRIANTAFYLLOU, Massachusetts Institute of Technology — Flapping foil actuation in nature, such as wings and flippers, often consist of highly complex joint kinematics which present an impossibly large parameter space for designing bioinspired mechanisms. Designers therefore often build a simplified model to limit the parameter space so an optimum motion trajectory can be experimentally found, or attempt to replicate exactly the joint geometry and kinematics of a suitable organism whose behavior is assumed to be optimal. We present a compromise: using a simple local fluids model to guide the design of optimized trajectories through a succession of experimental trials, even when the parameter space is too large to effectively search. As an example, we illustrate an optimization routine capable of designing asymmetric flapping trajectories for a large aspect-ratio pitching and heaving foil, with the added degree of freedom of allowing the foil to move parallel to flow. We then present PIV flow visualizations of the optimized trajectories. 8:52AM G5.00005 Flapping propulsion with tip pitch control1 , FRANCISCO HUERA-HUARTE, Universitat Rovira i Virgili & California Institute of Technology (Visiting Associate), MORTEZA GHARIB, California Institute of Technology — The effect of flexibility in the propulsion performance and efficiency of oscillating pitching foils has received a large amount of attention in the past years. Scientists have used simplified robotic models that mimic the kinematics of flying and swimming animals, in order to get inspiration to build more efficient engineering systems. Compliance is one of the aspects that has received more attention, as it seems to be a common feature in nature’s flyers and swimmers. Active or passive control elements are also common in nature. We will show how thrust generation in a pitching fin, can be greatly affected by controlling the tip pitch motion dynamically and independently of the fin itself. This is in fact a controlled local change of curvature of the end of the fin. A robotic system has been designed in a way that not only flapping amplitudes and frequencies can be controlled, but also the amplitudes and frequencies of the tip and the phase difference between the tip and the fin. We measured thrust forces and the vortex dynamics in the near wake of the system, by using planar DPIV (Digital Particle Image Velocimetry) in a wide variety of flapping situations with tip control. 1 Funding from Spanish Ministry of Science through grant DPI2012-37904 is gratefully acknowledged. 9:05AM G5.00006 Critical Point Matching and Distance Metrics of Unsteady Flow Separation from a Pitching Plate1 , FAEGHEH HOOMAN, PAUL KRUEGER, SMU — Unsteady flow separation is of interest for force and moment generation by flapping airfoils, but it is often difficult to determine how small differences in the motion lead to differences in the flow field and resulting forces. To better understand the flow evolution during unsteady separation in pitching maneuvers, analysis was performed of two numerical data sets for the pitch-up of a two-dimensional flat plate in a free stream flow with Re=1000 (data provided by Prof. J.D. Eldredge at UCLA). Flow fields were compared by finding the best match of first order critical points according to weighted physical location and topological characteristics. Weighting and smoothing helped eliminate outliers, especially after adding noise, and made the method robust. A total distant metric for matched critical points was defined to provide a global metric for identifying similarities and differences between flow fields. Comparisons of the flow evolution for the two data sets using the distance metric will be presented. 1 This material is based upon work supported by the National Science Foundation under Grant No. 1115139 9:18AM G5.00007 Effect of flexibility on the performance of an impulsively started foil , MEHDI SAADAT, HOSSEIN HAJ-HARIRI, University of Virginia — A computational study is conducted to investigate the effects of flexibility on the unsteady forces of an impulsively started foil. The flexibility is implemented into the model by allowing the rigid foil to rotate (pitch) about its leading edge. First, for a given initial angle of attack, the pitching motion of the foil is prescribed by imposing various constant angular velocities. It is shown that thrust window (i.e. the duration of which thrust is generated as defined by Wang 2000) and the value of maximum thrust is maximized at an optimum angular velocity. Next, the foil is allowed to rotate freely around its leading edge. The force balance of fluid added mass with the foil inertia predicts the solid-to-fluid density ratio required for achieving the optimum angular velocity, and maximizing the thrust widow and maximum thrust. Computations further confirm the optimum density ratio derived by the scaling analysis. We hypothesize that a flexible fin achieves its best performance when the fin is limp, and its solid-to-fluid density ratio is set to the optimum value. We provide comparison with data from biology. Wang, Z.J., “Vortex shedding and frequency selection in flapping flight,” J. Fluid Mech., 410, 323-341, 2000. 9:31AM G5.00008 Experimental study of the fluid and structure interaction for gravity driven falling plates , RUIJUN TIAN, FANGJUN SHU, New Mexico State University — Falling motion of thin plates and the induced flow field were investigated in this study. Time-resolved 2-D PIV measurements were conducted to investigate the dynamic development of the flow field induced by falling plates submerged in water. Two types of falling motions were observed, fluttering (sliding from side to side while descending) and tumbling (continuously rotating while falling downward and sliding to one direction), depending on the plate material and the physical dimensions, which forms a few governing non-dimensional parameters. The time-resolved PIV images, which also contain the plate location information, were further processed to extract the location and orientation of the plate. The data were then numerically differentiated to acquire the plate’s translational and angular speeds and accelerations. Thus, the instantaneous hydrodynamic force/moment on the plates and the surrounding flow field were correlated to perform empirical analysis on this classical unsteady fluid and structure interaction (FSI) problem. It is discovered that the leading edge vortex plays an important role since its development is drastically related to the dynamic features of falling plates, though it is still unclear if the vortex shedding causes or results from the plates’ movement. A theoretical model is proposed to simulate the dynamic features of the falling plates, which will be compared with the experimental data. 9:44AM G5.00009 Combined turning and propulsion of a flexible plate in viscous fluid , ALEXANDER ALEXEEV, PETER YEH, Georgia Institute of Technology — We use three dimensional computer simulations to study the flow and structural deformation of an oscillating elastic rectangular plate submerged in a viscous fluid. The elastic plate is actuated at the root near the first natural frequency and undergoes a combined sinusoidal plunging and twisting motion. This complex motion results in not only a forward propulsive force, but also a force perpendicular to the swimming direction. The latter force leads to turning. We find that the strength of the turning force depends on oscillation amplitudes as well as the phase difference between the plunging and twisting oscillations. Our simulations reveal an optimal phase difference and twisting amplitude that leads to maximum turning potential. These results can be used to design a basic mechanism for changing direction in a micro underwater autonomous vehicle actuated using flexible fins. 9:57AM G5.00010 The influence of circulation distribution on LEV development for an impulsively-started, spanwise-flexible profile , JAIME WONG, DAVID RIVAL, Queen’s University — As a spanwise-flexible profile is accelerated from rest, the profile bends thereby causing a component of the free-stream flow to align with the spanwise direction. For the rapid accelerations typical in biological swimming and flying, accelerating a profile from rest will simultaneously result in the formation of a leading-edge vortex (LEV). The spanwise flow resulting from profile bending rearranges the distribution of circulation in the LEV along the span of the wing via vorticity convection, which does not occur in an otherwise equivalent rigid case. The effect of this circulation redistribution on LEV detachment and force history is difficult to separate from other flexibility effects, such as the varying shear-layer feeding rate and local acceleration. Therefore, the current study utilizes cyber-physical fluid dynamics (CPFD) to simulate an impulsively-started spanwise-flexible profile in the absence of spanwise flow. Nominally two-dimensional CPFD results are combined in a blade-element scheme that replicates the distributed load on a flexible profile. In this way, the effect of spanwise flow on LEV development and detachment and the resulting force histories can be isolated from other flexibility effects. Monday, November 24, 2014 8:00AM - 10:10AM Session G6 Biofluids: Active Fluids III — 3010 - Eva Kanso, University of Southern California 8:00AM G6.00001 Copepod Behavior in “Cryptic Blooms” of Toxic Algae , A.C. TRUE, D.R. WEBSTER, M.J. WEISSBURG, J. YEN, Georgia Tech — Copepods, Acartia tonsa and Temora longicornis, were exposed to thin layers of exudates from the toxic dinoflagellate Karenia brevis (1 - 10,000 cells/mL) (i.e. models of “cryptic blooms” of toxic phytoplankton). Planar laser-induced fluorescence (PLIF) was used to quantify the spatiotemporal structure of the layer allowing for correlation of behavioral responses with toxin levels. Both species explicitly avoided the exudate layer and the vicinity of the layer. Measures of path kinematics (swimming speed, turn frequency) by location (in-layer vs. out-of-layer) and exposure (pre-contact vs. post-contact) revealed some similarities, but also significant differences, in trends for each species. A. tonsa significantly increases swimming speed and swimming speed variability in the exudate layer and post-contact, whereas T. longicornis slightly increases both in-layer and slightly reduces both post-contact. Both species increase turn frequency in-layer and post-contact with increasing K. brevis exudate concentration. Path fracticality indicates that A. tonsa trajectories became more diffuse/sinuous and T. longicornis trajectories became more linear/ballistic (trending effects). Regression analyses revealed that the rate of change of behavior with increasing exudate concentration for A. tonsa was thrice to fifty times that of T. longicornis. Toxic K. brevis can essentially eliminate top-down grazer control,another sinister means by which it gains a competitive advantage over the local phytoplankton taxa. 8:13AM G6.00002 Collective fluid mechanics of honeybee nest ventilation , NICK GRAVISH, SEAS & OEB, Harvard University, STACEY COMBES, OEB, Harvard University, ROBERT J. WOOD, SEAS, Harvard University, JACOB PETERS, OEB, Harvard University — Honeybees thermoregulate their brood in the warm summer months by collectively fanning their wings and creating air flow through the nest. During nest ventilation workers flap their wings in close proximity in which wings continuously operate in unsteady oncoming flows (i.e. the wake of neighboring worker bees) and near the ground. The fluid mechanics of this collective aerodynamic phenomena are unstudied and may play an important role in the physiology of colony life. We have performed field and laboratory observations of the nest ventilation wing kinematics and air flow generated by individuals and groups of honeybee workers. Inspired from these field observations we describe here a robotic model system to study collective flapping wing aerodynamics. We microfabricate arrays of 1.4 cm long flapping wings and observe the air flow generated by arrays of two or more fanning robotic wings. We vary phase, frequency, and separation distance among wings and find that net output flow is enhanced when wings operate at the appropriate phase-distance relationship to catch shed vortices from neighboring wings. These results suggest that by varying position within the fanning array honeybee workers may benefit from collective aerodynamic interactions during nest ventilation. 8:26AM G6.00003 Collective interaction of microscale matters in natural analogy: human cancer cells vs. microspheres , SUNGSOOK AHN, SANG JOON LEE, Pohang University of Science and Technology (POSTECH), POSTECH TEAM — Collective behaviors have been considered both in living and lifeless things as a natural phenomenon. During the ordering process, a sudden and spontaneous transition is typically generated between an order and a disorder according to the population density of interacting elements. In a cellular level collective behavior, the cells are distributed in the characteristic patterns according to the population density and the mutual interaction of the individual cells undergo density-dependent diffusive motion. On the other hand, density-controlled surface-modified hollow microsphere suspension induces an overpopulation via buoyancy which provides a driving force to induce an assembly. The collective behaviors of the cells and microspheres in a designed liquid medium are explained in terms of the deviation from the interparticle distance distribution and the induced strength to organize the particle position in a specific distance range. as a result, microscale particulate matters exhibit high resemblance in their pair correlation and dynamical heterogeneity in the intermediate range between a single individual and an agglomerate. Therefore, it is suggested that biological systems are analogically explained to be dominated by physically interactive aspects. 8:39AM G6.00004 Responding to flow: How phytoplankton adapt migration strategies to tackle turbulence , ANUPAM SENGUPTA, FRANCESCO CARRARA, ROMAN STOCKER, Massachusetts Institute of Technology — Phytoplankton are among the ocean’s most important organisms and it has long been recognized that turbulence is a primary determinant of their fitness. Yet, mechanisms by which phytoplankton may adapt to turbulence have remained unknown. We present experiments that demonstrate how phytoplankton are capable of rapid adaptive behavior in response to fluid flow disturbances that mimic turbulence. Our study organism was the toxic marine alga Heterosigma akashiwo, known to exhibit “negative gravitaxis,” i.e., to frequently migrate upwards against gravity. To mimic the effect of Kolmogorov-scale turbulent eddies, which expose cells to repeated reorientations, we observed H. akashiwo in a “flip chamber,” whose orientation was periodically flipped. Tracking of single cells revealed a striking, robust behavioral adaptation, whereby within tens of minutes half of the population reversed its direction of migration to swim downwards, demonstrating an active response to fluid flow. Using confocal microscopy, we provide a physiological rationalization of this behavior in terms of the redistribution of internal organelles, and speculate on the motives for this bet-hedging-type strategy. This work suggests that the effects of fluid flow – not just passive but also active – on plankton represents a rich area of investigation with considerable implications for some of earth’s most important organisms. 8:52AM G6.00005 A hybrid numerical-experimental study of fluid transport by migrating zooplankton aggregations , MONICA MARTINEZ, California Institute of Technology, JOHN DABIRI, Graduate Aeronautical Laboratories and Bioengineering, California Institute of Technology, JANNA NAWROTH, Harvard University, BRAD GEMMELL, Marine Biological Laboratory, SAMANTHA COLLINS, Polytechnic School — Zooplankton aggregations that undergo diel vertical migrations have been hypothesized to play an important role in local nutrient transport and global ocean dynamics. The degree of the contributions of these naturally occurring events ultimately relies on how efficiently fluid is transported and eventually mixed within the water column. By implementing solutions to the Stokes equations, numerical models have successfully captured the time-averaged far-field flow of self-propelled swimmers. However, discrepancies between numerical fluid transport estimates and field measurements of individual jellyfish suggest the need to include near-field effects to assess the impact of biomixing in oceanic processes. Here, we bypass the inherent difficulty of modeling the unsteady flow of active swimmers while including near-field effects by integrating experimental velocity data of zooplankton into our numerical model. Fluid transport is investigated by tracking a sheet of artificial fluid particles during vertical motion of zooplankton. Collective effects are addressed by studying different swimmer configurations within an aggregation from the gathered data for a single swimmer. Moreover, the dependence of animal swimming mode is estimated by using data for different species of zooplankton. 9:05AM G6.00006 Population dynamics in non-homogeneous environments , KIM M.J. ALARDS, FRANCESCA TESSER, FEDERICO TOSCHI, Eindhoven University of Technology — For organisms living in aquatic ecosystems the presence of fluid transport can have a strong influence on the dynamics of populations and on evolution of species. In particular, displacements due to self-propulsion, summed up with turbulent dispersion at larger scales, strongly influence the local densities and thus population and genetic dynamics. Real marine environments are furthermore characterized by a high degree of non-homogeneities. In the case of population fronts propagating in “fast” turbulence, with respect to the population duplication time, the flow effect can be studied by replacing the microscopic diffusivity with an effective turbulent diffusivity. In the opposite case of “slow” turbulence the advection by the flow has to be considered locally. Here we employ numerical simulations to study the influence of non-homogeneities in the diffusion coefficient of reacting individuals of different species expanding in a 2 dimensional space. Moreover, to explore the influence of advection, we consider a population expanding in the presence of simple velocity fields like cellular flows. The output is analyzed in terms of front roughness, front shape, propagation speed and, concerning the genetics, by means of heterozygosity and local and global extinction probabilities. 9:18AM G6.00007 Confining collective motion , DENIS BARTOLO, ENS Lyon, ANTOINE BRICARD, ESPCI and ENS Lyon, JEAN-BAPTISTE CAUSSIN, CHARLES SAVOIE, ENS Lyon, DEBASISH DAS, UCSD, OLESKAR CHEPIZHKO, FERNANDO PERUANI, Université de Nice, DAVID SAINTILLAN, UCSD — It is well established that geometrical confinement have a significant impact on the structure and the flow properties of complex fluids. Prominent examples include the formation of topological defects in liquid crystals, and the flow instabilities of viscoelastic fluids in curved geometries. In striking contrast very little is known about the macroscopic behavior of confined active fluids. In this talk we show how to motorize plastic colloidal beads and turn them into self-propelled particles. Using microfluidic geometries we demonstrate how confinement impacts their collective motion. Combining quantitative experiments, analytical theory and numerical simulations we show how a population of motile bodies interacting via alignement and repulsive interactions self-organizes into a single heterogeneous macroscopic vortex that lives on the verge of a phase separation. 9:31AM G6.00008 Density shocks in confined microswimmers , ALAN CHENG HOU TSANG, EVA KANSO, University of Southern California, BIODYNAMICS TEAM — Motile microorganisms are often subject to different types of boundary confinement in their natural environment, but the effects of confinement on their dynamics are poorly understood. We consider an idealized model of confined microswimmers restricted to move in a two-dimensional Hele-Shaw cell. We then impose two different types of boundary confinement: circular and sidewalls confinement. We study how boundaries trigger the emergence of global modes. In the case of circular confinement, the microswimmers can spontaneously organize themselves into a single vortex state when the radius of the circular boundary is below a certain critical value, reminiscent to what have been observed in recent experiments of bacterial suspensions. In the case of sidewalls confinement in a rectangular channel, the microswimmers form density shock, via interaction with the sidewalls and background flow. We show that, through controlling the strength of background flow, we can manipulate the density shock to form at the back or front of the swimmer clusters or the suppression of the shock which gives rise to a uniform traveling wave of swimmers. 9:44AM G6.00009 Rotating bacteria aggregate into active crystals , ALEXANDER PETROFF, The Rockefeller University, XIAO-LUN WU, University of Pittsburgh, ALBERT LIBCHABER, The Rockefeller University — The dynamics of many microbial ecosystems are determined not only by the response of individual bacteria to their chemical and physical environments but also the dynamics that emerge from interactions between cells. Here we investigate the collective dynamics displayed by communities of Thiovulum majus, one of the fastest known bacteria. We observe that when these bacteria swim close to a microscope cover slip, the cells spontaneously aggregate into a visually-striking two-dimensional hexagonal lattice of rotating cells. Each cell in an aggregate rotates its flagella, exerting a force that pushes the cell into the cover slip and a torque that causes the cell to rotate. As cells rotate against their neighbors, they exert forces and torques on the aggregate that cause the crystal to move and cells to hop to new positions in the lattice. We show how these dynamics arises from hydrodynamic and surface forces between cells. We derive the equations of motion for an aggregate, show that this model reproduces many aspects of the observed dynamics, and discuss the stability of these and similar active crystals. Finally, we discuss the ecological significance of this behavior to understand how the ability to aggregate into these communities may have evolved. 9:57AM G6.00010 Bacterial populations growth under co- and counter-flow condition , FRANCESCA TESSER, JOS C.H. ZEEGERS, HERMAN J.H. CLERCX, FEDERICO TOSCHI, Eindhoven University of Technology — For organisms living in a liquid ecosystem, flow and flow gradients play a major role on the population level: the flow has a dual role as it transports the nutrient while dispersing the individuals. In absence of flow and under homogeneous conditions, the growth of a population towards an empty region is usually described by a reaction diffusion equation. The solution predicts the expansion as a wave front (Fisher wave) proceeding at constant speed, till the carrying capacity is reached everywhere. The effect of fluid flow, however, is not well understood and the interplay between transport of individuals and nutrient opens a wide scenario of possible behaviors. In this work, we experimentally observe non-motile E. coli bacteria spreading inside rectangular channels in a PDMS microfluidic device. By use of a fluorescent microscope we analyze the dynamics of the population density subjected to different co- and counter-flow conditions and shear rates. Monday, November 24, 2014 8:00AM - 10:10AM Session G7 Biofluids: Aneurysms — 3012 - Vitaliy Rayz, University of California, Berkeley 8:00AM G7.00001 Vascular growth and remodeling coupled with fluid simulation in patient specific geometry , JIACHENG WU, SHAWN C. SHADDEN, University of California, Berkeley — In this talk, we propose a computational framework to couple vascular growth and remodeling (G&R) with fluid simulation in 3D patient specific geometry. Hyperelastic and anisotropic properties are considered for the vessel wall material. A constrained mixture model is used to represent multiple constituents in the vessel wall. The coupled simulation is divided into two time scales, the longer time scale for G&R and the shorter time scale for fluid dynamics simulation. G&R is simulated to determine the boundary of the fluid domain, the fluid simulation in turn generates wall shear stress and transmural pressure data that regulates G&R. To minimize required computation cost, fluid is only simulated when G&R causes significant vascular geometric change. This coupled model can be used to study the influence of the stress-mediated law parameters on the stability of the vascular tissue growth, and predict progression of vascular diseases such as aneurysm expansion. 8:13AM G7.00002 A Parallel Monolithic Approach for Fluid-Structure Interaction in a Cerebral Aneurysm1 , MEHMET SAHIN, ALI EKEN, Istanbul Technical University — A parallel fully-coupled approach has been developed for the fluid-structure interaction problem in a cerebral artery with aneurysm. An Arbitrary Lagrangian-Eulerian formulation based on the side-centered unstructured finite volume method is employed for the governing incompressible Navier-Stokes equations and the classical Galerkin finite element formulation is used to discretize the constitutive law for the Saint Venant-Kirchhoff material in a Lagrangian frame for the solid domain. The time integration method for the structure domain is based on the energy conserving mid-point method while the second-order backward difference is used within the fluid domain. The resulting large-scale algebraic linear equations are solved using a one-level restricted additive Schwarz preconditioner with a block-incomplete factorization within each partitioned sub-domains. The parallel implementation of the present fully coupled unstructured fluid-structure solver is based on the PETSc library. The proposed numerical algorithm is initially validated for several classical benchmark problems and then applied to a more complicated problem involving unsteady pulsatile blood flow in a cerebral artery with aneurysm as a realistic fluid-structure interaction problem encountered in biomechanics. 1 The authors acknowledge financial support from Turkish National Scientific and Technical Research Council through project number 112M107. 8:26AM G7.00003 Numerical predictions of hemodynamics following surgeries in cerebral aneurysms1 , VITALIY RAYZ, MICHAEL LAWTON, LOIC BOUSSEL, JOSEPH LEACH, GABRIEL ACEVEDO, VAN HALBACH, DAVID SALONER, UC San Francisco — Large cerebral aneurysms present a danger of rupture or brain compression. In some cases, clinicians may attempt to change the pathological hemodynamics in order to inhibit disease progression. This can be achieved by changing the vascular geometry with an open surgery or by deploying a stent-like flow diverter device. Patient-specific CFD models can help evaluate treatment options by predicting flow regions that are likely to become occupied by thrombus (clot) following the procedure. In this study, alternative flow scenarios were modeled for several patients who underwent surgical treatment. Patient-specific geometries and flow boundary conditions were obtained from magnetic resonance angiography and velocimetry data. The Navier-Stokes equations were solved with a finite volume solver Fluent. A porous media approach was used to model flow-diverter devices. The advection-diffusion equation was solved in order to simulate contrast agent transport and the results were used to evaluate flow residence time changes. Thrombus layering was predicted in regions characterized by reduced velocities and shear stresses as well as increased flow residence time. The simulations indicated surgical options that could result in occlusion of vital arteries with thrombus. Numerical results were compared to experimental and clinical MRI data. The results demonstrate that image-based CFD models may help improve the outcome of surgeries in cerebral aneurysms. 1 acknowledge R01HL115267 8:39AM G7.00004 Influence of transport on thrombogenic potential in cardiovascular flows1 , KIRK B. HANSEN, SHAWN C. SHADDEN, Univ of California - Berkeley — Intraluminal thrombus is a common complication inside aneurysms, as well as inside hearts with pumping or rhythmic deficiencies. The mechanisms of thrombus formation in these scenarios remain unclear. As opposed to stenotic thrombosis, where shear-induced platelet activation likely plays a major role, the shear stresses in aneurysmal flows are typically lower than in the surrounding vasculature. This suggests an alternative mechanism more akin to the stasis-driven coagulation typically seen on the venous side of the cardiovascular system. In this work, we investigate how transport properties of the complex flow features inherent to aneurysmal (or intracardiac) flows may affect the potential for thrombus formation. Patient-specific three-dimensional velocity fields are obtained using image-based computational fluid dynamics. These velocity fields are then coupled to a computational thrombosis model based on a system of reaction-advection-diffusion equations. This continuum-based model accounts for the transport and activation of platelets, as well as the transport and reaction of other chemical species involved in the coagulation cascade. Near-wall values of thrombin concentration are used as a metric for localized thrombogenic potential. 1 This work was supported by NSF grant 1354541 and NIH grant HL108272. 8:52AM G7.00005 Hemodynamic Changes in Treated Cerebral Aneurysms and Correlations with Long-Term Outcomes1 , PATRICK MCGAH, MICHAEL BARBOUR, MICHAEL LEVITT, LOUIS KIM, ALBERTO ALISEDA, University of Washington — The hemodynamic conditions in patients with cerebral aneurysms undergoing treatment, e.g. flow diverting stents or coil embolization, are investigated via computational simulations. Patient-specific 3D models of the vasculature are derived from rotational angiography. Patient-specific flow and pressure boundary conditions are prescribed utilizing intravascular pressure and velocity measurements. Pre-treatment and immediate post-treatment hemodynamics are studied in eight cases so as to ascertain the effect of the treatment on the intra-aneurysmal flow and wall shear stress. We hypothesize that larger reductions in intra-aneurysmal inflow and wall shear stress after treatment are correlated with an increased likelihood of aneurysmal occlusion and treatment success. Results indicate reductions of the intra-aneurysmal inflow and wall shear stress in all cases. Preliminary clinical six-month follow-up data, assessing if the treatment has been successful, shows that the cases with a persistent aneurysm had a smaller reduction in inflow and wall shear stress magnitude in the immediate post-treatment conditions. This suggests that CFD can be used to quantify a treatment’s probability of success by computing the change in pre- and- post-treatment hemodynamics in cerebral aneurysms. 1 NIH-NINDS 9:05AM G7.00006 Experimental Study of Fluid Flow in an Aneurysm for Varying Shape Indices1 , PAULO YU, CYRUS CHOI, VIBHAV DURGESH, Cal State Univ - Northridge — An aneurysm is an abnormal bulging of a blood vessel wall. A ruptured aneurysm can be severely debilitating or fatal. There is a lack of understanding of fluid flow parameters leading to aneurysm growth and rupture. Clinical studies have shown that certain aneurysm shape indices are strongly correlated to rupture. The overall goal of this study is to comprehensively characterize fluid dynamics parameters inside an aneurysm sac, for varying shape indices. As part of this work, two different idealized aneurysm glass models are used, and an in-house flow loop system has been developed to simulate constant and physiological pressure gradients. Index of refraction matching techniques have been used for accurate estimation of fluid flow parameters. Laser Doppler Velocimetry measurements are conducted for Reynolds number values from 10-200 to understand impact of inflow conditions on flow structures and parameters inside aneurysm sac. Particle Image Velocimetry measurements are performed on several horizontal and vertical planes inside aneurysm sac and show the presence of secondary fluid structures inside the sac, not observed in mid-plane measurements from earlier studies. The results show dependence of flow parameters/structures on aneurysm shape and inflow conditions. 1 This work is supported through the California State University Northridge Research Fellowship Program 2013-14 9:18AM G7.00007 FSI and CFD Modeling of Cerebral Aneurysm Model and Comparing to PIV Experiments , ZHAOPENG WANG, Indiana University Bloomington, QING HAO, Indiana University Purdue University Fort Wayne — Wall shear stress or strain is considered as an important factor for cerebral aneurysm growth and even rupture. The objective of present study is to evaluate wall shear stress in aneurysm sac and neck by Fluid Structure Interaction (FSI) and solid wall Computational Fluid Dynamics (CFD) approaches and compare the simulation results against Particle Image Velocimetry (PIV) experimental data from an elastic in vitro aneurysm model. The FSI and CFD simulation results showed that both approaches captured the flow patterns inside the aneurysm sac under pulsatile flow, that in diastole time period the flow inside the aneurysm sac was a stable circular clock-wise flow; when higher velocity entered into the aneurysm sac during systole and in a short diastole time period an anti-clock circular flow pattern emerged near the distal neck. Both approaches showed that the shear stress near the proximal neck experienced higher shear stress than the distal neck, while in the aneurysm dome the shear stress was always the lowest. In this study, we also showed that shear stress values at proximal neck and distal neck from FSI approach were lower than solid wall CFD approach. 9:31AM G7.00008 Experimental Study of a Thoracic Aortic Aneurysm Prior to and After Surgical Repair Hemodynamics , ANNA-ELODIE KERLO, Purdue University, STEVEN FRANKEL, Technion - Israel Institute of Technology, JUN CHEN, PAVLOS VLACHOS, Purdue University — Once a Thoracic Aortic Aneurysm (TAA) is detected, the risk of rupture is estimated based on the TAA diameter compared to the normal aortic diameter and its expansion rate. However, there are no reliable predictors that can provide accurate prognosis, and each aneurysm may progress differently. This work aims to assess the hemodynamic characteristics and flow structures associated with TAAs. The flow in a patient specific thoracic aortic aneurysm is compared to the same patient after treatment, in order to quantify the differences in the hydrodynamic forces acting on the aneurysm. Flow visualization with dye and Particle Image Velocimetry (PIV) are used to study flow features within both geometries. Local flow patterns are visualized to predict potential areas of recirculation and low shear stresses as they are associated with thrombogenicity. Understanding the differences in flow features between a thoracic aortic aneurysm and a normal aorta (or a TAA after surgical repair) may lead to a better understanding of disease mechanisms that will enable clinicians to better estimate the risk of rupture. 9:44AM G7.00009 Transluminal Attenuation Gradient for Thrombotic Risk Assessment in Kawasaki Disease Patients with Coronary Artery Aneurysms , NOELIA GRANDE GUTIERREZ, ANDREW KAHN, JANE BURNS, ALISON MARSDEN, UC San Diego — Kawasaki Disease (KD) can result in coronary aneurysms in up to 25% of patients if not treated early putting patients at risk of thrombus formation, myocardial infarction and sudden death. Clinical guidelines for administering anti-coagulation therapy currently rely on anatomy alone. Previous studies including patient specific modeling and computer simulations in KD patients have suggested that hemodynamic data can predict regions susceptible to thrombus formation. In particular, high Particle Residence Time gradient (PRTg) regions have shown to correlate with regions of thrombus formation. Transluminal Attenuation Gradient (TAG) is determined from the change in radiological attenuation per vessel length. TAG has been used for characterizing coronary artery stenoses, however this approach has not yet been used in aneurysmal vessels. The aim of this study is to analyze the correlation between TAG and PRTg in KD patients with aneurysms and evaluate the use of TAG as an index to quantify thrombotic risk. Patient specific anatomic models for fluids simulations were constructed from CT angiographic image data from 3 KD aneurysm patients and one normal control. TAG values for the aneurysm patients were markedly lower than for the non-aneurysmal patient (mean -18.38 vs. -2). In addition, TAG values were compared to PRTg obtained for each patient. Thrombotic risk stratification for KD aneurysms may be improved by incorporating TAG and should be evaluated in future prospective studies. 9:57AM G7.00010 Fractional Modeling of Viscoelasticity in Brain Aneurysms , YUE YU, Lehigh University, GEORGE KARNIADAKIS, Brown University — We develop fundamental new numerical methods for fractional order PDEs, and investigate corresponding models for arterial walls. Specifically, the arterial wall is a heterogeneous soft tissue with complex biomechanical properties, and its constitutive laws are typically derived using integer-order differential equations. However, recent simulations on 1D model have indicated that fractional order models may offer a more powerful alternative for describing arterial wall mechanics, because they are less sensitive to the parameter estimation compared with the integer-calculus-based models. We study the specific fractional PDEs that better model the properties of the 3D arterial walls, and for the first time employ them in simulating flow structure interactions for patient-specific brain aneurysms. A comparison study indicates that for the integer order models, the viscous behavior strongly depends on the relaxation parameters while the fractional order models are less sensitive. This finding is consistent with what is observed in the 1D models for arterial networks (Perdikaris & Karniadakis, 2014), except that when the fractional order is small, the 3D fractional-order models are more sensitive to the fractional order compared to the 1D models. Monday, November 24, 2014 8:00AM - 10:10AM Session G8 Focus Session: Superhydrophobicity and Drag Reduction I — 3001/3003 - Gareth McKinley, Massachusetts Institute of Technology 8:00AM G8.00001 Effect of fluctuating pressure on plastron stability of superhydrophobic surfaces1 , LINFENG PIAO, HYUNGMIN PARK, Seoul National University — In the present study, we theoretically predict the collapse transition (de- pinning from the edge or sagging touchdown) and breakdown of a plastron on superhydrophobic surfaces made up of micro-scale grates, under fluctuating pressure. Assuming a sinusoidally oscillating pressure, we constitute an oscillator equation, considering a gaseous diffusion across the interface together. The modeled equation is solved for a wide range of parameters for surface geometry and fluctuating pressure. The results show that the plastron collapses even before reaching the critical pressure (i.e., water depth of application) determined under a static pressure. Depending on the behavior of interface, we also classify transient and long-term regimes where the roles of dynamic pressure and gaseous diffusion are dominant, respectively. The dependence of plastron longevity on the surface geometry is found that the plastron on low gas-fraction surface (which breaks in long-term regime) lasts days while the one with high gas-fraction (> ∼ 70 − 90%), more susceptible to pressure fluctuation, lasts a shorter duration. Finally, we suggest that property of sidewalls in surface morphology is critical in the plastron longevity. 1 Supported by the NRF programs (NRF-2012M2A8A4055647, NRF-2013R1A1A1008373) of Korean government. 8:13AM G8.00002 Numerical study of wetting transition on patterned hydrophobic surfaces using the string method , WEIQING REN, National University of Singapore and IHPC — We study the wetting transition on micro-structured hydrophobic surfaces using the string method. On a patterned solid surface, a liquid droplet can exhibit the suspended Cassie-Baxter state, or impaled Wenzel state. We compute the transition states, the energy barriers, and the minimum energy paths for the wetting transition from the Cassie-Baxter state to the Wenzel state. Numerical results are obtained for the wetting of a hydrophobic surface textured with a square lattice of pillars. It is found that the wetting of the solid substrate occurs via infiltration of the liquid in a single groove, followed by lateral propagation of the liquid front. The propagation of the liquid front proceeds in a stepwise manner, and a zipping mechanism is observed during the infiltration of each layer. The minimum energy path for the wetting transition goes through a sequence of intermediate metastable states, whose wetted areas reflect the micro structure of the patterned surface. 8:26AM G8.00003 Studying the Microphysics of Superhydrophobic Surfaces using DNS1 , KARIM ALAME, KRISHNAN MAHESH, Univ of Minn - Minneapolis — DNS using the volume of fluid methodology will be used to study the microphysics of the gas-water interfaces in super-hydrophobic surfaces. The numerical method will be summarized along with relevant validation examples. The effect of pressure difference on an interface between solid walls will be discussed and contrasted to theory. Modes of interface failure will be presented. Simulations of channel flow with gas trapped in single longitudinal groove will be discussed and contrasted to results from approximate modeling approaches. Implications for air-layer drag reduction will be discussed. 1 Supported by Office of Naval Research 8:39AM G8.00004 The drag-reducing ingredients of superhydrophobic surfaces1 , YIXUAN LI, KRISHNAN MAHESH, University of Minnesota — The drag–reducing ingredients of superhydrophobic surfaces are studied for laminar and turbulent channel flow. Direct numerical simulation is used to examine the effects of micro–structure geometry and the state of the air–water interface, on drag reduction. Fully wetted simulations of the flow show how geometry alone yields an apparent slip to the external flow. An alternative to the commonly used zero–shear boundary condition is suggested for simulation of the interface. The amount of captured air is varied and its effect on net drag is quantified. The effect of meniscus curvature is considered and its effect on the flow is quantified. A local measure is introduced to examine the extent to which the flow inside the channel is affected. 1 Supported by Office of Naval Research 8:52AM G8.00005 Drag Reduction for Flow Past a Perfectly Hydrophobic Surface1 , GLEN MCHALE, Northumbria University, UK, MICHAEL I. NEWTON, Nottingham Trent University, UK, MORRIS R. FLYNN, University of Alberta, Canada, BRIAN R.K. GRUNCELL, NEIL D. SANDHAM, University of Southampton, UK, ANGELA BUSSE, Glasgow University, Scotland — We consider drag reduction for flow past a perfectly hydrophobic sphere (i.e. a vanishing Cassie solid surface fraction or with a Leidenfrost layer). At small Re number an exact analytical model for drag can be constructed for a sphere encapsulated in a layer of a gas (a “plastron”) [McHale, G. et al, Soft Matter 7 art. 10100, (2011)]. This predicts an optimum thickness for the gas layer for maximum drag reduction due to a competition between increased lubrication of the flow and increased cross-section for drag by the compound object (the solid plus its surface-retained layer of gas). Using numerical simulations for a perfectly hydrophobic solid sphere in water we show that the maximum drag reduction increases from 19% to 50% as the Re number increases to 100; this is due to suppression of flow separation and a narrower wake [Gruncell, B.R.K. et al, Phys. Fluids 25 art 043601, (2012)]. Introducing roughness into the simulations to model a superhydrophobic surface with a finite Cassie fraction results in less drag reduction because the vortex regime is no longer fully suppressed. Finally, we describe an analytical model of flow resistance through tubes or channels using similar boundary conditions to the flow past a gas-encapsulated sphere [Busse, A. et al, J. Fluid Mech. 727 488, (2013)]. 1 We acknowledge funding from the UK EPSRC (EP/G058318/1, EP/G069581/1 and EP/L026899/1) and the Canadian NSERC. 9:05AM G8.00006 The Effects of Superhydrophobic Surface on Near-wall Turbulence Structures and Drag Reduction1 , HYUNWOOK PARK, JOHN KIM, University of California, Los Angeles — Direct numerical simulations of a turbulent boundary layer (TBL) developing over a superhydrophobic surface (SHS) were performed. SHS was modeled through the shear-free boundary condition, assuming the air-water interface remained as a flat surface. It was found that SHS led to substantial drag reduction by weakening near-wall turbulence due to the lack of the shear over SHS. For the considered Reynolds number ranges and SHS geometries, it was found that the effective slip length normalized by viscous wall units was the key parameter and the effective slip length should be on the order of the buffer layer in order to have the maximum benefit of drag reduction. It was also found that the width of SHS, relative to the spanwise width of near-wall turbulence structures, was also a key parameter to the total amount of drag reduction. Similarities and differences between the present TBL and turbulent channel flows with SHS were also examined. In contrast to fully developed channel flows, the effective slip velocity and hence the effective slip length varied in the streamwise direction, implying that total drag reduction would depend on the streamwise length of a given SHS. This observation will be compared with recent experimental results.2 1 Supported 2 Park by ONR grant, N000141110503. et al., JFM 747 (2014) 722-734 9:18AM G8.00007 A numerical study of the effects of a superhydrophobic surface on nearwall turbulence characteristics1 , TAEYONG JUNG, HAECHEON CHOI, Seoul National University, JOHN KIM, University of California, Los Angeles — A superhydrophobic surface (SHS) in turbulent boundary layers can significantly affect near-wall turbulence, resulting in large skin-friction drag reduction. In this study, we performed direct numerical simulations of turbulent channel flow with SHS by solving both the main water flow and flow inside the air layer. The wall-parallel velocity and shear stress were maintained to be continuous across the interface between the air and water, while the interface was assumed to be flat. The Reynolds number considered was Re=5600 (based on the bulk velocity and channel height), and we varied the pitch length, gas fraction and air-layer thickness. It was found that these parameters had profound effects on the skin-friction drag, interfacial velocity and slip length. For example, with increasing the magnitudes of these parameters, the drag-reduction rate, interfacial velocity, and slip length increased. Also, near-wall vortical structures were significantly weakened, and the turbulence intensities were reduced near the SHS. At the SHS, streamwise and spanwise velocity (slip) fluctuations exist and their effects on the skin-friction drag will be discussed. 1 Supported by NRF program (NRF-2012M2A8A4055647) 9:31AM G8.00008 Direct numerical simulation of turbulent flows over superhydrophobic surfaces with periodic posts: effect of texture size1 , JONGMIN SEO, Stanford University, RICARDO GARCIA-MAYORAL , University of Cambridge, ALI MANI, Stanford University — Superhydrophobic surfaces submerged in water can produce slip on the wall and thus result in drag reduction by entrapping gas pockets between the roughness elements. This work aims to generate insights into the failure mechanism of such surfaces under turbulent conditions. We perform direct numerical simulations of channels with patterned slip/no-slip boundary conditions, for fixed gas fraction and texture wavelengths, L+ , ranging from 6 to 150 wall units, which include the regime of practical application. The rms pressure at the wall is found to have a fluctuating contribution, caused by the overlying turbulence, and a stationary contribution, caused by the stagnation of flow when encountering downstream solid posts. While the turbulence contribution remains essentially unmodified, the stationary pressure increases with the texture size, and can be responsible for the breakup of the entrapped gas bubbles. We present results revealing the scaling of the induced pressure and the consequent deformations of the air-water interface. 1 Supported by Office of Naval Research and the Kwanjeong Educational Scholarship Foundation 9:44AM G8.00009 Effective medium theory for drag-reducing micro-patterned surfaces in turbulent flows , ILENIA BATTIATO, San Diego State University — Many studies in the last decade have revealed that patterns at the microscale can reduce skin drag. Yet, the mechanisms and parameters that control drag reduction, e.g. Reynolds number and pattern geometry, are still unclear. We propose an effective medium representation of the micro-features, that treats the latter as a porous medium, and provides a framework to model turbulent flow over patterned surfaces. Our key result is a closed-form expression for the skin friction coefficient in terms of frictional Reynolds (or Karman) number in turbulent regime, the viscosity ratio between the fluid in and above the features, and their geometrical properties. We apply the proposed model to turbulent flows over superhydrophobic ridged surfaces. The model predictions agree with laboratory experiments for Reynolds numbers ranging from 3000 to 10000.1 1 Battiato, I., Eur. Phys. J. E (2014) 37: 19 DOI 10.1140/epje/i2014-14019-0 9:57AM G8.00010 Re-Entrant Structure for Robust Superhydrophobicity and Drag Reduction , HONG ZHAO, MOHAMED GAD-EL-HAK, Department of Mechanical and Nuclear Engineering, Virginia Commonwealth University, Richmond, VA 23284, USA — A re-entrant structure is required for superoleophobicity by effectively pinning low-surface-tension liquids from wetting the textures and forming a solid–liquid–air composite interface. In this work, we examine the contribution of a re-entrant structure to the robustness of superhydrophobicity and skinfriction reduction capabilities. Textured surfaces with wavy sidewall pillars provide re-entrant structures and are used as model surfaces. Gibbs energy analysis is conducted to study the pinning sites and wetting stability. The wetting robustness against pressure is characterized by breakthrough pressure, which is obtained by conservation of energy and force balance at the pinning sites. The slip length and slip velocity are evaluated through a shear stress and strain rate correlation, which is obtained using an Anton Paar rheometer. Gibbs energy analysis indicates that the breakthrough pressure provided by the wavy sidewall structure for water is about 18 times of that on the straight sidewall structure. This is mostly due to the energy barrier at the re-entrant structure. When a contact line advances onto and pins at the re-entrant structure, its slip performance degrades due to the increased no-slip fraction on the composite interface, but Cassie–Baxter state still remains. Monday, November 24, 2014 8:00AM - 9:57AM Session G9 Biofluids: Microswimmers I — 3014/3016 - Oscar Curet, Florida Atlantic University 8:00AM G9.00001 Locomotion in a turbulent world1 , M. KOEHL, Univ of California - Berkeley — When organisms swim or crawl in aquatic habitats, the water through which they travel is usually moving. Therefore, an important part of understanding how aquatic organisms locomote is determining how they interact with the fluctuating turbulent water currents through which they move. The research systems we have been using to address this question are microscopic marine animals swimming in turbulent, wavy water flow or crawling on surfaces in spatially-complex habitats exposed to such flow. Using a combination of field studies, wave-flume experiments, experiments in fluidic devices, and mathematical modeling, we have discovered that small organisms swimming or crawling in turbulent flow are not subjected to steady velocities. The shears, accelerations, and odor concentrations encountered by small swimmers and crawlers fluctuate rapidly, with peaks much higher than mean values. Although microscopic organisms swim slowly relative to ambient water flow, their locomotory behavior in response to the rapidly-fluctuating shears and odors they encounter can affect where they are transported by ambient water movement. Furthermore, the ability of small organisms to walk on surfaces without being dislodged by pulses of rapid flow constrains the microhabitats in which they can forage. 1 Supported by NSF grant #IOS-0842685. 8:13AM G9.00002 The Effect of Small Scale Turbulence on the Physiology of Microcystis aeruginosa cyanobacterium1 , ANNE WILKINSON, MIKI HONDZO, MICHELE GUALA, University of Minnesota — Microcystis aeruginosa is a single-celled blue-green alga, or cyanobacterium, that is responsible for poor water quality and microcystin production, which in high concentrations can be harmful to humans and animals. These harmful effects arise during cyanobacterium blooms. Blooms occur mainly in the summer when the algae grow uncontrollably and bond together to form colonies which accumulate on the surface of freshwater ecosystems. The relationship between fluid motion generated by wind and internal waves in stratified aquatic ecosystems and Microcystis can help explain the mechanisms of such blooms. We investigated the effect of small scale fluid motion on the physiology of Microcystis in a reactor with two underwater speakers. Different turbulent intensities were achieved by systematically changing the input signal frequency (30-50Hz) and magnitude (0.1-0.2V) to the speakers. The role of turbulence is quantified by relating energy dissipation rates with the cell number, chlorophyll amount, dissolved oxygen production/uptake, and pH. The results suggest that turbulence mediates the physiology of Microcystis. The findings could be instrumental in designing restoration strategies that can minimize Microcystis blooms. 1 This work was supported by the NSF Graduate Research Fellowship and University of Minnesota start-up funding. 8:26AM G9.00003 Response of Acartia tonsa to Burgers’ Vortex: Deconstructing TurbulenceCopepod Interactions , D.L. YOUNG, D.R. WEBSTER, J. YEN, Georgia Tech — In situ studies suggest that in many oceanic regimes, turbulence affects the vertical position of copepods primarily by changing their behavior, and only secondarily by altering their physical position. We test the hypothesis that fine-scale turbulence alters copepod behavior, presenting as alterations in directed movement and changes in swimming kinematics. To this end, we create two Burgers’ vortices, specifying the rotation rate and axial strain rate to correspond to turbulent vortices with size scale equaling the inverse wavenumber of the median viscous dissipation rate (i.e. r = 8.1η) for typical turbulent conditions in the coastal or near surface region (i.e., mean turbulent dissipation rates of 0.009 and 0.096 cm2 /s3 ). The vortex flow is quantified via tomo-PIV. Behavioral assays of Acartia tonsa are conducted, generating 3D trajectories for analysis of swimming kinematics and response to hydrodynamic cues. A. tonsa did not significantly respond to the vortex corresponding to dissipation rate of 0.009 cm2 /s3 , but drastically altered their swimming behavior in the presence of the 0.096 cm2 /s3 vortex, including increased relative swim speed, angle of alignment with the vortex axis, net-to-gross displacement ratio, and escape acceleration, along with decreased turn frequency (relative to stagnant control conditions). Further, A. tonsa escape location is preferentially in the core of the stronger vortex, indicating that the hydrodynamic cue triggering the distinctive escape behavior is vorticity. 8:39AM G9.00004 Comparison of Turbulence-Copepod Interaction: Temora longicornis vs. Acartia tonsa , N.H. DE JESUS-VILLANUEVA, University of Puerto Rico-Mayaguez, D.L. YOUNG, D.R. WEBSTER, J. YEN, Georgia Tech — The purpose of this study is to examine the behavioral response of the marine copepod Temora longicornis to a Burgers’ vortex intended to mimic the characteristics of a turbulent vortex that a copepod is likely to encounter in the coastal or near surface zone. Copepod behavioral assays were conducted for two turbulence levels corresponding to mean turbulent dissipation rates of 0.009 (Level 2) and 0.096 (Level 3) cm2 /s3 . The Burgers’ vortex parameters (i.e., circulation and axial strain rate) are specified to match a vortex corresponding to the median viscous dissipation rate for each target turbulence level. The behavioral response of T. longicornis compared to Acartia tonsa is of particular interest due to differences in swim style (cruiser vs. hop-sinker, respectively) and mechanosensory array morphology (planar vs. 3D, respectively). When exposed to these vortex flow treatments, T. longicornis exhibited a minimal behavioral response to the Level 2 vortex, but significantly altered their swimming behavior in the presence of the Level 3 vortex. Specifically, in the Level 3 vortex treatment T. longicornis increased relative swim speed, turn frequency, and escape acceleration while decreasing angle of alignment with the vortex axis and escape frequency (relative to stagnant control conditions). Histograms of escape jump location as a function of radius reveals no preferential escape location for T. longicornis, which contrasts the preferential escape location of A. tonsa in the vortex core. 8:52AM G9.00005 Response of a Motile/Non-Motile Escherichia coli Front to Hydrodynamic excitations1 , MAGALI BAABOUR, University Buenos Aeres (FUBA), CARINE DOUARCHE, Univ Paris Sud, Lab LPS, DOMINIQUE SALIN, Univ Pierre & Marie Curie, Lab FAST — In a recent study (Douarche et. al. PRL 102, 198101 (2009)), it has been shown that the motility of Escherichia coli (E. coli) is highly correlated to the oxygen level in a minimal medium: bacteria swim as long as they are provided with oxygen but reversibly transit to a non-motile state when they lack of it. Hence, when oxygen diffuses into an anaerobic sample of non-motile bacteria, a propagating front delineates a region of motile bacteria where oxygen is present from a region of non-motile ones where the oxygen is still not present. To study the response of this front to hydrodynamics excitation, we use the same fluorescent E. coli bacterial solution in rectangular cross section glass cells open to air (oxygen) at one inlet. After bacteria have consumed the oxygen of the solution, the presence of the oxygen only at the open edge of the sample leads to the formation of an analogous stationary front between motile and non-motile bacteria. We follow the response of this front to hydrodynamics flows such as an oscillating Poiseuille flow or natural convection. We analyze both the macroscopic behavior (shape and width) of the front as well as the microscopic displacements of individual bacteria. The dispersive behavior of this bacterial front is compared to the one of equivalent 1 Collaboration between Laboratories FAST and LPS, Univ Paris Sud and CNRS 9:05AM G9.00006 Gyrotactic cells subject to imposed 3D flows , NICHOLAS HILL, SCOTT RICHARDSON, ANDREW BAGGALEY, University of Glasgow — We examine the effect of imposed 3-dimensional test flows, specifically a Taylor–Green Vortex flow and an ABC flow, on the patterns and mixing of suspensions of gyrotactic swimming cells. Numerically solving the deterministic swimming trajectory equations for individual cells with random starting positions, we explore how the surrounding flow and the cell shape determine the long-time patterns. For certain parameter ranges these patterns often take the form of braided “plume-lie” structures, even when using the chaotic ABC flow. For various pattern configurations, analysis of the governing equations of motion reveals why they are formed, as analytical solutions of the equations for the swimming cell trajectories can be obtained. These patterns persist when small random perturbations (noise) are added to individual trajectories. 9:18AM G9.00007 Distribution of particle displacements due to swimming microorganisms1 , JEAN-LUC THIFFEAULT, University of Wisconsin - Madison — The experiments of Leptos et al. [Phys. Rev. Lett. 103, 198103 (2009)] show that the displacements of small particles affected by swimming microorganisms achieve a non-Gaussian distribution, which nevertheless scales diffusively. We use a simple model where the particles undergo repeated “kicks” due to the swimmers to explain the shape of the distribution as a function of the volume fraction of swimmers. The net displacement is determined by the self-convolution of the drift function caused by one swimmer, and a Poisson distribution for the frequency of interactions. The only adjustable parameter is the strength of the stresslet term in our spherical squirmer model. The effective diffusivity measured in the experiments is consistent with the model, with no further parameter adjustments. 1 Supported by NSF grant DMS-1109315 9:31AM G9.00008 Finding the best swimming sheet1 , TOM IVES, ALEXANDER MOROZOV, SUPA, School of Physics & Astronomy, University of Edinburgh, UK — Many microorganisms propel through fluid environments by undulating their bodies or long thin organelles (flagella). The particular waveform of the undulations can often be changed by the organism to adapt to particular environmental conditions. It has been proposed in the literature that this adaptation is driven by the desire to optimise the swimming efficiency. However, it remains an open question as to whether this is indeed the optimised quantity for microorganisms. We study propulsion in Newtonian fluids at zero inertia for a model organism, the so-called Taylor waving sheet. We develop a numerical method that allows us to calculate flow fields for sheets of arbitrary waverforms in the bulk and next to a wall. We perform optimisations of various quantities that can potentially be optimised by a swimming microorganisms (efficiency, speed, etc.) and present the optimal waveforms. We also present a simple analytical model that yields similar results. We conclude that various optimal waveforms are very similar, both in the bulk and next to a boundary, and one cannot claim that optimising the swimming efficiency is the strategy adopted by undulating microorganisms. 1 SUPA, School of Physics & Astronomy, University of Edinburgh, UK 9:44AM G9.00009 Understanding the detailed motion of a model bacterium , AKANKSHA THAWANI, MAHESH TIRUMKUDULU, Indian Inst of Tech-Bombay — Inspired by the motion of flagellated bacteria such as Escherichia coli and Bacillus subtilis, we have built a macroscopic model bacterium, in order to investigate the intricate aspects of their motion which cannot be visualized under a microscope. The flagellated rod shaped cells were approximated with a spherical head attached to a rigid metal helix, via a plastic hook. The motion of model bacterium was observed in a high viscosity silicone oil to replicate the low Reynolds number flow conditions. A significant wobble was observed even in the absence of an off-axis flagellum. We suspect that the flexibility in the hook connecting the head and flagellum is the cause for wobble, since wobble was observed to increase significantly with hook-flexibility. The motion of the model bacterium was predicted using the Slender Body theory of Lighthill, and was compared with the measured trajectories. Monday, November 24, 2014 8:00AM - 10:10AM Session G10 Microscale Flows: Emulsions and Interfaces — 3005 - Patrick Tabeling, ESPCI 8:00AM G10.00001 On tail formation during gravure printing of sessile drops , UMUT CEYHAN, S.J.S. MORRIS, University of California, Berkeley — Kitsomboonloha et al.(2012) study the deposition of femtolitre drops by the gravure method. The substrate (gravure plate) passes under a stationary blade; liquid placed on the substrate upstream of the blade fills the engraved wells as they enter the blade–substrate gap. Motion of the substrate beneath the blade removes the excess, leaving liquid–filled wells. The resulting pattern can then be printed. As a well leaves the blade, some liquid is, however, subtracted from it and left as a tail between the well and blade. Tails are undesirable because they reduce the sharpness of printed features. It was proposed that tails form by a 3–dimensional mechanism involving lateral wicking of liquid from the wells along the blade–substrate intersection. Here, lubrication theory is used to show that the effect can be understood within the context of plane flow. As a well passes under the trailing edge of the blade, capillary suction causes the meniscus to rise on the blade, but once the well has left, the increased drag exerted by the substrate pulls the meniscus down. Liquid dragged from the meniscus forms the tail. We conclude that tail formation is a problem in plane Stokes flow. 8:13AM G10.00002 Slip-accelerated falling drop along a vertical fiber , HSIEN-HUNG WEI, National Cheng Kung University, DAVID HALPERN, University of Alabama — Effects of wall slip on the motion of a falling drop along a vertical fiber are investigated theoretically. Using lubrication theory, we derive an interfacial evolution equation to describe how the drop’s travelling speed and height vary with the Bond number and the slip length. Our numerical results reveal that the drop can travel much faster than the one without slip due to the dramatic increase in the travelling speed with the slip length. The drop height is also found to rapidly increase with the slip length, which is due to enchanced capillary draining from the film into the drop. For Bond number above some critical value, however, capillary draining is suppressed and hence so is the drop height. We determine how the critical Bond number varies with the slip length. For a sufficiently large Bond number, the relevant Kuramoto-Sivashinsky equation is also derived to reveal how the suppression of the capillary instability is mediated by slip effects in the weakly nonlinear regime. 8:26AM G10.00003 Film flow over an inclined plate: effects of solvent properties and contact angles , RAJESH SINGH, JANINE GALVIN, National Energy Technology Laboratory, Albany — The liquid film behavior on the structured packing is a key aspect to the overall efficiency of the column. In this context, the effects of solvent properties and contact angle (γ) on the hydrodynamics of film flow are systematically investigated. Specifically, multiphase flow simulations for film flow over an inclined plate are carried out using volume of fluid method. A scaling analysis for film thickness and interfacial area was performed. Accordingly, a theory for film thickness and wetted area in terms of Kapitza number (Ka) is proposed. The advantage of the Ka is that it only depends on fluid properties and independent of flow parameters. Therefore the Ka becomes fixed for a given solvent and it decreases with increasing solvent viscosity. The results show that for a fully wetted plate the film thickness (δ) decreases with increasing Ka number as δ ∼ 1/Ka1/4 . For rivulet flow, the interfacial area (AIn ) is found to decrease with increasing Ka value. Indeed, scaling analysis shows the relation AIn ∼ 1/Ka1/2 . The effect of varying contact angle on the hydrodynamics of rivulet flows was also investigated. The contact angle has no impact on the film thickness for a fully wetted plate but strongly influences the interfacial area for the case of partially wetted plate. For rivulet flow the interfacial area increases with increasing contact angle and is holds the relation AIn ∼ 1/(1 − cos γ)m for a wide range of contact angle. The value of exponent m depends on the Ka number and shows two values, one for medium to high surface tension and another for low surface tension value. 8:39AM G10.00004 Fluid structure in the immediate vicinity of an equilibrium contact line from first principles and assessment of disjoining pressure models , ANDREAS NOLD, DAVID N. SIBLEY, Imperial College London, BENJAMIN D. GODDARD, The School of Mathematics and Maxwell Institute for Mathematical Sciences, The University of Edinburgh, SERAFIM KALLIADASIS, Imperial College London — Predicting the fluid structure at a three-phase contact line of macroscopic drops is of interest from a fundamental fluid dynamics point of view. However, exact computations for very small scales are prohibitive. As a consequence, coarse-grained quantities such as interface height and disjoining pressure profiles are used to model the interface shape. Here, we evaluate such coarse-grained models within a rigorous and self-consistent framework based on statistical mechanics, in particular with a Density Functional Theory (DFT) approach. We examine the nanoscale behavior of an equilibrium three-phase contact line in the presence of long-ranged intermolecular forces by employing DFT together with fundamental measure theory. Our analysis also enables us to evaluate the predictive quality of effective Hamiltonian models in the vicinity of the contact line. We compare the results for mean field effective Hamiltonians with disjoining pressures defined through the adsorption isotherm for a planar liquid film, and the normal force balance at the contact line [Phys. Fluids, 26, 072001, 2014]. Results are given for a variety of contact angles. An accurate description of the small-scale behavior of a three-phase conjunction is a prerequisite to understanding dynamic wetting phenomena. 8:52AM G10.00005 Influence of surface properties and miscibility upon displacement flow in microchannels1 , YU LU, EMILIA NOWAK, University of Birmingham, JAMES PERCIVAL, CHRIS PAIN, Imperial College London, MARK SIMMONS, University of Birmingham — Microfluidics have potential for a wide range of applications, yet successful operation will depend upon precise understanding of the fluid distribution in multiphase operations, particularly for cleaning of the system. Experiments and numerical studies have been conducted on the displacement process in a single microchannel geometry as a function of fluid dynamics and fluid properties. The microchannel has a near-semicircle cross-section of 205 µm width and 100 µm in depth. Miscible and immiscible fluid pairs (water/glycerol and water/silicon oil respectively) with different viscosity ratios and wettability of the channel walls are examined. Micro-Particle Image Velocimetry (µ-PIV) and Planar Laser Induced Fluorescence (PLIF) are used to obtain velocity fields and interface profile respectively. Flow patterns, interfacial instabilities, displacement efficiency and possible secondary flows are examined. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 9:05AM G10.00006 Dynamics of liquid bridges inside microchannels subject to pure oscillatory flows , MAJID AHMADLOUYDARAB, JALEL AZAIEZ, ZHANGXIN CHEN, University of Calgary — We report on 2D simulations of liquid bridges’ dynamics in microchannels of uniform wettability and subject to external oscillatory flows. The flow equations were solved using the Cahn-Hilliard diffuse-interface formulation and the finite element method with unstructured grid. It was found that regardless of the wettability properties of the microchannel walls, there is a critical frequency above which the bridge shows perpetual periodic oscillatory motion. Below that critical frequency, the liquid bridge ruptures when the channel walls are philic and detaches from the surface when they are phobic. This critical frequency depends on the viscosity ratio, oscillation amplitude and geometric aspect ratio of the bridge. It was also found that the flow velocity is out of phase with the footprint/throat lengths and that the latter two show a phase difference. These differences were explained in terms of the motion of the two contact lines on the substrates and the deformation of the fluid-fluid interfaces. To characterize the behavior of the liquid bridge, two quantitative parameters; the liquid bridge-solid interfacial length and the length of the throat of the liquid bridge were used. Variations of the interfacial morphology development of the bridge were analyzed to understand the bridge response. 9:18AM G10.00007 Confined Selective Withdrawal , ALVARO EVANGELIO, FRANCISCO CAMPO-CORTES, JOSE MANUEL GORDILLO, Universidad de Sevilla — It is well known that the controlled production of monodisperse simple and composite emulsions possesses uncountable applications in medicine, pharmacy, materials science and industry. Here we present both experiments and slender-body theory regarding the generation of simple emulsions using a configuration that we have called Confined Selective Withdrawal, since it is an improved configuration of the classical Selective Withdrawal. We consider two different situations, namely, the cases when the outer flow Reynolds number is high and low, respectively. Several geometrical configurations and a wide range of viscosity ratios are analyzed so that the physics behind the phenomenon can be fully understood. In addition, we present both experiments and theory regarding the generation of composite emulsions. This phenomenon is only feasible when the outer flow Reynolds number is low enough. In this case, we propose a more complex theory which requires the simultaneous resolution of two interfaces in order to predict the shape of the jet and the sizes of the drops formed. The excellent agreement between our slender-body approximation and the experimental evidence fully validates our theories. 9:31AM G10.00008 A semi-implicit finite element method for viscous lipid membranes1 , DIEGO RODRIGUES, Universidade de São Paulo, Brazil, ROBERTO AUSAS, Universidade de São Paulo, Brazil / IBM Research, Brazil, FERNANDO MUT, GUSTAVO BUSCAGLIA, Universidade de São Paulo, Brazil — We propose a robust simulation method for phospholipid membranes. It is based on a mixed three-field formulation that accounts for tangential fluidity (Boussinesq-Scriven law), bending elasticity (Canham-Helfrich model) and inextensibility. The unknowns are the velocity, vector curvature and surface pressure fields, all of which are interpolated with linear continuous finite elements. The method is semi-implicit, it requires the solution of a single linear system per time step. Conditional time stability is observed, with a time step restriction that scales as the square of the mesh size. Mesh quality and refinement are maintained by adaptively remeshing. Another ingredient is a numerical force that emulates the action of an optical tweezer, allowing for virtual interaction with the membrane. Extensive relaxation experiments are reported. Comparisons to exact shapes reveal the orders of convergence for position (5/3), vector curvature (3/2), surface pressure (1) and bending energy (2). Tweezing experiments are also presented. Convergence to the exact dynamics of a cylindrical tether is confirmed. Further tests illustrate the robustness of the method (six tweezers acting simultaneously) and the significance of viscous effects on membrane’s deformation under external forces. 1 The authors acknowledge the financial support received from grants #11/01800-5, #12/14481-8 and #12/23383-0, São Paulo Research Foundation (FAPESP). 9:44AM G10.00009 Numerical simulations of interacting surfactant-laden jets in microfluidic channels1 , GARVIT GOEL, IIT Delhi, JUNFENG YANG, JOAO CABRAL, OMAR MATAR, Imperial College London — We consider the dynamics of jets of surfactant solution in oil under microfluidic confinement. Previous experimental work has demonstrated the occurrence of “jetting” and “dripping” flow regimes depending on the choice of oil and water flow rates, viscosity ratio, and surfactant concentration. To take into account the influence of soluble surfactant on the behaviour of the jets, we present a computational fluid dynamics (CFD) approach which uses the Volume-of-Fluid method capturing the interface topology accurately with minimal mass loss. This approach accounts for sorption kinetics, Marangoni stresses, diffusion, and surface dilation. This method is incorporated into a CFD code to study the jetting and dripping regimes in a microfluidics channel. The modelling results are validated against experimental measurements. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1, and Grant number EP/J010502/1 9:57AM G10.00010 The fluid dynamics of microjet explosions caused by extremely intense X-ray pulses , CLAUDIU STAN, HARTAWAN LAKSMONO, RAYMOND SIERRA, PULSE Institute, SLAC National Accelerator Facility, DESPINA MILATHIANAKI, JASON KOGLIN, LCLS, SLAC National Accelerator Facility, MARC MESSERSCHMIDT, BioXFEL STC, GARTH WILLIAMS, LCLS, SLAC National Accelerator Facility, HASAN DEMIRCI, PULSE Institute, SLAC National Accelerator Facility, SABINE BOTHA, KAROL NASS, Max-Planck Institute for Medical Research, HOWARD STONE, Princeton University, ILME SCHLICHTING, ROBERT SHOEMAN, Max-Planck Institute for Medical Research, SEBASTIEN BOUTET, LCLS, SLAC National Accelerator Facility — Femtosecond X-ray scattering experiments at free-electron laser facilities typically requires liquid jet delivery methods to bring samples to the region of interaction with X-rays. We have imaged optically the damage process in water microjets due to intense hard X-ray pulses at the Linac Coherent Light Source (LCLS), using time-resolved imaging techniques to record movies at rates up to half a billion frames per second. For pulse energies larger than a few percent of the maximum pulse energy available at LCLS, the X-rays deposit energies much larger than the latent heat of vaporization in water, and induce a phase explosion that opens a gap in the jet. The LCLS pulses last a few tens of femtoseconds, but the full evolution of the broken jet is orders of magnitude slower – typically in the microsecond range – due to complex fluid dynamics processes triggered by the phase explosion. Although the explosion results in a complex sequence of phenomena, they lead to an approximately self-similar flow of the liquid in the jet. Monday, November 24, 2014 8:00AM - 10:10AM Session G11 Microscale Flows: Porous Media and Thermal Transport — 3007 - James Palko, University of Michigan 8:00AM G11.00001 Optimum design for effective water transport through a double-layered porous hydrogel inspired by plant leaves1 , HYEJEONG KIM, HYEONJEONG KIM, HYUNGKYU HUH, HYUNG JU HWANG, SANG JOON LEE, Pohang Univ of Sci & Tech — Plant leaves are generally known to have optimized morphological structure in response to environmental changes for efficient water usage. However, the advantageous features of plant leaves are not fully utilized in engineering fields yet, since the optimum design in internal structure of plant leaves is unclear. In this study, the tissue organization of the hydraulic pathways inside plant leaves was investigated. Water transport through double-layered porous hydrogel models analogous to mesophyll cells was experimentally observed. In addition, computational experiment and theoretical analysis were applied to the model systems to find the optimal design for efficient water transport. As a result, the models with lower porosity or with pores distributed widely in the structure exhibit efficient mass transport. Our theoretical prediction supports that structural features of plant leaves guarantee sufficient water supply as survival strategy. This study may provide a new framework for investigating the biophysical principles governing the morphological optimization of plant leaves and for designing microfluidic devices to enhance mass transport ability. 1 This study was supported by the National Research Foundation of Korea and funded by the Korean government. 8:13AM G11.00002 Flow-induced fiber deformation in a confined microchannel: in situ mechanical testing of gels , CAMILLE DUPRAT, Ladhyx, Ecole Polytechnique, HELENE BERTHET, PMMH, ESPCI, JASON WEXLER, Princeton University, OLIVIA DU ROURE, ANKE LINDNER, PMMH, ESPCI — Photopolymerized hydrogels are a functional template for micro-particle fabrication, microflowsensors and microbiology experiments. The control and knowledge of their mechanical properties are paramount to many applications. We have designed a novel robust method to determine these properties. We measure the deformation of a gel beam of precisely controlled shape, under a controlled flow forcing, which provides a direct measurement of the Young’s modulus of the gel upon its fabrication. We then use this method to determine the mechanical properties of the commonly used poly(ethylene glycol) diacrylate (PEGDA) under various experimental conditions. The mechanical properties of the gel can be highly tuned, yielding two orders of magnitude in the Young’s modulus. We provide a simple control parameter, the UV exposure time, to have a great control over the network properties, and rationalize these observations by studying the kinetics of the polymerization reaction. 8:26AM G11.00003 Characterization of Fluid Flow in Paper-Based Microfluidic Systems , NOOSHEEN WALJI, BRENDAN MACDONALD, University of Ontario Institute of Technology — Paper-based microfluidic devices have been presented as a viable low-cost alternative with the versatility to accommodate many applications in disease diagnosis and environmental monitoring. Current microfluidic designs focus on the use of silicone and PDMS structures, and several models have been developed to describe these systems; however, the design process for paper-based devices is hindered by a lack of prediction capability. In this work we simplify the complex underlying physics of the capillary-driven flow mechanism in a porous medium and generate a practical numerical model capable of predicting the flow behaviour. We present our key insights regarding the properties that dictate the behaviour of fluid wicking in paper-based microfluidic devices. We compare the results from our model to experiments and discuss the application of our model to design of paper-based microfluidic devices for arsenic detection in drinking water in Bangladesh. 8:39AM G11.00004 Flow control for a paper-based microfluidic device by adjusting permeability of paper , ILHOON JANG, GANGJUNE KIM, SIMON SONG, Dept. of Mechanical Convergence Engineering, Hanyang University, Korea — The paper-based microfluidics has attracted intensive attention as a prospective substitute for conventional microfluidic substrates used for a point-of-care diagnostics due to its superior advantages such as the cost effectiveness and production simplicity. Generally, a paper-based microfluidic device utilizes capillary force to drive a flow. Recent studies on flow control in such a device aimed at obtaining accurate and quantitative results by varying a channel geometry like width and length. According to the Darcy’s law describing a flow in a porous media like paper, a flow rate can be adjusted the permeability of paper. In this study, we investigate a flow control method by adjusting the permeability of paper. We utilize the wax printing for the adjustment and the fabrication of paper channels. A rectangular wax pattern was printed on one inlet channel of a Y-channel geometry. By varying the brightness of the wax pattern, a relationship between the flow rate and permeability changes due to the wax was investigated. As a result, we obtained an effective permeability contour with respect to the wax pattern length and brightness. In addition, we developed a paper-based micromixer of which the mixing ratio was controlled precisely by adjusting the permeability. 8:52AM G11.00005 Nanofluidic control by nanoporous materials using electrocapillary effects , YAHUI XUE, HUILING DUAN, Department of Mechanics and Engineering science, College of Engineering, Peking University, Beijing, JUERGEN MARKMANN, PATRICK HUBER, JOERG WEISSMUELLER, Institute of Materials Physics and Technology, Hamburg University of Technology, Hamburg — Electrocapillary techniques exhibit great advantages in nonmechanical electrofluidic manipulation, e.g., flow actuation in micro-/nano- channels. One issue of interest is the spontaneous imbibition of fluids in bodies with a nanoscale pores size. Contrary to previous studies we here use a metallic nanoporous body. This allows us to control the electrode potential at the solid-fluid interface. Nanoporous gold (NPG) with uniform pore- and ligament size of 45 nm was fabricated by dealloying an Ag75Au25 alloy. Spontaneous imbibition of aqueous electrolytes obeys the Lucas-Washburn law. Interestingly, the estimated tortuosity has the low value of 3.2 (3 is expected for an isotropic sponge). Electrocapillary effects were then used to manipulate the imbibition dynamics. As a result of the enhanced wetting by the electrocapillary effects, we observed an acceleration of the imbibition by 30%. When air as the pore fluid is replaced with cyclohexane, we show for aqueous electrolyte imbibition in nanoporous gold that the fluid flow can be reversibly switched on and off through electric potential control of the solid–liquid interfacial tension. Our findings demonstrate that the high electric conductivity along with the pathways for fluid/ionic transport render nanoporous gold a versatile, accurately controllable electrocapillary pump and flow sensor for minute amounts of liquids with exceptionally low operating voltages. 9:05AM G11.00006 Effect of Axial Fluid Conduction on Thermal Transport in Superhydrophobic Microchannels1 , ADAM COWLEY, DANIEL MAYNES, JULIE CROCKETT, Brigham Young Univ - Provo, FLUIDS AND THERMAL TRANS- PORT LAB - BYU TEAM — Convective heat transfer in a rib/cavity structured superhydrophobic microchannel is explored numerically. The cavities are assumed to be in the Cassie state (not wet) and the liquid meniscus is modeled as flat. The ribs are oriented perpendicular to the flow direction and are smaller than the channel hydraulic diameter. A constant heat flux condition is prescribed at the top of the ribs while the gas/liquid interface is approximated as adiabatic. The varied parameters include Peclet number, relative cavity size, and relative channel-wall spacing. The influence of fluid axial conduction is explored and it is found that axial conduction plays a significant role in superhydrophobic microchannels. Aggregate results are presented in the form of an average Nusselt number and the ratio of the temperature jump length to the hydrodynamic slip length. These results are compared to two previous studies: one where axial conduction was neglected and another where diffusion is assumed to be dominant. Overall, the results show that heat transfer is decreased for a superhydrophobic channel when compared to a classical smooth channel and that axial conduction exerts influence over a much larger range of parameters than prevails for classical no-slip channels. 1 This research was supported by the National Science Foundation (NSF) (Grant No. CBET-1235881) 9:18AM G11.00007 Fluorescence Thermometry Characterization of Microchannel Cooling Performance with Sidewall Heating , TAE JIN KIM, Stanford University, CARLOS HIDROVO, Northeastern University — Microchannel cooling of complex circuitry in microelectronics and LOC systems is an area of continued research that is constantly evolving. As such, it is important to properly evaluate the heat removal efficiency of the microchannels in near wall heating configurations. In this talk we evaluate the cooling efficiency of microchannels with microheaters embedded on the sidewalls. The microchannels are fabricated using soft lithography and the embedded joule heaters are created by filling molten low melting temperature alloys in two satellite microchannels and solidifying them. In order to assess the thermal transport rate, fluorescent images of the fluid mixed with two temperature-sensitive dyes were captured and pre-conditioned with an image-distortion correction algorithm. By taking their ratiometric value versus temperature measurements, results show that the heat removal efficiency initially increases as a function of Re and then plateaus at about 50 percent once Re is greater than 20. This behavior is the result of a decreasing advective resistance with increasing flow rate, where the ratio of the substrate-environment resistance to the wall-fluid convective resistance determines the ultimate performance of the cooling microchannel. 9:31AM G11.00008 Multiscale simulation of time-dependent thermal transpiration in largescale systems1 , DUNCAN A. LOCKERBY, ALEXANDER PATRONIS, University of Warwick, MATTHEW K. BORG, University of Strathclyde, JASON M. REESE, University of Edinburgh — We describe the development of an efficient hybrid continuum-molecular approach for simulating non-isothermal, lowspeed, internal rarefied gas flows, both in time and space. This is applied to transient flow in macro-scale Knudsen compressors, which is governed by both rarefied gas and continuum fluid dynamics. The method is an extension of the hybrid approach presented by Patronis et al. (2013) [J. Comp. Phys., 255, pp 558-571], which is based on the framework originally proposed by Borg et al. (2013) [J. Comp. Phys., 233, pp 400-413] for the simulation of micro/nano flows of high-aspect-ratio. The efficiency of the multiscale method allows the investigation of alternative Knudsen-compressor configurations to be undertaken. A comparison is made with published experimental data of the transient response (of pressure) in differentially heated reservoirs suddenly connected by a micro capillary. The multiscale simulation results compare very closely to the experimental data and are obtained at a fraction of the cost of a full kinetic or molecular solution. Recommendations for future development and opportunities are discussed. 1 This work is financially supported in the UK by EPSRC Programme Grant EP/I011927/1 and EPSRC grants EP/K038664/1 and EP/K038621/1. 9:44AM G11.00009 Controlling foam drainage in a 2D microchamber using thermocapillary stress , VINCENT MIRALLES, ESPCI, ISABELLE CANTAT, IPR, MARIE-CAROLINE JULLIEN, ESPCI, MMN TEAM, IPR TEAM — We investigate the drainage of a 2D microfoam in a vertical Hele-Shaw cell, and show that the Marangoni stress at the air-water interface generated by a constant temperature gradient applied in situ can be tuned to control the drainage. The temperature gradient is applied in such a way that thermocapillarity and gravity have an antagonist effect. We characterize the drainage over time by measuring the liquid volume fraction in the cell and find that thermocapillarity can overcome the effect of gravity, effectively draining the foam towards the top of the cell, or exactly compensate it, maintaining the liquid fraction at its initial value over at least 60 s. We quantify these results by solving the mass balance in the cell, and provide insight on the interplay between gravity, thermocapillarity and capillary pressure governing the drainage dynamics. Finally we use this model system to provide insight in the drainage dynamics for a more complex interfacial rheology, using insoluble surfactants inducing a solutocapillary effect. 9:57AM G11.00010 Volume-Of-Fluid Simulation for Predicting Two-Phase Cooling in a Microchannel , CATHERINE GORLE, Stanford University, PRITISH PARIDA, IBM Watson Research Center, FARZAD HOUSHMAND, MEHDI ASHEGHI, KENNETH GOODSON, Stanford University — Two-phase flow in microfluidic geometries has applications of increasing interest for next generation electronic and optoelectronic systems, telecommunications devices, and vehicle electronics. While there has been progress on comprehensive simulation of two-phase flows in compact geometries, validation of the results in different flow regimes should be considered to determine the predictive capabilities. In the present study we use the volume-of-fluid method to model the flow through a single micro channel with cross section 100 x 100 µm and length 10mm. The channel inlet mass flux and the heat flux at the lower wall result in a subcooled boiling regime in the first 2.5mm of the channel and a saturated flow regime further downstream. A conservation equation for the vapor volume fraction, and a single set of momentum and energy equations with volume-averaged fluid properties are solved. A reduced-physics phase change model represents the evaporation of the liquid and the corresponding heat loss, and the surface tension is accounted for by a source term in the momentum equation. The phase change model used requires the definition of a time relaxation parameter, which can significantly affect the solution since it determines the rate of evaporation. The results are compared to experimental data available from literature, focusing on the capability of the reduced-physics phase change model to predict the correct flow pattern, temperature profile and pressure drop. Monday, November 24, 2014 8:00AM - 9:57AM Session G12 Drops: Splashing, Stability and Breakup I — 3018 - Pierre Colinet, Univ Libre de Bruxelles 8:00AM G12.00001 To splash or not to splash? That is the question , GUILLAUME RIBOUX, JOSE MANUEL GORDILLO, Universidad de Sevilla — When a drop impacts a smooth, dry surface at a velocity above the so-called critical speed for drop splashing, the initial liquid volume losses its integrity, fragmenting into tiny droplets violently ejected radially and vertically outwards. Supported by experimental evidence, we obtained a theoretical criterium for the critical velocity for which a spherical liquid drop impacting onto a smooth dry surface produces the splash (Riboux G. & Gordillo J. M., Phys. Rev. Lett., 113, 024507, 2014). Our theory reveals that splashing is a consequence of the aerodynamic take-off experienced by the edge of the thin lamella which is ejected as a consequence of the impact. In this presentation, we apply our theory to describe what happens when the drop impact velocity is above the critical one. More precisely, we quantify the spatio-temporal evolution of the edge of the lamella, from which drops are ejected vertically and radially outwards once the rim destabilizes due to capillary effects. 8:13AM G12.00002 Numerical simulations of droplet breakup , JOMELA MENG, TIM COLONIUS, California Institute of Technology — The deformation and breakup of a liquid droplet in the flow behind a normal shock is simulated by solving the compressible NavierStokes equations using a multicomponent, shock- and interface-capturing algorithm. Fluids in the solver are modeled using the stiffened gas equation of state, which closes the system of governing equations. The interface is represented using volume fractions, which are evolved via an additional advection equation. Comparisons are made with experimental results in the literature for various metrics of deformation and breakup. As the post shock flow velocity is varied from low subsonic to slightly supersonic speeds, its effect on the breakup process and droplet acceleration are analyzed. It is shown that the transition does not alter the similarity of the unsteady acceleration and drag coefficient curves, which are successfully collapsed for the range of simulated shock Mach numbers. The effects of viscosity on droplet breakup are explored through comparisons with previous inviscid results. 8:26AM G12.00003 Droplet impact on highly viscous liquid: from experiments to numerics , ZHEN JIAN, Institut D’Alembert, CNRS & UPMC (Paris 06), France; Key Laboratory of Thermo-Fluid Science and Engineering of MOE, Xi’an Jiaotong University, China, GUY-JEAN MICHON, CHRISTOPHE JOSSERAND, STEPHANE ZALESKI, PASCAL RAY, Institut D’Alembert, CNRS & UPMC (Paris 06), UMR 7190, case 162, 4 place Jussieu, 75005 Paris, France, ZENGYAO LI, WENQUAN TAO, Key Laboratory of Thermo-Fluid Science and Engineering of MOE, Xi’an Jiaotong University, China — A numerical model is proposed to deal with the triple-phase impacting dynamics, of which a droplet of normal liquid impacts on a highly viscous liquid basin. Viscous effect is dominant during the dynamics as compared to the inertia and the surface tension. A liquid viscosity ratio ml is introduced to measure the viscosity deviation from a normal liquid as ml = µbasin /µdroplet,normal . Direct numerical simulations were executed with a code called Gerris. By increasing the liquid viscosity ratio ml , a continuous transition from L/L impact to L/S impact can be achieved. Two regimes are identified: wave-like regime and solidification regime. Experiments of droplet impacting on highly viscous liquid were also executed. Droplets of ethanol impact on a liquid basin of honey in a vacuum chamber where the gas pressure could be varied. A similarity to the impact on solid was observed, liquid basin performed as a solid and the complete suppression of splash was also observed by decreasing the gas pressure as reported for impacts on solid. Droplet shapes predicted by our simulations agree well with those observed in experiments. 8:39AM G12.00004 High Speed Drop Impact on Floating Oil Layers: Splash Behavior and Oily Marine Aerosol Production1 , DAVID MURPHY, CHENG LI, JOSEPH KATZ, Johns Hopkins University Department of Mechanical Engineering — Little is known about splash phenomena and marine aerosol formation occurring as high speed raindrops (We=ρv 2 d/σ > 2000) impact on thin crude oil slicks on seawater. Our experiments examine the effects of oil thickness and dispersant addition, which lowers the oil-air surface tension by 18% and oil-water interfacial tension by orders of magnitude. High speed imaging reveals that layer thickness and interfacial tension substantially impact splash behavior. In all high energy cases, a subsurface air cavity forms, and a supersurface crown with composition dependent on the layer thickness develops. When this crown closes, it generates upward and downward jets that contribute to oil entrainment. The initial raindrop impact ruptures only thin oil layers (< 200 µm). For thicker films, the crown comprises a short-lived upper oil film and a thicker lower section containing water and oil layers. Holographic microscopy shows a bimodal size distribution for airborne droplets ejected from ligaments on the crown rim, with peaks at 50 and 225 µm. The presence of oil increases the droplet production rate, as do increasing oil layer thickness and adding dispersant. Ejecta produced less than 0.3 ms after impact is another source of thousands of airborne microdroplets. 1 Sponsored by Gulf of Mexico Research Initiative (GoMRI) 8:52AM G12.00005 Dense suspension splash , KEVIN M. DODGE, IVO R. PETERS, JAKE ELLOWITZ, MARTIN H. KLEIN SCHAARSBERG1 , HEINRICH M. JAEGER, WENDY W. ZHANG, Department of Physics and the James Franck Institute, University of Chicago, Chicago, IL 60637 — Impact of a dense suspension drop onto a solid surface at speeds of several meters-per-second splashes by ejecting individual liquid-coated particles. Suppression or reduction of this splash is important for thermal spray coating and additive manufacturing. Accomplishing this aim requires distinguishing whether the splash is generated by individual scattering events or by collective motion reminiscent of liquid flow. Since particle inertia dominates over surface tension and viscous drag in a strong splash, we model suspension splash using a discrete-particle simulation in which the densely packed macroscopic particles experience inelastic collisions but zero friction or cohesion. Numerical results based on this highly simplified model are qualitatively consistent with observations. They also show that approximately 70% of the splash is generated by collective motion. Here an initially downward-moving particle is ejected into the splash because it experiences a succession of low-momentum-change collisions whose effects do not cancel but instead accumulate. The remainder of the splash is generated by scattering events in which a small number of high-momentum-change collisions cause a particle to be ejected upwards. 1 Current Address: Physics of Fluids Group, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands 9:05AM G12.00006 Drop impact of suspensions , M.-J. THORAVAL, F. BOYER, E. SANDOVAL NAVA, J.F. DIJKSMAN, D. LOHSE, J.H. SNOEIJER, Physics of Fluids, University of Twente — Drop impact studies have a wide range of applications, many of which involve complex fluids. We study here the liquid drop impact of a silver nano-particles dispersion on a solid glass surface. This dispersion is used for inkjet printing of functional electronic materials. When the impact velocity increases, the drop classically splashes into smaller droplets. However, it surprisingly stops splashing above a critical impact velocity. We combine high-speed imaging experiments with different characterizations of the dispersion to understand this transition to non-splashing. 9:18AM G12.00007 Barrel-Clamshell analog in the capillary bridges between two solid spheres , JAMES BIRD, TIMOTHY FARMER, Boston University — Sessile drops on a wire are known to adopt one of two topological configurations, referred to as a barrel or a clamshell, depending on the volume and contact angle. Here we report on an analogous topological transition for the capillary bridge between two contacting solid spheres. We characterize the transition by numerically computing the bridge shapes that minimize surface energy for a variety of contact angles and volumes. Finally, we are able to develop an exact solution to the non-axisymmetric bridge shapes by relying on symmetries of the geometry. 9:31AM G12.00008 Control of Drop Motion by Mechanical Vibrations , MICHAEL BESTEHORN, BTU Cottbus-Senftenberg, Cottbus, Germany — Since the first experimental observations of Michael Faraday in 1831 it is known that a vibrating liquid may show an instability of its flat free surface with respect to oscillating regular surface patterns. We study thin liquid films on a horizontal substrate in the long wave approximation. The films are parametrically excited by mechanical horizontal or inclined oscillations. Inertia effects are taken into account and the standard thin film formulation is extended by a second equation for the vertically averaged mass flux. The films can be additionally unstable by Van der Waals forces on a partially wetting substrate, leading to the formation of drops. These drops can be manipulated by the vibrations to move in a desired direction. Linear results based on a damped complex valued Mathieu equation as well as fully nonlinear results using a reduced model will be presented, for more details see [1,2]. [1] M. Bestehorn, Q. Han and A. Oron, Nonlinear pattern formation in thin liquid films under external vibrations, Phys. Rev. E 88, 023025 (2013). [2] M. Bestehorn, Laterally extended thin liquid films under external vibrations, Phys. Fluids 25, 114106 (2013). 9:44AM G12.00009 Equilibrium shapes of pendant monodisperse microbubbles suspension droplets , JUAN MANUEL FERNANDEZ, FRANCISCO CAMPO-CORTES, Universidad de Sevilla — The formation and stability of pendant droplets are a great value for both fundamental and engineering applications. In their pioneering work, Bashforth and Adams obtained the profile of a pendant pure liquid droplet by integrating the Young-Laplace equation. Since then, the stable and unstable conditions that govern the equilibrium of a pendant liquid droplet are well characterized. Here, we study the formation of droplets containing inside a suspension of monodisperse microbubbles. In this study, we present the different morphologies of these pendant multiphase droplets from the tip of a capillary tube of radius R for different average densities of the suspension droplet, defined as ρa = ρg αg + ρl αl where αg and αl are respectively the gas and liquid volume fractions. Experimental droplet profiles are compared with the theoretical predictions obtained by integrating the Young-Laplace equation. For low gas volume gas fraction (high liquid volume fraction), the suspension droplet shape (and consequently its maximum critical volume for stable equilibrium) is defined by the average Bond number, ρa gR2 /σ. However, for dense suspensions, αg > 0.7, the presence of microbubbles greatly changes the mode of drop formation. Monday, November 24, 2014 8:00AM - 10:10AM Session G13 Drops: General II — 3020 - Howard Stone, Princeton University 8:00AM G13.00001 Coffee-ring and uniform deposits from sessile nanofluid droplet evaporation1 , FEI DUAN, XIN ZHONG, Nanyang Technological University — Nanofluid droplet evaporation process has been investigated for the final deposits of particles for coffee-ring or uniform deposition. The 3-4 nm graphite nanoparticles are selected for preparing the nanofluids with surfactant or without the surfactant. The evaporation in term of contact angle, contact line, volume, spreading, etc, shows that the nanoparticles enhance the pinning effect and the evaporation rate, despite that the enhancement can be weakened as the nanoparticle concentration is higher in the samples without surfactant. In the sample with the surfactant, the variations of baseline, contact angle, volume and evaporation rate are abnormal at a certain surfactant concentration. Further discussion is conducted for the transition. The role of the surfactant influents the drying patterns from coffee ring to uniform deposition. The simulation is developed to help to understand the effect. 1 The authors acknowledge the support of A*Star Public Sector Funding (1121202010). 8:13AM G13.00002 Inverse Floatation1 , SAURABH NATH, Virginia Tech, ANISH MUKHERJEE, Georgia Tech, SOUVICK CHATTERJEE, Virginia Tech, RANJAN GANGULY, SWARNENDU SEN, ACHINTYA MUKHOPADHYAY, Jadavpur University, JONATHAN BOREYKO, Virginia Tech — We have observed that capillarity forces may cause floatation in a few non-intuitive configurations. These may be divided into 2 categories: i) floatation of heavier liquid droplets on lighter immiscible ones and ii) fully submerged floatation of lighter liquid droplets in a heavier immiscible medium. We call these counter-intuitive because of the inverse floatation configuration. For case (i) we have identified and studied in detail the several factors affecting the shape and maximum volume of the floating drop. We used water and vegetable oil combinations as test fluids and established the relation between Bond Number and maximum volume contained in a floating drop (in the order of µL). For case (ii), we injected vegetable oil drop-wise into a pool of water. The fully submerged configuration of the drop is not stable and a slight perturbation to the system causes the droplet to burst and float in partially submerged condition. Temporal variation of a characteristic length of the droplet is analyzed using MATLAB image processing. The constraint of small Bond Number establishes the assumption of lubrication regime in the thin gap. A brief theoretical formulation also shows the temporal variation of the gap thickness. 1 Jadavpur University, Jagadis Bose Centre of Excellence, Virginia Tech 8:26AM G13.00003 Vapor mediated droplet interactions - self-sensing droplet machines (Part 1) , NATE CIRA, ADRIEN BENUSIGLIO, MANU PRAKASH, Stanford University Department of Bioengineering — Reducing contact angle hysteresis is one strategy for making droplets mobile. Typically this involves carefully preparing a near-ideal surface. Here we show that a class of two-component droplets is self-motile on any high energy surface. Surprisingly, these binary droplets have characteristics of both completely and partially wetting fluids which precludes any hysteresis. This allows us to easily create mobile droplet systems with simple everyday materials like glass slides and sharpie. We build on the fundamental mechanisms and models we developed for this system and present multiple fluidic machines which take advantage of interactions between the droplets to autonomously execute complex tasks such as sorting and pattern formation. Time permitting we will run a live experiment to highlight the phenomena being discussed. 8:39AM G13.00004 Vapor mediated droplet interactions - models and mechanisms (Part 2) , ADRIEN BENUSIGLIO, NATE CIRA, MANU PRAKASH, Stanford University Department of Bioengineering — When deposited on clean glass a two-component binary mixture of propylene glycol and water is energetically inclined to spread, as both pure liquids do. Instead the mixture forms droplets stabilized by evaporation induced surface tension gradients, giving them unique properties such as negligible hysteresis. When two of these special droplets are deposited several radii apart they attract each other. The vapor from one droplet destabilizes the other, resulting in an attraction force which brings both droplets together. We present a flux-based model for droplet stabilization and a model which connects the vapor profile to net force. These simple models capture the static and dynamic experimental trends, and our fundamental understanding of these droplets and their interactions allowed us to build autonomous fluidic machines. 8:52AM G13.00005 The Effect of Maxwell Slip on Gravitational Collisions of Viscous Drops , MICHAEL ROTHER, University of Minnesota Duluth — When the gap between two spherical particles or liquid drops is comparable to the mean free path of surrounding fluid molecules, Stokes theory for the motion of the matrix fluid is no longer adequate. Use of the Maxwell slip approximation in this region shows that the resistance between two approaching surfaces decreases and that collision is possible, even for two solid particles. An important application of slip flow theory is to raindrop growth, where the mean free path of the air molecules is approximately 0.1 µm. Previous study of water drops in the atmosphere has treated small drops as solid spheres. In the current work, relative trajectories are calculated for two spherical liquid drops with exact methods for determining the hydrodynamic forces at finite Stokes number and low Reynolds number in gravitational flow. These constraints are met for drops between 10 and 30 µm in radius. In close approach, lubrication and attractive molecular forces are considered. In addition, Maxwell slip effects are determined exactly for motion along the drops’ line of centers by bispherical coordinate techniques. Collision efficiencies for liquid drops, including Maxwell slip with allowance for internal circulation in the drops, are compared to those for solid spheres. 9:05AM G13.00006 Rising motion of a suspension of drops in a linearly stratified fluid , SADEGH DABIRI, Purdue University, AMIN DOOSTMOHAMMADI, MORTEZA BAYAREH, University of Notre Dame, AREZOO ARDEKANI, Purdue University — Vertical variation of water temperature or salinity results in the generation of vertical density stratification in the water column in oceans and lakes and can affect the motion of particles, drops, and bubbles. Despite the broad body of research on the vertical motion of rigid particles in stratified fluids, the hydrodynamics of the vertical motion of deformable particles and drops in stratified fluids is currently poorly understood. In this manuscript, we report on the direct numerical simulations of a suspension of rising drops in a linearly stratified fluid. Our results show that density stratification suppresses both average rise velocity and velocity fluctuations of drops and results in an enhanced isotropy of velocity fluctuations. The combined effects of stratification, void fraction and deformability of drops on the motion of the suspension of drops are characterized. The results show that density stratification leads to an enhanced horizontal cluster formation. 9:18AM G13.00007 Drop Size and Velocity Distributions of the Spray of Aerated Liquid Injection in Gaseous Crossflow , KHALED SALLAM, ADEGBOYEGA ADEBAYO, Oklahoma State University — In this study an experimental investigation of the drop size and velocity distributions of the spray of an aerated liquid injection in subsonic crossflow is described. The test conditions for this study include injector exit diameter of 1 mm, crossflow Mach number of 0.3, momentum flux ratio of 5, and gas-to-liquid mass ratio of 8%. Double pulsed digital holography was used to investigate the spray characteristics at downstream distances of 50, and 100 jet diameter. The holograms are reconstructed into “slices” and analyzed using image-processing algorithms to yield information about the drop sizes and drop velocities. Four different drop size distributions are tested: the normal distribution, the Rosin-Rammler distribution, the log normal distribution, and the Simmons’ universal root-normal distribution. It was found that the log normal distribution best quantified the data obtained in this study. The drop streamwise and cross stream velocities at downstream distances of 50 jet diameter were found to be still evolving. The crossflow drop velocities converged to almost uniform velocity at a distance of 100 jet diameters downstream from the injector exit. 9:31AM G13.00008 On the Calculation of Planar Spray Characteristics from Point Sampled Data , KYLE BADE, RUDI SCHICK, Spraying Systems Co., SPRAY ANALYSIS AND RESEARCH SERVICES TEAM — An investigation into methods used to generate planar spray characteristics, such as mean drop size and velocity, from discrete measurement points is conducted. Two sprays are investigated, a hydraulic full cone spray, and an air-atomized multi-orifice spray, where an excessive number of sample-point locations are acquired using a Phase Doppler Interferometer (PDI) resulting in overly resolved spatial resolution, at a single axial plane. Due to the intentional spatial over-sampling, the data may be downsampled to determine the minimum number of required sample points to calculate reliable mean spray values. For each spray pattern, the influence of various spatial subsets of sampling points is investigated to determine the minimum number of required sample points for accurate planar results; normalized metric are developed to govern the number of sample points. In a related effort, meaningful average values are calculated using weighting methods and assessed for relative influence on the final calculations. Specifically, volume flux and discrete area weighting methods are developed and evaluated. evaluated. The results of this investigation allow a minimum number of data points to be processed into reliable planar spray characteristics. 9:44AM G13.00009 Dynamics of Rising Dispersant Laden Oil Drops in a Quiescent Environment , KHALIL CASTILLO-APONTE, Ithaca College, FARAZ MEHDI, JIAN SHENG, Texas Tech University — We study the dynamics of rising oil drops in a quiescent fluid chamber under uniform and density stratified conditions. Digital in-line holography allows for high resolution topological measurements and tracking of drop trajectories. Statistics of rising velocities, drop shapes and sizes are compared as functions of different oil-dispersant ratios. A conceptual model of an oil drop developing appendages and its subsequent breakdown into much smaller droplets is also discussed. 9:57AM G13.00010 The effects of aspect ratio on the flow invariants of droplets in an axisymmetric micro-tube , ADAM DEVORIA, KAMRAN MOHSENI, University of Florida — In this study the potential benefits of using a digitized flow (droplets) in place of a continuous flow are investigated. In particular, the aspect ratio (AR) of the droplets is varied and is an important parameter representing how different the droplet flow is from continuous flow. The flow within the droplets is measured with micro digital particle image velocimetry. The measurements are used to compute the flow invariants, namely circulation, hydrodynamic impulse, and kinetic energy in the droplet. It is found that the non-dimensionalized experimental invariants for low-AR droplets are increased above those for a corresponding segment of continuous flow. Also increased are the fluxes of the invariants, as well as the momentum flux. These increases above continuous flow go as the inverse of AR. For jetting applications, the implication is that using a digitized flow can increase the rate of generation of momentum and energy compared to a continuous jet, and furthermore that the droplet AR controls the amount of increase. Monday, November 24, 2014 8:00AM - 10:10AM — Session G14 Drops: Bouncing, Impact and Dynamic Surface Interactions II Lian, University of Louisville 3009/3011 - Yongsheng 8:00AM G14.00001 Resonant and antiresonant bouncing droplets , MAXIME HUBERT, DAMIEN ROBERT, HERVÉ CAPS, STÉPHANE DORBOLO, NICOLAS VANDEWALLE, University of Liège, GRASP TEAM — Droplets may bounce on a oscillating liquid surface. Because of the regeneration of the air layer between the drop and the surface, the bouncing droplets may never coalesce with the liquid underneath. We propose here a simple model for millimetric droplets of low viscosity, bouncing on a highly viscous bath. This model consists of two masses, linked together by a spring and a damper, bouncing onto a rigid and oscillating surface. We use this model to understand the shape of the bouncing threshold, the minimal bath acceleration required to sustain the bouncing dynamics. We show that the droplet is submitted to resonant and anti-resonant behaviors. We also show that those two phenomena are size-dependant and do not occur at the same frequencies for droplets of different radii. By means of resonance and antiresonance, we propose a new microfluidic technique to control precisely the droplets size, which only relies on the surface forcing parameters. 8:13AM G14.00002 Deceleration of free aqueous droplets skirting across the surface of a pool of the same fluid , JACOB HALE, CALEB AKERS, DePauw University — The non-coalescence of a free droplet atop a pool of the same fluid can be greatly enhanced when the drop has an initial horizontal velocity relative to the pool surface. The glancing impact and viscous interaction between the droplet and the pool impart a significant rotation to the droplet causing it to roll and thus entraining air between the two fluids. The translational speed of such a droplet is shown to decrease exponentially in time but with a time constant that increases linearly in time. This complex deceleration of the drop is in part due to the drop’s rotational deceleration, visualized with suspended, neutrally buoyant microbeads. The observed motion is described in terms of viscous dissipation of the rotating drop and a viscous shear force between the droplet and bath. 8:26AM G14.00003 Dynamics of bouncing droplets in annular cavities , ZACHARY LOUIS LENTZ, Department of Mechanical Engineering, UC Berkeley, MIR ABBAS JALALI, Department of Astronomy, UC Berkeley, MOHAMMAD-REZA ALAM, Department of Mechanical Engineering, UC Berkeley — In a cylindrical bath of silicon oil, vertically excited by a frequency of 45 Hz, we trace the motion of bouncing droplets as they fill an annular region. We compute the mean tangential and radial velocity components of the droplets and show that the maximum tangential velocity is larger than the maximum radial velocity by one order of magnitude. Velocity dispersions have almost equal levels in the radial and tangential directions, and their mean values are 1/4 times smaller than the mean tangential velocity. These results show that bouncing droplets undergo random motions within annular cavities determined by the interference patterns of self-induced circumferential waves. We derive analytical relations between the velocity dispersion and the wavelength of surface waves, and calculate the mean tangential velocity of droplets using the random kicks that they experience at the boundaries of the cavity by inward and outward traveling waves. 8:39AM G14.00004 Dynamic trapping of sliding drops on wetting defects , ANDREA CAVALLI, University of Twente, MICHIEL MUSTERD, TU Delft, DIETER ’T MANNETJE, DIRK VAN DEN ENDE, FRIEDER MUGELE, University of Twente — We present a numerical analysis of the dynamic interaction of a sessile sliding drop with a wetting defect. Our three-dimensional model, developed with OpenFOAM, allows us to describe inertial and viscous effects, as well as the internal degrees of freedom of the droplet. We observe that the ability of a drop to deform and stretch enhances the strength and range of the wetting defect, in comparison to a simplified analytic description. We further investigate the role of the strength, size and steepness of the defect in retaining the drop. Finally, we compare our simulations with trapping experiments on electrowetting obstacles, observing a quantitative agreement. This shows that the trapping of sliding drops follows a universal behavior, which is not significantly affected by the nature of the defect. 8:52AM G14.00005 Coalescence-induced jumping motion on non-wetting surfaces: The mechanism of low energy conversion efficiency , FANGJIE LIU, Department of Mechanical Engineering and Materials Science, Duke University, Durham, NC 27708, GIOVANNI GHIGLIOTTI , JAMES J. FENG, Department of Mathematics, University of British Columbia, Vancouver, BC, Canada V6T 1Z2, CHUAN-HUA CHEN, Department of Mechanical Engineering and Materials Science, Duke University, Durham, NC 27708 — When two drops coalesce on a non-wetting substrate such as a superhydrophobic surface, the merged drop spontaneously jumps away from the surface. The self-propelled jumping is powered by the surface energy released upon drop coalescence. However, the translational kinetic energy associated with the jumping is much smaller than the released surface energy. The mechanism of this low energy conversion efficiency is elucidated with 3D phase field simulations which have been experimentally validated. The coalescing drops can be viewed as a two-lobed perturbation to the merged drop with a larger equilibrium radius. The large-amplitude perturbation induces the capillary-inertial oscillation of the merged drop, and the symmetry of the oscillation is broken by the non-wetting substrate. Since the substrate intercepts only a small fraction of the merged drop, a small translational momentum is imparted by the symmetry-breaking substrate, giving rise to the low jumping velocity of 0.2 when nondimensionalized by the capillary-inertial velocity and consequently a low energy conversion efficiency of less than 4%. Other than this small fraction of translational kinetic energy, the majority of the kinetic energy is oscillatory and eventually dissipated. 9:05AM G14.00006 Wave-driven Frictionless Stick-Slip of a self-propelled drop , VINCENT BACOT, MATTHIEU LABOUSSE, Institut Langevin ESPCI Paristech, STÉPHANE PERRARD, YVES COUDER, Laboratoire Matière et Systèmes Complexes, UMR 7057 CNRS et Université Paris Diderot, EMMANUEL FORT, Institut Langevin ESPCI Paristech — Mechanical parts often move not smoothly but in jerks. This well-known phenomenon, called Stick-Slip, is usually associated with friction, since sudden variations of the friction between solid objects, associated with microscopic shear storage, account for the stop and go motion. We report the observation of a similar phenomenon, not controlled by friction, but by propulsion, through the storage of propulsion sources in a surface wave. Our system is made of a liquid droplet, bouncing on a vibrated bath of the same liquid. The vibration of the bath both inhibits the drop coalescence and sustains standing waves on the bath surface around each bouncing spot, for a certain amount of time. These standing waves act as propulsion sources for the drop which thus moves in the horizontal plane. We show how the dynamic interferences of the standing waves created along the way lead to transitory jerks upon starting the motion of the drop. We show how the inhomogeneous dynamic storage of propulsion sources can lead to a steady stop and go motion. In this wave piloted stick-slip motion, the roles of the propulsion and friction are reversed and the spatio-temporal periodicity is tunable. 9:18AM G14.00007 Air mediated dynamics of droplet impact on a smooth, solid surface1 , JOHN KOLINSKI, Harvard School of Engineering and Applied Sciences; Department of Physics of Complex Systems, Weizmann Institute of Science, L. MAHADEVAN, Harvard School of Engineering and Applied Sciences; Harvard University Department of Physics, SHMUEL RUBINSTEIN, Harvard School of Engineering and Applied Sciences; Department of Physics of Complex Systems, Weizmann Institute of Science — Before a falling drop can contact a solid surface, it must displace the air beneath it. Recent calculations and experiments show that as the drop approaches the surface, the air fails to drain, and instead compresses. As the air compresses, the pressure in the gas layer deforms the surface of the drop, thus inhibiting liquid-solid contact. Ultimately, the liquid droplet skates over a nanometer-thin film of air at a strikingly high velocity. These dynamics take place at fleeting timescales and diminutive length-scales, and are obscured by the bulk of the drop, making experimental observation difficult. We directly image the dynamics of the liquid-air interface, and use a novel form of TIR microscopy to study the dynamics and stability of the thin film of air beneath the drop. We show that the stability of the air film governs a novel transition in droplet impact events. 1 NSF GRFP, ISF grant number 1415/12 and Harvard MRSEC (DMR-0820484) 9:31AM G14.00008 Lift-Off Instability During the Impact of a Drop on a Solid Surface , SHMUEL RUBINSTEIN, Harvard University, JOHN KOLINSKI, HUJI, L. MAHADEVAN, Harvard University — We directly measure the rapid spreading dynamics succeeding the impact of a droplet of fluid on a solid, dry surface. Upon impact, the air separating the liquid from the solid surface fails to drain and wetting is delayed as the liquid rapidly spreads outwards over a nanometer thin film of air. We show that the approach of the spreading liquid front toward the surface is unstable and the spreading front lifts off away from the surface. Lift-off ensues well before the liquid contacts the surface, in contrast with prevailing paradigm where lift-off of the liquid is contingent on solid-liquid contact and the formation of a viscous boundary layer. Here I will discuss the dynamics of liquid spreading over a thin film of air and its lift-off away from the surface over a large range of fluid viscosities. 9:44AM G14.00009 Droplet impact patterns on inclined surfaces with variable properties1 , MICHAEL LOCKARD, G. PAUL NEITZEL, MARC K. SMITH, Georgia Inst of Tech — Bloodstain pattern analysis is used in the investigation of a crime scene to infer the impact velocity and size of an impacting droplet and, from these, the droplet’s point and cause of origin. The final pattern is the result of complex fluid mechanical processes involved in the impact and spreading of a blood drop on a surface coupled with the wetting properties of the surface itself. Experiments have been designed to study these processes and the resulting patterns for the case of a single Newtonian water droplet impacting a planar, inclined surface with variable roughness and wetting properties. Results for Reynolds numbers in the range of (9,000 – 27,000) and Weber numbers in the range of (300 – 2,600) will be presented. Transient video images and final impact patterns will be analyzed and compared with results from traditional bloodstain pattern-analysis techniques used by the forensics community. In addition, preliminary work with a new Newtonian blood simulant designed to match the viscosity and surface tension of blood will be presented. 1 Supported by the National Institute of Justice. 9:57AM G14.00010 Shaping/Launching Droplets Impacting on Wettability-Patterned Surfaces , M. ELSHARKAWY, University of Illinois at Chicago, T. SCHUTZIUS, Eidgenössische Technische Hochschule Zürich, G. GRAEBER, RWTH Aachen, J. ORELUK, University of California Berkeley, R. GANGULY, Jadavpur University, C. MEGARIDIS, University of Illinois at Chicago — We present experimental results of droplet impact on wettability-patterned surfaces specifically designed to perform various liquid handling tasks. Such surfaces are implemented for converting droplets from spheres to complex shapes (e.g., annuli or squares) and laterally launching the droplets even under orthogonal impact. The procedure harnesses the naturally occurring contact line pinning mechanisms at sharp wettability changes to influence droplet impact outcome, or even mobilize the fluid asymmetrically post impact. In the launching scenario, droplets impact orthogonally on a superhydrophobic surface and come in contact with a patterned hydrophilic region upon maximum spread. The end result is the launch of the droplet in the lateral direction due to contact line pinning on the hydrophilic region, and the resultant asymmetric disruption of the receding droplet dynamics. We analyze this phenomenon and explain the underlying physical conditions driving the lateral launch post impact. For such patterned surfaces, we show that there exist three possible regimes of dewetting, which depend on total contact time and location of droplet launch from the point of first contact. A model is put forth that predicts the horizontal launch velocity and relates the forces at play to discern between the three dewetting regimes. The study presents a new approach to control impacting droplet outcome and offers new insight on droplet impact behavior on wettability patterned surfaces. Monday, November 24, 2014 8:00AM - 9:57AM Session G15 Jets, Bridges and Rivulets — 3022/3024 - Alfonso Ganan-Calvo, Universidad de Sevilla 8:00AM G15.00001 Post-breakup solutions of Navier-Stokes and Stokes threads , JENS EGGERS, Univ of Bristol — We consider the breakup of a fluid thread, neglecting the effect of the outside fluid (or air). After breakup, the solution of the fluid equations consists of two threads, receding rapidly from the point of breakup. We show that the bulk of each thread is described by a similarity solution of slender geometry (which we call the thread solution), but which breaks down near the tip. Near the tip of the thread the thread solution can be matched to a solution of Stokes’ equation, which consists of a finger of constant spatial radius, rounded at the end. Very close to breakup, the thread solution balances inertia, viscosity, and surface tension (Navier-Stokes case). If however the fluid viscosity is large (as measured by the dimensionless Ohnesorge number), some time after breakup the thread solution consists of a balance of surface tension and viscosity only (Stokes case), and the thread profile can be described analytically. 8:13AM G15.00002 Dripping dynamics at high Bond numbers1 , MARIANO RUBIO-RUBIO, PALOMA TACONET, ALEJANDRO SEVILLA, Departamento de Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, Spain — When dispensing liquid from a vertically oriented injector under gravity, drops grow at the outlet until the surface tension forces can no longer balance their weight, and the pinch-off occurs. This dripping regime no longer exists above a critical flow rate, at which an abrupt transition to jetting takes place. These phenomena are governed by the liquid properties, the injector size and the injection flow rate, or non-dimensionally, by the Bond, Bo, Weber, W e, and Kapitza, Γ, numbers. Detailed accounts of the rich dynamics of the dripping regime and the transition leading to jetting can be found in the literature (e.g. Phys. Rev. Lett. vol. 93, 2004, and Phys. Fluids vol. 18, 2006), but only for two different values of Bo. Therefore, we present new experiments on the dripping dynamics and jetting transition for a wide range of both the liquid viscosity and the size of the injector, reaching values of Bo up to one order-of-magnitude larger than those present in the literature. Our results reveal the existence of new dynamics in the dripping regime not observed at small Bond numbers. In addition, we quantify the hysteresis present in the dripping-jetting transition, previously measured only for the inviscid case. 1 Supported by Spanish MINECO under project DPI 2011-28356-C03-02. 8:26AM G15.00003 Pinch-off of threads of nonhomogeneous Polymer solutions , VISHRUT GARG, SUMEET THETE, SANTOSH APPATHURAI, Purdue Univ, PRADEEP BHAT, The 3M Company, OSMAN BASARAN, Purdue Univ — Motivated by applications involving inkjet printing of complex fluids, we analyze the nonlinear dynamics of the deformation and breakup of polymeric liquid threads. Virtually all previous such studies have been restricted to situations in which the polymer concentration is uniform within the threads. Recently, Eggers (2014) has proposed that non-uniform polymer concentration can account for the blistering pattern that is sometimes seen during breakup of polymeric threads where at the incipience of pinch-off, the thread has the morphology of small drops that are separated by threads of highly concentrated polymer solution. Following Eggers’s approach but one in which he restricted his study to a linear stability analysis, we analyze the full nonlinear dynamics by solving simultaneously Cauchy’s momentum equation, the continuity equation, a convection-diffusion equation for the number density of polymers, and a constitutive equation for stress. The latter two equations account for the coupling between polymer concentration and the flow. As the thread profiles seen in experiments are typically quite slender, we expedite the analysis by solving these equations in the slender-jet limit by an approach based on the finite element method. 8:39AM G15.00004 The effect of a shear boundary layer on the stability of a capillary jet1 , ALFONSO GANAN-CALVO, Universidad de Sevilla, JOSE M. MONTANERO, Universidad de Extremadura, MIGUEL A. HERRADA, Universidad de Sevilla — The generic stabilization effect of a shear boundary layer over the free surface of a capillary jet is here studied from analytical (asymptotic), numerical and experimental approaches. In first place, we show the consistency of the proposed asymptotic analysis by a linear stability (numerical) analysis of the Navier-Stokes equations for a finite boundary layer thickness. We show how the convective-to-absolute instability transition departs drastically from the flat velocity profile case as the axial coordinate becomes closer to the origin of the boundary layer development. For large enough axial distances from that origin, Rayleigh’s dispersion relation is recovered. A collection of experimental observations is analyzed from the perspective provided by these results. We propose a systematic framework to the dynamics of capillary jets issued from a nozzle, either by direct injection into a quiescent atmosphere or in a co-flow (e.g. gas flow-focused jets), which exhibit peculiarities now definitely attributable in first order to the formation of shear boundary layers. 1 Partial support from the Ministry of Economy and Competitiveness, Junta de Extremadura, and Junta de Andalucia (Spain) through Grant Nos. DPI2010-21103, GR10047, P08-TEP-04128, and TEP-7465, respectively, is gratefully acknowledged. 8:52AM G15.00005 Energy-based Classification of Liquid Jet Dynamics: Experiments and Theory , BOWEN LING, ILENIA BATTIATO, San Diego State University — Experiments and dimensional analysis are used to study the dynamics of Newtonian fluid-air jets. Under quasi-static experimental conditions, new periodic phenomena are first captured by image recognition techniques. The former processes can be described through a newly defined dimensionless modified capillary number. We perform fit-free numerical simulations of appropriately simplified Navier-Stokes equations in different dynamical regimes. Results match the corresponding experimental data and are able to capture important dynamic properties of the system, including dripping-jetting and steady-chaos transitions. 9:05AM G15.00006 Breakup length of harmonically stimulated capillary jets – theory and experiments1 , FRANCISCO JAVIER GARCIA GARCIA, HELIODORO GONZALEZ GARCIA, University of Seville, JOSE RAFAEL CASTREJON-PITA, University of Cambridge, ALFONSO ARTURO CASTREJON-PITA, University of Oxford — A stream of liquid breaks up into several drops by the action of surface tension. Capillary breakup forms the basis of some modern digital technologies, especially inkjet printing (including 3D manufacturing). Therefore, the control and prediction of the breakup length of harmonically modulated capillary jets is of great importance, in particular in Continuous InkJet systems (CIJ). However, a theoretical model that rigorously takes into account the physical characteristics of the system, and that properly describes this phenomenon did not exist until now. In this work we present a simple transfer function, derived from first principles, that accurately predicts the experimentally obtained breakup lengths of pressure-modulated capillary jets. No fitting parameters are necessary. A detailed description of the theoretical model and experimental setup will be presented. 1 Spanish government (FIS2011-25161), Junta de Andalucia (P09-FQM-4584 and P11-FQM-7919), EPSRC-UK (EP/H018913/1), Royal Society and John Fell Fund (OUP). 9:18AM G15.00007 Laser-induced jet formation in liquid films , FREDERIK BRASZ, CRAIG ARNOLD, Princeton University — The absorption of a focused laser pulse in a liquid film generates a cavitation bubble on which a narrow jet can form. This is the basis of laser-induced forward transfer (LIFT), a versatile printing technique that offers an alternative to inkjet printing. We study the influence of the fluid properties and laser pulse energy on jet formation using numerical simulations and time-resolved imaging. At low energies, surface tension causes the jet to retract without transferring a drop, and at high energies, the bubble breaks up into a splashing spray. We explore the parameter space of Weber number, Ohnesorge number, and ratio of film thickness to maximum bubble radius, revealing regions where uniform drops are transferred. 9:31AM G15.00008 Stability of a rivulet in a co-flowing microchannel1 , MIGUEL A. HERRADA, AHMED S. MOHAMED, Universidad de Sevilla, JOSE M. MONTANERO, Universidad de Extremadura, ALFONSO GANAN-CALVO, Universidad de Sevilla — We here analyze the stability of a gas (liquid) rivulet on a hydrophobic (hydrophilic) strip along one of the inner sides of a quadrangular microfluidic channel where a liquid (gas) co-flows. The results essentially differ from those of co-flowing cylindrical capillary jets because the contact-line-anchorage conditions affect the rivulet’s instability nature. The temporal stability analysis shows that the rivulet becomes unstable not only for (unperturbed) contact angles larger than 90◦ (as can be expected) but also for values smaller than that angle. The maximum growth factor exhibits a non-monotonic dependence with respect to the Reynolds number (i.e., the viscosities). In fact, there are intervals of that parameter where the fluid system becomes unstable, while all the perturbations are damped outside that interval. The gaseous rivulet does not stabilize as the Reynolds number decreases, which means that it can be unstable even in the Stokes limit and for contact angles less than 90◦ . In addition, the stability of a flowing liquid rivulet is not determined by its contact angle exclusively (as occurs in the static case), but by the Reynolds number as well. Liquid rivulets with contact angles less than 90? can be unstable for sufficiently high Reynolds numbers. 1 Partial support from the Ministry of Science and Education, Junta de Extremadura, and Junta de Andalucı́a (Spain) through Grants Nos. DPI201021103, GR10047, and P08-TEP-04128, respectively, is gratefully acknowledged. 9:44AM G15.00009 Fast liquid transfer between two surfaces , HUANCHEN CHEN, TIAN TANG, University of Alberta, ALIDAD AMIRFAZLI, York University — Liquid transfer process between two surfaces typically ends by breaking of a stretched liquid bridge. The amount of liquid remaining on each of the surfaces relative to total volume is usually of interest in applications (e.g. offset or electronic printing, wet adhesion systems, etc.). Literature shows that depending on stretching velocity, U , surface wettability and liquid properties, the behaviour of the liquid bridge can be categorized into: quasi-static where the surface force dominates and dynamic where contributions from viscous and inertia forces are not negligible. Through a systematic experimental study, we demonstrate for the first time that the division of liquid between surfaces in the quasi-static regime is a constant which depends on the receding contact angles. In the dynamic regime (fast transfer), liquid division takes a complicated form. An analytical-empirical model is developed and verified by experimental results that can predict splitting of the liquid between two surfaces as a function of U, surface wettability and liquid viscosity. The model also successfully predicts an even split between surfaces at extremely high velocities as it was observed by us and others. Monday, November 24, 2014 8:00AM - 10:10AM Session G16 Free-Surface Flows IV: Instability — 2000 - Henri Lhuissier, University of Paris 8:00AM G16.00001 The effect of noncondensables on thermocapillary-buoyancy convection , TONGRAN QIN, ROMAN GRIGORIEV, Georgia Institute of Technology — We consider convection in a layer of volatile simple fluid with free surface subject to a horizontal temperature gradient in the presence of noncondensable gases, such as air, and driven by a combination of buoyancy and thermocapillary stresses. At ambient conditions a unicellular base flow becomes unstable as the temperature gradient is increased, developing a multicellular structure. Recent experimental studies showed that the composition of the gas phase has a significant effect on the convection pattern. In particular, although varying the average concentration of noncondensables over an experimentally accessible range has almost no effect on the average flow speed, the transition to multicellular convection is significantly delayed when noncondensables are evacuated. Using a combination of numerical simulations and linear stability analysis which account for heat and mass transport in the gas phase we show that this dependence is due mainly to the changes in thermocapillary stresses which are controlled by the variation in the composition of the gas phase that arises in response to evaporation and condensation. 8:13AM G16.00002 The effect of noncondensables on the stability of buoyancy-thermocapillary convection1 , YAOFA LI, ROMAN GRIGORIEV, MINAMI YODA, Georgia Institute of Technology — Buoyancy-thermocapillary convection is a well- known problem that is also of interest in evaporative cooling. Our fundamental understanding of convection and transport in the presence of phase change remains limited, however. Pathline visualizations and PIV were used to study convection in a confined layer of a pure volatile 0.65 cSt silicone oil driven by a horizontal temperature gradient at Marangoni numbers M a < 103 and Bond numbers BoD = O(1) below a sealed vapor space containing noncondensables (i.e., air) at concentrations ca = 11 mol% − 96%. At ca = 96% (i.e., ambient conditions), the results are in qualitative agreement with previous studies and a new linear stability analysis, with transitions from steady unicellular to partial multicellular to steady multicellular flow, then to oscillatory multicellular (OMC) flow as M a increases. In the OMC state, the cells oscillate near the heated end, but travel instead towards the cooled end. The results show that decreasing ca has a marked effect on the flow stability, increasing the critical M a for transition between different flow states. Indeed, only steady unicellular and partial multicellular flow states are observed at ca = 11% for these M a. 1 Supported by ONR. 8:26AM G16.00003 Two-dimensional Faraday instability with a spatially periodic bottom1 , NICOLAS PERINET, CLAUDIO FALCON, DFI-FCFM-Universidad de Chile, SEUNGWON SHIN, Hongik University, JALEL CHERGUI, DAMIR JURIC, LIMSICNRS — We study two-dimensional Faraday waves in a channel with rectangular obstacles on the lower boundary, varying the height and the length of the obstacles as well as the distance separating them to understand their influence on the wave patterns. The analysis is mainly numerical and performed by means of BLUE, a code based on a hybrid Front-Tracking/Level-set algorithm for Lagrangian tracking of arbitrarily deformable phase interfaces. In the absence of obstacles, the bifurcation diagram shows three distinct instabilities: the classical instability that leads to the formation of patterns, the sudden onset of temporal chaos and finally a high jump in the amplitude of the waves, the latter bifurcation showing hysteresis. We show that the presence of obstacles delays the primary threshold, inhibits the secondary instabilities and enriches the dynamics of the interface. In particular, obstacles add a new spatial large-scale stationary mode and harmonics resulting from its interaction with the classical resonant modes. 1 Powered@NLHPC: This research was partially supported by the supercomputing infrastructure of the NLHPC (ECM-02). Nicolas Perinet has received grant from CONICYT - FONDECYT/postdoctoral research project 3140522 8:39AM G16.00004 Faraday instability in deformable domains1 , GIUSEPPE PUCCI, Matières et Systèmes Complexes, Université Paris Diderot, CNRS UMR 7057, 10 rue Alice Domon et Léonie Duquet, 75205 Paris, France, MARTINE BEN AMAR, Laboratoire de Physique Statistique, Ecole Normale Superieure, UPMC Univ Paris 06, Universite Paris Diderot, CNRS, 24 rue Lhomond, 75005 Paris, France, YVES COUDER, Matières et Systèmes Complexes, Université Paris Diderot, CNRS UMR 7057, 10 rue Alice Domon et Léonie Duquet, 75205 Paris, France — We investigate the Faraday instability in floating liquid lenses, as an example of hydrodynamic instability that develops in a domain with flexible boundaries. We show that a mutual adaptation of the instability pattern and the domain shape occurs, as a result of the competition between the wave radiation pressure and the capillary response of the lens border. Two archetypes of behaviour are observed. In the first, stable shapes are obtained experimentally and predicted theoretically as the exact solutions of a Riccati equation, and they result from the equilibrium between wave radiation pressure and capillarity. In the second, the radiation pressure exceeds the capillary response of the lens border and leads to non-equilibrium behaviours, with breaking into smaller domains that have a complex dynamics including spontaneous propagation. 1 The authors are grateful to Université Franco-Italienne (UFI) for financial support. 8:52AM G16.00005 Instability of the capillary bridge , GOUNSETI PARE, JEROME HOEPFFNER, Institut Jean Le Rond d’Alembert — Capillary adhesion is a physical mechanism that maintains two bodies in contact by capillarity through a liquid ligament. The capillary bridge is an idealization of this capillary adhesion. In this study we first focus on the classical case of the stability of the capillary bridge. Secondly we study a slightly more complex configuration, imagining a flow in the capillary bridge as in the case of the dynamics of the neck of a liquid ligament, in its withdrawal under the effect of capillarity. Inspired by the experiments on soap films of Plateau, the configuration analyzed consists of an initially axisymmetric, mass of fluid held by surface tension forces between two parallel, coaxial, solid pipes of the same diameter. The results presented are obtained by numerical simulations using the free software, Gerris Flow Solver. We first focus on the capillary Venturi. In the static configuration the stability diagram of the capillary bridge obtained is in perfect agreement with the results of Lev A. Slobozhanin. In the dynamic case we develop a matlab code based on the one dimensional equations of Eggers and Dupont. The comparison of the bifurcation diagram obtained and the numerical simulations shows a good agreement. 9:05AM G16.00006 Stability of an unsupported multi-layer surfactant laden liquid curtain under gravity1 , DOMINIC HENRY, JAMAL UDDIN, University of Birmingham, UK, JEREMY MARSTON, Texas Tech University, USA, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology, KSA — The industrial process of curtain coating has long been an important method in coating applications, by which a thin liquid curtain is formed to impinge upon a moving substrate, the highly lucrative advantage being able to coat multiple layers simultaneously. We investigate the linear stability of an unsupported two-layer liquid curtain, which has insoluble surfactants in both liquids. We formulate the governing equations, simplified by making a thin film approximation, from which we obtain equations describing the steady state profiles. We then examine the response of the curtain to small perturbations about this steady state to identify conditions under which the curtain is unstable, finding the addition of surfactants stabilizes the curtain. Our results are then compared to experimental data, showing a favourable trend and therefore extending the work of Brown2 and Dyson et al.3 1 D.H. would like to thank EPSRC for their financial support and KAUST for funding the experimental work. Brown, J. Fluid Mech. 10, 297-305 (1960). 3 R.J. Dyson, J. Brander, C.J.W. Breward and P.D. Powell, J. Eng. Math. 64, 237-250 (2009). 2 D. 9:18AM G16.00007 Polygons in a Liquid Metal Free Surface Driven by Rotating Permanent Magnets1 , SERGIO CUEVAS, J. CARLOS DOMINGUEZ-LOZOYA, Universidad Nacional Autónoma de México, MICHEL RIVERO, Instituto Tecnologico de la Laguna, EDUARDO RAMOS, Universidad Nacional Autónoma de México — We report the appearance of an instability in a shallow liquid metal layer (GaInSn) driven by different arrays of rotating magnetized bars (6 cm × 1.27 cm × 1.27 cm) located at the bottom of a cylindrical plexiglas container with a diameter of 20.32 cm. The thickness of the fluid layer is 0.6 cm and the maximum analyzed rotation frequency is 7 Hz. We explored arrays with one, three, four, and five magnet bars placed radially and equidistantly. For specific magnet rotation frequencies, we observed the spontaneous breaking of the axial symmetry of the free surface which takes the form of an ellipse for the case of one rotating magnet, or a polygon with three, four, or five corners for the cases of three, four or five rotating magnets, respectively. The structures rotate uniformly with a speed about an order of magnitude lower that the rotating magnets. Similarities with instabilities observed with free surface hydrodynamic flows driven by a rotating bottom plate are discussed. 1 Work supported by CONACYT, Mexico under Project 131399. 9:31AM G16.00008 Impact of a viscoelastic jet , HENRI LHUISSIER, BAPTISTE NÉEL, LAURENT LIMAT, MSC, Université Paris Diderot — A jet of a Newtonian liquid impacting onto a wall at right angle spreads as a thin liquid sheet which preserves the radial symmetry of the jet. We observe that for a viscoelastic jet (solution of polyethylene glycol in water) this symmetry can break: close to the wall, the jet cross-section is faceted and radial steady liquid films (membranes) form, which connect the cross-section vertices to the sheet. The number of membranes increases with increasing viscoelastic relaxation time of the solution, but also with increasing jet velocity and decreasing distance from the jet nozzle to the wall. A mechanism for this surprising destabilization of the jet, which develops perpendicularly to the direction expected for a buckling mechanism, is presented that explains these dependences. The large-scale consequences of the jet destabilization on the sheet spreading and fragmentation, which show through the faceting of hydraulic jumps and suspended (Savart) sheets, will also be discussed. 9:44AM G16.00009 Experiments and non-parallel theory on the natural break-up of freely falling Newtonian liquid jets1 , PAULA CONSOLI-LIZZI, WILFRIED COENEN, ALEJANDRO SEVILLA, Dept. Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, Spain — The capillary break-up of liquid jets issuing from a needle at a constant flow rate is studied experimentally and theoretically. In particular, we focus on globally stable jets of a Newtonian liquid that are strongly stretched by gravity, so that the region close to the injector is highly non-parallel. In this regime, the use of parallel linear stability theory, based on a local dispersion relation between the frequency and the wavelength of travelling-wave disturbances, is questionable. We therefore propose a global linear frequency response analysis based on a one-dimensional formulation of the mass and momentum equations. Our model reveals that perturbations present large damping in the initial region of strong axial stretching, followed by a growth that eventually causes the break-up of the jet. Besides the break-up length, our model also allows for the prediction of the most amplified frequency. The theoretical predictions are compared with experimental observations, that comprise the natural break-up of stretched jets for a wide range of liquid viscosities, injector radii and flow rates. 1 Supported by Spanish MINECO under project DPI 2011-28356-C03-02 9:57AM G16.00010 Rill patterning on sloping snowpacks induced by Hortonian runoff , ELISA MANTELLI, CARLO CAMPOREALE, LUCA RIDOLFI, Politecnico di Torino — The morphological instability leading to rill formation over snowpacks is addressed in the present study. First, Hortonian saturation of a surface thin layer of snow is demonstrated to occur during the rising-intensity stage of rainfall events because the velocity of the water wavefront in the unsaturated snow is proportional to rainfall intensity. Therefore, a slowly downward-propagating shockwave is formed, behind which Hortonian saturation eventually occurs, and a turbulent water film moving along the maximum slope direction is allowed to develop above the snowpack surface. The linear stability analysis of the system made up of the water film and the saturated snow layer is then performed, and the dispersion relation obtained analytically. A spanwise morphological instability corresponding to rills is detected and investigated as a function of slope, friction coefficient, Reynolds number and wavenumber. The maximum instability wavelength is shown to have a purely hydrodynamic origin and to be originated by the interplay between pressure perturbation, free surface response and Reynolds stresses. Field work has been also performed, that confirms the validity of the presented model. Monday, November 24, 2014 8:00AM - 10:10AM Session G17 Nonlinear Dynamics III: Chaos — 2002 - Kevin A. Mitchell, University of Califonia, Merced 8:00AM G17.00001 Weakly Nonlinear Analysis and Chaotic Growth in Nanofilm Flows Directed by Thermocapillary Forces , CHENGZHE ZHOU, SANDRA M. TROIAN, California Institute of Technology, MC 128-95, Pasadena, CA 91125 — We examine the nonlinear response of nanofilm flows subject to interface deformation and patterned growth by thermocapillary and capillary forces. The governing interface evolution equation describes growth induced by an initial uniform and tranverse thermal gradient in the long wavelength limit.1,2 A bifurcation analysis via the method of multiple scales elucidates the influence of initial conditions, system geometry and material properties on the regions of stable and unstable flow. Investigation of the corresponding Ginzburg-Landau amplitude equation by finite element simulations reveals the existence of interesting spatiotemporal chaotic behavior during the later stages of patterned growth. Time permitting, we will discuss the possibility of tightly ordered symmetric growth by mode locking to spatially periodic external forcing,3 in analogy to behavior recently reported for the spatially forced Swift-Hohenberg equation in 1- and 2dimensions.4 1 M. Dietzel and S. M. Troian, Phys. Rev. Lett. 103 (7), 074501 (2009) Dietzel and S. M. Troian, J. Appl. Phys. 108, 074308 (2010) 3 N. Liu and S. M. Troian, preprint (2013) 4 Y. Mau, L. Haim, A. Hagberg and E. Meron, Phys. Rev. E 88, 032917 (2013) 2 M. 8:13AM G17.00002 Using tangles to quantify topological mixing of fluids1 , QIANTING CHEN, SULIMON SATTARI, KEVIN MITCHELL, Univ of California - Merced — Topological mixing is important in understanding complex fluid problems, ranging from oceanic transport to the design of micro-mixers. Typically, topological entropy (TE), the exponential growth rate of material lines, is used to quantify topological mixing. Computing TE from the direct stretching rate is computationally expensive and sheds little light on the source of the mixing. Previous work has focused on braiding by “ghost rods” (See, e.g. works by Boyland, Aref, Stremler, Tiffeault, and Finn). Following Grover et al. [Chaos 22,043135 (2012)], we study topological mixing in a two-dimensional lid-driven cavity flow. For a certain parameter range, the TE is dominated by a period-3 braid. However, this braid alone cannot explain all the TE within this range, nor the TE outside the range of existence of the braid. By contrast, we explain TE through the topology of intersecting stable and unstable manifolds, i.e. heteroclinic tangles, using homotopic lobe dynamics (HLD). In the HLD approach, stirring originates from “ghost rods” placed on heteroclinic orbits. We demonstrate that these heteroclinic orbits generate excess TE not accounted for in Grover et al. Furthermore, in the limit of utilizing arbitrarily long manifolds, the HLD technique converges to the true TE. 1 Supported by the US National Science Foundation under grant PHY-0748828. 8:26AM G17.00003 Chaotic scalar advection induced by a vortex pair interacting with a fixed vortex , EUGENE RYZHOV, Pacific Oceanological Institute — We examine passive scalar advection caused by a point-vortex pair, a structure consisting of two counter-rotating point vortices with equal strengths, encountering a fixed point vortex in a uniform fluid flow. Such an interaction is known to produce two distinct types of vortex motion. First is an unbounded regime as the pair moves unrestrictedly almost rectilinearly towards infinity after encountering the fixed vortex, although the path the pair moves in may be significantly altered by the interaction. Second is a bounded regime when the pair’s vortices periodically oscillate about the fixed vortex. The latter produces irregular advection of passive scalars as the pair acts as a periodic perturbation for the neighbouring scalars. In the unbounded case, the phenomenon of the exchanging of passive scalars between the pair’s vortex atmosphere, the area the pair drags along during its self-propagation, and the fixed vortex’s vicinity is numerically estimated. In the bounded regime, associated passive scalar transport is examined by means of Poincaré sections that reveal the regions of regular and chaotic transport. 8:39AM G17.00004 Lagrangian transport characteristics of a class of three-dimensional inlinemixing flows with fluid inertia1 , MICHEL SPEETJENS, ESUBALEW DEMISSIE, Eindhoven Univ of Tech, GUY METCALFE, CSIRO Materials Science & Engineering, HERMAN CLERCX, Eindhoven Univ of Tech — Laminar inline mixing is key to many industrial systems. However, insight into fundamental transport phenomena in case of 3D conditions and fluid inertia remains limited. This is studied for inline mixers with a cylindrical geometry. Said effects introduce three key features absent in simplified configurations: smooth transition between mixing cells; local upstream flow; symmetry breaking. Topological considerations imply a net throughflow region strictly separated from possible internal regions. The Lagrangian dynamics in this region admits representation by a 2D time-periodic Hamiltonian system. This establishes one fundamental kinematic structure for the present class of inline-mixing flows and implies universal behavior. All states follow from Hamiltonian breakdown of one common integrable state. Period-doubling bifurcation is the only way to eliminate transport barriers originating from the integrable state and thus necessary for global chaos. Important in a practical context is that a common simplification, i.e. cell-wise developed Stokes flow, retains these fundamental kinematic properties and deviates from the 3D inertial case essentially only in a quantitative sense. This substantiates its suitability for (at least first exploratory) studies on mixing properties. 1 Dutch Technology Foundation grant STW 11054 8:52AM G17.00005 Random fluctuations and resonances in near-integrable flows , DMITRI VAINCHTEIN, Temple University — Resonance phenomena, such as capture into resonance and scattering on resonance, are known to be major contributors to transport and mixing in near-integrable multi-scale flows. The long-time transport properties in such systems are described in terms of the evolution of the certain quantity, called adiabatic invariant. In the present talk we investigate the impact of different random fluctuations on adiabatic transport. This impact manifests itself in two ways: the statistical properties of the diffusion of the adiabatic invariant due to scattering are altered, and the fine properties of capture, such as the probability of capture and the input-output function, may change significantly. Using the Ekman pumping-driven flow in circular cells as example, we investigate the role these phenomena and obtain modifications to long-term diffusion equations derived before. 9:05AM G17.00006 Construction of an Optimal Background Profile for the KuramotoSivashinsky Equation using Semidefinite Programming , ANDREW WYNN, GIOVANNI FANTUZZI, Imperial College London — The Kuramoto-Sivashinsky (KS) equation has been derived in many physical contexts to describe systems whose dynamics are characterised by long-wavelength instability, for example flame-front instabilities or flow stability for thin liquid films on inclined planes. It is known that the KS equation has chaotic solutions if the governing parameter (typically the length L of the domain) is sufficiently large. Furthermore, numerical evidence suggests that the 1 asymptotic energy of the solution scales according to L 2 although, despite much effort, it has not yet been possible to prove such a bound analytically. We present a novel method of proving bounds on the asymptotic energy of the KS equation, by constructing a ‘background profile’ using Semidefinite Programming. The advantage of the method is that the background profile may be searched for automatically by solving a standard optimization problem, while coupling 3 the numerics to a careful mathematical analysis of the PDE ensures that the bounds hold analytically. The obtained scaling of L 2 agrees with the previous best results using the background profile method. Interestingly, the obtained profiles closely resemble the ‘viscous shock’ solutions which are known to exist for destabilized KS equations. 9:18AM G17.00007 Burning invariant manifolds in time-periodic and time-aperiodic vortex flows1 , SAVANNAH GOWEN2 , TOM SOLOMON, Bucknell University — We present experiments that study reaction fronts in a flow composed of a single, translating vortex. The fronts are produced by the excitable Belousov-Zhabotinsky (BZ) chemical reaction, and the vortex flow is driven magnetohydrodynamically by a radial current in a thin fluid layer above a Nd-Fe-Bo magnet. The magnet is mounted on a pair of perpendicular translation stages, allowing for controlled, two-dimensional movement of the magnet and the resulting vortex. We study reaction fronts that pin to the vortex for time-independent flows (produced by moving the vortex with a constant velocity) and for time-periodic and time-aperiodic flows produced by oscillating the vortex laterally. The steady-state front shape is analyzed in terms of burning invariant manifolds3 (BIMs) that act as one-way barriers against any propagating reaction fronts. For time independent and time-periodic flows, the location of the BIMs are calculated numerically and are compared with experimental images of the pinned reaction fronts. We investigate extensions of this BIM approach for analyzing fronts in time-aperiodic flows. 1 Supported by NSF Grants DMR-1004744, DMR-1361881 and PHY-1156964. address: Mt. Holyoke College, S. Hadley, MA. 3 J. Mahoney, D. Bargteil, M. Kingsbury, K. Mitchell and T. Solomon, Europhys. Lett. 98, 44005 (2012). 2 Current 9:31AM G17.00008 Defining Lagrangian coherent structures for reactions in time-aperiodic flows1 , KEVIN MITCHELL, JOHN MAHONEY, University of California, Merced — Recent theoretical and experimental investigations have highlighted the role of invariant manifolds, termed burning invariant manifolds (BIMs), as one-way barriers to reaction fronts propagating through a flowing medium. Originally, BIM theory was restricted to time-independent or time-periodic flows. The present work extends these ideas to flows with a general time-dependence, thereby constructing coherent structures that organize and constrain the propagation of reaction fronts through general flows. This permits a much broader and physically realistic class of problems to be addressed. Our approach follows the recent work of Farazmand, Blazevski, and Haller [Physica D 278-279, 44 (2014)], in which Lagrangian coherent structures (LCSs), relevant to purely advective transport, are characterized as curves of minimal Lagrangian shear. 1 Supported by the US National Science Foundation under grant CMMI-1201236. 9:44AM G17.00009 Chaotic mixing and front propagation in a three-dimensional flow1 , SARAH HOLLER, TOM SOLOMON, Bucknell University — We present experiments on passive mixing and on the behavior of the excitable Belousov-Zhabotinsky (BZ) chemical reaction in a time-independent, three-dimensional (3D) flow. The flow is composed of nested horizontal and vertical chains of vortices, a flow that has been shown2 numerically to produce chaotic mixing with a complicated structure of ordered and chaotic regions. We study mixing experimentally by tracking neutrally-buoyant tracer particles in 3D and by imaging the evolution of a fluorescent dye with a scanning laser technique. The same scanning technique enables us to image fronts of the Ruthenium-catalyzed BZ reaction in the same flow. We analyze the behavior of these fronts with an extension of a theory of burning invariant manifolds 3 that has been shown to predict accurately the locations of barriers that impede reaction fronts in 2D flows. 1 Supported by NSF Grants DMR-1004744, DMR-1361881 and PHY-1156964. Fogleman, M.J. Fawcett and T. H. Solomon, Phys. Rev. E 63, 02101(R) (2001). 3 J. Mahoney, D. Bargteil, M. Kingsbury, K. Mitchell and T. Solomon, Europhys. Lett. 98, 44005 (2012). 2 M.A. 9:57AM G17.00010 Front pinning in single vortex flows , JOHN MAHONEY1 , KEVIN MITCHELL, Univ of California Merced — We study fronts propagating in 2D fluid flows and show that there exist stable invariant front configurations for fairly generic flows. Here we examine the simple flow which combines a single vortex with an overall “wind.” We discuss how the invariant front can be derived from a simple 3D ODE. Existence of this front can then be understood in terms of bifurcations of fixed points, and the behavior of the invariant “sliding front” submanifold. Interestingly, the front bifurcation precedes the saddle-node bifurcation which gives rise to the vortex. This elementary structure has application in chemical reactor beds and laminar combustion in well-mixed fluids. 1 We request that this talk follow the related talks by our collaborators Tom Solomon, Savannah Gowen, and Sarah Holler. Monday, November 24, 2014 8:00AM - 10:10AM Session G18 Vortex Dynamics: Vortex Identification and Mechanisms — 2004 - Karen Mulleners, Das Institut fuer Turbomaschinen und Fluid-Dynamik 8:00AM G18.00001 Quantifying the reconnection process of two vortices1 , GUILLAUME BEARDSELL, U. Laval, Canada, LOUIS DUFRESNE, ETS, U. Quebec, Canada, GUY DUMAS, U. Laval, Canada — In this work, we use DNS to study the reconnection of two vortices. The Navier-Stokes equations are solved using a Fourier pseudospectral algorithm with triply periodic boundary conditions. The zero-circulation constraint, which was found to be problematic by Pradeep & Hussain (2004), is circumvented by solving the governing equations in a proper rotating frame. To quantify the reconnection of two vortices, an approach using vortex filaments is considered. This approach is first validated against the results of Hussain & Duraisamy (2011) for two parallel counter-rotating vortices. In this latter case, symmetries in the initial flow provide a simple way to compute the instantaneous rate of reconnection. Next, we study the interaction of orthogonal, unequal strength vortices for which only partial reconnection can occur. Typically, the weak vortex (Γ2 ) is seen to deform and wrap itself around the strong one (Γ1 ) to (partially) reconnect. For Reynolds numbers (Γ1 /ν) of the order of 103 and circulation ratios 0.1 ≤ Γ2 /Γ1 ≤ 0.9, we compute the instantaneous reconnection rate and observe the propagating vorticity structures. Particularly, we look at some of the topological features that can be well visualized with vortex filaments. 1 Financial support from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Fonds de Recherche du Québec - Nature et Technologies (FRQNT) is gratefully acknowledged by the authors 8:13AM G18.00002 Spectrum and Structure of Evolving Vortex Sheet , V. RAJESH, O.N. RAMESH, Indian Institute of Science — Shear layer structure and dynamics play an important role in understanding turbulent flows. The evolution of the Vortex sheet, considered as the inviscid/infinite Reynolds number approximation to the shear layer, has been studied in the literature for its many facets like Kelvin-Helmholtz instability, finite-time singularity and spiral structures formation. In the present work, a two-dimensional vortex sheet evolution is simulated using Krasny’s vortex blob method in high precision. The nonlinear stage of evolution and the resulting spectrum are validated against the recent results of Abid & Verga (2011). The obtained energy spectrum, showing similarities to two-dimensional turbulence spectrum, is investigated in detail. An attempt at simulating the evolution of three-dimensional disturbances on the vortex sheet and the effect of imposed strain is also presented. 8:26AM G18.00003 Eulerian and Lagrangian methods for vortex tracking in 2D and 3D flows1 , YANGZI HUANG, MELISSA GREEN, None — Coherent structures are a key component of unsteady flows in shear layers. Improvement of experimental techniques has led to larger amounts of data and requires of automated procedures for vortex tracking. Many vortex criteria are Eulerian, and identify the structures by an instantaneous local swirling motion in the field, which are indicated by closed or spiral streamlines or pathlines in a reference frame. Alternatively, a Lagrangian Coherent Structures (LCS) analysis is a Lagrangian method based on the quantities calculated along fluid particle trajectories. In the current work, vortex detection is demonstrated on data from the simulation of two cases: a 2D flow with a flat plate undergoing a 45o pitch-up maneuver and a 3D wall-bounded turbulence channel flow. Vortices are visualized and tracked by their centers and boundaries using Γ1 , the Q criterion, and LCS saddle points. In the cases of 2D flow, saddle points trace showed a rapid acceleration of the structure which indicates the shedding from the plate. For channel flow, saddle points trace shows that average structure convection speed exhibits a similar trend as a function of wall-normal distance as the mean velocity profile, and leads to statistical quantities of vortex dynamics. 1 Dr. Jeff Eldredge and his research group at UCLA are gratefully acknowledged for sharing the database of simulation for the current research. This work was supported by the Air Force Office of Scientific Research under AFOSR Award No. FA9550-14-1-0210 8:39AM G18.00004 Reconstruction of three-dimensional coherent structures in turbulent wakes using planar measurements1 , SERHIY YARUSEVYCH, CHRIS MORTON, University of Waterloo — The present study is focused on reconstructing the dynamics of dominant three-dimensional coherent structures in turbulent wakes of complex cylindrical geometries using time-resolved, planar Particle-Image-Velocimetry data. As a test case, the turbulent wake of a low aspect ratio dual step cylinder model is considered. The model consists of a large diameter cylinder (D) of low aspect ratio (L/D) attached to the mid-span of a small diameter cylinder (d). Experiments are performed in a water flume facility for Re D =2100, D/d =2, and L/D =1. The investigated model produces cellular vortex shedding, with distinct variations in the average shedding frequency along the span of the model, and the associated complex vortex interactions. Time-resolved velocity measurements are acquired simultaneously in two mutually orthogonal planes at multiple planes along the span of the model. The technique involves conditional averaging of the planar results to produce three-dimensional reconstructions of wake topology for a given planar alignment of the dominant spanwise vortex filaments. This is achieved by identifying velocity fields matching a given flow-based template. The results demonstrate that the proposed technique can successfully reconstruct the dominant wake vortex interactions and can be extended to other flows where traditional phase-averaging approaches are not applicable. 1 The authors gratefully acknowledge the Natural Sciences and Engineering Research Council of Canada (NSERC) for funding of this work. 8:52AM G18.00005 The Effect of Phase Averaging Techniques on Lagrangian Coherent Structures in the Wake of a Circular Cylinder1 , MATTHEW ROCKWOOD, MELISSA GREEN, Syracuse University — Experimental results of the wake of a circular cylinder were studied using Lagrangian coherent structures (LCS). The planar velocity data was collected at multiple Reynolds numbers over a range from 3,000 to 12,000 using a two-component DPIV measurement system. The data was phase averaged by binning velocity fields based on two reference quantities: vorticity centroid location in each snapshot, and pressure measurements on the cylinder surface. A Proper Orthogonal Decomposition (POD) was also applied to the velocity data to extract portions of the velocity field containing the most energy. Another set of phase averaged velocity fields were then generated using the vorticity centroid location of the POD reconstructed fields. The change in the LCS locations and the vortices identified using the Eulerian Q-criterion were found to be minimal. This investigation ensures that the most accurate, efficient phase averaging techniques are being used to study the LCS in the wake of the circular cylinder. 1 This work was supported by the Air Force Office of Scientific Research under AFOSR Award No. FA9550-14-1-0210. 9:05AM G18.00006 Identification of Vortex Breakdown in Bio-Inspired Wakes Using Proper Orthogonal Decomposition1 , ZACHARY BERGER, JUSTIN KING, MELISSA GREEN, Syracuse University — In this investigation, the flow field of a bio-inspired wake is studied using stereoscopic PIV at the mid-span and quarter-span of a trapezoidal pitching panel in a water channel. Threecomponent planar velocity fields are generated immediately downstream of the panel. Standard (4 Hz) PIV measurements require phase-averaging to extract relevant flow features with respect to the time scales of the flow. In order to gain more insight into the energy content of the flow field as well as quantitatively identify the vortex breakdown, reduced-order modeling in the form of proper orthogonal decomposition (POD) is applied to the data. Two component vector POD allows for the extraction of the most energetic, large scale structures which can be used to reconstruct a low-dimensional representation of the flow field. This can then be compared to phase-averaged data to readily quantify vortex breakdown as a function of spanwise location in order to construct a model of the three-dimensional wake structure. 1 This work was supported by the Office of Naval Research under ONR Award No. N00014-14-1-0418 9:18AM G18.00007 Helical vortices: viscous dynamics and instability1 , MAURICE ROSSI, UPMC-CNRS, CAN SELCUK, LIMSI, IVAN DELBENDE, LIMSI-CNRS, IJLRA-UPMC TEAM, LIMSI-CNRS TEAM — Understanding the dynamical properties of helical vortices is of great importance for numerous applications such as wind turbines, helicopter rotors, ship propellers. Locally these flows often display a helical symmetry: fields are invariant through combined axial translation of distance ∆z and rotation of angle θ = ∆z/L around the same z-axis, where 2πL denotes the helix pitch. A DNS code with built-in helical symmetry has been developed in order to compute viscous quasi-steady basic states with one or multiple vortices. These states will be characterized (core structure, ellipticity, ...) as a function of the pitch, without or with an axial flow component. The instability modes growing in the above base flows and their growth rates are investigated by a linearized version of the DNS code coupled to an Arnoldi procedure. This analysis is complemented by a helical thin-cored vortex filaments model. 1 ANR HELIX 9:31AM G18.00008 Evolution of mean flow and disturbances in strained vortices , YUJI HATTORI, IFS, Tohoku University — Evolution of disturbed strained vortices is studied by direct numerical simulation. We choose 2D flattened Taylor-Green vortices as a base flow and add a small wave packet which grows exponentially due to the elliptical instability. The evolution consists of three stages: the linear, non-linear, and turbulent stages. At the linear stage the wave packet located initially at the center of a vortex grows exponentially without significant change of the shape. At the nonlinear stage the wave packet collapses and small-scale structures develop. Concentration of vorticity in the mean flow, which is similar to the “expulsion of vorticity” in rotating turbulence, is observed before the transition to turbulence. Finally the flow becomes turbulent exhibiting the Kolmogorov energy spectrum although the mean flow is not far from the initial state. The mechanism behind the concentration of vorticity will be discussed in connection with angular momentum transfer and selective decay of inviscid invariants. 9:44AM G18.00009 Tomographic PIV Study of Hairpin Vortices1 , DANIEL SABATINO, TOBIAS ROSSMANN, Lafayette College — Tomographic PIV is used in a free surface water channel to quantify the flow behavior of hairpin vortices that are artificially generated in a laminar boundary layer. Direct injection from a 32:1 aspect ratio slot at low blowing ratios (0.1 < BR < 0.2) is used to generate an isolated hairpin vortex in a thick laminar boundary layer (485 < Reδ∗ < 600). Due to the large dynamic range of length and velocity scales (the resulting vortices have advection velocities 5X greater than their tangential velocities), a tailored optical arrangement and specialized post processing techniques are required to fully capture the small-scale behavior and long-time development of the flow field. Hairpin generation and evolution are presented using the λ2 criterion derived from the instantaneous, three-dimensional velocity field. The insight provided by the tomographic data is also compared to the conclusions drawn from 2D PIV and passive scalar visualizations. Finally, the three-dimensional behavior of the measured velocity field is correlated with that of a simultaneously imaged, passive scalar dye that marks the boundary of the injected fluid, allowing the examination of the entrainment behavior of the hairpin. 1 Supported by the National Science Foundation under Grant CBET-1040236 9:57AM G18.00010 Comparing wake structures behind a finite aspect ratio and an infinite span normal thin flat plate1 , ARMAN HEMMATI, DAVID H. WOOD, ROBERT J. MARTINUZZI, University of Calgary — The wake of an infinite span (2D) thin flat plate and that of a finite aspect ratio, AR = 3.2, plate, both normal to a uniform stream, are compared using DNS at Re=1200. For the 2D plate, three wake flow regimes are observed. Intervals of typical anti-symmetric Karman shedding (Regime M) are interrupted by intervals where the shear layer folding process first delayed (Regime L) and then accelerated, Regime H. The distinct flow patterns in these regimes have significant impact on lift and drag values, wake structure and instantaneous pressure loads. In contrast, only Regime M is observed for the AR=3.2 plate. The presence of two lateral shear layers appears to maintain the Karman shedding. Compared to the infinite plate: the mean recirculation region shrinks by 45% to 1.57H; the magnitude of the Reynolds Stresses drops significantly; Turbulent kinetic energy levels along the wake centerline and peak production and dissipation rates are significantly lower. Further, the three normal Reynolds stresses are comparable in magnitude. To better understand the impact of additional shear layers on the wake stability and resultant wake structures, the 3D structures will be reconstructed using DNS results. Pressure and stress distribution along the plate surfaces will also be examined. 1 This work is supported by AITF and NSERC fellowship grants. Monday, November 24, 2014 8:00AM - 10:10AM Session G19 Convection and Buoyancy-Driven Flows: Rotation — 2006 - Jin-Qiang Zhong, Tongji University 8:00AM G19.00001 Influence of the Prandtl number on the heat transport enhancement in rotating turbulent Rayleigh-Bénard convection1 , STEPHAN WEISS, PING WEI, GUENTER AHLERS, University of California, Santa Barbara — We present new Nusselt-number (Nu) measurements for slowly rotating turbulent thermal convection in cylinders with aspect ratio Γ = 1. By using compressed gasses and various liquids, we now have data in the Prandtl number (Pr) range 0.74 <Pr< 35.5 and for Rayleigh numbers (Ra) in the range 4×108 <Ra< 2×1011 . With these data we investigate in detail the effect of Pr and Ra on the heat-transport enhancement close to its onset. This enhancement takes place for rotation rates larger than a critical value, as expressed by the dimensionless inverse Rossby number (1/Ro), since only then vortices form, in which due to Ekman pumping fluid is transported from the thermal boundary layers into the turbulent bulk. We found that the critical inverse Rossby number (1/Roc ) decreases with increasing Pr, following a power law with exponent α = −0.40 ± 0.02. For larger rotation rates, the relative heat transport enhancement (N ur ) increases first linearly with a slope S = ∂N ur /∂(1/Ro). We show that also the slope S follows a power law S ∝ P rβ Raγ with β = −0.10 ± 0.06 and γ = −0.14 ± 0.04. We found that the maximum heat transport enhancement (up to 40% ) increases with increasing Pr and decreasing Ra. 1 This work was supported by NSF-Grant DMR11-58514. SW thanks the Deutsche Forschungsgesellschaft for financial support. 8:13AM G19.00002 Dynamics of the large-scale circulation in turbulent Rayleigh-Bénard convection with modulated rotation1 , JIN-QIANG ZHONG, SEBASTIAN STERL, HUI-MIN LI, Tongji University, Shanghai, China — We present measurements of the azimuthal rotation velocity θ̇ and thermal amplitude δ of the large-scale circulation (LSC) in turbulent Rayleigh-Bénard convection with modulated rotation. Both θ̇ and δ exhibit clear oscillations at the modulation frequency ω. Fluid acceleration driven by oscillating Coriolis force plays a role in determining the LSC rotations and causes an increasing phase lag in θ̇ when ω increases. The applied modulation also produces oscillatory boundary layers and the resulting time-varying viscous drag modifies δ periodically. Oscillation of θ̇ with the maximum amplitude occurs at an intermediate ω ⋆ . Such a resonance-like phenomena is interpreted as a result of the optimal coupling of δ to the sample rotation velocity. We show that an extended LSC model with a relaxation time for δ to response to modulated rotations provides predictions in close agreement with the experimental results. 1 Supported by NSFC Grant 11202151. 8:26AM G19.00003 Retrograde rotation of the large-scale circulation in turbulent rotating Rayleigh-Benard convection at large Rossby numbers up to 2001 , HUI-MIN LI, JIN-QIANG ZHONG, Tongji University, Shanghai, China — We examine the azimuthal rotation of the large-scale circulation (LSC) for turbulent Rayleigh-Benard convection in the present of week rotations about a vertical axis at angular velocities 1.0×10−3 ≤Ω≤0.1(rad/s). Over the entire Rossby-number range 1≤Ro≤200 studied, linear retrograde rotations of the LSC circulating plane are observed. With increasing Ro(∼1/Ω) the retrograde rotating velocity h − θ̇i decreases monotonically, but the ratio γ = h − θ̇i/Ω experiences a transition at Ro⋆ ≈80 above which γ increases sharply. We discuss the Ro-dependence of γ for Ro > Ro⋆ and show that a maximum ratio γmax = 0.36 is observed at Ro = 200, more than twice larger than other results reported before in a lower-Ro regime [1]. The experimental findings may shed new light to interpret the low precession rate under weak Coriolis force within the framework of the LSC models [2]. [1] J. E. Hart, S. Kittelman, and D. R. Ohlsen, Phys. Fluids 14, 955 (2002); J.-Q. Zhong and G. Ahlers, J. Fluid. Mech. 665, 300 (2010). [2] E. Brown and G. Ahlers, Phys. Fluids 18, 125108 (2006). 1 Supported by NSFC Grant 11202151. 8:39AM G19.00004 Lagrangian analysis of turbulent rotating convection , HADI RAJAEI, RUDIE KUNNEN, HERMAN CLERCX, Eindhoven University of Technology — This study focuses on exploring how the flow transition from one state to the other in rotating convection will affect the Lagrangian statistics of (fluid) particles. Up to now, the global parameters like the overall heat transfer or the wind Reynolds number are used to characterize the different turbulent states. However, it is obvious that the flow transition from weakly rotating Rayleigh-Benard (RB) to strongly rotating RB is also reflected in the Lagrangian dynamics of immersed tracer particles. We have employed 3D Particle Tracking Velocimetry (3D-PTV) in a water-filled cylindrical tank of equal height and diameter 200 mm. The measurements are performed in the central volume of 50 x 50 x 50 mm3 at a Rayleigh number Ra = 1.28e9 and Prandtl number Pr = 6.7. We are reporting the velocity and acceleration pdfs for different Rossby numbers and how transition from weakly rotating RB to strongly rotating RB affects the acceleration and velocity pdfs. 8:52AM G19.00005 Cell pattern transitions on a rotating convection induced by internal heat generation , YUJI TASAKA, YUDAI YAMAGUCHI, Hokkaido University, TAKATOSHI YANAGISAWA, JAMSTEC, YOSHIHIKO OISHI, YUICHI MURAI, Hokkaido University — We examined cell pattern formation on a rotating convection induced by internal heating. In this configuration with thermal insulation for the bottom boundary the thermal boundary layer whose separation provides convective motion exists only on the top boundary and the mechanism of the cell patter formation would be simpler than RBC. Flake visualization of the flow pattern indicated that there are three cell patterns on the marginal condition; irregular polygonal cells usually observed in internal heating convection, regular hexagonal cells, and time-dependent state as the results of competition of these patterns. These cell patterns are arranged by Rossby number. In the regular hexagonal cell pattern the fluid layer is occupied by regular hexagons like the rotating lattice observed in rotating RBC (Bajaj et al., 1998). But the cells do not show the continuous rotation unlike the rotating lattice but recreation of cells occurs intermittently in time and space. In a moderate condition, advecting local structures accompanied by sheet-like downward flows is observed instead of regular cell structures. This pattern is observed in a relatively large T a region and interruption of natural separation of the thermal boundary layer by thinning Ekman layer is an important factor. 9:05AM G19.00006 Effect of rotation on the temperature profile of turbulent convection with a Prandtl number P r = 12.31 , PING WEI, GUENTER AHLERS, Univ of California - Santa Barbara — We report on the influence of rotation about a vertical axis on the temperature profiles and the large-scale circulations (LSC) of turbulent Rayleigh-Bénard convection (RBC) in a cylindrical sample with aspect ratio Γ = D/L = 1.00 (D is the diameter and L the height). The working fluid was a fluorocarbon at a mean temperature Tm = 25◦ C with a Prandtl number P r = 12.3. The measurements covered the Rayleigh-number range 2 × 1010 ≤ Ra ≤ 2 × 1011 and the inverse Rossby number range 0 ≤ 1/Ro ≤ 9. With weak rotation the temperature in the fluid varied as A × ln(z/L) + B, where z is the distance from the bottom or top plate. For 1/Ro ≥ 1.2 we found that the temperature varied linearly with z. The temperature signature of the LSC along the sidewall was detectable up to 1/Ro ≃ 0.5. Retrograde rotation of the LSC was observed. The LSC temperature amplitude first decreased and then remained constant up to the critical inverse Rossby number 1/Roc for the onset of Ekman-vortex formation, and then decreased again. 1 Supported by NSF Grant DMR11-58514. 9:18AM G19.00007 Structure and local heat transport in the geostrophic regime of rotating Rayleigh-Benard convection , SERGIY GERASHCHENKO, SCOTT BACKHAUS, ROBERT ECKE, Los Alamos National Laboratory — We report experimental measurements of velocity fields and local temperature for rotating thermal convection in the geostrophic range with Rayleigh number 107 < Ra < 2 × 108 and Taylor number 109 < T a < 1010 (Ekman number 10−5 < Ek < 3 × 10−5 ). The fluid is water with Prandtl number P r ≈ 6. The velocity was obtained in a 2 cm × 2 cm area using particle tracking velocimetry, and the temperature in the middle of that area was measured using a thermistor. The simultaneous velocity and temperature data allow the local heat transport to be obtained. We also compute the vertical and lateral spatial correlation lengths, the probability distribution functions of temperature and velocity, and the spatial structure of the velocity field of localized convective structures - thermal plumes for the non-rotating system and Taylor columns p for convection with rotation. We present the dependence of these quantities for differing balances of buoyancy and rotation with Rossby number Ro = Ra/P rT a in the range 0.01 < Ro < 0.2 and provide a characterization of the state of geostrophic rotating thermal convection for the regime with Ra/Rac < 10 where Rac = 8.7T a4/3 . 9:31AM G19.00008 Natural Convection in a rotating multilayer spherical shell system with self gravity: A simplified global circulation model1 , FRANCISCO JAVIER LIRA RANGEL, RUBEN AVILA RODRIGUEZ, ARES CABELLO, Univ Nacl Autonoma de Mexico — The onset of thermal convection in rotating multilayer spherical shells is investigated. The system consist of six concentric shells. The first spherical gap has an aspect ratio equal to 0.35, the following four spherical gaps have different aspect ratio and the sixth gap has an aspect ratio equal to 0.8. The inner and the outer spherical gaps confine Boussinesq fluids while the middle spherical gaps are treated as a thermal conductor solid. The investigation is performed for Taylor numbers between 7.E4 and 1.E6 and Rayleigh numbers between 3.E3 and 1.E6. The convective patterns and the temperature fields are presented in the most inner and outer spherical gaps. Convection is driven by the temperature difference between the inner and outer spheres and a gravitational field wich varies like r and 1/r2 . The fluid equations are solved by using the spectral element method (SEM). The mesh is generated by using the cubed-sphere algorithm to avoid the singularity at the poles. To the knowledge of the autors the convection-conduction-convection problem presented in this paper has not been investigated previously. 1 Acknowledgment: DGAPA-PAPIIT project: IN117314-3 9:44AM G19.00009 Thermal convection in a rotating fluid sphere with self gravity, uniform heat source and precession1 , RUBEN AVILA, Universidad Nacional Autonoma de Mexico — The natural convection of a rotating fluid sphere with a self gravity field (which is proportional to the radius of the sphere) and with precessional motion is presented. The spherical bounding surface is maintained at a constant and uniform temperature which is lower than the temperature of the fluid. A constant and uniform heat source increases the temperature of the fluid confined in the sphere. The fluid sphere rotates and precesses with angular velocity vectors that form a certain inclination angle between them. The governing non-steady, three dimensional Navier-Stokes equations for an incompressible fluid, formulated in a Cartesian coordinate system (in the mantle reference frame) are solved by using the spectral element method. The influence of the Rayleigh number, the Ekman number and the Poincare number on the flow patterns, the temperature field and the heat transfer rate from the fluid sphere is presented. 1 DGAPA-PAPIIT project: IN117314-3 9:57AM G19.00010 A laboratory study of floating lenticular anticyclones1 , PATRICE LE GAL, IRPHE - Aix Marseille University - CNRS, HECTOR DE LA ROSA, ANNE CROS, RAÚL CRUZ-GOMEZ, Instituto de Astronomia y Meteorologia, Departamento de Fisica, Universidad de Guadalajara, MICHAEL LE BARS, IRPHE - Aix Marseille University - CNRS — Oceanic vortices play an important role in the redistribution of heat, salt and momentum in the oceans. Among these vortices, floating lenses or rings are often met in the meanders of warm currents. For instance the North Brazil Current rings are among the most intense and large anticyclonic vortices on Earth. In order to better describe these vortices, we propose here a laboratory study of these floating anticyclonic lenses. A blob of fresh water is slowly injected near the surface of a rotating layer of homogeneous salted water. Because of the opposite effects of rotation that tends to generate columnar structures and density stratification that spreads light water on the surface, the vortices take a finite size three dimensionnal typical shape. Visualization and PIV measurements of the shape, aspect ratios and vorticity profiles are compared to analytical predictions that use first a simple solid body rotation model and then a more realistic isolated Gaussian vorticity field inside the anticyclones. 1 This work was carried out within the framework of a bilateral cooperation between CNRS (France) and CONACYT (Mexico) Monday, November 24, 2014 8:00AM - 10:10AM Session G20 Acoustics III: Thermoacoustics — 2008 - Thierry Poinsot, CERFACS 8:00AM G20.00001 High-fidelity simulations of a standing-wave thermoacoustic-piezoelectric engine , JEFFREY LIN, CARLO SCALO, LAMBERTUS HESSELINK, Stanford University — We have carried out time-domain three-dimensional and one-dimensional numerical simulations of a thermoacoustic Stirling heat engine (TASHE). The TASHE model adopted for our study is that of a standing-wave engine: a thermal gradient is imposed in a resonator tube and is capped with a piezoelectric diaphragm in a Helmholtz resonator cavity for acoustic energy extraction. The 0.51m engine sustains 500Pa pressure oscillations with atmospheric air and pressure. Such an engine is interesting in practice as an external heat engine with no mechanically-moving parts. Our numerical setup allows for both the evaluation of the nonlinear effects of scaling and the effect of a fully electromechanically-coupled impedance boundary condition, representative of a piezoelectric element. The thermoacoustic stack is fully resolved. Previous modeling efforts have focused on steady-state solvers with impedances or nonlinear effects without energy extraction. Optimization of scaling and the impedance for power output can now be simultaneously applied; engines of smaller sizes and higher frequencies suitable for piezoelectric energy extraction can be studied with three-dimensional solvers without restriction. Results at a low-amplitude regime were validated against results obtained from the steady-state solver DeltaEC and from experimental results in literature. Pressure and velocity amplitudes within the cavities match within 2% difference. 8:13AM G20.00002 A nonlinear framework for the prediction of thermoacoustic oscillations , ALESSANDRO ORCHINI, MATTHEW JUNIPER, Univ of Cambridge — Thermoacoustic oscillations may occur in afterburners and gas turbines because of the interaction of unsteady heat release and acoustic waves. These oscillations lead to structural damage and deteriorate system efficiency. We present a low-order thermoacoustic model for premixed flames that exploits the fact that the main nonlinearity of this instability is due to the unsteady heat release. We describe the flame dynamics using the G-equation model, and some of the features of a Low Order ThermoAcoustic Network (LOTAN) to describe, in an efficient way, the system’s acoustics. The advantage of the latter approach, with respect to the more diffused Galerkin decomposition of the acoustic equations, is that mean flow effects, temperature variations, and cross sectional area changes of the combustion chamber can be easily included. With the resulting nonlinear network, we can analyze the stability of thermoacoustic systems both in the frequency and time domain, and determine the frequency, amplitude and stability of limit cycle oscillations. In the time domain, we can also predict the location of Neimark-Sacker bifurcations, which lead to quasi-periodic oscillations and more elaborate dynamical behavior. Numerical continuation is proved to be an efficient tool to achieve this goal. 8:26AM G20.00003 An Acoustically Consistent Investigation of Combustion Instabilities in a Dump Combustor , VIJAYA KRISHNA RANI, SARMA RANI, University of Alabama in Huntsville — An acoustically consistent, linear modal analysis-based analytical method is presented to predict the longitudinal and transverse combustion instabilities in a 2-D cartesian dump combustor. At first, rigorous acoustical analysis is performed. Novel, acoustically consistent jump or matching conditions are developed and applied at the duct cross-sectional interface(s), with distinct forms for the purely axial and non-axial modes. The effects of uniform and non-uniform mean flow, cross-sectional area ratio, as well as of different types of boundary conditions on the duct acoustic modes are investigated. Subsequent to the acoustic analysis, combustion instabilities of a 2-D, cartesian dump combustor are investigated. The instability analysis employs the developed acoustically consistent jump conditions, instead of the conventional mass, momentum, and energy balance-based conditions. Effects of the fluctuating heat-release source term in the acoustic wave equation are incorporated directly into the longitudinal wavenumber, obviating the need for a separate energy matching condition across the flame. A detailed investigation of the parametric space and boundary conditions affecting combustion instabilities is undertaken. 8:39AM G20.00004 Numerical Study of Energy Conversion of the Taconis Oscillations in an Axisymmetric Closed Tube , KATSUYA ISHII, Nagoya Univ, SHIZUKO ADACHI, Tokyo International Univ, HIROYUKI HAYASHI, Nagoya Univ — This paper studies spontaneous thermoacoustic oscillations of a helium gas in a closed cylindrical tube by solving the axisymmetric compressible Navier-Stokes equations. The wall temperature of the hot part near both ends (300K) and that of the cold central part (20K) are fixed. Numerical simulations are done for various values of the length ratio of the hot part to the cold part between 0.2 and 5.0. It is found that there are three groups of oscillation states, which are the fundamental mode and the second mode of a standing wave, and the oscillation with a shock wave. The states in each group have distinguished features of the vortical flow field and the temperature distribution. The evolution of the Lagrangian time derivative of entropy is analyzed to understand the energy conversion mechanism which maintains the nonlinear thermoacoustic oscillations. 8:52AM G20.00005 Nonlinear saturation of thermoacoustic oscillations in annular combustion chambers , GIULIO GHIRARDO, MATTHEW JUNIPER, University of Cambridge — Continuous combustion systems such as aeroplane engines can experience self-sustained pressure oscillations, called thermoacoustic oscillations. Quite often the combustion chamber is rotationally symmetric and confined between inner and outer walls, with a fixed number of burners equispaced along the annulus, at the chamber inlet. We focus on thermoacoustic oscillations in the azimuthal direction, and discuss the nonlinear saturation of the system towards 2 types of solutions: standing waves (with velocity and pressure nodes fixed in time and in space) and spinning waves (rotating waves, in clockwise or anti-clockwise direction). We neglect the effect of the transverse velocity oscillating in the azimuthal direction in the combustion chamber, and focus the model on the nonlinear effect that the longitudinal velocity, just upstream of each burner, has on the fluctuating heat-release response in the chamber. We present a low-order analytical framework to discuss the stability of the 2 types of solutions. We discuss how the stability and amplitudes of the 2 solutions depend on: 1) the acoustic damping in the system; 2) the number of injectors equispaced in the annulus; 3) the nonlinear response of the flames. 9:05AM G20.00006 Multiple space-scale global analysis for hydrodynamic/thermoacoustic instability in low Mach number combustion chambers , LUCA MAGRI, OUTI TAMMISOLA, University of Cambridge, YEE CHEE SEE, MATTHIAS IHME, Stanford University, MATTHEW JUNIPER, University of Cambridge — We propose a method to reduce the complexity of the reacting compressible Navier-Stokes equations for global/sensitivity analyses of thermo-acoustic systems. We use multiple space-scale analysis and consider a low Mach number. We assume that reacting hydrodynamic phenomena evolve at small space scales whereas acoustics evolve at larger space scales, a common situation in thermo-acoustics. The reacting hydrodynamics (RH) is governed by the reacting low Mach number equations, and the acoustics (AC) by the reacting Euler equations. The RH feeds into the AC via the heat release by the flame and the AC, in turn, feed back into the RH via the acoustic-pressure gradient (Klein’s limit). These two coupling terms enable the thermo-acoustic system to be linearized around time-averaged LES flows and studied as an eigenproblem. We perform global, adjoint and sensitivity analyses, investigating the reciprocal influence of RH/AC interactions and suggest strategies for open-loop control. The analysis is applied to a dump combustor and a complex industrial combustor (Meier’s). 9:18AM G20.00007 Forced response of self-excited thermoacoustic oscillations: lock-in, bifurcations, chaos and open-loop control , KARTHIK KASHINATH, Lawrence Berkeley National Laboratory, MATTHEW JUNIPER, University of Cambridge — This study aims to identify synchronization phenomena in thermoacoustics and to explore the possibility of open-loop control of self-excited thermoacoustic oscillations using weak periodic perturbations. We examine the response of a self-excited system of a premixed flame in a duct to harmonic forcing. When the system oscillating periodically is forced, we find that: (i) at low forcing amplitudes, the system responds at the natural and forced frequencies and linear combinations of these; (ii) above a critical forcing amplitude, the system locks into the external forcing; (iii) the bifurcations leading up to lock-in and the critical forcing amplitude required depend on the proximity of the forcing frequency to the natural frequency. When the system oscillating quasi-periodically is forced, we find that (i) if the forcing frequency is the same as one of the two characteristic frequencies of the torus attractor, then lock-in occurs at a critical amplitude via a saddle-node bifurcation; (ii) if the forcing frequency is not equal to either of the two characteristic frequencies of the torus attractor, then the torus breaks down and more elaborate behavior is noticed. When the system oscillating chaotically is forced close to the most dominant frequency in its spectrum, we find that it is possible to establish a stable periodic oscillation. Finally, we find that in some cases low-amplitude forcing can achieve lock-in and the amplitude of oscillations in the system are decreased by up to 70% of the unforced amplitude. 9:31AM G20.00008 Change of criticality of a thermoacoustic system , SUJITH RAMAN, AKHIL K T, GOPALAKRISHNAN E A, IIT Madras — The presence of an unsteady heat source in a confinement can result in high amplitude pressure oscillations when the unsteady heat release rate is in phase with inherent pressure fluctuations. This phenomenon is termed as thermoacoustic instability. Previous studies have shown that this phenomenon is nonlinear in nature and that the transition to instability is through a Hopf bifurcation. The nature of criticality of the Hopf bifurcation becomes significant from the point of view of controlling and mitigating these undesirable oscillations. We consider a prototypical thermoacoustic system, an electrically heated horizontal Rijke tube, to investigate the nature of criticality during transition. Heater power is varied in a quasi-steady manner while keeping other system parameters constant. A subcritical transition to thermoacoustic instability is observed when the mass flow rate was above a critical value, whereas a supercritical transition is observed below it. This change in nature of bifurcation is ascertained rigorously using renormalization group method (RG). 9:44AM G20.00009 Engine core noise analysis using an hybrid modeling approach , JEFFREY O’BRIEN, JEONGLAE KIM, MATTHIAS IHME, Stanford University - Center for Turbulence Research — As aircraft engines become progressively quieter through the reduction of jet noise, the acoustic contributions of components upstream of the jet, especially the combustor, must be reduced to produce still quieter engines. Combustion noise can be broken down into two components: direct and indirect noise. Direct noise refers to pressure fluctuations that are generated directly by turbulent combustion, while indirect noise describes acoustics that stem from the interaction between the entropy fluctuations generated inside a combustor and the downstream flow path. This study analyzes the effects of both types of core noise. An LES simulation of a model swirl combustor is performed in order to generate representative pressure and entropy fluctuations which are then fed into a moving-mesh RANS calculation of a high pressure turbine stage. The evolution of these fluctuations through the turbine stage is analyzed and the ”chopping” effect of the turbine on the fluctuations is characterized. Additionally, the turbine output will be fed into a fully compressible jet noise calculation to assess how the entropy fluctuations are affected by the flow path and alter the acoustic behavior of the jet. 9:57AM G20.00010 Adjoint sensitivity analysis of thermoacoustic instability in a nonlinear Helmholtz solver1 , MATTHEW JUNIPER, LUCA MAGRI, University of Cambridge — Thermoacoustic instability is a persistent problem in aircraft and rocket engines. It occurs when heat release in the combustion chamber synchronizes with acoustic oscillations. It is always noisy and can sometimes result in catastrophic failure of the engine. Typically, the heat release from the flame is assumed to equal the acoustic velocity at a reference point multiplied by a spatially-varying function (the flame envelope) subject to a spatially-varying time delay. This models hydrodynamic perturbations convecting down the flame causing subsequent heat release perturbations. This creates an eigenvalue problem that is linear in the acoustic pressure but nonlinear in the complex frequency, omega. This can be solved as a sequence of linear eigenvalue problems in which the operators are updated with a new value of omega after each iteration. Adjoint methods find the sensitivity of each eigenmode to all the parameters simultaneously and are well suited to thermoacoustic problems because there are a few interesting eigenmodes but many influential parameters. The challenge here is to express the sensitivity of the eigenvalue at the final iteration to an arbitrary change in the parameters of the first iteration. This is a promising new technique for the control of thermoacoustics. 1 European Research Council Grant number 2590620 Monday, November 24, 2014 8:00AM - 10:10AM Session G21 Separated Flows — 2010 - Kenneth Christensen, Notre Dame University 8:00AM G21.00001 Bluff-body wakes in streamwise oscillating flow , KENNETH GRANLUND, TIMOTHY CLEAVER, Air Force Research Laboratory — The wake topology from bluff bodies in freestream are sensitive to streamwise fluctuations and the wake can lock-on to multiples or fractions of the forcing frequency. Here we asses the amplitude level receptivity to such lock-on in nominally 2-dimensional flow at Reynolds number 2 · 105 . The influence of cross-section geometry on the wake is studied for circular, triangular and rectangular with varying slenderness ratio. The experiments are performed in a water tunnel, qualitatively and quantitatively observed with PLIF and high-speed PIV. 8:13AM G21.00002 Investigation of Bio-Inspired High Lift Devices for Stall Mitigation1 , ESTEBAN HUFSTEDLER, BEVERLEY J. MCKEON, California Institute of Technology — A passive upper-surface flap has been shown to increase the lift on a wing after stall and reduce the severity of stall at a wide range of Reynolds numbers. Experiments at Re=20,000 have been conducted that examined the forces and flow fields around an airfoil with passively moving and static upper-surface flaps. Force measurements confirm the reported post-stall lift-enhancing effect. Particle image velocimetry measurements display the interaction of a significant region of reversed flow with the flap in the lift-enhancing regime. Application of proper orthogonal decomposition techniques to the velocity field data leads to identification of relevant timescales in the separated region and a quantification of the intermittency of vortex shedding that occurs after stall. 1 The support of Airbus for this work is gratefully acknowledged. 8:26AM G21.00003 Formation of Three-Dimensional Stall Cells on Two-Dimensional Airfoils , VICTOR SIVANERI, BURAK TUNA, EDWARD DEMAURO, MICHAEL AMITAY, Rensselaer Polytechnic Institute — Stall cells are a pattern of threedimensional mushroom-shaped structures that form within the separated region of stalled, thick airfoils within a certain range of Reynolds numbers. The occurrence and number of stall cells are dependent on the wing camber, aspect ratio, angle of attack, and Reynolds number. While much work within the literature has been conducted to visualize and measure this phenomenon, to date a comprehensive explanation for their existence remains elusive. The present work aims to identify these structures, quantify them, and understand the mechanisms by which they are formed. This was conducted using oil flow visualization and stereoscopic particle image velocimetry (SPIV) on a two-dimensional NACA 0015 airfoil, pitched to 18◦ angle of attack, at Reynolds numbers ranging from 160,000 to 400,000. Oil flow visualization was used to qualitatively identify the signature of the stall cells on the airfoil surface and resolve the associated skin friction vector fields. In addition, SPIV measurements were taken in order to quantify the flow field in the presence and absence of stall cells within the region of separated flow above the surface of the airfoil. Results showed that the stall cells are highly sensitive to Reynolds number, with evidence of an apparent bi-stable state existing at a Reynolds number of 320,000. 8:39AM G21.00004 Effect of rib length on characteristics of separation and reattachment , JACQUES W. VAN DER KINDERE, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — Ribs reproduce key elements in engineering. Their aerodynamics can be detrimental to vehicles, or harnessed favorably in motors, and heat exchangers. The flow around such obstacle includes separation upstream of the obstacle, separation and reattachment on the top surface, and separation downstream. The interaction between these different recirculation regions is affected by the obstacle’s length. This study examines experimentally how the interaction between different recirculation regions evolves with rib length. The rib is submerged in a fully turbulent boundary layer (δ/H = 1.37, where δ and H are respectively incoming boundary layer thickness and rib height), and the Reynolds number based on rib height is ReH = 20, 000. Particle Image Velocimetry synchronized with pressure measurements was carried out on the flow past ribs of different lengths. The length of the rib (distance between the two vertical faces) varied between L = 0.1H and L = 8H. Results from this experiment will be used to compare the mean recirculation lengths of the different separation regions. Pressure distribution within the separation regions will also be examined and compared. Finally, the interaction between the different shear layers will be examined and contrasted across all cases. 8:52AM G21.00005 Near-wake characteristics of a rotating dimpled sphere1 , JOOHA KIM, HAECHEON CHOI, Seoul National University — In this study, we investigate the characteristics of flow around a rotating dimpled sphere in the subcritical, critical and supercritical Reynolds number (Re) regimes. The experiment is performed in a wind tunnel at Re = 0.3 × 105 – 2.4 × 105 and the spin ratio (α; ratio of surface velocity to the free-stream velocity) of 0 (no spin) – 2.6. We directly measure the drag and lift forces and the velocity field in the near wake using PIV and smoke visualization. In the subcritical Re regime, the wake of a stationary dimpled sphere shows large-scale wavy structures and the hairpin-shaped vortices are shed changing its azimuthal orientation quasi-randomly in time. As Re increases from subcritical to critical regime, the recirculation bubble length decreases significantly and the drag coefficient reduces rapidly to about 0.23. The wavelength of wake also decreases and the shedding orientation of the hairpin-shaped vortices becomes fixed in time. In the supercritical regime, both the recirculation bubble length and drag coefficient remain almost constant, whereas the wavelength of wake decreases further. With rotation, the recirculation bubbles disappear at very small α in the critical and supercritical regimes, resulting in a faster increase in the lift coefficient with α than that in the subcritical regime. 1 Supported by 2011-0028032, 2014M3C1B1033980. 9:05AM G21.00006 Flight trajectory of a rotating golf ball with grooves1 , MOONHEUM BAEK, JOOHA KIM, HAECHEON CHOI, Seoul National University — Dimples are known to reduce drag on a sphere by the amount of 50% as compared to a smooth surface. Despite the advantage of reducing drag, dimples deteriorate the putting accuracy owing to their sharp edges. To minimize this putting error but maintain the same flight distance, we have devised a grooved golf ball (called G ball hereafter) for several years. In this study, we modify the shape and pattern of grooves, and investigate the flow characteristics of the G ball by performing wind-tunnel experiments at the Reynolds numbers of 0.5 × 105 − 2.5 × 105 and the spin ratios (ratio of surface velocity to the free-stream velocity) of 0 – 0.6 that include the real golf-ball velocity and rotational speed. We measure the drag and lift forces on the rotating G ball and compare them with those of a smooth ball and two well-known dimpled balls. The lift-to-drag ratio of the G ball is much higher than that of a smooth ball and is in between those of the two dimpled balls. The trajectories of flying golf balls are computed. The flight distance of G ball is almost the same as that of one dimpled ball but slightly shorter than that of the other dimpled ball. The fluid-dynamic aspects of these differences will be discussed at the talk. 1 Supported by 2011-0028032, 2014M3C1B1033980. 9:18AM G21.00007 Prediction of flow separation from aircraft tails using a RSM turbulence model1 , ANDREA MASI, Airbus - University of Cambridge, JEREMY BENTON, Airbus, PAUL G. TUCKER, University of Cambridge — Enhancing engineers’ capability to predict flow separation would generate important benefits in aircraft design. In this study the attention is focused on the vertical tail plane (VTP), which consists of a fixed part (the fin) and a moveable control surface (the rudder). For standard two-engine aircraft configurations, the size of the VTP is driven by the condition of loss of an engine during takeoff and low speed climb: in this condition the fin and the rudder have to be sufficient in size to balance the aircraft. Due to uncertainties in prediction of VTP effectiveness, aircraft designers keep to a conservative approach, risking specifying a larger size for the VTP than it is probably necessary. Uncertainties come from difficulties in predicting the separation of the flow from the surfaces of the aircraft using current CFD techniques, which are based on the use of RANS equations with eddy viscosity turbulence models. The CFD simulations presented in this study investigate the use of a RSM turbulence model with RANS and URANS. The introduction of a time-dependency gives benefits in the accuracy of the flow solution in presence of massive flow separation. This leads to the investigation of hybrid RANS/LES techniques with the aim of improving the solution of the detached flow. 1 EU FP7 project ANADE (Grant Agreement Number 289428) 9:31AM G21.00008 Numerical Dissipation and Subgrid Scale Modeling for Separated Flows at Moderate Reynolds Numbers1 , FRANCOIS CADIEUX, JULIAN ANDRZEJ DOMARADZKI, University of Southern California — Flows in rotating machinery, for unmanned and micro aerial vehicles, wind turbines, and propellers consist of different flow regimes. First, a laminar boundary layer is followed by a laminar separation bubble with a shear layer on top of it that experiences transition to turbulence. The separated turbulent flow then reattaches and evolves downstream from a nonequilibrium turbulent boundary layer to an equilibrium one. In previous work, the capability of LES to reduce the resolution requirements down to 1% of DNS resolution for such flows was demonstrated (Cadieux et al, JFE 136-6). However, under-resolved DNS agreed better with the benchmark DNS than simulations with explicit SGS modeling because numerical dissipation and filtering alone acted as a surrogate SGS dissipation. In the present work numerical viscosity is quantified using a new method proposed recently by Schranner et al. and its effects are analyzed and compared to turbulent eddy viscosities of explicit SGS models. The effect of different SGS models on a simulation of the same flow using a non-dissipative code is also explored. 1 Supported by NSF 9:44AM G21.00009 Mechanisms Of Pressure Distributions Within Laminar Separation Bubble At Different Reynolds Numbers1 , DONGHWI LEE, University of Tokyo, SOSHI KAWAI, TAKU NONOMURA, AKIRA OYAMA, KOZO FUJII, Institute of Space and Astronautical Science, JAXA — Large-eddy simulation around 5% thickness flat plate at Re = 5, 000, 6, 100, 11, 000 and 20, 000 are performed and the physical mechanisms of the pressure distributions (Cp ) in laminar separation bubbles are analyzed. Depending on the Reynolds number, a gradual pressure recovery and plateau pressure distribution are observed as experiments by Anyoji et al. [AIAA paper 2011-0852]. The causes of the pressure distributions are quantitatively shown by deriving the pressure gradient (momentum budget) equation from the steady momentum equation. From the results, we identify that the viscous diffusion term near the surface has a major contribution to the pressure gradients, and a different growth of the separated shear layer relying on the Reynolds numbers affects the viscous stress near the surface. The gradual pressure recovery at the lower Reynolds numbers is caused by the progressive development of separated shear layer due to the viscous stress which makes a non-negligible viscous stress. On the other hand, a thin laminar separated shear layer is created at the higher Reynolds numbers because of the relatively small viscous diffusion effects, which results in a negligible shear stress distribution. It makes dp/dx ≈ 0 and the plateau pressure distribution is generated. 1 Asahi Glass Scholarship 9:57AM G21.00010 Flow-induced instabilities of a flexibly-mounted rigid plate placed in water1 , PARIYA POURAZARM, YAHYA MODARRES-SADEGHI, MATTHEW LACKNER, University of Massachusetts Amherst — Flow-induced instabilities of a flat rigid plate placed in water with either one degree of freedom (torsional) or two degrees of freedom (transverse and torsional) are studied. The onset of dynamic instability for each configuration was pinpointed and the post-critical behavior of the system in both 1 DoF and 2 DoF cases was investigated. For the 1DoF case, for all flow velocities higher than the critical, a periodic motion was observed. The amplitude of oscillations increased with increasing flow. For the 2DoF system, a period-doubling route to chaos was observed. The post-critical periodic oscillations were followed by period-2 and later on period-4 oscillations, which led to chaotic oscillations at higher flow velocities. Flow visualizations showed that for periodic oscillations, one vortex was shed in each cycle, and for period-2 oscillations, two vortices were shed in each cycle. 1 The support provided by the Wind Technology Testing Center, a part of the Massachusetts Clean Energy Center is acknowledged. Monday, November 24, 2014 8:00AM - 10:10AM Session G22 Instability: Transition to Turbulence — 2012 - Jim Riley, University of Washington 8:00AM G22.00001 Coherent structures in stratified plane Couette flows , DANIEL OLVERA, RICH KERSWELL, University of Bristol — Wall–bounded shear flows typically follow a subcritical transition scenario where finite amplitude solutions unconnected to the basic flow play a key role. Edge tracking has been very useful in finding some of these by following the laminar–turbulent boundary in phase space for many canonical shear flows. However, it has yet to be used to probe stratified flows. We will discuss the results of edge tracking in stably–stratified plane Couette flow over large and small domains. 8:13AM G22.00002 Critical Reynolds number for global instability of channel flow in subcritical scenario1 , TAKAHIRO ISHIDA, TAKAHIRO TSUKAHARA, Department of Mechanical Engineering, Tokyo University of Science — We perform direct numerical simulations for the transitional pressure-driven channel flow and investigate the critical Reynolds number for global instability (ReG ). In the channel flow, the critical Reynolds number relevant to local (infinitesimal) instability (ReL ) is known as 5772, which is based on the channel half width (h/2) and the channel centerline velocity in laminar flow, by the linear stability theory. However, the understanding of ReG is still unresolved because it is inherently non-linear. In this study, temporal progress of a turbulent spot that grows from finite disturbance in the laminar flow is analyzed. We use the small and the large computational boxes for identifying the lower critical number, which are Lx *Ly *Lz =6.4h*h*3.2h and 102.4h*h*51.2h, respectively. For the small box, we determine ReG |S is 1400. The flow regime observed in the small box is only either laminar flow or fully-developed turbulence. As for the large box, we obtain ReG |L of 875 because of the emergence of the transitional structure named “turbulent stripe,” or a coexistence of oblique turbulent and laminar region. Because the spatial size of turbulent stripe is much larger than the small box, ReG |S indicates the critical Reynolds number for the flow without large-scale intermittency. Therefore, we found that the existence of turbulent stripe caught in large box would decrease the ReG value compared to those proposed by previous studies. 1 The first author was supported by Grant-in-Aid for JSPS Fellows (26.7477). 8:26AM G22.00003 Nonlinear optimal perturbations of stratified plane Couette flow , T.S. EAVES, DAMTP, University of Cambridge, C.P. CAULFIELD, BP Institute & DAMTP, University of Cambridge — The stability properties of shear flows have received wide attention due to the important engineering applications of understanding how and when turbulence might emerge in a given flow geometry. Research has recently focused on identifying “minimal seeds,” i.e. the initial perturbations to a laminar state with the smallest initial perturbation energy E0 = Ec that ultimately trigger the transition to turbulence. In unstratified plane Couette flow, Rabin, Caulfield & Kerswell (J. Fluid Mech. 2012 712) identified both such a minimal seed, and other “nonlinear optimal perturbations” (NLOPS) with E0 < Ec which maximised the gain in kinetic energy over some finite time while the flow still remained laminar. We use the same variational method of “nonlinear adjoint looping” to identify NLOPS and minimal seeds in stably stratified plane Couette flow, where a constant (stabilising) density difference is maintained across the flow. We also identify the mechanisms through which such perturbations may transiently gain both kinetic and potential energy as the bulk Richardson number is varied, identifying how stratification changes the qualitative characteristics of the optimal perturbations. 8:39AM G22.00004 Singularity of Navier-Stokes Equations Leading to Turbulent Transition , HUA-SHU DOU1 , Zhejiang Sci-Tech University, FLUID MECHANICS RESEARCH TEAM — As is well known, there is discontinuity during the transition from laminar flow to turbulence in the time-averaged Navier-Stokes equations. In other words, singularity may implicitly exist in the Navier-Stokes equations. Transition of a laminar flow to turbulence must be implemented via the singularity. However, how the singularity of Navier-Stokes equations is related to the turbulent transition is not understood. In this study, the singularity possibly hidden in the Navier-Stokes equation is exactly derived by mathematical treatment. Then, it is found that for pressure driven flows, the singularity of Navier-Stokes equations corresponds to the inflection point on the velocity profile. Since the rate of amplification to a disturbance at the inflection point is infinite, the laminar flow is able to involve into turbulence at this point firstly at a sufficient high Reynolds number. This is the reason why turbulent spot is formed at the location of inflection point. It is further demonstrated that the existence of the singularity in the time-averaged Navier-Stokes equations is the necessary and sufficient condition for the turbulent transition in pressure driven flows. These results agrees well with the findings from the recent proposed energy gradient method. 1 Professor in Fluid Mechanics; AIAA Associate Fellow 8:52AM G22.00005 The competition of convective and absolute instabilities in rotating-disk flow transition1 , SHINTARO IMAYAMA, P. HENRIK ALFREDSSON, R.J. LINGWOOD2 , Linné FLOW Centre, KTH Mechanics, Stockholm, Sweden — The main objective of this experimental study is to investigate laminar-turbulent transition mechanisms in the rotating-disk boundary-layer flow. Lingwood (1995) found that the flow becomes locally absolutely unstable above a critical Reynolds number and suggested that absolutely unstable travelling waves triggered nonlinearity leading to transition. However, the growth of convectively unstable stationary vortices is also a possible alternative route if the surface roughness of the disk is sufficiently large. The convectively unstable stationary vortices are attributed to an inviscid crossflow mechanism. Flow-visualization studies and hot-wire measurements of the rotating-disk boundary layer typically capture 28-32 stationary vortices in the transition regime (e.g. Imayama et al. 2014). The hot-wire measurements presented here were performed on a smooth glass disk with a diameter of 474 mm. To excite stationary vortices disk-shaped roughness elements with a diameter of 2 mm and a height of 5 micron were put on the disk at a radial position of 110 mm. In the presentation, the details of the convectively unstable stationary vortices in the rotating-disk boundary layer are shown and compared with travelling waves and similarities/differences in the turbulent transition discussed. 1 This 2 Also work is supported by the Swedish Research Council (VR) and the Linné FLOW Centre. at University of Cambridge, UK. 9:05AM G22.00006 Fully localised nonlinear energy growth optimals in pipe flow , CHRIS PRINGLE, Coventry University, ASHLEY WILLIS, University of Sheffield, RICH KERSWELL, University of Bristol — In wall-bounded shear flows such as pipe flow, transition to turbulence remains a problem of great theoretical and practical importance. The transition is typically abrupt, occurs at flow rates for which the underlying base flow is stable, and is triggered by disturbance amplitudes much smaller that the ensuing turbulent state. Progress has recently been made in identifying the smallest perturbation capable of triggering turbulence (the minimal seed) using energy growth optimals, but only in small periodic domains. Here we present a new fully-localised (non-periodic in the streamwise direction) energy growth optimal for pipe flow. The perturbation approaches the experimentally-relevant minimal seed for transition in long pipes. 9:18AM G22.00007 Transition to sustained turbulence in pipe flow: a second order phase transition? , MUKUND VASUDEVAN, BJÖRN HOF, The Institute of Science and Technology Austria — In a recent study, the critical point for sustained turbulence in a pipe was estimated to be Re ≈ 2040, by balancing the times scales for turbulence growth and decay processes. This work brought into focus the spatio-temporal aspects of the transition and suggested the possibility that the transition is a second order non-equilibrium phase transition. The present contribution aims to experimentally characterize the transition to sustained turbulence in pipe flow in greater detail and explore the analogy to a phase transition. However, the long time scales near the critical point (∼ 107 advective time units) pose a challenge in realizing this. We circumvent this problem by constructing a set-up with a quasi-periodic pipe, that exploits the memoryless nature of the turbulence spreading and decay processes in the vicinity of the critical point. In conjunction with an accurate control of the Reynolds number, it is then possible to monitor the spatio-temporal dynamics for arbitrarily long times, and obtain quantities such as the equilibrium turbulent fraction. We present evidence to support the idea that the transition to sustained turbulence in pipe flow is a phase transition of second order and provide first estimates of some of the associated critical exponents. 9:31AM G22.00008 Investigation of Turbulent Laminar Patterns in Poiseuille-Couette flow , QUOC NGUYEN, The University of Oklahoma, DIMITRIOS PAPAVASSILIOU, The University of Oklahoma & NSF — Laminar-turbulent intermittency has recently been observed in the transitional regime of pipe ...[1-2] and plane Couette flow ...[3-7]. While many works focus on behavior of these patterns in plane Couette flow, little attention has been paid to Poiseuille flow and transition from Couette to Poiseuille flow. In this study, we investigate behavior of turbulent laminar patterns in Poiseuille-Couette flow, including pure Poiseuille and Couette flows at two limits. Direct Numerical Simulation (DNS) is used to simulate a Poiseuille-Couette channel at a size of 16πh x 2h x 2πh (corresponding to a resolution of 512x129x128 in x, y and z directions), with periodic boundary condition applied in the x and z directions (h is half of the channel height). The Reynolds number is 300, and the flow is at transitional regime in all simulations. Behavior of laminar turbulent patterns as the flow goes from Couette to Poiseuille flow will be presented in details. This would shed some light on the effect of different types of flow on these patterns, as well as how these patterns vary from fully Poiseuille flow to fully Couette flow. Bibliography .1. Moxey D & Barkley D (2009), PNAS 107(18). 2. Samanta D, Lozar AD, & Hof B (2011). J. Fluid Mech. 681. 3. Barkley D & Tuckerman LS (2005) Phys. Rev. Lett. 94. 4. Duguet Y & Schlatter P (2013) Phys. Rev. Lett. 110. 5. Philip J & Manneville P (2011). Phys. Rev. E 83. 6. Tuckerman LS & Barkley D (2011) Phys. Fluids 23. 7. Shi L, Avila M, & Hof B (2013) Phys. Rev. Lett. 110. 9:44AM G22.00009 Effects of Longitudinal Grooves on the Stability of Channel Flow , H. VAFADAR MORADI, JERZY M. FLORYAN, University of Western Ontario — The travelling wave instability in a channel with small-amplitude longitudinal grooves of arbitrary shape has been studied. The disturbance velocity field is always three-dimensional with disturbances which connect to the two-dimensional waves in the limit of zero groove amplitude playing the critical role. The presence of grooves destabilizes the flow if the groove wave number β is larger than βtran ≈ 4.22, but stabilizes the flow for smaller β. It has been found that βtran does not depend on the groove amplitude. The dependence of the critical Reynolds number on the groove amplitude and wave number has been determined. Special attention has been paid to the drag-reducing long wavelength grooves, including the optimal grooves. It has been demonstrated that such grooves slightly increase the critical Reynolds number, i.e., such grooves do not cause an early breakdown into turbulence. 9:57AM G22.00010 Secondary instability and tertiary states in rotating plane Couette flow , CONOR DALY, University of Cambridge, TOBIAS SCHNEIDER, École Polytechnique Fédérale de Lausanne, PHILIPP SCHLATTER, Kungliga Tekniska Högskolan, NIGEL PEAKE, University of Cambridge — Recent experimental studies have shown rich transition behaviour in rotating plane Couette flow (RPCF). In this paper we study the transition in supercritical RPCF theoretically by determination of various equilibria and periodic orbit tertiary states via Floquet analysis on secondary Taylor vortex solutions. Two new tertiary states are discovered which we name oscillatory wavy vortex flow (oWVF) and skewed vortex flow (SVF). We present the bifurcation routes and stability properties of these new tertiary states, alongside a bifurcation procedure whereby a set of defected wavy twist vortices are approached. Monday, November 24, 2014 8:00AM - 10:10AM Session G23 Geophysical Fluid Dynamics: Internal Waves — 2001 - Scott Wunsch, John Hopkins University 8:00AM G23.00001 Internal Waves Generated by Unsteady Impulsive Forcing - Laboratory Experiments , KARA SHIPLEY, ALAN BRANDT, MATTHEW PAOLETTI, JHU/APL — Internal waves are generated in laboratory experiments using impulsive forcing to further the understanding of unsteady source mechanisms. Impulsive forcing events, unlike steady or periodic forcing, are both transient and broadband, and have been the focus of only a limited number of fundamental studies. The experiments presented here examine the dynamics of the release of a homogeneous heavy fluid into a fluid with a density-stratified layer above a region of constant density. The miscible forcing volume is visualized utilizing fluorescent dye, which allows for measurements of the plume energy flux, while the internal wave field is characterized by measuring the density fluctuations with an array of conductivity probes. In all cases, the uniformly stratified region has a buoyancy frequency of N = 1 rad/s, the depth of which varied 12-50% of the total fluid depth. The descending plume entrains ambient fluid and subsequently rebounds to an equilibrium level. The effects of varying the density and volume of the forcing fluid on the energy flux of the radiated internal waves are presented. 8:13AM G23.00002 Internal waves generated by unsteady impulsive forcing - numerical simulations1 , MATTHEW PAOLETTI, KARA SHIPLEY, ALAN BRANDT, Johns Hopkins University — Numerical simulations of the generation of internal waves by an unsteady impulse are presented. While extensive work has examined the generation of internal waves by steady flow, such as winds over mountains, or periodic flow, an example being tidal flow over bathymetry, internal waves can also be generated by transient events like those produced by local instabilities. The studies presented here focus on the generation of internal waves by the release of a patch of miscible fluid of constant density into a stably stratified water column. The fluid descends owing to its initial momentum, spreads in the lateral direction, and vertically displaces the isopycnals, leading to the generation of internal waves. The transfer of energy from the impulse to the internal wave field is characterized by the energy flux of the radiated internal waves. While the impulse is initially axisymmetric, the effects of the three-dimensional nature of the turbulent evolution are examined by comparing the results of two-dimensional and three-dimensional numerical simulations. 1 Supported by the Office of Navel Research 8:26AM G23.00003 Internal Wave Generation in Evanescent Regions , ALLISON LEE, JULIE CROCKETT, Brigham Young Univ - Provo — Internal waves are well known to be generated by flow over topography in regions where the excitation frequency is less than the local buoyancy frequency of the fluid. Linear theory indicates that if the excitation frequency is greater than the buoyancy frequency, then internal waves will be evanescent. These evanescent waves’ amplitudes decrease exponentially away from the region of generation. However, under certain conditions, the wave stress generated by flow over topography in a region that is weakly stratified can be transported by evanescent waves into a region of higher stratification, far from the generation region, and internal waves will begin to propagate away. Further exploration of this theory is undertaken with physical experiments performed with varying stratifications and multiple types of topography. 8:39AM G23.00004 Parametric subharmonic instability (PSI): from internal plane waves to realistic beams , THIERRY DAUXOIS, CNRS & ENS de Lyon, BAPTISTE BOURGET, SYLVAIN JOUBAUD, PHILIPPE ODIER, HÉLÈNE SCOLAN, ENS de Lyon — We study experimentally the parametric subharmonic instability, which corresponds to the destabilization of a primary plane wave and the spontaneous emission of two secondary waves, of lower frequencies and different wave vectors. We show how, using a time-frequency analysis and a Hilbert transform, one can characterize precisely the instability. Moreover, we present results showing the crucial importance of the finite width when considering beams. Experiments and numerical results will be discussed in relation with a new theoretical approach. The latter brings new insights on energy transfers in the ocean where internal waves with finite size beams are dominant. 8:52AM G23.00005 PSI and turbulence during the interaction of the internal wave beam with upper ocean pycnocline , BISHAKHDATTA GAYEN, Australian National University, SUTANU SARKAR, University of California San Diego — Three-dimensional numerical simulations are performed to investigate the interaction of a semidiurnal internal wave (IW) beam with the nonuniform stratification of an upper ocean pycnocline. During the initial stage of the interaction, higher harmonics originate after reflection of the IW beam at the caustic and are trapped in the pycnocline, while at later time the incoming beam undergoes a parametric subharmonic instability (PSI) inside the pycnocline, that exhibit exponential growth with a rate of 2/3 day−1 . During PSI small vertical waves form resulting in wave steepening and produce convective overturns. Convective instability initiates transition to turbulence while shear production maintains it. Turbulence is characterized by examining the temporal evolution of its production and dissipation. 9:05AM G23.00006 Internal waves incident on a sheared ocean pycnocline , SCOTT WUNSCH, HWAR KU, Johns Hopkins University — Internal waves are commonplace in the oceans. Near the surface, they interact with a sharply increasing density gradient (pycnocline) as well as near-surface currents. Here, fully nonlinear numerical simulations are used to study internal waves incident on a sheared pycnocline. Linear analysis of the unstable modes of a sheared pycnocline above a stably stratified fluid reveals that two types of instabilities may occur. One is the well-known Holmboe instability, while the other is a longer wavelength Kelvin-Helmholtz mode which couples more strongly to incident internal waves. Both types of instabilities are seen in the simulations, and the nonlinear evolution of each is explored. Possible implications of these results for oceanic internal waves are considered. 9:18AM G23.00007 Energetics of nonlinear harmonic generation during the incidence of an internal wave beam on a model oceanic pycnocline , ANIL AKSU, DIAMESSIS PETER, Cornell University, SCOTT WUNSCH, Johns Hopkins University, Applied Physics Laboratory — An energetic analysis of the interaction of a numerically simulated IWB with a model ocean pycnocline is presented. The focus is on the nonlinear generation of harmonics. The analysis consists of a) monitoring the transfer of the primary beam’s energy into higher harmonics along the beam path and b) evaluating how any energy trapped inside the pycnocline is distributed across different wave frequencies propagating within it. The majority of the analysis is performed on a dataset spanning a wide range of pycnocline strengths and thicknesses restricted to an IWB propagating at 45◦ from the horizontal. For such an angle, internal wave refraction is the primary driver of nonlinear harmonic generation. Moreover, all resulting harmonics remain trapped within the pycnocline. Preliminary results from additional simulations with shallower angles of IWB incidence are also analyzed. When the incidence angle is less than 30 degrees, IWB reflection is an additional important mechanism of harmonic generation and lower harmonics are able to radiate back out of the pycnocline. 9:31AM G23.00008 Harmonic Generation of Internal Waves Reflected from a Slope1 , BRUCE RODENBORN, Centre College, DANIEL KIEFER, Center for Nonlinear Dynamics, University of Texas at Austin, HEPENG ZHANG, Jiao Tong University, HARRY L. SWINNEY, Center for Nonlinear Dynamics, University of Texas at Austin — Internal wave reflection from a uniform sloping boundary is often analyzed using linear or a weakly nonlinear inviscid theory.2 Under these assumptions for a linearly stratified fluid, Thorpe3 and Tabaei et al.4 derived predictions for the boundary angle where second harmonic generation should be most intense. We previously conducted experiments and simulations that found the angle that maximizes second harmonic generation is given instead by an empirical geometric relationship between the wave beam and boundary angles.5 We used integrated kinetic energy as a measure of beam intensity, but the method of Lee et al.6 determines the energy flux of the second harmonic wave in the experiments. We compare results using this new method to weakly nonlinear theories and our empirical prediction. 1 Texas Advanced Computing Center Dauxois and W.R. Young, J. Fluid Mech. 390, 271-295 (1999) 3 S. A. Thorpe, J. Fluid Mech., 178, 279-302 (1987) 4 A. Tabaei, T. R. Akylas and K. G. Lamb, J. Fluid Mech. 526, 217-243 (2005) 5 B. E. Rodenborn, D. Kiefer, H. P. Zhang, and H. L. Swinney. Phys. Fluids, 23(2), 2011. 6 Frank M. Lee, M. S. Paoletti, H. L. Swinney, P. J. Morrison. arXiv:1401.2484. 2 T. 9:44AM G23.00009 Suppression of tidal conversion by virtual seafloor1 , HARRY L. SWINNEY, LIKUN ZHANG, University of Texas at Austin — We examine in numerical simulations how the conversion of tidal energy into internal gravity wave energy is suppressed by wave interference between adjacent ridges of steep topography [L.K. Zhang and H.L. Swinney, Phys. Rev. Lett. 112, 104502 (2014)]. Simulations for both periodic and random steep topographies reveal that the time-averaged wave energy radiated upwards arises only from the portion of the ridges above an elevated “virtual seafloor.” We find that the average radiated wave power can be predicted by linear theory for weak topography by replacing the actual floor with the virtual floor. The virtual floor concept is used to extend linear theory to predict the energy conversion rate for steep topography. This nonlocal modification of linear theory should be useful in estimating the energy flux generated by tidal flow over the global seafloor. 1 Supported by ONR MURI Grant N000141110701 (WHOI). Also, LZ is supported by the 2013-14 ASA F. V. Hunt Postdoctoral Research Fellowship. 9:57AM G23.00010 Determination of internal wave power from synthetic schlieren data , FRANK M. LEE, MICHAEL ALLSHOUSE, P.J. MORRISON, HARRY L. SWINNEY, Univ of Texas, Austin — Internal waves are generated in the ocean by tidal flow over bottom topography, and they are of considerable interest because of their significant contribution to the energy budget of the ocean. One way of measuring internal waves produced in the laboratory setting is by a technique called “synthetic schlieren,” whereby the perturbation density field is obtained from the change in index of refraction in the fluid. However, the usual computation of power requires the velocity and pressure, or under certain assumptions, the stream function [Lee et al., “Experimental determination of radiated internal wave power without pressure field data,” Phys. Fluids 26, 046606, (2014)]. We present a method for computing the radiated internal wave power that uses only the perturbation density field, assuming the flow is sufficiently 2-dimensional, and we demonstrate the method using data from simulations and experiments. Monday, November 24, 2014 8:00AM - 9:57AM Session G24 Granular Flows: Mixing, Segregation and Separation Superieure de Lyon — 2003 - Sylvain Joubaud, Ecole Normale 8:00AM G24.00001 Continuum modeling of diffusion and dispersion in dense granular flows1 , IVAN C. CHRISTOV, Theoretical Division & Center for Nonlinear Studies, Los Alamos National Laboratory, HOWARD A. STONE, Mechanical & Aerospace Engineering, Princeton University — Continuum modeling of granular flows remains a challenge of modern statistical physics. Granular materials do not perform Brownian motion, yet diffusion and shear dispersion can be observed in such systems when agitation causes inelastic collisions between particles. In a number of canonical flow regimes (e.g., in a rotating container or down an incline), granular materials can behave like fluids. We formulate and solve the granular counterparts to two basic fluid mechanics problems: diffusion of a pulse and shear dispersion of a pulse for dense granular materials in rapid flow. We provide a theory to account for the concentration-dependent diffusivity of bidisperse granular mixtures, and we give an asymptotic argument for the self-similar behavior of such a diffusion process for which an exact self-similar analytical solution does not exist. For shear dispersion, we show that the effective dispersivity of the depth-averaged concentration of the dispersing powder varies as the Péclet number squared, as in classical Taylor–Aris dispersion of molecular solutes. The calculation is extended to generic shear profiles, showing a significant enhancement for convex profiles due to the shear-rate dependence of the diffusivity of granular materials. 1 ICC was supported by NSF Grant DMS-1104047 and the U.S. DOE through the LANL/LDRD Program; HAS was supported by NSF Grant CBET1234500. 8:13AM G24.00002 Multiscale modelling of multi-component granular mixtures , DEEPAK TUNUGUNTLA, ANTHONY THORNTON, Mathematics of Computational Science and Multiscale Mechanics, STEFAN LUDING, THOMAS WEINHART, Multiscale Mechanics — Efforts to extract, or “coarse grain,” continuum fields (macroscopic dynamics) from microscopic data have existed for decades. We present novel coarse graining expressions for the stress fields for discrete mechanical systems and illustrate their application to segregating granular chute flow. These expressions are applicable near boundaries/interfaces and also to multi-component granular mixtures. Boundary interaction forces are taken into account in a self-consistent way and thus allow for construction of a continuous stress and interaction force field, avoiding problems many other methods have near boundaries. Similarly, stress and drag forces can be determined for individual constituents/components of a mixture. The resolution and shape of the coarse-graining function used in the formulation can be chosen freely, such that both microscopic and macroscopic effects can be studied. The method does not require temporal averaging and thus can be used to investigate time-dependent flows as well as static and steady situations. Furthermore, discrete element simulations of granular mixtures are presented to illustrate the strength of the new boundary/mixture treatment and show that the coarse graining scale (i.e. resolution) is independent of the size of the components (for spheres). 8:26AM G24.00003 Modeling Size Polydisperse Granular Flows1 , RICHARD M. LUEPTOW, CONOR P. SCHLICK, AUSTIN B. ISNER, PAUL B. UMBANHOWAR, JULIO M. OTTINO, Northwestern University — Modeling size segregation of granular materials has important applications in many industrial processes and geophysical phenomena. We have developed a continuum model for granular multi- and polydisperse size segregation based on flow kinematics, which we obtain from discrete element method (DEM) simulations. The segregation depends on dimensionless control parameters that are functions of flow rate, particle sizes, collisional diffusion coefficient, shear rate, and flowing layer depth. To test the theoretical approach, we model segregation in tri-disperse quasi-2D heap flow and log-normally distributed polydisperse quasi-2D chute flow. In both cases, the segregated particle size distributions match results from full-scale DEM simulations and experiments. While the theory was applied to size segregation in steady quasi-2D flows here, the approach can be readily generalized to include additional drivers of segregation such as density and shape as well as other geometries where the flow field can be characterized including rotating tumbler flow and three-dimensional bounded heap flow. 1 Funded by The Dow Chemical Company and NSF Grant CMMI-1000469. 8:39AM G24.00004 Asymmetric flux models for particle-size segregation in granular avalanches , PARMESH GAJJAR, NICO GRAY, Univ of Manchester — Particle-size segregation commonly occurs in dense shallow flows of grains down an incline, through the combined processes of kinetic sieving and squeeze expulsion. Recent experimental observations suggest that a single small particle can percolate downwards through a matrix of large particles faster, than a single large particle can be levered upwards through a matrix of fines. In this work, this asymmetry is modelled using a segregation flux that is dependent only on the small particle concentration. The flux function is asymmetric about its maximum point, differing from the symmetric quadratic form used in recent models of particle size-segregation, and a cubic flux function is used in this work for illustration. Exact solutions are presented for steady non-diffuse flow in two dimensions with both a homogeneously mixed and normally graded inflow, as well as for a steady-state breaking wave. The new asymmetric flux results in a concentration dependence on both the distance to fully segregate, and the length of the breaking wave. 8:52AM G24.00005 A mixture theory for size and density segregation in granular free-surface flows , ANTHONY THORNTON, DEEPAK TUNUGUNTLA, University of Twente — In the past years much work has been undertaken on developing mixture theory continuum models to describe kinetic-sieving driven size-segregation [1-3]. We propose an extension to these models that allows their application to bidisperse flows over inclined channels, with particles varying in density and size [4]. Our model incorporates both a recently proposed explicit formula, for how the total pressure is distributed among different species of particles, of Marks et al. [2], which is one of the key elements of mixture theory-based kinetic sieving models and a shear rate-dependent drag. The resulting model is used to predict the range of particle sizes and densities for which the mixture segregates. The prediction of no segregation in the model is benchmarked by using discrete particle simulations, and good agreement is found when a single fitting parameter is used which determines whether the pressure scales with the diameter, surface area or volume of the particle. [1] [2] [3] [4] Gray and Thornton. Proc. Royal Soc. A, 461:1447-1473, 2005. Marks, Rognon, and Einav. JFM, 690:499-511, 2012. Fan and Hill.NJP, 13(9):095009, 2011. Tunuguntla, Bokhove and Thornton. JFM, 749:99-112, 2014. 9:05AM G24.00006 Evolution of a dynamic suspension created by the invasion of an air flow in a granular bed , TESS HOMAN, VALERIE VIDAL, SYLVAIN JOUBAUD, Laboratoire de Physique-ENS de Lyon — We experimentally investigate the behavior of an immersed granular bed when perturbed by an air inflow from a single inlet at the bottom of a 2D cell. In particular, we focus on quasi-suspensions, meaning that the grains are slightly heavier than the fluid. We observe the creation of a dynamic suspension. We characterize the evolution of the local packing fraction, the percentage of particles mixed in the dynamic suspension and the shape of the “dead zone,” i.e. a region where the grains remain motionless. In particular, we study the influence of the air flow-rate or injection pressure. We complement the study by considering the effect of the density difference between the grains and the fluid, the initial height of the fluid or the height of the bed. 9:18AM G24.00007 Properties of the hydrodynamic profiles of an air-fluidized granular gas1 , FRANCISCO VEGA REYES, Universidad de Extremadura — We study the properties of a non-uniform steady flow in a granular gas that is fluidized by air in turbulent flow. Our granular gas is composed of identical inelastic spheres and is confined between two infinite parallel walls. We show that this system can be accurately described by Navier-Stokes hydrodynamics, even for high inelasticity. We also analyze the properties of segregation of a granular impurity immersed in this granular gas. We focus on the case of flows with uniform heat-flux. We compare air-fludized granular flows with sheared granular gases at uniform heat flux. We find that both types of flows show important similarities in hydrodynamic properties like temperature profile, thermal conductivity, of thermal diffusion coefficient. However, we show that Navier-Stokes hydrodynamics only applies in the case of an air-fluidized granular flow. After solving at Navier-Stokes order the theoretical hydrodynamic profiles for an air-fluidized granular gas with uniform heat flux, we show that they exhibit good agreement with computer simulations of the corresponding the kinetic equation (direct simulation Monte Carlo method). This agreement is independent of the degree of inelasticity of the granular gas, contrary to what would be expected. 1 Financial support from the Spanish Government through grants FIS2010-12587 (partially financed by FEDER funds and by Junta de Extremadura through Grant No. GRU10158) and FIS2013-42840 9:31AM G24.00008 Dynamics of density-inverted/Leidenfrost state in a vibrofluidized bed , ISTAFAUL ANSARI, MEHEBOOB ALAM, Jawaharlal Nehru Ctr Adv Sci — Akin to the original Leidenfrost state, a dense, compact layer of particles can be supported by a dilute gaseous region of fast moving particles underneath it in vertically shaken granular materials– this is dubbed granular Leidenfrost state (gLS). Previous experiments and simulations have noted that the gLS is a stationary state that bifurcates from a time-periodic bouncing bed state with increasing shaking intensity. Here we report a novel unsteady behavior of the gLS in experiments on vertically shaken vibrofluidized bed of a monolayer of spherical balls. With the help of high speed imaging, we track the height of the interface (that separates the dense cloud of particles from the dilute gaseous region) as well as the top surface of the bed at various time instants of the oscillation cycle. Both these quantities are found to vary sinusoidally with time but with different amplitudes and a phase-lag and their oscillation frequencies closely match the frequency of the shaker. The amplitude difference and the phase-lag between the top-surface and interface motions are two distinguishing features of the “oscillatory” gLS. The transition from synchronized oscillatory motion to a probable “steady” gLS with increasing shaking intensity seems to be subtle. 9:44AM G24.00009 Component morphology, size, and compositional impact on pharmaceutical powder blend flowability , DAVID GOLDFARB, HIROTAKA NAKAGAWA, STEPHEN CONWAY, Merck Research Laboratories — Through analysis of particle morphology, particle size, and compositional influences, we present experimental case studies revealing unexpected transitions in flowability and cohesion of pharmaceutical powder blends. We explore interactions between the needle-like API (Active Pharmaceutical Ingredient) and the more spherical remaining components (excipients) in the blend to explain these transitions, and optimal concentrations are identified. A range of particle sizes, aspect ratios (for API), and compositions were examined. Surprisingly, under certain conditions, a blend with a low API concentration exhibits less cohesive flowability properties than a placebo blend containing no API. Effective volume and coordination number models are tested by investigation of particle geometry, particle contact, and Van der Waals force factors. These results should translate both to the improved understanding of mixed component morphology systems and to a novel approach towards pharmaceutical product formulation optimization. Monday, November 24, 2014 8:00AM - 10:10AM Session G25 Turbulence: Theory I — 2005 - Charles Meneveau, Johns Hopkins University 8:00AM G25.00001 The nature of turbulence at high Mach numbers , SHARATH GIRIMAJI, Texas A&M University — We begin by listing the features of a real turbulence flow field and identifying the quintessential characteristics. We contrast these with the main features of wave-like motion. We examine the competition between wave-like and real turbulence characteristics in two elementary flow cases: (i) direct numerical simulation (DNS) of a decaying flow field that is initially anisotropic and purely dilatational; and (ii) DNS and linear analysis of a homogeneously sheared velocity field which is initially entirely solenoidal. The first case examines the non-linear aspects while the latter study addresses linear processes as well. Return-to-isotropy, potential-to-dilatational energy partition and broad-bandedness of the energy spectra are examined. Important, but not necessarily conclusive, arguments are presented. 8:13AM G25.00002 On the direct and inverse energy transfer in 2-dimensional and 3dimensional turbulent flows and in turbulent models1 , GANAPATI SAHOO, LUCA BIFERALE, MASSIMO DE PIETRO, Department of Physics, University of Rome Tor Vergata, Rome, Italy — In this seminar, I will discuss a few important open problems in “Fully Developed Turbulence” concerning its most idealized realization, i.e. the case of statistically homogeneous and isotropic flows. I will discuss the importance of inviscid conserved quantities in relation to the most striking statistical properties shown by all turbulent flows: the growth of small-scales, strongly non-Gaussian fluctuations, including the presence of anomalous scaling laws. By using unconventional numerical methodology, based on a Galerkin decimation of helical Fourier modes [1-3], I will argue that some phenomena characterizing homogeneous and isotropic flows might be important also for a much larger spectrum of applications, including flows with geophysical and astrophysical relevance as for the case of rotating turbulence and/or conducting fluids. Results about both real 3D decimated Navier-Stokes equations and dynamical models of it will be presented. [1] L. Biferale, S. Musacchio and F. Toschi, Phys. Rev. Lett. 108 164501(2012). [2] L. Biferale, E.S. Titi, J. Stat. Phys. 151, 1089(2013). [3] L. Biferale, S. Musacchio and F. Toschi, J. Fluid Mech. 730, 309(2013). 1 Supported by ERC Advanced Grant (N. 339032) “NewTURB.” 8:26AM G25.00003 Turbulence under Fractal Fourier Decimation1 , LUCA BIFERALE, University of Rome “Tor Vergata”, ALESSANDRA LANOTTE, Cnr-Isac, SHIVA MALAPAKA, University of Rome “Tor Vergata”, FEDERICO TOSCHI, Technical University of Eindhoven — We present a systematic investigation of 3D turbulent flows evolved on a highly decimated set of Fourier modes. In particular, we investigate the change in small-scales intermittency when the flow is constrained to excite only a fractal set of modes but keeping the symmetries of the original 3D Navier-Stokes equations. 1 Partially supported by ERC Grant No 339032. 8:39AM G25.00004 Depletion of nonlinearity in two-dimensional turbulence , ANDREY PUSHKAREV, WOUTER BOS, LMFA - CNRS, Ecole Centrale de Lyon, France, ROBERT RUBINSTEIN, Newport News, VA, USA — The strength of the nonlinearity is measured in decaying two-dimensional turbulence, by comparing its value to that found in a Gaussian field. It is shown how the nonlinearity drops following a two-step process. First a fast relaxation is observed on a timescale comparable to the time of formation of vortical structures, as also observed in 3 dimensions [1], then at long times the nonlinearity relaxes further during the phase when the eddies merge to form the final dynamic state of decay. Both processes seem roughly independent of the value of the Reynolds number. [1] Bos, W. J. T., & Rubinstein, R. (2013). On the strength of the nonlinearity in isotropic turbulence. Journal of Fluid Mechanics, 733, 158-170. 8:52AM G25.00005 Angular statistics of fluid particle trajectories in turbulence , WOUTER BOS, LMFA, CNRS, Ecole Centrale de Lyon, Université de Lyon, France, BENJAMIN KADOCH, IUSTI-CNRS, Aix-Marseille University, Marseille, France, KAI SCHNEIDER, M2P2-CNRS & CMI Aix-Marseille University, Marseille, France — The angle between subsequent particle displacement increments is evaluated as a function of the time lag, following a recent proposition by Burov et al. [1]. First, the link between the investigated angle and the curvature of the trajectories is explained. Subsequently we compare the Lagrangian trajectories in two-dimensional periodic and wall-bounded turbulent flows. We show that at long times the probability density function of the angles carries the signature of the confining domain if finite size effects are present. At short times, the PDF of the cosine of the angle is given by a power law with a well defined exponent, reminiscent of the close to Gaussian character of the velocity field. [1] Burov, S., Tabei, S. A., Huynh, T., Murrell, M. P., Philipson, L. H., Rice, S. A. & Dinner, A. R. (2013). Distribution of directional change as a signature of complex dynamics. Proc. Natl. Acad. Sci., 110(49), 19689-19694. 9:05AM G25.00006 Backwards Two-Particle Dispersion in a Turbulent Flow1 , THEODORE DRIVAS, Johns Hopkins University — We derive an exact equation governing two-particle backwards mean-squared dispersion for both deterministic and stochastic tracer particles in turbulent flows. For the deterministic trajectories, we probe consequences of our formula for short time and arrive at approximate expressions for the mean squared dispersion which involve second order structure functions of the velocity and acceleration fields. For the stochastic trajectories, we analytically calculate an exact t3 contribution to the squared separation and additionally compute the average dispersion using direct numerical simulation (DNS) results of incompressible homogeneous isotropic turbulence. We find that this exactly calculable term accounts for almost all of the observed behavior. We argue that this contribution also appears to describe the asymptotic Richardson-like behavior for deterministic paths and present DNS results to support this claim. 1 Partially supported by NSF Grant No. CDI-II: CMMI 0941530 at Johns Hopkins University. 9:18AM G25.00007 Energy Spectra in Weakly Compressible and Isothermal Turbulence , GUOWEI HE, YUFENG DONG, LNM, Institute of Mechanics, Chinese Academy of Sciences — The universal scaling of energy spectra of velocity fluctuations is fundamentally important to understand turbulent flows. For incompressible turbulence, the universal scaling -5/3 of energy spectra is originally proposed by Kolmogorov, based on dimensional analysis. This empirical result is further derived from the Navier-Stokes equations, using the two-point closure approaches. However, for compressible turbulence, the dimensional analysis is difficult to be conducted due to nonlinear coupling of velocity, density and pressure. In this paper, we will use a two-point closure approach, EDQNM, to derive the universal scaling of energy spectra for compressible and isothermal turbulence. In the EDQNM equations, the eddy-damping rates are determined by the recently developed swept-wave model for space-time correlations (Phys. Rev. E 88 021001R, 2013). The leading term in the eddy-damping rates leads to the -7/3 scaling for dilatational energy spectra, while the sub-leading one leads to the -3 scaling. The former implies that dilatational components are dominated by acoustic-wave time scales; the latter implies that dilatational components dominated by local straining time scales. Our DNS result appears to favor the -7/3 scaling. This study clarifies the possible scaling of compressible energy spectra in terms of space-time correlations. 9:31AM G25.00008 Local behavior of streamlines in turbulent flows , JONAS BOSCHUNG, FABIAN HENNIG, NORBERT PETERS, Institute for Combustion Technology, RWTH Aachen University — Although streamlines have often been used mainly to visualize flow fields, they have been studied in recent years to some extent in the search for a better, more intuitive description and decomposition of the flow field. Streamlines seem a good candidate, as they are tangential to the velocity field and thus are prescribed by its structure. Similarly to the Q-R-classification of flow topologies, it is possible to classify the behavior of streamlines in an absolute sense by the unit vector gradient tensor and its first and second invariant H and K. The invariants are found to have a physical interpretation, inasmuch as they are a measure for the local net convergence or divergence of the streamlines and its rate of change, respectively. The joint pdf of H and K is evaluated for different Reynolds-numbers from 119 to 330. It is found that streamlines expand rapidly while shrinking gently. As the local flow behavior is determined by the invariants, several quantities are conditioned on H and K in order to relate them to the structure of the flow. 9:44AM G25.00009 Symmetry-plane models of 3D Euler fluid equations: Analytical solutions and finite-time blowup using infinitesimal Lie-symmetry methods1 , MIGUEL D. BUSTAMANTE, Complex and Adaptive Systems Laboratory, School of Mathematical Sciences, University College Dublin — We consider 3D Euler fluids endowed with a discrete symmetry whereby the velocity field is invariant under mirror reflections about a 2D surface known as the “symmetry plane.” This type of flow is widely used in numerical simulations of classical/magnetic/quantum turbulence and vortex reconnection. On the 2D symmetry plane, the governing equations are best written in terms of two scalars: vorticity and stretching rate of vorticity. These determine the velocity field on the symmetry plane. However, the governing equations are not closed, because of the contribution of a single pressure term that depends on the full 3D velocity profile. By modelling this pressure term we propose a one-parameter family of sensible models for the flow along the 2D symmetry plane. We apply the method of infinitesimal Lie symmetries and solve the governing equations analytically for the two scalars as functions of time. We show how the value of the model’s parameter determines if the analytical solution has a finite-time blowup and obtain explicit formulae for the blowup time. We validate the models by showing that a particular choice of the model’s parameter corresponds to a well-known exact solution of 3D Euler equations [Gibbon et al., Physica D 132:497 (1999)]. We discuss practical applications. 1 Supported by Science Foundation Ireland (SFI) under Grant Number 12/IP/1491. 9:57AM G25.00010 Joint Statistics of Finite Time Lyapunov Exponents in Isotropic Turbulence1 , PERRY JOHNSON, CHARLES MENEVEAU, Johns Hopkins University — Recently, the notion of Lagrangian Coherent Structures (LCS) has gained attention as a tool for qualitative visualization of flow features. LCS visualize repelling and attracting manifolds marked by local ridges in the field of maximal and minimal finite-time Lyapunov exponents (FTLE), respectively. To provide a quantitative characterization of FTLEs, the statistical theory of large deviations can be used based on the so-called Cramér function. To obtain the Cramér function from data, we use both the method based on measuring moments and measuring histograms (with finite-size correction). We generalize the formalism to characterize the joint distributions of the two independent FTLEs in 3D. The “joint Cramér function of turbulence” is measured from the Johns Hopkins Turbulence Databases (JHTDB) isotropic simulation at Reλ = 433 and results are compared with those computed using only the symmetric part of the velocity gradient tensor, as well as with those of instantaneous strain-rate eigenvalues. We also extend the large-deviation theory to study the statistics of the ratio of FTLEs. When using only the strain contribution of the velocity gradient, the maximal FTLE nearly doubles in magnitude and the most likely ratio of FTLEs changes from 4:1:-5 to 8:3:-11, highlighting the role of rotation in de-correlating the fluid deformations along particle paths. 1 Supported by NSF Graduate Fellowship (DGE-1232825), a JHU graduate Fellowship, and NSF grant CMMI-0941530. CM thanks Prof. Luca Biferale for useful discussions on the subject. Monday, November 24, 2014 8:00AM - 10:10AM Session G26 Rough Wall Boundary Layers I — 2007 - Karen Flack, United States Naval Academy 8:00AM G26.00001 Turbulent transport of momentum and scalars in urban-like geometries , QI LI, ELIE BOU-ZEID, Department of Civil and Environmental Engineering, Princeton University, US, WILLIAM ANDERSON, Department of Mechanical Engineering, University of Texas at Dallas, US, SUE GRIMMOND, Department of Meteorology, University of Reading, UK — A numerical study is carried out using large-eddy simulations to investigate the mechanisms of turbulent transport of momentum and passive scalars over urban-like geometries. The immersed boundary method is used to represent buildings; this induces “ringing,” i.e. the Gibbs phenomenon associated with the use of spectral discretization in domains with sharp discontinuities. We present a new approach to reduce this ringing and improve the numerical accuracy of the method. Topological parameters, such as the frontal area index and the plan area index, were varied to examine their impact on turbulence and transport characteristics. The heterogeneity of the surface is shown to increase both the heteorogeneity and anisotropy of the flow, and to significantly modulate the efficiencies of momentum and passive scalar transport. 8:13AM G26.00002 Optimizing the determination of roughness parameters of urban canopies , AUVI RAHMAN, PABLO HUQ, University of Delaware — We present an optimization procedure to determine the roughness parameters for an urban canopy. The mean velocity profile above an urban canopy is described by the log law via the roughness parameters: zero-plane displacement height d, roughness length z0 , and friction velocity u∗ . Traditionally these parameters are obtained from a single mean velocity profile. We have devised a new procedure which is akin to the bootstrap or jackknife resampling method where multiple mean velocity profiles are generated from a single mean velocity profile. Each of the generated profiles are then best fit to the log law and sets of d, z0 , and u∗ are estimated. These sets of values show distinct clusters when plotted against the relative sensitivity of the log law to the zero-plane displacement height d. A single representative or optimal value of the roughness parameters are then obtained automatically by utilizing a standard clustering procedure. Application of this method is also presented for field and laboratory data. 8:26AM G26.00003 Turbulent boundary layer flow over broad-banded roughness , GENO PAWLAK, PAYAM AGHSAEE, University of California San Diego, SAEED MAZROUEI, STEFANO LEONARDI, University of Texas Dallas, KRISHNAKUMAR RAJAGOPALAN, MARCELO KOBAYASHI, University of Hawaii at Manoa — The response of the boundary layer to a regular roughness is often parameterized in terms of the length scales defining the roughness. Difficulty arises in the case of broad-banded and highly irregular roughness distributions such as over coral reefs or urban canopies where the length scale that determines the response of the boundary layer is not clear. Here we use a spectral description for roughness to create idealized two-dimensional irregular roughness profiles, using square waves as a basis function. Laboratory experiments along with Direct Numerical Simulations (DNS) are used to examine the hydrodynamic response to the broad-banded roughness and flow characteristics are related to geometric characteristics of the boundary. The simulations and experiments show that the nature of the flow over two-dimensional irregular walls can be determined as a function of the hydrodynamic origin, which, in turn, can be determined as a function of a mean cavity shape. Results are interpreted in terms of the spectral characteristics of the roughness. The contribution of the various spectral components to the total drag is analyzed for each case. The roughness spectrum influences the flow through the shape of the cavities on the wall and can provide some guidance in predicting the nature of the flow. 8:39AM G26.00004 Effect of surface morphology on drag and roughness sublayer in flows over regular roughness elements , MARCO PLACIDI1 , BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — The effects of systematically varied roughness morphology on bulk drag and on the spatial structure of turbulent boundary layers are examined by performing a series of wind tunnel experiments. In this study, rough surfaces consisting of regularly and uniformly distributed LEGOTM bricks are employed. Twelve different patterns are adopted in order to methodically examine the individual effects of frontal solidity (λF , frontal area of the roughness elements per unit wall-parallel area) and plan solidity (λP , plan area of roughness elements per unit wall-parallel area), on both the bulk drag and the turbulence structure. A floating element friction balance based on Krogstad & Efros (2010) was designed and manufactured to measure the drag generated by the different surfaces. In parallel, high resolution planar and stereoscopic Particle Image Velocimetry (PIV) was applied to investigate the flow features. This talk will focus on the effects of each solidity parameter on the bulk drag and attempt to relate the observed trends to the flow structures in the roughness sublayer. 1 Currently at City University London 8:52AM G26.00005 Near-wall 3D velocity measurements above biomimetic shark skin denticles using Digital In-line Holographic Microscopy , MOSTAFA TOLOUI, DAVID BRAJKOVIC, JIARONG HONG, University of Minnesota — Digital In-line Holography is employed to image 3D flow structures in the vicinity of a transparent rough surface consisting of closely packed biomimetic shark skin denticles as roughness elements. The 3D printed surface replicates the morphological features of real shark skin, and the denticles have a geometrical scale of 2 mm, i.e. 10 times of the real ones. In order to minimize optical aberrations near the fluid-roughness interface and enable flow measurements around denticles, the optical refractive index of the fluid medium is maintained the same as that of the denticle model in an index-matched flow facility using NaI solution as the working fluid. The experiment is conducted in a 1.2 m long test section with 50 mm × 50 mm cross section. The sampling volume is located in the downstream region of a shark skin replica of 12” stretch where the turbulent flow is fully-developed and the transitional effect from smooth to the rough surface becomes negligible. Several instantaneous realizations of the 3D velocity field are obtained and are used to illustrate turbulent coherent structures induced by shark-skin denticles. This information will provide insights on the hydrodynamic function of shark’s unique surface ornamentation. 9:05AM G26.00006 The impact of algal biofilms on skin-friction in a turbulent channel flow1 , MICHAEL SCHULTZ, KAREN FLACK, CECILY STEPPE, U.S. Naval Academy, JESSICA WALKER, University of Tasmania — Experiments were carried out in a fully-developed, turbulent channel flow facility over a wide Reynolds number range. The wall shear stress was determined using the bulk flow rate and the streamwise pressure gradient in the downstream section of the channel. A biofilm dominated by three species of diatoms developed on acrylic test surfaces exposed for four days in a brackish tidal environment at the United States Naval Academy. The resulting biofilm had an average thickness of 200 µm. This biofilm had a significant effect on the flow showing a doubling of the skin-friction compared to the hydraulically-smooth condition at the highest Reynolds number. Scale up of the present results to ship scale indicates that this biofilm would generate an 18% powering penalty for a mid-sized naval ship at cruising speed. 1 This research was funded by ONR. 9:18AM G26.00007 Numerical simulation of adverse-pressure-gradient boundary layer with or without roughness , POUYA MOTTAGHIAN, JUNLIN YUAN, UGO PIOMELLI, Queen’s University — Large-eddy and direct numerical simulations are carried out on flat-plate boundary layer over smooth and rough surfaces, with adverse pressure gradient.The deceleration is achieved by imposing a wall-normal freestream velocity profile, and is strong enough to cause separation at the wall. The Reynolds number based on momentum thickness and freestream velocity at inlet is 600. Numerical sandgrain roughness is applied based on an immersed boundary method, yielding a flow that is transitionally rough. The turbulence intensity increases before separation, and reaches a higher value for the rough case, indicating stronger mixing. Roughness also causes higher momentum deficit near the wall, leading to earlier separation. This is consistent with previous observation made on rough-wall flow separation over a ramp. In both cases, the turbulent kinetic energy peaks inside the shear layer above the detachment region, with higher values in the rough case; it then decreases approaching the reattachment region. Near the wall inside the separation bubble, the near-zero turbulent intensity indicates that the turbulent structures are lifted up in the separation region. Compared with the smooth case, the shear layer is farther from the wall and the reattachment length is longer on the rough wall. 9:31AM G26.00008 DNS of turbulent channel flow over hemispherical roughness , SICONG WU, University of Illinois at Urbana-Champaign, KENNETH CHRISTENSEN, University of Notre Dame, CARLOS PANTANO, University of Illinois at UrbanaChampaign — Turbulent channel flows over certain rough surfaces have been studied using direct numerical simulation (DNS) in recent years. Most of these previous studies have focused on roughness with closely packed cubic or cylindrical ribs and it is well documented that the near-wall flow is strongly affected by the roughness but the outer region is relatively unaffected, in agreement with previous experiment evidence. In this study, DNS of turbulent channel flow with hexagonally packed hemispheres on a wall is performed. The roughness height k/h is about 10 and the average spacing between hemispheres from center to center is of the order of 6 times of the roughness height. The friction Reynolds number is approximately 400 and the simulation employs the NEK5000 solver, an incompressible Navier-Stokes code based on spectral elements. We will discuss detailed turbulence statistics in the near-wall region and forces on the rough surface, with the aim of improving understanding of flow physics to guide development of reliable LES models. 9:44AM G26.00009 Effects of roughness on accelerating boundary layer , JUNLIN YUAN, UGO PIOMELLI, Queen’s University — Large-eddy simulation is carried out on a rough-wall boundary layer with favourable pressure gradient (FPG) to study the combined effects of FPG and roughness. The acceleration is strong enough to start relaminarization on a smooth wall; the fully rough regime is achieved in the FPG region. Unlike the flow over a smooth wall, where FPG causes significant Reynolds-stress anisotropy and decoupling between the inner and outer layers, on the rough wall, the interaction between inner and outer layers is amplified near the roughness crest by FPG, due to the increased Reynolds shear stress associated with strong sweeping events, which cause large fluxes of turbulent kinetic energy towards the wall. Spatial variations of time-averaged velocities in the roughness sublayer are observed. They scale with friction velocity and the roughness length scale and, in the FPG region, they increase in magnitude due to the work of the mean flow against the form drag. The spatial disturbances of time-averaged Reynolds stresses play an important role, by sustaining the vertical turbulent motions through a production mechanism, and subsequently lead to the increase in Reynolds shear stress. As a result, relaminarization is not achieved on the rough wall. 9:57AM G26.00010 Direct numerical simulation of a turbulent pipe with systematically varied three-dimensional roughness , LEON CHAN, MICHAEL MACDONALD, DANIEL CHUNG, NICHOLAS HUTCHINS, ANDREW OOI, University of Melbourne — Direct Numerical Simulations (DNS) are conducted at low to medium Reynolds numbers for a turbulent pipe flow with roughness. The roughness, which is comprised of three-dimensional sinusoidal elements, causes a downward shift in the mean velocity profile known as the Hama roughness function ∆U + . In engineering applications, ∆U + (which is related to the coefficient of drag Cf ) is an important parameter as it is used to quantify the increase in drag and the decrease in efficiency. To have a better understanding of roughness and how it affects the flow, a range of numerical studies were conducted where the roughness height h+ , wavelength λ+ and Reynolds number of the flow are varied. For the range of cases simulated, it is found that the roughness average height ka+ (which is proportional to h+ ) is strongly correlated to the roughness function ∆U + whereas λ+ has a weaker influence on the flow. Results from simulations of more complicated surfaces comprised of two superimposed modes of different wavelength are also presented. Analysis of the turbulence statistics convincingly supports Townsend’s outer-layer hypothesis for all of the cases simulated. Monday, November 24, 2014 8:00AM - 10:10AM Session G27 Turbulent Boundary Layers IV — 2009 - Stefano Leonardi, University of Texas at Austin 8:00AM G27.00001 Unsteady boundary layer detachment in planar flows at large Reynolds number , ROMAIN NGUYEN VAN YEN, Freie Universität Berlin, MARIE FARGE, ENS, CNRS, Paris, MATTHIAS WAIDMANN, RUPERT KLEIN, Freie Universität Berlin, KAI SCHNEIDER, Aix Marseille Univerité — We study a vortex dipole impinging onto a wall with two different models: Navier-Stokes equations and Euler’s equation coupled with Prandtl’s boundary layer equation. The solutions in the limit of large Reynolds number Re are computed by DNS performed using a high-order compact finite differences scheme with no-slip boundary conditions. For both models we first observe the formation on the wall of two opposite-sign boundary layers whose thickness scales in Re−1/2 , as predicted by Prandtl in 1904. At a later time tD the solution of the Navier-Stokes equation shows that the boundary layers suddenly collapse down to thickness as fine as Re−1 , as predicted by Kato in 1984, then detach from the wall and roll up into strongly dissipative new dipoles that are ejected away from the wall into the bulk flow. In contrast, at the same time tD Prandtl’s solution becomes singular and the boundary layers can no more be computed, while Euler’s solution gives two opposite sign-vortices that slip along the wall in opposite directions without detaching. 8:13AM G27.00002 Direct Numerical Simulation of an Adverse Pressure Gradient Turbulent Boundary Layer at the Verge of Separation1 , VASSILI KITSIOS, CALLUM ATKINSON, Monash University, JUAN SILLERO, BORRELL GUILLEM, Universidad Politécnica de Madrid, AYSE GUNGOR, Istanbul Technical University, JAVIER JIMENÉZ, Universidad Politécnica de Madrid, JULIO SORIA, Monash University — We investigate the structure of an adverse pressure gradient (APG) turbulent boundary layer (TBL) at the verge of separation. The intended flow is generated via direct numerical simulation (DNS). The adopted DNS code was previously developed for a zero pressure gradient TBL. Here the farfield boundary condition (BC) is modified to generate the desired APG flow. The input parameters required for the APG BC are initially estimated from a series of Reynolds Averaged Navier-Stokes simulations. The BC is implemented into the DNS code with further refinement of the BC performed. The behaviour of the large scale dynamics is illustrated via the extraction of coherent structures from the DNS using analysis of the velocity gradient tensor and vortex clustering techniques. 1 The authors acknowledge the research funding from the Australian Research Council and European Research Council, and the computational resources provided by NCI and PRACE. 8:26AM G27.00003 Evolution of the Reynolds shear stresses in highly accelerated turbulent boundary layers , GUILLERMO ARAYA, LUCIANO CASTILLO, FAZLE HUSSAIN, Texas Tech University — Turbulent boundary layers subjected to severe acceleration or strong Favorable Pressure Gradients (FPG) are of great fundamental and technological importance; examples of the latter include nozzle design, underwater bodies and drag reduction applications. Scientifically, they pose great interest from the point of view of scaling laws, the complex interaction between the outer and inner regions, and relaminarization phenomena. Direct Numerical Simulations (DNS) of highly accelerated turbulent boundary layers are performed by means of the Dynamic Multi-scale Approach (DMA) recently developed by [Araya et al. JFM, vol. 670, pp. 581-605, 2011]. It is shown that the Reynolds shear stress monotonically decreases and exhibits a logarithmic layer in the meso-layer region during the laminarization process. In addition, the local maxima of streamwise velocity fluctuations in wall units remain almost constant in the very strong FPG region, which prevents the flow to become completely laminar. Furthermore, the re-distribution of Reynolds shear stresses due to sweeps and ejections in the FPG region is performed and a physical mechanism is proposed. 8:39AM G27.00004 Structure of relaminarizing turbulent boundary layers , O. RAMESH, SAURABH PATWARDHAN, Indian Institute of Science — Relaminarization of a turbulent boundary layer in a strongly accelerated flow has received a great attention in recent times. It has been found that such relaminarization is a general and regularly occurring phenomenon in the leading-edge region of a swept wing of an airplane (van Dam et. al., 1993). In this work, we investigate the effect of initial Reynolds number on the process of relaminarization in turbulent boundary layers. The experimental and numerical investigation of relaminarizing turbulent boundary layers undergoing same history reveals that the boundary layer with higher initial Reynolds number relaminarizes at a lower pressure gradient value compared to the one with lower Reynolds number. This effect can be explained on the inviscid theory proposed earlier in the literature. Further, various parameter criteria proposed to predict relaminarization, are assessed and the structure of relaminarizing boundary layers is investigated. A mechanism for stabilization of near-wall low speed streaks is proposed. 8:52AM G27.00005 Pressure measurements in a rapidly sheared turbulent wall layer1 , SOURABH DIWAN, JONATHAN MORRISON, Imperial College London — The aim of the present work is to improve understanding of the role of pressure fluctuations in the generation of coherent structures in wall-bounded turbulent flows, with particular regard to the rapid and slow source terms. The work is in part motivated by the recent numerical simulations of Sharma et al. (Phy. Fluids, 23, 2011), which showed the importance of pressure fluctuations (and their spatial gradients) in the dynamics of large-scale turbulent motions. Our experimental design consists of first generating a shearless boundary layer in a wind tunnel by passing a grid-generated turbulent flow over a moving floor whose speed is matched to the freestream velocity, and then shearing it rapidly by passing it over a stationary floor further downstream. Close to the leading edge of the stationary floor, the resulting flow is expected to satisfy the approximations of the Rapid Distortion Theory and therefore would be an ideal candidate for studying linear processes in wall turbulence. We carry out pressure measurements on the wall as well as within the flow – the former using surface mounted pressure transducers and the latter using a static pressure probe similar in design to that used by Tsuji et al. (J. Fluid. Mech. 585, 2007). We also present a comparison between the rapidly sheared flow and a more conventional boundary layer subjected to a turbulent free stream. 1 We acknowledge the financial support from EPSRC (grant no. EP/I037938) 9:05AM G27.00006 The effects of localized blowing on pressure-velocity correlation , CAN LIU, GUILLERMO ARAYA, LUCIANO CASTILLO, Department of Mechanical Engineering, Texas Tech University, STEFANO LEONARDI, Department of Mechanical Engineering, University of Texas at Dallas — It is well known that wall pressure fluctuations are footprints of the large coherent motions existent in the outer region of the boundary layer. In this investigation, spatial-temporal correlations of the pressure and velocity fields are computed in a spatially-developing turbulent channel flow with five-blowing jets located at the bottom wall and along the spanwise direction. Direct numerical simulations are performed at a friction Reynolds number of 394. The main purpose behind the present study is to assess the influence of perturbing blowing jets on the large scale structures of the turbulent channel flow. Furthermore, the key role of pressure fluctuations on the energy redistribution among the velocity components is scrutinized by computing the energy budgets of tke and Reynolds stresses, and a physical mechanism is proposed to explain the outer peak on turbulence production due to localized blowing. 9:18AM G27.00007 Quantifying Hairpin Vortex Generation1 , RIJAN MAHARJAN2 , DANIEL SABATINO, Lafayette College — Hairpin vortices are artificially generated via fluid injection through a streamwise oriented slot into an otherwise laminar boundary layer in a free-surface water channel. Injection through the 32:1 aspect ratio slot is intended to approximate the behavior of a low speed streak along with its neighboring streamwise vortices that spawn naturally occurring hairpins in fully turbulent boundary layers. A parametric study is performed by varying the slot streamwise location, the average injection flow rate and injection duration. Hairpins are examined for boundary layer conditions between 485 < Reδ∗ < 600 and blowing ratios up to 0.2. Cross-stream 2D-PIV is primarily used to characterize the injection profile and the strength of the initial streamwise vortices as well as establish the strength and structure of the resulting hairpin for each condition. The role of the streamwise vorticity in the generation of the hairpin is examined. Threshold conditions which will yield hairpins that have sufficient strength to autogenerate secondary hairpins are also considered. 1 Supported 2 Now by the National Science Foundation under Grant CBET-1040236 at Yale University 9:31AM G27.00008 Evolution of Lagrangian structures in the K-type temporal transition in channel flow , YAOMIN ZHAO, YUE YANG, SHIYI CHEN, State Key Laboratory for Turbulence and Complex Systems, Peking University — We report a Lagrangian study on the evolution of hairpin vortices in the K-type temporal transition in a channel flow. Based on the Eulerian velocity field from the direct numerical simulation, a backward-particle-tracking method is used to solve the Lagrangian scalar transport equation, and Lagrangian material surfaces are extracted as isosurfaces of the Lagrangian scalar. As an approximation of the Helmholtz vorticity theorem, a Lagrangian surface, which is initially a vortex surface, can be approximately as a vortex surface before significant vortex reconnections in a time evolution (Yang and Pullin, J. Fluid. Mech., 2010). Thus, by tracking the evolution of Lagrangian material surfaces in the early transitional phase, the dynamics of hairpin vortices can be studied in a Lagrangian framework. In the present study, the Lagrangian surface evolves from a streamwise-spanwise vortex sheet to a Λ-shaped bulge, and then rolls up into a hairpin-shaped structure. The dynamical evolution of the Lagrangian hairpin vortex is analysed in consecutive times. With the comparison of the coherent structures identified by the Eulerian criteria (e.g., ‘λ2 -criterion’), differences between Lagrangian and Eulerian structures are discussed. 9:44AM G27.00009 DNS of the flow around a wall-mounted square cylinder under various inflow conditions , RICARDO VINUESA, PHILIPP SCHLATTER, JOHAN MALM, DAN S. HENNINGSON, KTH Mechanics, CATHERINE MAVRIPLIS, University of Ottawa — The flow around a wall-mounted square cylinder is investigated by means of DNS. The effect of inflow conditions is assessed by considering two different cases with matching Reθ ≃ 1000 at the obstacle: the first case is a fully-turbulent zero pressure gradient boundary layer, and the second one is a laminar boundary layer with prescribed Blasius inflow profile. An auxiliary simulation carried out with the pseudo-spectral code SIMSON is used to obtain time-dependent inflow conditions which are then fed into the main simulation where the actual flow around the cylinder is computed. This main simulation is performed, for both laminar and turbulent inflows, with the spectral element code Nek5000. Transition to turbulence is observed in the laminar case, induced by the recirculation bubble produced at the obstacle. In both cases we find the same Strouhal number St = 0.1, in good agreement with available experimental measurements, although the two wakes exhibit structural differences associated with turbulent wall-normal transport and spanwise fluctuations. In the turbulent case the streamwise fluctuations modulate the horseshoe vortex formed around the cylinder. Additional insight on the differences between both wakes is achieved by means of a POD study of the flow. 9:57AM G27.00010 Turbulent structures in a wall jet , SHIBANI BHATT, SRAVAN ARTHAM, REDA MANKBADI, EBENEZER GNANAMANICKAM, Embry-Riddle Aeronautical Univ — Wall Jets are special shear layers, in that they have two shear layers which arise from two different instability mechanisms, to form a single turbulent layer. The three-dimensional wall jet in addition has an inherent secondary flow which adds to the complexity of the flow physics. These jets finds wide use in heat transfer applications such as film cooling. However, the wall jet has received little attention when compared to the turbulent boundary layer, fully developed channel flow or a free jet. As part of this study hot-wire measurements were carried out in a three-dimensional wall jet. The eventual goal of this work is to study inner-outer interactions using the wall jet as the two shear layers affords the possibility of independent control. Velocity statistics as well as spectra derived from velocity measurements are presented. The wall jet is shown to have structures of longer wavelength in the outer free-jet region which arise from the inviscid instability of the free-jet layer. The near wall region of the wall jet, which arises from a viscous instability, is populated with finer structures similar to that seen in a turbulent boundary layer. The implications of these observations towards studying inner-outer interactions is also discussed. Monday, November 24, 2014 8:00AM - 10:10AM Session G28 Turbulence: Mixing I — 2011 - Filippo Coletti, University of New Mexico 8:00AM G28.00001 The influence of coherent structures on the turbulent dispersion of a passive scalar plume1 , CHRISTINA VANDERWEL, STAVROS TAVOULARIS, University of Ottawa — We investigated the influence of coherent structures on the dispersion of a passive scalar by studying instantaneous measurements of a plume of dye released in uniformly sheared flow generated in a water tunnel. Measurements were performed using simultaneous stereo particle image velocimetry and planar laser-induced fluorescence to obtain instantaneous concentration and velocity maps in cross-sections normal to the flow direction. Coherent vortices were observed to effectuate scalar transport by inducing motions which displaced dyed fluid. Dye was observed to preferentially congregate within vortex cores and far away from vortices, whereas regions adjacent to vortices were less likely to contain dye. A conditional eddy analysis demonstrated that counter-rotating vortex pairs associated with hairpin vortices were responsible for both large Reynolds stress events and large scalar flux events. This observation was supported by the fact that the Reynolds stress was found to be correlated with the scalar flux. 1 Supported by NSERC 8:13AM G28.00002 Turbulent Mixing of Jet in Crossflow with Compound Angle Injection , KEVIN RYAN, Stanford University, FILIPPO COLETTI, University of Minnesota, CHRISTOPHER ELKINS, JOHN EATON, Stanford University — A dominant feature governing the development of the jet in crossflow is a pair of longitudinal vortices that originate at the point of injection. These vortices cause a distortion of the jet and promote mixing of the jet and mainstream fluid. The vortex structure is significantly altered for jets with compound angle injection, with respect to jets with no skew relative to the mainstream flow. In skewed geometries, a single dominant vortex controls the development of the jet and mixing of the jet fluid with the mainstream. The 3D velocity and concentration fields were measured for a compound angle jet injected with a skew angle of 30 degrees relative to the incoming flow. Measurements were conducted using magnetic resonance imaging (MRI) techniques using water as the working fluid. The development of the vorticity was investigated at the point of injection. The effect of the single dominant vortex on the turbulent mixing of the jet fluid with the mainstream was evaluated using the scalar concentration field. Spreading of the jet due to turbulent mixing is shown to be highly asymmetric. Results obtained for the skewed jet were compared to an angled jet in crossflow with no skew. 8:26AM G28.00003 Turbulent Scalar Flux Modeling for an Inclined Jet in Crossflow: An Analysis of the Error Incurred by Various Modeling Assumptions , JULIA LING, Sandia National Lab, KEVIN RYAN, JOHN EATON, Stanford University — In order to use Reynolds-Averaged Navier Stokes (RANS) solvers to determine a passive scalar concentration distribution, it is necessary to model the turbulent scalar fluxes. Various models have been proposed for these turbulent scalar fluxes, each of which relies on a different set of basic assumptions. The gradient diffusion hypothesis assumes that the turbulent scalar fluxes can be modeled as diffusive fluxes with an isotropic diffusivity. A fixed turbulent Schmidt number model relies on the Reynolds analogy between momentum and scalar transport. Higher order algebraic closures, such as the Generalized Gradient Diffusion Hypothesis (GGDH), break away from the isotropic assumption inherent in a fixed turbulent Schmidt number model. The error from each of these modeling assumptions was isolated and evaluated through the analysis of a high-fidelity Large Eddy Simulation (LES) for an inclined jet in crossflow configuration. The LES velocity field and Reynolds stresses were fed into a RANS solver and the scalar distribution was calculated using various turbulent scalar flux models. This methodology removed the compounding effect of errors in the velocity field and Reynolds stresses, enabling the direct evaluation of the isotropic assumption and the Reynolds analogy. It was shown that neither of these assumptions were appropriate for this flow. However, it was also demonstrated that the largest source of error in the scalar flux modeling was due to poorly tuned model coefficients, not to any particular modeling assumption. 8:39AM G28.00004 Flow and turbulence structure in a shallow mixing layer developing over a flat surface at high Reynolds numbers , GOKHAN KIRKIL, Kadir Has University — Results of a high resolution Detached Eddy Simulation (DES) are used to characterize the evolution of a shallow mixing layer developing between two parallel streams in a long open channel with a flat surface at a high Reynolds number (Re=250,000). The study discusses the influence of Reynolds number on the development of the mixing layer as well as the vertical non-uniformity in the mixing layer structure and provides a quantitative characterization of the growth of the large-scale coherent structures with the distance from the splitter plate. Mixing layer growth rate and its change in the vertical direction are compared with experiments and a simulation at Re=16,000. Power spectra of the horizontal velocity components are examined for the presence of a -3 and -5/3 subranges at streamwise locations away from the splitter plate. Passive scalar is introduced at the tip of the splitter plate close to the free surface to estimate the size of the mixing structures based on mass transport. The effect of the Reynolds number on the shift of the centerline of the mixing layer is quantified. 8:52AM G28.00005 Effect of background turbulence on the mixing of a passive scalar within a turbulent jet1 , ALEJANDRO PÉREZ ALVARADO, Department of Mechanical Engineering, McGill University, SUSAN GASKIN, Department of Civil Engineering and Applied Mechanics, McGill University, LAURENT MYDLARSKI, Department of Mechanical Engineering, McGill University — The vast majority of the research on turbulent jets has studied those emitted into quiescent (or laminar) backgrounds. Yet, most jets in environmental or engineering applications are emitted into turbulent backgrounds. It is therefore imperative to fully understand the underlying physics of jets emitted into turbulent environments. The present investigation builds on previous work (Khorsandi, Gaskin and Mydlarski, J. Fluid Mech., 2013, which studied the effect of background turbulence on the velocity field of a turbulent jet) and examines the mixing of a (high-Schmidt-number) passive scalar within a turbulent jet that is emitted into a turbulent environment. A quasi-homogeneous and isotropic, zero-mean-flow turbulent background was generated by means of a random jet array. The concentration field within the turbulent jet was measured using planar laser induced fluorescence. We will present results pertaining to the evolution of the statistical moments of the scalar field, as well as its characteristic length, relative that of a jet emitted into a quiescent background, and as a function of the intensity of the background turbulence. The role of background turbulence on the jet entrainment mechanism will also be addressed. 1 Support for this work was provided by NSERC. 9:05AM G28.00006 Mixing efficiency in two-dimensional turbulent and chaotic flows , BENJAMIN KADOCH, IUSTI-CNRS, Aix-Marseille University, Marseille, France, WOUTER BOS, LMFA-CNRS, Ecole Centrale de Lyon, Université de Lyon, Ecully, France, KAI SCHNEIDER, M2P2-CNRS & CMI Aix-Marseille University, Marseille, France — We investigate the mixing in a flow generated by a circular rod, describing a figure-eight-shaped motion in a two-dimensional circular vessel. The vessel, the moving rod, and the equations of motion are modeled using a volume penalization method imbedded in a classical Fourier pseudo-spectral code as described in [1]. The influence of the Peclet number on the mixing efficiency is measured for different Stokes and turbulent regimes. Here, the mixing efficiency is measured by evaluating the decay of passive scalar fluctuations for a given energy injection rate. The Stokes regime shows results similar to the ones obtained in [2] for chaotic mixing. For instance, the passive scalar variance decays following a powerlaw, related to the presence of unmixed fluid near the fixed walls, which acts as a reservoir for the mixing away from the wall. For the turbulent regimes, however, the detachment of vorticity in the boundary layer more efficiently injects the unmixed fluid into the domain. [1] B. Kadoch, D. Kolomenskiy, K. Schneider and P. Angot. J. Comput. Phys., 231, 4365-4383, 2012. [2] E. Gouillart, O. Dauchot, B. Dubrulle, S S. Roux, and J.-L. Thiffeault. Phys. Rev. E 78, 026211, 2008 9:18AM G28.00007 Statistics and Scaling Laws of Turbulent Mixing at High Reynolds Numbers , MICHAEL GAUDING, MARKUS HEMPEL, CHRISTIAN HASSE, TU Freiberg — We examine the turbulent mixing of passive scalars with imposed mean gradient and varying diffusivities by means of direct numerical simulation. The transport mechanism within the turbulent cascade is altered when differential diffusion is present. In order to analyze this effect, we derive from first principles an equation in correlation space that quantifies differential diffusion. This equation captures the balance between inter-scale transport, diffusive transport, scalar dissipation, as well as a transport that originates from unequal diffusivities between the involved scalars. This equation is not closed but each term can be analyzed by means of direct numerical simulation. To this end, direct numerical simulations have been conducted with Taylor based Reynolds number varying between 88 and 529. The Schmidt number is varied between 1/8 and 1. 9:31AM G28.00008 Intermittency and universality of small scales of passive scalar in turbulence1 , TOSHIYUKI GOTOH, TAKESHI WATANABE, Nagoya Institute of Technology — Recent experiments and Direct Numerical Simula- tions (DNSs) suggest that the small scale statistics of passive scalar may not be as “universal” as in the velocity case. To address this problem, we study the moments of scalar increment in steady turbulence at Rλ > 800 by using DNS up to the grid points of 40963 . In order for the scalar and turbulent flow to be as faithful as possible to the assumptions that would be made in theories, Scalar 1 and Scalar 2 are simultaneously convected by the identical isotropic turbulent flow but excited by two different methods. Scalar 1 is excited by the random scalar injection which is isotropic, Gaussian and white in time at low wavenumber band, while Scalar 2 is excited by the uniform mean scalar gradient. The moments of two scalars as functions of the separation vector are expanded in terms of the Legendre polynomials to extract the scaling exponents of the moments up to the 4th anisotropic sector for Scalar 2. It is found that the exponents of the isotropic sectors seem to have the same values at separation distances in the narrow range over which the 4/3 law holds simultaneously for two scalars. The exponents of the anisotropic sectors and the cumulants of the moments will also be reported. 1 HPCI, JHPCN, Grant-in-Aid for Sci. Res. No.24360068, Ministry of Edu. Sci., Japan 9:44AM G28.00009 Optimal mixing of a passive scalar by supercritical 3D plane Poiseuille flow1 , LUKAS VERMACH, Cambridge Centre for Analysis, University of Cambridge, C.P. CAULFIELD, BPI & DAMTP, University of Cambridge — We consider a passive zero-mean scalar field organised into two layers of different concentration, in a 3D plane channel subjected to a constant along-stream pressure gradient. We employ a fully nonlinear adjoint-looping approach to identify the optimal initial perturbation of the velocity field with given initial energy which yields maximal mixing by a target time horizon, in the sense of minimisation of the spatially-integrated variance of the concentration field. Foures et.al. (JFM, 2014) considered 2D plane Poiseuille flow at a sufficiently low (subcritical) Re ∼ 500 to not be subject to flow instabilities, and demonstrated that the initial perturbation which maximizes the time-averaged energy gain of the flow leads to weak mixing, and is qualitatively different from the optimal initial “mixing” perturbation which exploits classical Taylor dispersion. We generalise this study to the optimisation of mixing three-dimensional flows at a range of significantly higher (supercritical) Reynolds numbers, showing how the potential triggering of “strong” flow instabilities modifies the structure of the optimal initial mixing perturbation qualitatively. 1 This work was supported by the UK Engineering and Physical Sciences Research Council (EPSRC) grant EP/H023348/1 for the University of Cambridge Centre for Doctoral Training, the Cambridge Centre for Analysis. 9:57AM G28.00010 Normalizations of High Taylor Reynolds Number Power Spectra1 , ALEJANDRO PUGA, TIMOTHY KOSTER, JOHN C. LARUE, University of California, Irvine — The velocity power spectrum provides insight in how the turbulent kinetic energy is transferred from larger to smaller scales. Wind tunnel experiments are conducted where high intensity turbulence is generated by means of an active turbulence grid modeled after Makita’s 1991 design (Makita, 1991) as implemented by Mydlarski and Warhaft (M&W, 1998). The goal of this study is to document the evolution of the scaling region and assess the relative collapse of several proposed normalizations over a range of Rλ from 185 to 997. As predicted by Kolmogorov (1963), an asymptotic approach of the slope (n) of the inertial subrange to −5/3 with increasing Rλ is observed. There are three velocity power spectrum normalizations as presented by Kolmogorov (1963), Von Karman and Howarth (1938) and George (1992). Results show that the Von Karman and Howarth normalization does not collapse the velocity power spectrum as well as the Kolmogorov and George normalizations. The Kolmogorov normalization does a good job of collapsing the velocity power spectrum in the normalized high wavenumber range of 0.0002 ≤ κλ ≤ 0.4 while the George normalization does a better job in the normalized mid-wavenumber range of 15 ≤ κλ ≤ 25. 1 University of California, Irvine Research Fund Monday, November 24, 2014 8:00AM - 10:10AM — Session G29 Experimental Techniques: Flow Visualization and Quantitative Imaging 2014 - Chris Elkins, Stanford University 8:00AM G29.00001 High Spatial Resolution Femtosecond Laser Measurements of Turbulence1 , MATTHEW EDWARDS, ARTHUR DOGARIU, RICHARD MILES, Princeton University — Experimental study of turbulence at length scales smaller than the Taylor microscale can provide unique information about isotropy, homogeneity, and intermittency. We use femtosecond laser electronic excitation tagging (FLEET), an experimentally simple and unseeded molecular tagging method, to probe small-scale turbulent structures in air and nitrogen, measuring velocity with better than 100-micron spatial resolution. At these scales, the density perturbation from the tagging process may influence measured turbulence parameters. Here, we quantify the effect of small perturbations during measurement on the observed statistics of turbulence and explore the spatial limits at which FLEET can be employed. 1 This research has been supported by AFOSR and an NSF Fellowship 8:13AM G29.00002 Multi-photon Molecular Tagging Velocimetry with Femtosecond Excitation (FemtoMTV)1 , MANOOCHEHR KOOCHESFAHANI, SHAHRAM POUYA, ALEXANDER VAN RHIJN, MARCOS DANTUS, Michigan State University — We report results from first MTV measurements in water under nonlinear resonant femtosecond excitation of a phosphorescent supramolecule. Both two-photon and three-photon absorption processes are examined and the feasibility of measurements is demonstrated by single component velocimetry in a simple jet flow. The new capabilities enabled by FemtoMTV include elimination of the need for short wavelength UV excitation source and UV optical access in flow facilities, and potential for high rep-rate flow imaging. 1 This work was supported by the Air Force Office of Scientific Research, Grant number FA9550-13-1-0034. 8:26AM G29.00003 Spatiotemporal evolution of a laser-induced shock wave measured by the background-oriented schlieren technique , YOSHIYUKI TAGAWA, SHOTA YAMAMOTO, MASAHARU KAMEDA, Tokyo Univ of Agri & Tech — We investigate the spatiotemporal evolution of a laser-induced shock wave in a liquid filled thin tube. In order to measure pressure distribution at shock front, we adopt the background-oriented schlieren (BOS) technique. This technique provides two- or three-dimensional pressure field in a small region with a simple setup. With an ultra high-speed video camera and a laser stroboscope, we successfully capture the spatial evolution of the shock every 0.2 µs. We find an angular variation of the pressure at the shock front. The maximum pressure is in the direction of the laser shot while the minimum value is in the perpendicular direction. We compare the temporal evolution of the pressure measured by BOS technique with those obtained by another method, i.e. pressure estimation from the shock front position. Overall trend from both methods show a good agreement. The pressure from the shock front position exists between the maximum and minimum values from BOS technique. It indicates that our quantification method can measure more detailed pressure field in twoor three-dimensions. Our results might be used for the efficient generation systems for the microjet, which can be applicable for needle free injection devices. 8:39AM G29.00004 New applications of focusing schlieren imaging , MICHAEL HARGATHER, STEWART YOUNGBLOOD, New Mexico Tech — Focusing schlieren is a refractive imaging technique that visualizes refractive disturbances in a limited depth of field. Whereas traditional schlieren visualizes refractive disturbances along an entire optical path, focusing schlieren can be used to see inside of a refractive flow or to eliminate disturbances outside of a defined test section. The basic optical layout and design of a focusing schlieren system are reviewed. Comparisons between traditional schlieren and focusing schlieren images are presented to highlight the ability to selectively image refractive disturbances. The imaging technique is applied to measuring quantitative density fields with low- and high-speed applications. Additional applications to refractive feature tracking and schlieren image velocimetry are presented. 8:52AM G29.00005 Resolving gas-liquid interface geometry using light field imaging , ALEXANDER JAFEK, Brigham Young University, JESSE BELDEN, Naval Undersea Warfare Center, TADD TRUSCOTT, Brigham Young University — We present a novel approach for reconstructing the geometry of a three-dimensional specular gas-liquid interface from an image captured by a light-field camera. Whereas the scanning of a diffuse surface can be accomplished with a simple projector-camera system, the local reconstruction of a specular surface is non-unique and requires a more constrained sampling method. In our set-up, a known array of laser points is reflected by the unknown specular surface onto the image plane of a light-field camera. For each illuminated pixel, possible surfaces are generated that are defined by a depth location and local surface normal vector. We show that when the aperture is sufficiently small we can find the exact location and orientation of the local surface. Further, we present an algorithm that allows us to reconstruct a reflective surface from images that are taken with wider apertures. The algorithm searches the possible surfaces for points and normal vectors that are most consistent with each other based on input parameters. We present our simulated results with experimental validation. 9:05AM G29.00006 Fabrication of temperature sensitive hollow micro capsule for the flow visualization , SATOSHI SOMEYA, FUMIO TAKEMURA, TETSUO MUNAKATA, National Institute of Advanced Industrial Science and Technology — Temperature and oxygen sensitive hollow micro capsules were fabricated using the bubble template method. The micro bubbles were nucleated in droplets of a dichloromethan solution of polymer. The polymer covered micro bubbles were suspended in aqueous solution. The dichloromethan solution of temperature and oxygen sensitive dye was dissolved into the polymer solution and the temperature and oxygen sensitive dye was incorporated into the capsule shell. Using the bubble template method, large amount of hollow micro capsules could be formed with very high number of density. The diameter of capsules was 1 ∼ 3 micro meter and the specific gravity of capsules was 1.01 g/cm3 . They seemed to be suitable as tracer particles for the PIV measurement. The temperature sensitivity and the oxygen sensitivity of fluorescence intensity from the functional capsules were measured using a spectrometer. The effect of excitation wavelength on these sensitivities and the quenching due to large excitation intensity were also evaluated. The temperature sensitivity was about -2%/◦ C and the fluorescence intensity was stable and no quenching was detected in 20 minutes, even under the intense excitation of 1W. 9:18AM G29.00007 Characterization of Scalar Mixing in Dense Gaseous Jets Using X-Ray Computed Tomography1 , JARED DUNNMON, TAE WOOK KIM, ANTHONY KOVSCEK, REBECCA FAHRIG, MATTHIAS IHME, Stanford Univ — An experimental technique based on X-Ray Computed Tomography (XCT) is used to characterize scalar mixing of gaseous jets at Reynolds numbers up to 20,000. In this study, the mixing of a krypton jet with ambient air is considered. The high radiodensity of the krypton gas enables non-intrusive volumetric measurements of gas density and mixture composition based on spatial variations in x-ray attenuation. Comparisons to theoretical and computational results are presented, and the viability of this diagnostic technique is assessed. Important aspects of x-ray attenuation theory and practice are considered in data processing and their impacts on future development of this technique are discussed. 1 Support from DoD through the NDSEG Fellowship Program and from NIH through S10 Shared Instrumentation Grant S10RR026714-01 are gratefully acknowledged. 9:31AM G29.00008 Measurement of density gradient across wind turbine interface , VIRGILIO GOMEZ, AMELIA TAYLOR, ARQUIMEDES RUIZ-COLUMBIE, SUHAS POL, CARSTEN WESTERGAARD, LUCIANO CASTILLO, Texas Tech University — The wake of a field installed model turbine was visualized using a large-scale shadowgraph apparatus. To enable a large field of view a focused shadowgraph apparatus was used where the camera lens and the light source axis were aligned. A retroreflective screen is used as a back plane to reflect the image back to the camera. Sonic anemometer measurements of velocity and temperature were obtained at points overlapping the field of view. As much as 2% change in temperature has been observed within wake, enough to cause measurable index of refraction fluctuations. Schlieren method will be used to directly measure the density gradient across the wake interface. These measurements will be used to explain the dynamics at the wake interface for different atmospheric boundary layer stability (stratification) conditions. 9:44AM G29.00009 Water turbulence effects on vapor blanket present in a flat-ended cylindrical probe at high temperature and, cooled by forced convection , ALBERTO CERVANTES GARCIA, Universidad Michoacana de San Nicolás de Hidalgo, MARTIN HERREJÓN ESCUTIA, Instituto Tecnológico de Morelia, GILDARDO SOLORIO DÍAZ, Universidad Michoacana de San Nicolás de Hidalgo, HÉCTOR JAVIER VERGARA HERNÁNDEZ, Instituto Tecnológico de Morelia, ALICIA AGUILAR CORONA, Universidad Michoacana de San Nicolás de Hidalgo, JOSE ROBERTO ZENIT CAMACHO, Universidad Nacional Autónoma de México — In this work, the effect of turbulent flow on the vapor blanket, which originates in cooling by with water of a flat-end cylindrical probe was studied to observe the flow patterns around the vapor layer and its effect on heat extraction. The experiments to visualize the vapor blanket were carried out in an experimental device using flat-end cylindrical probes made with AISI 304 stainless steel. The probe was heated up to 915 ◦ C and plunged into a tube of plexiglass in which water was flowing. For helping to visualize the streamlines within the fluid, polyamide particles were added to the flow and illuminated with a sheet of laser light to visualize a slice of fluid flow pattern near the probe surface. Water velocities were considered: 0.2 m/s, 0.4 m/s and 0.6 m/s. During each experiment, thermal response data was acquired at 10 Hz. Results show that the vapor layer around the cylinder is stable at water 0.2 m/s, however, when the water velocity is increased, the flow becomes more turbulent, and the vapor layer becomes unstable, and vapor is entrained by eddies originate in the water. All this is reflected in the thermal histories. 9:57AM G29.00010 Development of Naphthalene PLIF for Making Quantitative Measurements of Ablation Products Transport in Supersonic Flows1 , CHRISTOPHER COMBS, NOEL CLEMENS, The University of Texas at Austin — Ablation is a multi-physics process involving heat and mass transfer and codes aiming to predict ablation are in need of experimental data pertaining to the turbulent transport of ablation products for validation. Low-temperature sublimating ablators such as naphthalene can be used to create a limited physics problem and simulate ablation at relatively low temperature conditions. At The University of Texas at Austin, a technique is being developed that uses planar laser-induced fluorescence (PLIF) of naphthalene to visualize the transport of ablation products in a supersonic flow. In the current work, naphthalene PLIF will be used to make quantitative measurements of the concentration of ablation products in a Mach 5 turbulent boundary layer. For this technique to be used for quantitative research in supersonic wind tunnel facilities, the fluorescence properties of naphthalene must first be investigated over a wide range of state conditions and excitation wavelengths. The resulting calibration of naphthalene fluorescence will be applied to the PLIF images of ablation from a boundary layer plug, yielding 2-D fields of naphthalene mole fraction. These images may help provide data necessary to validate computational models of ablative thermal protection systems for reentry vehicles. 1 Work supported by NASA Space Technology Research Fellowship Program under grant NNX11AN55H Monday, November 24, 2014 8:00AM - 10:10AM Session G30 Aerodynamics: Flapping and Flexible Wings — 2016 - 8:00AM G30.00001 Influence of Ground Effect on Low Aspect Ratio Membrane Wings , ROBERT BLEISCHWITZ, PhD-student, ROELAND DE KAT, Research Fellow, BHARATHRAM GANAPATHISUBRAMANI, Professor, University of Southampton — Inspired by the current interest of membrane wings for Micro Air Vehicles (M AV s) and hard limits in aerodynamic performance for wings in moderate Reynolds number regimes, an experimental wind tunnel study is conducted at a Reynolds number of approximately 65,000 to determine the aeromechanics of flexible, low aspect ratio (AR) membrane wings (AR ≤ 2) in the vicinity of the ground. Pitch angle α and height over ground h/c is varied with a traverse system. Flexible membrane wings are compared with rigid flat plates. A rolling road is used to impose the ground effect and the boundary layer leading up to the road is removed using a suction system. Time-averaged lift, drag and pitch moment changes are captured with a 6-axis force transducer and its effects are interpreted in terms of the membrane motions obtained using Direct-Image-Correlation (DIC) technique. Flow-structure-ground interactions are examined and the membrane dynamics are compared to results obtained outside of ground effect. Ultimately, understanding the ground effect on flexible membrane wings at moderate Reynolds numbers could help to design Wing-in-Ground M AV s with extended range and reduced energy consumption. 8:13AM G30.00002 Lift on Flexible and Rigid Cambered Wings at High Incidence1 , ANYA JONES, PETER MANCINI, University of Maryland, KENNETH GRANLUND, MICHAEL OL, U.S. Air Force Research Laboratory — The effects of camber and camber change due to elastic deflection of a membrane wing were investigated for wings in rectilinear translation with parameter variations in wing incidence and acceleration. Direct force and moment measurements were performed on a rigid flat plate wing, rigid cambered wings, and a membrane wing. Features in the force histories were further examined via flow visualization by planar laser illumination of fluorescent dye. Below 10 degrees of incidence, Wagner’s approximation accurately predicts the time-evolution of lift for the rigid wings. At higher incidence, flow separation results in force transients, and the effect of wing camber is no longer additive. Both the rigid flat plate and rigid cambered wings reach peak lift at a 35 degree angle of attack, whereas the flexible wing experiences stall delay and reaches peak lift at 50 degrees. Due to the aeroelasticity of the flexible membrane, flow over the suction surface remains attached for much higher incidence angles than for the rigid wings. For incidence angles less than 30 degrees, the peak lift of the flexible wing is lower than that of its rigid counterparts. Beyond 30 degrees, the flexible wing experiences an aeroelastically induced stall delay that allows lift to exceed the rigid analogs 1 This work was supported by the Air Force Office of Scientific Research (AFOSR) Summer Faculty Fellowship Program and the U.S. Army Research Laboratory under the Micro Autonomous Systems and Technology (MAST) program. 8:26AM G30.00003 Characteristics of the flow around tandem flapping wings , LUKE MUSCUTT, BHARATHRAM GANAPATHISUBRAMANI, GABRIEL WEYMOUTH, The University of Southampton, THE UNIVERSITY OF SOUTHAMPTON TEAM — Vortex recapture is a fundamental fluid mechanics phenomenon which is important to many fields. Any large scale vorticity contained within a freestream flow may affect the aerodynamic properties of a downstream body. In the case of tandem flapping wings, the front wing generates strong large scale vorticity which impinges on the hind wing. The characteristics of this interaction are greatly affected by the spacing, and the phase of flapping between the front and rear wings. The interaction of the vorticity of the rear wing with the shed vorticity of the front wing may be constructive or destructive, increasing thrust or efficiency of the hind wing when compared to a wing operating in isolation. Knowledge of the parameter space where the maximum increases in these are obtained is important for the development of tandem wing unmanned air and underwater vehicles, commercial aerospace and renewable energy applications. This question is addressed with a combined computational and experimental approach, and a discussion of these is presented. 8:39AM G30.00004 Aerodynamics of flapping insect wing in inclined stroke plane hovering with ground effect , KRISHNE GOWDA V, S. VENGADESAN, Indian Institute of Technology - Madras — This work presents the time-varying aerodynamic forces and the unsteady flow structures of flapping insect wing in inclined stroke plane hovering with ground effect. Two-dimensional dragonfly model wing is chosen and the incompressible Navier-Stokes equations are solved numerically by using immersed boundary method. The main objective of the present work is to analyze the ground effect on the unsteady forces and vortical structures for the inclined stroke plane motions. We also investigate the influences of kinematics parameters such as Reynolds number (Re), stroke amplitude, wing rotational timing, for various distances between the airfoil and the ground. The effects of aforementioned parameters together with ground effect, on the stroke averaged force coefficients and regimes of force behavior are similar in both normal (horizontal) and inclined stroke plane motions. However, the evolution of the vortex structures which produces the effects are entirely different. 8:52AM G30.00005 Flapping flight: effect of asymmetric kinematics , NAKUL PANDE, SIDDHARTH KRITHIVASAN, SREENIVAS K.R., Jawaharlal Nehru Centre for Advanced Scientific Research — Flapping flight has received considerable attention in the past with its relevance in the design of micro-air vehicles. In this regard, asymmetric flapping of wings offers simple kinematics. Nevertheless, it leads to symmetry-breaking in the flow field and generation of sustained lift. It has been observed previously with flow visualization experiments and Discrete Vortex Method (DVM) simulations that if the down-stroke time period is lesser than the up-stroke time, there is a net downward momentum imparted to the fluid. This is seen as a switching the flow field from a four-jet (symmetric) to a two-jet (asymmetric) configuration when the stroke-time ratio is progressively varied. This symmetry breaking has been studied experimentally using Particle Image Velocimetry (PIV) across a range of Reynolds Numbers and asymmetry ratios. Results are also corroborated with results from 3-D numerical simulations. Study helps in shedding light on the effectiveness of asymmetric kinematics as a lift generation mechanism. 9:05AM G30.00006 Effect of wing flexibility on phasing of tandem wings in forward flight , VISHAL NAIDU1 , JOHN YOUNG2 , JOSEPH LAI3 , University of New South Wales — The dragonfly with two pairs of wings in tandem uses different phases between the wing pairs to suit the needs of the flight. Previous studies to understand the effect of phasing in forward flight are based on rigid wings. This is in contrast to the highly flexible dragonfly wings, with varying spanwise and chordwise flexibility. Here, we study flexible flapping wing simulations using Fluid Structure Interaction (FSI) in forward flight, at an advance ratio of 0.3 and Reynolds number of approximately 1300. The FSI simulations are carried out for phase 90◦ (hindwing leading), 0◦ (in-phase) and 180◦ (anti-phase). The performance of flexible wings will be compared with that of the rigid wings and the effect of flexibility will be discussed. 1 PhD Student Lecturer 3 Professor 2 Senior 9:18AM G30.00007 High amplitude surging and plunging motions at low Reynolds number , JEESOON CHOI, TIM COLONIUS, Caltech, DAVID WILLIAMS, IIT, CALTECH COLLABORATION, IIT COLLABORATION — Aerodynamic forces and flow structures associated with high amplitude oscillations of an airfoil in the streamwise (surging) and transverse (plunging) direction are investigated in two-dimensional simulations at low Reynolds number (Re=102 ∼ 103 ). While the unsteady aerodynamic forces for low-amplitude motions were mainly affected by the leading-edge vortex (LEV) acting in- or out-of phase with the quasi-component of velocity, large-amplitude motions involve complex vortex interactions of LEVs and trailing-edge vortices (TEVs) with the moving body. For high-amplitude surging, the TEV, instead of the LEV, induces low-pressure regions above the airfoil during the retreating portion of the cycle near the reduced frequency, k=0.5, and enhances the time-average forces. The time required for the LEV to convect along the chord becomes an intrinsic time scale, and for plunging motions, there is a sudden change of flow structure when the period of the motion is not long enough for the LEV to convect through the whole chord. 9:31AM G30.00008 Using adjoint-based optimization to study wing flexibility in flapping flight1 , MINGJUN WEI, MIN XU, New Mexico State University, HAIBO DONG, University of Virginia — In the study of flapping-wing flight of birds and insects, it is important to understand the impact of wing flexibility/deformation on aerodynamic performance. However, the large control space from the complexity of wing deformation and kinematics makes usual parametric study very difficult or sometimes impossible. Since the adjoint-based approach for sensitivity study and optimization strategy is a process with its cost independent of the number of input parameters, it becomes an attractive approach in our study. Traditionally, adjoint equation and sensitivity are derived in a fluid domain with fixed solid boundaries. Moving boundary is only allowed when its motion is not part of control effort. Otherwise, the derivation becomes either problematic or too complex to be feasible. Using non-cylindrical calculus to deal with boundary deformation solves this problem in a very simple and still mathematically rigorous manner. Thus, it allows to apply adjoint-based optimization in the study of flapping wing flexibility. We applied the “improved” adjoint-based method to study the flexibility of both two-dimensional and three-dimensional flapping wings, where the flapping trajectory and deformation are described by either model functions or real data from the flight of dragonflies. 1 Supported by AFOSR 9:44AM G30.00009 Flow Structure and Force Variation with Aspect Ratio for a Two-Degreeof-Freedom Flapping Wing1 , MATTHEW BURGE, JAMES FAVALE, MATTHEW RINGUETTE, State University of New York at Buffalo — We investigate experimentally the effect of aspect ratio (AR) on the flow structure and forces of a two-degree-of-freedom flapping wing. Flapping wings are known to produce complex and unsteady vortex loop structures, and the objective is to characterize their variation with AR and how this influences the lift force. Previous results on rotating wings demonstrated that changes in AR significantly affect the three-dimensional flow structure and lift coefficient. This is primarily due to the relatively greater influence of the tip vortex for lower AR. At Reynolds number of order O(103 ) we test wings of AR = 2-4, values typically found in nature, with simplified planform shapes. The lift force is measured using a submersible transducer at the base of the wing in a glycerin-water mixture. The qualitative, three-dimensional vortex loop structure for different ARs is obtained using multi-color dye flow visualization. Guided by this, quantitative three-component flow information, namely vorticity, the Q-criterion, and circulation, is acquired from stereoscopic particle image velocimetry in key planes. Of interest is how these parameters and the vortex loop topology vary with AR, and their connection to features in the unsteady force signal. 1 This work is supported by the National Science Foundation, Award Number 1336548, supervised by Dr. Dimitrios Papavassiliou. 9:57AM G30.00010 Modeling unsteady forces and pressures on a rapidly pitching airfoil1 , NICOLE K. SCHIAVONE, SCOTT T.M. DAWSON, CLARENCE W. ROWLEY, Princeton University, DAVID R. WILLIAMS, Illinois Institute of Technology — This work develops models to quantify and understand the unsteady aerodynamic forces arising from rapid pitching motion of a NACA0012 airfoil at a Reynolds number of 50 000. The system identification procedure applies a generalized DMD-type algorithm to time-resolved wind tunnel measurements of the lift and drag forces, as well as the pressure at six locations on the suction surface of the airfoil. Models are identified for 5-degree pitch-up and pitch-down maneuvers within the overall range of 0–20 degrees. The identified models can accurately capture the effects of flow separation and leading-edge vortex formation and convection. We demonstrate that switching between different linear models can give accurate prediction of the nonlinear behavior that is present in high-amplitude maneuvers. The models are accurate for a wide-range of motions, including pitch-and-hold, sinusoidal, and pseudo-random pitching maneuvers. Providing the models access to a subset of the measured data channels can allow for improved estimates of the remaining states via the use of a Kalman filter, suggesting that the modeling framework could be useful for aerodynamic control applications. 1 This work was supported by the Air Force Office of Scientific Research, under award No. FA9550-12-1-0075. Monday, November 24, 2014 8:00AM - 10:10AM Session G31 CFD: Applications — 2018 - Andy Nonaka, Lawrence Berkely National Laboratory 8:00AM G31.00001 CFD Aided Design and Production of Hydraulic Turbines1 , ALPER KAPLAN, HUSEYIN CETINTURK, GIZEM DEMIREL, ECE AYLI, KUTAY CELEBIOGLU, SELIN ARADAG, TOBB University of Economics and Technology, ETU HYDRO RESEARCH CENTER TEAM — Hydraulic turbines are turbo machines which produce electricity from hydraulic energy. Francis type turbines are the most common one in use today. The design of these turbines requires high engineering effort since each turbine is tailor made due to different head and discharge. Therefore each component of the turbine is designed specifically. During the last decades, Computational Fluid Dynamics (CFD) has become very useful tool to predict hydraulic machinery performance and save time and money for designers. This paper describes a design methodology to optimize a Francis turbine by integrating theoretical and experimental fundamentals of hydraulic machines and commercial CFD codes. Specific turbines are designed and manufactured with the help of a collaborative CFD/CAD/CAM methodology based on computational fluid dynamics and five-axis machining for hydraulic electric power plants. The details are presented in this study. 1 This study is financially supported by Turkish Ministry of Development. 8:13AM G31.00002 Application of Computational Fluid Dynamics Model to Disinfection Reactors in Water Reclamation Plants , ANDREA HELMNS, PABLO TEXEIRA, EMIN ISSAKHANIAN, JOSE SAEZ, Loyola Marymount Univ — California’s current drought has renewed public interest in recycled water from Water Reclamation Plants (WRPs). It is critical that the recycled water meets public health standards. This project consists of simulating the transport of an instantaneous conservative tracer through the chlorine contact tanks at two WRPs in California, where recycled water regulations stipulate a minimum 90-minute modal contact time during disinfection at peak dry weather design flow. Computational Fluid Dynamics (CFD) is used to model the turbulent flow, transport, and contact time of a conservative solute for several real operating scenarios. Given as-built drawings and operation parameters, the chlorine contact tanks are modeled to match actual geometries and flow conditions. The turbulent flow solutions are used as the basis to model the transport of a turbulently diffusing conservative tracer added instantaneously to the inlet of the reactors. This tracer simulates the transport through advection and dispersion of chlorine in the WRPs. Breakthrough curves of the tracer at the outlet are used to determine the modal contact times. 8:26AM G31.00003 Large-eddy simulation of turbulent flows around a fin-tube heat exchanger enclosed by a compartment , CHANGKEUN SON, Dept. Mech. Eng., Sogang Univ., Seoul, South Korea, SIMON SONG, JEESOO LEE, Dept. Mech. Convergence Eng., Hanyang Univ., Seoul, South Korea, SEONGWON KANG, Dept. Mech. Eng., Sogang Univ., Seoul, South Korea — The main objective of the present study is to analyze heat transfer and flow characteristics of a heat exchanger in an industrial application using high-fidelity simulation techniques. Large-eddy simulations (LES) were performed to investigate the turbulent flows around a fin-tube heat exchanger enclosed by a compartment. The complex geometry of the compartment poses a difficulty in a simulation as the local Re number is about two orders of different magnitude, and generates various scales of the 3-D vortices and complex flow patterns. Careful tests with both grid resolution and turbulent inflow boundary condition were performed in order to compare our results to the measured data from a MRV experiment as well as the results from RANS simulations. From interaction of the flow structures such as the 3-D vortices, a few interesting flow phenomena were observed which are different from a plain fin-tube heat exchanger, such as helical flows and a jet stream observed behind the fin-tube region. Also, performance of the heat exchanger was analyzed using the data from plain fin-tube heat exchangers. Based on this analysis, a numerical technique for heat exchanger was devised and tested to show a possibility of reducing the computational cost significantly, using a porous media model. 8:39AM G31.00004 Control of Flow Structure in Square Cross-Sectioned U Bend using Numerical Modeling1 , MEHMET METIN YAVUZ, YIGITCAN GUDEN, Middle East Technical University (METU) — Due to the curvature in U-bends, the flow development involves complex flow structures including Dean vortices and high levels of turbulence that are quite critical in considering noise problems and structural failure of the ducts. Computational fluid dynamic (CFD) models are developed using ANSYS Fluent to analyze and to control the flow structure in a square cross-sectioned U-bend with a radius of curvature Rc /D=0.65. The predictions of velocity profiles on different angular positions of the U-bend are compared against the experimental results available in the literature and the previous numerical studies. The performances of different turbulence models are evaluated to propose the best numerical approach that has high accuracy with reduced computation time. The numerical results of the present study indicate improvements with respect to the previous numerical predictions and very good agreement with the available experimental results. In addition, a flow control technique is utilized to regulate the flow inside the bend. The elimination of Dean vortices along with significant reduction in turbulence levels in different cross flow planes are successfully achieved when the flow control technique is applied. 1 The project is supported by Meteksan Defense Industries, Inc. 8:52AM G31.00005 Development of an Open Source Image-Based Flow Modeling Software SimVascular1 , ADAM UPDEGROVE, University of California, Berkeley, JAMESON MERKOW, DANIELE SCHIAVAZZI, University of California, San Diego, NATHAN WILSON, University of California, Los Angeles, ALISON MARSDEN, University of California, San Diego, SHAWN SHADDEN, University of California, Berkeley — SimVascular (www.simvascular.org) is currently the only comprehensive software package that provides a complete pipeline from medical image data segmentation to patient specific blood flow simulation. This software and its derivatives have been used in hundreds of conference abstracts and peer-reviewed journal articles, as well as the foundation of medical startups. SimVascular was initially released in August 2007, yet major challenges and deterrents for new adopters were the requirement of licensing three expensive commercial libraries utilized by the software, a complicated build process, and a lack of documentation, support and organized maintenance. In the past year, the SimVascular team has made significant progress to integrate open source alternatives for the linear solver, solid modeling, and mesh generation commercial libraries required by the original public release. In addition, the build system, available distributions, and graphical user interface have been significantly enhanced. Finally, the software has been updated to enable users to directly run simulations using models and boundary condition values, included in the Vascular Model Repository (vascularmodel.org). In this presentation we will briefly overview the capabilities of the new SimVascular 2.0 release. 1 National Science Foundation 9:05AM G31.00006 Effect of Flow Rate Controller on Liquid Steel Flow in Continuous Casting Mold using Numerical Modeling1 , KADIR ALI GURSOY, MEHMET METIN YAVUZ, Middle East Technical University — In continuous casting operation of steel, the flow through tundish to the mold can be controlled by different flow rate control systems including stopper rod and slide-gate. Ladle changes in continuous casting machines result in liquid steel level changes in tundishes. During this transient event of production, the flow rate controller opening is increased to reduce the pressure drop across the opening which helps to keep the mass flow rate at the desired level for the reduced liquid steel level in tundish. In the present study, computational fluid dynamic (CFD) models are developed to investigate the effect of flow rate controller on mold flow structure, and particularly to understand the effect of flow controller opening on meniscus flow. First, a detailed validation of the CFD models is conducted using available experimental data and the performances of different turbulence models are compared. Then, the constant throughput casting operations for different flow rate controller openings are simulated to quantify the opening effect on meniscus region. The results indicate that the meniscus velocities are significantly affected by the flow rate controller and its opening level. The steady state operations, specified as constant throughput casting, do not provide the same mold flow if the controller opening is altered. Thus, for quality and castability purposes, adjusting the flow controller opening to obtain the fixed mold flow structure is proposed. 1 Supported by Middle East Technical University (METU) BAP (Scientific Research Projects) Coordination 9:18AM G31.00007 Geometric VOF-PLIC simulations of Hollow Cone Sprays , THOMAS NELSON, MICHAEL BENSON, BRETT VANPOPPEL, US Military Acad, LUIS BRAVO, Army Research Lab, USMA TEAM SPRAY COLLABORATION, ARL VEHICLE COMBUSTION LAB COLLABORATION, STANFORD UNIVERSITY COLLABORATION — This work examines a Computational Fluid Dynamics (CFD) approach to provide temporally resolved simulations of a novel pressure swirl atomizer presently studied at Stanford University [1]. In a pressure swirl atomizer, the liquid spreads out to form an air-cored vortex within the nozzle and an emerging thin annular film. Due to instabilities the film breaks up to form a hollow cone spray. The numerical simulations focus on the near field nozzle flow physics and primary atomization of the spray. An incompressible flow formulation is adopted with a geometric unsplit Volume of Fluid (VOF) method to track the interface between two immiscible fluids in interfacial flow simulations. Here, the interface is modeled via an advection equation implicitly tracked using a discrete indicator function, f, with values representing the volume fraction of the tagged fluid within a cell. An Adaptive Mesh Refinement (AMR) scheme is also employed to efficiently capture the shear layers near the liquid-gas interface. The study is carried out for two atomizers with 2mm and 3mm diameters at intermediate Re = 2.6-3.9x103, We=0.11-0.17x105. An in depth comparison is then provided between the CFD results and measurements obtained via shadowgraphy and CT scans. [1] P.A. Vazques, J. Eaton, R. Fahrig, et al, ILASS, 2014. 9:31AM G31.00008 Supercritical fluid mixing in Diesel Engine Applications1 , LUIS BRAVO, U.S. Army Research Laboratory, PETER MA, Stanford University, MATTHEW KURMAN, MICHAEL TESS, U.S. Army Research Laboratory, MATTHIAS IHME, Stanford University, CHOL-BUM KWEON, U.S. Army Research Laboratory — A numerical framework for simulating supercritical fluids mixing with large density ratios is presented in the context of diesel sprays. Accurate modeling of real fluid effects on the fuel air mixture formation process is critical in characterizing engine combustion. Recent work (Dahms, 2013) has suggested that liquid fuel enters the chamber in a transcritical state and rapidly evolves to supercritical regime where the interface transitions from a distinct liquid/gas interface into a continuous turbulent mixing layer. In this work, the Peng Robinson EoS is invoked as the real fluid model due to an acceptable compromise between accuracy and computational tractability. Measurements at supercritical conditions are reported from the Constant Pressure Flow (CPF) chamber facility at the Army Research Laboratory. Mie and Schlieren optical spray diagnostics are utilized to provide time resolved liquid and vapor penetration length measurement. The quantitative comparison presented is discussed. 1 Oak Ridge Associated Universities (ORAU) 9:44AM G31.00009 An Evaluation of a Phase-Lag Boundary Condition for Francis Hydroturbine Simulations Using a Pressure-Based Solver1 , ALEX WOUDEN, JOHN CIMBALA, Pennsylvania State University, BRYAN LEWIS, Halliburton — While the periodic boundary condition is useful for handling rotational symmetry in many axisymmetric geometries, its application fails for analysis of rotor-stator interaction (RSI) in multi-stage turbomachinery flow. The inadequacy arises from the underlying geometry where the blade counts per row differ, since the blade counts are crafted to deter the destructive harmonic forces of synchronous blade passing. Therefore, to achieve the computational advantage of modeling a single blade passage per row while preserving the integrity of the RSI, a phase-lag boundary condition is adapted to OpenFOAM R software’s incompressible pressure-based solver. The phase-lag construct is accomplished through restating the implicit periodic boundary condition as a constant boundary condition that is updated at each time step with phase-shifted data from the coupled cells adjacent to the boundary. Its effectiveness is demonstrated using a typical Francis hydroturbine modeled as single- and double-passages with phase-lag boundary conditions. The evaluation of the phase-lag condition is based on the correspondence of the overall computational performance and the calculated flow parameters of the phase-lag simulations with those of a baseline full-wheel simulation. 1 Funded in part by DOE award number: DE-EE0002667 9:57AM G31.00010 About the prediction of Organic Rankine Cycles performances integrating local high-fidelity turbines simulation and uncertainties , PIETRO CONGEDO, Inria Bordeaux, DANTE DE SANTIS, Stanford Univ, GIANLUCA GERACI, Stanford Univ and Inria Bordeaux — Organic Rankine Cycles (ORCs) are of key-importance when exploiting energy systems with a high efficiency. The variability of renewable heat sources makes more complex the global performance prediction of a cycle. The thermodynamic properties of the complex fluids used in the process are another source of uncertainty. The need for a predictive and robust simulation tool of ORCs remains strong. A high-order accurate Residual Distribution scheme has been recently developed for efficiently computing a turbine stage on unstructured grids, including advanced equations of state in order to take into account the complex fluids used in ORCs. Advantages in using high-order methods have been highlighted, in terms of number of degrees of freedom and computational time used, for computing the numerical solution with a greater accuracy compared to lower-order methods, even for shocked flows. The objective of this work is to quantify the numerical error with respect to the various sources of uncertainty of the ORC turbine, thus providing a very high-fidelity prediction in the coupled physical/stochastic space. Monday, November 24, 2014 8:00AM - 10:10AM Session G32 Particle-Laden Flows: Particle-Turbulence Interaction — 2020 - Rodney O. Fox, Iowa State University 8:00AM G32.00001 Laminar-turbulent transition and inertial shear-thickening of particle suspensions , LUCA BRANDT, IMAN LASHGARI, FRANCESCO PICANO, Linné FLOW Centre and SeRC, KTH Mechanics, Stockholm, Sweden, WIM-PAUL BREUGEM, Laboratory for Aero & Hydrodynamics, TU-Delft, Delft, The Netherlands — When a suspension of rigid particles is considered instead of a pure fluid, the particle-fluid interactions significantly alter the bulk behavior of the flow unexplained effects appear in the transitional regime. These are important in several environmental and industrial applications. The aim of this study is to characterize the flow regimes of suspensions of finite-size rigid particles in a viscous fluid at finite inertia. We explore the system behavior as function of the particle volume fraction and the Reynolds number. Unlike single phase flows where a clear distinction exists between the laminar and the turbulent states, three different regimes can be identified in the presence of a particulate phase, with smooth transitions between them. At low volume fractions, the flow becomes turbulent when increasing the Reynolds number, transitioning from the laminar regime dominated by viscous forces to the turbulent regime characterized by enhanced momentum transport by turbulent eddies. At larger volume fractions, we identify a new regime characterized by an even larger increase of the wall friction. The wall friction increases with the Reynolds number (inertial effects) while the turbulent transport is unaffected, as in a state of intense inertial shear-thickening. 8:13AM G32.00002 Multiphase turbulence in vertical wall-bounded collisional gas-particle flows , RODNEY O. FOX1 , Department of Chemical and Biological Engineering, Iowa State University, JESSE CAPECELATRO, OLIVIER DESJARDINS, Sibley School of Mechanical and Aerospace Engineering, Cornell University — Wall-bounded particle-laden flows are common in many environmental and industrial applications, and are often turbulent. In vertical flows, strong coupling between the phases leads to the spontaneous generation of dense clusters that fall due to gravity at the walls, while dilute suspensions of particles rise in the central region. Sustained volume fraction and velocity fluctuations caused by the clusters result in the production of fluid-phase turbulent kinetic energy, referred to as cluster-induced turbulence (CIT). To better understand the nature of CIT in wall-bounded flows, Eulerian-Lagrangian simulations of statistically stationary three-dimensional gas-solid flows in vertical pipes are performed. To extract useful information consistent with Eulerian turbulence models, a separation of length scales is introduced to decompose correlated and uncorrelated granular motion. To accomplish this, an adaptive spatial filter is employed on the particle data with an averaging volume that varies with the local particle-phase volume fraction. Radial profiles of turbulence statistics are generated from the Eulerian-Lagrangian results. Details on the nature of the turbulence are described, as well as the challenges they present to turbulence modeling. 1 Marie-Curie Senior Fellow, Ecole Centrale Paris 8:26AM G32.00003 Ineractions of Turbulence and Sediment Particles in an Open Channel Flow1 , PEDRAM PAKSERESHT, SOURABH APTE, Oregon State University, JUSTIN FINN, None — Interactions of glass particles in water in a turbulent open channel flow over a smooth bed is examined using direct numerical simulation (DNS) together with Lagrangian Discrete-Element-Model (DEM) for particles. Unlike several studies on wall-bounded turbulent flows with particles, in this work, the gravity is perpendicular to the mean flow, resulting in interesting dynamics between the destabilizing lift forces on the particles and stabilizing effects of graivty. The turbulent Reynolds number (Reτ ) is 710 corresponding to the experimental observations of Righetti & Romano (JFM, 2004). Particles of size 100 microns with volume loading of 10−3 result in a single layer of non-touching particles at the bottom wall. The entrainment and deposition mechanisms of particles and their interactions with the near wall turbulence structure are studied in detail. For the particle concentration studied, the particles affect the flow field in both the outer as well as inner region of the wall layer where particle inertia and concentration are higher. The effect of these interactions on the wall events is being explored. 1 Supported by NSF project number 1133363 8:39AM G32.00004 Fluid-particle characteristics in fully-developed cluster-induced turbulence1 , JESSE CAPECELATRO, OLIVIER DESJARDINS, Cornell University, RODNEY FOX, Iowa State University — In this study, we present a theoretical framework for collisional fluid-particle turbulence. To identify the key mechanisms responsible for energy exchange between the two phases, an Eulerian-Lagrangian strategy is used to simulate fully-developed cluster-inudced turbulence (CIT) under a range of Reynolds numbers, where fluctuations in particle concentration generate and sustain the carrier-phase turbulence. Using a novel filtering approach, a length-scale separation between the correlated particle velocity and uncorrelated granular temperature (GT) is achieved. This separation allows us to extract the instantaneous Eulerian volume fraction, velocity and GT fields from the Lagrangian data. Direct comparisons can thus be made with the relevant terms that appear in the multiphase turbulence model. It is shown that the granular pressure is highly anisotropic, and thus additional transport equations (as opposed to a single equation for GT) are necessary in formulating a predictive multiphase turbulence model. In addition to reporting the relevant contributions to the Reynolds stresses of each phase, two-point statistics, integral length/timescales, averages conditioned on the local volume fraction, and PDFs of the key multiphase statistics are presented and discussed. 1 The research reported in this paper is partially supported by the HPC equipment purchased through U.S. National Science Foundation MRI grant number CNS 1229081 and CRI grant number 1205413. 8:52AM G32.00005 ABSTRACT WITHDRAWN — 9:05AM G32.00006 The effect of the dissipation of energy on the hydrodynamics of the gas-particle fluidized beds , D.J. BERGSTROM, MOHAMMAD REZA HAGHGOO, Department of Mechanical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, R.J. SPITERI, Department of Computer Science, University of Saskatchewan, Saskatoon, SK, Canada — The flow structure in dense gas-particle fluidized beds is strongly affected by the dissipation of kinetic energy through particle collisions with each other and the wall. The energy dissipation reduces the kinetic energy of the particles. Consequently, larger clusters will be formed, and this in turn leads to the formation of larger bubbles. Therefore, it is insightful to investigate the instantaneous dissipation of energy in a fluidized bed in order to have a better understanding of the hydrodynamics of the particle phase. Visualization of the dissipation term will also clarify how much the walls contribute to the dissipation of energy in the overall system. In this study, a two-fluid model is used for the numerical simulation of an engineering-scale bubbling fluidized bed. The MFiX code is used to perform the simulations. A modified SIMPLE algorithm for multiphase flows is employed that uses a higher-order discretization scheme to accurately compute bubble shapes and the deferred correction method to enhance numerical stability. The results of the three-dimensional simulation are in good agreement with the limited experimental data. The dissipation of the kinetic energy of the particles is evaluated using the model relations based on the simulated particle velocity fields. 9:18AM G32.00007 Lagrangian statistics of inertial particles in near-wall turbulence , JUNGHOON LEE, CHANGHOON LEE, Yonsei University — Despite many studies regarding particle-laden turbulence in near-wall turbulence, detailed investigation on the Lagrangian nature of particles is very rare. In our study, inertial particle trajectories suspended in turbulent channel flow were calculated via direct numerical simulation with Lagrangian particle tracking. Since particles smaller than the Kolmogorov length scale and their dilute suspension are addressed in this study, one-way coupled simulations with point-particle approach are performed. By ensemble-averaging over a number of particles, we investigate Lagrangian statistics, such as particle dispersion, velocity and acceleration autocorrelation and probability density function of expected particle position, of particles released at several different distances from the wall for a wide range of Stokes numbers. In addition, the effects of gravity and lift on the Lagrangian statistics are investigated. Plausible physical explanations are provided. 9:31AM G32.00008 Rotational response of suspended particles to turbulent flow: laboratory and numerical synthesis1 , EVAN VARIANO, University of California, Berkeley, LIHAO ZHAO, Norwegian Technical National University, MARGARET BYRON, University of California, Berkeley, GABRIELE BELLANI, University of Bologna, YIHENG TAO, University of California, Berkeley, HELGE ANDERSSON, Norwegian Technical National University — Using laboratory and DNS measurements, we consider how aspherical and inertial particles suspended in a turbulent flow act to “filter” the fluid-phase vorticity. We use three approaches to predict the magnitude and structure of this filter. The first approach is based on Buckingham’s Pi theorem, which shows a clear result for the relationship between filter strength and particle aspect ratio. Results are less clear for the dependence of filter strength on Stokes number; we briefly discuss some issues in the proper definition of Stokes number for use in this context. The second approach to predicting filter strength is based on a consideration of vorticity and enstrophy spectra in the fluid phase. This method has a useful feature: it can be used to predict the filter a priori, without need for measurements as input. We compare the results of this approach to measurements as a method of validation. The third and final approach to predicting filter strength is from the consideration of torques experienced by particles, and how the “angular slip” or “spin slip” evolves in an unsteady flow. We show results from our DNS that indicate different flow conditions in which the spin slip is more or less important in setting the particle rotation dynamics. 1 Collaboration made possible by the Peder Sather Center 9:44AM G32.00009 Three Dimensional Tracking of Two-Particle Dispersion in a Turbulent Jet , STEPHANIE PAUSTIAN, CHIN HEI NG, ALBERTO ALISEDA, ME Department, University of Washington — We present experimental measurements of two-particle dispersion in a turbulent shear flow. The baseline flow is a well-characterized high Reynolds number (up to approximately 200,000) turbulent submerged round jet. Injection is performed in the self-similar part of the jet, where careful PIV measurements of the mean flow field and turbulent second order moments have been obtained and validated against theory and other experiments. Three-dimensional tracking of two particles, injected simultaneously at two different radial or axial positions in the jet, is obtained from two-camera high-speed shadowgraphy. The influence of mean shear and turbulence fluctuations on dispersion is analyzed for initial positions where a mean velocity shear is superimposed on turbulent fluctuations. Injection of different particles, liquid droplets and gas bubbles is performed to analyze the influence of density ratio, particle inertia and buoyancy on the dispersion ratio. The versatility of the experimental facility also allows for the experimental investigation of the Reynolds number effect on dispersion, ranging from below the mixing transition (approximately 3,000) to very high values (>100,000). 9:57AM G32.00010 Application of Medical Magnetic Resonance Imaging for Particle Concentration Measurement , DANIEL BORUP, CHRISTOPHER ELKINS, JOHN EATON, Stanford Univ — Particle transport and deposition in internal flows is important in a range of applications such as dust aggregation in turbine engines and aerosolized medicine deposition in human airways. Unlike optical techniques, Magnetic Resonance Imaging (MRI) is well suited for complex applications in which optical access is not possible. Here we present efforts to measure 3D particle concentration distribution using MRI. Glass particles dispersed in water flow reduce MRI signal from a spin-echo or gradient-echo scanning sequence by decreasing spin density and dephasing the spins present in the fluid. A preliminary experiment was conducted with a particle streak injected at the centerline of a turbulent round pipe flow with a U bend. Measurements confirmed that signal strength was related to particle concentration and showed the effects of gravitational settling and turbulent dispersion. Next, measurements of samples in a mixing chamber were taken. Particle volume fraction was varied and sensitivity to particle/fluid velocity was investigated. These results give a relationship between MRI signal, particle volume fraction, MRI sequence echo time, and spin relaxation parameters that can be used to measure local particle volume fraction in other turbulent flows of interest. Monday, November 24, 2014 8:00AM - 10:10AM Session G33 Vortex Dynamics: General — 2022 - Geoffrey Spedding, University of Southern California 8:00AM G33.00001 A numerical study of the laminar necklace vortex system and its effect on the wake for a circular cylinder , GOKHAN KIRKIL, Kadir Has University, GEORGE CONSTANTINESCU, University of Iowa — Large Eddy Simulation is used to investigate the structure of the laminar horseshoe vortex (HV) system and the dynamics of the necklace vortices as they fold around the base of a circular cylinder mounted on the flat bed of an open channel for Reynolds numbers defined with the cylinder diameter, D, smaller than 4,460. The study concentrates on the analysis of the structure of the HV system in the periodic breakaway sub-regime which is characterized by the formation of three main necklace vortices. For the relatively shallow flow conditions considered in this study (H/D 1, H is the channel depth), at times, the disturbances induced by the legs of the necklace vortices do not allow the SSLs on the two sides of the cylinder to interact in a way that allows the vorticity redistribution mechanism to lead to the formation of a new wake roller. As a result, the shedding of large scale rollers in the turbulent wake is suppressed for relatively large periods of time. Simulation results show that the wake structure changes randomly between time intervals when large-scale rollers are forming and are convected in the wake (von Karman regime), and time intervals when the rollers do not form. 8:13AM G33.00002 A Generic Mechanism for Splitting and Shedding of Vortices and Recirculation Regions , KEVIN CASSEL, MICHAEL BOGHOSIAN, Illinois Institute of Technology — Vortex shedding is a common feature in many high-Reynolds number internal and external flows. While several mechanisms have been put forth to explain this important phenomenon in specific settings, a general framework that unifies these mechanisms and applies in a broad class of flows has not been forthcoming. A surprisingly simple minimal flow unit is identified in the present study and shown to apply in a wide variety of settings in which vortices or recirculation regions are found to split and shed in the vicinity of smooth surfaces. There are two necessary conditions for this mechanism: 1) a region of low momentum fluid (as found at the center of a slowly moving vortex or recirculation region), and 2) a pressure or body force having a particular structure acting on the region of low momentum. While the impetuous for the pressure or body force may vary, its action on the vortex or recirculation region is generic. The basic framework is illustrated and causality established through calculation of several simple model problems, and computational results for some canonical flows, such as shedding behind a circular cylinder and flow over a forward-facing step, are used to illustrate how the generic mechanism can be identified in real flows. 8:26AM G33.00003 Wake states and forces associated with a cylinder rolling down an incline under gravity1 , FARAH YASMINA HOUDROGE, MARK THOMPSON, KERRY HOURIGAN, FLAIR, Monash University, THOMAS LEWEKE, IRPHE, CNRS/Universités Aix-Marseille — The flow around a cylinder rolling along a wall at a constant velocity was recently investigated by Stewart et al. (JFM, 643, 648, 2010). They showed that the wake structure varies greatly as the Reynolds number was increased, and that the presence of the wall as well as the imposed motion of the body have a strong influence on the dominant wake structure and the wake transitions when the body is placed in free stream. In this work, attention is given to the flow dynamics and the fluid forces associated with a cylinder rolling down an incline under the influence of gravity. Increasing the inclination angle or the Reynolds number is shown to destabilize the wake flow. For a body close to neutrally buoyancy, the formation and shedding of vortices in its wake result in fluctuating forces and a final kinematic state in which the body’s velocity is not constant. The non-dimensionalization of the main equations allows us to determine the essential parameters that govern the problem’s dynamics. Furthermore, through numerical simulations we analyse in more detail the time-dependant fluid forces and the different structures of the wake in order to gain a better understanding of the physical mechanisms behind the motions of the fluid and the body. 1 This research was supported by an Australian Research Council Discovery Project Grant DP130100822. We also acknowledge computing time support through National Computing Infrastructure projects D71 and N67. 8:39AM G33.00004 Shear-thinning effects on vortex breakdown in swirling pipe flows: experiments and simulations , DAVID DENNIS, University of Liverpool, TOM PETIT, University of Liverpool and Ecole Centrale de Nantes, DEACON THOMPSON, ROBERT POOLE, University of Liverpool — Laminar pipe flow with a controllable wall swirl has been studied both numerically and experimentally to explore the behaviour of inelastic shear-thinning fluids. The pipe consists of two smoothly joined sections that can be rotated independently about the same axis. The circumstances of flow entering a stationary pipe from a rotating pipe (decaying swirl) and flow entering a rotating pipe from a stationary pipe (growing swirl) have been investigated. A numerical parametric study using a simple power law model is conducted and reveals the axial length of the recirculation region is increased for shear-thinning fluids and decreased for shear-thickening (in comparison to the Newtonian reference). The critical swirl ratio required to induce the breakdown at a range of Reynolds numbers and extent of shear-thinning is investigated and a method of scaling is presented that collapses all the data for all fluids (shear-thickening, Newtonian and shear-thinning) onto a single universal curve. Experimental visualisations using an aqueous solution of Xantham Gum (shear-thinning) confirm the conclusions drawn from the numerical results. 8:52AM G33.00005 Wall-separation and vortex-breakdown zones in a solid-body rotation flow in a rotating pipe , ZVI RUSAK, Renssealer Polytechnic Institute, SHIXIAO WANG, Auckland University, New Zealand — The axisymmetric dynamics of perturbations on a solid-body rotation flow with a uniform axial velocity in a rotating, finite-length circular pipe is studied via global analysis methods and numerical simulations. We first describe the bifurcation diagram of steady-state solutions of the flow problem as a function of the swirl ratio ω. We prove that the base columnar flow is a unique steady-state solution when ω is below a critical level, ω1 . This state is asymptotically stable and a global attractor of the flow dynamics. However, when ω > ω1 , we reveal, in addition to the base columnar flow, the co-existence of states that describe swirling flows around either centerline stagnant breakdown zones or wall pseudo-stagnant zones. The base columnar flow is a min-max point of the energy functional that governs the problem while the swirling flows with wall-separation and breakdown zones are global and local minimizer states and attractors of the flow dynamics. We also find additional min-max states that are transient attractors of the flow dynamics. The wall-separation states have same chance to appear as that of the breakdown states and there is no hysteresis loop between these states. 9:05AM G33.00006 The effect of aspect ratio on vortex rings within the wake of impulsivelystarted flat plates1 , JOHN FERNANDO, DAVID RIVAL, Queen’s University — Vortex pinch-off has been the focus of many studies since it was first observed for vortices produced via piston-cylinder arrangements. Minimal work has been performed on other vortex generation methods. The current study investigates vortex rings behind impulsively-started circular, square, and elliptical flat plates. Preliminary force and PIV measurements show temporal/spatial similarities between vortex growth in the wake of the circular and square plates. Forces and vortex evolution are also shown to be strongly coupled; the presence of stable wake vortex rings results in a reduction of plate drag. For all three plates, pinch-off is initiated by the formation of a positive pressure gradient on the leeward side of the plate, which terminates mass transport to the vortex. It is hypothesized that an increase in aspect ratio (AR) from unity results in isolated vortex lines with non-uniform vorticity along the leading edges. Strong spanwise velocity gradients and stretching near the plate tips facilities vortex detachment. Results from experiments on rectangular plates with varying ARs are discussed and the effect of stretching and tilting in the tip region is investigated. 1 The United States Air Force Office of Scientific Research 9:18AM G33.00007 Vortex Formation Behind an Inclined 2-Dimensional Thin Flat Plate1 , MERAJ MOHEBI, DAVID H. WOOD, ROBERT J. MARTINUZZI, Univ of Calgary — Stereo Particle Image Velocimetry was used to measure the turbulent wake of a 2D flat plate inclined relative to a uniform stream as a heuristic model for airfoils and wind turbine blades at high incidence. Phase Averaging was performed to study the vortex dynamics and relate these to the force characteristics. Below 90◦ , immediately behind the plate, rounder and more organized trailing edge vortices form which possess higher circulation and are associated with higher Reynolds stresses than the counter-rotating, weaker and elongated leading edge vortices. The quasi-periodically shed vortices on the sides of the wake decay in strength at different rates to reach a circulation ratio of -1 within a distance less than 5 chords downstream of the plate for all angles. This equalization of vortex strength is related to an increase in turbulence diffusion, due to mostly-incoherent 3-dimensionality which progressively increases as the inclination angle is reduced, and convective transfer of vorticity between counter-rotating vortices. The wake experiences a sudden change in vortex formation mechanism at around 40◦ . At this angle, the frequency analysis on the signals of a pair of micro-pressure transducers in the wake also shows a discontinuity in the trends. 1 This work was supported by NSERC Discovery grants to R. J. Martinuzzi and D. H. Wood. 9:31AM G33.00008 Experimental study of periodic flow effects on spanwise vortex1 , CRUZ DANIEL GARCIA MOLINA, ERICK JAVIER LOPEZ SANCHEZ, GERARDO RUIZ CHAVARRIA, Facultad de Ciencias Universidad Nacional Autonoma de Mexico, ABRAHAM MEDINA OVANDO, Escuela Superior de Ingenieria Mecanica Electrica, Azcapotzalco, Instituto Politecnico Nacional, Mexico — We present an experimental study about the spanwise vortex produced in a flow going out of a channel in shallow waters. This vortex travels in front of the dipole. The velocity field measurement was done using the PIV technique, and DPIVsoft (https://www.irphe.fr/∼meunier/) was used for data processing. In this case the flow has a periodic forcing to simulate ocean tides. The experiment was conducted in a channel with variable width and the measurements were made using three different values of the aspect ratio width-depth. We present results of the position, circulation of this spanwise vortex and the flow inversion effect. The change of flow direction modify the intensity of the vortex, but it does not destroy it. The vertical components of the velocity field contributes particle transport. 1 G. Ruiz Chavarria, E. J. Lopez Sanchez and C. D. Garcia Molina acknowledge DGAPA-UNAM by support under project IN 116312 (Vorticidad y ondas no lineales en fluidos) 9:44AM G33.00009 Flow Interference between a Square (Upstream) and a Circular Cylinder: Flow Pattern Identification , NITHIN S. KUMAR, AJITH KUMAR R, JAYALAKSHMI MOHAN, Department of Mechanical Engg, AMRITA University, Amritapuri Campus, Kerala, India — In this paper, flow interference between an upstream square cylinder and a circular cylinder of equal size is studied in tandem arrangement. The main objective of this invesigation is to identify the possible flow patterns at different spacing ratios, L/B where L is the centre-to-centre distance between the cylinders and B is the characteristic dimension of the bodies. All the experiments are conducted in a water channel and the test Reynolds number is 2100 (based on B). L/B is varied from 1.0 to 5.0. The flow visualization experiments are videographed and then analyzed frame-by-frame to capture the finer details of the flow patterns. Flow over single square and circular cylinders is analyzed first. Then, flow interference between two square cylinders is investigated. Subsequently, flow over a square-circular configuration is investigated. No such systematic studies are reported so far. Different flow patterns are observed for the square-circular configuration. Additionally, the time of persistence of each flow pattern have been recorded over a sufficiently long period of time to see the most dominant flow pattern. The schedule of occurrence of flow patterns have also been studied during this investigation. This study bears considerable practical relevance in the context of possible interference effects occurring in engineering structures such as buildings, bridges etc. 9:57AM G33.00010 Flow Interference between a Circular (Upstream) and a Square Cylinder: Flow Pattern Identification , JAYALAKSHMI MOHAN, AJITH KUMAR R, NITHIN S KUMAR, Department of Mechanical Engineering, AMRITA University, Amritapuri Campus, Kerala, India — In this paper, flow interference between an upstream circular cylinder and a square cylinder of equal size is studied in tandem arrangement. The main objective of this invesigation is to identify the possible flow patterns at different spacing ratios, L/B where L is the centre-to-centre distance between the cylinders and B is the characteristic dimension of the bodies. All the experiments are conducted in a water channel and the test Reynolds number is 2100 (based on B). L/B is varied from 1.0 to 5.0. The flow visualization experiments are videographed and then analyzed frame-by-frame to capture the finer details of the flow patterns. Flow over single square and circular cylinders is analyzed first. Then, flow interference between two circular cylinders is investigated. Subsequently, flow over a circular-square configuration is investigated. No such studies are reported so far. Different flow patterns are observed for the circular-square configuration. Additionally, the time of persistence of each flow pattern have been recorded over a sufficiently long period of time to see the most dominant flow pattern. The schedule of occurrence of flow patterns have also been studied during this investigation. This study is very much relevant in the context of possible interference effects occuring in engineering structures such as buildings, heat exchanger tubes etc. Monday, November 24, 2014 8:00AM - 10:10AM Session G34 Stratified and Premixed Flames — 2024 - Peter Hamlington, University of Colorado 8:00AM G34.00001 Analysis of lift-off height and structure of n-heptane tribrachial flames in laminar jet configuration , STEFANO LUCA, FABRIZIO BISETTI, Clean Combustion Research Center, KAUST — A set of lifted tribrachial n-heptane flames in a laminar jet configuration is simulated. The simulations are performed using finite rate chemistry and detailed transport, and aim at investigating the geometry and the structure of tribrachial flames. Varying the inlet velocity of the fuel, different stabilization heights are obtained, and the dependence on the inlet velocity is compared with experimental data. The results of the simulations show that when the stabilization height decreases, resulting in larger velocity and mixture fraction gradients at the tribrachial point, the tilt of the flame increases, while the heat release rate and radius of curvature decrease. A detailed analysis of the flame geometry, compared to unstretched premixed flames is performed, focusing on differential diffusion effects, flame stretch, and transport of heat and mass from the burnt gases to the flame front. Our analysis seems to indicate that for a flame that stabilizes further downstream positive stretch along the rich premixed wing leads to an increase in the rate of chemical reaction in the whole flame. 8:13AM G34.00002 Structure of a laminar triple flame of a jet fuel surrogate , KRITHIKA NARAYANASWAMY, PERRINE PEPIOT, Cornell University — Triple flames are found in jet flames and play an important role in the stabilization and thereby lift-off height of lifted jet flames. In this study, 2D laminar triple flames burning jet fuel are simulated using finite rate chemistry and detailed transport of species. The jet fuel is represented by using a surrogate mixture, comprised of n-dodecane, methyl-cyclohexane, and m-xylene. The chemical kinetics of this multi-component surrogate are described using a reduced model derived from a well-validated detailed mechanism. The structure of the simulated triple flames is explored by examining the reactivity of the different hydrocarbons in the multi-component fuel and the radical profiles. The heat release profiles of the lean and rich branches of the triple flame are compared to their unstretched 1D counterparts to identify similarities. Varying the composition of the components in the multi-component surrogate is found to result in little differences in the laminar flame speed predictions. The simulations are repeated for different fuel mixtures in order to investigate the effect of the surrogate fuel composition on the combustion process. 8:26AM G34.00003 Direct numerical simulations of flow-chemistry interactions in statistically stationary turbulent premixed flames , HONG G. IM, PAUL G. ARIAS, King Abdullah University of Science and Technology, SWE- TAPROVO CHAUDHURI, India Institute of Science, Bangalore, CHUNG K. LAW, Princeton University, KAUST COLLABORATION, INDIA INSTITUTE OF SCIENCE, BANGALORE COLLABORATION — The effects of Damkohler number and Karlovitz number on the flame dynamics of three-dimensional statistically planar turbulent premixed flames are investigated by direct numerical simulation incorporating detailed chemistry and transport for a hydrogen-air mixture. The mean inlet velocity was dynamically adjusted to ensure a stable flame within the computational domain, allowing the investigation of time-averaged quantities of interest. A particular interest was on understanding the effects of turbulence on the displacement speed of the flame relative to the local fluid flow. The results show that the displacement speed dynamics in response to turbulent eddies depends strongly on the specific choice of the iso-surfaces in the progress variable. As such, the statistical distribution of the flame speed versus the strain/curvature relations shows a significant sensitivity on the definition of the flame speed. Further analysis is conducted to examine the behavior of the alignment between the flame surface and the strain rate eigenvectors. The results for the reference conditions are compared against different parametric conditions in order to assess their effects on flame-flow interaction characteristics. 8:39AM G34.00004 The turbulent flame speed for low-to-moderate turbulence intensities , MOSHE MATALON, NAVIN FOGLA, University of Illinois at Urbana Champaign, FRANCESCO CRETA, University of Rome La Sapienza — Premixed flame propagation in two-dimensional turbulent flows is examined within the context of a hydrodynamic model. The flame is treated as a surface of density discontinuity and propagates against a turbulent flow of prescribed intensity and scale. A hybrid Navier-Stokes/interface capturing technique is used to describe the flow field throughout the entire domain and track the highly-fluctuating flame front which is allowed to attain folded conformations and form pockets of unburned gases that detach from the main flame surface and are rapidly consumed. A parametric study is conducted to examine the effects of the turbulence parameters: intensity and scale, and the combustion parameters: thermal expansion and mixture composition (or Markstein length). Markstein lengths are varied in order to span both, the Darrieus-Landau (DL) instability-free subcritical and the DL instability-affected supercritical regimes. Scaling laws for the turbulent flame speed, exhibiting explicit dependence on the system parameters, are proposed for low-to-moderate turbulence intensities. 8:52AM G34.00005 Impact of Karlovitz number on vortex evolution through a premixed flame , CHRIS BRADLEY, BROCK BOBBITT, GUILLAUME BLANQUART, California Institute of Technology — As a canonical test case of premixed turbulent combustion, the vortex-flame interaction is investigated for the transformation of vorticity through the flame. This is analyzed as a function of the length and velocity scale of the vortex, which may be related to the Karlovitz number in premixed turbulent combustion. This analysis is performed using theoretical analysis of the vorticity equation and results from Direct Numerical Simulations. The vorticity is found to transform with different behavior due to the variable importance of viscous dissipation, dilatation, and baroclinic torque. The importance of these affects are shown to be based on the velocity and length scale of the vortex in relation to the velocity and length scale of the flame. The conditions under which these effects are dominant is outlined and confirmed through comparison of the theoretical and simulation results. 9:05AM G34.00006 Bivariate chemistry model for the simulation of high Karlovitz premixed turbulent flames in the absence of differential diffusion , BRUNO SAVARD, GUILLAUME BLANQUART, California Institute of Technology — In recent work (Savard et al., Proc. Comb. Inst. (2014)), it was shown that the structure, i.e. the dependence of the species mass fractions on temperature, of a high Karlovitz n-heptane/air turbulent premixed flame was similar to that of an unstretched one-dimensional flame in the absence of differential diffusion. The mean profiles of the species chemical source terms conditional to temperature were also found to be very close to those of a corresponding unstretched one-dimensional flame. However, while minimal deviation from the one-dimensional solution was found in the flame structure, large relative fluctuations around the mean (close to the one-dimensional solution) were found in the species source terms. Accordingly, conventional tabulated chemistry with only one progress variable transported is shown to adequately represent the flame structure and the mean species source terms, but not the source terms fluctuations. In this work, a bivariate chemistry model (i.e. a model based on two variables) that captures these species source terms fluctuations is presented. The model is first developed for unity Lewis number flames, but an extension to flames with differential diffusion is discussed. 9:18AM G34.00007 Differing effects of viscosity and density changes on turbulence in high Karlovitz premixed flames , BROCK BOBBITT, GUILLAUME BLANQUART, California Institute of Technology — The change in turbulence characteristics through a premixed flame is an important phenomenon in premixed turbulent combustion as the characteristics behind the preheat zone are the relevant quantities for the effects on the reaction zone. Both the fluid properties of viscosity and density are altered through the preheat zone which induce changes to the turbulence. As the viscosity can increase by a factor of 30 and density can decrease by a factor of 6 in many hydrocarbon/air flames, these effects are significant. This work focuses on the individual effects of each of these fluid properties to better understand the coupling of the turbulence and flame. This is analyzed through high Karlovitz number Direct Numerical Simulations of a n-heptane/air flame and varying the relative importance of density and viscosity. The viscosity change relates to viscous dissipation while the density change relates to dilation and baroclinic torque in the vorticity equation. 9:31AM G34.00008 3D DNS of Turbulent Premixed Flame with over 50 Species and 300 Elementary Reactions , MASAYASU SHIMURA, BASMIL YENERDAG, YOSHITSUGU NAKA, Tokyo Institute of Technology, YUZURU NADA, The University of Tokushima, MAMORU TANAHASHI, Tokyo Institute of Technology — Three-dimensional direct numerical simulation of methane-air premixed planar flame propagating in homogenous isotropic turbulence is conducted to investigate local flame structure in thin reaction zones. Detailed kinetic mechanism, GRI-Mech 3.0 which includes 53 species and 325 elementary reactions, is used to represent methane-air reaction, and temperature dependences of transport and thermal properties are considered. For a better understanding of the local flame structure in thin reaction zones regime, distributions of mass fractions of major species, heat release rate, temperature and turbulent structures are investigated. Characteristic flame structures, such as radical fingering and multi-layered-like flame structures, are observed. The most expected maximum heat release rate in flame elements is lower than that of laminar flame with same mixture. To clarify mechanism of the decrease in local heat release rate, effects of strain rates tangential to flame front on local heat release rate are investigated. 9:44AM G34.00009 Spectral Kinetic Energy Transfer Through a Premixed Flame Brush , COLIN A.Z. TOWERY, Univ of Colorado - Boulder, ALEXEI Y. POLUDNENKO, Naval Research Laboratory, PETER E. HAMLINGTON, Univ of Colorado - Boulder — Turbulence-flame interactions are of fundamental importance for understanding and modeling premixed turbulent reacting flows. These interactions can result in nonlinear feedback leading to large changes in both the turbulence and flame. Recent computational studies have indicated, however, that not all scales of turbulent motion are affected equally. Small-scale motions appear to be suppressed while larger-scale motions are unaffected or even enhanced. In order to determine the scale-dependence of turbulence-flame interactions, direct numerical simulations of statistically planar, premixed flames have been performed and analyzed. Two-dimensional kinetic energy spectra, conditioned on the planar-averaged fuel mass-fraction, are measured through the flame brush and compared to both compressible and incompressible non-reacting flow spectra. Changes in the spectra with respect to fuel mass-fraction are then connected to the dynamics of the kinetic energy spectrum transport equation. Particular focus is placed on understanding triadic velocity, pressure, and dilatation interactions, including the characterization of backscatter due to heat release and compressibility. Finally, the implications of these results for modeling practical premixed combustion problems are outlined. 9:57AM G34.00010 High spatial resolution PIV and CH-PLIF measurements of a Shear Layer Stabilized Flame , CHRISTOPHER FOLEY, IANKO CHTEREV, JERRY SEITZMAN, TIM LIEUWEN, Georgia Institute of Technology — In practical combustors, flames stabilize in thin shear layers with very high strain rates, which alter the flame burning rate - either enhancing or diminishing reaction rates, and even leading to extinction. Therefore, the bulk velocity that provides stable operation in these combustors is limited, presumably due to the associated maximum stretch rate that the flame is able to withstand. The focus of this work is to develop a deeper understanding of the interaction between flow and flame for a shear layer stabilized, premixed flame. This study consists of planar, high resolution, simultaneous PIV and CH-PLIF measurements, in a 8 x 6 mm plane with 0.11 mm and 0.16 mm PIV vector and CH-PLIF image resolution, respectively, of the flame stabilization region in a swirling jet. The hydrodynamic strain induced stretch rate along the high CH concentration layer of the flame front is calculated from these measurements. In addition, this study elucidates the unsteady behavior of the flame in the thin shear layer. The measured flame stretch is highly spatially and temporally dependent, and dominated by contributions from normal and shear strain terms of axial velocity. Although normal strain is much greater than shear, the near horizontal flame orientation results in neither strain term dominating flame stretch. Furthermore, the flame angle changes the sign of the shear strain contributions as observed experimentally, an important implication for reduced order modeling approaches. Monday, November 24, 2014 8:00AM - 10:10AM Session G35 Compressible Flow III: Explosions and Shock Focusing — 2001A - Veronica Eliasson, University of Southern California 8:00AM G35.00001 On the Stability of Ionizing Shocks in Monatomic Gases1 , HAI LE, ANN KARAGOZIAN, UCLA, MARCO PANESI, University of Illinois at Urbana-Champaign, JEAN-LUC CAMBIER, Air Force Research Laboratory — Prior work by our group demonstrates the use of a collisional-radiative model in reproducing the correct steady-state shock structure of ionizing shocks in monatomic gases.2 In this presentation, we report on time dependent calculations of ionizing shock flows, which reveal additional physical phenomena arising from the unsteadiness and non-linear wave coupling between convection and kinetics. The observed phenomena are similar to instabilities often seen in gaseous detonations.3 The present model also takes into account radiative heat losses and radiation transport, which result in a reduction in the shock velocity and precursor effects. The latter phenomena may be important at high shock velocities, and are being investigated in detail. 1 Distribution A: Approved for public release, distribution unlimited. Supported by AFOSR grant 12RZ06COR (PM: Dr. F. Fahroo) H. P., et al., Bull. Am. Phys. Soc. 57, 17 (2012) 3 Cole, L. K., et al., Combust. Sci. Technol. 184, 1502-1525 (2012) 2 Le, 8:13AM G35.00002 Modeling Gas-Dynamic Effects in Shock-Tubes for Reaction-Kinetic Measurements , KEVIN GROGAN, QING WANG, MATTHIAS IHME, Stanford University — Accurate chemical kinetic models are pivotal for characterizing the effects of new fuel compositions on existing propulsion systems and for developing future combustion technologies. Shock-tube facilities remain invaluable for providing detailed information about ignition delay times, extinction limits, and species time histories for the development and validation of reaction mechanisms. However, viscous and heat transfer effects along the shock-tube wall introduce variations of the thermodynamic state behind the reflected shock wave, thereby affecting the reaction kinetics being measured. These effects have been countered experimentally by the use of driver inserts, extended shock-tube diameters, and the dilution of the test gas. To assist with the design of driver inserts and the selection of operating conditions, a low-order one-dimensional model is developed and compared to two-dimensional Unsteady-Favre-Averaged-Navier-Stokes (UFANS) models as well as experimental data. This model is shown to give accurate predictions of the gas-dynamics in shock-tubes at a computationally efficient cost. 8:26AM G35.00003 Shock wave control using liquid curtains , BRENDAN COLVERT, XINGTIAN TAO, VERONICA ELIASSON, Univ of Southern California — The effectiveness of a planar wall of liquid as a blast mitigation device is examined using a shock tube and a customdesigned and -built shock test chamber. Experimental data collection methods being used include high-speed schlieren photography and high-frequency pressure sensors. During the relevant shock interaction time periods, the liquid-gas interface is examined to determine its effect on shock waves. The characteristic quantities that reflect these effects include reflected-to-incident shock strength ratio, transmitted-to-incident shock strength ratio, transmitted and reflected impulse, and peak pressure reduction. These parameters are examined for correlations to incident wave speed, liquid mass, liquid density, and liquid viscosity. Initial results have been obtained that show a correlation between fluid mass and peak pressure reduction. More experiments are being performed to further explore this relationship as well as examine the effects of altering the other parameters such as liquid-gas interface geometry and using dilatant fluids. 8:39AM G35.00004 Explosive-driven shock wave interaction with a propane flame , NANCY CANAFAX, MICHAEL HARGATHER, New Mexico Tech, PAUL GIANNUZZI, New Mexico Tech EMRTC, GRAHAM DOIG, UNSW Australia — Experiments were performed to analyze the interaction of an explosively driven shock wave and a propane flame. A 30 gram explosive charge was detonated at one end of a shock tube with a 3 m length and 0.6 m diameter to produce a shock wave which propagated down the tube and out into the atmosphere. A propane flame source was positioned at various locations outside of the shock tube to investigate the flame response to different strength shock waves. Retroreflective shadowgraph imaging with a high-speed digital camera was used to visualize the shock wave motion and flame response. The explosively driven shock tube was shown to produce a repeatable shock wave and a large vortex ring. Digital streak images show the shock wave and vortex ring expansion and propagation throughout the field of view. The high-speed shadowgraph images show that the shock wave extinguishes the propane flame by pushing it off of the fuel source. Even a weak shock wave was found to be capable of extinguishing the propane flame. 8:52AM G35.00005 Analysis of shock wave propagation from explosives using computational simulations and artificial schlieren imaging , CHRISTOPHER ARMSTRONG, MICHAEL HARGATHER, New Mexico Tech — Computational simulations of explosions are performed using the hydrocode CTH and analyzed using artificial schlieren imaging. The simulations include one and three-dimensional free-air blasts and a confined geometry. Artificial schlieren images are produced from the density fields calculated via the simulations. The artificial schlieren images are used to simulate traditional and focusing schlieren images of explosions. The artificial schlieren images are compared to actual high-speed schlieren images of similar explosions. Computational streak images are produced to identify time-dependent features in the blast field. The streak images are used to study the interaction between secondary shock waves and the explosive product gas contact surface. 9:05AM G35.00006 An experimental study of shock wave reflection over non-Newtonian liquid wedges1 , HONGJOO JEON, CHRISTOPHER DOUGHERTY, RYAN MILLER, VERONICA ELIASSON, University of Southern California — An experimental investigation of the reflection of a planar shock wave over different density liquid wedges was performed by means of an angled shock tube. The goal is to find a transition criterion between regular reflection (RR) and irregular reflection (IR). The shock tube can be rotated to any angle between the horizontal and vertical planes for various impact media. The reflection of the oblique shock wave for different wedges was visualized using the shadowgraph and schlieren techniques. Previous research by Ben-Dor et al. (1987) conducted different types of reflecting solid conditions and Takayama et al. (1989) investigated a similar experiment with a nonsolid reflecting surface. Motivated by the previous work, we undertook a series of shock tube experiments where both Newtonian and non-Newtonian liquids were used to form a wedge for a shock wave to impact. Shear-thickening materials, such as a water-cornstarch mixture, or similar suspensions, could potentially be utilized to protect soldiers and other high-risk personnel from impacts. Results show that, for both a water-cornstarch and ballistic gelatin sample, the detachment angle at which the RR transitions to an IR was different from those of a solid and water. 1 This work is funded by NSF grant #CBET-1437412. 9:18AM G35.00007 A Comparison of Numerical and Experimental Results of Passive Shock Wave Attenuation in Two-Dimensional Ducts , QIAN WAN, MONICA NGUYEN, VERONICA ELIASSON, University of Southern California — The study of shock wave attenuation has drawn much attention in shock wave area. One of the common ways to attenuate shock waves is to arrange multiple obstacles to block the propagation path of the shock wave. We propose an arrangement of the obstacles by placing the square or cylinder shaped obstacles along the outline of a logarithmic spiral curve, taking advantage of its ability of collecting the incident shock wave to its focal region. We simulated the process of the shock wave passing through these arrangements using the Euler equations of gas dynamics. Then, to validate the numerical results, we present corresponding experiments under the same initial conditions. Results show that the numerical and experimental methods agree well, and that placing obstacles along a logarithmic spiral curve can effectively attenuate the transmitted shock waves. 9:31AM G35.00008 Geometrical shock dynamics, formation of singularities and topological bifurcations of converging shock fronts1 , NUGZAR SURAMLISHVILI, JENS EGGERS, School of Mathematics, University of Bristol, UK, MARCO FONTELOS, Instituto de Ciencias Matematicas, C/Serrano 123, 28006, Spain — We are concerned with singularities of the shock fronts of converging perturbed shock waves. Our considerations are based on Whitham’s theory2 of geometrical shock dynamics. The recently developed method of local analysis3 is applied in order to determine generic singularities. In this case the solutions of partial differential equations describing the geometry of the shock fronts are presented as families of smooth maps with state variables and the set of control parameters dependent on Mach number, time and initial conditions. The space of control parameters of the singularities is analysed, the unfoldings describing the deformations of the canonical germs of shock front singularities are found and corresponding bifurcation diagrams are constructed. 1 Research is supported by the Leverhulme Trust, grant number RPG-2012-568. Whitham, Linear and Nonlinear Waves, (John Wiley & Sons, 1974). 3 J. Eggers and M. A. Fontelos, Panoramas et Synthèses, 38, 69 (2013). 2 G.B. 9:44AM G35.00009 Simulations on shock focusing effects of multiple munitions using Euler equations and geometrical shock dynamics , SHI QIU, VERONCIA ELIASSON, USC — Shock propagation effects from single or multiple munitions onto a specified target and its surroundings have been explored. The results of a single blast wave with fixed energy, E, was compared to that of N multiple blast waves, each with energy E/N and placed in specific geometrical patterns around the intended target. The intention is to increase the severe conditions at the target area while simultaneously reduce collateral damage. Simulations using the Euler equations with a Godunov scheme have been used to study the dynamics of the shock waves. Results show that multiple munitions generate a coalesced shock front that eventually forms a polygonal converging shock, which reconfigures during propagation towards the target. In order to further study this phenomenon, an approach based on Whitham’s theory of geometrical shock dynamics (GSD) has been implemented. In GSD, the motion of the converging shock is computed independent of the flow field behind the shock. Hence, the scheme is efficient and inexpensive and can be used to further analyze the shock focusing effects based on initial location of individual munitions. Results from both simulations will be presented and optical configurations will be discussed. 9:57AM G35.00010 Stochastic analysis and robust optimization for a converging shock wave experimental setting , DANTE DE SANTIS, Stanford Univ, GIANLUCA GERACI, Stanford Univ and Inria Bordeaux, PIETRO CONGEDO, Inria Bordeaux — The efficient generation of ultrahigh pressure is one of the key issues in research related to high energy density physics, as for example Inertial Confinement Fusion reactions. In order to create more stable converging shock configurations, recently, it has been proposed to shape the shock front by the means of obstacles. Such polygonal-shaped shocks are expected to be less sensitive to external disturbances that circular ones, but at the same time obstacles produce a loss of energy during the focusing process of the shocks. The aim of this work is to perform a robust shape optimization of the obstacles by taking into account several experimental uncertainties, thus yielding a more stable and efficient shock configuration. For this purpose, both inviscid and viscous turbulent solvers are coupled with a Polynomial Chaos method. This analysis allows estimating the variability of the maximal temperature and energy. Finally, obstacle shape is optimized in order to maximize the energy concentration and thus provide useful remarks for improving the experience. The effect of neglecting viscous terms in the optimization process is also investigated. Monday, November 24, 2014 8:00AM - 9:57AM Session G36 Jets I — Alcove A - Tobias Rossmann, Lafayette College 8:00AM G36.00001 The influence of upstream boundary conditions on swirling flows undergoing vortex breakdown1 , LOTHAR RUKES, MORITZ SIEBER, KILIAN OBERLEITHNER, OLIVER PASCHEREIT, Chair of Fluid Dynamics, Hermann-Foettinger-Institut, TU Berlin — Swirling jets undergoing vortex breakdown are common in research and technology. In part this is because swirling jets are widely used to anchor the flame position in gas turbines. Recently, the benefit in terms of flashback safety of axial air injection via a center body in the upstream mixing tube of a simplified premixed burner was demonstrated, Reichel (ASME Turbo Expo 2014). However, the presence of a center body alone alters the upstream boundary conditions for the downstream swirling flow. This study investigates how different upstream conditions modify the downstream swirling jet in a more generic setup. A swirling jet facility is used, consisting of a swirler, a pipe, a nozzle and an unconfined part. The focus lies on two large-scale flow features: the precessing vortex core (PVC) and the recirculation bubble. The flow field is measured with Particle Image Velocimetry and proper orthogonal decomposition is conducted to extract the dominant coherent structures. Additionally, a feature tracking approach is used to track the instantaneous shape and position of the recirculation bubble. We find that different center bodies modify the inflow profiles of the unconfined part of the flow in a specific way. This leads to significant differences in the large scale dynamics. 1 Financial support from the German Science Foundation is gratefully acknowledged. 8:13AM G36.00002 Experimental investigation of the influence of temperature differences on the precessing vortex core in swirling jets , MORITZ SIEBER, LOTHAR RUKES, KILIAN OBERLEITHNER, C. OLIVER PASCHEREIT, Chair of Fluid Dynamics, Hermann-Foettinger-Institut, TU Berlin — Swirling jets undergoing vortex breakdown are commonly used in gas turbine combustors. The vortex breakdown is accompanied by a meandering motion of the vortex core around the jet axis. This is referred to as the precessing vortex core, or short PVC. Extensive research has been done on the occurrence of the PVC in isothermal swirling jets. It was demonstrated that the PVC is a global instability mode. Measurements of the isothermal flow in gas turbine combustors usually show the presence of the PVC. However, recent investigations at our institute revealed that the PVC may be supressed in the reacting flow, depending on the flame position. This feature of non-isothermal swirling jets is of particular interest, because the PVC is known to be a robust structure that is hard to suppress in general. A subsequent theoretical investigation of the flow showed that the suppression of the PVC is related to a change of the hydrodynamic stability. This is again related to the temperature distribution within the flow. In the presented work this phenomenon is experimentally investigated in a swirling jet, where temperature differences are generated by electric heating. Therefore, the influence of the temperature can be investigated separately from the combustion. The experimental investigations consistently show that the PVC is strongly reduced by imposing temperature differences on the flow field. These characteristics are obtained by particle image velocimetry and proper orthogonal decomposition. 8:26AM G36.00003 Axial Reynolds Stress Budget for Turbulent Swirling and Non-Swirling Jets , SARA TOUTIAEI, JONATHAN NAUGHTON, University of Wyoming — The terms of the axial Reynolds stress budget were studied for turbulent swirling and non-swirling jets. Laser Doppler anemometry (LDA) was used for acquiring measurements at locations where the two jets were at similar stages of development. A large number of data (∼50,000 samples) was obtained at each measurement point in order to achieve high accuracy for the third order moments. The mean velocity and Reynolds stress results were consistent with previous studies. Production, convection and turbulent transport terms of Axial Reynolds stress equation were used as a means to investigate the differences in the two jets. Higher production in the swirling jet contributes to the higher Reynolds stress magnitude compared with the non swirling jet. In general, production exhibits higher values compared with the turbulence transport for both jet cases. Near the center of the jet where production has near zero values, turbulence transport has higher magnitudes. The turbulence transport is thus moving turbulence away from where it is largely produced toward the center of the jet. The faster development of turbulence transport term coupled with higher production in the swirling jet is found to be responsible for its faster growth compared to the non-swirling jet. 8:39AM G36.00004 Acoustic response of an isothermal coaxial swirling jet , SAPTARSHI BASU, SANTHOSH RUDRASETTY, Department of Mechanical Engineering, Indian Institute of Science, Bangalore, India — This experimental study concerns acoustic response of internal recirculation zone (IRZ) in an unconfined isothermal coaxial swirling jet (geometric swirl number: 2-3). Two IRZ modes with characteristic modified Rossby number Ro m >1 and Ro m ≤ 1 are considered. It is observed that as the amplitude of excitation is increased till a critical magnitude the IRZ topology (time-mean streamline plot obtained from PIV) with Ro m ≤ 1 fans-out/widens. At super-critical amplitude the IRZ transits back and finally undergoes a transverse shrinkage. However, fanning-out is absent in Ro m >1 IRZ mode as this flow is dominated by pressure deficit due to entrainment (at the interface between central and co-annular jet) effect when compared to radial pressure gradient due to rotational influence. Thus the central jet with high kinetic pressure exists in Ro m >1mode (swirl fails to penetrate till the central axis and impart recirculation) but not in Ro m ≤ 1. The imparted acoustics fails to disrupt IRZ in Ro m >1against the high kinetic pressure of the central stream, failing to impart a fan-out. 8:52AM G36.00005 Instability wavepackets in optimally forced subsonic jets , ONOFRIO SEMERARO, LUTZ LESSHAFFT, LADHYX - Ecole Polytechnique - F-91128, Palaiseau - France — Jets are known to be very receptive to ambient perturbations, due to their strong convective instability. Coherent wavepackets are formed as a result, which may, as recent experiments suggest, represent the dominant source of jet noise. We model these wavepackets as the linear flow response to a harmonic forcing input that yields the highest amplification in a fully non-parallel setting. Axisymmetric turbulent jets are considered. Mean flows are taken from numerical simulations as well as from experiments, characterized by high subsonic Mach numbers (M a = 0.84 and M a = 0.9) and high Reynolds numbers. The formalism relies on singular mode decomposition of the linear resolvent operator, based on the fully compressible Navier-Stokes equations. Two different objectives are used for the optimization: the maximum energy of the near-field wavepacket and the maximum radiated acoustic power. The effects of turbulence are modeled through a turbulent viscosity formulation. The predicted acoustic radiation will be compared against simulation and experiment, and the influence of the chosen turbulent viscosity model will be discussed. 9:05AM G36.00006 Analysis of a steady laminar stagnation flow and its self-similarity properties , GIANFRANCO SCRIBANO, FABRIZIO BISETTI, King Abdullah University of Science and Technology — The velocity field in a steady laminar stagnation flow is analyzed experimentally and numerically. The flow configuration is characterized by a stagnation plane formed between two streams flowing from opposite directions. This configuration is used in the study of flames and condensing aerosols. The flow is characterized geometrically by the nozzle diameter D and the separation H between the nozzles. Together with the bulk velocity U, the separation H is used to define the Reynolds number. Particle Image Velocimetry is used to measure the velocity field and simulations are conducted to further characterize the flow. For this analysis, we explore values of H/D equal to 0.5, 1, 1.5, and 2, and values of the Reynolds number equal to 300, 600, 900, and 1200. The analysis is repeated for four nozzles having identical shape and diameters D equal to 7.5, 15, 30, and 35 mm. Our results show that the non-dimensional velocity fields are parametrized well by Re and H/D for all the diameters and that the simulations agree with the PIV data very well. The non-dimensional fields depend mostly on H/D, while the influence of Re is negligible for Re > 300, in accordance with theoretical results. The parameter H/D plays an important role in influencing the flow inside and outside the nozzle. 9:18AM G36.00007 Global stability analysis of electrified jets1 , JAVIER RIVERO-RODRIGUEZ, MIGUEL PÉREZSABORID, Universidad de Sevilla — Electrospinning is a common process used to produce micro and nano polymeric fibers. In this technique, the whipping mode of a very thin electrified jet generated in an electrospray device is nhanced in order to increase its elongation. In this work, we use a theoretical Eulerian model that describes the kinematics and dynamics of the midline of the jet, its radius and convective velocity. The model equations result from balances of mass, linear and angular momentum applied to any differential slice of the jet together with constitutive laws for viscous forces and moments, as well as appropriate expressions for capillary and electrical forces. As a first step towards computing the complete nonlinear, transient dynamics of the electrified jet, we have performed a global stability analysis of the forementioned equations and compared the results with experimental data obtained by Guillaume et al [2011] and Guerrero-Millán et al [2014]. 1 The support of the Ministry of Science and Innovation of Spain (Project DPI 2010-20450-C03-02) is acknowledged. 9:31AM G36.00008 Study on a liquid jet with cavitation bubbles1 , AKIHITO KIYAMA, YOSHIYUKI TAGAWA, Tokyo Univ of Agri & Tech — A focused liquid jet is important in medical applications such as needle-free drug injection systems. A method for generating a liquid jet by laser-induced shock wave is proposed. However, there are some problems. Hence we examine another method for generating a focused liquid jet. We drop a liquid filled test tube on the rigid plate, leading to the emergence of a jet. Within certain experimental conditions, the jet velocity in our experiment agrees well with the semiempirical relation proposed by Tagawa, et al., (2012, Phys. Rev. X) and Peters, et al. (2013, J. Fluid Mech.). In other conditions, we find that the jet velocity remarkably increases. In order to understand the jet velocity increment, we use two high-speed cameras: One records motion of a jet. Another films cavitation bubbles inside a liquid bath. We categorize jets into three types based on their shape and the existence of cavitation bubbles. We find that the jet with cavitation bubbles is much faster than that without cavitation bubbles. For elucidating the mechanism of jet velocity increment, we discuss the effect of pressure wave, which propagates in a liquid bath. We propose a model for describing these phenomena and verify it experimentally. 1 JSPS KAKENHI 26709007 9:44AM G36.00009 Characterization of far-field jet flows from complex nozzles via Particle Tracking Velocimetry , JIN-TAE KIM, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, ALEX LIBERZON, School of Mechanical Engineering, Tel Aviv University, Israel, CARLO ZUNIGA ZAMALLOA, University of Illinois at Urbana-Champaign, LEONARDO P. CHAMORRO, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — Understanding the advection, diffusion and mixing of turbulence and scalars of jet flows under various geometric configurations and Reynolds numbers is of high relevance in environmental and engineering applications. In this experimental study, we characterize the far-field turbulence of jet flows in the proximity of twelve rotor diameters downstream of a series of complex nozzle geometries. The jet flows are released into a still body of water from a series of nozzles of different cross sections but with common hydraulic diameter dh= 0.01 m at a Reynolds number Re= U0*dh/ν approx. 7000, where U0 is the flow velocity at the outlet of the jet and ν is the kinematic viscosity of the flow. The system is closed-loop and seeded with particles of 100 µm diameter. Results are analyzed from Lagrangian and Eulerian frames of references via 3D particle tracking velocimetry (OpenPTV, www.openptv.net). Lagrangian features of the particles are characterized in terms of the nozzle geometries and high-order turbulence statistics are obtained at various planes within the interrogation volume, which mimics 3D PIV. Monday, November 24, 2014 10:30AM - 12:40PM Session H1 General Fluid Dynamics II — 3000 - Sigurdur Thoroddsen, King Abdullah University of Science and Technology 10:30AM H1.00001 Crown sealing and buckling instability during sphere impact , JEREMY MARSTON, Texas Tech University, TADD TRUSCOTT, Brigham Young University, MOHAMMAD MANSOOR, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology — We present results from an experimental investigation of the classical crown splash and sealing phenomena observed during the impact of spheres onto quiescent liquid pools for a range of ambient pressures. A 6-metre tall vacuum chamber was used to provide the required ambient conditions from atmospheric conditions down to 1/16th of an atmosphere. We pay particular attention to the above-surface crown formation and ensuing dynamics, including the buckling instability of the crown just before seal. In addition, we have observed the very rapid motions of the ejecta formed immediately after impact, using ultra-high-speed imaging at frame rates over 400,000 fps, which reveals qualitative differences at reduced pressures. 10:43AM H1.00002 Dripping like Pollock , BERNARDO PALACIOS, SANDRA ZETINA, ROBERTO ZENIT, Universidad Nacional Autonoma de Mexico, CHRIS MCGLINCHEY, Museum of Modern Art, New York — In this investigation we have reproduced, in a controlled manner, the dripping technique used by Jackson Pollock to creat abstract paintings.We drip a fluid jet on top of a horizontal surface, varying the height from the substrate, the liquid flow rate and the displacement speed of the nozzle. We are able to reproduce the coiling and dripping instabilities which characterize the characteristic patterns of Pollock’s paintings. We also found that the non Newtonian properties of the paints are of great importance to create these patterns. Since the fluid jets are rapidly stretched, the sudden increase of extensional viscosity plays an important role to produce the characteristic Pollock patterns. Some preliminary results will be shown and discussed. 10:56AM H1.00003 Entropic Lattice Boltzmann Methods for Fluid Mechanics: Thermal, Multi-phase and Turbulence , SHYAM CHIKATAMARLA, F. BOESCH, N. FRAPOLLI, A. MAZLOOMI, I. KARLIN, ETH Zurich — With its roots in statistical mechanics and kinetic theory, the lattice Boltzmann method (LBM) is a paradigm-changing innovation, offering for the first time an intrinsically parallel CFD algorithm. Over the past two decades, LBM has achieved numerous results in the field of CFD and is now in a position to challenge state-of-the art CFD techniques [1]. Major restyling of LBM resulted in an unconditionally stable entropic LBM which restored Second Law (Boltzmann H theorem) in the LBM kinetics and thus enabled affordable direct simulations of fluid turbulence [2, 3]. In this talk, we shall review recent advances in ELBM as a practical, modeling-free tool for simulation of complex flow phenomenon. We shall present recent simulations of fluid turbulence including turbulent channel flow, flow past a circular cylinder, creation and dynamics of vortex tubes, and flow past a surface mounted cube. Apart from its achievements in turbulent flow simulations, ELBM has also presented us the opportunity to extend lattice Boltzmann method to higher order lattices which shall be employed for turbulent, multi-phase and thermal flow simulations. A new class of entropy functions are proposed to handle non-ideal equation of state and surface tension terms in multi-phase flows. It is shown the entropy principle brings unconditional stability and thermodynamic consistency to all the three flow regimes considered here. Acknowledgements: ERC Advanced Grant “ELBM” and CSCS grant s437 are deeply acknowledged. References: [1] Chikatamarla et al, J.Fluid.Mech, 656 (2010); Physica.A,392(2013) [2] Chikatamarla and Karlin, Phys.Rev.Lett., 010201 (2006); Phys.Rev.Lett, 190601 (2006). [3] Karlin et al, Europhys. Letters, 47, no. 2, p. 182, 1999. 11:09AM H1.00004 Validation of X-Ray CT-measured Liquid Concentration against LIF , TYLER SOWELL, ZACHARY LEE, MICHAEL BENSON, BRET VAN POPPEL, THOMAS NELSON, United States Military Academy, PABLO VASQUEZ GUZMAN, REBECCA FAHRIG, JOHN EATON, WALDO HINSHAW, Stanford University, MATTHEW KURMAN, MICHAEL TESS, CHOL-BUM KWEON, U.S. Army Research Laboratory — Dense spray near the nozzle exit requires more research. X-Ray Computed Tomography (CT) technique has shown that it could capture spray patterns similar to that of a conventional Shadowgraphy; however, liquid density measured with the X-Ray CT technique lacks validation. Thus, the objective of the current study is to validate liquid density measured with the X-Ray CT technique against that of the conventional Laser-Induced Fluorescence (LIF) method. Water solution with 150 parts per billion (ppb) of Rhodamine WT dye is sprayed into a cold spray chamber by using a pressure swirl atomizer. An Nd:YAG laser with a light-sheet optics is used to fluoresce Rhodamine WT dye in water spray and a high-speed CMOS camera with a filter is employed to measure quantitative liquid concentrations approximately thirty (30) nozzle diameters downstream. The intensity of fluorescence correlates linearly with the amount of Rhodamine WT dye in the water, enabling mass distribution measurement of the liquid spray. As the X-Ray CT technique measures liquid mass distribution, the X-Ray CT measured spray density can be validated by the proven conventional LIF method. 11:22AM H1.00005 Effects of Coral Colony Properties on the Dissipation of Wave Energy , ANNE STAPLES, LEILA AZADANI, Virginia Tech — About 40% of the world’s population lives in coastal areas, therefore coastal protection is of particular importance given the increasing frequency of superstorms like Katrina and Sandy, and associated storm surges and flooding. Coral reefs are recognized to provide coastal regions with excellent protection against high-energy wave impacts. The hydrodynamic roughness of coral reefs caused by the presence of several multiscale coral colonies plays a crucial role in the dissipation of wave energy. Here, we design a series of experiments to study the effects of the properties of corals on wave attenuation. As a first step, prototypes of two species (elkhorn and staghorn) of natural branching corals are produced using 3D printing technologies. Wave tank experiments are then performed on the 3D printed samples and the natural corals. The effects of parameters such as coral geometry and coral stiffness on wave dissipation are investigated. In order to study the effect of the geometry of the corals, experiments are performed for both species of corals. The effect of coral stiffness is investigated by using different additive manufacturing materials, which gives different flexibilities to the coral models. Preliminary results from these experiments will be presented. 11:35AM H1.00006 Steady and Unsteady Size-Dependent Couple Stress Creeping Flow , GARY DARGUSH, ALI HADJESFANDIARI, AREZOO HAJESFANDIARI, HAOYU ZHANG, University at Buffalo, State University of New York — As an inevitable consequence of non-central forces acting at the atomic level, couple-stresses appear within the framework of continuum mechanics, and force-stress must be considered as a general non-symmetric tensor. Based upon recent theoretical work, the couple-stress tensor is shown to be skew-symmetric, as is its energy conjugate mean curvature rate tensor. Within this fully consistent couple stress continuum theory applied to fluid flow, there appears a material length scale l that becomes increasingly important as the characteristic geometric dimension of the problem reduces. In the present work, we study the effects of this theory for creeping incompressible flows by developing fundamental point force and point couple solutions, along with boundary integral representations for both the steady and unsteady cases. In addition, we develop the corresponding boundary element methods and solve several problems that highlight the size-dependent nature of these flows, which may be most relevant to a range of micro- and nano-scale technologies. 11:48AM H1.00007 Numerical simulation of the induced magnetic field within a rotating concentric annulus with self gravity1 , ARES CABELLO, RUBEN AVILA, UNAM — In order to study the GEODYNAMO is necessary to know the behavior of the natural convection of the electrical conducting fluid confined in a rotating spherical shell. In this work, the convective patterns within this geometry are presented. Natural convection is induced by a temperature difference between the inner sphere and outer sphere and a gravitational field which varies like 1/r3 . The patterns presented are known as Busse cells and are moving around the rotational axis. The magnetic fields induced by previously mentioned convective patterns are presented. These magnetic fields are obtained by solving the equations of MHD. The free-divergence magnetic field is obtained by using a Lagrange multiplier scheme. All the equations are solved based on a spectral element method (SEM). To avoid the singularity at the poles, the cubed-sphere algorithm is used to generate the mesh. The obtained magnetic fields are similar to the results reported by other research groups. 1 Thanks to DGAPA-PAPIIT project: IN117314-3. 12:01PM H1.00008 Turbulent flows interacting with groups of obstacles , SONIA TADDEI, COSTANTINO MANES, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — The interaction between a turbulent incoming flow and patches of obstacles (with circular cross section in plan view with diameter D) that contain a number of individual cylinders (Nc is the number of cylinders and d is their diameter) with different void-fractions (φ = Nc d2 /D 2 ) are studied. Streamwise-spanwise plane PIV measurements, at the mid-height of the patches, of the wakes generated by the different void-fractions show that the three-dimensionality of the patches and the incoming turbulence lead to different results compared to the laminar 2D cases available in literature. In particular, for void-fraction φ > 0.1, no steady recirculation region is detected behind the obstacles, and even for lower φ, its streamwise length is drastically reduced. Furthermore, for higher φ (> 0.15), the wakes are not comparable with the one of a solid cylinder with the same height and diameter, as it happens for the laminar 2D cases. Results from vertical PIV measurements along the symmetry plane of the patches will also be discussed. 12:14PM H1.00009 Transitioning from a single-phase fluid to a porous medium: a boundary layer approach , MOHIT P. DALWADI, S. JON CHAPMAN, JAMES M. OLIVER, SARAH L. WATERS, University of Oxford — Pressure-driven laminar channel flow is a classic problem in fluid mechanics, and the resultant Poiseuille flow is one of the few exact solutions to the Navier-Stokes equations. If the channel interior is a porous medium (governed by Darcy’s law) rather than a single-phase fluid, the resultant behaviour is plug flow. But what happens when these two flow regions are coupled, as is the case for industrial membrane filtration systems or biological tissue engineering problems? How does one flow transition to the other? We use asymptotic methods to investigate pressure-driven flow through a long channel completely blocked by a finite-length porous obstacle. We analytically solve for the flow at both small and large Reynolds number (whilst remaining within the laminar regime). The boundary layer structure is surprisingly intricate for large Reynolds number. In that limit, the structure is markedly different depending on whether there is inflow or outflow through the porous medium, there being six asymptotic regions for inflow and three for outflow. We have extended this result to a wide class of 3D porous obstacles within a Hele-Shaw cell. We obtain general boundary conditions to couple the outer flows, and find that these conditions are far from obvious at higher order. 12:27PM H1.00010 Swirling flows with imposed radial flow - a model for a cold accretion disk? , RICH KERSWELL, Bristol University — There is a lot of current interest in exploring whether Keplerian-type flows (dI/dr > 0 but dΩ/dr < 0 where I is the angular momentum and Ω the angular velocity) can harbour nontrivial hydrodynamic flows in order to explain the inferred presence of turbulence in cold accretion disks (e.g. Balbus, Nature 2011). Invariably, the very small (accreting) inflow is neglected in accretion disk models. I will discuss how this could be a dangerous omission by building upon recent work (Gallet et al. 2010, Ilin & Morgulis 2013) which shows linear instability of otherwise-stable Taylor-Couette flows when radial flow is imposed. Monday, November 24, 2014 10:30AM - 12:40PM Session H2 Surface Tension Effects: General — 3002 - John Lister, University of Cambridge 10:30AM H2.00001 Modelling two-phase slug self-propulsion in a capillary , MATHIEU SELLIER, IRSHAD KHODABOCUS, VOLKER NOCK, University of Canterbury, CLAUDE VERDIER, CNRS — We present a numerical study of the flow of a droplet of two miscible fluids juxtaposed in a capillary tube. We show that the asymmetry of the system results in the spontaneous motion of the composite droplet which can be of potential use in microfluidics applications or for transport in porous media. We also show that the droplet motion is sustained until the miscible fluids have become fully mixed. The proposed numerical model is implemented in COMSOL Mutiphysics using the Laminar Two-Phase Flow Phase Field Method coupled with an advection-diffusion chemical concentration equation. The results are validated using experimental data from Reference 1 and explained in the context of a simplified phenomenological model. An important benefit of the simulation is the ability to investigate the transient behaviour composite droplet and the development of the internal flow features in the system which could allow the development of optimized systems. [1] Bico, J., Quéré, D., Liquid trains in a tube, Europhys. Lett., 51(5), 546 (2000) 10:43AM H2.00002 Viscous peeling with capillary suction , GUNNAR PENG, JOHN LISTER, University of Cambridge — If an elastic tape is stuck to a rigid substrate by a thin film of viscous fluid and then peeled off by pulling at a small angle to the horizontal, then both viscous and capillary forces affect the peeling speed (McEwan and Taylor, 1966). If there is no capillary meniscus (e.g. if the peeling is due to viscous fluid being injected under the tape), then the peeling speed is given by a Cox–Voinov-like law, and is an increasing function of the peeling angle. We show that, with a meniscus present, the effect of the capillary forces is to suck down the tape, reducing the effective peeling angle and hence the peeling speed. When surface tension dominates and the peeling speed tends to zero, the system transitions to a new state whose time-evolution can be described by a system of coupled ordinary differential equations. These asymptotic results are confirmed by numerical calculations. Similar results hold for the peeling-by-bending of elastic beams, with “angle” replaced by “curvature” (i.e. bending moment). 10:56AM H2.00003 Painting Pictures with Whisky , HYOUNGSOO KIM, FRANÇOIS BOULOGNE, EUJIN UM, IAN JACOBI, HOWARD STONE, Princeton University — Have you ever looked at the dried mark of whisky on the glass? While the whisky evaporates, various solid components inside the whisky are deposited with a peculiar pattern, which creates a beautiful picture. This particle patterning is induced by the solutal Marangoni effect. We investigate this effect on both the flow behavior and the particle deposition patterns in binary-mixture droplet evaporation by varying the concentration ratio between ethanol and water. To visualize the particle and fluid motion, we perform Particle Image Velocimetry. We observe that at the beginning stage complex circulating flow patterns occurred, which are triggered by the surface tension gradient, i.e. Marangoni effect. Ethanol first evaporates due to the lower vapor pressure compared to water. When the ethanol has vanished, a radial flow pattern is observed. Furthermore, we find that as the initial ethanol concentration increases, the mobility of the receding contact line increased. At high ethanol concentrations, the contact line kept receding so as to draw groups of particles that deposited in an annular pattern. We thank Ernie Button for sharing with us many beautiful images of whisky after it had dried. 11:09AM H2.00004 Capillary Flows along Open Channel Conduits: the Open-Star Section1 , MARK WEISLOGEL, YONGKANG CHEN, THANH NGUYEN, JOHN GEILE, Portland State University, MICHAEL CALLAHAN, NASA Johnson Space Center — Capillary rise in tubes, channels, and grooves has received significant attention in the literature for over 100 years. In yet another incremental extension of related work, a transient capillary rise problem is solved for spontaneous flow along an interconnected array of open channels forming what is referred to as an “open-star” section. This geometry possesses several attractive characteristics including passive phase separations and high diffusive gas transport rates. Despite the complex geometry, novel and convenient approximations for capillary pressure and viscous resistance enable closed form predictions of the flow. As part of the solution, a combined scaling approach is applied that identifies unsteady-inertial-capillary, convective-inertial-capillary, and visco-capillary transient regimes in a single parameter. Drop tower experiments are performed employing 3-D printed conduits to corroborate all findings. 1 NASA NNX09AP66A, Glenn Research Center 11:22AM H2.00005 Characteristics of air entrainment during dynamic wetting failure along a planar substrate , SATISH KUMAR, ERIC VANDRE, University of Minnesota, MARCIO CARVALHO, PUC-Rio — We report results of experiments characterizing the onset of air entrainment during dynamic wetting failure along a planar substrate (J. Fluid Mech. 747 (2014) 119). Using high-speed video, dynamic contact line (DCL) behavior is recorded as a tape substrate is drawn through a bath of a glycerol/water solution. Air entrainment is identified by triangular air films that elongate from the DCL above a critical substrate speed. Meniscus confinement between the substrate and a stationary plate delays air entrainment to higher speeds for a wide range of liquid viscosities. Liquid pressurization moves the meniscus near a sharp corner, changing its shape and further postponing air entrainment. Meniscus shapes recorded near the DCL demonstrate that smaller entrained air films appear in the more viscous solutions. Regardless of size, air films become unstable to thickness perturbations and ultimately rupture, leading to entrainment of air bubbles. Recorded critical speeds and air-film sizes compare well to predictions from a hydrodynamic model for dynamic wetting failure, indicating that strong air stresses near the DCL trigger the onset of air entrainment. The results suggest strategies for postponing air entrainment, which often limits the maximum speed of industrial coating processes. 11:35AM H2.00006 Shallow flows over surfaces of patterned wettability1 , MORGANE GRIVEL, DAVID JEON, MORTEZA GHARIB, Caltech — Our previous work showed that surfaces with spatially patterned wetting properties induce passive displacements of shallow flows. Polycarbonate plates were patterned with hydrophobic and hydrophilic stripes, and a thin, rectangular water jet impinged on the patterned surface. We reported development of intriguing roller structures at the hydrophobic-hydrophilic interfaces. In our present work, we study the effect of varying the stripes’ width, spacing, and orientation on the dynamics of these roller structures. Specifically, we are interested in the vortex generation and air entrainment by the rollers. We report quantitative results to this effect. We will also discuss potential uses of this technique for modifying contact line dynamics and bow waves near ships. 1 This work is supported by the Office of Naval Research (grant # ONR- N00014-11-1-0031) and by NSF-GRFP. 11:48AM H2.00007 Capillary-Inertial Colloidal Catapult upon Drop Coalescence , ROGER CHAVEZ, FANGJIE LIU, Duke Univ, JAMES FENG, University of British Columbia, CHUAN-HUA CHEN, Duke Univ — To discharge micron-sized particles such as colloidal contaminants and biological spores, an enormous power density is needed to compete against the strong adhesive forces between the small particles and the supporting surface as well as the significant air friction exerted on the particles. Here, we demonstrate a colloidal catapult that achieves such a high power density by extracting surface energy released upon drop coalescence within an extremely short time period, which is governed by the capillary-inertial process converting the released surface energy into the bulk inertia of the merged drop. When two drops coalesce on top of a spherical particle, the resulting capillary-inertial oscillation is perturbed by the solid particle, giving rise to a net momentum eventually propelling the particle to launch from the supporting surface. The measured launching velocity follows a scaling law that accounts for the redistribution of the momentum of the merged drop onto the particle-drop complex, and is therefore proportional to the capillary-inertial velocity characterizing the coalescing drops. The interfacial flow process associated with the colloidal catapult is elucidated with both high-speed imaging and phase-field simulations. 12:01PM H2.00008 Translational and rotational diffusion of Janus nanoparticles at liquid interfaces , HOSSEIN REZVANTALAB, SHAHAB SHOJAEI-ZADEH, Rutgers University — We use molecular dynamics simulations to understand the thermal motion of nanometer-sized Janus particles at the interface between two immiscible fluids. We consider spherical nanoparticles composed of two sides with different affinity to fluid phases, and evaluate their dynamics and changes in fluid structure as a function of particle size and surface chemistry. We show that as the amphiphilicity increases upon enhancing the wetting of each side with its favored fluid, the in-plane diffusivity at the interface becomes slower. Detail analysis of the fluid structure reveals that this is mainly due to formation of a denser adsorption layer around more amphiphilic particles, which leads to increased drag acting against nanoparticle motion. Similarly, the rotational thermal motion of Janus particles is reduced compared to their homogeneous counterparts as a result of the higher resistance of neighboring fluid species against rotation. We also incorporate the influence of fluid density and surface tension on the interfacial dynamics of such Janus nanoparticles. Our findings may have implications in understanding the adsorption mechanism of drugs and protein molecules with anisotropic surface properties to biological interfaces including cell membranes. 12:14PM H2.00009 Driven interfacial particles acting as capillary dipoles , AARON DOERR, STEFFEN HARDT, Tech Univ Darmstadt — Solid particles attached to fluid-fluid interfaces exhibit a number of interesting phenomena such as the formation of regular crystal-like structures. The underlying particle-particle interactions as well as their various origins have been the subject of many studies mainly covering static situations. By contrast, the case of driven particles moving along a fluid-fluid interface is still widely unexplored. By means of perturbation methods we demonstrate that such particles cause a dipolar interfacial deformation which decays algebraically with distance from the particle center. In our study, we focus on particles at interfaces between two fluids of high viscosity ratio, equilibrium contact angles close to 90◦ , and a pinned three-phase contact line. It is shown that the moving particles change their orientation with respect to the interface normal at zero velocity, similar to the occurrence of a trim angle in ship hydrodynamics. The corresponding interfacial deformation gives rise to an direction-dependent particle-particle interaction which can be approximated via linear superposition in the case of large separations relative to the particle diameter. 12:27PM H2.00010 Collapse and sinking of self-assembled sphere clusters on a liquid-liquid interface , STEVEN JONES, NIKI ABBASI, ABHINAV AHUJA, VIVIAN TRUONG, SCOTT TSAI, Ryerson University — The self-assembly of objects on a liquid-liquid interface is a phenomenon that has attracted a lot of attention. When a single settling sphere has insufficient gravitational energy to break through a liquid-liquid interface, multiple spheres can self-assemble on the interface, by capillary and buoyancy forces, to form a cluster. If a sphere cluster’s gravitational energy overcomes the interfacial tension energy barrier of the liquid-liquid interface, the cluster will sink through the interface. Here we show with experiments that small spheres approaching an oil-water interface will self-assemble into clusters, and at a critical size pass through the interface. We demonstrate that the size of a sphere cluster at the time of interface breakthrough can be controlled by altering the sphere radius and the liquid-liquid interfacial tension. We also find that the critical cluster size changes depending on the way the spheres are deposited: spheres deposited into a monolayer raft configuration will sink though the interface as a larger cluster than spheres stacked into a spheroidal geometry. We find that the number of spheres in each sinking cluster scales with power-laws of the Bond number, and we observe different power-laws for raft and stack cluster configurations. Monday, November 24, 2014 10:30AM - 12:40PM Session H3 Porous Media Flows V: Theory — 3004 - Markus Schmuck, Heriot-Watt University 10:30AM H3.00001 Feedback-induced phase transitions in active porous media1 , SAMUEL OCKO, Massachusetts Institute of Technology, L. MAHADEVAN, Harvard University — We consider a reduced-complexity model for an active porous medium where flow and resistance are coupled to each other i.e. the porous medium is modified by the flow and in turn modifies the flow. Using numerical simulations, we show that this results in both channelization and wall-building transitions depending on the form of the feedback. A continuum model allows us to understand the qualitative features of the resulting phase diagram, and suggests ways to realize complex architectures using simple rules in engineered systems. 1 Human Frontiers Science Program grant RGP0066/2012- TURNER 10:43AM H3.00002 Dynamics of clogging in drying porous media1 , C. NADIR KAPLAN, L. MAHADEVAN, Harvard University — Drying in porous media pervades a range of phenomena from brine evaporation arrested in porous bricks, causing efflorescence, i.e. salt aggregation on the surface where vapor leaves the medium, to clogging of reservoir rocks via salt precipitation when carbon dioxide is injected for geological storage. During the process of drying, the permeability and porosity of the medium may change due to the solute accumulation as a function of the particle concentration, in turn affecting the evaporation rate and the dynamics of the fluid flow imposed by it. To examine the dynamics of these coupled quantities, we develop a multiphase model of the particulate flow of a saline suspension in a porous medium, induced by evaporation. We further provide dimensional arguments as to how the salt concentration and the resulting change in permeability determine the transition between efflorescence and salt precipitation in the bulk. 1 This research was supported by the Air Force Office of Scientific Research (AFOSR) under Award FA9550-09-1-0669-DOD35CAP and the Kavli Institute for Bionano Science and Technology at Harvard University. 10:56AM H3.00003 Simplified model for fouling of a pleated membrane filter1 , PEJMAN SANAEI, LINDA CUMMINGS, New Jersey Inst of Tech — Pleated filter cartridge are widely used to remove undesired impurities from a fluid. A filter membrane is sandwiched between porous support layers, then pleated and packed in to an annular cylindrical cartridge. Although this arrangement offers a high ratio of surface filtration area to volume, the filter performance (measured, e.g., by graph of total flux versus throughput for a given pressure drop), is not as good as a flat filter membrane. The reasons for this difference in performance are currently unclear, but likely factors include the additional resistance of the porous support layers upstream and downstream of the membrane, the pleat packing density (PPD) and possible damage to the membrane during the pleating process. To investigate this, we propose a simplified mathematical model of the filtration within a single pleat. We consider the fluid dynamics through the membrane and support layers, and propose a model by which the pores of the membrane become fouled (i) by particles smaller than the membrane pore size ; and (ii) by particles larger than the pores.We present some simulations of our model, investigating how flow and fouling differ between not only flat and pleated membranes, but also for support layers of different permeability profiles. 1 NSF DMS-1261596 11:09AM H3.00004 Interception efficiency in flow of power-law fluids past confined porous bodies , SETAREH SHAHSAVARI, GARETH MCKINLEY, Massachusetts Inst of Tech-MIT — Understanding the flow of power-law fluids through porous media is important for a wide range of filtration and sedimentation processes. In this study, the mobility of power-law fluids through porous media is investigated numerically and we use parametric studies to systematically understand the individual roles of geometrical characteristics, rheological properties as well as flow conditions. In addition, an analytical solution is presented that can be used as a modified Darcy law for generalized Newtonian fluids. Building on this modified Darcy law, the incompressible laminar flow of power-law and Carreau fluids past a confined porous body is modeled numerically. From the simulations we calculate the flow interception efficiency, which provides a measure of the fraction of streamlines that intercept a porous collector. Finally, the interception efficiency of power-law fluids are compared with the case of a Newtonian fluid. The focus of this work is principally for flow of inelastic fluids in fibrous media; however, the methodology can also be extended to other porous media. 11:22AM H3.00005 On the determination of a generalized Darcy equation for yield stress fluid in porous media using a LB TRT scheme1 , LAURENT TALON, THIBAUD CHEVALIER, lab. FAST, UPMC, CNRS UMR7608 — Non-Newtonian fluids have practical applications in very different domains. Indeed, polymer mixture, paints, slurries, colloidal suspensions, emulsions, foams or heavy oil present complex rheologies. Among the large number of different non-Newtonian fluids an important class of behavior is represented by the yield-stress fluids, viz. fluids that require a minimum of stress to flow. Yield stress fluids are usually modelled as a Bingham fluid or by the Herschel-Bulkley equation. However, simulating flow of a Bingham fluid in porous media still remains a challenging task as the yield stress may significantly alter the numerical stability and precision. In the present work, we use a Lattice-Boltzmann TRT scheme to determine this type of flow in a synthetic porous medium or fracture. Different pressure drops ∆P have been applied in order to derive a generalization of the Darcy’s equation. Three different scaling regimes can be distinguished when plotting the dimensionless flow rate q as function of the distance to the critical pressure ∆P − ∆Pc . In this presentation, we will investigate the importance of the heterogeneities on those flowing regimes. 1 ANR-12-MONU-0011 11:35AM H3.00006 New upscaled equations for multiphase flows in porous media based on a phase field formulation for general free energies , MARKUS SCHMUCK, Maxwell Institute for Mathematical Sciences and Department of Mathematics, Heriot-Watt University, Edinburgh, UK, MARC PRADAS, Department of Chemical Engineering, Imperial College, London, UK, GRIGORIOS A. PAVLIOTIS, Department of Mathematics, Imperial College, London, UK, SERAFIM KALLIADASIS, Department of Chemical Engineering, Imperial College, London, UK — Based on thermodynamic and variational principles we formulate novel equations for mixtures of incompressible fluids in strongly heterogeneous domains, such as composites and porous media, using elements from the regular solution theory. Starting with equations that fully resolve the pores of a porous medium, represented as a periodic covering of a single reference pore, we rigorously derive effective macroscopic phase field equations under the assumption of periodic and strongly convective flow. Our derivation is based on the multiple scale method with drift and our recently introduced splitting strategy for Ginzburg-Landau/Cahn-Hilliard-type equations [1]. We discover systematically diffusion-dispersion relations (including TaylorAris-dispersion) as in classical convection-diffusion problems. Our results represent a systematic and efficient computational strategy to macroscopically track interfaces in heterogeneous media which together with the well-known versatility of phase field models forms a promising basis for the analysis of a wide spectrum of engineering and scientific applications such as oil recovery, for instance. [1] M. Schmuck, M. Pradas, G.A. Pavliotis and S. Kalliadasis, Nonlinearity 26:3259-3277 2013. 11:48AM H3.00007 Oscillation-Free Methods for Modeling Fluid-Porous Interfaces Using Segregated Solvers on Unstructured Grids , MILOS STANIC, University of Twente, MARKUS NORDLUND, ARKADIUSZ KUCZAJ, Philip Morris International R&D, Philip Morris Products S.A., EDOARDO FREDERIX, BERNARD GEURTS, University of Twente — Porous media flows can be found in a large number of fields ranging from engineering to medical applications. A volume-averaged approach to simulating porous media is often used because of its practicality and computational efficiency. Derivation of the volume-averaged porous flow equations introduces additional porous resistance terms to the momentum equation. When discretized these porous resistance terms create a body force discontinuity at the porous-fluid interface, which may lead to spurious oscillations if not accounted for properly. A variety of numerical techniques has been proposed to solve this problem, but few of them have concentrated on collocated grids and segregated solvers, which have wide applications in academia and industry. In this work we discuss the source of the spurious oscillations, quantify their amplitude and apply interface treatments methods that successfully remove the oscillations. The interface treatment methods are tested in a variety of realistic scenarios, including the porous plug and Beaver-Joseph test cases and show excellent results, minimizing or entirely removing the spurious oscillations at the porous-fluid interface. This research was financially supported by Philip Morris Products S.A. 12:01PM H3.00008 Description of multiphase flows in porous media using an effective convective Cahn-Hilliard equation , RAJAGOPAL VELLINGIRI, MARC PRADAS, Department of Chemical Engineering, Imperial College London, UK, MARKUS SCHMUCK, School of Mathematical and Computer Sciences and the Maxwell Institute for Mathematical Sciences, Heriot-Watt University, UK, SERAFIM KALLIADASIS, Department of Chemical Engineering, Imperial College London, UK — Immiscible two-phase flows in porous media find a variety of applications such as microfluidics, oil extraction from reservoirs and chromatography, to name but a few. In this study, we investigate the dynamics of interfaces in porous media using an effective convective Cahn-Hilliard equation which was derived in [1] from a Stokes-Cahn-Hilliard equation for microscopic heterogeneous domains by means of a homogenization methodology. We consider different types of microstructures, including periodic and non-periodic, observing that the macroscopic model is able to retain the microscopic features, hence indicating that our formulation provides an efficient and systematic computational framework to track interfaces in porous media. [1] M. Schmuck, M. Pradas, G.A. Pavliotis and S. Kalliadasis, 2013 “Derivation of effective macroscopic Stokes–Cahn–Hilliard equations for periodic immiscible flows in porous media,” Nonlinearity 26 3259-3277. 12:14PM H3.00009 Continuum approach for aerothermal flow through ablative porous material using discontinuous Galerkin discretization. , PIERRE SCHROOYEN1 , PHILIPPE CHATELAIN, Universite catholique de Louvain, Institute of Mechanics, Materials and Civil Engineering, KOEN HILLEWAERT, Cenaero, THIERRY E. MAGIN2 , von Karman Institute for Fluid Dynamics, Aeronautics and Aerospace Department — The atmospheric entry of spacecraft presents several challenges in simulating the aerothermal flow around the heat shield. Predicting an accurate heat-flux is a complex task, especially regarding the interaction between the flow in the free stream and the erosion of the thermal protection material. To capture this interaction, a continuum approach is developed to go progressively from the region fully occupied by fluid to a receding porous medium. The volume averaged Navier-Stokes equations are used to model both phases in the same computational domain considering a single set of conservation laws. The porosity is itself a variable of the computation, allowing to take volumetric ablation into account through adequate source terms. This approach is implemented within a computational tool based on a high-order discontinuous Galerkin discretization. The multi-dimensional tool has already been validated and has proven its efficient parallel implementation. Within this platform, a fully implicit method was developed to simulate multi-phase reacting flows. Numerical results to verify and validate the methodology are considered within this work. Interactions between the flow and the ablated geometry are also presented. 1 Supported 2 Supported by Fund for Research Training in Industry and Agriculture by the European Research Council Starting Grant #259354 12:27PM H3.00010 Propagation of viscous currents on porous substrate with finite entry pressure1 , ROIY SAYAG, Dept. of Environmental Physics (BIDR) and Dept. of Mechanical Engineering, Ben-Gurion University, Israel, JEROME A. NEUFELD, BP Institute, Dept. of Earth Science, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, UK — We study the propagation of viscous gravity currents over a thin porous substrate with finite capillary entry pressure. Near the origin, where the current is deep, propagation of the current coincides with leakage through the substrate. At the nose of the current, where the depth reduces below a critical threshold, drainage is absent. Consequently the flow can be characterised by the evolution of the drainage front and the fluid front at the nose. We analyze this flow using numerical and analytical techniques combined with laboratory-scale experiments. We find that at early times the position of both fronts is proportional to t1/2 , similar to an axisymmetric gravity current without drainage. At later time the growing effect of drainage inhibits spreading. However, as the drainage front approaches a steady position at which the horizontal flux in the current is nonzero the asymptotic propagation of the fluid front approaches a similarity solution ∝ t1/2 , implying a diminishing impact of the draining domain on the propagating nose. 1 BIDR, BGU, Israel and DAMTP, Univ. of Cambridge, UK Monday, November 24, 2014 10:30AM - 12:40PM Session H4 Bubbles: Microbubbles and Nanobubbles — 3006 - Xue Hua Zhang 10:30AM H4.00001 Collapse of Surface Nanobubbles , LONGQUAN CHEN, CHON U CHAN, MANISH ARORA, CLAUSDIETER OHL, School of Physical and Mathematical Science, Nanyang Technological University — Surface nanobubbles are nanoscopic gaseous domains that entrap on immersed solid surfaces in water. They are surprisingly stable and are difficult to be distinguished from polymeric/hydrophobic drops and solid particles (contamination). Here, we report a comparative study of contact line motion across surface nanobubbles, polymeric drops and solid particles. We show that surface nanobubbles spontaneously collapse once contact line touches them while a fast jump process and a pinning process are observed on polymeric drops and on solid particles, respectively. These distinct contact line dynamics provide a new approach to identify surface nanobubbles. The collapse of surface nanobubbles demonstrates their gaseous property and also indicates that they are metastable. The collapse process last few milliseconds with a characteristic speed of 0.1 mm/s, which is much longer and slower than that of hydrodynamic phenomena. We further show that the collapse phenomenon can be explained with a microscopic contact line dynamics. 10:43AM H4.00002 Four-dimensional visualization of rising microbubbles , JUNG HO JE, JI WON JUNG, JAEYEON PYO, Dept. Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Korea, JAE-HONG LIM, Pohang Accelerator Laboratory, Korea — Four-dimensional imaging, which indicates imaging in three spatial dimensions as a function of time, provides useful evidence to investigate the interactions of rising bubbles. However, this has been largely unexplored for microbubbles, mostly due to problems associated with strong light scattering and shallow depth of field in optical imaging. Here, we developed tracking x-ray microtomography that is capable of visualizing rising microbubbles in four dimensions. Bubbles are tracked by moving the in-situ cell to account for their rise velocity. The sizes, shapes, time-dependent positions, and velocities of individual rising microbubbles are clearly identified, despite substantial overlaps between bubbles in the field of view. Our tracking x-ray microtomography affords opportunities for understanding bubble-bubble (or particle) interactions at microscales – important in various fields such as microfluidics, biomechanics, and floatation. 10:56AM H4.00003 On the generation of mini-clusters of microbubbles using water electrolysis , ANA MEDINA-PALOMO, ELENA IGUALADA-VILLODRE, JAVIER RODRIGUEZ-RODRIGUEZ, Carlos III University of Madrid — The interest on microbubbles and their behavior under ultrasound excitation has increased over the last years. Several phenomena can be observed when microbubbles interact with an ultrasound field. For instance, they can oscillate at their natural frequency, translate in the direction of the acoustic pulse (due to the well-known Bjerknes force) or coalesce (due to the secondary Bjerknes force). To study these effects, it is convenient to have an isolated bubble or a cloud consisting of a few bubbles. Using electrolysis we are able to produce mini-clusters of bubbles with controlled parameters, namely, bubble number and size distribution. We achieve this control using voltage pulses of well-defined properties. The most remarkable characteristics of this technique are its low cost and ease of implementation. We illustrate the applications of the technique with some academic examples, like the validation of the expressions for the primary and secondary Bjerknes forces. Funded by the Spanish Ministry of Economy and Competitiveness through grant DPI2011-28356-C03-02. 11:09AM H4.00004 Microfluidic pinball made of quasi-2D microbubbles: on the collective dynamics of confined bubbles pulsating under ultrasound , FLORE MEKKI-BERRADA, PIERRE THIBAULT, PHILIPPE MARMOTTANT, LIPhy — The pulsation properties of air bubbles under ultrasound have received much attention since the development of sonoporation and contrast agents. Spherical bubbles are well known to induce streaming when excited by ultrasound. Here we study how the vibration of very confined bubbles pinned to pits (assuming a quasi-2D “pancake” shape) influences the streaming inside a microfluidic channel. For a single bubble, 20 to 70 µm in radius, we observe the well-known parametric instability, giving rise to a shape deformation, and sketch a phase diagram of existence of the surface modes. We also evidence very active out-of-plane fluid circulations located near the bubble that are correlated with the surface modes. In the case of a bubble pair, the interaction results in an additional bipolar surface mode. We demonstrate that a long-range multipolar recirculating flow occurs from a combination of phase-lagged vibration modes. Using a large triangular lattice of these microbubbles, we obtain a unique acoustic bubble “pinball” driving fluid and particles in complex paths, the constructive interference between vibration modes leading to the elaborate in-plane microstreaming vortices. This work gives a new insight in bubbles efficiency to trigger local and non-local mixing in laminar flows. 11:22AM H4.00005 Lifetime of surface nanodroplets and surface nanobubbles , XUEHUA ZHANG, RMIT, Melbourne, DETLEF LOHSE, University of Twente — Surface nanodroplets are nanoscopic emulsion droplets (e.g. of oil) on (hydrophobic) substrates immersed in water. Correspondingly, surface nanobubbles are nanoscopic gaseous domains on water-immersed substrates. Both can form through local oversaturation of the water with oil or gas, respectively. Such local oversaturation can be achieved through solvent exchange. Here we study the lifetime of such surface nanodroplets and nanobubbles in clean and degassed water, showing how both dissolve over time. We highlight pinning effect which considerably extend the lifetime of both surface nanodroplets and nanobubbles and reveal stick-slip motion of the three phase contact line during the dissolution process. We also discuss collective effects which extend the lifetime too. 11:35AM H4.00006 Experimental microbubble generation by sudden pressure drop and fluidics1 , FERNANDO FRANCO GUTIERREZ, Universidad Michoacana de San Nicolas de Hidalgo, BERNARDO FIGUEROA ESPINOZA, Universi- dad Nacional Autonoma de Mexico, ALICIA AGUILAR CORONA, JESUS VARGAS CORREA, GILDARDO SOLORIO DIAZ, Universidad Michoacana de San Nicolas de Hidalgo — Mass and heat transfer, as well as chemical species in bubbly flow are of importance in environmental and industrial applications. Microbubbles are well suited to these applications due to the large interface contact area and residence time. The objective of this investigation is to build devices to produce microbubbles using two methods: pressure differences and fluidics. Some characteristics, advantages and drawbacks of both methods are briefly discussed, as well as the characterization of the bubbly suspensions in terms of parameters such as the pressure jump and bubble equivalent diameter distribution. 1 The authors acknowledge the support of Consejo Nacional de Ciencia y Tecnologı́a 11:48AM H4.00007 Nucleation of interfacial nanobubbles via solventless exchange , BENG HAU TAN, MANISH ARORA, CLAUS-DIETER OHL, Nanyang Technological University — Interfacial nanobubbles are flat spherical caps of gas that attach on wetted hydrophobic surfaces. Nanobubbles are typically nucleated by wetting an atomically smooth surface with a water-solvent exchange. The bubbles appear when water is flushed into the system, but dissolve in ethanol. Although there is abundant evidence to suggest the bubbles are gaseous (for instance with infrared spectroscopy [1], water-solvent exchanges by themselves cannot rule out the possibility that the structures are organic contaminants rather than gaseous bubbles, e.g. [2]. We report an AFM study of nanobubbles on HOPG using an exchange of saturated water and degassed water. Nanobubbles nucleated by our solventless technique are smaller in radius and height than with the standard solvent exchange technique. The nanobubbles disappear on a second exchange with degassed water. Since the exchange is free of organic solvent, we rule out organic contamination. Moreover, since the exchange affects only the dissolved gas in the liquid, the appearance and disappearance of the bubbles by successive exchange can be conclusively linked to the gas. [1] X H Zhang, A Quinn, W A Ducker, Langmuir (2008), 24, 4756. [2] R P Berkelaar et al, Soft Matter (2014), 10, 4947 12:01PM H4.00008 Stability of gas supersaturation in water: Implication of the existence of bulk nanobubbles? , TATSUYA YAMASHITA, KEITA ANDO, Department of Mechanical Engineering, Keio University — While nanobubbles sitting at solid surfaces are well known to exist for hours or even days, the existence of nanobubbles in the bulk of water is yet under discussion. However, recent molecular dynamics simulations suggest that the thermodynamic stability of bulk nanobubbles can possibly be supported by gas supersaturation in the liquid phase by having bubbles closely populated. Here, we demonstrate the production of supersaturated water by a commercial fine bubble generator. Micron and submicron bubbles are continuously injected in the system of tap water circulation for the water to be gas saturated through bubble dissolution. Dissolved oxygen measurements show gas supersaturation in the water which stays for a couple of days. To further support this observation, we examine the diffusion driven growth of millimeter-sized gas bubbles nucleated at glass surfaces. The growth rate is found to agree with the (extended) theory of Epstein and Plesset, meaning that the water is indeed supersaturated with gases. We speculate that a large number of nanobubbles in the bulk or attached at the surface of floating particles may possibly exist in the supersaturated tap water. 12:14PM H4.00009 Adherent nanoparticles-mediated micro- and nanobubble nucleation , CHON U CHAN, LONG QUAN CHEN, Division of Physics and Applied Physics, School of Physical and Mathematical Science, ALEXANDER LIPPERT, Lam Research AG, Villach, Austria, MANISH ARORA, CLAUS-DIETER OHL, Division of Physics and Applied Physics, School of Physical and Mathematical Science — Surface nanobubbles are commonly nucleated through water-ethanol-water exchange. It is believed that the higher gas solubility in ethanol and exothermic mixing leads to a supersaturation of gas in water. However details of the nucleation dynamic are still unknown. Here we apply the exchange process onto a glass surface deposited with nanoparticles and monitor the dynamics optically at video frame rates. During exchange bubbles of a few micron in diameter nucleate at the sites of nanoparticles. These microbubbles eventually dissolve in ethanol but are stable in water. This agrees with the nucleation process observed for surface nanobubbles. Also we find a reduction of surface attached nanobubbles near the particles, which might be due to gas uptake from the microbubble growth. Finally, high speed recordings reveal stick-slip motion of the triple contact line during the growth process. We will discuss possibilities of utilizing the findings for contamination detection and ultrasonic cleaning. 12:27PM H4.00010 Dynamics of Sub-Micron Bubbles Growing in a Wedge in the Low Capillary Number Regime1 , MICHAEL NORTON, University of Pennsylvania, JEUNG PARK, IBM T. J. Watson Research Center, SUNEEL KODAMBAKA, University of California Los Angeles, FRANCES ROSS, IBM T. J. Watson Research Center, HAIM BAU, University of Pennsylvania — Using a hermeticallysealed liquid cell, we observed the growth and migration of bubbles (tens to hundreds of nanometers in diameter) with a transmission electron microscope. The internal pressure of the liquid caused the thin silicon nitride membranes that comprise the cell’s observation windows to bow outward, creating spatial gradients in the liquid cell’s height. As a result, growing bubbles migrated in the direction of increasing cell height. To better understand the migration dynamics, we developed a simple, two-dimensional model to predict the translational velocity of a bubble that makes contact with both wedge surfaces as a function of the bubble growth rate and wedge opening angle. The model is valid in the asymptotic limit of zero capillary number and relies on a phenomenological relationship between the contact line velocity and the dynamic contact angle. The theoretical predictions are compared with experimental observations. 1 MN was supported, in part, by the Nano/Bio Interface Center through the National Science Foundation NSEC DMR08-32802. HHB and FR were supported, in part, by grants 1129722 and 1066573 from the National Science Foundation. Monday, November 24, 2014 10:30AM - 12:40PM Session H5 Biofluids: Swimming — 3008 - Leif Ristroph, Courant Institute 10:30AM H5.00001 Optimal Swimming with a Burst-and-Coast Behaviour1 , EMRE AKOZ, KEITH MOORED, Lehigh University — Swimming animals are typically assumed to be continuously adding power to the fluid throughout a period of motion. On the other hand, animals have been observed using a non-continuously powered motion described as a burst-and-coast or burst-and-glide behavior. When animals use a non-continuously powered motion it is estimated that their cost of transport is reduced by as much as 45%. However, there are competing mechanisms in the literature that lead to this conclusion. The present study aims to identify the underlying mechanism of burst-and-coast energy savings and to quantify the scaling of optimal motions. A two-dimensional boundary element method approach is used to quantify the performance and wake structure of a free-swimming pitching panel operating with a burst-and-coast behavior. 1 Supported by the Office of Naval Research under Program Director Dr. Bob Brizzolara, MURI grant number N00014-14-1-0533. 10:43AM H5.00002 Stability may not compromise Maneuverability in Aquatic Periodic Locomotion1 , EVA KANSO, FANGXU JING, University of Southern California — Most aquatic vertebrates swim by lateral flapping of their bodies and caudal fins. While much effort has been devoted to understanding the flapping kinematics and its influence on the swimming efficiency, little is known about the stability (or lack of) of periodic swimming. It is believed that stability limits maneuverability and body designs/flapping motions that are adapted for stable swimming are not suitable for high maneuverability and vice versa. Here, we consider an idealized model of a planar elliptic body undergoing prescribed periodic heaving and pitching in a perfect fluid. We show that periodic locomotion depends on several parameters including the aspect ratio of the body and the amplitude and phase of the prescribed flapping. We then study the stability of periodic locomotion using Floquet theory. We find that interesting trends of switching between stable and unstable motions emerge and evolve as we continuously vary the parameter values. This suggests that, when it comes to live organisms, maneuverability and stability need not be thought of as disjoint properties, rather the organism may manipulate its motion in favor of one or the other depending on the task at hand. 1 This work is partially supported by the National Science Foundation through the CAREER award CMMI 06-44925 and the grant CCF 08-11480 10:56AM H5.00003 Stereoscopic Particle Image Velocimetry Used to Study the Wake Patterns of an Ideal Anguilliform Swimming Motion , BRANDON TARAVELLA, J. BAKER POTTS, University of New Orleans, New Orleans, LA, MATTHEW STEGMEIR, TSI Inc. Shoreview, MN — The University of New Orleans recently acquired a self-contained stereoscopic particle image velocimetry system for use in their 125 ft long towing tank. This system is being used to study the wake flow behind an anguilliform swimming robot that swims with an ideal motion that is theorized not to produce any trailing vortices. The presentation will describe the particulars of the SPIV system along with details of installation of the SPIV system within the towing tank. The calibration routine will be discussed in detail and results of the free-flow runs will be discussed. Preliminary results from the anguilliform swimming motion will also be presented. 11:09AM H5.00004 The hydrodynamics and kinematics of sea lion swimming , MEGAN C. LEFTWICH, CHEN FRIEDMAN, George Washington University — A highly interactive, non-research, female sea lion was used for studying its thrust production mechanisms at the Smithsonian National Zoo in Washington, DC. Videography was used for flipper kinematics extraction by tracing the flipper center line and studying the flipper shape throughout the thrust phase. Acceleration from rest was studied with respect to flipper angular rate and flipper shape by digitizing the videos using 10 points spanning root to tip. Resulting functions reveal spanwise camber of up to 32%, with instantaneous angular rates as high as 20 rad/sec, generating thrust values in the range of 150-680 N. The sea lion flipper was scanned using several 3D scanning techniques to generate a 3D model which will be used to reproduce a scaled robotic flipper for testing in a controlled laboratory setting. Techniques included two highly accurate structured light based 3D scanner, an image based software, capable of generating 3D meshes, and a kinect based scanner. A silicone mold of the flipper was also created for reference and comparison. The 3D models are used to extract several section airfoils which aid in both modeling the flipper computationally and designing foreflipper based robotic platforms. 11:22AM H5.00005 A Study of Kinematics Modeling and the Computational Optimization of the Human Underwater Undulatory Kick by Comparison of Swimmers and Body Orientations1 , XIAORAN ZHU, Western Albemarle High School, GENG LIU, YAN REN, HAIBO DONG, University of Virginia, FLOW SIMULATION RESEARCH GROUP TEAM — Underwater Undulatory Swimming (UUS), better known as the underwater dolphin kick, is the most important technique in competitive swimming. Faster than three of the four strokes in swimming, UUS is permitted in the 15m after dives and turns. In this study, we compared the UUS of a college-level swimmer and a younger swimmer. 3D human models were built and reconstructed using stereo-videos for identifying key components of undulatory kick kinematics with respect to strongly flexing joints. A gradient-based optimizer and an immersed boundary method based CFD solver was then used to study the hydrodynamic performance of each swimmer. Optimal settings of current kinematic models will help us to understand the efficiency of the observed undulatory kick mechanisms and further improvements of the human UUS strategy. 1 This work is supported by NSF CEBT-1313217 and UVa HooS-STER program 11:35AM H5.00006 On the efficient swimming of a ray-inspired underwater vehicle Part I: Experimental study on swimming optimization of control and fin structure1 , JIANZHONG ZHU, MERVYN LOPEZ, VENTRESS WILLIAMS, University of Virginia, THEOPHILUS ALUKO, University of Maryland Baltimore County, HAIBO DONG, HILARY BARTSMITH, University of Virginia — Batoid fish such as manta and cownose rays are among the most agile and energy efficient swimming creatures. These capabilities arise from flapping and bending their dorsally flattened pectoral fins. To assess this contribution, this study focuses on the study of a bio-inspired underwater vehicle—the MantaBot—where biological design criteria are applied. The MantaBot consists of two parts: a rigid body rendered from a CT scanning image of a cownose ray and two flexible fins driven by tensegrity actuators. The experiments were conducted in a water tank where the MantaBot was attached to a rail for rectilinear swimming. Three stereo-videos were taken and digitized to measure the 3D kinematics. Results showed that the fins conduct deformations in both spanwise and chordwise directions during steady swimming. Optimal operation conditions were determined for fastest swimming by surveying a wide range of parameters. Contributions of thrust generation and amplitude hindrance of various portions of the fin volume were examined. Additionally, fin tip structure, material and bending properties were studied for optimal swimming. 1 This research was supported by the Office of Naval Research (ONR) under the Multidisciplinary University Research Initiative (MURI) Grant N0001408-1-0642 and Grant N00014-14-1-0533. 11:48AM H5.00007 The effect of flexibility on ribbon-fin-based propulsion , HANLIN LIU, BEVAN TAYLOR, EVAN LATSHAW, OSCAR CURET, Florida Atlantic Univ — Ribbon-fin-based propulsion has the potential to improve the maneuverability of underwater vehicles navigating in complex environments. In this type of propulsion a series of rays are used to send traveling waves along an elongated fin. The use of flexible rays could further enhance the propulsive efficiency of undulating ribbon fins. In this work, we characterize the mechanical behavior and performance of a robotic undulating ribbon fin with different ray flexibilities. We tested the physical model in a water tunnel. In a series of experiments we measure the propulsive force, power consumption and swimming speed of the robotic fin for different ray flexural stiffness, wave frequencies and flow conditions. We found that an increase in flexibility decreases both thrust production and power consumption. Flexible rays could improve or worsen the propulsive performance compared to a rigid counterpart depending on the actuation parameters. We present the result concerning the different performance between rigid and flexible fins. 12:01PM H5.00008 On the efficient swimming of a ray-inspired underwater vehicle. Part II: Computational analysis of fin hydrodynamics1 , GENG LIU, YAN REN, JIANZHOU ZHU, HILARY BART-SMITH, HAIBO DONG, Department of Mechanical and Aerospace Engineering, University of Virginia — High-fidelity numerical simulations are being used to examine the key hydrodynamic features and thrust performance of the fin of a manta ray-inspired underwater vehicle (MantaBot) which is moving at a constant forward velocity. The numerical modeling approach employs a parallelized DNS immersed boundary solver for low-Reynolds number flows past highly deformable bodies such as fish pectoral fins and insect wings. The three-dimensional, time-dependent fin kinematics is obtained via a stereo-videographic technique. The primary objectives of the CFD effort are to quantify the thrust performance of the MantaBot fin with different bending stiffness as well as to establish the mechanisms responsible for thrust production. Simulations show that the bending angle and bending rate of the fin play important roles in thrust producing. A distinct system of connected vortices produced by the deformable fins is also examined in detail for understanding the thrust producing mechanisms. 1 This research was supported by the Office of Naval Research (ONR) under the Multidisciplinary University Research Initiative (MURI) Grant N0001414-1-0533. 12:14PM H5.00009 A fish-like robot : Mechanics of swimming due to constraints , PHANINDRA TALLAPRAGADA, RIJAN MALLA, Clemson University — It is well known that due to reasons of symmetry, a body with one degree of actuation cannot swim in an ideal fluid. However certain velocity constraints arising in fluid-body interactions, such as the Kutta condition classically applied at the trailing cusp of a Joukowski hydrofoil break this symmetry through vortex shedding. Thus Joukowski foils that vary shape periodically can be shown to be able to swim through vortex shedding. In general it can be shown that vortex shedding due to the Kutta condition is equivalent to nonintegrable constraints arising in the mechanics of finite-dimensional mechanical systems. This equivalence allows hydrodynamic problems involving vortex shedding, especially those pertaining to swimming and related phenomena to be framed in the context of geometric mechanics on manifolds. This formal equivalence also allows the design of bio inspired robots that swim not due to shape change but due to internal moving masses and rotors. Such robots lacking articulated joints are easy to design, build and control. We present such a fish-like robot that swims due to the rotation of internal rotors. 12:27PM H5.00010 Fast Computation of Fully Resolved Neuromechanically Simulated Locomotion1 , NAMU PATEL, Department of Engineering Sciences and Applied Mathematics, Northwestern University, NEELESH A. PATANKAR, Department of Mechanical Engineering, Northwestern University — In fish, caudally propagating waves of neural activity produce muscle bending moments. These moments, coupled with forces due to the body’s elastic properties and forces due to fluid-body interactions, determine the deformation kinematics for swimming. Fully resolved simulations of neurally-activated swimming can be used to decode activation patterns underlying observed behaviors in a swimming animal. These computations are expensive; the time stepping requirement is onerous due to the canonically used explicit coupling between the elastic body and the fluid. To overcome this barrier, we use our prior result that deformation kinematics closely follow the preferred kinematics due to muscle activation when a swimmer has a sufficiently stiff body. Thus, we can impose the preferred deformation kinematics directly on the body immersed in the fluid. In this way, the need to solve the elastic equations is eliminated. Here, we couple physiochemical and physiomechanical equations to a constraint-based self-propulsion formulation. With this method, we demonstrate how different behaviors, such as turning, emerge from varying the neural signal. 1 This work is supported by NSF: CBET-0828749, CMMI-0941674, CBET- 1066575, and DGE-0903637 Monday, November 24, 2014 10:30AM - 12:40PM Session H6 Biofluids: Artificial Active Microswimmers — 3010 - John Brady, California Institute of Technology 10:30AM H6.00001 Magnetic Helical Microswimmers in Poiseuille Flow , ALPEREN ACEMOGLU, SERHAT YESILYURT, Sabanci University — We analyze the motion of artificial magnetic microswimmers which mimic the swimming of natural organisms at low Reynolds numbers. Artificial magnetic microswimmers consist of a rigidly connected helical tail and a magnetic head. Magnetic swimmers are actuated with three orthogonal electromagnetic coil pairs. The swimmer motion is examined in the laminar flow which is introduced to channel with syringe pump. We recorded videos for forward (pusher-like swimming / in the head direction) and backward (puller-like swimming / in the tail direction) motion of swimmers. Swimmers have non-stable helical trajectories for forward motion and stable straight trajectories for backward motion. The flow effects on trajectories are observed for swimmers with different geometric parameters in the circular channels. Experiment results show that helical wavelengths of the trajectories are affected with the flow. Additionally, the flow has more pronounced effect on the trajectories of the swimmers in wide channels. Moreover, circular confinement in narrow channels leads to more stable trajectories; in wide channels swimmers follow complex trajectories. A CFD model is used to compare experiments with simulations and to analyze the effects of hydrodynamic interactions. 10:43AM H6.00002 Remote control of self-assembled microswimmers , NICOLAS VANDEWALLE, GALIEN GROSJEAN, ALEXIS DARRAS, GUILLAUME LAGUBEAU, MAXIME HUBERT, GEOFFROY LUMAY, GRASP, Institute Physics B5a, Sart Tilman, University of Liege, B4000 Liege, Belgium — Physics governing the locomotion of microorganisms and other microsystems is dominated by viscous damping. An effective swimming strategy involves the non-reciprocal and periodic deformations of the considered body. Herein, we show that a magnetocapillary-driven self-assembly, composed of three soft-ferromagnetic beads, is able to swim along a liquid-air interface when driven by an external magnetic field. Moreover, the system can be fully controled, opening ways to explore low Reynolds number swimming and to create micromanipulators in various applications. 10:56AM H6.00003 Self-propulsion via natural convection , AREZOO ARDEKANI, University of Notre Dame, MATTHIEU MERCIER1 , MICHAEL ALLSHOUSE2 , THOMAS PEACOCK, Massachusetts Institute of Technology — Natural convection of a fluid due to a heated or cooled boundary has been studied within a myriad of different contexts due to the prevalence of the phenomenon in environmental systems such as glaciers, katabatic winds, or magmatic chambers; and in engineered problems like natural ventilation of buildings, or cooling of electronic components. It has, however, hitherto gone unrecognized that boundary-induced natural convection can propel immersed objects. We experimentally investigate the motion of a wedge-shaped object, immersed within a two-layer fluid system, due to a heated surface. The wedge resides at the interface between the two fluid layers of different density, and its concomitant motion provides the first demonstration of the phenomenon of propulsion via boundary-induced natural convection. Established theoretical and numerical models are used to rationalize the propulsion speed by virtue of balancing the propulsion force against the appropriate drag force. We successfully verified the influence of various fluid and heat parameters on the predicted speed. 1 now 2 now at IMFT (Institut de Mécanique des Fluides de Toulouse). at Center for Nonlinear Dynamics, University of Texas, Austin. 11:09AM H6.00004 Autophoretic self-propulsion of homogeneous particles , SEBASTIEN MICHELIN, LadHyX - Ecole Polytechnique, ERIC LAUGA, GABRIELE DE CANIO, DAMTP - University of Cambridge — Phoretic mechanisms such as diffusiophoresis exploit short-ranged interactions between solute molecules in the fluid and a rigid wall to generate local slip velocities in the presence of solute gradients along the solid boundary. This boundary flow can result in macroscopic fluid motion or phoretic migration of inert particles. These mechanisms have recently received a renewed interest to design self-propelled “autophoretic” systems able to generate the required solute gradients through chemical reaction at their surface. Most existing designs rely on the asymmetric chemical treatment of the particle’s surface to guarantee symmetry-breaking and the generation of a net flow. We show here, however, that chemical asymmetry is not necessary for flow generation and that homogeneous particles with asymmetric geometry may lead to self-propulsion in Stokes flow. Similarly, this principle can be used to manufacture micro-pumps using channel walls with uniform chemical properties. 11:22AM H6.00005 Artificial Rheotaxis , JEREMIE PALACCI, CSMR, NYU, USA, STEFANO SACANNA, Dpt of Chemistry, NYU, USA, ANAIS ABRAMIAN, ENS de Lyon, France, JEREMIE BARRAL, CNS, NYU, USA, KASEY HANSON, ALEXANDER GROSBERG, DAVID PINE, PAUL CHAIKIN, CSMR, NYU, USA — Self-propelled micro-particles are intrinsically out-of-equilibrium. This renders their physics far richer than that of passive colloids while relaxing some thermodynamical constraints and give rise to the emergence of complex phenomena e.g. collective behavior, swarming. . . I will show that we can design microparticles with features usually observed for living microorganisms, the sensing of their environment or rheotaxis, the migration in a shear flow. We quantitatively describe the phenomenon and show that we can use a flow to control and assemble the particles. These self propelled particles realize a step forward in the design of advanced biomimetic systems. 11:35AM H6.00006 Confined Swimming of Bio-Inspired Magnetic Microswimmers in Rectangular Channels , FATMA ZEYNEP TEMEL, Brown University, SERHAT YESILYURT, Sabanci University — Bio-inspired microswimmers have great potential for medical procedures in conduits and vessels inside the body; hence, controlled swimming in confined spaces needs to be well understood. In this study, analysis of swimming modes of a bio-inspired microswimmer in a rectangular channel at low Reynolds number is performed with experimental and computational studies. A left-handed magnetic helical swimmer (MHS), having 0.5 mm diameter and 2 mm length, is used in experiments by utilizing rotating magnetic field actuation obtained by electromagnetic coil pairs. Three motion modes are observed in experiments depending on the rotation frequency: (i) lateral motion under the effect of gravity and surface traction at low frequencies, (ii) lateral motion under the effect of gravity and fluid forces at transition frequencies, and (iii) circular motion under the effect of fluid forces at high frequencies. Translational and angular velocities of the MHS are calculated using CFD simulations to investigate the motion modes. In addition, induced flow fields for different radial positions of the MHS are studied. Results demonstrate the significance of rotation frequency, flow fields and pressure distribution on swimming modes and behaviour of the MHS inside rectangular channels. 11:48AM H6.00007 Geometric optimization of helical tail designs to calibrate swimming velocities of microswimmers , EBRU DEMIR, SERHAT YESILYURT, Sabanci University — Artificial microswimmers present both a solution and a challenge as alternative tools to be used in medical applications, namely, drug delivery and minimally invasive surgeries. Achieving desired amount of controlled displacement of microswimmers at desired velocities plays an important role in determining the success of such applications. In this study, a non-dimensionalised CFD model is utilised to investigate the effects of various geometrical parameters on swimming velocities of microswimmers with helical tails in cylindrical confinements, such as helix wavelength, helical body thickness, and diameter. To this end, a “one wavelength long” helical tail is placed inside a cylindrical channel of the same length with periodic boundary conditions applied to both ends, constituting an infinite helix model. As the channel diameter is kept constant, a parametric study of abovementioned geometric identities is conducted to observe the change in the swimming velocities. Furthermore, effects of helix-channel eccentricity and helix rotation about the longitudinal axis on swimming velocity of a dimensionally optimized helix are investigated to reveal near wall effects. The results are found to be in good agreement with the theoretical models existing in the literature. 12:01PM H6.00008 Interactions between particles in a magnetocapillary self-assembly , GUILLAUME LAGUBEAU, ALEXIS DARRAS, GALIEN GROSJEAN, GEOFFROY LUMAY, MAXIME HUBERT, NICOLAS VANDEWALLE, GRASP, Physics Department, University of Liège, B-4000 Liège, Belgium, GRASP TEAM — When particles are suspended at air-water interfaces in the presence of a vertical magnetic field, dipole-dipole repulsion competes with capillary attraction. This interaction was used recently to control self-assembling particles, as well as to create low Reynolds swimming systems. Although the equilibrium properties of the magnetocapillary interaction is understood, the dynamics was unclear. In the present report, we emphasize the rich behavior of two/three particles driven by this interaction. We propose a model for describing the motion driven by an external field, being the basis for developing swimming strategies and other elaborated collective behaviors along liquid-air interfaces. 12:14PM H6.00009 Nonlocal slender body theory for active and passive particles above a wall , KYLE R. STEFFEN, CHRISTEL HOHENEGGER, Univ of Utah — Active suspensions, such as collections of motile particles or swimming microorganisms, have been the subject of much research over the past decade. A recent model proposed by Saintillan and Shelley (2007, 2012) models the motion of particles in free space using a local slender body theory, where the motile force is due to an imposed shear stress at the particle surface and the dynamics of the slender particle is approximated by relating its velocity to the force along its centerline. Because interactions between suspended particles and a fixed wall are inherently nonlocal, the local drag model is not enough. Motivated by the work of Tornberg et al. (2004, 2006) and Götz (2006), we present a nonlocal slender body theory for a slender body above a stationary planar boundary. We consider both the case of a rigid fiber and of a motile swimmer including an active shear stress. Simulating the resulting dynamics of multiple particles requires the solution of a system of coupled integral equations for the force density. As opposed to the case of a straight fiber in free space, the resulting system is not diagonalizable using Legendre polynomials. We consider direct simulations of a small number of particles. 12:27PM H6.00010 Effective diffusion of confined active Brownian swimmers1 , MARIO SANDOVAL, LEONARDO DAGDUG, Universidad Autonoma Metropolitana — We find theoretically the effect of confinement and thermal fluctuations, on the diffusivity of a spherical active swimmer moving inside a two-dimensional narrow cavity of general shape. The explicit formulas for the effective diffusion coefficient of a swimmer moving inside two particular cavities are presented. We also compare our analytical results with Brownian Dynamics simulations and we obtain excellent agreement. 1 L.D. thanks Consejo Nacional de Ciencia y Tecnolo- gia (CONACyT) Mexico, for partial support by Grant No. 176452. M. S. thanks CONACyT and Programa de Mejoramiento de Profesorado (PROMEP) for partially funding this work under Grant No. 103.5/13/6732. Monday, November 24, 2014 10:30AM - 12:27PM Session H7 Biofluids: Cardiovascular Fluid Mechanics I — 3012 - Alison Marsden, University of California, San Diego 10:30AM H7.00001 Coupling 1D Navier Stokes equation with autoregulation lumped parameter networks for accurate cerebral blood flow modeling , JAIYOUNG RYU, University of California, Berkeley, XIAO HU, University of California, San Francisco, SHAWN C. SHADDEN, University of California, Berkeley — The cerebral circulation is unique in its ability to maintain blood flow to the brain under widely varying physiologic conditions. Incorporating this autoregulatory response is critical to cerebral blood flow modeling, as well as investigations into pathological conditions. We discuss a one-dimensional nonlinear model of blood flow in the cerebral arteries that includes coupling of autoregulatory lumped parameter networks. The model is tested to reproduce a common clinical test to assess autoregulatory function - the carotid artery compression test. The change in the flow velocity at the middle cerebral artery (MCA) during carotid compression and release demonstrated strong agreement with published measurements. The model is then used to investigate vasospasm of the MCA, a common clinical concern following subarachnoid hemorrhage. Vasospasm was modeled by prescribing vessel area reduction in the middle portion of the MCA. Our model showed similar increases in velocity for moderate vasospasms, however, for serious vasospasm (∼ 90% area reduction), the blood flow velocity demonstrated decrease due to blood flow rerouting. This demonstrates a potentially important phenomenon, which otherwise would lead to false-negative decisions on clinical vasospasm if not properly anticipated. 10:43AM H7.00002 Automated tuning for parameter identification in multiscale coronary simulations , JUSTIN TRAN, DANIELE SCHIAVAZZI, ABHAY RAMACHANDRA, ANDREW KAHN, ALISON MARSDEN, Univ of California - San Diego — Computational simulations of coronary flow can provide non-invasively obtained information on hemodynamics and wall mechanics that can aid in treatment planning and improve understanding of disease progression. In this study, patient-specific geometry of the aorta and coronary arteries is constructed from CT scans and combined with finite element flow simulations. Lumped parameter networks are coupled as boundary conditions at the inlet and outlets and calculate global hemodynamic quantities. These tools have potential for clinical impact in identifying optimal geometries for Coronary Artery Bypass Grafts, in determining the risk of re-stenosis in saphenous vein grafts, or for studying other coronary diseases. Despite advances in simulation methods, clinical adoption of these tools is currently hindered by the lack of tools for uncertainty quantification. In current simulations, results are reported as single values without confidence intervals. These simulations also do not account for uncertainties in modeling assumptions, nor the uncertainties in the clinical measurements. This study will take the first step in quantifying these uncertainties. Distributions of the modeling parameters will be inferred through inverse Bayesian estimation and propagated through the model to determine parameter sensitivity and quantify confidence in simulation results. Quantification of these uncertainties is a crucial step towards acceptance of coronary flow simulations in the clinical community. 10:56AM H7.00003 A patient-specific CFD-based study of embolic particle transport for stroke1 , DEBANJAN MUKHERJEE, SHAWN C. SHADDEN, Univ of California - Berkeley — Roughly 1/3 of all strokes are caused by an embolus traveling to a cerebral artery and blocking blood flow in the brain. A detailed understanding of the dynamics of embolic particles within arteries is the basis for this study. Blood flow velocities and emboli trajectories are resolved using a coupled Euler-Lagrange approach. Computer model of the major arteries is extracted from patient image data. Blood is modeled as a Newtonian fluid, discretized using the Finite Volume method, with physiologically appropriate inflow and outflow boundary conditions. The embolus trajectory is modeled using Lagrangian particle equations accounting for embolus interaction with blood as well as vessel wall. Both one and two way fluid-particle coupling are considered, the latter being implemented using momentum sources augmented to the discretized flow equations. The study determines individual embolus path up to arteries supplying the brain, and compares the size-dependent distribution of emboli amongst vessels superior to the aortic-arch, and the role of fully coupled blood-embolus interactions in modifying both trajectory and distribution when compared with one-way coupling. Specifically for intermediate particle sizes the model developed will better characterize the risks for embolic stroke. 1 American Heart Association (AHA) Grant: Embolic Stroke: Anatomic and Physiologic Insights from Image-Based CFD 11:09AM H7.00004 Novel Non-invasive Estimation of Coronary Blood Flow using Contrast Advection in Computed Tomography Angiography , PARASTOU ESLAMI, JUNG-HEE SEO, Johns Hopkins University, AMIRALI RAHSEPAR, RICHARD GEORGE, ALBERT LARDO, Johns Hopkins School of Medicine, RAJAT MITTAL, Johns Hopkins University — Coronary computed tomography angiography (CTA) is a promising tool for assessment of coronary stenosis and plaque burden. Recent studies have shown the presence of axial contrast concentration gradients in obstructed arteries, but the mechanism responsible for this phenomenon is not well understood. We use computational fluid dynamics to study intracoronary contrast dispersion and the correlation of concentration gradients with intracoronary blood flow and stenotic severity. Data from our CFD patient-specific simulations reveals that contrast dispersions are generated by intracoronary advection effects, and therefore, encode the coronary flow velocity. This novel method- Transluminal Attenuation Flow Encoding (TAFE) - is used to estimate the flowrate in phantom studies as well as preclinical experiments. Our results indicate a strong correlation between the values estimated from TAFE and the values measured in these experiments. The flow physics of contrast dispersion associated with TAFE will be discussed. This work is funded by grants from Coulter Foundation and Maryland Innovation Initiative. The authors have pending patents in this technology and RM and ACL have other financial interests associated with TAFE. 11:22AM H7.00005 Numerical study of blood flow and bruits from a realistic arterial stenosis , JAEYONG JEONG, DONGHYUN YOU, POSTECH — The arterial stenosis is a major cause of fatal cardiovascular diseases in developed countries. It is well known that a stenosed artery generates distinct sounds called bruits. Many researchers have been trying to use bruits to diagnose how severely an artery is stenosed without using an invasive method. The previous research revealed that more intensified acoustic fluctuations with higher frequency contents are induced by blood flow for more severely constricted arteries. However, most previous research has been conducted on two-dimensional configurations of artery with a variety of simplifications, which may exclude some of the crucial aspects in real stenosed arteries. In the present study, the generation and propagation of bruits from a realistic stenosed artery is simulated and analyzed in detail using a hydrodynamic/acoustic splitting method, where the flow field in a lumen is predicted by solving the incompressible Navier-Stokes equations using an immersed boundary method, while the acoustic field is predicted by linearized perturbed compressible equations. 11:35AM H7.00006 Direct numerical simulation of a pulsatile flow in a coronary artery1 , JORGE BAILON-CUBA, HEATHER HAYENGA, STEFANO LEONARDI, University of Texas at Dallas — A direct numerical simulation of the blood flow in a coronary artery has been performed. A pulsatile, turbulent flow, inside a branchless, rigid cylindrical artery with non-slip conditions has been considered. The blood is assumed to be a Newtonian fluid. As a fundamental component of the coronary geometry, several cross-sectional shapes of the arterial lumen, as a function of the streamwise coordinate-z, are being included using the immersed boundary method, with a simple transversal wavy wall, as the most simple case. A preliminary set of simulations has being run, with two time varying flow rate functions. Results include flow velocities, pressure gradients and wall shear stress (WSS) distribution, and their comparison with other CFD and experimental results. In particular, WSS is important due to the significant role that it plays in the early formation of coronary artery disease (CAD). It has been found that waviness on the wall increases the instantaneous streamwise velocity, w(y), and its fluctuations, hw′2 i(y), and more drastically the WSS. 1 The numerical simulations were performed on the Extreme Science and Engineering Discovery Environment (XSEDE) under Grant No. CTS070066. 11:48AM H7.00007 A Numerical Multiscale Framework for Modeling Patient-Specific Coronary Artery Bypass Surgeries , ABHAY B. RAMACHANDRA, Department of Mechanical and Aerospace Engineering, UCSD, ANDREW KAHN, Department of Medicine, UCSD, ALISON MARSDEN, Department of Mechanical and Aerospace Engineering, UCSD — Coronary artery bypass graft (CABG) surgery is performed to revascularize diseased coronary arteries, using arterial, venous or synthetic grafts. Vein grafts, used in more than 70% of procedures, have failure rates as high as 50% in less than 10 years. Hemodynamics is known to play a key role in the mechano-biological response of vein grafts, but current non-invasive imaging techniques cannot fully characterize the hemodynamic and biomechanical environment. We numerically compute hemodynamics and wall mechanics in patient-specific 3D CABG geometries using stabilized finite element methods. The 3D patient-specific domain is coupled to a 0D lumped parameter circulatory model and parameters are tuned to match patient-specific blood pressures, stroke volumes, heart rates and heuristic flow-split values. We quantify differences in hemodynamics between arterial and venous grafts and discuss possible correlations to graft failure. Extension to a deformable wall approximation will also be discussed. The quantification of wall mechanics and hemodynamics is a necessary step towards coupling continuum models in solid and fluid mechanics with the cellular and sub-cellular responses of grafts, which in turn, should lead to a more accurate prediction of the long term outcome of CABG surgeries, including predictions of growth and remodeling. 12:01PM H7.00008 Modelling Brain Temperature and Cerebral Cooling Methods1 , STEPHEN BLOWERS, PRASHANT VALLURI, Institute of Materials and Processes, University of Edinburgh, IAN MARSHALL, Neuroimaging Sciences, Centre for Clinical Brain Sciences, University of Edinburgh, PETER ANDREWS, BRIDGET HARRIS, Critical Care Unit, NHS Lothian, Centre for Clinical Brain Sciences, University of Edinburgh, MICHAEL THRIPPLETON, Neuroimaging Sciences, Centre for Clinical Brain Sciences, University of Edinburgh — Direct measurement of cerebral temperature is invasive and impractical meaning treatments for reduction of core brain temperature rely on predictive mathematical models. Current models rely on continuum equations which heavily simplify thermal interactions between blood and tissue. A novel two-phase 3D porous-fluid model is developed to address these limitations. The model solves porous flow equations in 3D along with energy transport equation in both the blood and tissue phases including metabolic generation. By incorporating geometry data extracted from MRI scans, 3D vasculature can be inserted into a porous brain structure to realistically represent blood distribution within the brain. Therefore, thermal transport and convective heat transfer of blood are solved by means of direct numerical simulations. In application, results show that external scalp cooling has a higher impact on both maximum and average core brain temperatures than previously predicted. Additionally, the extent of alternative treatment methods such as pharyngeal cooling and carotid infusion can be investigated using this model. 1 Acknowledgement: EPSRC DTA 12:14PM H7.00009 Determining an Effective Shear Modulus in Tubular Organs for FluidStructure Interaction , ROBERT CHISENA, JAMES BRASSEUR, FRANCESCO COSTANZO, Penn State U, HANS GREGERSEN, JINGBO ZHAO, Aalborg Hospital — Fluid-structure interaction (FSI) is central to the mechanics of fluid-filled tubular organs such as the intestine and esophagus. The motions of fluid chyme are driven by a muscularis wall layer of circular and longitudinal muscle fibers. The coupled motions of the fluid and elastic solid phases result from a local balance between active and passive muscle stress components, fluid pressure, and fluid viscous stresses. Model predictions depend on the passive elastic response of the muscularis layer, which is typically parameterized with an average isotropic elastic modulus (EM), currently measured in vivo and in vitro with estimates for total hoop stress within a distension experiment. We have shown that this approach contains serious error due to the overwhelming influence of incompressibility on the hydrostatic component. We present a new approach in which an effective shear modulus, containing only deviatoric contributions, is measured to overcome this serious error. Using in vitro measurements from pig intestines, we compare our new approach to the current method, showing vastly different predictions. We will also report on our current analysis which aims to determine the influence of residual stress on the EM measurements and comment on it use in FSI simulations. Monday, November 24, 2014 10:30AM - 12:40PM Session H8 Focus Session: Superhydrophobicity and Drag Reduction II — 3001/3003 - Steven Ceccio, University of Michigan 10:30AM H8.00001 Longevity of underwater superhydrophobic surfaces for drag reduction1 , MUCHEN XU, CHANG-JIN “CJ” KIM, Univ of California - Los Angeles — The superhydrophobic (SHPo) surfaces capable of drag reduction are usually metastable under water and undergo wetting transition from dewetted (Cassie-Baxter) to wetted state (Wenzel). On the other hand, the SHPo surfaces capable of staying dewetted indefinitely under water unfortunately provide little drag reduction. In order to develop drag-reducing SHPo surfaces for underwater applications some day, it is critical to understand the wetting transition of SHPo surfaces. However, unlike the case of droplets in air, the wetting transition of SHPo surfaces under water is complicated and not fully understood. Based on our recent report, where ∼ 100 microns-wide trenches maintained the dewetted state indefinitely (measured >50 days), we will explain why the wetting transition occurs much easier in reality than the theoretical predictions. We are also expanding the longevity study from the current static condition to flow conditions including turbulent boundary-layer flows. 1 Supported by the Office of Naval Research (ONR) (Grant No. N000141110503) and National Science Foundation (Grant No. 1336966) 10:43AM H8.00002 Mechanically Robust Superhydrophobic Surfaces for Turbulent Drag Reduction1 , KEVIN GOLOVIN, MATHEW BOBAN, CHARLOTTE XIA, ANISH TUTEJA2 , University of Michigan — Superhydrophobic surfaces (SHS) resist wetting by keeping a thin air layer within their texture. Such surfaces have been shown to reduce skin friction during laminar and transitional flows. However, turbulent boundary layer flows exhibit high shear stresses that damage the fragile microstructure of most SHS, and it is yet unclear to what extent these surfaces can reduce drag. Moreover, the increasing pressure fluctuations and decreasing wall unit length experienced during turbulent flow makes designing mechanically robust SHS with the correct roughness scales a challenge. In this work we evaluate many different SHS in terms of their hydrophobicity, mechanical durability and roughness. Whereas even commercially available SHS lose their superhydrophobic properties after slight mechanical abrasion, our novel coatings survive up to 200x longer. Moreover, we evaluate how the roughness of such surfaces changes with mechanical abrasion, and we design SHS with the correct roughness to display optimal drag reduction in turbulent boundary layer flows. 1 Funding from ONR Investigator 2 Principle 10:56AM H8.00003 Self-limiting electrochemical recovery of dewetted state underwater: visualization and performance1 , RYAN FREEMAN, CHANG-JIN “CJ” KIM, University of California, Los Angeles — While superhydrophobic (SHPo) surfaces have garnered much interest with their potential drag reducing ability, they are only effective with a sustained gas layer. Unfortunately, the gas inevitably depletes from passive surfaces in reality due to defects in microstructures and hydrophobic coatings, fluctuations and drifts in the environment, and other factors, making an active gas recovery mechanism necessary. So far, the only practical solution has been the self-limiting electrochemical recovery, which requires no external control and consumes a minimal amount of energy. Here we present direct visualization of gas recovery by electrolysis of water, highlighting the effect of trench geometry and hydrostatic pressure on recovery. A novel fabrication process is developed to prepare the semi-active SHPo surfaces that are optically clear to enable side view observation of gas restoration. Mindful that electrolysis requires energy, the power being expended to recover fully wetted SHPo surfaces of various sizes is measured and evaluated. The study is being expanded to flows in a water tunnel, demonstrating the sustainability of the gas over a range of flow conditions. 1 Supported by the National Science Foundation (Grant No. 1336966). 11:09AM H8.00004 High resolution velocity measurements within a turbulent boundary layer over super-hydrophobic surface1 , HANGJIAN LING, Johns Hopkins University, SIDDARTH SRINIVASAN, Massachusetts Institute of Technology, JOSEPH KATZ, Johns Hopkins University, GARETH MCKINLEY, Massachusetts Institute of Technology — Using dual view digital holographic microscopy (DHM), high resolution velocity measurements within a turbulent boundary layer (TBL) over a super-hydrophobic surface (SHS) studied its potential application for drag reduction. The 50×152 mm2 (spanwise×streamwise) SHS was created by spray-coating a dispersion of perfluorodecyl polyhedral oligomeric silsesquioxanes (F-POSS) in a poly (methyl methacrylate) binder. A porous base was used for replenishment of entrained air. In water tunnel experiments, the entrainment rate of air from the SHS increased with velocity, but was presumably replenished through the porous wall. Typical reconstructed fields of the 2.6×4.5×2.4 mm3 DHM sample contained more than 34,000 (2µm) particles. Particle tracking and ensemble averaging gave the mean velocity profiles at a resolution that enabled direct calculation of wall shear stress τw from velocity gradients. Over a smooth wall, the sample covered the viscous, buffer and part of the log layers ( ν/uτ =14 & 5 µm at 2 & 6 m/s). The τw on the SHS was reduced by 19% and 47% at 2 and 6 m/s, respectively, clearly proving drag reduction in a TBL. The upward shifted velocity profile may facilitate measurements of slip velocity. 1 Sponsored by ONR. 11:22AM H8.00005 Sustainable Drag Reduction in Turbulent Taylor-Couette Flows using Sprayable Superhydrophobic Surfaces , SIDDARTH SRINIVASAN, JUSTIN KLEINGARTNER, JONATHAN GILBERT, ANDREW MILNE, ROBERT COHEN, GARETH MCKINLEY, Massachusetts Institute of Technology — We demonstrate a reduction in the measured inner wall shear stress in moderately turbulent Taylor-Couette (TC) flows by depositing sprayable superhydrophobic (SH) microstructures on the inner moving rigid surface rotor. The surface morphology and the liquid meniscus are characterized using confocal microscopy from which we determine the initial overall wetted solid fraction. We find that the magnitude of drag reduction on our SH coating in turbulent TC flow becomes progressively larger at higher Reynolds numbers up to a maximum of 22% at Re = 8 × 104 . We show that the mean skin friction coefficient Cf in the presence of the SH coating can be expressed by a modified Prandtl-von Karman type relationship of the form (Cf /2)−1/2 = M ln Re(Cf /2)1/2 + N + (b/∆r)Re(Cf /2)1/2 . From this relationship we extract an effective slip length of b = 19µm which remains constant provided the air-layer is not depleted. Thus, a single value of the slip length b is shown to account for the observed drag reduction over the entire range of Re. Finally, we show that the dimensionless effective slip length b+ = b/δν is the key parameter that governs the drag reduction, and scales as b+ ∼ Re1/2 in the limit of high Reynolds number. 11:35AM H8.00006 Measurements of drag reduction by SLIPS1 , MOHAMED A. SAMAHA2 , JESSICA SHANG, MATTHEW FU, KAREN WANG, HOWARD STONE, ALEXANDER SMITS, MARCUS HULTMARK, Princeton University — Slippery liquid infused porous surfaces (SLIPS) consist of an omniphobic lubricant impregnated into a micro/nanoscale textured substrate. These surfaces have been shown to repel a wide range of liquids. Several techniques to fabricate such surfaces are available in the literature. Here, we report on drag reduction and slip-length measurements using a parallel plate rheometer. Skin-friction measurements of different working fluids are performed on SLIPS with fluorinated boehmite substrates infused with different lubricants. The measurements are refined by considering the evaporation effect of the working fluids. The experiments are performed for different viscosity ratios, N (viscosity of working fluid to that of the lubricant). The effect of the gap height and strain rate on the drag reduction is also investigated. The results show that drag-reduction behavior is influenced by the viscosity ratio and the lubricant-film thickness. The observed drag reduction exists even for very thin film thicknesses. Furthermore, drag reduction is observed for different working fluids even with those having low surface tension such as ethanol. 1 Supported 2 Present under ONR Grants N00014-12-1-0875 and N00014-12-1-0962 (program manager Ki-Han Kim). address: Department of Mechanical Engineering, Rochester Institute of Technology, Dubai campus (RIT Dubai). 11:48AM H8.00007 Liquid infused surfaces in turbulent channel flow1 , MATTHEW FU, HOWARD STONE, ALEXANDER SMITS, IAN JACOBI, MOHAMED SAMAHA, JASON WEXLER, JESSICA SHANG, BRIAN ROSENBERG, LEO HELLSTRÖM, YUYANG FAN, KAREN WANG, KEVIN LEE, MARCUS HULTMARK, Princeton University — A turbulent channel flow facility is used to measure the drag reduction capabilities and dynamic behavior of liquid-infused micro-patterned surfaces. Liquid infused surfaces have been proposed as a robust alternative to traditional air-cushion-based superhydrophobic surfaces. The mobile liquid lubricant creates a surface slip with the outer turbulent shear flow as well as an energetic sink to dampen turbulent fluctuations. Micro-manufactured surfaces can be mounted flush in the channel and exposed to turbulent flows. Two configurations are possible, both capable of producing laminar and turbulent flows. The first configuration allows detailed investigation of the infused liquid layer and the other allows well resolved pressure gradient measurements. Both of the configurations have high aspect ratios 15-45:1. Drag reduction for a variety of liquid-infused surface architectures is quantified by measuring pressure drop in the channel. Flow in the oil film is simultaneously visualized using fluorescent dye. 1 Supported under ONR Grants N00014-12-1-0875 and N00014-12-1-0962 (program manager Ki-Han Kim) 12:01PM H8.00008 Drag on a liquid-infused superhydrophobic cylinder1 , JESSICA SHANG, Princeton University, ALEXANDER SMITS, Princeton University; Monash University, HOWARD STONE, Princeton University — We examine the effect of liquid-infused superhydrophobic surfaces on the separation over a circular cylinder for Reynolds numbers 400 < ReD < 1700. Two superhydrophobic surfaces are compared with a smooth untreated surface. A thin lubricant film (1-20 microns in thickness) is applied to a surface with isotropic nanoscale texture and also to a surface with 50 µm-deep, 65 µm-wide triangular grooves aligned with the flow. The viscosity and thickness of the lubricant are varied. With a superhydrophobic surface, the drag increases by 0 to 5%; greater drag is experienced by the microstructured surface. Drag does not appear to depend on the thickness of the overlying lubricant. In contrast to superhydrophobic surfaces with gas-filled cavities, liquid-infused surfaces produce no change in the Strouhal number. The source of the drag increase is rationalized using the structure of the measured velocity fields near the cylinder. 1 Supported by ONR N00014-12-1-0875 12:14PM H8.00009 Evaluation of Drag Reduction via Superhydrophobic Surfaces and Active Gas Replenishment in a Fully-developed Turbulent Flow , JAMES W. GOSE, KEVIN GOLOVIN, STEVEN L. CECCIO, MARC PERLIN, ANISH TUTEJA, Univ of Michigan - Ann Arbor — The development of superhydrophobic surfaces (SHS) for skin-friction drag reduction in the laminar regime has shown great promise. A team led by the University of Michigan is examining the potential of similar SHS in high-speed naval applications. Specifically, we have developed a recirculating facility to investigate the reduction of drag along robustly engineered SHS in a fully-developed turbulent boundary layer flow. The facility can accommodate both small and large SHS samples in a test section 7 mm (depth) x 100 mm (span) x 1200 mm (length). Coupled with an 11.2 kilowatt pump and a 30:1 contraction, the facility is capable of producing an average flow velocity of 20 m/s, yielding a height based (7 mm) Reynolds number of 140,000. The SHS tested were designed for large-scale application. The present investigation shows skin-friction drag reduction for various sprayable and chemically developed SHS that were applied over a 100 mm (span) x 1100 mm (length) area. The drag measurement methods include pressure drop across the test specimen and PIV measured boundary layers. Additional SHS investigations include the implementation of active gas replenishment, providing an opportunity to replace gas-pockets that would otherwise be disrupted in traditional passive SHS due to high shear stress and turbulent pressure fluctuations. Gas is evenly distributed through a 90 mm (span) x 600 mm (length) sintered porous media with pore sizes of 10 to 100 microns. The impact of the active gas replenishment is being evaluated with and without SHS. 12:27PM H8.00010 Turbulent flow drag reduction on hybrid riblet superhydrophobic surfaces1 , JULIE CROCKETT, RICHARD PERKINS, DANIEL MAYNES, Brigham Young University — We investigate characteristics of turbulent flow in a mini-scale channel where one of the walls is structured with riblets, superhydrophobic microribs, or a hybrid surface that has both structure types present. Individually, large scale riblets, approximately 80 microns tall with 160 micron spacing, provide drag reduction through damping spanwise turbulent motions, and superhydrophobic surfaces, with nearly an order of magnitude smaller features, provide drag reduction through apparent slip at the wall. It is postulated that the combination of the structures will yield a more significant drag reduction than either alone. Experiments were conducted in a rectangular channel with one wall comprised of superhydrophobic features, riblets, or the combination of the two and for channel Reynolds numbers ranging from 4500 to 20000. The velocity profile, turbulent statistics, and shear stress profile are observed using PIV measurements. In addition friction factor and turbulence production are extracted from the PIV data. Modest drag reductions were observed for both the superhydrophobic and riblet surfaces. The combined surfaces showed the greatest drag reduction and turbulence production was significantly reduced for these surfaces. 1 NSF Grant No. 1066356 Monday, November 24, 2014 10:30AM - 12:40PM Session H9 Biofluids: Microswimmers II - Boundary Effects — 3014/3016 - James Bird, Boston University 10:30AM H9.00001 Surfing wavy surfaces: Bacteria-surface interactions in flow , GASTÓN L. MIÑO, VASILY KANTSLER, ROMAN STOCKER, MIT — Complex processes occur when microbes interact with surfaces, from mixture enhancement and motion rectification to biofilm formation. Microbe-surface interactions frequently occur in flowing fluids, and flow has recently been shown to have itself unexpected consequences on the dynamics of motile microbes. Here we report on microfluidic experiments in which the interactions of Escherichia coli bacteria with wavy surfaces was quantified in the presence of fluid flow, a model system for naturally occurring topography of many real surfaces. We quantify surface interactions in terms of incident and scattering angles over a range of flow conditions, and compare results to the observations for a microchannel with straight walls. 10:43AM H9.00002 Bacterial encountering with oil droplet1 , JIAN SHENG, MEHDI MOLAEI, Texas Tech University — Encountering of microorganisms with rising oil droplets in aqueous environments is the first and one of the critical steps in the biodegradation of crude oil. Several factors such as droplet sizes, rising velocity, surfactant, and motility of bacteria are expected to affect the encounter rate. We establish well controlled microfluidic devices by applying layer-by-layer technique that allows us to produce horizontal micro droplets with different sizes. The encounter rates of passive particles, motile and non-motile bacteria with these droplets are measured by high speed microscopy. The effects of mobility and motility of these particles on encounter rates are assessed quantitatively. Meanwhile, we visualize reorientation of the particle due to flow filed around the oil droplet. Results show that the motile bacteria have higher probabilities to interact with an oil droplet compare to the passive particles. Ongoing analyses focus on the effect of shear rates, angular dispersion, curvatures of streamlines, and the swimming velocity of bacteria. The ratios of the encounter area to the entire droplet surface at various flow regimes will also been measured. 1 GoMRI 10:56AM H9.00003 Active oil-water interfaces: buckling and deformation of oil drops by bacteria , GABRIEL JUAREZ, ROMAN STOCKER, MIT — Bacteria are unicellular organisms that seek nutrients and energy for growth, division, and self-propulsion. Bacteria are also natural colloidal particles that attach and self-assemble at liquid-liquid interfaces. Here, we present experimental results on active oil-water interfaces that spontaneously form when bacteria accumulate or grow on the interface. Using phase-contrast and fluorescence microscopy, we simultaneously observed the dynamics of adsorbed Alcanivorax bacteria and the oil-water interface within microfluidic devices. We find that, by growing and dividing, adsorbed bacteria form a jammed monolayer of cells that encapsulates the entire oil drop. As bacteria continue to grow at the interface, the drop buckles and the interface undergoes strong deformations. The bacteria act to stabilize non-equilibrium shapes of the oil-phase such wrinkling and tubulation. In addition to presenting a natural example of a living interface, these findings shape our understanding of microbial degradation of oil and may have important repercussions on engineering interventions for oil bioremediation. 11:09AM H9.00004 Direct measurement of cell concentrations in bubble films prior to rupture , PETER WALLS, JAMES BIRD, Boston Univ — Pathogens or other solid particulates suspended in a liquid can attach to the interface of a bubble as it rises to the surface. When the bubble eventually ruptures at the free surface, these particulates can be ejected into film or jet droplets, such as those linked to the respiratory irritation experienced by shoreline residents during red tide events. Previous studies have demonstrated that the particulate concentration in these aerosols can be significantly higher than in the original liquid. However, the evolution from enriched film to enriched droplets is not entirely understood. Here we develop a physical model for the concentration enrichment by considering the concentration of particulates in the bubble film. In addition, we experimentally measure particulate concentration in the bubble film prior to rupture. The observed concentrations are consistent with the developed model. 11:22AM H9.00005 Microorganism Billiards , COLIN WAHL, JOSEPH LUKASIK, SAVERIO SPAGNOLIE, JEAN-LUC THIFFEAULT, University of Wisconsin-Madison — The presence of boundaries can have many different consequences on the locomotion of microorganisms. Recent experiments and numerical simulations have shown that certain types of microorganisms have a particular interaction with a wall: either through active (flagellar contact with the surface) or passive (hydrodynamic) interactions, the body rotates away from the surface and then departs at a critical angle. We explore the billiard-like motion of such a body as it swims in confined domains. The dynamics of swimming inside a regular polygon is characterized, where stable periodic or unstable chaotic trajectories are determined by the angle of departure. We also explore the dynamics of swimming in an array of obstacles. The results may provide insight on entrapment and sorting of microorganisms and other active particles. 11:35AM H9.00006 Attraction of undulatory swimmers, such as nematodes, to surfaces1 , JINZHOU YUAN, DAVID RAIZEN, HAIM BAU, University of Pennsylvania — Nematodes play a significant role in the ecosystem; agriculture; human, animal, and plant disease; and medical research. The interactions between nematodes and surfaces may play an important role in nematodes’ life cycle and ability to invade a host. We studied the effect of a surface on the dynamics of low-Reynolds number, undulating swimmers such as Caenorhabditis (C.) elegans – both wild type and touch-insensitive. The experiments demonstrated that swimmers located far from a surface selected randomly their direction of motion. In contrast, surface-proximate swimmers rotated towards, collided with, and swam along the surface for considerable time intervals, periodically contacting the surface with their anterior. Likewise, swimmers in a swarm were present at higher concentrations close to the surface. Both resistive force theory-based calculations and symmetry arguments predict that short range hydrodynamic torque, resulting from the interaction between the swimmer-induced flow field and the surface, rotate the swimmer towards the surface. We conclude that the surface attraction and following results from the interplay between short-range hydrodynamic and steric forces and is genotype-independent. 1 The work was supported, in part, by NIH NIA 5R03AG042690-02 and NBIC NSF NSEC DMR08-32802. 11:48AM H9.00007 Motion of motile bacteria near a solid surface under shear flows1 , MEHDI MOLAEI, JIAN SHENG, Texas Tech University — Shear is known to affect microorganism locomotion in several ways that includes rheotaxis, upstream motility, and periodic motion namely Jeffery orbits, which are crucial biological processes in biofilm formation. We investigate the effect of shear flow on the motility of E.coli by employing microfluidic devices, high speed microscopy, and digital holography microscopy. Digital holography enables us to track the bacteria and obtain 3D swimming trajectories; meanwhile, high speed microscopy at different distances from the surfaces of the microfluidics allows us to visualize fast occurring phenomena such as cell reorientation by shear or tumbling events and subsequently to quantify the angular dispersion of active particle suspension. The result shows that Jeffery orbital motion for motile E. coli is diametric different than that for passive bacteria. The results show that shear promotes bacterial re-orientation/tumbling near a solid surface whereas the tumbling is suppressed near a solid surface under quiescent flow condition. Ongoing analyses focus on determining whether this enhancement is the results of Jeffery orbital motion by the flow shear or the hydrodynamic interactions of bacteria with a solid surface. 1 NSF,GoMRI 12:01PM H9.00008 Hydrodynamic entrapment, scattering, and escape of swimming bodies near colloidal particles , SAVERIO SPAGNOLIE, Univ of Wisconsin-Madison, GREGORIO MORENO FLORES, Pontificia Universidad Catolica de Chile, DENIS BARTOLO, Ecole Normale Supérieure de Lyon, ERIC LAUGA, Cambridge University — Microorganisms and other self-propelling bodies in viscous fluids are known to traverse complex trajectories in the presence of boundaries, due to passive hydrodynamic and other physical effects. Motivated by the experimental findings of Takagi et al. on self-propulsion in a field of colloidal particles, we derive the far-field hydrodynamic interaction between model “pusher” and “puller” dipole swimmers and no-slip spherical bodies of varying size. Using the analytical estimates for the swimming trajectories, we predict the critical colloid size or dipole strength for which hydrodynamic entrapment occurs, the scattering dynamics for near-obstacle interactions, and the consequences of Brownian fluctuations. The dynamics include billiard-like motion between colloids, intermittent periods of entrapped/orbiting states near single colloids, and apparently randomized escape behavior. We envision applications of the theory to techniques for sorting microorganisms or other self-propelled swimmers, and to the behavior of motile suspensions in inhomogeneous environments. 12:14PM H9.00009 Swimming near an interface in a viscoelastic fluid , SHAHRZAD YAZDI, Department of Chemical Engineering, The Pennsylvania State University, AREZOO ARDEKANI, School of Mechanical Engineering, Purdue University, ALI BORHAN, Department of Chemical Engineering, The Pennsylvania State University — Given the versatility of their natural habitats, microorganisms often encounter the presence of confining boundaries while moving in polymeric solutions. Some examples include swimming of spermatozoa in the mammalian reproductive tract or bacteria in extracellular polymeric matrices during biofilm formation. It has been shown that both confinement and fluid elasticity can have significant impacts on the locomotion of microswimmers. However, the combined effect of these environmental conditions has not been fully understood yet. In this work, we present a fully resolved solution of a low-Reynolds-number microorganism swimming near an interface in a viscoelastic fluid. The kinematics of locomotion for a squirmer in a viscoelastic fluid is compared to its Newtonian counterpart using a perturbation analysis. The results suggest that extracellular polymers dramatically alter the swimming hydrodynamics, and in general increase the residence time of the microswimmer near a no-slip boundary that can consequently facilitate its adhesion rate. The emergence of a limit cycle can also enhance cell-cell communication in the form of quorum sensing and consequently biofilm formation. 12:27PM H9.00010 Effect of solid boundaries on a motile microorganism in a viscoelastic fluid1 , ALIREZA KARIMI, GAOJIN LI, University of Notre Dame, AREZOO ARDEKANI, Purdue University — Microorganisms swimming in viscoelastic fluids are ubiquitous in nature; this includes biofilms grown on surfaces, Helicobacter pylori colonizing in the mucus layer covering the stomach and spermatozoa swimming through cervical mucus inside the mammalian female reproductive tract. Previous studies have focused on the locomotion of microorganisms in an unbounded viscoelastic fluid. However in many situations, microorganisms interact with solid boundaries and their hydrodynamic interaction is poorly understood. In this work, we numerically study the effect of solid boundaries on the swimming behavior of an archetypal low-Reynolds number swimmer, called “squirmer,” in a viscoelastic fluid. A Giesekus constitutive equation is used to model both viscoelasticity and shear-thinning behavior of the background fluid. We found that the time a neutral squirmer spends in the close proximity of the wall increases with polymer relaxation time and reaches a maximum at Weissenberg number of unity. A pusher is found to be trapped near the wall in a viscoelastic fluid, but the puller is less affected. 1 This publication was made possible, in part, with support from NSF (Grant No. CBET- 1150348-CAREER) and Indiana Clinical and Translational Sciences Institute Collaboration in Biomedical/Translational Research (Grant No. TR000006) from NIH. Monday, November 24, 2014 10:30AM - 12:40PM Session H10 Microscale Flows: Interfaces and Wetting — 3005 - Xin Yong, Binghamton University 10:30AM H10.00001 Reentry to the two-thirds power law for the surfactant-laden Bretherton problem in a slippery tube , DAVID HALPERN, University of Alabama, HSIEN-HUNG WEI, National Cheng Kung University — Recent reports on the clean-interface Bretherton problem show that the well-known two-thirds power law can break down due to wall slip (Liao et al. Phys. Rev. Lett. 111, 136001, 2013; Li et al. J. Fluid Mech. 741, 200-227, 2014). Instead, the film thickness can vary quadratically with the capillary number Ca for Ca below some critical value, corresponding to the situation where slip effects are strong. Here we find that the presence of insoluble surfactant completely changes the above result. Specifically, combined effects of surfactant and wall slip can not only make the strong-slip quadratic law disappear, but also completely suppress the usual Marangoni film thickening along the two-thirds law, making the film behave as if surfactant and wall slip were absent. How to test the above finding experimentally is also discussed. 10:43AM H10.00002 Coupling Molecular Dynamics to Continuum Computational Fluid Dynamics to simulate Superspreading at the macro-scale1 , EDWARD SMITH, PANAGIOTIS THEODORAKIS, ERICH MULLER, RICHARD CRASTER, OMAR MATAR, Imperial College London — Superspreading surfactants are widely researched, due to their fascinating properties and their many potential applications. However, the mechanism behind superspreading is still poorly understood. Karapetsas et al. (JFM, 2011) demonstrated that surfactant absorption at the contact line is of critical importance by a simple constitutive law in a continuum solver. Molecular dynamics (MD) has the ability to elucidate the details of this mechanism, replacing the constitutive law with explicit modelling of the surfactant, fluid and solid interactions at the contact line. However, MD is computationally-expensive and usually limited to nano-scale problems. We couple both continuum and molecular models in a single simulation so that the mechanism at the contact-line can be explicitly simulated at the molecular scale, while the continuum model can be employed throughout the remaining domain. This allows simulations on scales which approach macroscale experiments while maintaining the vital molecular details. Here, the required coupling techniques to implement the proposed solution are discussed: obtaining continuum boundary conditions by averaging molecules, applying constraint force to the molecular region, addition or removal of molecules and software for coupled simulation on HPC. 1 EPSRC Grant number EP/J010502/1 10:56AM H10.00003 Characteristic Structure of Forced Wetting , MENGFEI HE, SIDNEY NAGEL, University of Chicago — As a solid plate is lowered vertically into a tank of liquid, the plate will entrain some of the surrounding air. The contact line between the gas, liquid, and solid will be pushed below the original surface height of the liquid. When the dipping velocity surpasses a critical speed, a transition takes place. At that point the contact becomes elongated and, in the final steady state, form a V-shaped cavity of air surrounded on one side by the solid plate and the other side by the liquid [1]. Using interference imaging, we find that there is a characteristic structure to the thickness of the entrained air layer. Not only is there a thick region of air at the edge of the cavity, but there is also characteristic V-shaped regions at the two top corners. The thick region around the edge is reminiscent of the ridge structures observed in dewetting [2]. The non-uniformity of the air pocket geometry suggests a non-uniformity of the air flow distribution, which further suggests a new instability related to the air pocket dynamics. [1] T D Blake and K J Ruschak, Nature (London) 282, 489(1979) [2] J H Snoeijer, G Delon, M Fermigier and B Andreotti, PRL 96, 174504(2006) 11:09AM H10.00004 Dynamics of a capillary invasion in a closed-end capillary , HOSUB LIM, Sungkyunkwan University, ANUBHAV TRIPATHI, Brown University, JINKEE LEE, Sungkyunkwan University — The position of fluid invasion in an open capillary increases as the square root of time and ceases when the capillary and hydrostatic forces are balanced, when viscous and inertia terms are negligible. Although this fluid invasion into open-end capillaries has been well described, detailed studies of fluid invasion in closed-end capillaries have not been explored thoroughly. Thus, we demonstrated, both theoretically and experimentally, a fluid invasion in closed-end capillaries, where the movement of the meniscus and the invasion velocity are accompanied by adiabatic gas compression inside the capillary. Theoretically, we found the fluid oscillations during invasion at short time scales by solving the one dimensional momentum balance. This oscillatory motion is evaluated in order to determine which physical forces dominate the different conditions, and is further described by a damped driven harmonic oscillator model. However, this oscillating motion is not observed in the experiments. This inconsistency is due to the following; first, a continuous decrease in the radius of the curvature caused by decreasing the invasion velocity and increasing pressure inside the close-ended capillary, and second, the shear stress increase in the short time scale by the plug like velocity profile within the entrance length. dℓ The viscous term of modified momentum equation can be written as K 8µℓ by using the multiplying factor K, which represents the increase of shear stress. r 2 dt c The K is 7.3, 5.1 and 4.8 while capillary aspect ratio χc is 740, 1008 and 1244, respectively. 11:22AM H10.00005 Overflow cascades on liquid-infused surfaces , IAN JACOBI, Technion & Princeton University, JASON WEXLER, HOWARD STONE, Princeton University — The shear-driven dewetting of liquid-infused, micro-patterned surfaces is shown to exhibit a complex cascade of overflow, droplet generation and liquid displacement behaviors. Because liquid-infused surfaces are important in systems as varied as free-surface microfluidic devices and high Reynolds number drag-reducing coatings, understanding the dewetting mechanism is crucial to designing substrates capable of retaining infused liquid or, alternatively, dispensing it in a controlled way. Shear flow experiments on a variety of liquid-infused surface architectures are performed and the interfacial dynamics are characterized at macro- and microscopic scales. Analysis of the different stages of the dewetting cascade is then used to develop substrate design criteria for enhanced liquid control under a variety of shear flow conditions. 11:35AM H10.00006 Acoustic spreading of thin films of water: balancing capillary, viscous, and vibrational mechanisms , OFER MANOR, GENNADY ALTSHULER, Chemical Engineering, Technion - Israel Institute of Technology, Haifa, Israel, SMALL SCALE TRASPORT LABORATORY TEAM — Substrate vibrations at frequencies comparable to HF radio frequencies and in contact with liquid generate flow at submicron length scales that may result in spreading of liquid films. This spreading mechanism is thought as a way of manipulating liquids on microfluidic platforms. In previous studies we used silicon oil as a model liquid; silicon oil spread easily and smoothly as long as the oil and substrate vibrations are in contact. Water films under similar conditions, however, were observed to spread to a minute extent and only under high power levels that further render intense capillary instabilities. In this presentation we use theory and experimental evidence to discuss the physical mechanisms associated with acoustic spreading of water films. We highlight mechanisms associated with acoustic spreading of arbitrary liquids, and we show the various influences of these mechanisms on liquid spreading is encapsulated within one dimensionless number whose value determines whether spreading is to take place. We further elucidate the discrepancy, observed in earlier literature, between the response of oil and water to acoustic excitation and highlight an intermediate parametric region, where precise manipulation of water spreading is achieved by carefully balancing the governing mechanisms. 11:48AM H10.00007 Drag reduction on liquid infused superhydrophobic surfaces1 , JEONG-HYUN KIM, JONATHAN ROTHSTEIN, Univ of Mass - Amherst — The drag reduction on liquid infused superhydrophobic surfaces was measured through a microchannel. The microfluidic device consisted of two halves, a superhydrophobic surface and a microchannel, respectively. The superhydrophobic surface was created from a silicon wafer with ridge patterns 30 to 60 microns in width and spacing generated by a standard photolithography. A low viscosity, immiscible, incompressible silicone oil was filled to the gaps of the superhydrophobic surfaces. Several microchannels varying in size from 100 to 200 microns were fabricated from PDMS with an inlet, outlet and two pressure ports. After flow coating the superhydrophobic surface with a uniform film of oil, the two halves were aligned and clamped together and the pressure drop measured. A systematic study on drag reduction and slip length was performed by varying the viscosity ratio between the water and oil phase between 0 to 50. Several aqueous glycerin solutions with different viscosity were prepared. The slip length, pressure drop, and longevity of the oil phase were studied as a function of surface geometry, capillary number and the dispense volume. 1 NSF CBET-1334962 12:01PM H10.00008 A Microfluidic Platform for Interfacial Electrophoretic Deposition , YOUNG SOO JOUNG, JEFFREY MORAN, ANDREW JONES, Massachusetts Institute of Technology, ERIC BAILEY, University of Maryland, CULLEN BUIE, Massachusetts Institute of Technology — Composite membranes of hydrogel and carbon nanotubes (CNTs) are fabricated using electrophoretic deposition (EPD) at the interface of two immiscible liquids in microfluidic channels. Microfluidic channels, which have two parallel electrodes at the walls, are used to create electric fields across the interface of oil and water continuously supplied into the channels. Depending on the Reynolds (Re) and Weber (We) numbers of oil and water, we observe different formations of the interface. Once we find the optimal Re and We to create a planar interface in the channel, we apply an electric field across the interface for EPD of CNTs and/or silver (Ag) nanorods dispersed in water. During EPD, particles migrate to the oil/water interface, where cross-linking of polymers is induced to form composite hydrogel membranes. Properties of the composite hydrogel films are controlled by electric fields, CNT concentrations, and both Re and We numbers, allowing for continuous production. This fabrication method is effective to create composite polymer membranes placed in microfluidic devices with tunable electrical, mechanical, and biological properties. Potential applications include fabrication of doped hydrogels for drug delivery, conductive hydrogels for biological sensing, and electron permeable membranes for water splitting and osmotic power generation. 12:14PM H10.00009 Influence of spatial variation of phenomenological parameters on the modeling of boundary conditions for flows with dynamic wetting , YUKA HIZUMI, TAKESHI OMORI, YASUTAKA YAMAGUCHI, TAKEO KAJISIMA, Department of Mechanical Engineering, Osaka University — For reliable prediction of multiphase flows in micro- and nanoscales, continuum models are expected to account for small scale physics near the contact line (CL) region. Some existing works (for example the series of papers by the group of Qian and Ren) have been successful in deriving continuum models and corresponding boundary conditions which reproduce well the molecular dynamics (MD) simulation results. Their studies, however, did not fully address the issue of adsorption layer especially in the CL region, and it is still not clear if general conclusion can be deduced from their results. In the present study we investigate in detail the local viscosity and the corresponding stress tensor formulation in the solid-liquid interface and in the CL region of immiscible two-phase Couette flows by means of MD simulation. The application limit of the generalized Navier boundary condition and the continuum model with uniform viscosity is addressed by systematic coarse-graining of sampling bins. 12:27PM H10.00010 Molecular-like hierarchical self-assembly of monolayers of mixtures of particles1 , P. SINGH, M. HOSSIAN, S. GURUPATHAM, K. SHAH, A. AMAH, M. JANJUA, S. NUDURUPATI, I. FISCHER, NJIT, N. AUBRY, Northeastern University — We present a technique that uses an externally applied electric field to self-assemble monolayers of mixtures of particles into molecular-like hierarchical arrangements on fluid-liquid interfaces. The arrangements consist of composite particles (analogous to molecules) which are arranged in a pattern. The structure of a composite particle depends on factors such as the relative sizes of the particles and their polarizabilities, and the electric field intensity. If the particles sizes differ by a factor of two or more, the composite particle has a larger particle at its core and several smaller particles forming a ring around it. The number of particles in the ring and the spacing between the composite particles depends on their polarizabilities and the electric field intensity. Approximately same sized particles form chains (analogous to polymeric molecules) in which positively and negatively polarized particles alternate. 1 The work was supported by National Science Foundation Monday, November 24, 2014 10:30AM - 12:40PM Session H11 Microscale Flows: Locomotion — 3007 - David Saintillian, University of California at San Diego 10:30AM H11.00001 Investigation of fluid flow and pumping due to a bacterial flagellum in its various polymorphic forms rotating above a no-slip boundary , JAMES MARTINDALE, HENRY FU, University of Nevada - Reno — Recently, the fabrication of magnetically actuated rotating bacterial flagella attached to a planar substrate has been achieved. An array of such flagella may have applications as a microscale pump. In order to understand pumping, velocity, and other flow properties of anchored bacterial flagella rotating above a no-slip plane in the Stokes flow regime, a model of a single flagellar filament under constant torque near a no-slip boundary is considered for various polymorphic forms. The method of regularized Stokeslets, whose code is verified by examining a classical problem, is employed to create a benchmark case which can in turn be used to justify a slender body theory approach that drastically decreases computational cost. We investigate the flow for all 11 polymorphic forms of bacterial flagella, as well as the effect of tilt on several flow metrics for the various polymorphic forms. 10:43AM H11.00002 Enhanced Monopropellant Fuel Decomposition by High Aspect Ratio, Catalytic CNT Structures for Propulsion of Small Scale Underwater Vehicles , KEVIN MARR, Department of Mechanical Engineering, Brigham Young University, Provo, UT, JONATHAN CLAUSSEN, Department of Mechanical Engineering, Iowa State University, Ames, IA, BRIAN IVERSON, Department of Mechanical Engineering, Brigham Young University, Provo, UT — Both maneuverability and efficiency for reagentbased propulsion systems of small-scale exploratory devices, such as autonomous underwater vehicles (AUVs), is largely dependent on their maximum fuel decomposition rate. Reagent-based systems, however, require large catalyst surface area to fuel volume ratios in order to achieve the fuel decomposition rates necessary for locomotion. This work demonstrates the utility of platinum-coated, carbon nanotube (CNT) scaffolds as high surface area catalysts for decomposition of hydrogen peroxide (H2 O2 ) in a flowing environment. Usage of these functionalized microchannels ensures that both the maximum distance between fuel and catalyst is only half the microchannel diameter, and that the fuel concentration gradient increases due to boundary-layer thinning. These conditions facilitate intimate contact between fuel and catalyst and, therefore, faster decomposition rates. Electrochemical testing revealed that electroactive surface area to volume ratios of approximately 61.4 cm−1 can be achieved for samples fabricated using a static Pt deposition scheme. Thrust measurements were taken using a small-scale submersible which indicated a maximum thrust of 0.114 N using 50 weight percent H2 O2 exposed to eight inline 2.867 cm2 Pt-CNT scaffolds. 10:56AM H11.00003 Self-sustained motion of microcapsules on a substrate controlled via the repressilator regulatory network , HENRY SHUM, VICTOR YASHIN, ANNA BALAZS, University of Pittsburgh — We design microcapsules that undergo self-induced motion in a fluid along a substrate and are able to collectively self-organize when controlled by a biomimetic signaling network. Three microcapsules act as localized sources of distinct chemicals that diffuse through the fluid. The production rate of each chemical is modulated by a regulatory network known as the repressilator: each species represses the production of the next in a cycle. We show that this system can exhibit sustained oscillations. We then allow the diffusing species to adsorb onto the substrate, altering the surface interaction energy. Gradients in surface energy lead to motion of the microcapsules. We find that regulation via the repressilator gives rise to qualitatively different outcomes. Chemical oscillations can facilitate aggregation of the microcapsules and the aggregate can undergo sustained translational or oscillatory motion. Numerical simulation of the fluid flow, microcapsule dynamics and concentration fields is achieved by a combination of the lattice Boltzmann, immersed boundary and finite difference methods. We assess the role of hydrodynamic interactions by comparison with a simplified model that assumes a constant drag coefficient relating the force on a microcapsule to its velocity. 11:09AM H11.00004 Collective dynamics and mixing in a suspension of micro-rotors , ENKELEIDA LUSHI, Brown University, KYONGMIN YEO, IBM Research, PETIA VLAHOVSKA, Brown University — We investigate theoretically and computationally the dynamics of many interacting micro-rotors suspended in fluid. As a particle rotates due to intrinsic or external torques, it disturbs the surrounding fluid and the motion of neighbouring particles. It can be shown that the motion of less than four point rotors is periodic and above that number their trajectories can become chaotic, a dynamics reminiscent to that of 2D point-vortices. If the full hydrodynamical interactions and lubrication effects between the particles are accounted for in a finite domain, a richer dynamics emerges. We exploit this coupled dynamics between micro-rotors and the structure of the generated fluid flows to mix a passive dye field or passive sphere particles also immersed in the fluid. The efficiency of the mixing for a variety of parameters will be discussed as well as experimental realizations. 11:22AM H11.00005 Rotation axes, bistability, and controllability of rigid achiral magnetically rotated microswimmers , FARSHAD MESHKATI, HENRY FU, University of Nevada, Reno — We investigate magnetically actuated microswimmers through analytical and numerical schemes which are applicable to arbitrary rigid geometries. We examine the dynamics of a simple nonhelical, achiral, rigid swimmer composed of three connected colloidal beads. We consider magnetic fields that can rotate either perpendicular to its rotation axis, or at a nonperpendicular angle to its rotation axis. We find the steady rotating orbits of the swimmer and evaluate them for stability. We show that certain experimental conditions, determined by magnetic field strength, rotation frequency, and angle of field relative to rotation axis, can result in more than one stable orbit. We compare this to experimental observations of bistability of helical swimmers. We scrutinize the dependence of the rotation axis of the swimmer on experimental parameters and compare it to the experimental observations of wobbling in the literature. Finally, we show that the controllability of these types of swimmers can be improved by manipulating the angle between the direction of the magnetic field and its axis of rotation. 11:35AM H11.00006 Design of helical magnetically rotated microswimmers for controllability , HENRY FU, University of Nevada, Reno — Microswimmers or microrobots have recently received much attention due to their possible applications in microscale sensing and actuation, including many biomedical applications such as drug delivery, in vivo diagnostics, and tissue manipulation. We have developed a modeling framework to describe the dynamics of rigid microswimmers that can be propelled through bulk fluid (rather than only near surfaces) when rotated by an external magnetic field. Here, this modeling framework is used to identify stable steady rotating orbits of the helical microswimmers under development by many research groups. I investigate how the swimming properties depend on the magnetization direction and geometry (pitch and radius) of the helix. In general, these swimmers have nonlinear dependence of velocity on frequency due to changes in the rotation axis of the swimmer as frequency is changed. However, a linear dependence would enhance velocity control and precise positioning of these swimmers. I identify magnetization directions which keep the rotation axis constant as a function of frequency, hence lead to linear velocity-frequency dependence. I also identify helical geometries which lead to maximal swimming velocities and rotation axes closest to the helical axis of the swimmer. 11:48AM H11.00007 Geometrical Performance of Electrocatalytic Nanomotors1 , AMIR NOURHANI, PAUL E. LAMMERT, VINCENT H. CRESPI, Phys. Dept., Penn State, ALI BORHAN, Chem. Eng. Dept., Penn State — We provide a general analytical expression for the speed of electrocatalytic nanomotors in terms of surface cation flux, interfacial potential and physical properties of motor environment in the linear regime and thin electric diffuse layer. We model the motor geometry by a prolate spheroid which covers a range of geometries from sphere to rod-shape and slender bodies. We obtain a functional that turns the surface cation flux distribution into a motive utility factor. For a spherical motor the kernel of the functional reduces to the first Legendre polynomial of the first kind and with increase in the aspect ratio of the motor, the kernel tends to give more significance to the cation flux near the ends of the motor and the motor velocity becomes less sensitive to the flux distribution around the equator of the spheroid. 1 This work was supported by the NSF under Grant No. DMR-0820404 through the Penn State Center for Nanoscale Science. 12:01PM H11.00008 Designing a bio-inspired self-propelling hydrogel micro-swimmer1 , SVETOSLAV NIKOLOV, PETER YEH, ALEXANDER ALEXEEV, Georgia Institute of Technology — Artificial micro-swimmers have found numerous applications in microfluidics, drug delivery systems, and nanotechnology. In our current research we use dissipative particle dynamics to design and optimize a self-propelling hydrogel micro-swimmer with an X-shaped flat geometry and bi-layered hydrogel structure. The two polymeric layers that bind to each other have identical material properties but distinctive chemical responses to external stimuli. In the presence of outside stimuli one of the layers swells where the other remains passive resulting in hydrogel bending. Our simulations demonstrate that under periodic applications of an external stimulus this actuation routine is capable of creating time-irreversible motion in a low Reynolds number environment. Initially, when the external stimulus is introduced a forward stroke is initiated, as the swimmer first expands and then bends. When the outside stimulus is removed the forward stroke is terminated and a backward stroke begins, as the swimmer contracts and then straightens. Propulsion results due to the difference in momentum exchange between the forward and backward strokes. We use our simulations to probe how alterations in the material properties of the bi-layered hydrogel can affect swimming performance. 1 Support from NSF CAREER Award (DMR-1255288) is gratefully acknowledged. 12:14PM H11.00009 Phase behavior of monolayer suspensions of counter rotating rotors , KYONGMIN YEO, IBM Research, ENKELEIDA LUSHI, PETIA VLAHOVSKA, Brown University — The dynamics of monolayer suspensions of counter-rotating spherical rotors is investigated by using the force coupling method. The motions of the suspended rotors are confined to the horizontal plane perpendicular to the axis of rotation. The suspensions are equally divided by two species of spherical particles, which are rotating under equal-magnitude opposite-sign torques. Unlike the previous results in non-hydrodynamic limit, it is shown that the conversion rate of the rotational kinetic energy to the translational kinetic energy increases slowly with the increase in volume fraction (φ) and eventually exhibits a sharp drop around a critical volume fraction (φ ≃ 0.54). A closer investigation of suspension microstructure reveals that the rotors of the same torque start to form a cluster for φ ≥ 0.30. Around the critical volume fraction, hexagonal structures emerge in the suspensions and the particle mobility is significantly hindered by the caging effects. 12:27PM H11.00010 Acoustophoretic particle motion in a square glass capillary1 , RUNE BARNKOB, ALVARO MARIN, MASSIMILIANO ROSSI, CHRISTIAN J. KÄHLER, Bundeswehr University Munich — Acoustofluidics applications often use complex resonator geometries and complex acoustic actuation, which complicates the prediction of the acoustic resonances and the induced forces from the acoustic radiation and the acoustic streaming. Recently, it was shown that simultaneous actuation of two perpendicular half-wave resonances in a square channel can lead to acoustic streaming that will spiral small particles towards the pressure nodal center (Antfolk, Anal. Chem. 84, 2012). This we investigate in details experimentally by examining a square glass capillary with a 400-µm microchannel acoustically actuated around its 2-MHz half-wave transverse resonance. The acoustic actuation leads to the formation of a half-wave resonance in both the vertical and horizontal direction of the microchannel. Due to viscous and dissipative losses both resonances have finite widths, but are shifted in frequency due to asymmetric actuation and fabrication tolerances making the channel not perfectly square. We determine the resonance widths and shift by measuring the 3D3C trajectories of large particles whose motion is fully dominated by acoustic radiation forces, while the induced acoustic streaming is determined by measuring smaller particles weakly influenced by the acoustic radiation force. 1 DFG KA 1808/16-1 Monday, November 24, 2014 10:30AM - 12:40PM Session H12 Drops: Splashing, Stability and Breakup II — 3018 - Jose Manuel Gordillo, Universidad de Sevilla 10:30AM H12.00001 Fragmentation dynamics in the droplet bag breakup regime , VARUN KULKARNI, PAUL SOJKA, Purdue University — The closing stages of a droplet bag breakup event is marked by the appearance of several topological changes in the drop shape, followed by its fragmentation owing to hydrodynamics instabilities. In the present work we examine this breakup event, which occurs when a drop enters a continuous jet air stream. The deformed drop before eventual fragmentation is comprised of two main features: a bag and a bounding rim. Our investigation discusses the mechanism of rim/ bag breakup and the ensuing drop size distribution. The role of two possible instabilities, Plateau−Rayleigh and Rayleigh−Taylor, in rim breakup is examined and the dominant role of the Plateau−Rayleigh instability is revealed. In contrast, the Rayleigh−Taylor instability is seen to explain the disintegration of the bag well. The effects of viscosity and air jet velocity are also investigated. The formation of secondary features, such as nodes on the rim and holes on the bag, are also discussed. To conclude, a simple scaling argument based on the characteristic time scales of these instabilities is presented to explain the commonly observed early bursting of the bag, vis-à-vis the rim. 10:43AM H12.00002 Faraday instability of a spherical drop , A. EBO ADOU, PMMH-CNRS-ESPCI, UPMC, France, LAURETTE TUCKERMAN, PMMH-CNRS-ESPCI, France, SEUNGWON SHIN, Hongik Univ., Seoul, Korea, JALEL CHERGUI, DAMIR JURIC, LIMSI-CNRS, France — A liquid drop subjected to an oscillatory radial force comprises a spherical version of the Faraday instability, with a subharmonic response which is half of the forcing frequency. The time-dependent shape of the drop and the velocity field in and around it are calculated using BLUE, a code based on a hybrid Front Tracking/Level-set algorithm for Lagrangian tracking of arbitrarily deformable phase interfaces. We compare this shape with the spherical harmonic selected at onset, calculated by adapting the Floquet stability analysis of Kumar and Tuckerman to a spherical geometry. We interpret the shape in light of theoretical results by Busse, Matthews and others concerning pattern formation in the presence of O(3) symmetry. 10:56AM H12.00003 Drop deformation and breakup in a partially filled horizontal rotating cylinder1 , ANDREW WHITE, CAROLINE PEREIRA, HYAQUINO HYACINTHE, THOMAS WARD, Iowa State University — Drop deformation and breakup due to shear flow has been studied extensively in Couette devices as well as in gravity-driven flows. In these cases shear is generated either by the moving wall or the drop’s motion. For such flows the drop shape remains unperturbed at low capillary number (Ca), deforms at moderate Ca, and can experience breakup as Ca → 1 and larger. Here single drops of NaOH(aq) will be placed in a horizontal cylindrical rotating tank partially filled with vegetable oil resulting in 10−2 < Ca < 101 . It will be shown that the reactive vegetable oil-NaOH(aq) system, where surfactants are produced in situ by saponification, can yield lower minimum surface tensions and faster adsorption than non-reactive surfactant systems. Oil films between the wall and drop as well as drop shape will be observed as rotation rates and NaOH(aq) concentration are varied. Results will be presented in the context of previous work on bubble and drop shapes and breakup. 1 NSF CBET #1262718 11:09AM H12.00004 Moving and deforming a liquid drop by pulsed laser irradiation , ALEXANDER L. KLEIN, CLAAS WILLEM VISSER, Physics of Fluids, Faculty of Science & Technology, University of Twente, The Netherlands, HENRI LHUISSIER, Matiere et Systemes Complexes, Université Paris Diderot, France, EMMANUEL VILLERMAUX, Aix-Marseille Université, IRPHE, France, CHAO SUN, DETLEF LOHSE, HANNEKE GELDERBLOM, Physics of Fluids, Faculty of Science & Technology, University of Twente, The Netherlands — The impact of a focused laser pulse onto a liquid drop can be so violent that the drop strongly deforms and eventually explodes. We studied the drop dynamics that results from this laser impact experimentally, in order to understand the time evolution of the drop and find the underlying driving mechanism. The high reproducibility of the dynamics allowed us to use stroboscopic illumination with short, ns exposure times. Combining this technique with high-speed imaging we captured key details of the laser impact and drop deformation. The laser impact ablates the front the drop while the remainder of the drop acquires a velocity of several m/s. The drop expands radially into a disk-like shape with a velocity of the same order of magnitude, before instabilities develop and the drop fragments. A parameter study of the time-resolved drop shape and velocity as a function of the laser energy is presented. 11:22AM H12.00005 How a laser pulse deforms a liquid drop , HANNEKE GELDERBLOM, WILCO BOUWHUIS, ALEXANDER L. KLEIN, DETLEF LOHSE, Physics of Fluids, Faculty of Science & Technology, University of Twente, The Netherlands, EMMANUEL VILLERMAUX, Aix-Marseille Université, IRPHE, France, HENRI LHUISSIER, Matiere et Systemes Complexes, Université Paris Diderot, France, JACCO H. SNOEIJER, Physics of Fluids, Faculty of Science & Technology, University of Twente, The Netherlands — When a liquid drop is hit by a ns laser pulse it experiences a strong pressure kick. As a consequence, the drop is propelled forward and deforms into a thin sheet that eventually becomes unstable and fragments. We aim to understand how the drop motion, deformation and fragmentation depend on the laser-pulse properties and drop characteristics. On the time scale of the laser pulse, where the drop dynamics is purely inertial, an analytical expression for the internal velocity field is obtained. The output of this inertial model is then used as input for a later-stage model that describes the surface-tension limited expansion of the liquid sheet. In the intermediate regime, where the drop is not a sheet yet, its shape evolution is investigated with a boundary integral method. The drop deformation dynamics described by these models is the starting point to study the subsequent drop fragmentation. 11:35AM H12.00006 Yield stress fluid droplet impact on coated horizontal surfaces , RANDY EWOLDT, BRENDAN BLACKWELL, MARC DEETJEN, Univ of Illinois - Urbana — Yield stress fluids, including gels and pastes, are effectively fluid at high stress and solid at low stress. Droplet impacts on a solid surface can create localized lumps and craters, or extended splash events featuring long lifetime ejection sheets. Here we experimentally study liquid-solid impact of yield stress fluids on pre-coated horizontal surfaces. Under critical splash conditions sheet breakup occurs, and ejected droplets can be nonspherical and threadlike due to the inability of capillary stresses to deform material above a certain lengthscale. The presence of a yield stress also allows complex contours forming on the surface to be stable at long times. Droplet size, impact velocity, surface coating thickness, and rheological material properties are varied. We identify regime maps of the stick/splash transition and quantify behavior with measures such as crater diameter, deposition thickness, impact event timescale, and radial extent of material deposition. The results are characterized as a function of appropriate dimensionless parameters in a manner that supports rheological fluid design for specific applications. 11:48AM H12.00007 Multi-Scale Simulation of Atomization with small drops represented by Lagrangian Point-Particle Model1 , YUE LING, STÉPHANE ZALESKI, Université Pierre et Marie Curie, INSTITUT JEAN LE ROND D’ALEMBERT TEAM — Numerical simulation is conducted to investigate the drop formation and evolution in gas-assisted atomization. The atomizer consists of two parallel planar jets: the fast gas jet and the slow liquid jet. Due to the shear between gas and liquid streams, the liquid-gas interface is unstable, and this eventually leads to full atomization. A fundamental challenge in atomization simulations is the existence of multiple length scales involved. In order to accurately capture both the gas-liquid interface instability and the drop dynamics, a multi-scale multiphase flow simulation strategy is proposed. In the present model, the gas-liquid interface is resolved by the Volume-of-Fluid (VOF) method, while the small drops are represented by Lagrangian point-particle (LPP) models. Particular attention is paid on validating the coupling and conversion between LPP and VOF. The present model is validated by comparing with direct numerical simulation (DNS) results and also experimental data. The simulation results show complex coupling between the interface instability and the turbulent gas jet, which in turn influence the formation and evolution of the drops formed in atomization. 1 ANR-11-MONU-0011 12:01PM H12.00008 Investigation on Liquid Atomization Mechanism in Sparkling Fireworks1 , CHIHIRO INOUE, The University of Tokyo — The physics behind the beauty of sparkling fireworks is a mystery over 300 years. There are two types of liquid atomization phenomena; the ejection of streaks of light from a mother fireball, and the spreading streaks burst downstream to produce pine needle-like streaks of light. In the present study, the mechanism of the atomization process in sparkling fireworks is investigated by using a high-speed video camera. It is clarified that bursting bubbles on the mother fireball is essential for the ejection of droplets, which will be streaks of light. The secondary bursting of the light streaks is due to the sudden expansion and catastrophic bursting of spreading droplets. The results of temperature variance of spreading droplets and those of the TG-DTA-MS are also discussed. 1 This study was supported by ILASS-Japan. 12:14PM H12.00009 Instability of a large Leidenfrost drop under confinement , PASCAL RAUX, GUILLAUME DUPEUX, CHRISTOPHE CLANET, DAVID QUERE, Ladhyx, Ecole Polytechnique, Palaiseau, France and PMMH, ESPCI, 10 rue Vauquelin, Paris, France — A Leidenfrost drop confined between two hot plates is unstable, when large enough. After a short delay necessary to build a vapor pocket at its center, it forms a ring which rapidly expands and eventually bursts. We analyze this instability, and show that the ring size increases in a non-linear manner, as a function of time, in agreement with the experiments. 12:27PM H12.00010 Break-up of droplets in a concentrated emulsion flowing through a narrow constriction , MINKYU KIM, LIAT ROSENFELD, SINDY TANG, Stanford Univ, TANG LAB TEAM — Droplet microfluidics has enabled a wide range of high throughput screening applications. Compared with other technologies such as robotic screening technology, droplet microfluidics has 1000 times higher throughput, which makes the technology one of the most promising platforms for the ultrahigh throughput screening applications. Few studies have considered the throughput of the droplet interrogation process, however. In this research, we show that the probability of break-up increases with increasing flow rate, entrance angle to the constriction, and size of the drops. Since single drops do not break at the highest flow rate used in the system, break-ups occur primarily from the interactions between highly packed droplets close to each other. Moreover, the probabilistic nature of the break-up process arises from the stochastic variations in the packing configuration. Our results can be used to calculate the maximum throughput of the serial interrogation process. For 40 pL-drops, the highest throughput with less than 1% droplet break-up was measured to be approximately 7,000 drops per second. In addition, the results are useful for understanding the behavior of concentrated emulsions in applications such as mobility control in enhanced oil recovery. Monday, November 24, 2014 10:30AM - 12:40PM Session H13 Drops: Electric Fields — 3020 - William Ristenpart, University of California, Davis 10:30AM H13.00001 Leidenfrost droplets in an electric field , SANDER WILDEMAN, CHAO SUN, DETLEF LOHSE, University of Twente — In a recent video broadcast dubbed the “Knitting Needle Experiment,” astronaut Don Petit aboard the ISS demonstrated how weightless water droplets can be made to orbit a statically charged Teflon rod. We study the earthly analogue of mobile droplets in an electric field, whereby the mobility is ensured by a thin vapor film sustained between the droplet and a hot plate (the Leidenfrost effect). We find that in a strong vertical electric field the droplet starts to bounce progressively higher, defying gravitational attraction. From its trajectory we can deduce the temporal evolution of the charge on the droplet. The measurements show that the charge starts high and then decreases in a step-like manner as the droplet evaporates. The discharge trend is predicted well by treating the droplet as a dielectric sphere in electrical contact with the hot plate, but the mechanism by which definite lumps of charge are transferred through the vapor film is still an open question. 10:43AM H13.00002 Leidenfrost state suppression at ultrahigh temperatures , ARJANG SHAHRIARI, JILLIAN WURZ, VAIBHAV BAHADUR, University of Texas at Austin — The Leidenfrost effect is the formation of a vapor layer between a liquid and an underlying hot surface which severely degrades heat transfer and results in surface temperature overshoots. We demonstrate and analyze electrostatic suppression of the Leidenfrost state at ultrahigh surface temperatures. A concentrated electric field across the vapor layer can attract liquid towards the surface and promote wetting. This principle is successful even at ultrahigh temperatures. Elimination of the vapor layer increases heat dissipation capacity by more than one order of magnitude. Heat removal capacities exceeding 500 W/cm2 are reported, which is a significant advancement in boiling heat transfer. The underlying science can be understood via a multiphysics analytical model which captures the coupled electrical-fluid-heat transport phenomena underlying Leidenfrost state suppression. The physical insights gained are used to devise and demonstrate a novel electrostatic suppression technique which does not need any surface modifications. Overall, this work uncovers the physics underlying dryout prevention and demonstrates electrically tunable boiling heat transfer with ultralow power consumption. 10:56AM H13.00003 Electrohydrodynamics of a particle-covered drop1 , MALIKA OURIEMI, IFPEN, France, PETIA VLAHOVSKA, Brown University — We study the dynamics of a drop nearly-completely covered with a particle monolayer in a uniform DC electric field. The weakly conducting fluid system consists of a silicon oil drop suspended in castor oil. A broad range of particle sizes, conductivities, and shapes is explored. In weak electric fields, the presence of particles increases drop deformation compared to a particle-free drop and suppresses the electrohydrodynamic flow. Very good agreement is observed between the measured drop deformation and the small deformation theory derived for surfactant-laden drops (Nganguia et al, 2013). In stronger electric fields, where drops are expected to undergo Quincke rotation (Salipante and Vlahovska, 2010), the presence of the particles greatly decreases the threshold for rotation and the stationary tilted drop configuration observed for clean drop is replaced by a spinning drop with either a wobbling inclination or a very low inclination. These behaviors resemble the predicted response of rigid ellipsoids in uniform electric fields. At even stronger electric fields, the particles can form dynamic wings or the drop implodes. The similar behavior of particle–covered and surfactant–laden drops provides new insights into understanding stability of Pickering emulsions. 1 supported by NSF-CBET 1437545 11:09AM H13.00004 Equilibrium Shapes and Instability of Liquid Drops in Electric Field , ASGHAR ESMAEELI, Southern IL Univ-Carbondale — Electrohydrodynamics of liquid drops is currently the focus of increased attention because of its relevance in a host of processes such as micro- and bio-fluidics. In a weak electric field the drop acquires an equilibrium shape, deforming to an oblate or a prolate spheroid. However, beyond a critical electric field strength it will disintegrate through tip-streaming or bulbous breakup. The modes and mechanisms of drop disintegration has been reasonably well-studied, assuming the drop viscosity to be the same as that of the ambient. However, there are some evidences that suggest the viscosity ratio can dramatically affect the dynamics, even leading to new breakup modes. The goal of this study is to provide further insight regarding this issue through computational simulations. To this end, we use a front tracking/finite difference scheme in conjunction with Taylor leaky-dielectric model to solve the governing electrohydrodynamic equations. 11:22AM H13.00005 Electric field induced deformation of sessile drops1 , LINDSEY CORSON, University of Strathclyde, Glasgow, UK, COSTAS TSAKONAS, Nottingham Trent University, Nottingham, UK, BRIAN DUFFY, NIGEL MOTTRAM, University of Strathclyde, Glasgow, UK, CARL BROWN, Nottingham Trent University, Nottingham, UK, STEPHEN WILSON, University of Strathclyde, Glasgow, UK — The ability to control the shape of a drop with the application of an electric field has been exploited for many technological applications including measuring surface tension, producing an optical display device, and optimising the optical properties of microlenses. In this work we consider, both theoretically and experimentally, the deformation of pinned sessile drops with contact angles close to either 0◦ or 90◦ resting on the lower substrate inside a parallel plate capacitor due to an A.C. electric field. Using both asymptotic and numerical approaches we obtain predictive equations for the static and dynamic drop shape deformations as functions of the key experimental parameters (drop size, capacitor plate separation, electric field magnitude and contact angle). The asymptotic results agree well with the experimental results for a range of liquids. 1 We gratefully acknowledge the financial support of EPSRC via research grants EP/J009865 and EP/J009873. 11:35AM H13.00006 Immersed Interface Method for Drop Electrohydrodynamic , HERVE NGANGUIA, YUAN-NAN YOUNG, New Jersey Institute of Technology, ANITA LAYTON, Duke University, WEI-FAN HU, MING-CHIH LAI, National Chiao Tung University — A numerical scheme based on the immersed interface method (IIM) is developed to simulate the dynamics of an axisymmetric viscous drop under an electric field. In this work, the IIM is used to solve both the fluid velocity field and the electric potential field. Detailed numerical studies on the numerical scheme shows second-order convergence. Moreover, the numerical scheme is further validated by the good agreement with published analytical models, and results from the Boundary Integral method. The IIM code is used to investigate inertia effects and/or time-varying electric field on drop electro-deformation. Results from the simulations illustrate how the inertia effects and time dependence of the electric field affect the electro-deformation of a viscous leaky dielectric drop. 11:48AM H13.00007 A high throughput droplet based electroporation system1 , BYEONGSUN YOO, MYUNGMO AHN, POSTECH, DOJIN IM, Pukyong Natioanl Univ., INSEOK KANG, POSTECH — Delivery of exogenous genetic materials across the cell membrane is a powerful and popular research tool for bioengineering. Among conventional non-viral DNA delivery methods, electroporation (EP) is one of the most widely used technologies and is a standard lab procedure in molecular biology. We developed a novel digital microfluidic electroporation system which has higher efficiency of transgene expression and better cell viability than that of conventional EP techniques. We present the successful performance of digital EP system for transformation of various cell lines by investigating effects of the EP conditions such as electric pulse voltage, number, and duration on the cell viability and transfection efficiency in comparison with a conventional bulk EP system. Through the numerical analysis, we have also calculated the electric field distribution around the cells precisely to verify the effect of the electric field on the high efficiency of the digital EP system. Furthermore, the parallelization of the EP processes has been developed to increase the transformation productivity. 1 This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (Grant Number: 2013R1A1A2011956). 12:01PM H13.00008 Electric-Field-Assisted Droplet Dispensing on Immiscible Fluids1 , TAEWOONG UHM, Department of Chemical Engineering, POSTECH, JIWOO HONG, SANG JOON LEE, Department of Mechanical Engineering, POSTECH, IN SEOK KANG, Department of Chemical Engineering, POSTECH — Dispensing tiny droplets is a basic and crucial process in numerous practical applications, such as printed electronics, DNA microarray, and digital microfluidics. The precise positioning with demanded size of droplets is the main issue of dispensing tiny droplets. Furthermore, capability of dispensing charged droplets on the immiscible fluids could bring out more utilities. In this work, we demonstrate the droplet dispensing on immiscible fluids by means of electrical charge concentration (ECC). This results from the fact that the droplet is generated by electric force caused by electric induction between the surface of droplet and the immiscible fluid. The temporal evolution of the droplet-dispensing process was observed consecutively with a high-speed camera. In addition, the relationship between the size of dispensed droplet and the parameters, such as physical properties of fluids and electrical field strength, is established. 1 This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (Grant Number: 2013R1A1A2011956). 12:14PM H13.00009 Measurement of Charge Transfer to Aqueous Droplets In High Voltage Electric Fields , ERIC ELTON, ETHAN ROSENBERG, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — When water droplets contact electrodes in insulating oils, the electrodes impart a net charge to the droplet. Under sufficiently high field strengths, the droplet moves back and forth between electrodes, in effect “bouncing” between them. Although the droplets clearly acquire charge, the exact mechanism by which charge transfer occurs remains unclear. Here we present evidence that the charge transfer process for a given droplet varies strongly with the number of previous bounces. Simultaneous high speed video and high resolution electrometry were used to quantify the effect of droplet composition, ionic strength, electrode material, and applied voltage. We show that bounce-to-bounce variation in charge transfer is a strong function of ionic strength and pH, and we establish that the amount of charge transferred systematically drifts in magnitude with time. Taken together, the results suggest that electrochemical reactions play a key role in modulating the charge transferred to the drop. 12:27PM H13.00010 Levitation of oil droplets over an electrode in oscillatory electric fields , SCOTT BUKOSKY, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — Application of an oscillatory electric field causes immiscible oil droplets in water to aggregate, a phenomenon believed to result from induced electrohydrodynamic (EHD) fluid flows. Recently it has also been shown that rigid colloids exhibit a distinct bifurcation of their equilibrium height over the electrode in response to low frequency electric fields. Here we report that oil droplets also exhibit a bifurcation in their equilibrium height in response to oscillatory fields. Optical and confocal microscopy observations show that at low applied frequencies (< 100 Hz) a large fraction of droplets levitates up to several microns away from the electrode. We investigate the impact of the electric field properties and droplet size on the levitation, and we discuss the implications of this height bifurcation phenomenon for separation of emulsified oils from solution via a contactless electrostatic precipitation process. Monday, November 24, 2014 10:30AM - 12:40PM — Session H14 Drops: Bouncing, Impact and Dynamic Surface Interactions III 3009/3011 - Francois Blanchette, University of California, Merced 10:30AM H14.00001 Studying gas-sheared liquid film in horizontal rectangular duct with LIF technique: droplets deposition and bubbles entrapment1 , ANDREY CHERDANTSEV, University of Nottingham, and Kutateladze Institute of Thermophysics, Russia, DAVID HANN, BARRY AZZOPARDI, University of Nottingham — High-speed laser-induced fluorescence technique is applied to study gas-sheared liquid film in horizontal rectangular duct (width 161 mm). Instantaneous distributions of film thickness over an area of 50*20 mm are obtained with frequency 10 kHz and spatial resolution 40 µm. The technique is also able to detect droplets entrained from film surface and gas bubbles entrapped by the liquid film. We focus on deposition of droplets onto film surface and dynamics of bubbles. Three scenarios of droplet impact are observed: 1) formation of a cavern, which is similar to well-known process of normal droplet impact onto still liquid surface; 2) “ploughing,” when droplet is sinking over long distance; 3) “bouncing,” when droplet survives the impact. The first scenario is often accompanied by entrainment of secondary droplets; the second by entrapment of air bubbles. Numerous impact events are quantitatively analyzed. Parameters of the impacting droplet, the film surface before the impact, the evolution of surface perturbation due to impact and the outcome of the impact (droplets or bubbles) are measured. Space-time trajectories of individual bubbles have also been obtained, including velocity, size and concentration inside the disturbance waves and in the base film region. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 10:43AM H14.00002 Simulation of Droplet Collision using Moment of Fluid Method , YONGSHENG LIAN, University of Louisville, YONGSHENG LIAN TEAM, YISEN GUO TEAM — Binary droplet collisions were numerically studied. For the present method the interfaces between different phases were captured using the moment of fluid method, a directionally split cell integrated semi-Lagrangian method was used to calculate interface and momentum advection, a projection method was used to calculate pressure, and a block structured adaptive mesh refinement method was used to locally increase the resolution in the regions of interest. Both head-on collisions and oblique collisions were investigated. Droplets of same material and different materials were considered. The effects of droplet size ratio, dimensionless impact parameter, and the Weber number on the collision outcome were systematically investigated. Our results showed that the method can accurately predict the droplet bouncing, reflective separation and coalescence. 10:56AM H14.00003 Impact and rebound behaviors of polymer solution droplets with different concentrations and molecular weights , HYUNG KYU HUH, SANG JOON LEE, POSTECH — The spreading and rebounding behaviors of diluted polyehtyleneoxide(PEO) solution droplets impacted on a Teflon-coated surface are experimentally investigated using a high-speed imaging technique. The maximum spreading of PEO droplets are well fitted in a single curve, regardless of the polymer concentration. Additional energy dissipation by polymer additives is increased during the retraction phase of droplets, as the polymer concentration increases. Polymer solution of high molecular weight dissipates more energy, compared to that of low molecular weight. There is no significant effect on the energy dissipation, when the polymer concentration is smaller than 0.03wt%. Polymer residue composed of small satellite droplets is optically observed after retraction of droplet contact line. Contact-line velocity on the residue area is decreased, because the residue works as an additional friction on the surface. The friction coefficient of polymer solution is increased linearly as the reduced concentration of polymer solution increases. A semi-empirical model is derived to estimate the rebound tendency of PEO droplets as a function of the maximum spreading factor, the retraction velocity and the reduced concentration. 11:09AM H14.00004 Impact of liquid drops on a moving surface1 , CHRISTOPHE PIRAT, HENRI LASTAKOWSKI, FRANCOIS BOYER, ANNE-LAURE BIANCE, CHRISTOPHE YBERT, ILM, Univ. Lyon1 CNRS — When a liquid drop impacts on a solid surface, it is well known that, depending on the impact velocity, liquid and surface properties, it can experience various, rich and complex dynamics. In this experimental study, we focus on the specific case of a water drop that impacts on a smooth surface having a tangential velocity. For a rather high surface velocity, a thin layer of air intercalates between the surface and the drop throughout the spreading, leading to a low friction condition similar to what is observed for a leidenfrost impact. For a low surface velocity, the surface can carry the full drop away. We report here on the intermediate regime: a partial rebound observed when a droplet detaches “upstream” from the rest of the drop. We study the threshold below which the surface cannot pull the liquid film without breaking-up anymore. Two distinct situations are observed, depending on the relative strength of capillary and viscous effects. 1 The authors would like to thank ANR for funding through the FREEFLOW project (contract number ANR-11BS04-001-02) 11:22AM H14.00005 Vortex rings in drop impact on liquid pool , JI SAN LEE, SU JI PARK, Dept. Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Korea, BYUNG MOOK WEON, Schoold of Advanced Materials Science and Engineering, Sungkyunkwan University, Korea, KAMEL FEZZAA, Advanced Photon Source, Argonne National Laboratory, USA, JUNG HO JE, Dept. Materials Science and Engineering, Pohang University of Science and Technology (POSTECH), Korea — Since Thomson and Newall’s pioneering work in 19th century, the formation of a vortex ring by drop impact has attracted many scientists over a century because of fundamental interests involved as well as importance in fluid mixing and mass transport process. However, the origin of vorticity and the dynamics of vortex ring are for the most part unexplored and not clearly understood yet. In this study, unprecedented dynamic features of vortex rings in drop impact are unveiled by applying ultrafast X-ray imaging. We reveal that capillary waves contribute to the generation of multiple vortex rings along the wall of the drop. Each ring shows different vorticity, specifically different dependency on Reynolds number. Finally we build up a phase diagram for the multiplicity of vortex rings, which shows a novel analogy with external jetting phenomena in drop impact on liquid pool. 11:35AM H14.00006 Thin sheet break-up in droplet-pool impact events1 , SHAHAB MIRJALILI, ALI MANI, Stanford University — Many experiment have shown that during the impact of a droplet of the size of a few millimeters on a pool of the same liquid with a velocity of a few meters per second, a thin sheet of gas is entrapped delaying the contact of the two liquid bodies. It has also been demonstrated that the break-up of this sheet, which happens in very small time scales, can lead to the generation of micro-bubbles. Given the very small scales involved, this problem is cumbersome to study numerically. In this work, we have undertaken this task by tackling the problem in 2-D. First, we use a relatively cheap boundary element simulation to find the evolution of the profiles prior to impact. After identifying the regimes of interest, and the relevant parameters and scales, diffuse interface CFD calculations are done and the process of sheet breakup and bubble generation is resolved via this approach. Parameter dependence studies are performed using these tools and statistics such as thin film thickness, length and micro-bubble distributions are presented. Finally, a linear stability analysis of thin gas sheet is performed and using the data from the two aforementioned approaches, thin gas sheet breakup is examined in the context of hydrodynamic instabilities. 1 Supported by ONR. 11:48AM H14.00007 Droplet Impacting on Liquid Film: Evolution of Entrapped Air Layer , XIAOYU TANG, ABHISHEK SAHA, CHUNG K. LAW, Princeton Univ — In this work we experimentally studied the dynamics of droplet impacting films of various thicknesses within a range of h/R ≤ 1 (h: film thickness; R: droplet radius). High speed imaging and color interferometry enabled the investigation of the evolution of the air layer trapped between two surfaces, which plays a critical role in determining the collision outcome. Globally, two distinct regimes of impaction outcome, namely bouncing and merging, are observed at low and high impact inertia, respectively. Impaction with high inertia depletes the air layer before the droplet significantly deforms, resulting in permanent merging. At low impact inertia, however, color interferometry shows the existence of three phases prior to bouncing. Upon impaction, droplet endures significant deformation trapping the air layer between the interfaces, hence increasing the resistance force. This phase is characterized by fast deformation of the air film, followed by a period of pseudo equilibrium, with minimal changes in the interfacial air-film profile. The droplet also lacks kinetic energy to penetrate further into the film, resulting in a temporary balance between the droplet weight and air-film pressure. The deformed droplet eventually relaxes towards spherical shape to destroy the equilibrium. Fast change occurs in air-layer and pressure distribution favoring the droplet bouncing. 12:01PM H14.00008 Water Entry by a Train of Droplets , CLAUS-DIETER OHL, XIN HUANG, CHON U CHAN, Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore, PHILIPP ERHARD FROMMHOLD, Christian Doppler Laboratory for Cavitation and Micro-Erosion, Third Institute of Physics, Georg-August-Universität Göttingen, Germany, ALEXANDER LIPPERT, Lam Research AG, Villach, Austria — The impact of single droplets on a deep pool is a well-studied phenomenon which reveals reach fluid mechanics. Lesser studied is the impact of a train of droplet and the accompanied formation of largely elongated cavities, in particular for well controlled droplets. The droplets with diameters of 20-40 µm and velocities of approx. 20 m/s are generated with a piezo-actuated nozzle at rates of 200-300 kHz. Individual droplets are selected by electric charging and deflection and the impact is visualized with stroboscopic photography and high-speed videos. We study in particular the formation and shape of the cavity as by varying the number of droplets from one to 64. The cavities reach centimetres in length with lateral diameters of the order of 100 of micrometres. 12:14PM H14.00009 Radial jetting during the impact of compound drops , JIA MING ZHANG, ER QIANG LI, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology — Here we report radial jetting behavior during the impact of compound droplet onto a dry solid surface. The size and number of the inner droplets was precisely controlled by a microfluidic device. With the help of high-speed video imaging from both side view and bottom view, intricate and regular horizontal jetting patterns were recorded. The radial jets are formed due to the interaction between the inner droplets and the outer liquid film, and the jet velocity is much higher than the drop impact velocity. The number of inner droplets and their position within the outer droplet were shown to be very important parameters which governed the generation and pattern of the jets. Other parameters such as droplet impact velocity, inner/outer liquid viscosity, density and interfacial tension have also been varied and used to analyze the jetting dynamics. Entrapment of minute air bubbles [1] was also clearly observed. [1] S. T. Thoroddsen, K. Takehara and T. G. Etoh, “Bubble entrapment through topological change,” Phys. Fluids, 22, 051701 (2010). 12:27PM H14.00010 A model for wave-droplet interaction in a confined environment1 , TRISTAN GILET, University of Liege, FRANCOIS BLANCHETTE, University of California, Merced — A walker is a droplet bouncing on a liquid surface and propelled by the waves that it generates. This macroscopic wave-particle association exhibits behaviors reminiscent of quantum particles. The horizontal trajectory of a single walker becomes chaotic when it is subject to horizontal confinement. Recent experiments (D. Harris et al., Phys. Rev. E 2013) reveal that the statistics of the walker position is shaped by the eigenmodes of the cavity in which it is confined, similarly to a quantum particle in a box. In this talk, we introduce a model of the coupling between a droplet and a confined surface wave. The resulting iterated map captures many features of the walker dynamics under confinement. These features include the time decomposition of the chaotic trajectory in quantized eigenstates, and the droplet statistics being shaped by the wave. It suggests that deterministic wave-particle coupling expressed in its simplest form can account for some quantum-like behaviors. 1 This research is part of the ARC project QUANDROPS funded by the FWB (Belgium). Monday, November 24, 2014 10:30AM - 12:40PM Session H15 Drops: Elastic Surfaces and Fibers — 3022/3024 - Paul H. Steen, Cornell University 10:30AM H15.00001 Elastocapilllarity in insect adhesion: the case of beetle adhesive hair1 , SOPHIE GERNAY, TRISTAN GILET, University of Liege, PIERRE LAMBERT, Universite Libre de Bruxelles, WALTER FEDERLE, University of Cambridge — The feet of many insects are covered with dense arrays of hair-like structures called setae. Liquid capillary bridges at the tip of these micrometric structures are responsible for the controlled adhesion of the insect on a large variety of substrates. The resulting adhesion force can exceed several times the body weight of the insect. The high aspect-ratio of setae suggests that flexibility is a key ingredient in this capillary-based adhesion mechanism. There is indeed a strong coupling between their elastic deformation and the shape of the liquid meniscus. In this experimental work, we observe and quantify the local deflection of dock beetle seta tips under perpendicular loading using interference microscopy. Our results are then interpreted in the light of an analytic model of elastocapillarity. 1 This research has been funded by the FRIA/FNRS and the Interuniversity Attraction Poles Programme (IAP 7/38 MicroMAST) initiated by the Belgian Science Policy Office. 10:43AM H15.00002 Capillary stretching of elastic fibers , SUZIE PROTIERE, CNRS - Institut Jean le Rond d’Alembert, HOWARD A. STONE, MAE - Princeton University, CAMILLE DUPRAT, LadHyX - Ecole Polytechnique — Fibrous media consisting of constrained flexible fibers can be found in many engineered systems (membranes in filters, woven textile, matted paper). When such materials interact with a liquid, the presence of liquid/air interfaces induces capillary forces that deform the fibers. To model this interaction we study the behaviour of a finite volume of liquid deposited on two parallel flexible fibers clamped at both ends. A tension along the fibers is imposed and may be varied. We show that the system undergoes various morphological changes as the interfiber distance, the elasticity and the tension of the fibers are varied. For a certain range of parameters, the liquid spreads along the fibers and pulls them together, leading to the “zipping” of the fibers. This capillary adhesion can then be enhanced or reduced by changing the tension within the fibers. We will show that balancing stretching and capillary forces allows the prediction of this transition as well as the conditions for which detachment of the fibers occurs. These results may be used to prevent the clogging of fibrous membranes or to optimize the capture of liquids. 10:56AM H15.00003 Tunable Transport of Drops on a Vibrating Fiber , ALISON BICK, ALBAN SAURET, FRANCOIS BOULOGNE, HOWARD STONE, Princeton Univ — Transport of liquid drops on a fibrous medium is common in engineering systems such as fog harvesting and textile cleaning. The control of the drop movement on fibrous media can make these engineering systems more efficient. We investigated how to move drops along a single inclined fiber by controlling fiber vibration. Drop motion: static, sliding or falling, depends on the fiber inclination angle, drop volume, and the distance of the drop from the vibrating source. Specifically, by vibrating the fiber the transition between the three drop motion states can be controlled. By defining the response of drop movement to vibration frequency, we can model the drop movement transition. This knowledge is directly useful for controlling drop movement on the fiber. In particular, we experimentally demonstrated that vibration frequency can be used to transport a drop along a fiber. 11:09AM H15.00004 Compound droplet manipulations on fiber arrays1 , FLORIANE WEYER, MARJORIE LISMONT, LAURENT DREESEN, NICOLAS VANDEWALLE, University of Liege — Recent works demonstrated that fibers are the basis of an open digital microfluidics. Indeed, various processes such as droplet motion, fragmentation, trapping, releasing, mixing and encapsulation can be constructed on fiber arrays. However, addressing a large number of tiny droplets resulting from the mixing of several liquid components is still a challenge. Here we show that it is possible to manipulate tiny droplets reaching a high level of complexity. Wetting droplets are known to glide along vertical fibers. When a droplet reaches an horizontal fiber, it sticks at the crossing if capillary overcomes gravity. Otherwise, the droplet continues its way, crosses the node and leaves a tiny residue. Therefore, a vertical fiber decorated with a series of horizontal fibers will retain residual droplets at the successive nodes. An oil droplet, sliding on the vertical fiber, is able to collect the residues. Thus a multicompound droplet is created. The volume of the residual droplets has been studied and seems to be related to the diameters of both vertical and horizontal fibers. Moreover, the conditions under which the residues are released have been investigated in order to understand the formation of such a fluidic object. 1 F. Weyer is financially supported by an FNRS grant. This work is also supported by the FRFC 2.4504.12. 11:22AM H15.00005 Dynamic Contact Angle of a Soft Solid , STEFAN KARPITSCHKA, University of Twente, SIDDHARTHA DAS, University of Maryland, BRUNO ANDREOTTI, Univ. Paris-Diderot, JACCO SNOEIJER, University of Twente — The wetting motion of a liquid on a rigid solid is a multi-scale problem in which viscous effects at microscopic scales modify the macroscopic liquid contact angle. Here we show that a contact line moving on a soft elastic substrate also leads to a dynamic contact angle, but this time, in the solid: the initially flat solid surface is deformed elastically into a sharp ridge whose angle depends on the contact line velocity. We predict the dynamic solid contact angle for generic viscoelastic rheologies. The solid angle provides a mechanism for changing the liquid contact angle, which is corroborated by dynamic wetting experiments of water on silicone gels. Our dynamical theory for soft wetting also captures the growth or decay of the wetting ridge, as recently accessed experimentally. 11:35AM H15.00006 Dynamic wetting on soft substrates studied by x-ray imaging , SU JI PARK, JUNG HO JE, Dept. Materials Science and Engineering, Pohang University of Science and Technology (POSTECH) — When a droplet sits on a soft surface, the surface tension of the droplet deforms the underlying material, creating a wetting ridge. Wetting ridge formation affects not only static wetting but also dynamic wetting behaviors. However, the underlying mechanisms are still largely unexplored mostly due to limitations in observation. Here, we directly visualize wetting ridges in real-time during spreading of a liquid drop using x-ray microscopy with high spatial and temporal resolutions. We clearly show that ridge-growth dynamics is closely linked to spreading behaviors. Interestingly, we reveal that the bending of a ridge cusp enhances the pinning force. We believe that our results would shed light on understanding of dynamic wetting behaviors on soft solids (e.g. contact angle hysteresis or evaporation) and be potentially important to interpret complex biological processes on or in soft tissues (e.g. cell-substrate interactions). 11:48AM H15.00007 Receding Contact Line on a Soft Gel: Dip-Coating Geometry Investigation , TADASHI KAJIYA, Max Plank Institute for Polymer Research, PHILIPPE BRUNET, LAURENT ROYON, ADRIAN DAERR, MATHIEU RECEVEUR, LAURENT LIMAT, MSC, UMR 7057 CNRS, Univ. Paris Diderot — We investigated the behavior of a liquid contact line receding on a soft gel surface (SBS-paraffin). To realize a well-defined geometry with an accurate control of velocity, a dip-coating setup was implemented. As the elastic modulus of the gel is small enough, a significant deformation takes place near the contact-line, which in turn influences the wetting behaviour. Depending on the translation velocity, the contact line exhibits different regimes of motions. Continuous motions are observed at high and low velocities, meanwhile two types of stick-slip motions, periodic and erratic, appear at intermediate velocities. We conjecture that the observed transitions could be explained in terms of the competition between different frequencies, i.e., the frequency f of the strain field variation induced by the contact line motion and the frequency fcross = 1/τgel related to the material relaxation. Finally, we propose a qualitative modeling which predicts the existing range of the stick-slip regimes. Therein, we consider the continuous spectrum associated with the surface deformation that ranges from the meniscus size to the elasto-capillary length: 1/lcap < 1/l < 1/le . 12:01PM H15.00008 Wrapping a liquid drop with a thin elastic sheet , JOSEPH PAULSEN, VINCENT DÉMERY, BENNY DAVIDOVITCH, CHRIS SANTANGELO, THOMAS RUSSELL, NARAYANAN MENON, University of Massachusetts Amherst — We study the wrapping of a liquid drop by an initially-planar ultrathin (∼ 100 nm) circular sheet. These elastic sheets can completely relax compressive stresses by forming wrinkles [1]. In the experiment, we find that when a small fraction of the drop is covered, the overall shape of the sheet (i.e. averaging over the wrinkles) is axisymmetric. As we shrink the drop further, the sheet develops radial folds that break the axisymmetry of the sheet and the drop. Our data are consistent with a model where the sheet selects the shape that minimizes the exposed liquid surface area. We thus identify a “geometric wrapping” regime, where the partially-wrapped shape depends only on the relative radii of the sheet and the drop; the global breaking of axisymmetry is independent of the elastic energy of the deformed sheet. This regime requires that bending energy is negligible compared to surface energy, in contrast to the “capillary origami” regime [2] where the static shape of the drop comes from a balance of bending and capillary forces. [1] King et al., PNAS 109, 2012. [2] Py et al., PRL 98, 2007. 12:14PM H15.00009 Dynamics of fluid driven fracturing of magnetic elastica , JEROME NEUFELD, BP Institute, Department of Earth Sciences, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, THOMAS LE REUN, École Normale Supérieure de Lyon — The dynamic spreading of viscous fluids between magnetic elastic sheets provides an novel experimental system in which to examine the interplay between bending, in-plane tension, and a form of “magnetic” fracturing. A suite of constant volume, and fixed flux injections of viscous fluid demonstrate the quasi-static and dynamic responses of the magnetic, elastic sheets. For fixed fluid volume, magnetic adhesion leads to static solutions in both the tensional (thin membrane) and bending (thick membrane) limits, directly analogous to to the sessile capillary drop. These static tensional and bending solutions are also observed when the fluid injection flux is small. For larger fluid fluxes, new dynamic solutions emerge as reflected in the rates of deformation and spreading. This new experimental system provides a laboratory in which to repeatably study the dynamics of fluid driven fracturing of elastica. 12:27PM H15.00010 Elastic membrane for needle-free drop deposition , PRASHANT WAGHAMRE, SUSHANTA MITRA, Department of Mechanical Engineering, University of Alberta — Contact angle measurements on low-energy surfaces (like superhydrophobic, etc.) are often a challenge as the needle-drop-surface combination does not allow to detach the drop readily. Here we present a new technique to achieve a “needle-free” drop by bringing the drop in contact with an elastic membrane, kept between the needle-drop assembly and the characterizing substrate. The detachment of the drop from the needle is achieved by retracting the needle-drop assembly at a finite speed and allowing the drop to receive the elastic energy of the soft flexible membrane. Such interaction of the drop with the elastic membrane permits the drop to get repelled from the elastic membrane and gets deposited on a characterizing substrate. The repelling behavior of the drop can be controlled by appropriately selecting the mechanical and wetting properties of the elastic membrane. This technique not only provides a needle-free drop deposition that is independent of the physical properties of the liquid and the needle but it also allows achieving the drop size that is independent of the needle diameter. Monday, November 24, 2014 10:30AM - 12:40PM Session H16 Free-Surface Flows V: General — 2000 - Palaniswamy Ananthakrishnan, Florida Atlantic University 10:30AM H16.00001 Long-wave runup in Lagrangian framework: application to lake tsunamis , LOUIS-ALEXANDRE COUSTON, Univ of California - Berkeley, CHIANG C. MEI, Massachusetts Institute of Technology, MOHAMMAD-REZA ALAM, Univ of California - Berkeley — Coastal settlements and infrastructures are exposed and vulnerable to long waves because such waves can climb up sloping beaches without breaking. Thus accurate long-wave runup predictions are of significant importance for effective mitigation and evacuation. Although numerical models are now routinely used for long-wave propagation, calculating the runup remains an arduous task. Operational and research models usually rely on the Eulerian description of the fluid. Yet, runup processes at sloping shores can involve large horizontal stretches of the fluid domain, thereby requiring a varying number of computational nodes based on ad-hoc criteria. Here we argue that a convenient alternative is the less used Lagrangian coordinates system in which the fluid flow is described by following the trajectory of each fluid particle as an unknown function of the initial position. An immediate advantage of the Lagrangian formulation is that the free-surface and the moving shoreline become explicitly part of the solution and defined by their initial positions. As a case study we consider 3D landslide tsunamis in lakes, and present numerical results that highlight the significance of nonlinearity and wave superposition. 10:43AM H16.00002 Pressure Predictions and Run-Up on a Vertical Wall and Sloping Beach , JANNETTE FRANDSEN, CAROLINE SÉVIGNY, RÉGIS XHARDÉ, INRS-ETE, University of Quebec, Canada — This paper presents large scale experiments of water wave impact on wall. This study is concerned with advancing knowledge on rapidly varying pressure magnitude and distributions on different types of sea/river/harbor walls. The experiments are conducted in the new Quebec Coastal Physics Laboratory (QCPL), Canada. The flume has a depth and a width of 5 m and is 120 m long. It is designed for modeling the interactions of waves, currents and sediment transport. The wall has a test area of 1.2 × 2.4 m. The outer regions of the wall are made of steel to span the entire width of the tank. The wall is designed to behave as a rigid plate. The geometric model to full scale is about 1:4. Sensors are mounted along the flume, beach slope and wall to monitor hydrodynamics parameters. The incoming waves evolve over a flat bed to climb the final 25 m on a beach with a constant slope of 1:10. Broad- and narrow-banded spectra representing operational and storm events were investigated. The initial results are promising. Details of the underlying mechanism of various types of breaking and impact on the wall will be presented. 10:56AM H16.00003 Experimental validation of a Fluid-Structure interaction model for simulating offshore floating wind turbines1 , ANTONI CALDERER, CHRIST FEIST, St. Anthony Falls Lab, University of Minnesota, KELLEY RUEHL, Sandia National Lab, MICHELE GUALA, FOTIS SOTIROPOULOS, St. Anthony Falls Lab, University of Minnesota — A series of experiments reproducing a floating wind turbine in operational sea conditions, conducted in the St. Anthony Falls Lab. wave facility, are employed to validate the capabilities of the recently developed FSI-Levelset-CURVIB method of Calderer, Kang and Sotiropoulos (JCP 2014) to accurately predict turbine-wave interactions. The numerical approach is based on solving the Navier-Stokes equations coupled with the level set method, which is capable of carrying out LES of two-phase flows (air and water) with complex floating structures and waves. The investigated floating turbine is a 1:100 Froude scaled version of the 13.2 MW prototype designed by Sandia National Lab; it is installed on a cylindrical barge style platform which is restricted to move with two degrees of freedom, heave and pitch in the vertical plane defined by the direction of the propagating 2D waves. The computed turbine kinematics as well as the free surface elevation results are compared with the experimental data for different free decay tests and wave conditions representative of the Maine and the Pacific North West coasts. The comparison shows promising results indicating the validity of the model for simulating operational floating turbines. 1 This work is supported by the US Department of Energy (DE-EE0005482), the University of Minnesota IREE program, and the Minnesota Supercomputing Institute 11:09AM H16.00004 A 3D GPU-accelerated MPI-parallel computational tool for simulating interaction of moving rigid bodies with two-fluid flows , ASHISH PATHAK, MEHDI RAESSI, University of Massachusetts Dartmouth — We present a 3D MPI-parallel, GPU-accelerated computational tool that captures the interaction between a moving rigid body and two-fluid flows. Although the immediate application is the study of ocean wave energy converters (WECs), the model was developed at a general level and can be used in other applications. Solving the full Navier-Stokes equations, the model is able to capture non-linear effects, including wave-breaking and fluid-structure interaction, that have significant impact on WEC performance. To transport mass and momentum, we use a consistent scheme that can handle large density ratios (e.g. air/water). We present a novel reconstruction scheme for resolving three-phase (solid-liquid-gas) cells in the volume-of-fluid context, where the fluid interface orientation is estimated via a minimization procedure, while imposing a contact angle. The reconstruction allows for accurate mass and momentum transport in the vicinity of three-phase cells. The fast-fictitious-domain method is used for capturing the interaction between a moving rigid body and two-fluid flow. The pressure Poisson solver is accelerated using GPUs in the MPI framework. We present results of an array of test cases devised to assess the performance and accuracy of the computational tool. 11:22AM H16.00005 Free Surface and Flapping Foil Interactions1 , PALANISWAMY ANANTHAKRISHNAN, Florida Atlantic University — Flapping foils for station-keeping of a near-surface body in a current is analyzed using a finite-difference method based on boundary-fitted coordinates. The foils are hinge-connected to the aft of the body and subject to pitch oscillation. Results are obtained for a range of Strouhal number, Froude number, unsteady frequency parameter τ , Reynolds number and the depth of foil submergence. Results show that at low Strouhal number (St < 0.1) and sub-critical unsteady parameter τ < 0.25, the flapping generates drag instead of thrust. At high Strouhal number and super-critical value of the unsteady parameter (τ > 0.25) flapping generates high thrust with low efficiency. Thrust and efficiency are found to decrease with decreasing submergence depth of the foil. At the critical τ = 0.25 and shallow submergence of the foil, the standing wave generated above the foil continues to grow until breaking; both the thrust and efficiency of the foil are reduced at the critical τ . The necessary conditions for optimal thrust generation by a flapping foil underneath the free surface are found to be (i) Strouhal number in the range from 0.25 to 0.35, (ii) unsteady parameter τ > 0.25 and (iii) the maximum angle of attack less than 15o for the flat-plate foil. 1 Supported by the US Office of Naval Research through the Naval Engineering Education Center (NEEC) Consortium of the University of Michigan, Ann Arbor. 11:35AM H16.00006 Fluid flow and the bending, buckling and wrinkling of floating elastica , FINN BOX, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, UK., JEROME NEUFELD, BP Institute, University of Cambridge, UK. — Bending, buckling and wrinkling of floating elastic sheets may be induced by localised viscous flows leading to a complex interplay between flow and deformation. Here we present an experimental investigation illuminating the rich behaviour that arrises due to the coupling between deformation and viscous flow. An understanding of this fluid-structure interaction is applicable over a wide-range of length-scales, from the loading of the Indian subcontient by the Tibetan plateau to the deformation of nanoscale elastic sheets by fluid deposition. We explore the limits of small and large deformations through a range of fluid fluxes and by altering the ratio of spreading to ambient fluid densities, in order to characterise the bending-dominated and tension-dominated spreading regimes. We identify regimes for which the fluid motion is similar to that of a gravity current spreading over a rigid surface and for which the propagation of the fluid is dominated by the elastic deformation of the sheet. The coupling between flow and tension-induced wrinkling of the sheet, which occurs for large deformations, is also explored. 11:48AM H16.00007 New experimental technique for the measurement of the velocity field in thin films falling over obstacles , JULIEN R. LANDEL, ANA DAGLIS, DAMTP, University of Cambridge, HARRY MCEVOY, DSTL, STUART B. DALZIEL, DAMTP, University of Cambridge — We present a new experimental technique to measure the surface velocity of a thin falling film. Thin falling films are important in various processes such as cooling in heat exchangers or cleaning processes. For instance, in a household dishwasher cleaning depends on the ability of a thin draining film to remove material from a substrate. We are interested in the impact of obstacles attached to a substrate on the velocity field of a thin film flowing over them. Measuring the velocity field of thin falling films is a challenging experimental problem due to the small depth of the flow and the large velocity gradient across its depth. We propose a new technique based on PIV to measure the plane components of the velocity at the surface of the film over an arbitrarily large area and an arbitrarily large resolution, depending mostly on the image acquisition technique. We perform experiments with thin films of water flowing on a flat inclined surface, made of glass or stainless steel. The typical Reynolds number of the film is of the order of 100 to 1000, computed using the surface velocity, the film thickness and the kinematic viscosity of the film. We measure the modification to the flow field, from a viscous-gravity regime, caused by small solid obstacles, such as three-dimensional hemispherical obstacles and two-dimensional steps. We compare our results with past theoretical and numerical studies. This material is based upon work supported by the Defense Threat Reduction Agency under Contract No. HDTRA1-12-D-0003-0001. 12:01PM H16.00008 Stilling Basin Performance Analysis by ADV , SOBHAN ALEYASIN, University of Manitoba, NIMA FATHI, PETER VOROBIEFF, University of New Mexico — The outlet flow from dams, channels, and pipes, as well as the river flow, can cause damage to the bed of the river or channel and cause scouring of structures such as the saddles of bridges, because of the huge amount of the kinetic energy carried by the flow. One of the ways to dissipate this energy is via the use of stilling basins, which are structures that calm the flow. Here we present a study of one type of stilling basins for pipe outlets based on a widely used standard1. During the study, splitters and cellular baffles were placed in the stilling basin, and their locations were changed to assess their effect on the flow dissipation. Velocity at several locations in the basin was measured via acoustic Doppler velocimetry (ADV) for different Froude numbers to investigate the effect of flow rate and inlet velocity. Based on the findings of the experiments, we make several suggestions regarding the efficiency and geometry of stilling basins. 12:14PM H16.00009 Damping of liquid sloshing by foams: from everyday observations to liquid transport , ALBAN SAURET, FRANCOIS BOULOGNE, JEAN CAPPELLO, HOWARD STONE, Princeton Univ — When a liquid-filled container is set in motion, the free surface of the liquid starts to slosh, i.e. oscillate. Such effects can be observed when a glass of water is handled carelessly and the fluid sloshes or even spills over the rim of the container. However, beer does not slosh as readily, which suggests that the presence of foam could be used to damp sloshing. In this work, we study experimentally the effect on sloshing of liquid foam placed on top of a liquid bath in a Hele-Shaw cell. We generate a monodisperse 2D liquid foam and track its motion. The influence of the foam on the sloshing dynamics is characterized: 2 to 3 layers of bubbles are sufficient to significantly damp the oscillations. For more than 5 layers of bubbles, the original vertical motion of the foam becomes mainly horizontal. We rationalize our experimental findings with a model that describes the foam contribution to the damping coefficient. This study motivated by everyday observations has promising applications in numerous industrial applications such as the transport of liquid in cargoes. 12:27PM H16.00010 RANS-VOF Modeling of Stratified Turbulent Flow in a Straight Rectangular Duct , CHANDRIMA JANA, URMILA GHIA, University of Cincinnati, LEONID TURKEVICH, National Institute of Occupational Safety and Health — Turbulent, stratified flow of air and water in a straight rectangular duct (aspect ratio 2:1) is investigated. Turbulent flow in straight rectangular ducts exhibits secondary currents or vortices, generated by the anisotropy of the Reynolds stresses near the boundaries. Although these secondary motions are small in comparison with the streamwise motions, they influence the flow and scalar transport, and are challenging to predict accurately. Near the two-fluid interface, the turbulence structures are modified due to their interaction with the interface. The present work simulates the two-fluid flow in the duct in order to capture the structure of the secondary vortices. The turbulent flow is modeled using a Reynolds-Averaged Navier-Stokes (RANS) formulation, along with an anisotropic Reynolds Stress Model (RSM). The air-water interface is tracked using the VOF (Volume-of-Fluid) formulation. The Reynolds stresses are tracked near the solid boundaries and in the vicinity of the air-water interface. The structure of the secondary vortices in the corner formed by the interface and the solid side wall differs from that in the corner formed by the solid duct base and the solid side wall. Monday, November 24, 2014 10:30AM - 12:40PM Session H17 Nonlinear Dynamics IV: Model Reduction — 2002 - Clancy Rowley, Princeton University 10:30AM H17.00001 On the relationship between Koopman Mode Decomposition and Dynamic Mode Decomposition , IGOR MEZIC, HASSAN ARBABI, UCSB — We discuss several issues in theory and applications of Koopman modes in fluid mechanics. We show an explicit relationship between a basic – companion matrix - version of the Dynamic Mode Decomposition (DMD) and the Koopman Mode Decomposition (KMD) of dynamical systems, that allows for estimates of validity of approximation of Koopman modes by DMD modes. As a side result we link the recently introduced Generalized Laplace Analysis and the inverse of the Vandermonde matrix. Using these theoretical results, a new method for computation of Koopman modes is presented that avoids inversion of the Vandermonde matrix. Application of this method to analysis of dynamic stall are shown. 10:43AM H17.00002 Applications of Koopman Operator Theory to Model Reduction in Fluid Mechanics , HASSAN ARBABI, IGOR MEZIĆ, University of California, Santa Barbara — We discuss some applications of the Koopman operator theory to the problems in fluid mechanics. These applications involve the Koopman mode decomposition (KMD), which describes the nonlinear evolution of the flow field observables, such as velocity or vorticity field, in terms of a linear expansion - analogous to the normal mode analysis in linear oscillations. By applying KMD to the incompressible flow in a 2D rectangular cavity, we identify the spectrum of the flow with the associated global modes, both in periodic and aperiodic regime. We also apply KMD to in-vivo measurements of the blood velocity field inside human’s heart, and extract the Koopman modes and frequencies based on the assumption of evolution on an attractor. The dominant Koopman modes are then combined to create a low-dimensional model for both the cavity and heart flow. The mesochronic analysis shows that those reduced models capture the mixing topology with an accuracy comparable to that of the original data. Comparison in the L2 -norm also shows that the reduced models obtained by KMD could give a more accurate representation of the flow field compared to POD. 10:56AM H17.00003 Extending Dynamic Mode Decomposition: A Data–Driven Approximation of the Koopman Operator1 , MATTHEW WILLIAMS, IOANNIS KEVREKIDIS, CLARENCE ROWLEY, Princeton University — In recent years, Koopman spectral analysis has become a popular tool for the decomposition and study of fluid flows. One benefit of the Koopman approach is that it generates a set of spatial modes, called Koopman modes, whose evolution is determined by the corresponding set of Koopman eigenvalues. Furthermore, these modes are valid globally, and not only in some small neighborhood of a fixed point. A popular method for approximating the Koopman modes and eigenvalues is Dynamic Mode Decomposition (DMD). In this talk, we show that DMD approximates the Koopman eigenfunctions, but uses linear monomials to do so; this may be limiting in certain applications. We then introduce an extension of DMD, which we refer to as Extended DMD (EDMD), that uses a richer set of user determined basis functions to approximate the Koopman eigenfunctions. We demonstrate the impact this difference has on the eigenvalues and modes by applying DMD and EDMD to some simple example problems. Although the algorithms for DMD and EDMD appear to be similar, modifications like the ones we will present can be important if the resulting eigenvalues, eigenfunctions, and modes are to accurately approximate those of the Koopman operator. 1 Work supported by the NSF (DMS-1204783) 11:09AM H17.00004 Efficient simulation of detached flows at high Reynolds number , JOSE M. VEGA, VICTOR ASENSIO, RAUL HERRERO, ETSI Aeronauticos, Universidad Politecnica de Madrid, FERNANDO VARAS, EI Telecomunicacion, Universidad de Vigo — A method is presented for the computationally efficient simulation of quasi-periodic detached flows in multi-parameter problems at very large Reynolds numbers, keeping in mind a variety of applications, including helicopter flight simulators, control and certification of unmanned aerial vehicles, control of wind turbines, conceptual design in aeronautics, and civil aerodynamics. In many of these applications, the large scale flows (ignoring the smaller turbulent scales) are at most quasi-periodic, namely the Fourier transform exhibits a finite set of concentrated peaks resulting from the nonlinear passive interaction of periodic wakes. The method consists in an offline preprocess and the online operation. In the preprocess, a standard CFD solver (such as URANS) is used in combination with several ingredients such as an iterative combination proper orthogonal decomposition and fast Fourier transform. The online operation is made with a combination of high order singular value decomposition and interpolation. The performance of the method is tested considering the ow over a fairly complex urban topography, for various free stream intensities and orientations, seeking real time online simulations. 11:22AM H17.00005 Stability Analysis of Non-Newtonian Rotational Flow with Hydromagnetic Effect , NARIMAN ASHRAFI, None — Stability of the magnetorheological rotational flow in the presence of a magnetic excitation in the tangential direction is examined. The conservation of mass and momentum equations for an isothermal Carreau fluid between coaxial cylinders are numerically solved while mixed boundary conditions are assumed. In the absence of magnetic excitation, the base flow loses its radial flow stability to the vortex structure at a critical Taylor number. The emergence of the vortices corresponds to the onset of a supercritical bifurcation. The Taylor vortices, in turn, lose their stability as the Taylor number reaches a second critical number corresponding to the onset of a Hopf bifurcation. The tangential magnetic field turns out to be a controlling parameter as it alters the critical points throughout the bifurcation diagram. Also, the effect of the Hartmann number, the Deborah number and the fluid elasticity on the flow parameters were investigated. 11:35AM H17.00006 Dynamics and Control of a Reduced Order System of the 2-d NavierStokes Equations1 , NEJIB SMAOUI, MOHAMED ZRIBI, Kuwait University — The dynamics and control problem of a reduced order system of the 2-d Navier-Stokes (N-S) equations is analyzed. First, a seventh order system of nonlinear ordinary differential equations (ODE) which approximates the dynamical behavior of the 2-d N-S equations is obtained by using the Fourier Galerkin method. We show that the dynamics of this ODE system transforms from periodic solutions to chaotic attractors through a sequence of bifurcations including a period doubling scenarios. Then three Lyapunov based controllers are designed to either control the system of ODEs to a desired fixed point or to synchronize two ODE systems obtained from the truncation of the 2-d N-S equations under different conditions. Numerical simulations are presented to show the effectiveness of the proposed controllers. 1 This research was supported and funded by the Research Sector, Kuwait University under Grant No. SM02/14. 11:48AM H17.00007 Network-theoretic approach to model vortex interactions1 , ADITYA NAIR, KUNIHIKO TAIRA, Florida State University — We present a network-theoretic approach to describe a system of point vortices in two-dimensional flow. By considering the point vortices as nodes, a complete graph is constructed with edges connecting each vortex to every other vortex. The interactions between the vortices are captured by the graph edge weights. We employ sparsification techniques on these graph representations based on spectral theory to construct sparsified models of the overall vortical interactions. The edge weights are redistributed through spectral sparsification of the graph such that the sum of the interactions associated with each vortex is maintained constant. In addition, sparse configurations maintain similar spectral properties as the original setup. Through the reduction in the number of interactions, key vortex interactions can be highlighted. Identification of vortex structures based on graph sparsification is demonstrated with an example of clusters of point vortices. We also evaluate the computational performance of sparsification for large collection of point vortices. 1 Work supported by US Army Research Office (W911NF-14-1-0386) and US Air Force Office of Scientific Research (YIP: FA9550-13-1-0183). 12:01PM H17.00008 Sparsified-dynamics modeling of discrete point vortices with graph theory1 , KUNIHIKO TAIRA, ADITYA NAIR, Florida State University — We utilize graph theory to derive a sparsified interaction-based model that captures unsteady point vortex dynamics. The present model builds upon the Biot-Savart law and keeps the number of vortices (graph nodes) intact and reduces the number of inter-vortex interactions (graph edges). We achieve this reduction in vortex interactions by spectral sparsification of graphs. This approach drastically reduces the computational cost to predict the dynamical behavior, sharing characteristics of reduced-order models. Sparse vortex dynamics are illustrated through an example of point vortex clusters interacting amongst themselves. We track the centroids of the individual vortex clusters to evaluate the error in bulk motion of the point vortices in the sparsified setup. To further improve the accuracy in predicting the nonlinear behavior of the vortices, resparsification strategies are employed for the sparsified interaction-based models. The model retains the nonlinearity of the interaction and also conserves the invariants of discrete vortex dynamics; namely the Hamiltonian, linear impulse, and angular impulse as well as circulation. 1 Work supported by US Army Research Office (W911NF-14-1-0386) and US Air Force Office of Scientific Research (YIP: FA9550-13-1-0183). 12:14PM H17.00009 Global Model Reduction for the Aerodynamics of Coupled FluidStructure Systems1 , HAOTIAN GAO, MINGJUN WEI, New Mexico State University — We have recently developed a global approach for model order reduction of dynamic problems involving coupled fluid-structure systems. The approach is based on but different from traditional POD-Galerkin projection method, which is usually applied on fluid flow with fixed solid boundaries (or infinite domain). To consider moving boundaries/structures, instead, we work on a modified Navier-Stokes equation for the combined fluid-solid domain where body forcing terms are added for the description of solid motion. Then, POD modes can be easily computed in the combined fluid-solid domain, and so is the Galerkin projection. However, our earlier model required time-consuming integration at every time steps to count for the contribution from solid motion. In the current work, we decompose the solid motion to base functions and reduce the integration time from the number of time steps to a much lower number of representative modes of solid motion. A separate dynamic equation is developed to describe the evolution of these modes of solid motion to further simplify the process and allow fully-coupled fluid-structure interaction to be considered. The accuracy and efficiency of the new approach are demonstrated in both canonical cases (e.g. oscillatory cylinder) and practical applications. 1 Supported by ARL (MAST & AHPCRC) 12:27PM H17.00010 Solitary states in the Taylo-Couette system with a radial temperature gradient1 , CLÉMENT SAVARO, ARNAUD PRIGENT, INNOCENT MUTABAZI, LOMC, UMR6294, CNRS-Université du Havre — The vertical TaylorCouette system with a radial temperature gradient exhibits a rich variety of states since the base flow state is a combination of the circular Couette flow and an axial baroclinic flow. Two main control parameters characterize the flow: the Taylor number (T a) for the rotation and the Grashof number (Gr) for the temperature difference. For small values of Gr, the critical state is the Taylor vortices, and for large values of Gr, the critical states appear either in form of helicoidal vortices or modulated waves. For a fixed value of Gr, increasing T a leads to the appearance of higher instability modes where helicoidal vortices or traveling waves bifurcate into contrarotating vortices. A special attention will be focused on the states observed for |Gr| > 1500 and T a ≃ 12 when the base state bifurcates to a state of modulated wave. A small increase of T a leads to the appearance of a solitary wave which is superimposed to the modulated wave state. Using visualization technique and particle image velocimetry (PIV) coupled with liquid crystal thermography (TLC), we have measured the amplitude of the solitary structure from velocity and temperature fields. The spatial and temporal localizations give the signature of the solitary wave. 1 Supported by the French National Research Agency (ANR) through the program Investissements d’Avenir (ANR-10 LABX-09-01), LABEX EMC3. Monday, November 24, 2014 10:30AM - 12:40PM Session H18 Vortex Dynamics: Dipoles, Pairs and Instabilities — 2004 - Ellen Longmire, University of Minnesota 10:30AM H18.00001 Interaction of a vortex dipole with a deformable cantilevered plate , EUGENE ZIVKOV, SERHIY YARUSEVYCH, SEAN PETERSON, University of Waterloo — The coupled interaction of a vortex dipole impacting the tip of a deformable cantilevered plate is investigated both numerically and experimentally. Numerically, a strongly coupled fluid-structure interaction code is used to simulate the impact at three dipole Reynolds numbers, Re = 500, 1500, and 3000. These Reynolds numbers are representative of flows over small-scale energy harvesting devices, and the plate properties model an ionic polymer-metal composite. Of particular interest is the vortex dynamics and the attendant plate response, with the underlying implications to energy harvesting. As the dipole approaches the plate, secondary vortical structures are generated at the plate, with finer structures present at higher Reynolds number. The dipole breaks up after the initial impact, which is followed by complex vortex interactions of secondary structures. The initial impact produces the largest plate deflection, followed by a more complex response attributed to plate interaction with multiple secondary vortices. The plate response to the initial impact is not strongly dependent upon the Reynolds number. However, the secondary vortex dynamics, and the associated plate loadings, exhibit strong Reynolds number dependence. To validate the numerical results, a similar dipole-plate interaction is modelled experimentally and characterized using flow visualization and time resolved, planar particle image velocimetry. 10:43AM H18.00002 Numerical Study of Transport in Viscous Vortex-Dipole Flows1 , LING XU, University of Notre Dame, ROBERT KRASNY, University of Michigan — Material transport in viscous vortex-dipole flows is studied numerically using a high order finite difference method and the Lamb-dipole as the basic unit in the initial condition. The vorticity field and streamline pattern are displayed, and material curves are tracked in order to visualize particle trapping and detrapping in the dipole. Results are presented for evolution of one dipole, and the interactions of two and three dipoles. 1 Association for Women in Mathematics 10:56AM H18.00003 Interaction of monopolar and dipolar vortices with a shear flow: a numerical study , LEON KAMP, VITOR MARQUES ROSAS FERNANDES, GERT-JAN VAN HEIJST, HERMAN CLERCX, Eindhoven University of Technology — Interaction of large-scale flows with vortices is of fundamental and widespread importance in geophysical fluid dynamics and also, more recently for the dynamics of fusion plasma. More specifically the interplay between two-dimensional turbulence constituted by a collection of unsteady eddies and so-called zonal flows has gained considerable interest because of the relevance for transport and associated barriers. We present numerical results on the interaction of individual monopolar and dipolar vortices with typical sheared channel flows (Couette and Poiseuille). Contrary to monopolar vortices, dipolar ones tend to retain their compactness while propagating through the shear flow along curved pathways without much distortion. Simulations on the interaction of a driven turbulent field with mentioned channel flows are used to explore the suppression of turbulence and turbulent transport and the pronounced role played by the boundaries on these. 11:09AM H18.00004 Interaction of dipolar vortices with a shear flow: experimental results , VITOR MARQUES ROSAS FERNANDES, LEON KAMP, HERMAN CLERCX, GERT-JAN VAN HEIJST, Eindhoven University of Technology — Interaction of large-scale flows with vortices is of fundamental and widespread importance in geophysical fluid dynamics and also, more recently for the dynamics of fusion plasma. More specifically the interplay between two-dimensional turbulence constituted by a collection of unsteady eddies and so-called zonal flows has gained considerable interest because of the relevance for transport and associated barriers. We present an experimental study with a two-fluid-layer setup of the interaction of Lamb-like dipolar vortices with a quasi-two-dimensional channel flow that is driven electromagnetically. Dipoles are injected into the sheared flow perpendicularly and obliquely. Using particle image velocimetry we evaluate the evolution of the dipolar vorticity. Results are confronted with two-dimensional numerical simulations. Dipoles turn out to be quite robust structures despite the shearing action imposed by the background flow. 11:22AM H18.00005 Embbeded dipolar vortices driven by Lorentz forces in a shallow liquid metal layer1 , CINTHYA G. LARA, SERGIO CUEVAS, Universidad Nacional Autonoma de Mexico — We present an experimental and numerical study of the vortex pattern that results from the action of a localized Lorentz force in a thin liquid metal layer (GaInSn) contained in a square box. The fluid motion is generated by the interaction of a uniform D.C. current and a non-uniform magnetic field produced by square-shaped permanent magnet much smaller that the container. Unlike the simple vortex dipole created by a localized Lorentz force in a layer of electrolyte, a more complex vortex pattern is formed in a liquid metal layer. Experiments show the appearance of two “embedded” vortex dipoles with a quasi-stagnat zone in the region of highest magnetic field intensity. The observed pattern can be explained by noticing that the localized magnetic field acts as a magnetic obstacle for the imposed flow. Using the Ultrasonic Doppler Velocimetry technique, we obtained the velocity profiles along the symmetry axis. We developed a quasi-two-dimensional numerical model that takes into account the effect of the boundary layers adhered to the bottom wall, the Hartmann friction and the induced effects. Numerical simulations show a satisfactory qualitative and quantitative agreement with the experimental results. 1 Work supported by CONACYT, Mexico under Project 131399. C. G. Lara acknowledges a grant from CONACYT. 11:35AM H18.00006 Deflection of a vortex pair by a flat plate , MONIKA NITSCHE, JASON ARCHER, University of New Mexico — We investigate the inviscid evolution of a counterrotating vortex pair in the presence of a flat plate. The plate is positioned downstream of the initial vortex pair position, centered on its trajectory. If the plate lies normal to the incoming vortex trajectory, the vortices travel around the plate and leave on the opposite side without changing direction. If the plate is inclined relative to the incoming vortex pair, the vortices are deflected and leave the plate at an angle. Changes in the outgoing angle are highly sensitive to changes in the plate inclination, which under certain conditions lead to singular behaviour. The observations are applied to separate an incoming stream of vortex pairs. 11:48AM H18.00007 Interaction regimes of unequal viscous vortex pairs in the presence of background shear , PATRICK FOLZ, KEIKO NOMURA, University of California - San Diego — The interaction of two co-rotating viscous vortices in linear background shear is investigated through two-dimensional numerical simulations. In general, equal co-rotating viscous vortices will merge if brought within a critical separation distance. The mutually induced strain causes core detrainment which eventually leads to mutual entrainment and the flow transforming into a single vortex with combined strength. Unequal vortices, depending on the degree of asymmetry, may or may not merge depending on the relative timing of core detrainment and core destruction. When background shear is present, advective motion of the vortices is altered. With sufficiently strong adverse shear, the vortices will separate. Merger may be enhanced or inhibited by favorable or adverse shear respectively. Prior studies of interacting invsicid pairs identified several interaction regimes based on a merging efficiency, i.e., the circulation of the final vortex (or vortices) relative to the initial circulation. Here, a similar method is developed for viscous flows, and is used to objectively identify the observed interaction outcomes. A categorization of possible interactions is presented. 12:01PM H18.00008 Evolution of Vortex Pairs Subject to the Crow Instability in Wall Effect1 , DANIEL ASSELIN, C.H.K. WILLIAMSON, Cornell University — In this research, we examine the effect of a solid boundary on the dynamics and instabilities of a pair of counter-rotating vortices. An isolated vortex pair is subject to both a short-wave elliptic instability and a long-wave Crow (1970) instability. Near a wall, the boundary layer that forms between the primary vortices and the wall can separate, leading to the generation of secondary vorticity. In the present study, we are examining the long-wave Crow instability as it is modified by interaction with a wall. Several key features of the flow are observed. Strong axial flows cause fluid containing vorticity to move from the “troughs” of the initially wavy vortex tube to the “peaks.” This process is associated with distinct differences in vortex concentration at the peak and the trough, which lead to the establishment of an axial pressure gradient. Furthermore, the primary and secondary vortices interact to form additional small-scale vortex rings. The exact number and orientation of these small-scale rings is highly dependent on the extent to which the Crow instability has developed prior to interaction with the ground. Finally, significant changes to the vortex dynamics, including circulation, core size, and topology, are also observed during and after interaction with the boundary. 1 This work was supported by the Office of Naval Research under ONR Award No. N00014-12-1-0712. 12:14PM H18.00009 Kelvin-Helmhotz instability and Bénard-Von Karman vortex street in a confined geometry , LUC LEBON, PAUL BONIFACE, MATHIEU RECEVEUR, LAURENT LIMAT, CNRS / Univ. Paris 7 — We have experimentally investigated the appearance of Kelvin-Helmhotz vortices in a confined geometry: in a closed rectangular tank a tape is pulled at high speed on the water surface. This induces a flow in the same direction as the tape, and by conservation a backward flow in the opposite direction. With an appropriate choose of the experiment parameters (water height, tape speed) the backward flow takes place on the sides of the tank: this creates a strong shear that can induces a Kelvin-Helmhotz instability on each side of the tank. As long as the tape width stays small enough compared to the tank width, we can observe the appearance of well organized vortex rows on each sides of the tank. In this case, the vortex rows are coupled like a Bénard-Von Karman vortex street, but without the classical forcing of a wake behind an obstacle. All our experiments are in agreement with a theoretical prediction by Rosenhead which extended the Bénard-Von Karman vortex street stability calculation to a confined geometry. Our work seems to be one of the first experimental verification of this 80 years old model. 12:27PM H18.00010 First instabilities of the wake behind a rotating sphere , JOSE EDUARDO WESFREID, PMMH (ESPCI-CNRS) Paris, France, MACIEJ SKARYSZ, Warsaw University of Technology, Faculty of Power and Aeronautical Eng., SOPHIE GOUJONDURAND, PMMH (ESPCI-CNRS) Paris, France, JACEK ROKICKI, Warsaw University of Technology, Faculty of Power and Aeronautical Eng. — The wake behind a sphere, rotating about an axis aligned with the streamwise direction, has been experimentally investigated in a water tunnel using LIF visualizations and PIV measurements. The measurements focused on the evolution of the flow regimes that appears depending of two control parameters, namely the Reynolds number Re and the dimensionless rotation or swirl rate Ω which is the ratio of the maximum azimuthal velocity of the body to the free stream velocity. In the present investigation, we covers the range of Re smaller than 400 and Ω from 0 and 1.5 . Different wakes regimes such as an axisymmetric base flow, a low frequency frozen state, and an single and double helicoidal mode are represented in the (Re, Ω) parameter plane. Monday, November 24, 2014 10:30AM - 12:40PM Session H19 Convection and Buoyancy-Driven Flows: Turbulence — 2006 - Rudie Kunnen, Technische Universiteit Eindhoven 10:30AM H19.00001 An investigation of transitional Phenomena from Laminar to Turbulent Natural Convection using Compressible Direct Numerical Simulation , CHUNGGANG LI, MAKOTO TSUB- OKURA, RIKEN Advanced Institute for Computational Science, COMPLEX PHENOMENA UNIFIED SIMULATION RESEARCH TEAM — The complete transition from laminar to turbulent natural convection in a long channel is investigated using compressible direct numerical simulation (DNS). Numerical methods of Roe scheme with precontioning and dual time stepping are used for addressing the flow field which is low speed but the density is variable. During the transient development, there are four stages which are laminar, unstable process, relaminarization and turbulence can be obviously identified. After reaching the quasi steady state, the laminar, transition and turbulence simultaneously coexist in the same flow field. Additionally, the comparisons of the statistics with the experimental data are also well consistent. 10:43AM H19.00002 Covariant Lyapunov Vectors of Chaotic Rayleigh-Bénard Convection , MU XU, MARK PAUL, Virginia Tech — The complex dynamics of large spatially extended systems that are driven far-from-equilibrium are central to many important challenges. Much of the difficulty is rooted in the fact that the dynamics are extremely high dimensional. Progress has been made using Lyapunov exponents and vectors that have been computed using frequent Gram-Schmidt reorthonormalizations. However, a significant disadvantage of this approach is that the directions of all of the Lyapunov vectors, except the leading order vector, is lost due to the reorthonormalizations. However, it is well known that there exists a set of vectors intrinsic to the dynamics which satisfy the so-called Oseledec splitting and are called the covariant Lyapunov vectors. Recently, algorithms have become available to compute the spectrum of covariant Lyapunov vectors for large spatially extended systems. In this talk, we use the covariant Lyapunov vectors to explore the chaotic dynamics of Rayleigh-Bénard convection in a large rectangular domain. Knowledge of the covariant Lyapunov vectors allows us to probe fundamental features of the dynamics such as the degree of hyperbolicity, the spatiotemporal features of the spectrum of Lyapunov vectors, and the possible splitting of the dynamics into physical and isolated modes. 10:56AM H19.00003 Analysis of vortical structures in turbulent natural convection , SANGRO PARK, CHANGHOON LEE, Yonsei University — Natural convection of fluid within two parallel walls, Rayleigh-Bénard convection, is studied by direct numerical simulation using a spectral method. The flow is in soft turbulence regime with Rayleigh number 106 , 107 , 108 , Prandtl number 0.7 and aspect ratio 4. We investigate the relations between thermal plumes and vortical structures through manipulating the evolution equations of vorticity and velocity gradient tensor. According to simulation results, horizontal vorticity occurs near the wall and changes into vertical vorticity by vertical stretching of fluid element which is caused by vertical movement of the thermal plume. Additionally, eigenvalues, eigenvectors and invariants of velocity gradient tensor show the topologies of vortical structures, including how vortical structures are tilted or stretched. Difference of velocity gradient tensor between inside thermal plumes and background region is also investigated, and the result indicates that thermal plumes play an important role in changing the distribution of vortical structures. The results of this study are consistent with other researches which suggest that vertical vorticity is stronger in high Rayleigh number flows. Details will be presented in the meeting. 11:09AM H19.00004 Multi-Scale Coherent Structure Interactions in Rayleigh-Benard Convection , PHILIP SAKIEVICH, YULIA PEET, RONALD ADRIAN, Arizona State Univ — Rayleigh-Benard convection (RBC) is characterized by a rich set of coherent structures. One of the most notable and widely recognized structures in RBC is the large scale circulations, or roll-cells. Roll-cells are identified by large circulatory currents that can span the boundaries of the domain. For domains with aspect ratios (AR) of less than two there is generally only one roll-cell present, but as the AR grows the number of roll-cells increase. Currently little is known about the physical dynamics of multiple roll-cell interactions and their effects on the smaller scale structures such as thermal plumes and waves. In the current presentation we present visualizations from a direct numerical simulation of turbulent RBC in a wide AR cell. We identify multiple roll-cells and track the evolution of smaller scale coherent structures as they develop inside the larger scale roll-cells. In this simulation a cylindrical domain with an AR of 6.3 is used with Prandtl and Rayleigh numbers of 6.3 and 9.6*107 respectively. The spectral element code Nek5000 is used for simulation. 11:22AM H19.00005 Anomalous scaling of temperature structure functions in turbulent thermal convection1 , PENGER TONG, Hong Kong University of Science and Technology, XIAOZHOU HE, Max Planck Institute for Dynamics and Self Organization, XIAODONG SHANG, South China Sea Institute of Oceanology, CAS — The scaling properties of the temperature structure function (SF) are investigated in turbulent Rayleigh-Benard convection [1]. The measured SFs are found to exhibit good scaling in space and time and the resulting SF exponent is obtained both at the center of the convection cell and near the sidewall. It is found that the difference in the functional form of the measured SF exponents at the two locations in the cell is caused by the change of the geometry of the most dissipative structures in the (inhomogeneous) temperature field from being sheet-like at the cell center to filament-like near the sidewall. The experiment thus provides direct evidence showing that the universality features of turbulent cascade are linked to the degree of anisotropy and inhomogeneity of turbulent statistics. [1] “Test of the anomalous scaling of passive temperature fluctuations in turbulent Rayleigh-Benard convection with spatial inhomogeneity,” Xiaozhou He, Xiao-dong Shang and Penger Tong, J. Fluid Mech. 753, 104 (2014). 1 This work was supported by the Research Grants Council of Hong Kong SAR. 11:35AM H19.00006 A New Parameterization of N u-Ra Relation in Turbulent RayleighBénard Convection1 , JUN CHEN, ZI-PING CHE, ZHEN-SU SHE, State Key Lab. for Turbulence & Complex Sys., Dept. Mech. & Engg. Sci., College of Engg., Peking Univ. — Nusselt-Rayleigh relation is a key subject in the study of turbulent Rayleigh-Bénard convection (RBC). She et al. introduced Structural Ensemble Dynamics(SED) theory to study wall-bounded turbulence, which yields a multi-layer model of velocity and temperature profiles for RBC system. Here,
we report a result of this study, i.e. a new parameterization of Nusselt number(Nu) as a function of Rayleigh number(Ra): N u = αRa1/7 exp γRaβ . The parameters (α, β and γ) are supposed to be slowly varying with Ra and other physical parameters, in particular Prandtl number(Pr). Analysis of a set of experimental data with Ra = 108 ∼ 1012 and P r = 0.7 ∼ 7.0 shows that this parameterization is efficient, yielding an accurate description of Nu-Ra with errors bounded within 1%. This parameterization surprisingly reveals two distinct states as α varies, with transition at α = 1. Then, an analytic model linking the variation of the three parameters is proposed, yielding a uniform description for the enormous empirical Nu-Ra data, significantly more accurate than the well-known Grossmann-Lohse (GL) model. In conclusion, the SED theory emphasizing the internal profiles provides a viable description of the RBC system. 1 National Natural Sciences Foundation of China, Grant No. 11372362 11:48AM H19.00007 Temperature power spectra of turbulent Rayleigh-Bénard convection with a Prandtl number P r = 12.3 ∗ 1 , GUENTER AHLERS, PING WEI, UCSB, Sante Barbara, CA, USA, XIAOZHOU HE, MPIDS, Göttingen, Germany — We report on measurements of power spectra of temperature fluctuations in turbulent Rayleigh-Bénard convection in a cylindrical sample with aspect ratio Γ = D/L = 0.50 (D is the diameter and L the height) as a function of the distance z from the bottom or top plate. The working fluid was a fluorocarbon at a mean temperature Tm = 25.00o C with a Prandtl number P r = 12.3, and the Rayleigh number was Ra ≃ 4 × 1011 . Consistent with many previous investigations, there was a low-frequency range, spanning about a factor of twenty, where the spectra could be described by a power law P (f ) = P0 f −α . Contrary to the finding by He et al.2 for P r ≃ 0.8 of a universal spectrum with α = 1.0 in the near-wall range z/L ≤ 0.1 and α ≃ 1.5 for z/L = 0.5, we found that α varied with z from about 0.6 near the plate (z/L ≃ 0.01) to about 1.1 at the cell center (z/L = 0.5). Along the sample center line and for z/L ≤ 0.1 α could be described well by α = α0 ln(z/L) + α1 with α0 ≃ 0.2 and α1 ≃ 1.5. 1 Supported 2 X. by NSF Grant DMR11-58514. He, D.P.M. vanGils, E. Bodenschatz, and G.Ahlers, Phys. Rev. Lett. 112, 174501 (2014). 12:01PM H19.00008 Anisotropic turbulent temperature probability densities in high-Ra thermal convection1 , XIAOZHOU HE, DENNIS P. M. VAN GILS, EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Self-Organization, Goettingen, Germany, GUENTER AHLERS, Department of Physics, University of California, USA — We present systematic measurements of conditional diffusion r(x) = hẌ|X = xi and dissipation q(x) = h(Ẋ)2 |X = xi of the normalized temperature fluctuations X = (T − T̄ )/σ in turbulent Rayleigh-Bénard convection (RBC) at several radial positions where the flow is anisotropic. The data cover the Rayleigh-number range 1013 ≤ Ra ≤ 1015 for a Prandtl number P r ≃ 0.80. The sample was a right-circular cylinder with aspect ratio Γ ≡ D/L = 0.50 (D = 1.12 m is the diameter and L = 2.24 m is the height). We compared experimental forms of q(x) and r(x) with previous investigations based on the “fluctuation-dissipation” relation for isotropic flow.2 We derived a general form for the temperature probability-density function (PDF). Similar analyses have also been extended to the study of the temperature time derivative, and to the temperature increment in the time domain. Good agreements are found between experimental temperature probability densities and predicted PDF forms. 1 Supported 2 Emily by the Max Planck Society, the Volkswagenstiftung, the DFG Sonderforschungsbereich SFB963, and NSF Grant DMR11-58514. S. C. Ching, Phys. Rev. Lett. 70, 283 (1993) 12:14PM H19.00009 Scale-by-scale energy budget in turbulent convection , RUDIE KUNNEN, HERMAN CLERCX, Eindhoven University of Technology — Turbulent free convection is driven by buoyancy. A footprint of buoyancy is thus expected in the energy cascade. The existence of this so-called Bolgiano–Obukhov (BO) scaling is a long-standing open question. We use DNS of Rayleigh–Bénard convection in a horizontally periodic domain to address this question. Moderate Rayleigh numbers 2.6 × 106 and 2.5 × 107 are applied, at three different Prandtl numbers 1, 3 and 10. We show that the length scale bounding the convective scaling regime from below, the Bolgiano scale LB , is typically large relative to the domain size. Scale-by-scale energy budgets are calculated based on Yakhot’s equivalent of Kolmogorov’s isotropic four-fifths law for convection. They reveal that buoyancy is active on many scales, obscuring the classical Kolmogorov scaling for scales smaller than LB . Only at very large separations a buoyancy-dominated scaling range could exist. Close to the plates, where LB is smaller, anisotropy complicates the detection of scaling. 12:27PM H19.00010 Turbulence production in low-Pr number convection flows1 , JOERG SCHUMACHER, Technische Universitaet Ilmenau, JANET SCHEEL, Occidental College Los Angeles — Convection at very low Prandtl numbers can be considered in some sense as Terra Incognita given the detailed investigations for P r ∼ 1 or P r > 1 and the challenges in studying these turbulent flows in simulation and experiment. Laboratory experiments for P r < 10−1 have to be conducted in liquid metals such as gallium at P r = 0.021 and sodium at P r = 0.005 both of which are opaque. High-resolution direct numerical simulations are therefore the only tool to unravel the detailed three-dimensional mechanisms of turbulence generation in low-Prandtl number flows and to compare to convection flows at P r ∼ 1. We therefore analyze flows for which Rayleigh and Prandtl numbers are chosen such that the same Grashof number results. Analysis on enstrophy production due to vortex stretching and temperature gradient are discussed together with statistics of local strain. 1 This work is supported by a grant within the INCITE program of DOE, grant HIL07 of the Juelich Supercomputing Centre and the Deutsche Forschungsgemeinschaft Monday, November 24, 2014 10:30AM - 12:40PM Session H20 Flow Control: Drag Reduction — 2008 - Jens Fransson, KTH Royal Institute of Technology 10:30AM H20.00001 Cylinder Drag Reduction Using Cross-Section Variation From Circle to Ellipse , A. BOUABDALLAH, Université des Sciences et de la Tech nologie Houari Boumediene, Algiers, Algeria, H. OUALLI, M. MEKADEM, H. CHETITAH, C. BOULAHBAL, École Militaire Polytechnique, Algiers, Algeria, M. GAD-EL-HAK, Virginia Commonwealth University, Richmond, Virginia, USA — Vortices in the wake of blunt bodies are responsible for significant portion of the drag. An active flow control strategy is designed to inhibit the shedding of such vortex structures. Last year, we presented a numerical study to investigate the effect of periodic cross-section variations on the shed vortices. We extend the research to experiment using a cylinder with finite length. The controlling frequency range is extended up to 40 times the natural shedding frequency f0 . Amplitude and frequency modulations are the key parameters directly affecting the efficiency of the system and the topology of the flow, which permanently adjusts in response to the superimposed pulsatile motion exhibiting a cascade of bifurcations accompanied by shedding modes shifting from the natural mode 2S to 2P, 2T, and 2C. Optimal operational conditions are identified and the results show that drag drastically drops to zero then negative, i.e. thrust, with complete suppression of the vortex shedding. Von Kármán vortices are no longer shed, but rather pulled out in small-scale, weakened vorticity packets released from the lateral cylinder wall. For a deforming amplitude of 100% and an exciting frequency of 20f0 , the negative drag reaches 14 times its value for an uncontrolled cylinder. 10:43AM H20.00002 Disc actuators for turbulent drag reduction , DANIEL J. WISE, CLAUDIA ALVARENGA, PIERRE RICCO, The University of Sheffield, SHEFFIELD FLUID MECHANICS GROUP COLLABORATION — The response of a turbulent channel flow to flush-mounted steadily rotating discs is investigated numerically. The effect on drag reduction of the discs arrangement is studied at a Reynolds number of Rb =5600, based on the bulk velocity and channel height. The flow exhibits complex dependence on the positioning of the discs. For low disc-tip velocity the drag reduction scales linearly with the percentage actuated area, whereas for higher tip velocity the drag reduction may be higher than predicted from this coverage scaling. Therefore by halving the number of discs increased drag reduction can be found. The departure from linear scaling is found to relate to the presence of stationary-wall regions upstream of discs. This is explained by the impingement of the disc boundary layer onto the areas of unactuated wall. For some arrangements tubular, streamwise-elongated structures occur between discs. The criteria for their creation are elucidated through the employment of the Fukagata-Iwamoto-Kasagi identity and flow visualisations. Improvements in the performance of the disc actuators are found with inspiration from the laminar solution to the disc flow. Through the introduction of novel half-disc and annular configurations, a maximum drag reduction of 26% is obtained. 10:56AM H20.00003 Application of opposition control to disks actuators for turbulent drag reduction1 , SOHRAB KHOSH AGHDAM, PIERRE RICCO, MEHDI SEDDIGHI, Univ of Sheffield — Direct numerical simulations have been performed to study an active control technique in which the steadily rotating flush-mounted of either disks or rings is combined with the opposition control approach of Choi et al. (J. Fluid Mech., 262: 75-110). Active blowing and suction are applied such that the controlled velocity on the wall has the same amplitude but opposite sign to that of the optimum detection plane at y + = 15. Various configurations are examined using the out-of-phase v and w control, and the steadily rotating discs and rings. The net drag reduction and turbulence structure of the controlled cases are studied and compared with those when individual controls are applied. The results show that when the “optimum” rotating disks case of Ricco and Hahn (J. Fluid Mech., 722: 267-290) are used together with the opposition control of wall-normal velocity, an additional drag reduction of 5% is achieved. Moreover, in the particular case where the disks are replaced with the rings configuration, the drag reduction is better improved reaching a value of 30%. 1 Airbus Group 11:09AM H20.00004 Utilization of transient growth disturbances for drag reduction in boundary layers1 , JENS H.M. FRANSSON, KTH Royal Institute of Technology — Over the last decade wind tunnel experiments2,3 have shown that steady streamwise elongated streaks, produced by the lift-up mechanism, are able to reduce skin-friction drag by delaying transition to turbulence in flat plate boundary layers. Steady streaks may be generated by passive devices such as circular roughness elements or miniature vortex generators (MVGs), the latter being the more effective device. The optimal streak amplitude to accomplish the stabilizing boundary-layer effect is around 30% of the free-stream velocity (considering an integrated amplitude definition). On the basis of a parametrical study, by varying boundary layer as well as geometrical parameters of the MVGs, a streak amplitude scaling founded on empiricism has been proposed, which is necessary when applying the control strategy in new flow configurations. Different types of disturbances have successfully been damped and the possibility of extending the laminar boundary layer even further by mounting a second array of MVGs downstream of the first one has been accomplished. A review of the AFRODITE program results and future work will be presented. 1 ERC is gratefully acknowledged for their financial support of the AFRODITE program. et al. PRL 96, 064501, 2006. 3 Shahinfar et al. PRL 109, 074501, 2012. 2 Fransson 11:22AM H20.00005 Direct numerical simulation of a compressible turbulent channel flow with uniform blowing and suction through isothermal walls1 , YUKINORI KAMETANI, Kungliga Tek Hogskolan KTH, KOJI FUKAGATA, Keio University — High-speed transports such as aircrafts and bullet trains support human activity in the modern society. In such applications, the turbulent friction drag is the major contributor to the energy loss. Kametani and Fukagata (J. Fluid Mech., 2011) investigated by means of direct numerical simulation (DNS) the drag reduction effect by blowing and the turbulence stabilization effect by suction in an incompressible spatially developing turbulent boundary layer, and quantitatively discussed different contributions to those effects. In this study, DNS of a compressible turbulent channel with uniform blowing and suction through the isothermal walls is performed. The Reynolds number based on the bulk mass flow rate, the viscosity on the wall and the channel half width is set to be 3000. The bulk Mach number is set to be 0.8 and 1.5 to compare the results in subsonic and supersonic cases. The drag reduction (enhancement) effect was confirmed on the blowing (suction) wall. As the Mach number increases, however, the control efficiency of blowing is found to be deteriorated because of the increased density near the wall. 1 Japan Aerospace Exploration Agency, Japan Society for the Promotion of Science 11:35AM H20.00006 The impact of superhydrophobic surface texture on channel-flow turbulence and drag , THOMAS JELLY, SEO YOON JUNG, Imperial College London, TAMER ZAKI, Johns Hopkins University, Imperial College London — Fully-developed turbulent channel flow past streamwise-aligned superhydrophobic surface (SHS) textures is simulated at a fixed bulk Reynolds number, Re = 2,800 (Jelly et al., Phys. Fluids, 2014). The influence of the spanwise-repeated surface pattern is examined using phase-averaged statistics of the flow which is decomposed into mean, periodic and stochastic motions. Relative to a reference no-slip channel flow, the mean skin-friction coefficient is reduced by 21.6%. At particular phases, however, the skin friction far exceeds the reference value. The contributions to drag are examined and are attributed to changes in the primary flow, the presence of a secondary flow of Prandtl’s second type, and changes in the Reynolds stresses. Each of these contributions is quantified, and the largest performance penalties are examined in detail. Finally, an eduction algorithm is used to identify near-wall turbulence structures and to quantify the changes in their strength and population density by the SHS texture. 11:48AM H20.00007 Reinforcement of steady streamwise streaks for consecutive transition delay1 , SOHRAB S. SATTARZADEH, JENS H.M. FRANSSON, Linné Flow Centre, KTH Mechanics — Miniature vortex genrators (MVGs) are recently proven efficient as passive control devices to delay the turbulence transition on a flat plate boundary layer by modulating the base flow in the spanwise direction, through generating steady streamwise elongated streaks, and hence reducing the skin-friction drag2 . As the MVGs are localized in the streamwise direction, a shortcoming of the passive laminar control is the recovery of the two-dimensional boundary layer which force the control effects to fade away. In the present study we show that by placing a second array of MVGs downstream of the first one the streamwise extent of the control can be prolonged by reinforcing the steady streaks in the streamwise direction. The reinforced passive control strategy results in consecutive turbulence transition delay with obtaining a net skin-friction drag reduction of 65%, for the present measurement conditions, compared to the smooth plate boundary layer. 1 Support from the European Research Council (ERC) is acknowledged. S., Sattarzadeh, S. S., Fransson, J. H. M., Talamelli, A. Phys. Rev. Lett. 109, 074501, 2012. 2 Shahinfar, 12:01PM H20.00008 Flow Through Surface Mounted Continuous Slits , A. TARIQ, M.A. ALI, Department of Mechanical & Industrial Engineering Indian Institute of Technology, Roorkee 247 667, INDIA, M. GAD-EL-HAK, Department of Mechanical & Nuclear Engineering Virginia Commonwealth University, Richmond, VA 23284, USA — Ribs are used inside certain gas-turbine blades as passive devices to enhance heat transfer. Slits in those ribs are utilized to control the primary shear layer. The role of secondary flow through a continuous slit behind a surface mounted rib is investigated herein in a rectangular duct using hotwire anemometry and particle image velocimetry. Changing the open-area-ratio and the slit’s location within the rib dominate the observed shear layer. The behavior of discrete Fourier modes of the velocity fluctuations generated by different configurations is explored. Two distinct flow mechanisms are observed in the rib’s wake. Both mechanisms are explained on the basis of large-scale spectral peak in the shear layer. The results show the successful impact of changing the open-area-ratio by manipulating the small-scale vortices at the leeward corner of the rib, which is suspected to be the potential cause of surface “hot spots” in a variety of engineering devices with heat transfer. Eventually, the size and location of the slit are seen to be an additional parameter that can be used to control the fluid flow structures behind rib turbulators. 12:14PM H20.00009 A Study of Laminar Drag Reducing Grooves , A. MOHAMMADI, JERZY M. FLORYAN, University of Western Ontario — The performance of grooves capable of reducing shear drag in laminar channel flow driven by a pressure gradient has been analyzed. Only grooves with shapes that are easy to manufacture have been considered. Four classes of grooves have been studied: triangular grooves, trapezoidal grooves, rectangular grooves and circular-segment grooves. Two types of groove placements have been considered: grooves that are cut into the surface (they can be created using material removal techniques) and grooves that are deposited on the surface (they can be created using material deposition techniques). It has been shown that the best performance is achieved when the grooves are aligned with the flow direction and are symmetric. For each class of grooves there exists an optimal groove spacing which results in the largest drag reduction. The largest drag reduction results from the use of trapezoidal grooves and the smallest results from the use of triangular grooves for the range of parameters considered in this work. Placing the same grooves on both walls increases the drag reduction by up to four times when comparing with grooves on one wall only. The predictions remain valid for any Reynolds number as long as the flow remains laminar. 12:27PM H20.00010 The momentum balance in upward turbulent channel flow laden with microbubbles1 , YOICHI MITO, SHOTA IKEDA, Kitami Institute of Technology — The influence of the addition of microbubbles as dispersed gas phase on fully-developed turbulent flow in a vertical channel in which the liquid is flowing upward with a constant pressure gradient or a constant rate has been examined by using direct numerical simulation to calculate the liquid velocities seen by the microbubbles and the point force method to consider the influence of the microbubbles on the liquid. The microbubbles are represented by solid spheres and are released from uniformly distributed point sources. The streamwise momentum balance shows that the influence of the addition of the microbubbles on the drag of the liquid flow appears as a function of volume fraction, Froude number and friction velocity that results from distribution of the microbubbles. The experimental conditions are chosen such that zero to two hundred percent of the pressure gradient of the single-phase flow is added by the addition of the microbubbles. The drag decreases by the addition of the microbubbles, whereas the frictional drag increases with the increases in the accumulation of the microbubbles on walls, which attenuates the effect of reducing drag. Changes in the liquid turbulence are not clearly seen except for what are due to the changes in the bulk Reynolds number. 1 This work was supported by JSPS KAKENHI Grant Number 26420097. Monday, November 24, 2014 10:30AM - 12:40PM Session H21 Acoustics IV: Aeroacoustics — 2010 - Justin Jaworski, Lehigh University 10:30AM H21.00001 Acoustic structures in the near-field from clustered rocket nozzles , ANDRES CANCHERO, Univ of Texas, Austin, CHARLES E. TINNEY1 , The University of Texas at Austin, NATHAN E. MURRAY2 , The University of Mississippi, JOSEPH H. RUF, NASA Marshall Space Flight Center — The plume and acoustic field produced by a cluster of two and four rocket nozzles is visualized by way of retroreflective shadowgraphy. Steady state and transient operations (startup/shutdown) were conducted in the fully-anechoic chamber and open jet facility of The University of Texas at Austin. The laboratory scale rocket nozzles comprise thrust-optimized parabolic contours, which during start-up, experience free shock separated flow, restricted shock separated flow and an end-effects regime prior to flowing full. Shadowgraphy images with synchronized surveys of the acoustic loads produced in close vicinity to the rocket clusters and wall static pressure profiles are first compared with several RANS simulations during steady operations. A Proper Orthogonal Decomposition of various regions in the shadowgraphy images is then performed to elucidate the prominent features residing in the supersonic annular flow region, the acoustic near field and the interaction zone that resides between the nozzle plumes. POD modes are used to detect propagation paths of the acoustic waves and shock cell structures in the supersonic shear layer. Spectral peak frequencies on the propagation paths are associated with the shock cell length, which are responsible for generating broadband shock noise. 1 Aerospace 2 National Engineering & Engineering Mechanics Center for Physical Acoustics 10:43AM H21.00002 Properties and Localizations of Acoustic Sources in High Speed Jet1 , PINQING KAN, JACQUES LEWALLE, ZACHARY BERGER, MATTHEW BERRY, MARK GLAUSER, Syracuse Univ, SYRACUSE UNIVERSITY TEAM — Jet noise has become one major concern for aircraft engine design in recent decades. The problem is to identify the near-field (NF) structures that produce far-field (FF) noise and develop noise control and reduction strategies. We developed an algorithm to identify the events that are responsible for NF and FF cross-correlations. Two sets of experimental data from Mach 0.6 jets are analyzed. They consist of 10kHz TRPIV measurement and pressure sampling in both near- and far-field. Several NF diagnostics (velocity, vorticity, Q criterion, etc.) are calculated to represent the 2D velocity fields. The main contributors between these NF diagnostics and FF pressure are extracted as Diagnostic-Microphone (DM) events. The NF localization of DM event clusters will be compared to the NF triangulation of MM events, which were acquired using FF signals alone. In the time-frequency domain, the events are short wave packets, distorted by ambient perturbations. As a result, the matching of DM to MM events at physical lags is particularly difficult. We will report on different algorithms using time, frequency and space information to improve the reliability of the matches. We will also relate the event localization to the NF flow fields that correspond to FF “loud” POD modes (Low et al. 2013 and Berger et al. 2014). 1 This work is supported by Spectra Energies LLC, Syracuse University MAE Department and the Glauser group at Syracuse University. 10:56AM H21.00003 End-effects-regime in full scale and lab scale rocket nozzles , RAYMUNDO ROJO, CHARLES TINNEY, University of Texas at Austin, Aerospace Engineering & Engineering Mechanics, WOUTIJN BAARS, University of Melbourne, JOSEPH RUF, NASA Marshall Space Flight Center — Modern rockets utilize a thrust-optimized parabolic-contour design for their nozzles for its high performance and reliability. However, the evolving internal flow structures within these high area ratio rocket nozzles during start up generate a powerful amount of vibro-acoustic loads that act on the launch vehicle. Modern rockets must be designed to accommodate for these heavy loads or else risk a catastrophic failure. This study quantifies a particular moment referred to as the “end-effects regime,” or the largest source of vibro-acoustic loading during start-up [Nave & Coffey, AIAA Paper 1973-1284]. Measurements from full scale ignitions are compared with aerodynamically scaled representations in a fully anechoic chamber. Laboratory scale data is then matched with both static and dynamic wall pressure measurements to capture the associating shock structures within the nozzle. The event generated during the “end-effects regime” was successfully reproduced in the both the lab-scale models, and was characterized in terms of its mean, variance and skewness, as well as the spectral properties of the signal obtained by way of time-frequency analyses. 11:09AM H21.00004 The Structure and Noise Reduction Capacity of Owl Down , JUSTIN JAWORSKI, Lehigh University, IAN CLARK, NATHAN ALEXANDER, WILLIAM DEVENPORT, Virginia Polytechnic Institute and State University, CONOR DALY, NIGEL PEAKE, University of Cambridge, STEWART GLEGG, Florida Atlantic University — Many species of owl rely on specialized plumage to reduce their self-noise levels and enable hunting in acoustic stealth. In contrast to the leading-edge comb and compliant trailing-edge fringe attributes of owls, the aeroacoustic impact of the fluffy down material on the upper wing surface remains largely speculative as a means to eliminate aerodynamic noise across a broad range of frequencies. Photographic analysis of the owl down reveals a unique forest-like structure, whereby the down fibers rise straight up from the wing surface and then bend into the flow direction to form a porous canopy, with an open area fraction of approximately 70%. Experimental measurements demonstrate that the canopy feature reduces dramatically the turbulent pressure levels on the wing surface by up to 30dB, which affects the roughness noise characteristic of the down in a manner consistent with the theory of flows over and through vegetation. Mathematical models developed for the turbulence noise generation by the down fibers and for the mixing-layer instability above the porous canopy furnish a theoretical basis to understand the influence of the down geometric structure on its self-noise signature and noise suppression characteristics. 11:22AM H21.00005 Wall Modeled Large Eddy Simulation of Airfoil Trailing Edge Noise1 , JOSEPH KOCHEEMOOLAYIL, SANJIVA LELE, Stanford Univ — Large eddy simulation (LES) of airfoil trailing edge noise has largely been restricted to low Reynolds numbers due to prohibitive computational cost. Wall modeled LES (WMLES) is a computationally cheaper alternative that makes full-scale Reynolds numbers relevant to large wind turbines accessible. A systematic investigation of trailing edge noise prediction using WMLES is conducted. Detailed comparisons are made with experimental data. The stress boundary condition from a wall model does not constrain the fluctuating velocity to vanish at the wall. This limitation has profound implications for trailing edge noise prediction. The simulation over-predicts the intensity of fluctuating wall pressure and far-field noise. An improved wall model formulation that minimizes the over-prediction of fluctuating wall pressure is proposed and carefully validated. The flow configurations chosen for the study are from the workshop on benchmark problems for airframe noise computations. The large eddy simulation database is used to examine the adequacy of scaling laws that quantify the dependence of trailing edge noise on Mach number, Reynolds number and angle of attack. Simplifying assumptions invoked in engineering approaches towards predicting trailing edge noise are critically evaluated. 1 We gratefully acknowledge financial support from GE Global Research and thank Cascade Technologies Inc. for providing access to their massivelyparallel large eddy simulation framework. 11:35AM H21.00006 Direct numerical simulation and reduced-order modeling of the soundinduced flow through a cavity-backed circular under a turbulent boundary layer , QI ZHANG, DANIEL BODONY, University of Illinois at Urbana-Champaign — Commercial jet aircraft generate undesirable noise from several sources, with the engines being the most dominant sources at take-off and major contributors at all other stages of flight. Acoustic liners, which are perforated sheets of metal or composite mounted within the engine, have been an effective means of reducing internal engine noise from the fan, compressor, combustor, and turbine but their performance suffers when subjected to a turbulent grazing flow or to high-amplitude incident sound due to poorly understood interactions between the liner orifices and the exterior flow. Through the use of direct numerical simulations, the flow-orifice interaction is examined numerically, quantified, and modeled over a range of conditions that includes current and envisioned uses of acoustic liners and with detail that exceeds experimental capabilities. A new time-domain model of acoustic liners is developed that extends currently-available reduced-order models to more complex flow conditions but is still efficient for use at the design stage. 11:48AM H21.00007 Computation of noise from separated flows using large eddy simulation1 , ZANE NITZKORSKI, KRISHNAN MAHESH, University of Minnesota — We investigate noise production from turbulent flow over bluff bodies using the Ffowcs-Williams and Hawkings (FW-H) acoustic analogy. We propose a dynamic end cap methodology to account for volumetric contributions to the far-field sound within the context of the FW-H acoustic analogy. The quadrupole source terms are correlated over multiple planes to obtain a convection velocity that is then used to determine a corrective convective flux at the FW-H porous surface. The proposed approach is first demonstrated for a convecting potential vortex. It is then applied to compute the noise from a cylinder at ReD =89k, and a 45 degree beveled trailing edge at Rec =1.9M. We compare our results for base flow and acoustic data to available computations and experiments. We demonstrate insensitivity of the end cap correction approach to end plane location and spacing, discuss the effect of dynamic convection velocity, and show better performance than commonly used end cap corrections. Finally, we discuss some physical mechanisms that generate the far-field noise. 1 Office of Naval Research 12:01PM H21.00008 Cross-stream ejection in the inter-wheel region of aircraft landing gears1 , PHILIP MCCARTHY, ALIS EKMEKCI, University of Toronto — The reduction of aircraft noise is an important challenge currently faced by aircraft manufacturers. During approach and landing, the landing gears contribute a significant proportion of the aircraft generated noise. It is therefore critical that the key noise sources be identified and understood in order for effective mitigation methods to be developed. For a simplified two-wheel nose landing gear, a strong cross stream flow ejection phenomena has been observed to occur in the inter-wheel region in presence of wheel wells. The location and orientation of these flow ejections causes highly unsteady, three dimensional flow between the wheels that may impinge on other landing gear components, thereby potentially acting as a significant noise generator. The effects of changing the inter-wheel geometry (inter-wheel spacing, the wheel well depth and main strut geometry) upon the cross-stream ejection behaviour has been experimentally investigated using both qualitative flow visualisation and quantitative PIV techniques. A summary of the key results will be presented for the three main geometrical parameters under examination and the application of these findings to real life landing gears will be discussed. 1 Thanks to Messier-Bugatti-Dowty and NSERC for their support for this project. 12:14PM H21.00009 Sound-turbulence interaction in transonic boundary layers , LUDOVIC LELOSTEC, Department of Aeronautics and Astronautics, Stanford, CA, 94305, CARLO SCALO, Center of Turbulence Research, Stanford, CA, 94305, SANJIVA LELE, Department of Aeronautics ans Astronautics, Stanford, CA, 94305 — Acoustic wave scattering in a transonic boundary layer is investigated through a novel approach. Instead of simulating directly the interaction of an incoming oblique acoustic wave with a turbulent boundary layer, suitable Dirichlet conditions are imposed at the wall to reproduce only the reflected wave resulting from the interaction of the incident wave with the boundary layer. The method is first validated using the laminar boundary layer profiles in a parallel flow approximation. For this scattering problem an exact inviscid solution can be found in the frequency domain which requires numerical solution of an ODE. The Dirichlet conditions are imposed in a high-fidelity unstructured compressible flow solver for Large Eddy Simulation (LES), CharLESx . The acoustic field of the reflected wave is then solved and the interaction between the boundary layer and sound scattering can be studied. 12:27PM H21.00010 An exact and dual-consistent formulation for high-order discretization of the compressible viscous flow equations , RAMANATHAN VISHNAMPET, DANIEL BODONY, JONATHAN FREUND, University of Illinois at Urbana-Champaign — Finite-difference operators satisfying a summation-by-parts property enable discretization of PDEs such that the adjoint of the discretization is consistent with the continuous-adjoint equation. The advantages of this include smooth discrete-adjoint fields that converge with mesh refinement and superconvergence of linear functionals. We present a high-order dual-consistent discretization of the compressible flow equations with temperature-dependent viscosity and Fourier heat conduction in generalized curvilinear coordinates. We demonstrate dual-consistency for aeroacoustic control of a mixing layer by verifying superconvergence and show that the accuracy of the gradient is only limited by computing precision. We anticipate dual-consistency to play a key role in compressible turbulence control, for which the continuous-adjoint method, despite being robust, introduces adjoint-field errors that grow exponentially. Our dual-consistent formulation can leverage this robustness, while simultaneously providing an exact sensitivity gradient. We also present a strategy for extending dual-consistency to temporal discretization and show that it leads to implicit multi-stage schemes. Our formulation readily extends to multi-block grids through penalty-like enforcement of interface conditions. Monday, November 24, 2014 10:30AM - 12:40PM Session H22 Instability: Richtmyer-Meshkov I — 2012 - Praveen Ramaprabhu, University of North Carolina at Charlotte 10:30AM H22.00001 Numerical simulations of the Single-mode, Doubly-shocked RichtmyerMeshkov (RM) Instability , VARAD KARKHANIS, PRAVEEN RAMAPRABHU, University of North Carolina at Charlotte — We describe results from numerical simulations of a single-mode, doubly-shocked material interface between gases of different densities. The time interval between the shocks was varied to observe interfacial growth due to Richtmyer-Meshkov Instability initialized with different amplitudes. The simulations were performed with low and high density ratio fluids (A = 0.15 and A = -0.99), where the latter case is relevant to ejecta formation. We compare the growth rates from our simulations after the first and second shocks with linear, nonlinear [1] and ejecta models [2,3]. In the heavy to light configuration (A = -0.99), we observe two consecutive phase inversions following each shock. We have also investigated the effect of variations in the initial interface perturbation to include sine, chevron, sawtooth, and square-wave form, and find our results to be of relevance to machined target experiments. [1] Guy Dimonte and P. Ramaprabhu, Phys. Fluids 22, 014104 (2010). [2] W. T. Buttler et al., J. Fluid Mech., 703 (2012). [3] Guy Dimonte et al., J. Appl. Phys. 113, 024905 (2013). 10:43AM H22.00002 Numerical Simulations of the single-mode Richtmyer-Meshkov instability in a spherically convergent geometry , ISMAEL DJIBRILLA BOUREIMA, PRAVEEN RAMAPRABHU, University of North Carolina at Charlotte, LABAKANTA MANDAL, Budge Budge Institute of Technology, MANORANJAN KHAN, Jadavpur University — We investigate the linear and nonlinear development of the single-mode Richtmyer-Meshkov instability in a spherically convergent geometry. The three-dimensional simulations were performed using the astrophysical FLASH code [1], with a resolution of 2560 x 512 x 512 in the radial, azimuthal and polar p directions. The perturbations at the interface 2l+1 between an outer shell of light fluid and an inner shell of heavy fluid were specified according to η (t, θ) = al (t) Pl cos (θ) where al (t) is the angular 2 displacement, and Pl is the Legendre polynomial of mode number l. A third, fictitious layer formed an external shell to sustain the incident shock. The modification of the linear growth rate due to convergence was compared with the linear model of [2], while the growth rates during the nonlinear stage were verified against a potential flow model described in [3]. The validity of the above results is examined at different Atwood numbers. [1] Fryxell, B. et al., Astrophys. J. Suppl., 131 (1), 2000, 273. [2] Mikaelian, Karnig O., Phys. Rev. A 42 (6), 1990, 3400. [3] Mandal, L. et al., arXiv preprint arXiv: 1109.5363 (2011). 10:56AM H22.00003 Numerical simulations of a chemically reacting Richtmyer-Meshkov turbulent mixing layer , HILDA VARSHOCHI, NITESH ATTAL, PRAVEEN RAMAPRABHU, Department of Mechanical Engineering and Engineering Sciences, University of North Carolina at Charlotte — We report on results from detailed numerical simulations that capture the evolution through the RichtmyerMeshkov instability of a multi-mode interface that initially separates a fuel (H2 ) and a corresponding oxidizer (O2 ). The three-dimensional simulations were carried out at a resolution of 512 x 512 x 3072 using a modified version of the FLASH code, capable of handling detailed H2 -O2 combustion chemistry [1], temperature-dependent equation of state, and temperature-dependent molecular transport properties. The perturbation interface was initialized with “alphagroup” [2] type perturbations, and impacted by a Mach 1.2 incident shock travelling from the light (H2 ) to heavy (O2 ) fluid. We track several quantities through the linear, non-linear and turbulent stages of evolution, and make comparisons with the corresponding non-reacting flowfield from a separate set of simulations. The turbulent mixing layer is also subjected to reshock, which dramatically increases the combustion efficiency at the interface. [1] Attal, N. et al., submitted to Computers and Fluids for review. [2] Dimonte, G. et al., Phys. Fluids, 16, p. 1668, 2004. 11:09AM H22.00004 Energy dynamics in the Richtmyer-Meshkov instability induced turbulent mixing flow , ZUOLI XIAO, HAN LIU, State Key Laboratory for Turbulence and Complex Systems, College of Engineering,Peking University — The Richtmyer-Meshkov instability (RMI) induced turbulent mixing flow in a shock tube is numerically investigated by using direct numerical simulation based on an effective in-house high-order turbulence solver (HOTS). The energy transfer and transport characteristics are studied both before and after re-shock. The celebrated Kolmogorov -5/3 spectrum can be observed in a long inertial subrange during the development of the turbulent mixing zone (TMZ). Insight is taken into the underlying mechanism by evaluating the energy-budget equations. A posteriori analysis of the influence of subgrid scales on resolved motions also gives a consistent picture of energy transfer in the RMI-induced turbulent mixing. Moreover, the kinetic energy cascade in the TMZ is discussed by using Favre filtering approach in physical space. A nonlinear vortex-stretching model for the subgrid-scale stress serves to explain the underlying mechanism of the energy cascade in the RMI-induced turbulence. 11:22AM H22.00005 Linear Simulations of the MHD Richtmyer-Meshkov Instability in Cylindrical Geometry , ABEER BAKSH, RAVI SAMTANEY, KAUST, VINCENT WHEATLEY, University of Queensland — Numerical simulations and analysis indicate that the Richtmyer-Meshkov instability (RMI) is suppressed in ideal magnetohydrodynamics (MHD) in Cartesian slab geometry. Motivated by the presence of hydrodynamic intstabilities in inertial confinement fusion and suppression by means of a magnetic field, we investigate the RMI via linear MHD simulations in cylindrical geometry. The physical setup is that of a Chisnell-type converging shock interacting with a density interface with either axial or azimuthal (2D) or a combination of both (3D) perturbations. The linear stability is examined in the context of an initial value problem (with a time-varying base state) wherein the linearized ideal MHD equations are solved by an extension of the numerical method proposed by Samtaney (J. Comput. Phys. 2009). Linear simulations in the absence of a magnetic field, indicate that RMI growth rate during the early time period similar to that observed in Cartesian geometry. However, this RMI phase is short-lived and followed by a Rayleigh-Taylor growth phase with an accompanied exponential increase in the perturbation amplitude. 2p We examine several strengths of the magnetic field (characterized by β = B 2 ) and observe a significant suppression of the instability for β ≈ 2. 11:35AM H22.00006 The magnetohydrodynamic Richtmyer-Meshkov instability in twodimensional implosions , WOUTER MOSTERT, VINCENT WHEATLEY, The University of Queensland, DALE PULLIN, California Institute of Technology, RAVI SAMTANEY, King Abdullah University of Science and Technology — We present numerical results showing the behaviour of the magnetohydrodynamic Richtmyer-Meshkov instability in two-dimensional implosions in the presence of an externally applied seed magnetic field. An initially perturbed cylindrical density interface is accelerated from the outside by a set of imploding magnetohydrodynamic shocks, themselves generated by a cylindrical Riemann problem. We characterize the process of shock refraction with the density interface and examine the subsequent growth of the interface perturbations, comparing with the zero-field case. We test two candidate seed magnetic field configurations: a uniform-strength, unidirectional field; and a field with a saddle point at the domain origin. Both field configurations show suppression of interface perturbation growth, with the latter exhibiting the least asymmetry in the implosion. 11:48AM H22.00007 Instability evolution in shock-accelerated inclined heavy gas cylinder1 , DELL OLMSTEAD, PATRICK WAYNE, PETER VOROBIEFF, DANIEL DAVIS, C. RANDALL TRUMAN, The University of New Mexico — A heavy gas cylinder interacts with a normal or oblique shockwave at Mach numbers M ranging from 1.13 to 2.0. The angle between the shock front and cylinder axis is varied between 0 and 30◦ , while the Atwood numbers A range from 0.25 (SF6 -N2 mix) to 0.67 (pure SF6 ). The evolution of the column is imaged in two perpendicular planes with Planar Laser Induced Fluorescence (PLIF). For oblique shock interactions, the nature of the flow is fully three-dimensional, with several instabilities developing in separate directions. In the plane that captures a cross-section of the column, Richtmyer-Meshkov instability (RMI) leads to formation of a pair of counter-rotating vortex columns. A uniform scaling appears to govern the primary instability growth in this plane across the M and A ranges, when the length scale is normalized by a product of the minimum streamwise scale after shock compression and M0.5 . In the vertical plane through the column, Kelvin-Helmholtz vortices form with regular spacing along the column. The dominant wavelength of the structures in the vertical plane also appears to scale with the minimum compressed streamwise length. 1 This research is supported by the US DOE National Nuclear Security Administration (NNSA) grant DE-NA0002220. 12:01PM H22.00008 Droplet tracer characterization in shock-driven multiphase flow1 , FRANCISCO VIGIL, MIQUELA TRUJILLO, PETER VOROBIEFF, C. RANDALL TRUMAN, The University of New Mexico — Small glycol droplets have long been introduced into shock-accelerated gas as a tracer, to track the evolution of Richtmyer-Meshkov instability (RMI). However, it was observed that droplets are not passive tracers when shock-accelerated - to the extent that their introduction itself can lead to vortex formation. Because of the complex interplay between the droplets and surrounding gas, it is imperative to know the droplet size and population density. The absence of this knowledge has led to differences between results from numerical models, Planar Laser-Induced Fluorescence (PLIF) measurements, and Mie scattering observations. To gain a better understanding of the droplet velocity and inertial flow fields, a more involved study of droplet sizing is required. A Malvern laser diagnostic system is used to determine the sizes of the glycol droplets seeded into the flow. A series of tests are performed to analyze differences in glycol droplet size and population distribution that result from changing gaseous mediums in the test section. These measurements facilitate better quantification of the velocity fields in shock accelerated flow and improve interpretation of results from Mie scattering. 1 This research is supported by the US DOE National Nuclear Security Administration (NNSA) grant DE-NA0002220. 12:14PM H22.00009 Vortex formation in oblique shock interaction with a heavy gas column1 , PATRICK WAYNE, DELL OLMSTEAD, C. RANDALL TRUMAN, PETER VOROBIEFF, The University of New Mexico, SANJAY KUMAR, University of Texas - Brownsville — In an oblique shockwave interaction with a column of heavy gas, we observe both the expected counter-rotating vortex pairs (same as caused by normal shockwaves) and periodic co-rotating vortices that vary with Mach number. We study the effects of oblique shock interaction with a column of acetone-infused sulfur-hexafluoride (SF6 ) gas. Visualization of the shock-accelerated gas column is accomplished via Planar Laser-Induced Fluorescence (PLIF) imaging. The shock tube itself is inclined at a 30◦ angle, while the initial conditions (ICs) are introduced into the test section vertically. Because of the inclined angle, the normal shock propagates down the shock tube and impacts the ICs at a 30◦ down-angle, producing an oblique shock. Vertical plane PLIF images reveal vorticity deposition between the SF6 column and the surrounding air leading to Kelvin-Helmholtz instability. The evolving vortices cascade down the entire vertical length of the gas column, and interact with the counter-rotating vortex structures along the column. The most interesting aspect of this discovery is that these small-scale instabilities exhibit periodic behavior and, according to preliminary data, this behavior is Mach number dependent. 1 This research is supported by the US DOE National Nuclear Security Administration (NNSA) grant DE-NA0002220. 12:27PM H22.00010 Transition in Hypersonic Boundary Layers: Role of Dilatational Waves , CHUANHONG ZHANG, YIDING ZHU, QING TANG, HUIJING YUAN, JIEZHI WU, SHIYI CHEN, CUNBIAO LEE, State Key Laboratory of Turbulence and Complex Systems, Collaborative Innovation Center of Advanced Aero-Engine, Peking University, MOHAMED GAD-EL-HAK, Department of Mechanical & Nuclear Engineering, Virginia Commonwealth University — Transition and turbulence production in a hypersonic boundary layer is investigated in a Mach 6 quiet wind tunnel using Rayleigh-scattering visualization, fast-response pressure measurements, and particle image velocimetry. A previously undiscovered unusual behavior of the second instability mode is noticed. Very high frequency dilatational waves are observed to grow rapidly followed by very fast annihilation. The second instability mode is a key modulator of the hypersonic laminar-to-turbulence transition, and the bulk viscosity plays an important role in that dynamical process. At its peak, the second mode strongly interacts with the first instability mode to directly promote a rapid growth of the latter and immediate transition to turbulence. This interaction can be explained by a nonlinear coupling of vorticity and dilatation in the interior of the boundary layer, combined with a viscous linear coupling at the wall. Our study of transition in hypersonic flows suggests that more attention should be given to the inviscid dilatational waves and their coupling with transverse vortical structures. Monday, November 24, 2014 10:30AM - 12:40PM — Session H23 Geophysical Fluid Dynamics: Eddy/Wave-Mean Flow Interactions 2001 - Pascale Lelong, Northwest Research Associates 10:30AM H23.00001 Near-inertial waves within an anticyclonic eddy in the Mediterranean Sea: Observations and numerical simulations1 , PASCALE LELONG, Northwest Res Assoc, PASCALE BOURUET-AUBERTOT, YANNIS CUYPERS, LOCEAN, Université Pierre et Marie Curie, Paris, CYPRUS EDDY MODELING COLLABORATION — One of the objectives of the BOUM field experiment, conducted in the Mediterranean Sea during the Summer of 2008, was to investigate the impact of submesoscale ocean dynamics on biogeochemical cycles. Analysis of data collected in the permanent, warm-core, anticyclonic Cyprus eddy provides a case study for near-inertial wave generation and turbulence in the presence of an eddy. Observations reveal the presence of near-inertial oscillations over the entire profile, from the mixed layer to below the base of the eddy. We present the results of a parallel LES numerical study with a Boussinesq pseudo-spectral code which was designed to explain the observed near-inertial signal. Two generation mechanisms are discussed: (i) inertial pumping at the base of the mixed layer following a wind event and (ii) adjustment of the eddy with possible trapping at the base of the eddy. Our numerical study confirms the role of anticyclonic eddies in influencing the propagation of wind-driven inertial oscillations into the thermocline. 1 NSF-Physical Oceanography 10:43AM H23.00002 Internal wave generation by a distributed vortex , SURUPA SHAW, JOHN MCHUGH, University of New Hampshire — Internal wave generation in a continuously stratified fluid by a mature vortex pair and by a distributed line vortex is considered using direct numerical simulations. For large Froude number, the distributed vorticity for the line vortex quickly rolls up and forms a vortex pair, approximately matching the case that is initiated as a vortex pair. However for small Froude number, both cases disintegrate into internal waves. Recent results show that both cases exhibit a strong vertical oscillation with a frequency that depends on the buoyancy frequency N . The wave generation causes the initial energy to spread in the radial direction, however after several oscillations at the buoyancy frequency the spreading is slow and the overall size of the structures become approximately constant. The internal wave generation is identified by distinct radial structures in contours of energy flux. 10:56AM H23.00003 Anisotropic mesoscale eddy transport in ocean general circulation models , SCOTT RECKINGER, BAYLOR FOX-KEMPER, Brown University, SCOTT BACHMAN, University of Cambridge, FRANK BRYAN, JOHN DENNIS, GOKHAN DANABASOGLU, National Center for Atmospheric Research — In modern climate models, the effects of oceanic mesoscale eddies are introduced by relating subgrid eddy fluxes to the resolved gradients of buoyancy or other tracers, where the proportionality is, in general, governed by an eddy transport tensor. The symmetric part of the tensor, which represents the diffusive effects of mesoscale eddies, is universally treated isotropically. However, the diffusive processes that the parameterization approximates, such as shear dispersion and potential vorticity barriers, typically have strongly anisotropic characteristics. Generalizing the eddy diffusivity tensor for anisotropy extends the number of parameters from one to three: major diffusivity, minor diffusivity, and alignment. The Community Earth System Model (CESM) with the anisotropic eddy parameterization is used to test various choices for the parameters, which are motivated by observations and the eddy transport tensor diagnosed from high resolution simulations. Simply setting the ratio of major to minor diffusivities to a value of five globally, while aligning the major axis along the flow direction, improves biogeochemical tracer ventilation and reduces temperature and salinity biases. These effects can be improved by parameterizing the oceanic anisotropic transport mechanisms. 11:09AM H23.00004 Effects of Submesoscale Turbulence on Tracer Evolution in the Oceanic Mixed Layer , KATHERINE SMITH, SPENCER ALEXANDER, University of Colorado - Boulder, LUKE VAN ROEKEL, Northland College, BAYLOR FOX-KEMPER, Brown Univeristy, PETER HAMLINGTON, University of Colorado - Boulder — Ocean tracers such as CO2 , nutrients, and plankton evolve mainly in the mixed layer where light and air-sea gas exchange occur. It is known from prior studies there can be substantial heterogeneity in tracer distributions due to vertical and horizontal turbulent mixing across a range of scales. The contribution of submesoscale turbulence to these distributions is not entirely understood, particularly in the sub-kilometer range where both large-scale, nearly 2D and small-scale, 3D turbulence are active, resulting in dynamical complexity from which heterogeneity can arise. In this talk, results from large eddy simulations of a large temperature front evolving are used to examine effects of multi-scale turbulence on idealized tracer distributions from scales 20km to 5m. Simulations include the effect of Langmuir turbulence by solving the wave-averaged Boussinesq equations with an imposed Stokes drift velocity. Tracers with different source and boundary conditions are examined to understand the role of both small-scale, near-surface vertical mixing and larger-scale upwelling motions typically associated with submesoscale eddies. Tracer evolution is characterized using spectra, multi-scale fluxes, and probability distribution functions, and implications of the results are outlined. 11:22AM H23.00005 Estimates of Lagrangian particle transport by wave groups: forward transport by Stokes drift and backward transport by the return flow , TON S. VAN DEN BREMER, PAUL H. TAYLOR, University of Oxford — Although the literature has examined Stokes drift, the net Lagrangian transport by particles due to of surface gravity waves, in great detail, the motion of fluid particles transported by surface gravity wave groups has received considerably less attention. In practice nevertheless, the wave field on the open sea often has a group-like structure. The motion of particles is different, as particles at sufficient depth are transported backwards by the Eulerian return current that was first described by Longuet-Higgins & Stewart (1962) and forms an inseparable counterpart of Stokes drift for wave groups ensuring the (irrotational) mass balance holds. We use WKB theory to study the variation of the Lagrangian transport by the return current with depth distinguishing two-dimensional seas, three-dimensional seas, infinite depth and finite depth. We then provide dimensional estimates of the net horizontal Lagrangian transport by the Stokes drift on the one hand and the return flow on the other hand for realistic sea states in all four cases. Finally we propose a simple scaling relationship for the transition depth: the depth above which Lagrangian particles are transported forwards by the Stokes drift and below which such particles are transported backwards by the return current. 11:35AM H23.00006 Effects of swell on dispersion of oil plumes within the ocean mixed layer1 , BICHENG CHEN, Department of Meteorology, Pennsylvania State University, DI YANG, Department of Mechanical Engineering, Johns Hopkins University, MARCELO CHAMECKI, Department of Meteorology, Pennsylvania State University, CHARLES MENEVEAU, Department of Mechanical Engineering, Johns Hopkins University — Oil plumes from deep-water blowouts rise through the ocean and reach the ocean mixed layer (OML), where dispersion is strongly affected by Langmuir turbulence generated by interactions between the wind forcing and the wave regime. The wind-driven wave field is approximately aligned with wind direction. However, the swell wave can have an arbitrary orientation relative to the local wind. We used large-eddy simulation (LES) to study the influences of the misalignment between wind and wave field on the transport and dispersion of oil plumes in the OML. Results show that the plume response to these forcing is strongly dependent on the size of the oil droplets. For the large oil droplets, the center line of the time-averaged surface plume tends to follow the mean surface current direction; for small droplets, the change of orientation of center line with wave direction is smaller than that of large droplets. Vertical eddy diffusivity calculated from LES data is compared to closures currently used in ocean models (such as the KPP model employed in HYCOM). The magnitude of the eddy diffusivity changes by a factor of two as the misalignment between swell and wind changes, and it is typically much larger than predicted by KPP. 1 This study is supported by a Gulf of Mexico Research Initiative research grant. 11:48AM H23.00007 Oil droplet plume evolution in Langmuir turbulence: a Large Eddy Simulation study1 , DI YANG, Department of Mechanical Engineering, Johns Hopkins University, BICHENG CHEN, MARCELO CHAMECKI, Department of Meteorology, Pennsylvania State University, CHARLES MENEVEAU, Department of Mechanical Engineering, Johns Hopkins University — When the oil plumes from deep water blowouts reach the ocean mixed layer (OML), their fates on the sea surface are highly affected by the interactions with wind and wave-generated Langmuir turbulence in the OML. In this study, we use large eddy simulations (LES) to quantify the complex oil dispersion phenomena. We find that although the instantaneous surface oil slick patterns are very complex, the time-averaged surface oil plume can be parameterized as a Gaussian-type plume. The centerline of the surface plume is inclined clockwise (in the Northern Hemisphere) with respect to the wind and wave direction due to Ekman transport. The initial width of the mean surface plume and the inclination angle increase as the droplet size decreases. The surface plume width grows downstream, with a growth rate that varies non-monotonically with oil droplet size. Using LES data, we evaluate the eddy viscosity and eddy diffusivity following the K-profile parameterization (KPP) framework. We also evaluate stress-strain misalignments caused by Stokes drift and evaluate means of parameterizing these effects. Improvements to the KPP model will be discussed. 1 This study is supported by a Gulf of Mexico Research Initiative research grant. 12:01PM H23.00008 Langmuir circulation in shallow water waves , W.R.C. PHILLIPS, Swinburne Univ of Tech, ALBERT DAI, National Taiwan University — The instability of shallow water waves on a moderate shear to Langmuir circulation (LC) is considered. In such instances the shear can significantly affect the drift giving rise to profiles markedly different from the simple Stokes drift. Since drift and shear are instrumental in the instability to LC, of key interest is how that variation in turn affects onset to LC. The initial value problem describing the wave-mean flow interaction is crafted from scratch and includes a consistent set of free-surface boundary conditions. The problem necessitates that Navier Stokes be employed side by side with a set of mean-field equations; these are seen to reduce to the well known CL-equations, albeit with different time and velocity scales. Typical shear driven and pressure driven flows are considered. Shear driven flow is found to be destabilizing while pressure driven are stabilizing to LC. It is further found that multiple layers, as opposed to a single layer, of LC can form, with the most intense circulations at the ocean floor. LC can also extend into a region of flow beyond which instability applies thus deepening the mixed layer. Two preferred spacings occur, one closely in accord with observation for small aspect ratio LC. 12:14PM H23.00009 LES of full-depth Langmuir circulation with surface cooling , RACHEL WALKER, ANDRES E. TEJADA-MARTINEZ, University of South Florida, CHESTER E. GROSCH, Old Dominion University — Results are presented from large-eddy simulations (LES) of full-depth Langmuir circulation (LC) in the presence of surface cooling in a domain representative of the shallow coastal ocean. LC consists of counter-rotating vortices aligned roughly in the direction of the wind and generated by the interaction of the wind-driven shear with the Stokes drift velocity induced by surface gravity waves. In LES of open channel flow (without LC), surface cooling has been found to lead to the development of full-depth convection cells similar in structure to LC. As such, in the current simulations unstable stratification is imposed by a constant surface cooling flux and an adiabatic bottom wall to assess the impact of cooling-induced buoyancy on the strength of the wind and wave-driven LC. The surface cooling flux will be quantified by the value of the Rayleigh number, representative of surface buoyancy forcing relative to wind shear forcing. The impact of the convection on LC will be assessed by analysis of mean velocity, root mean square of velocity, and budgets of Reynolds stress components. It is intended that results may assist in determining the dominant mechanism in large-scale cell structure development when both LC and surface cooling are present. 12:27PM H23.00010 DNS of scalar transport across a wind-driven air-water interface1 , AMINE HAFSI, ANDRES TEJADA-MARTINEZ, Univ of South Florida, FABRICE VERON, Univ of Delaware — When wind blows over an initially quiescent air-sea interface, it first generates short capillary waves which in time coexist with longer waves as part of a broad spectrum of waves. The interaction between the wind-driven waves and shear current on the waterside leads to Langmuir turbulence characterized by Langmuir circulation (LC) consisting of counter rotating vortices roughly aligned in the direction of the wind. The typical length scale of LC ranges from several centimeters when short capillary waves first appear up to tens of meters when the spectrum of waves broadens. Results are presented from direct numerical simulation (DNS) of a coupled air-water interface driven by an air flow with free stream speed of 5 m/s. The evolution of the air-water interface starting from rest and the accompanying development of centimeter-scale Langmuir turbulence on the waterside during the first 20 seconds of simulation are investigated. Emphasis is placed on the impact of the Langmuir turbulence on scalar transfer from the airside to the waterside, in particular the transfer velocity which is a measure of scalar transfer efficiency. Simulations are made with a finite volume discretization employing the volume of fluid method for interface tracking. 1 Support from National Science Foundation is gratefully acknowledged. Monday, November 24, 2014 10:30AM - 12:40PM Session H24 Granular Flows: Locomotion and Drag — 2003 - Ho-Young Kim, Seoul National University 10:30AM H24.00001 Switch of states of a short chain in response to vibrations , YU-CEN SUN, JUNG-REN HUANG, National Taiwan Normal University, Institute of Physics, CHIAO-YU TAO, JIH-CHIANG TSAI, Academia Sinica, Institute of Physics — We study experimentally the dynamics of a short ball chain confined in a quasi-2D vertical channel under different vibrational strengths(VS). For a substantial range of VS, the chain maintains period-1 bouncing with the channel, but also undergoes transitions from a uniform response to various states of excitations as VS increases. In the transitional zone, we find that the unexcited and excited states exhibit bistability and switch spontaneously at fixed values of VS. This coexistence of different states explains the stocastic switch of ratcheting behaviors we reported previously in Phys. Rev. Lett. 112, 058001 (2014) where a spatial gradient of vibration is imposed. 10:43AM H24.00002 Photoelastic gelatin spheres for investigation of locomotion in granular media , SEYED AMIR MIRBAGHERI, ERICSON CENICEROS, MEHDI JABBARZADEH, ZEPHYR MCCORMICK, HENRY FU, University of Nevada, Reno — We describe a force measurement method in granular media which uses highly-sensitive photoelastic gelatin spheres and its application to measuring forces exerted as animals burrow through granular media. The method is applicable to both freshwater and marine organisms. We fabricate sensitively photoelastic gelatin spheres and describe a calibration method which relates forces applied to gelatin spheres with photoelastic signal. We show that photoelastic gelatin spheres can detect forces as small as 1 microNewton, and quantitatively measure forces with up to 60 microNewton precision, a two order of magnitude improvement compared to methods using plastic disks. Gelatin spheres can be fabricated with a range of sizes to investigate a variety of granular media. Finally, we used the calibrated gelatin spheres in a proof-of-principle experiment to measure forces during earthworm locomotion. 10:56AM H24.00003 A Nondimensional Model for Axial Digging in Granular Materials , HAO LI, DAWN WENDELL, ANETTE HOSOI, PAWEL ZIMOCH, MIT — We investigate the mechanics of thin diggers in a packing of granular materials. Experiments are conducted with diggers of varying thickness and force-depth data is recorded. Accounting for buckling and drag force, we propose a continuum model thats predicts an optimal digger thickness that maximizes digging depth. Model predictions are compared to experimental data. This model is refined when digger thickness approaches the grain scale to account for stochasticity. 11:09AM H24.00004 Helical swimming in granular media , ROBERTO ZENIT, ELSA DE LA CALLEJA, FRANCISCO GODINEZ, Universidad Naclonal Autonoma de Mexico — In nature, many organisms are capable of swimming in sand by performing an undulatory motion. Recently, Goldman and collaborators showed that a modification of the low-Re number resistive force theory can be used to explain the phenomena. In this investigation we use a self-propelled magnetically-driven swimmer with a helical tail to further investigate the swimming performance in sand. We successfully produced devices that effectively swam in sand the the rotating action of a helical tail. We measured the swimming speed for a range of rotational speeds and tail geometries. Preliminary results will shown and discussed. 11:22AM H24.00005 Legless locomotion in lattices , PERRIN SCHIEBEL, DANIEL I. GOLDMAN, Georgia Institute of Technology — Little is known about interactions between an animal body and complex terrestrial terrain like sand and boulders during legless, undulatory travel (e.g. snake locomotion). We study the locomotor performance of Mojave shovel-nosed snakes (Chionactis occipitalis, ≈ 35 cm long) using a simplified model of heterogeneous terrain: symmetric lattices of obstacles. To quantify performance we measure mean forward speed and slip angle, βs , defined as the angle between the instantaneous velocity and tangent vectors at each point on the body. We find that below a critical peg density the presence of granular media results in high speed (≈ 60 cm/s), low average slip (βs ≈ 6◦ ) snake performance as compared to movement in the same peg densities on hard ground (≈ 25 cm/s and βs ≈ 15◦ ). Above this peg density, performance on granular and hard substrates converges. Speed on granular media decreases with increasing peg density to that of the speed on hard ground, while speed on hard ground remains constant. Conversely, βs on hard ground trends toward that on granular media as obstacle density increases. 11:35AM H24.00006 Self-burrowing seeds: drag reduction in granular media , WONJONG JUNG, SUNG MOK CHOI, Seoul National University, WONJUNG KIM, Sogang University, HO-YOUNG KIM, Seoul National University — We present the results of a combined experimental and theoretical investigation of drag reduction of self-burrowing seeds in granular media. In response to environmental changes in humidity, the awn (a tail-like appendage of seed) of Pelargonium carnosum exhibits coiling-uncoiling deformation which induces the thrust and rotary motions of the head of the seed against the surface of the soil. Using various sizes of glass beads that mimic the granular soil, we measure the thrust forces required for the seed of Pelargonium carnosum to penetrate into granular media with and without rotation. Our quantitative measurements show that the rotation of the seed remarkably reduces the granular drag as compared to the drag against the non-spinning seed. This leads us to conclude that the hygroscopically active awns of Pelargonium carnosum enables its seed to dig into the relatively coarse granular soils. 11:48AM H24.00007 A Model for Solid-Solid drag in Bidisperse Gas-solid Flows1 , ERIC MURPHY, SHANKAR SUBRAMANIAM, Iowa State University — Computational models for gas-solid mixtures often require closures for interphase momentum and energy transfer. One of the most important interactions for polydisperse systems is a so-called solid-solid drag, i.e. the momentum transfer between different particulate phases traveling at different mean velocities. Modeling of these, and additional terms has been a focus of the granular physics community for nearly three decades and is no easy task. Flows of bidisperse particles are often high Mach number, Ma>>1. As a result, many theories developed for low Mach number applications using the Chapman-Enskog(CE) theory are not strictly applicable. Still, many other analytic moment methods did not properly couple granular temperature and slip between particulate phases. We have developed a moment theory for the slip and temperature evolution employing the pseudo-Liouville operator technique, which correctly accounts for the coupling between phasic slip and temperatures. The theory is compared with other existing moment models for solid-solid drag. It is found that the drag model is a weighted sum of terms arising in both (CE) and existing moment theories. Additionally, new phase specific temperature evolution terms are obtained that shed light on phenomena such as non-equipartition of energy in bidisperse granular gases. Lastly, we explore some of the segregation behavior implied by the model for homogeneous gas-solid flows with bidisperse particles. 1 This work was supported through DOE award number DE-FE0007260 and NSF grants CMMI 0927660 and CBET 1134500. 12:01PM H24.00008 Giant drag reduction due to interstitial air in sand , DEVARAJ VAN DER MEER, University of Twente, The Netherlands, TESS HOMAN, Laboratoire de Physique, ENS Lyon, France — When an object impacts onto a bed of very loose, fine sand, the drag it experiences depends on the ambient pressure in a surprising way: Drag is found to increase significantly with decreasing pressure. We use a modified penetrometer experiment to investigate this effect and directly measure the drag on a sphere as a function of both velocity and pressure. We observe a drag reduction of over 90% and trace this effect back to the presence of air in the pores between the sand grains. Finally, we construct a model based on the modification of grain-grain interactions that is in full quantitative agreement with the experiments. 12:14PM H24.00009 Modulation of orthogonal body waves enables versatile maneuverability in limbless locomotion1 , DANIEL GOLDMAN, Georgia Tech, A COLLABORATION2 — Limbless organisms can create different motions by modulating axial undulations that pass through their bodies. Sidewinding snakes generate horizontal and vertical waves, with a phase offset of π/2, resulting in posteriorly-propagating alternating regions of static contact with the substrate and elevated motion, resulting in a “stepping” motion of body segments. We have discovered that sidewinder rattlesnakes (it Crotalus cerastes) are quite maneuverable and possess at least two turning methods: “differential turning” and “reversal turning.” In differential turning, the amplitude of the horizontal wave changes along the body length, resulting in turns of average 25.6 ± 12.9, maximum 86.1◦ per cycle. In reversal turning, the vertical wave’s phase rapidly changes by π, resulting in a sudden, large change in movement direction (average 77.8 ± 27.4, maximum 160.5◦ per cycle) without body rotation. We applied these control mechanisms to a 16-link snake robot capable of sidewinding on sand. By modulation of horizontal wave amplitude gradient along the body, we replicated differential turning, and by producing a π phase shift in the vertical wave, we replicated a reversal turn. More complex wave modulations lead to enhanced robot maneuverability. 1 Work 2 H. supported by NSF PoLS. C. Astley, C. Gong, M. Serrano, H. Marvi, H. Choset, J. Mendelson, and D. L. Hu 12:27PM H24.00010 Design of a Localized Fluidization Burrowing Robot , DANIEL DORSCH, AMOS WINTER, MIT — This presentation will focus on the critical fluid and granular mechanics principles that drove the design of RoboClam 2.0, a self-actuated, radially expanding underwater burrowing device. RoboClam 2.0 was inspired by the Atlantic razor clam, Ensis directus, which burrows by contracting its valves and fluidizing the surrounding soil to reduce burrowing drag. This contraction results in a localized fluidized region occurring 1—5 body radii away from the animal. Moving through a fluidized, rather than static, soil requires energy that scales linearly with depth, rather than depth squared. In addition to providing an advantage for the animal, localized fluidization may yield significant value to engineering applications such as subsea robot anchoring and pipe installation. RoboClam 2.0 is sized to be an anchoring platform for autonomous underwater vehicles. We will present the scaling relationships that can be used to design RoboClam derivatives for different size scales and applications. The critical speed, displacement and force with which the device must contract to create fluidization are calculated based on soil parameters. These parametric relationships allow for choosing actuators of appropriate size and power output for desired burrowing performance. Monday, November 24, 2014 10:30AM - 12:40PM Session H25 Turbulence Theory: Wall-Bounded Flows — 2005 - Javier Jimenez, Universidad Politecnica de Madrid 10:30AM H25.00001 A minimal support for turbulence in a restricted nonlinear (RNL) model , VAUGHAN THOMAS, DENNICE F. GAYME, Johns Hopkins University, BRIAN FARRELL, Harvard University, PETROS IOANNOU, National and Kapodistrian University of Athens — In this work we explore the range of streamwise varying perturbations that can support self-sustaining turbulence in a restricted nonlinear (RNL) model of plane Couette flow. The RNL model partitions the dynamics of the flow field into a streamwise averaged mean flow and streamwise varying perturbations about that mean. The resulting system is a minimal representation of self-sustaining turbulence in which only a small number of streamwise varying perturbations interact with the mean flow. In the current work, we show that there is a minimum and maximum streamwise wavelength associated these streamwise perturbations. We also demonstrate that RNL turbulence can also be supported when the dynamics are further restricted to a single streamwise varying perturbation. This minimal RNL system possesses an upper and lower limit on the wavelengths associated with the single streamwise varying perturbation that is able to support RNL turbulence, i.e. when restricted to a perturbation whose wavelength is outside of this range, the RNL system returns to a laminar state. 10:43AM H25.00002 Structure and spectra of self-sustaining turbulence in a restricted nonlinear model1 , DENNICE F. GAYME, VAUGHAN THOMAS, Johns Hopkins Unversity, BRIAN FARRELL, Harvard University, PETROS IOANNOU, University of Athens — In this work we study a restricted nonlinear (RNL) model for plane Couette flow. This model is derived directly from the Navier Stokes equations and permits higher resolution studies of the dynamical system associated with the stochastic structural stability theory (S3T) model, which is a second order approximation of the statistical state dynamics of the flow. The RNL system was previously shown to exhibit self-sustaining turbulence that closely resembles DNS of turbulence but has the computational advantage of being supported by a small number of streamwise modes. Here, we further examine the structures underlying RNL turbulence. In particular, we focus on the roll and streak structures that are known to be critical in the self-sustaining process of wall-turbulence. We compare the RNL structures to those obtained from DNS by examining the temporal spectra of their streak and roll energies as well as the spectral densities of these structures at different wall-normal positions. The results show close correspondence between the structure and spectra of the rolls and streaks as well as agreement between the mean velocity profiles obtained from RNL simulations and DNS. 1 NSF support from AGS-1246929 and ATM-0736022 (to B.F.F) is gratefully acknowledged 10:56AM H25.00003 Using Synchronization to study the self-sustaining process in plane Couette flow turbulence1 , BRIAN FARRELL, Harvard University, PETROS IOANNOU, National and Kapodistrian University of Athens, DENNICE GAYME, VAUGHAN THOMAS, Johns Hopkins University — We show that separate realizations of turbulence in restricted nonlinear (RNL) simulations of plane Couette flow can be synchronized by linearly relaxing only the stream wise averaged components of the flow. The RNL system is obtained directly from the Navier-Stokes (NS) system by decomposing the dynamics into stream wise mean and perturbation equations and neglecting the perturbation-perturbation nonlinearity in the latter. Previous work demonstrated that the RNL system self-sustains turbulence with a mean flow as well as structural and dynamical features consistent with DNS. Using synchronization we verify that the self-sustaining process (SSP) operating in the RNL system is the parametric Lyapunov mechanism previously demonstrated to operate in the closely related stochastic structural stability theory (S3T) system. 1 NSF support from ATM-0736022 (to BF) is gratefully acknowledged 11:09AM H25.00004 Approximation of traveling wave solutions in wall-bounded flows using resolvent modes1 , BEVERLEY MCKEON, California Institute of Technology, MICHAEL GRAHAM, University of Wisconsin-Madison, RASHAD MOARREF, California Institute of Technology, JAE SUNG PARK, University of Wisconsin-Madison, ATI SHARMA, University of Southampton, ASHLEY WILLIS, Unversity of Sheffield — Significant recent attention has been devoted to computing and understanding exact traveling wave solutions of the NavierStokes equations. These solutions can be interpreted as the state-space skeleton of turbulence and are attractive benchmarks for studying low-order models of wall turbulence. Here, we project such solutions onto the velocity response (or resolvent) modes supplied by the gain-based resolvent analysis outlined by McKeon & Sharma (JFM, 2010). We demonstrate that in both pipe (Pringle et al, Phil. Trans. R. Soc. A, 2009) and channel (Waleffe, JFM, 2001) flows, the solutions can be well-described by a small number of resolvent modes. Analysis of the nonlinear forcing modes sustaining these solutions reveals the importance of small amplitude forcing, consistent with the large amplifications admitted by the resolvent operator. We investigate the use of resolvent modes as computationally cheap “seeds” for the identification of further traveling wave solutions. 1 The support of AFOSR under grants FA9550-09-1-0701, FA9550-12-1-0469, FA9550-11-1-0094 and FA9550-14-1-0042 (program managers Rengasamy Ponnappan, Doug Smith and Gregg Abate) is gratefully acknowledged. 11:22AM H25.00005 A restricted nonlinear-dynamics model for turbulent channel flows1 , ADRIÁN LOZANO-DURÁN, JAVIER JIMÉNEZ, Universidad Politécnica de Madrid, BRIAN F. FARRELL, Harvard University, PETROS J. IOANNOU, MARIOS A. NIKOLAIDIS, NAVID C. CONSTANTINOU, University of Athens — The dynamics of the formation of very-large scale structure in turbulent plane Poiseuille flow is studied by restricting the nonlinearity in the Navier–Stokes (NS) equations to interactions between the streamwise-averaged flow and perturbations. Using comparisons with DNS, we show that this restricted nonlinear dynamics (RNL) supports essentially realistic turbulence at Reτ = 900, despite the naturally occurring severe reduction in the set of streamwise wavenumbers supporting the turbulence. Using statistical diagnostics we verify that there are similar selfsustaining processes (SSP) underlying turbulence in the RNL and in the NS dynamics, separate manifestations of which operate in the buffer and outer layers. In the buffer layer, the SSP supports the familiar roll-streak mechanism of wall-bounded turbulence, while the outer-layer streaks in the RNL are probably the streamwise elongated structures referred to as VLSI. It is argued that the formation of the roll-streak structure is a universal mechanism that can be fruitfully studied in the minimal dynamics of RNL. 1 Funded by Multiflow project of the ERC, Navid Constantinou acknowledges the support of the Alexander S. Onassis Public Benefit Foundation. Brian Farrell was supported by NSF AGS-1246929. 11:35AM H25.00006 Identifying structure models in real turbulence1 , JAVIER JIMÉNEZ, U. Politécnica Madrid — Even when a model for the structures of a turbulent flow makes theoretical sense, it is important to test whether those structures are present in the flow, as well as how approximately and how often they appear. How that can be done is explored by tracking a linear transient-growth model for the logarithmic-layer in medium-size channel simulations (Reτ =1000–2000). The predicted linearized behavior is found in the evolution of ‘minimal’ Fourier modes of the wall-normal velocity, but only during bursting events accounting for about half of the total elapsed time. In particular, if a wavefront tilt angle is defined for each mode, periods of increasing forward tilt correspond to amplitude bursts. It is mostly during those periods that the tilt is well defined but, even then, the linearly most amplified perturbations do not describe the flow well. The flow evolution is explained by the model, but nonlinear initial conditions remain important for the fluctuation profiles. Quantitative measures for the level of approximation are defined and reported. 1 Funded by the Multiflow project of the ERC 11:48AM H25.00007 Wavenumber-frequency spectra in the logarithmic layer of wall turbulence1 , MICHAEL WILCZEK, RICHARD J.A.M. STEVENS, CHARLES MENEVEAU, Johns Hopkins University — We study space-time cor- relations of wall-bounded turbulence in terms of wavenumber-frequency spectra of the streamwise velocity component. The spectra are obtained from Large Eddy Simulations, which provide a full space-time record of the flow. We find that the frequency distributions exhibit a Doppler shift, which is a consequence of mean flow advection, as well as a considerable Doppler broadening, consistent with the Kraichnan-Tennekes random sweeping hypothesis. For wall-bounded turbulence, both of these effects vary with the wall distance and are closely related to the logarithmic behavior of the mean velocity profile and the velocity fluctuation profiles. We incorporate these observations into a simple analytical model for the wavenumber-frequency spectrum based on an advection equation featuring advection of the small-scale velocity fluctuations with a mean and a large-scale random-sweeping velocity. The model is found to be in very good agreement with the LES data. Potential applications of the model spectrum, e.g., to quantify the spatio-temporal structure of fluctuations in wind energy conversion, will be discussed. 1 Supported by DFG grant WI 3544/2-1, “Fellowships for Young Energy Scientists’ ’ (YES!) of FOM, and the US National Science Foundation grant IIA 1243482. 12:01PM H25.00008 Structures and scaling laws of turbulent Couette flow1 , MARTIN OBERLACK, VICTOR AVSARKISOV, Dept. Mech. Eng., TU Darmstadt, SERGIO HOYAS, Motores Termicos, Univ. Politecnica de Valencia, ANDREAS ROSTECK, Dept. Mech. Eng., TU Darmstadt, JOSE P. GARCIA-GALACHE, Motores Termicos, Univ. Politecnica de Valencia, ANDY FRANK, Dept. Mech. Eng., TU Darmstadt — We conducted a set of large scale DNS of turbulent Couette flow with the two key objectives: (i) to better understand large scale coherent structures and (ii) to validate new Lie symmetry based turbulent scaling laws for the mean velocity and higher order moments. Though frequently reported in the literature large scale structures pose a serious constraint on our ability to conduct DNS of turbulent Couette flow as the largest structures grow with increasing Re#, while at the same time Kolmogorov scale decreases. Other than for the turbulent Poiseuille flow a too small box is immediately visible in low order statistics such as the mean and limited our DNS to Reτ = 550. At the same time we observed that scaling of the mean is peculiar as it involves a certain statistical symmetry which has never been observed for any other parallel wall-bounded turbulent shear flow. Symmetries such as Galilean group lie at the heart of fluid dynamics, while for turbulence statistics due to the multi-point correlation equations (MPCE) additional statistical symmetries are admitted. Most important, symmetries are the essential to construct exact solutions to the MPCE, which with the new above-mentioned special statistical symmetry led to a new turbulent scaling law for the Couette flow. 1 DFG grant no; KH 257/2-1 12:14PM H25.00009 Local dissipation scales in strongly inhomogeneous turbulent shear flows , KHANDAKAR MORSHED, LAKSHMI DASI, Colorado State University — We have previously shown that the distribution of dissipation scales, Q(η), is dependent on the shear-dissipation Reynolds number Res ≡ hǫi/(S 2 ν) along the streamwise homogeneous direction. This dependency is further explored through a generalized theoretical framework linking Q(η), with the two-point correlation function, as well as the Reynolds stress tensor orientation relative to the mean axes of principal deformation. Time-resolved particle image velocimetry measurements were performed in a stationary turbulent flow past a backward facing step at Reynolds numbers 13,600, 9,000, and 5,500 based on the maximum velocity and step size. Q(η) were measured in all directions at different points in the measurement region with varying mean shear magnitude. Results show anisotropic Q(η) distributions strongly reflecting the anisotropy in the two-point correlation function and the Reynolds stress tensor. Based on these results the previous Res dependency is generalized to inhomogeneous directions while discussing the physical role of strong mean shear in inducing local anisotropy. 12:27PM H25.00010 The evolution of the very large scale motions in pipe flow1 , LEO HELLSTRÖM, Princeton University, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton, ALEXANDER SMITS, Princeton University, Monash University — We present a dual-plane snapshot POD analysis of turbulent pipe flow at a Reynolds number of 94,000. The high-speed PIV data were simultaneously acquired in two planes, a cross-stream plane (2D-3C) and a streamwise plane (2D-2C) on the pipe centerline. The two light sheets were orthogonally polarized, allowing particles situated in each plane to be distinguished. The dual-plane data were conditionally-averaged based on the occurrence/intensity of a given cross-stream snapshot POD mode. The conditionally-averaged modes reveal the streamwise extent and evolution of that particular cross-stream snapshot POD mode. A complex structure consisting of both wall-attached and detached large-scale structures is associated with the most energetic modes. The temporal evolution of these large-scale structures is examined using the time-shifted correlation of the cross-stream snapshot POD coefficients, identifying the low energy intermediate modes responsible for the transition between the large-scale modes. 1 Supported under ONR Grant N00014-13-1-0174 and ERC Grant No. 277472 Monday, November 24, 2014 10:30AM - 12:27PM Session H26 Rough Wall Boundary Layers II — 2007 - Michael Schultz, United States Naval Academy 10:30AM H26.00001 A study of transient channel flow in a transitionally rough regime , MEHDI SEDDIGHI, SHUISHENG HE, Universiy of Sheffield, TOM O’DONOGHUE, DUBRAVKA POKRAJAC, University of Aberdeen, ALAN VARDY, University of Dundee — DNS has been used to investigate the transient behaviour of turbulence following a rapid flow acceleration from an initially turbulent flow in a channel with a smooth top wall and a roughened bottom wall made of close-packed pyramids. Simulations have been performed at various flow conditions in the transitionally rough regime with equivalent roughness heights (ks+ ) ranging from 12 to 42. It is shown that the transient responses of the flow over the smooth and rough walls are practically independent of each other. Also, the nature of the process over the rough wall varies strongly as the influence of the roughness increases during the early stages of the acceleration. Whereas the transient flow over the smooth-wall undergoes a process strikingly similar to laminar-turbulent bypass transition, the corresponding behaviour over the rough wall depends on the wall condition. When the equivalent roughness height of the final flow condition is below ∼30, bypass-like transition dominates, although the roughness induces early transition. When ks+ > 30, however, the rough-wall flow undergoes a highly transient process resembling roughness induced transition. 10:43AM H26.00002 Properties of the advective transport contribution to the inertial mean dynamics of rough-wall boundary layers , RACHEL EBNER, University of New Hampshire, JOSEPH KLEWICKI, University of New Hampshire; University of Melbourne — Measurements and scaling analyses are conducted to clarify the combined effects of roughness and Reynolds number on momentum transport in rough-wall turbulent boundary layers. Experiments employing a four element (“Foss style”) spanwise vorticity probe cover nearly a decade in Reynolds number, and nearly three decades in sand grain roughness, ks+ . Here we leverage the expression that decomposes the Reynolds stress gradient into the difference of two velocity-vorticity correlations, i.e., −∂uv/∂y = vωz − wωy . The present analyses focus on the first term on the left hand side, vωz , in the logarithmic layer and outer regions, as it is known from smooth-wall studies that this advective transport mechanism is the largest contributor to −∂uv/∂y in the domain where the mean dynamics are inertially dominated. Streamwise correlation maps and length scales associated with the spectra and correlations of v and ωz are used to clarify the scaling behaviors of the motions underlying −∂uv/∂y. The results are shown to further support the combined roughness Reynolds number description of Mehdi et al. 2013, J. Fluid Mech. 731, 682 10:56AM H26.00003 Characteristics of Large-Scale Motions in a Turbulent Boundary Layer Overlying Complex Roughness , J.M. BARROS, U.S. Naval Academy, K.T. CHRISTENSEN, Univ. of Notre Dame — The characteristics of large-scale motions in a turbulent boundary layer overlying complex roughness are explored with high-frame-rate stereo PIV measurements in the wallnormal–spanwise plane. It was previously reported that the single-point turbulence statistics of this flow display strong spanwise heterogeneity, particularly spanwise-alternating low- and high-momentum flow pathways in the mean flow bounded by large-scale streamwise-oriented roll cells and marked by enhanced Reynolds stresses and turbulent kinetic energy. These patterns were interpreted as the imprints of roughness-induced turbulent secondary flows owing to the streamwise elongation and spanwise heterogeneity of the topography. Frequency spectra of all three velocity components at fixed wall-normal location also display strong dependence on spanwise position, principally that of the streamwise velocity. In particular, the roughness promotes enhanced turbulent kinetic energy content of the large-scale motions and smaller-scale motions, coupled with strong spanwise dependence in the energy content of the very-large-scale motions when compared to smooth-wall flow. Modifications of Reynolds shear stress content as a function of scale are also explored from the three-component velocity measurements. 11:09AM H26.00004 Comparison of the coherent structure of rough and smooth wall turbulent boundary layers at high Reynolds number , DOUGAL SQUIRE, CHARITHA DE SILVA, University of Melbourne, MICHAEL SCHULTZ, United States Naval Academy, NICHOLAS HUTCHINS, IVAN MARUSIC, University of Melbourne — A comparison of structural aspects of the log- and wake-regions of smooth and rough wall zero pressure gradient turbulent boundary layers is presented at a friction Reynolds number of approximately 12,000. The roughness consists of P36 sandpaper, installed over the 54 m2 working section in a continuous sheet. The results from four measurements are discussed, consisting of two eight-camera PIV arrangements above each surface. The field of view of both arrangements captures the full wall-normal extent of the boundary layer, but differs in the streamwise direction; one camera array captures a streamwise domain that spans approximately twice the boundary layer thickness; the other has a narrower streamwise extent in order to obtain an enhanced spatial resolution in the order of the Kolmogorov microscale. Combined, the two arrangements enable investigation of structural features with reasonably large streamwise dimension—using the large field of view data—and provide well resolved information on the wall-normal structure of the boundary layer—using the narrow field of view data. Generally, the data in the inertial dominated region confirm that the studied smooth and rough wall bounded flows are structurally similar, providing support for the outer-layer similarity hypothesis. 11:22AM H26.00005 Perturbation of roughness-induced secondary flow in a turbulent boundary layer overlying complex roughness , G. PATHIKONDA, Univ. of Illinois, K.T. CHRISTENSEN, Univ. of Notre Dame — Recent experiments investigating flow over more heterogeneous and organized roughness have revealed spanwise inhomogeneity in turbulence statistics interpreted as roughness-induced secondary flow induced by streamwise elongation and spanwise heterogeneity of the topography itself. In particular, spanwise alternating regions of low- and high-momentum pathways in mean streamwise velocity have been observed, each flanked by streamwise oriented counter-rotating roll cells, for flow over both the complex roughness investigated herein and organized roughness reported in the literature. We explore perturbation of this roughness-induced secondary flow as a means of studying its origin and persistence. Spanwise–wall-normal stereo PIV measurements of flow over complex roughness are made, first with an incident smooth-wall turbulent boundary layer upstream of the roughness followed by perturbation of this incident smooth-wall turbulent boundary layer with organized hemispherical roughness elements prior to transition to the complex roughness (with the hemisphere scale being distinct from that of the complex roughness). Hot-wire measurements are also made to capture the energy distribution/re-distribution at various flow scales in both flow conditions. 11:35AM H26.00006 Turbulent secondary flows in high Reynolds number boundary layers induced by streamwise-elongated complex roughness , WILLIAM ANDERSON, University of Texas at Dallas, JULIO BARROS, United States Naval Academy, KENNETH CHRISTENSEN, University of Notre Dame — It has been reported that complex roughness with a predominant streamwise elongation induces secondary mean flow heterogeneities in the above turbulent boundary layer (Mejia-Alvarez and Christensen, 2013: Phys. Fluids 25:115109, MAC; Nugroho et al., 2013: Int. J. Heat Fluid Flow 41:90-102). These mean secondary flows exist as transverse variations of mean streamwise velocity (so-called low- and high-momentum pathways, MAC) and are flanked by mean counter-rotating, boundary layer-scale circulations (Christensen and Barros, 2014: J. Fluid Mech. 748:R1). In related work, we have used large-eddy simulation to model turbulent boundary layer flow over a suite of topographies composed of “strips” of high and low roughness length (drag imposed with the equilibrium logarithmic law); in all cases, we observe the formation of highand low-momentum pathways (Willingham et al., 2013: Phys. Fluids 26:025111.). Here, we investigate turbulence statistics from large-eddy simulation such as magnitudes and spatial gradients of Reynolds stresses and turbulence kinetic energy, to discern underlying physical processes responsible for the secondary flows. We demonstrate that elevated production of turbulence above “high” roughness necessitates the mean circulations by virtue of turbulent kinetic energy production-dissipation non-equilibrium. We propose that the mean flow is Prandtl’s secondary flow of the second kind. 11:48AM H26.00007 Boundary-layer structure of shallow free-surface flows with high relative roughness , OLIVIER EIFF, EMMA FLORENS, FRÉDÉRIC MOULIN, U. of Toulouse / CNRS — The boundary-layer structure of shallow free-surface flows over very rough walls is investigated with particle image velocimetry (PIV) both within the canopy and above, without disturbing the flow, by gaining complete optical access. This enabled reliable estimates of the double-averaged mean and turbulence profiles to be obtained by minimizing and quantifying the usual errors introduced by limited temporal and spatial sampling. It is shown that poor spatial sampling can lead to erroneous vertical profiles in the roughness sublayer. In order to better define and determine the roughness-sublayer height, a methodology based on the measured spatial dispersion is proposed which takes into account temporal sampling errors. The results reveal values well below the usual more ad hoc estimates. Then, the double-averaged statistics were used to investigate the effect of low relative submergence of the roughness elements on the friction velocity and the logarithmic law. The measurements show that the dispersive stresses are necessary to estimate correctly the total shear stress above the canopy top. The logarithmic law is shown to persist for submergence ratios at least as high as 0.33, even though the roughness sublayer largely extends into it. A dependence of the roughness length on submergence is observed, but not for the displacement height. 12:01PM H26.00008 Investigation of Wall Shear Stress Behavior for Rough Surfaces with Blowing1 , JACOB HELVEY, COLBY BORCHETTA, University of Kentucky, MARK MILLER, Princeton University, ALEXANDRE MARTIN, SEAN BAILEY, University of Kentucky — We present an experimental study conducted in a turbulent channel flow wind tunnel to determine the modifications made to the turbulent flow over rough surfaces with flow injection through the surfaces. Hot-wire profile results from a quasi-two-dimensional, sinusoidally-rough surface indicate that the effects of roughness are enhanced by momentum injection through the surface. In particular, the wall shear stress was found to show behavior consistent with increased roughness height when surface blowing was increased. This observed behavior contradicts previously reported results for regular three-dimensional roughness which show a decrease in wall shear stress with additional blowing. It is unclear whether this discrepancy is due to differences in the roughness geometry under consideration or the use of the Clauser fit to estimate wall shear stress. Additional PIV experiments are being conducted for a three-dimensional fibrous surface to obtain Reynolds shear stress profiles. These results provide an additional method for estimation of wall-shear stress and thus allow verification of the use of the Clauser chart approach for flows with momentum injection through the surface. 1 This research is supported by NASA Kentucky EPSCoR Award NNX10AV39A, and NASA RA Award NNX13AN04A 12:14PM H26.00009 The structure of turbulence overlying impermeable and permeable rough walls , T. KIM, G. BLOIS, J. BEST, Univ. of Illinois, K.T. CHRISTENSEN, Univ. of Notre Dame — Turbulent flow overlying complex topographies, both impermeable and permeable, occur across a broad range of scales in both natural and engineering environments. Permeability of the wall introduces a higher degree of both structural and conceptual complexity, with previous studies suggesting that interactions between the turbulent free flow and pore flow occur along the permeable interface and play a defining role in momentum exchange across the interface. Here we employ a Refractive-Index-Matching (RIM) technique in order to access the flow across the permeable interface with the particle image velocimetry (PIV) method, resulting in unimpeded optical access to the fluid flow at and within a permeable bed. Cubic-packed hemispheres are studied in both impermeable and permeable configurations, with models cast by an acrylic resin whose refractive index matched that of the working fluid (aqueous sodium iodide). The statistical and structural features of the flow in the near-wall region of the impermeable case and the interfacial region of the permeable case are compared to understand the role of permeability in driving momentum exchange processes as a function of Reynolds number. Comparisons to recent numerical simulations are also made. Monday, November 24, 2014 10:30AM - 12:27PM Session H27 Turbulence: Theory II — 2009 - Robert Moser, University of Texas at Austin 10:30AM H27.00001 Turbulence at high resolution: intense events in dissipation, enstrophy and acceleration1 , P.K YEUNG, X.M. ZHAI, Georgia Tech, K.R. SREENIVASAN, New York Univ — Access to the Blue Waters supercomputer under the NSF Track 1 Petascale Resource Allocations program has allowed us to conduct an 81923 simulation of forced isotropic turbulence, with Taylor-scale Reynolds number close to 1300, and grid spacing at about 1.5 Kolmogorov scales. Extreme fluctuations in dissipation and enstrophy (over 10,000 times the mean) are observed, and found to scale similarly and occur together. Conditional sampling based on both dissipation and enstrophy shows that such extreme events in these variables are directly associated with strong intermittency in the fluid particle acceleration, which reaches values well beyond 100 standard deviations. An attempt is made to characterize in detail the formation of events of intense dissipation and enstrophy, including the transport, production and dissipation terms in the dissipation and enstrophy transport equations, as well as the nature of local flow conditions in principal strain-rate axes. Statistics of dissipation and enstrophy averaged over 3D sub-domains of linear size in the inertial range are also available. Both high Reynolds number and good small-scale resolution are important factors in these results. 1 Supported by NSF Grant ACI-1036170 10:43AM H27.00002 Energy Spectra of Higher Reynolds Number Turbulence by the DNS with up to 122883 Grid Points , TAKASHI ISHIHARA, JST CREST, Nagoya University, YUKIO KANEDA, Aichi Institute of Technology, KOJI MORISHITA, MITSUO YOKOKAWA, Kobe University, ATSUYA UNO, RIKEN AICS — Large-scale direct numerical simulations (DNS) of forced incompressible turbulence in a periodic box with up to 122883 grid points have been performed using K computer. The maximum Taylor-microscale Reynolds number Rλ , and the maximum Reynolds number Re based on the integral length scale are over 2000 and 105 , respectively. Our previous DNS with Rλ up to 1100 showed that the energy spectrum has a slope steeper than −5/3 (the Kolmogorov scaling law) by factor 0.1 at the wavenumber range (kη < 0.03). Here η is the Kolmogorov length scale. Our present DNS at higher resolutions show that the energy spectra with different Reynolds numbers (Rλ > 1000) are well normalized not by the integral length-scale but by the Kolmogorov length scale, at the wavenumber range of the steeper slope. This result indicates that the steeper slope is not inherent character in the inertial subrange, and is affected by viscosity. 10:56AM H27.00003 The turbulent cascade of individual eddies1 , CECILIA HUERTAS-CERDEIRA, ADRIÁN LOZANO-DURÁN, JAVIER JIMÉNEZ, U. Politécnica Madrid — The merging and splitting processes of Reynolds-stress carrying structures in the inertial range of scales are studied through their time-resolved evolution in channels at Reλ = 100 − 200. Mergers and splits coexist during the whole life of the structures, and are responsible for a substantial part of their growth and decay. Each interaction involves two or more eddies and results in little overall volume loss or gain. Most of them involve a small eddy that merges with, or splits from, a significantly larger one. Accordingly, if merge and split indexes are respectively defined as the maximum number of times that a structure has merged from its birth or will split until its death, the mean eddy volume grows linearly with both indexes, suggesting an accretion process rather than a hierarchical fragmentation. However, a non-negligible number of interactions involve eddies of similar scale, with a second probability peak of the volume of the smaller parent or child at 0.3 times that of the resulting or preceding structure. 1 Funded by the Multiflow project of the ERC 11:09AM H27.00004 The transfer of kinetic energy in turbulent flows1 , JOSE I. CARDESA, JAVIER JIMENEZ, Universidad Politecnica de Madrid — We study the statistics of the point-wise inter-scale energy transfer across a given filter width in direct numerical simulations of homogeneous isotropic turbulence, homogeneous shear flow and turbulent channels. This is first done for the classical subgrid-scale (SGS) dissipation found in the kinetic energy equation for the filtered velocity field. It is then compared with an analogous term T arising in the equation for the residual (small-scale) velocity field. T can take several expressions, and we report on the one which minimises its variance. For all flows, the SGS dissipation exhibits a negative skewness which increases with the filter width, while T has a positive skewness which decreases with filter width. This is consistent with the SGS dissipation being an average energy sink for the large scales, while T is an average energy source for the small ones. The different dependence on filter width of the mean and standard deviations of these two quantities is explored, and joint probability density functions based on the two quantities are investigated to understand the observed discrepancies between forward scatter and backscatter events. 1 Funded by the ERC Multiflow project 11:22AM H27.00005 The reversible 3D turbulent cascade1 , ALBERTO VELA-MARTÍN, JAVIER JIMÉNEZ, Universidad Politécnica de Madrid — It has been known for some time that the dynamic Smagorinsky LES model is reversible. If the sign of the velocities in an isotropic turbulence simulation is inverted after it has decayed for some time, it evolves back to its original state, recovering its energy and other turbulent quantities. We use this reverse evolution, during which the cascade transfers energy from the small to the large scales, to gain new insights into the behavior and reversible features of the inertial energy range. The dynamics in the plane of the Q-R topological invariants are studied for the forward and backward evolutions, as well as the structure of the Lyapunov exponents in both regimes. Considerable differences are found. In particular, the Q-R pdf of the inverse evolution is reversed, with a stable Vieillefosse tail along negative R, and a main lobe in which vortex compression predominates. The contribution of the different terms in the equation is computed for both cases, both with and without an LES model. 1 Funded by Multiflow project of the ERC 11:35AM H27.00006 A Phenomenological Theory of Rotating Turbulence , YASIR BIN BAQUI, PETER DAVIDSON, University of Cambridge — We present direct numerical simulations of statistically-homogeneous, freely-decaying, rotating turbulence in which the Rossby number, Ro = u⊥ /2Ωℓ⊥ , is of order unity. The initial condition consists of fully-developed turbulence in which Ro is sufficiently high for rotational effects to be weak. However, as the kinetic energy falls, so also does Ro, and quite quickly we enter a regime in which the Coriolis force is relatively strong and anisotropy grows rapidly, with ℓ⊥ << ℓ// . This regime occurs when Ro ∼ 0.4 and is characterised by an almost constant perpendicular integral scale, ℓ⊥ ∼ constant, a rapid linear growth in the integral scale parallel to the rotation axis, ℓ// ∼ ℓ⊥ Ωt, and a slow decline in the value of Ro. We observe that the rate −5/3  of dissipation of energy scales as ε ∼ u3 ℓ// and that both the perpendicular and parallel energy spectra exhibit an k−5/3 inertial range; E(k⊥ ) ∼ ε2/3 k⊥ and E(k// ) ∼ −5/3 ε2/3 k// . We show that these power-law spectra have nothing to do with Kolmogorov’s theory and are not a manifestation of traditional −7/5 critical balance theory, as this requires ε ∼ u3 /ℓ⊥ and E(k// ) ∼ (ε4/5 /Ω2/5 )k// . Finally, we develop a spectral theory of the inertial range that assumes that the observed linear growth in anisotropy, ℓ// /ℓ⊥ ∼ Ωt, also occurs on a scale-by-scale basis all the way down to the Zeman scale. 11:48AM H27.00007 Filtered linear forcing: a technique for simulating high Reynolds number turbulence in physical space , JOHN PALMORE, OLIVIER DESJARDINS, Cornell University — Low waveshell spectral forcing has been proven to be a simple and effective manner to generate isotropic turbulence in a periodic domain. This simplicity is lost for flow problems with complex boundary conditions such as resolved particle flows, fluid-fluid flows with interfaces, and wall-bounded flows. Lundgren’s linear forcing in physical space is a straightforward and easy-to-implement method to tackle these problems; however, the use of this method results in a halving of the large turbulence length scale. The technique that will be presented in this talk applies a low-pass filter to the source term used in linear forcing. It is shown to recover the scale resolution of low waveshell spectral forcing which translates to an approximately 60 percent increase in the attainable Reynolds number for a given computation domain. The characteristics of homogeneous isotropic turbulence generated using filtered linear forcing will be discussed. Finally, extension of this idea to scalar forcing will be presented. 12:01PM H27.00008 Grid generated turbulence in the near-field , RICARDO SALAZAR, JUAN ISAZA, Universidad EAFIT, ZELLMAN WARHAFT, Cornell University — Using a conventional bi-planar turbulence-generating grid, we confirm the recent findings (Valente & Vassilicos, Phys. Rev. Lett., vol. 108, 2012, art. 214503) that show there is a turbulence decay region close to the generating grid that departs from the “classical” turbulence decay (Comte-Bellot & Corrsin, J. Fluid Mech., vol. 25, 1966, pp. 657–682). In this “near field” region, the turbulence energy decays more rapidly than in the far field and it exhibits unusual scaling properties. Based on the velocity decay laws, we show that for our conventional grid, the near field extends from x/M ∼ 6 to x/M ∼ 12 where x is the downstream distance from the grid and M is the mesh size. However, other statistics (velocity derivatives and length scales ratios) indicate that the extent of the initial period depends on the grid mesh Reynolds number, RM , extending further for higher values of RM . In the near field the turbulence approaches isotropy both at the large and small scales but there still is inhomogeneity in the derivative statistics. The derivative skewness also departs from values observed at comparable Reynolds numbers in the far field decay region, and in other turbulent flows at comparable Reynolds numbers. We do not believe that the near field scaling violates Kolmogorov phenomenology, which applies to systems that are not affected by proximity to initial and boundary conditions. These conditions are not met close to the grid. 12:14PM H27.00009 Turbulent wakes of irregular objects , MARTIN OBLIGADO, JOHN CHRISTOS VASSILICOS, Imperial College London — Recently, flow regions with non-equilibrium high Reynolds number turbulence at odds with usual Richardson-Kolmogorov phenomenology have been discovered in a number of turbulent flows, in particular axisymmetric and self-preserving turbulent wakes of plates with irregular edges. These regions are characterised by streamwise evolutions of the mean flow profiles which have only recently been documented, see PRL 111, 144503 (2013). One of the main differences between the equilibrium and the non-equilibrium predictions involves the momentum thickness. We therefore have carried out experiments with bluff bodies that have various different chord lengths in the direction of the flow. We performed wind tunnel anemometry measurements of wakes generated by bluff plates with simple square edge peripheries and by bluff plates with irregular edge peripheries which allow the formation of jet-wake flows. The wakes generated by the irregular plates become axisymmetric much earlier than the wakes generated by regular ones, irrespective of chord length. Furthermore, the non-equilibrium wake scalings are present in the case of the bluff plates with irregular edges, again irrespective of chord length. Monday, November 24, 2014 10:30AM - 12:40PM Session H28 Turbulence: Mixing II — 2011 - Julia Ling, Stanford University 10:30AM H28.00001 DNS of Turbulent Mixing Layers Between Two Fluids of Large Density Difference , JON BALTZER, DANIEL LIVESCU, Los Alamos National Laboratory — In numerous practical applications, shear layers exist between fluids of strongly differing densities. At high Atwood numbers, the large variations in density introduce important effects that have recently been observed in other flows (e.g., Livescu & Ristorcelli, J. Fluid Mech., 605, 145-180, 2008). To investigate the inertial variable density effects on the instability growth and structure of mixing layers, we first omit the buoyancy effects and perform very large Direct Numerical Simulations of planar mixing layers between two miscible fluids, each with different density. We consider initial disturbances for the DNS in light of linear stability analysis using a new, generalized form of the Orr-Sommerfeld equations, which includes the variable density effects. Based on the most unstable modes obtained from this analysis, DNS domain sizes are varied to accommodate different extents of mode pairing. The results display the overall statistical effects on the turbulence and mixing, as well as the structural differences, that occur as Atwood number is varied. In particular, the asymmetries introduced by the differences in the densities of the mixing layer streams are highlighted. 10:43AM H28.00002 ABSTRACT WITHDRAWN — 10:56AM H28.00003 Favre-Averaged Turbulence Statistics in Variable Density Mixing of Buoyant Jets , JOHN CHARONKO, KATHY PRESTRIDGE, Los Alamos National Lab — Variable density mixing of a heavy fluid jet with lower density ambient fluid in a subsonic wind tunnel was experimentally studied using Particle Image Velocimetry and Planar Laser Induced Fluorescence to simultaneously measure velocity and density. Flows involving the mixing of fluids with large density ratios are important in a range of physical problems including atmospheric and oceanic flows, industrial processes, and inertial confinement fusion. Here we focus on buoyant jets with coflow. Results from two different Atwood numbers, 0.1 (Boussinesq limit) and 0.6 (non-Boussinesq case), reveal that buoyancy is important for most of the turbulent quantities measured. Statistical characteristics of the mixing important for modeling these flows such as the PDFs of density and density gradients, turbulent kinetic energy, Favre averaged Reynolds stress, turbulent mass flux velocity, density-specific volume correlation, and density power spectra were also examined and compared with previous direct numerical simulations. Additionally, a method for directly estimating Reynolds-averaged velocity statistics on a per-pixel basis is extended to Favre-averages, yielding improved accuracy and spatial resolution as compared to traditional post-processing of velocity and density fields. 11:09AM H28.00004 Alignment of principal strain rates, vorticity, and scalar gradients in a turbulent nonpremixed jet flame , ANTONIO ATTILI, FABRIZIO BISETTI, King Abdullah University of Science and Technology — The alignment of vorticity and gradients of conserved and reactive scalars with the eigenvectors of the strain rate tensor (i.e., the principal strains) is investigated in a direct numerical simulation of a turbulent nonpremixed flame achieving a Taylor’s scale Reynolds number in the range 100 ≤ Reλ ≤ 150 [Attili et al. Comb. Flame, 161, 2014]. The vorticity vector displays a pronounced tendency to align with the direction of the intermediate strain. These alignment statistics are in almost perfect agreement with those in homogeneous isotropic turbulence [Ashurst et al. Physics of Fluids 30, 1987] and differ significantly from the results obtained in other nonpremixed flames in which vorticity alignment with the most extensive strain was observed [Boratav et al. Physics of Fluids 8, 1996]. The gradients of conserved and reactive scalars align with the most compressive strain. It is worth noting that conditioning on the local values of the mixture fraction does not affect the statistics. Our results suggest that turbulence overshadows the effects of heat release and chemical reactions. This may be due to the larger Reynolds number achieved in the present study compared to that in previous works. 11:22AM H28.00005 Modeling Reaction Rates in Variable-Density Turbulent Mixing , NICHOLAS DENISSEN, RAYMOND RISTORCELLI, Los Alamos National Laboratory — Modeling reactions in variable-density turbulent mixing is important for multiphysics applications such as Inertial Confinement Fusion (ICF). Reynolds–Averaged Navier–Stokes (RANS) models are an important tool in this research, and work is ongoing to improve their fidelity in complex flows. Connecting these models to the underlying multi-material mixing and thermonuclear reaction rates is essential. This talk describes the BHR family of turbulence models developed at Los Alamos National Laboratory (LANL), and the information they provide for characterizing turbulent mixing. Some exact relationships for reaction rates in variable-density turbulence will be presented and connected to the lower-order moments provided by variable-density RANS models. These provide model equations for initially pre-mixed and initially separated reactants. The strengths and limitations will be discussed, and examples will be shown from ICF-like hydrodynamic simulations in the LANL ASC code FLAG to assess the impact of these models. 11:35AM H28.00006 Prediction of scalar gradient distributions under stretching and random aggregation processes: application to mixing in turbulent and porous media flows , TANGUY LE BORGNE, University of Rennes 1, France, PETER HUCK, Aix Marseille University, France, MARCO DENTZ, CSIC Barcelona, Spain, EMMANUEL VILLERMAUX, Aix Marseille University, France — Scalar gradients play a key role in controlling mixing and reaction processes in natural and industrial flow systems. The stretching action of flow fields naturally organizes scalar fields into lamellar structures, whose elongation and aggregation determine the evolution of concentration distributions. In this context, the prediction of scalar gradient distributions requires quantifying the spatial correlation of concentration fields. For general stretching and aggregation processes, we derive theoretical predictions of the temporal evolution of the concentration increment PDFs over any spatial increments. This framework is shown to provide accurate predictions of concentration gradient distributions for a range of flow systems, including turbulent and porous media flows. In particular, the theory links intermittent scalar field properties to their random additive nature and consequent spatial organization. We argue that the analysis of the distribution of concentration increments over different spatial increments may be considered as a deconstruction of the basic lamella assemblage, revealing the elementary structures building concentration distributions in heterogeneous flows. 11:48AM H28.00007 Hierarchical parcel-swapping (HiPS) representation of turbulent flow and mixing , ALAN KERSTEIN, Danville, CA — An economical representation of effects of turbulence on the time-evolving structure of diffusive scalar fields is obtained by introducing a hierarchical (tree) network connecting fluid parcels, with effects of turbulent advection represented by swapping pairs of sub-trees at rates determined by turbulence time scales associated with the sub-trees [1]. The fluid parcels reside at the base of the tree. The tree structure partitions the fluid parcels into adjacent pairs (or more generally, p-tuples). Adjacent parcels intermix at rates governed by diffusion time scales based on molecular diffusivities and parcel sizes. This simple procedure efficiently accomplishes long-standing objectives of turbulent mixing model development, such as generating physically based time histories of fluid-parcel nearest-neighbor encounters and the associated spatial structure of turbulent scalar fields. With the introduction of velocity components as well as scalars, this hierarchical parcel-swapping (HiPS) formulation becomes a self-contained flow simulation, as illustrated by its application to fully developed channel flow [2]. [1] A. R. Kerstein, J. Stat. Phys. 153, 142-161 (2013). [2] A. R. Kerstein, J. Fluid Mech. 750, 421-463 (2014). 12:01PM H28.00008 Backward tracking and Lagrangian passive scalar mixing in turbulence simulations1 , D. BUARIA, P.K YEUNG, Georgia Tech, B.L. SAWFORD, Monash University, Australia — In many environmental problems the dispersion of contaminants with finite molecular diffusivity is closely related to the trajectories of molecules which undergo Brownian motion relative in the field. This invokes a Lagrangian view of mixing, which asks, for instance, how a pair of molecules far apart at earlier times may come together and cause a high local concentration of the diffusing material or property. We have implemented an efficient and statistically robust approach to extract backward statistics via the post-processing of trajectory data stored in direct numerical simulations with many millions of fluid particles and diffusing molecules. Results are obtained at Taylor scale Reynolds number 140 to 400 and Schmidt numbers from 0.001 to 1000. As expected the contrast between forward and backward dispersion is greater at higher Reynolds number where nonlinear turbulent transport is stronger. Subject to sampling, the Lagrangian data obtained agree well with Eulerian results on the production and dissipation of the variance of a passive scalar driven by a uniform mean gradient in isotropic turbulence. Extensions to multi-particle and multi-molecule clusters are also briefly addressed. 1 Supported by NSF Grant CBET-1235906. 12:14PM H28.00009 Turbulent Diapycnal Mixing In Stratified Shear Flows: Parameterizations of Mixing Efficiency and Diapycnal Diffusivity , HESAM SALEHIPOUR, W. RICHARD PELTIER, University of Toronto — Motivated by the importance of diapycnal mixing in geophysical fluids, we study the inhomogeneously stratified and sheared turbulence that is engendered by the breaking of a Kelvin-Helmholtz wave and will direct connections to the homogeneously stratified case. We employ DNS method to focus on the high-Re regime and investigate a wide range of Ri and P r. This talk will consist of three related topics: (1) linear stability analysis of the transition process to understand the sequence of secondary instabilities that govern transition at increasing values of P r at a fixed high-Re. (2) A “recipe” for irreversible mixing which illuminates the fundamental physical mechanisms that are involved in rising the background potential energy and represents mixing as being composed of both (i) turbulent buoyancy flux and (ii) flow anisotropy due to energy-containing coherent eddies. (3) The precise derivation of diapycnal diffusivity, Kρ , with no simplifying assumptions which depends on an exact definition of mixing efficiency (rather than approximating it in terms of the flux Richardson number). We will propose robust parameterizations in terms of buoyancy Reynolds number, Reb , Ri and P r for a wide range of these parameters extending Reb = O(103 ). 12:27PM H28.00010 Effect of velocity ratio on coherent-structure dynamics in turbulent free shear layers1 , SAIKISHAN SURYANARAYANAN, RODDAM NARASIMHA, Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore — The relevance of the vortex-gas model to the large scale dynamics of temporally evolving turbulent free shear layers has been established by extensive simulations (Phys.Rev.E 89, 013009, 2014). The effects of velocity ratio (r = U2 /U1 ) on shear layer dynamics are revealed by spatially evolving vortex-gas shear-layer simulations using a computational model based on Basu et al (Appl.Math.Modelling 19, 1995), but with a crucial improvement that ensures conservation of global circulation. The simulations show that the initial conditions and downstream boundaries can significantly affect the flow over substantial part of the domain, but the equilibrium spread rate is a universal function of r, and is within the experimental scatter. The spread in the r = 0 limit is higher than Galilean-transformed temporal value. The present 2D simulations at r = 0 show continuous growth of structures, while merger-dominated evolution is observed for r = 0.23 (and higher). These two mechanisms were observed across the same two values of r in the experiments of D’Ovidio & Coats (J.Fluid Mech 737, 2013), but the continuous growth was instead attributed to mixing-transition and 3D. The 2D mechanisms responsible for the observed continuous growth of structures are analyzed in detail. 1 Supported in part by RN/Intel/4288 and RN/DRDO/4124. Monday, November 24, 2014 10:30AM - 12:40PM Session H29 Aerodynamics: Fluid-Structure Interaction — 2014 - 10:30AM H29.00001 Numerical investigation of a flexible plate in a uniform flow , XI-YUN LU, CHAO TANG, University of Science and Technology of China — The dynamics of a flexible plate in a uniform flow with different flow directions have been studied numerically by an immersed boundary-lattice Boltzmann method for the fluid flow and a finite element method for the plate deformation. A series of distinct states of the plate deformation are identified, including straight, flapping, deflected, and deflected-flapping modes which depend mainly on the bending stiffness of the plate. The effects of the flow direction and the aspect ratio of the plate on dynamics of the fluid-plate system and elastic strain energy of the plate are investigated. The vortical structures around the plate are analyzed to elucidate the correlation of the evolution of vortices with the plate deformation. It is obtained that the flow-induced flapping mode can efficiently produce elastic strain energy for harvesting fluid kinetic energy. The results obtained in this study provide physical insight into the understanding of the mechanisms on the dynamics of the fluid-plate system and the conversion of fluid kinetic energy to elastic strain energy. 10:43AM H29.00002 Fluid-solid-electric couplings and efficiency of piezoelectric flags1 , YIFAN XIA, SEBASTIEN MICHELIN, LadHyX - Ecole Polytechnique, OLIVIER DOARE, UME - ENSTA ParisTech — The spontaneous and self-sustained flapping of a flexible plate in an axial flow can be used for energy harvesting applications by placing piezoelectric patches on its surface, that periodically deform with the plate, generating an electrical current. These piezoelectric elements also introduce a feedback of the output circuit on the fluid-solid dynamics and may modify its flapping behavior. To better understand the dynamics of these piezoelectric flags, the resulting energy transfers and their harvesting efficiency, numerical simulations of the fluid-solid-electric problem were carried out using an explicit description of the energy harvesting mechanism and simple output circuits. In the case of purely resistive circuits, a tuning mechanism is identified between the circuit’s time scale and the flapping frequency. When the circuit possesses its own dynamics (e.g. inductive/resonant circuits), a lock-in mechanism is observed that leads to an effective control of the flapping frequency by the output circuit over a large range of parameters and a significant increase in the energy harvesting performance of the device. 1 Supported by ANR (Grant # ANR-12-JS09-0017). 10:56AM H29.00003 On the wake dynamics of flapping inverted flags1 , ANVAR GILMANOV, FOTIS SOTIROPOULOS, Univ of Minn - Minneapolis, JULIA COSSE, MORY GHARIB, California Institute of Technology — As recently shown experimentally by Kim et al. (JFM, 2013), when a flexible flag with a fixed trailing edge (an inverted flag) is exposed to a uniform inflow it can exhibit complex structural response and rich fluid-structure interaction (FSI) dynamics. We employ a new FSI numerical method to carry out large-eddy simulation (LES) of inverted flags in the range where large-amplitude flapping instabilities have been found experimentally. The numerical method integrates the curvilinear immersed boundary (CURVIB) FSI method of Borazjani et al. (JCP, 2008) with the thin-shell, rotation-free, finite-element (FE) formulation of Stolarski et al. (Int. JNME, 2013) and is able to simulate FSI of flexible thin bodies undergoing oscillations of arbitrarily large amplitude. The dynamic Smagorinsky model is employed for subgrid scale closure and a wall model is employed for reconstructing velocity boundary conditions. Comparisons with the experimental data show that the simulations are able to capture the structural response of the flag with very good accuracy. The computed results are analyzed to elucidate the structure and dynamics of the massively separated, unsteady flow shed off the flag edges. 1 This work is supported by the US Department of Energy and the Minnesota Supercomputing Institute. 11:09AM H29.00004 The Effect of Aspect Ratio and Angle of Attack on the Transition Regions of the Inverted Flag Instability1 , JULIA COSSE, Caltech, JOHN SADER, University of Melbourne, BOYU FAN, DAEGYOUM KIM, MORY GHARIB, Caltech — The inverted flag instability occurs when a pliable plate is held parallel to a free-stream, with the leading edge free to move and the trailing edge clamped. Large-amplitude flapping is observed across a slim band of non-dimensional wind speeds. This specific boundaries of this flapping band vary greatly, depending on both the aspect ratio and the angle of attack of the plate with respect to the incoming flow. In addition, both periodic and aperiodic flapping modes exist. The frequency of the plate motion was analyzed and was found to be consistent with vortex-induced vibration. 1 This research is supported by the Gordon and Betty Moore Foundation 11:22AM H29.00005 Dynamic Chord-wise Tip Curvature on Flexible Flapping Plates1 , NATHAN MARTIN, MORTEZA GHARIB, Caltech — The aerodynamic characteristics of long rectangular flapping plates are strongly influenced by the interaction between tip and edge vortices. This has led to the development of many tip actuation mechanisms to independently bend or rotate the tip towards the root of the plate in the span-wise direction. In our current work, the influence of dynamically altering the chord-wise curvature of the tip on the generation of aerodynamic forces is investigated; for this case, the two free corners of the flat plate bend towards each other. The parameters of actuation timing, maximum curvature, Reynolds number, flexibility, and tip speed are independently varied to determine their influence. These results will further the fundamental understanding of unsteady aerodynamics. 1 This material is based upon work supported by the National Science Foundation Graduate Research Fellowship and the Gordon and Betty Moore foundation. 11:35AM H29.00006 Free-standing inflatable solar chimney: experiment and theory1 , PETER VOROBIEFF, ANDREA MAMMOLI, NIMA FATHI, The University of New Mexico, VAKHTANG PUTKARADZE, The University of Alberta — Solar chimneys (or solar updraft towers) offer an attractive way to use solar energy for production of baseload power. In a power plant of this type, sunshine heats the air under a wide greenhouse-like roofed collector surrounding the central base of a tall chimney. The heated air drives an updraft flow through the tower, whose energy is harvested with turbines. For a sufficiently large plant of this type, the thermal mass of the heated ground under the collector is sufficient to drive the flow even when the sun is down. The primary challenge in building the solar chimney power plant is the construction of the chimney that generates the updraft, which must be very tall (hundreds of meters for a commercial-sized plant). Here we present a study of an inflatable chimney which is a self-supporting, deformable, free-standing stack of gas-filled tori. The structure is stabilized via a combination of shape, overpressure, and buoyancy. Theoretical considerations suggest that filling the tori with air rather than with a light gas may be advantageous for stability. The chimney shape is optimized for deformation under wind loading. A prototype chimney has demonstrated the viability of the concept, with experimental results in good agreement with theoretical predictions. 1 This research is partially supported by the UNM Research Allocations Comittee (RAC) and UNM Center for Emerging Energy Technologies (CEET). 11:48AM H29.00007 Small-Scale Vortical Motions Effected by Aeroelastic Fluttering of a SelfOscillating Reed in a Channel Flow , SOURABH JHA, PABLO HIDALGO, ARI GLEZER, Georgia Inst of Tech — The formation, shedding, and advection of a hierarchy of small-scale vortical motions effected by an aeroelastically fluttering reed cantilevered across the span of a square channel are investigated experimentally at low (laminar or transitional) Reynolds numbers using high-resolution particle image velocimetry (PIV) and hot-wire anemometry. Formation and advection of vorticity concentrations along the surface of the reed are induced by concave/convex surface undulations associated with structural vibration modes of the reed. These modes lead to alternate time-periodic shedding of CW and CCW vortical structures having cross stream scales that are commensurate with the cross stream amplitude of the reed motion. The evolution of these vortices in the vicinity of the reed is strongly affected by interactions with the wall boundary layers that engender vorticity filaments spanning the entire height of the channel. These reciprocal interactions between the reed and the embossing channel flow leads to the evolution of small scale motions of decreasing scales that is characterized by enhanced dissipation and a distribution of spectral components that are reminiscent of a turbulent flow even at the low Reynolds number of the base flow. Supported by AFOSR. 12:01PM H29.00008 Flow-induced instabilities of shells of revolution conveying fluid , GARY HAN CHANG, Umass Amherst — In the present work, we study flow-induced instabilities of an axis-symmetric shell of revolution with an arbitrary non-uniform cross-section due to uniform or pulsatile flow. We consider a fully-coupled fluid-structure interaction model, which we solve using a method that combines the Galerkin technique with the boundary element method (BEM). Several modes in the axial direction have been used in the numerical solution and the mode number in the circumferential direction has been chosen as n = 5. As the flow velocity is increased, the system loses its stability through divergence and the shell buckles. We have also conducted experiments on shells of revolution, made of silicon rubber, conveying fluid in order to observe their flow-induced instabilities. Experimental results show that thin shells of revolution conveying fluid lose their stability by divergence with asymmetric mode-shapes, in agreement with our theoretical results. 12:14PM H29.00009 Flapping Dynamics of an Inverted Flexible Foil in a Uniform Axial Flow , PARDHA SARADHI GURUGUBELLI VENKATA, RAJEEV K. JAIMAN, National University of Singapore — This work presents a numerical study on selfinduced flapping dynamics of an inverted flexible foil in uniform flow. The inverted foil considered in this study is clamped at the trailing edge and the leading edge is allowed to oscillate. A high-order coupled FSI solver based on CFEI formulation has been used to present the flapping response results for a wide range of nondimensional bending rigidity using a fixed Reynolds number of 1000 and a mass-ratio of 0.1. As a function of bending rigidity four flapping regimes have been discovered: fixed point, inverted limit-cycle oscillation, deflected flapping, and flipped flapping. The inverted foil configuration undergoes flapping motion more readily and experiences very large amplitude oscillations than the conventional foil. A wide variety of vortex wakes with a maximum of 14 vortices per oscillation cycle have been observed. The inverted limit-cycle flapping generate novel 4P+6S (14 vortices) and 2P+6S (10 vortices) wakes. On the other hand, the flipped flapping regime is characterized by a von Kármán wake. We also observe that inverted foil can extract 1000 times more energy from the surrounding fluid compared to the conventional foil 12:27PM H29.00010 Empirical parametric study of fluid-structure interaction at high Reynolds number1 , GRANT DOWELL, MICHAEL MCPHAIL, MICHAEL KRANE, CENGIZ CAMCI, Penn State University — An experimental parametric study of a fluid-structure interaction is presented, in order to identify appropriate conditions for in-depth measurements for validation of fully-coupled fluid-structure interaction solvers. The structure is a rigid, square cross-section beam, to which is attached a thin, flexible, rectangular membrane. The rigid beam is mounted perpendicular to the incident flow, and the flexible element is mounted to the rigid element, parallel to the flow direction, so that it interacts with wake of the rigid element. Time-resolved flexible element motion is captured from two directions using high-speed video, for a range of flexible element aspect ratio, stiffness, and Reynolds number based on flow speed and rigid element dimension. Image processing was then used to characterize the frequency and amplitude of flexible element flap and twist modes. 1 Acknowledge NAVSEA Code 073R Monday, November 24, 2014 10:30AM - 12:40PM Session H30 Wind Turbines: Vertical Axis — 2016 - Matthias Kinzel, California Institute of Technology 10:30AM H30.00001 Wake visualization behind multiple VAWTs in a wind tunnel using sPIV , COLIN PARKER, MEGAN C. LEFTWICH, George Washington University — This work visualizes the wake behind multiple vertical axis wind turbines (VAWTs). The flow is visualized in a wind tunnel behind scaled model VAWTs driven at constant rotational velocity. The wake is visualized using stereo particle imaging velocimetry (sPIV) at the mid-plane downstream of the turbines. Syncing the sPIV system with the rotation of the turbine allows images to be taken at known phase angles. These images are then averaged to see the phase-averaged wake behind the VAWTs. Moving downstream, the averaged wake structure can be tracked by phase matching. Initially, data was taken in the near wake behind a single VAWT. As the blade turns normal to, and then back towards the free-stream, a vortex structure is shed into the wake and moves downstream. The out-of-plane velocity corresponding to this vortex pair shows the structures to be highly three-dimensional. Phase averaged wakes show distinct structures behind the turbine that move downstream with the free stream. Next, we measured the wake interactions behind a two turbine system. In this setup, a pair of counter rotating VAWTs is placed in the wind tunnel. We can vary the spacing and orientation between the counter rotating pair to compare changes in the downstream wind profile. 10:43AM H30.00002 A numerical investigation of the wake structure of vertical axis wind turbines , ELIAS BALARAS, ANTONIO POSA, MEGAN LEFTWICH, The George Washington University — Recent field-testing has shown that vertical axis wind turbines (VAWT) in wind farm configurations have the potential to reach higher power densities, when compared to the more widespread horizontal axis turbines. A critical component in achieving this goal is a good understanding of the wake structure and how it is influenced by operating conditions. In the present study the Large-Eddy Simulation technique is adopted to characterize the wake of a small vertical axis wind turbine and to explore its dependence on the value of its Tip Speed Ratio (TSR). It will be shown that its wake significantly differs from that of a spinning cylinder, often adopted to model this typology of machines: the displacement of the momentum deficit towards the windward side follows the same behavior, but turbulence is higher on the leeward side. An initial increase of the momentum deficit is observed moving downstream, with central peaks in the core of the near wake for both momentum and turbulent kinetic energy, especially at lower TSRs. No back-flow is produced downstream of the turbine. The interaction between blades is stronger at higher values of the TSR, while the production of coherent structures is enhanced at lower TSRs, with large rollers populating the leeward side of the wake. 10:56AM H30.00003 Velocity measurements in the wake of laboratory-scale vertical axis turbines and rotating circular cylinders , DANIEL ARAYA, JOHN DABIRI, California Institute of Technology — We present experimental data to compare the wake characteristics of a laboratory-scale vertical-axis turbine with that of a rotating circular cylinder. The cylinder is constructed to have the same diameter and height as the turbine in order to provide a comparison that is independent of the tunnel boundary conditions. Both the turbine and cylinder are motor-driven to tip-speed ratios based on previous experiments. An analysis of the effect of the motor-driven flow is also presented. These measurements are relevant for exploring the complex structure of the vertical axis turbine wake relative to the canonical wake of a circular cylinder. 2D particle image velocimetry is used to measure the velocity field in a two-dimensional plane normal to the axis of rotation. This velocity field is then used to compare time-averaged streamwise velocity, phase-averaged vorticity, and the velocity power spectrum in the wake of the two configurations. The results give insight into the extent to which solid cylinders could be used as a simplified model of the flow around vertical axis turbines in computational simulations, especially for turbine array optimization. 11:09AM H30.00004 In Situ Measurements of the Flow around a Single Vertical-Axis Wind Turbine , IAN BROWNSTEIN, DANIEL ARAYA, MATTHIAS KINZEL, JOHN DABIRI, Caltech — Laboratory studies of model vertical-axis wind turbines (VAWTs) are typically unable to match both the Reynolds number (Re) and tip speed ratio (TSR) of full-scale wind turbines. In order to match both relevant parameters, a quantitative flow visualization method was developed to take in situ measurements of the flow around full-scale VAWTs. An apparatus was constructed to deploy a horizontal sheet of smoke upstream of the turbine at the mid-span of the rotor. Quantitative results were obtained by tracking the evolution of this smoke sheet using a PIV algorithm. This method will be demonstrated through a comparative study of three- and five-bladed VAWTs at the Field Laboratory for Optimized Wind Energy (FLOWE) in Lancaster, CA. Additionally, results will be presented in comparison with previous laboratory studies to help determine the dependence of the flow physics on Re and TSR. 11:22AM H30.00005 Simulations of Vertical Axis Wind Turbine Farms in the Atmospheric Boundary Layer , SEYED HOSSEIN HEZAVEH, ELIE BOU-ZEID, Civil and Environmental Engineering Department, Princeton University, MARK LOHRY, LUIGI MARTINELLI, Mechanical and Aerospace Engineering Department, Princeton University — Wind power is an abundant and clean source of energy that is increasingly being tapped to reduce the environmental footprint of anthropogenic activities. The vertical axis wind turbine (VAWT) technology is now being revisited due to some important advantages over horizontal axis wind turbines (HAWTS) that are particularly important for farms deployed offshore or in complex terrain. In this talk, we will present the implementation and testing of an actuator line model (ALM) for VAWTs in a large eddy simulation (LES) code for the atmospheric boundary layer, with the aim of optimizing large VAWT wind farm configurations. The force coefficients needed for the ALM are here obtained from blade resolving RANS simulations of individual turbines for each configuration. Comparison to various experimental results show that the model can very successfully reproduce observed wake characteristic. The influence of VAWT design parameters such as solidity, height to radius ratio, and tip speed ratio (TSR) on these wake characteristics, particularly the velocity deficit profile, is then investigated. 11:35AM H30.00006 Comparison between Vertical-Axis Wind Turbine Arrays and Plant Canopies1 , MATTHIAS KINZEL, DANIEL ARAYA, JOHN DABIRI, California Institute of Technology — We present experimental results from three different full scale arrays of vertical-axis wind turbines (VAWTs) under natural wind conditions. One array consists of a row of four single turbines while the other two are made up of nine counter rotating turbine pairs. The wind velocities throughout the turbine arrays are measured using a portable meteorological tower with seven, vertically-staggered, three-component ultrasonic anemometers. Furthermore, the power output of each turbine is measured simultaneously with the free stream wind velocity and direction. These measurements yield detailed understanding of the aerodynamics inside the VAWT arrays and the resulting power productions. Quadrant hole analysis is employed to gain a better understanding of the vertical energy transport at the top of the VAWT array. Results comparing the energy transport and the responsible mechanisms between the larger turbine arrays and the four single turbines configuration will be presented. Furthermore, results are compared to the flow in urban and plant canopies. Emphasis is given to the flow physics in the adjustment region of the canopy, i.e. the region where the flow transitions from an atmospheric surface layer to a canopy flow. 1 This project is funded by the Gordon and Betty Moore Foundation through Grant 2645. 11:48AM H30.00007 Mean and turbulent flow development through an array of rotating elements1 , ANNA CRAIG, Stanford University, JOHN DABIRI, California Institute of Technology, JEFFREY KOSEFF, Stanford University — The adjustment of an incoming boundary layer profile as it impacts and interacts with an array of elements has received significant attention in the context of terrestrial and aquatic canopies and more recently in the context of horizontal axis wind farms. The distance required for the mean flow profile to stabilize, the energy transport through the array, and the structure of the turbulence within the array are directly dependent upon such factors as the element height, density, rigidity/flexibility, frontal area distribution, element homogeneity, and underlying topography. In the present study, the mean and turbulent development of the flow through an array of rotating elements was examined experimentally. Element rotation has been shown to drastically alter wake dynamics of single and paired elements, but the possible extension of such rotation-driven dynamics had not previously been examined on larger groups of elements. Practically, such an array of rotating elements may provide insight into the flow dynamics of an array of vertical axis wind turbines. 2D particle image velocimetry was used along the length of the array in order to examine the effects of an increasing ratio of cylinder rotation speed to streamwise freestream velocity on flow development and structure. 1 Work supported by a NSF Graduate Research Fellowship & Stanford Graduate Fellowship to A.E.C, by funding to J.O.D. from ONR N000141211047 and the Gordon and Betty Moore Foundation through Grant GBMF2645, and by funding from the EFML. 12:01PM H30.00008 Numerical investigation of the self-starting of a vertical axis wind turbine1 , HSIEH-CHEN TSAI, TIM COLONIUS, California Institute of Technology — The immersed boundary method is used to simulate the incompressible flow around two-dimensional airfoils at sub-scale Reynolds number in order to investigate the self-starting capability of a vertical-axis wind turbine (VAWT). By investigating a single blade fixed at various angle of attacks, the leading edge vortex (LEV) is shown to play an important role in the starting mechanism for both flat-plate and NACA 0018 blades. Depending on the angle of attack of the blade, as the LEV grows, the corresponding low pressure region results in a thrust in the tangential direction, which produces a positive torque to VAWT. Due to the characteristics of the blades, a NACA 0018 blade produces a larger thrust over a wider range of angle of attacks than a flat-plate blade. Therefore, a VAWT with NACA 0018 blades can self-start more easily than one with flat-plate blades. Moreover, by investigating the starting torque of three-bladed VAWTs fixed at various orientations, the optimal orientation that produces the largest torque to start both VAWTs is with a blade parallel to the flow and facing downstream. The simulations are also compared to results from companion water-tunnel experiments at Caltech. 1 This project is supported by Caltech FLOWE center/Gordon and Betty Moore Foundation. 12:14PM H30.00009 Investigation of implementation of stators on vertical axis wind turbines , AARON ALEXANDER, ARVIND SANTHANAKRISHNAN, Oklahoma State University — Vertical Axis Wind Turbines (VAWT) have historically suffered from an inability to self-start and, especially on Savonius rotors, low efficiencies due to drag on the returning blade. A few VAWT studies have examined the use of stators to direct the flow onto the power producing side of the rotor thus preventing drag on the returning side, yet all of the designs studied allow the air to exit on the downstream side of the entering flow. This study investigates an alternative stator design for extracting more wind energy by trapping the incoming flow into a rising vortex within the stator enclosure. The flow is then allowed to exit above the stator. The current study compared the performance of a generic Savonius rotor in a 7 m/s free stream flow with the same rotor in two different stator designs. The first stator design allows the flow to escape in the downstream direction. The second stator design utilizes the same stator shape, but forces the air to remain trapped until it can exit above the stators. The initial evaluation of the results was conducted using Computational Fluid Dynamics (CFD) package Star-CCM+ set up with an unsteady k-ε model at a Reynolds number of about 1,400,000. Experimental comparisons with scale models will be presented. 12:27PM H30.00010 Flow–blade interaction in a Vertical Axis Wind Turbine , ROBERTO DOMINGUEZ, SAUL PIEDRA, EDUARDO RAMOS, Universidad Nacional Autonoma de Mexico — We present an analysis of the interaction between an incoming wind and three airfoils symmetrically located, and free to rotate around a common axis. The geometrical configuration considered is a two dimensional model of Vertical Axis Wind Turbine. The model is based in the conservation equations of the fluid coupled with the Newton-Lagrange equations for the interaction with the airfoils. The presence of the rigid body in the fluid is simulated using immersed boundary conditions. The interaction of the wind with the airfoil located further upstream generates a force on the airfoil and vortices that are swept downstream and collide with the other airfoils. This effect generates a complex interplay of dynamical forces whose resultant is a torque that sets the system in motion. We describe the flow around the airfoils and examine the efficiency of the system as a function of geometric variables. Our conclusions are potentially useful for the design of VAWT’s. Monday, November 24, 2014 10:30AM - 12:40PM Session H31 CFD: General — 2018 - Kambiz Salari, Lawrence Livermore National Laboratory 10:30AM H31.00001 von Neumann Stability Analysis of Numerical Solution Schemes for 1D and 2D Euler Equations , SANTOSH KONANGI, NIKHIL KUMAR PALAKURTHI, URMILA GHIA, University of Cincinnati — A von Neumann stability analysis is conducted for numerical schemes for the full system of coupled, density-based 1D and 2D Euler equations, closed by an isentropic equation of state. The governing equations are discretized on a staggered grid, which permits equivalence to finite-volume discretization. Presently, first-order accurate spatial and temporal finite-difference techniques are analyzed. The momentum convection term is treated as explicit, semi-implicit or implicit. Density upwind bias is included in the spatial operator of the continuity equation. By combining the discretization techniques, ten solution schemes are formulated. For each scheme, unstable and stable regimes are identified through the stability analysis, and the maximum allowable CFL number is predicted. The predictions are verified for selected schemes, using the Riemann problem at incompressible and compressible Mach numbers. Very good agreement is obtained between the analytically predicted and “experimentally” observed CFL values for all cases, thereby validating the analysis. The demonstrated analysis provides an accurate indication of stability conditions for the Euler equations, in contrast to the simplistic conditions arising from model equations, such as the wave equation. 10:43AM H31.00002 Simulation of a supercritical fluid flow with extremely high temperature gradients1 , SATOKO KOMURASAKI, Nihon Univ - Tokyo — Eruption of geothermally heated water from the hydrothermal vent in deep oceans of depth over 2,000 meters is numerically simulated. The hydrostatic pressure of water is assumed to be over 200 atmospheres, and temperature of heated water is occasionally more than 300◦ C. Under these conditions, a part of heated water can be in the supercritical state, and the physical properties can change significantly by the temperature. Particularly, thermal diffusivity at the critical temperature becomes so small which prevents heat diffusion and the temperature gradients can become high. The compressible Navier-Stokes equations are solved using a method for the incompressible equations under the constant pressure. The equations are approximated by the multidirectional finite difference method, and for the highly-unsteady-flow computation, KK scheme is used to stabilize the high-accuracy computation. To treat high temperature gradients in the computation, the energy equation is solved which is derived by transformation of thermodynamic variable φ into ϕ that is ϕ = −sgnφ · log(1 − φ · sgnφ). Solving the equation about ϕ instead of φ allows the sharp boundaries of φ to be properly preserved in the computation. 1 This work was partially supported by Grant-in-Aid for Scientific Research from MEXT/JSPS (26610119). 10:56AM H31.00003 Direct numerical simulation of steady state, three dimensional, laminar flow around a wall mounted cube1 , ANASTASIOS LIAKOS, United States Naval Academy, NIKOLAOS MALAMATARIS, George Mason University / ATEI of Thessaloniki — The topology and evolution of flow around a surface mounted cubical object in three dimensional channel flow is examined for low to moderate Reynolds numbers. Direct numerical simulations were performed via a home made parallel finite element code. The computational domain has been designed according to actual laboratory experimental conditions. Analysis of the results is performed using the three dimensional theory of separation. Our findings indicate that a tornado-like vortex by the side of the cube is present for all Reynolds numbers for which flow was simulated. A horse-shoe vortex upstream from the cube was formed at Reynolds number approximately 1266. Pressure distributions are shown along with three dimensional images of the tornado-like vortex and the horseshoe vortex at selected Reynolds numbers. Finally, and in accordance to previous work, our results indicate that the upper limit for the Reynolds number for which steady state results are physically realizable is roughly 2000. 1 Financial support of author NM from the Office of Naval Research Global (ONRG-VSP, N62909-13-1-V016) is acknowledged. 11:09AM H31.00004 Shock and turbulence simulations using observable Euler and NavierStokes equations , KAMRAN MOHSENI, DOUG LIPINSKI, University of Florida — The observable divergence theorem enables a systematic derivation of high-wavenumber regularized PDEs from conservation laws. Application of this theorem to the conservation of mass, momentum, and energy produces the inviscidly regularized observable Euler equations or, after adding physical dissipation, the observable Navier-Stokes equations. This talk will first present the derivation of the observable Euler equations from basic principles and then we report results for performance of the observable Euler and observable Navier-Stokes equations in several canonical problems involving multi-dimensional shocks and/or turbulence and their interactions. The results were compared with several previously published data using Stan, Stan-I, hybrid, WENO, ADPDIS3D, and shock fitting techniques. The observable equations consistently performs as well as the best methods. 11:22AM H31.00005 Tabulated Chemistry Simulations of Thermodiffusive Instabilities in Lean Premixed Hydrogen/Air Flames , JASON SCHLUP, GUILLAUME BLANQUART, Caltech — Determining how unstable laminar flames transition from an initial perturbed planar flame to a cellular structure is an important step in understanding turbulent flame propagation and their physical mechanisms. While Direct Numerical Simulations of the turbulent reacting-flow equations complete with detailed chemical models would be ideal, the computational expense for such large scale simulations is prohibitive. To this end, tabulated chemistry models are used in this work to capture the important physical mechanisms of unsteady laminar flames. Two dimensional numerical simulations of lean hydrogen/air premixed flames are performed for a variety of domain sizes and grid resolutions. A one dimensional hydrogen/air flame serves as the initial profile, which is perturbed using a sinusoidal disturbance in the transverse direction. Additionally, detailed chemistry simulations are performed as a comparison metric for the tabulated chemistry results. Finally, the tabulated chemistry results are compared to experimental data of spherically expanding flames. 11:35AM H31.00006 Rotating Lid-Driven Cubical Cavity (RLDCC) Flow , NAGANGUDY PANCHAPAKESAN, Indian Institute of Technology Madras, Chennai, JITENDRA KUMAR, ADA, Bangalore — Fluid motion in a cubical cavity geometry driven by a rotating lid was simulated using OpenFoam software. The flow structure observed is compared with cylindrical cavity driven by rotating lid and the evolution of the flow with Reynolds number is presented. The critical Reynolds number for transition to oscillatory flow is estimated. The flow structure around the critical Reynolds number is visualized and the effects of parameters on the structure is presented. 11:48AM H31.00007 Efficient Modeling of Multicomponent Diffusive Mixing , ANDY NONAKA, Lawrence Berkeley National Laboratory, AMIT BHATTACHARJEE, Courant Institute of Mathematical Sciences, ALEJANDRO GARCIA, San Jose State University, JOHN BELL, Lawrence Berkeley National Laboratory, ALEKSANDAR DONEV, Courant Institute of Mathematical Sciences — We have developed a low Mach number hydrodynamics code appropriate for modeling diffusive mixing of an arbitrary number of fluids with different densities and transport properties. Our low Mach number formulation eliminates acoustic waves and allows for an advective CFL time step constraint. Unlike models for incompressible flow which eliminate acoustic waves by imposing that the divergence of the velocity be zero, in this formulation the divergence is determined by the mixing of the fluids. We couple the divergence constraint to an implicit viscosity treatment using a newly-developed staggered-grid, finite-volume Stokes solver with a projection-method based preconditioner. Our code supports multiple time-stepping schemes suitable for both inertial and large Schmidt number regimes. The code has been implemented in the highly-scalable BoxLib software framework publicly available at Lawrence Berkeley Laboratory. The code also contains modules for thermal fluctuations using stochastic forcing terms as proposed by Landau and Lifshitz. We have successfully used the code to replicate multi-mode instabilities observed in experiments. 12:01PM H31.00008 Towards numerical simulations of fluid-structure interactions for investigation of obstructive sleep apnea , CHIEN-JUNG HUANG, SUSAN M. WHITE, SHAO-CHING HUANG, SANJAY MALLYA, JEFF D. ELDREDGE, Univ of California - Los Angeles — Obstructive sleep apnea(OSA) is a medical condition characterized by repetitive partial or complete occlusion of the airway during sleep. The soft tissues in the airway of OSA patients are prone to collapse under the low pressure loads incurred during breathing. The numerical simulation with patient-specific upper airway model can provide assistance for diagnosis and treatment assessment. The eventual goal of this research is the development of numerical tool for air-tissue interactions in the upper airway of patients with OSA. This tool is expected to capture collapse of the airway in respiratory flow conditions, as well as the effects of various treatment protocols. Here, we present our ongoing progress toward this goal. A sharp-interface embedded boundary method is used on Cartesian grids for resolving the air-tissue interface in the complex patient-specific airway geometries. For the structure simulation, a cut-cell FEM is used. Non-linear Green strains are used for properly resolving the large tissue displacements in the soft palate structures. The fluid and structure solvers are strongly coupled. Preliminary results will be shown, including flow simulation inside the 3D rigid upper airway of patients with OSA, and several validation problem for the fluid-structure coupling. 12:14PM H31.00009 How grid quality affects solution accuracy , JOHN RHOADS, Pointwise, Inc — Finite volume simulations ranging from RANS to LES inherently require some discretization of the fluid region being examined, which is in many ways dictated by the numerics employed to represent Navier-Stokes. However, for generic finite volume codes, the presence of poor quality elements can lead to difficulty in obtaining a converged solution and, perhaps even worse, significant non-physical artifacts in a converged solution. A fundamental set of examples in two and three dimensions will be presented in order to demonstrate these effects and how to avoid them. Additionally, results for a benchmark geometry from a case study on grid effects will also be discussed. 12:27PM H31.00010 Multiscale Modeling Facilitated by the Dual Domain Material Point Method1 , DUAN ZHANG, TILAK DHAKAL, Los Alamos National Laboratory — Many current multiscale methods are based on Eulerian descriptions, which have well-known difficulties when applied to problems of large material deformation and history dependency. Unfortunately, many problems requiring the consideration of multiscale and non-equilibrium thermodynamic effects are in this category. In this talk, a general multiscale approach is introduced to study material responses at different scales. The up-scaling approach uses the ensemble averaging technique. The required closures for the averaged equations are expressed in terms of lower scale quantities, which can be evaluated directly using numerical simulations following the motion of the material. The required history tracking can be achieved efficiently using the dual domain material point (DDMP) method because of the Lagrangian nature of the material points. The DDMP method requires only communications between material points and mesh nodes and no communication between material points. Therefore the response of the material represented by each material point can be numerically simulated independently in parallel computers with high efficiency. Applications of this approach to materials undergoing rapid and large deformation are demonstrated. 1 Work sponsored by DOE. Monday, November 24, 2014 10:30AM - 12:40PM Session H32 Particle-Laden Flows: Non-Spherical Particles — 2020 - Alfredo Soldati, Universita Degli Studi di Udine 10:30AM H32.00001 Consequences of a double zero eigenvalue for the rotational motion of a prolate spheroid in shear flow , TOMAS ROSEN, Linne FLOW Centre and Wallenberg Wood Science Center, KTH Mechanics, Stockholm, Sweden, ARNE NORDMARK, KTH Mechanics, Stockholm, Sweden, CYRUS K. AIDUN, G.W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA, FREDRIK LUNDELL, Linne FLOW Centre and Wallenberg Wood Science Center, KTH Mechanics, Stockholm, Sweden — The rotational motion of a single prolate spheroidal particle in a linear shear flow provides fundamental knowledge necessary to understand both rheology and orientation distributions of suspensions including elongated particles. In this work, we present results from both direct numerical simulations using the lattice Boltzmann method and stability analysis using Comsol Multiphysics. It has been found that particle and fluid inertia cause different stable rotational states.1,2 The transitions between these originate from the fact that the Log-rolling particle (particle aligned with vorticity) has a double zero eigenvalue for a certain choice of particle Reynolds number Rep = ReP F and solid-to-fluid density ratio α = αDZ . The consequence is that particles with α > αDZ , will always go to a planar rotation (symmetry axis perpendicular to vorticity), while lighter particles can assume a stable periodic or chaotic state which is non-planar. Since αDZ is decreasing with aspect ratio, we find further that only planar states exist for particles of low aspect ratio (length/width). 1 Rosén 2 Rosén et al., J. Fluid Mech. 738, 563 (2013). et al., J. Fluid Mech., submitted (2014). 10:43AM H32.00002 Chaotic orbits tracked by a 3D asymmetric immersed solid at high Reynolds numbers using a novel Gerris-Immersed Solid (DNS) Solver , PEI SHUI, Institute for Materials and Processes, School of Engineering, The University of Edinburgh, United Kindom, STÉPHANE POPINET, National Institute of Water and Atmospheric Research, P.O. Box 14-901, Kilbirnie, Wellington, New Zealand, PRASHANT VALLURI, Institute for Materials and Processes, School of Engineering, The University of Edinburgh, United Kindom, RAMA GOVINDARAJAN, TIFR-Hyderabad, Narsingi, Hyderabad 500075, India — The motion of a neutrally buoyant ellipsoidal solid with an initial momentum has been theoretically predicted to be chaotic in inviscid flow by Aref (1993). On the other hand, the particle could stop moving when the damping viscous force is strong enough. This work provides numerical evidence for 3D chaotic motion of a neutrally buoyant general ellipsoidal solid and suggests criteria for triggering this motion. The study also shows that the translational/rotational energy ratio plays the key role on the motion pattern, while the particle geometry and density aspect ratios also have some influence on the chaotic behaviour. We have developed a novel variant of the immersed solid solver under the framework of the Gerris flow package of Popinet et al. (2003). Our solid solver, the Gerris Immersed Solid Solver (GISS), is capable of handling 6 degree-of-freedom motion of particles with arbitrary geometry and number in three-dimensions and can precisely predict the hydrodynamic interactions and their effects on particle trajectories. The reliability and accuracy have been checked by a series of classical studies, testing both translational and rotational motions with a vast range of flow properties. 10:56AM H32.00003 Rotational motion of elongated particles in isotropic turbulent flow: statistical perspective1 , LIHAO ZHAO, HELGE ANDERSSON, Norwegian University of Science and Technolog, EVAN VARIANO, University of California, Berkeley — We consider the rotational motion of non-spherical particles in turbulent flow, comparing the statistics of particles’ angular velocity to the corresponding quantities computed in the fluid phase. We use numerical (DNS) and laboratory measurements for particles that are both larger and smaller than the Kolmogorov lengthscale. The particles are spheroids or rods, with aspect ratios between 1 and 10. We will discuss the subtleties of defining a meaningful Stokes number for these particles, focusing on the effect of asphericity and the fact that our interest is in rotation and not translation. Comparing the probability density function of angular velocity between fluid and particle phase indicates that the angular velocity of particles has a narrower distribution than that of the fluid phase, and that. particles do respond to extreme events in the fluid phase. The first four moments of the PDFs are analyzed, and these show that the “filtering” effect is very similar between DNS and lab experiments, despite differences in particle sizes and mass. We propose a nondimensional curve for predicting the magnitude of the filtering effect, and discuss the implications of this curve for the definition of Stokes number, as discussed earlier. 1 This work has been supported by grants from the Peder Sather Center for Advanced Study at UC Berkeley and from the Research Council of Norway (Contract No. 213917/F20). 11:09AM H32.00004 Flow of a flexible fiber past an obstacle , HECTOR LOPEZ, JEAN PIERRE HULIN, HAROLD AURADOU, FAST, Université Paris Sud, Orsay, France, VERONICA D’ANGELO, Grupo de Medios Porosos, FIUBA, Argentina — The transport of flexible biological or man made fibers by a flow is of interest in view of their potential applications in many different industrial fields. Here we study the deformation and transport of elastic fibers in a viscous Hele-Shaw flow with curved streamlines. The variations of the global velocity and orientation of the fiber follow closely those of the local flow velocity. The ratios of the curvatures of the fibers by the corresponding curvatures of the streamlines reflect a balance between elastic and viscous forces: this ratio is shown experimentally to be determined by a dimensionless Sperm number (Sp) combining the characteristic parameters of the flow (transverse velocity gradient, viscosity, fiber diameter/cell gap ratio) and those of the fiber (diameter, effective length, Young’s modulus). The effective length is either the fiber length (short fiber case) or the characteristic size of the obstacle (long fiber case). For low values of Sp the ratio of the curvatures increases linearly with Sp. For values higher than 250, the fiber and the streamlines have the same curvature. 11:22AM H32.00005 Particle motion inside Ekman and Bödewadt boundary layers1 , MATIAS DURAN MATUTE, STEVEN VAN DER LINDEN, GERTJAN VAN HEIJST, Eindhoven University of Technology — We present results from both laboratory experiments and numerical simulations of the motion of heavy particles inside Ekman and Bödewadt boundary layers. The particles are initially at rest on the bottom of a rotating cylinder filled with water and with its axis parallel to the axis of rotation. The particles are set into motion by suddenly diminishing the rotation rate and the subsequent creation of a swirl flow with the boundary layer above the bottom plate. We consider both spherical and non-spherical particles with their size of the same order as the boundary layer thickness. It was found that the particle trajectories define a clear logarithmic spiral with its shape depending on the different parameters of the problem. Numerical simulations show good agreement with experiments and help explain the motion of the particles. 1 This research is funded by NWO (the Netherlands) through the VENI grant 863.13.022 11:35AM H32.00006 The effect of shear flow on the rotational diffusivity of a single axisymmetric particle , BRIAN LEAHY, Department of Physics, Cornell University, DONALD KOCH, Chemical and Biomolecular Engineering, Cornell University, ITAI COHEN, Department of Physics, Cornell University — Colloidal suspensions of nonspherical particles abound in the world around us, from red blood cells in arteries to kaolinite discs in clay. Understanding the orientation dynamics of these particles is important for suspension rheology and particle self-assembly. However, even for the simplest case of dilute suspensions in simple shear flow, the orientation dynamics of Brownian nonspherical particles are poorly understood at large shear rates. Here, we analytically calculate the time-dependent orientation distributions of particles confined to the flow-gradient plane when the rotary diffusion is small but nonzero. For both startup and oscillatory shear flows, we find a coordinate change that maps the convection-diffusion equation to a simple diffusion equation with an enhanced diffusion constant, simplifying the orientation dynamics. For oscillatory shear, this enhanced diffusion drastically alters the quasi-steady orientation distributions. Our theory of the unsteady orientation dynamics provides an understanding of a nonspherical particle suspension’s rheology for a large class of unsteady flows. For particles with aspect ratio 10 under oscillatory shear, the rotary diffusion and intrinsic viscosity vary with amplitude by a factor of ≈ 40 and ≈ 2, respectively. 11:48AM H32.00007 Non-spherical aerosol transport under oscillatory shear flows at lowReynolds numbers , LIHI SHACHAR BERMAN, YANN DELORME, PHILIPP HOFEMEIER, STEVEN FRANKEL, JOSUE SZNITMAN, None — Most airborne particles are intrinsically non-spherical. In particular, non-spherical particles with high aspect ratios, such as fibers, are acknowledged to be more hazardous than their spherical counterparts due to their ability to penetrate into deeper lung regions, causing serious pulmonary diseases. Not only do particle properties such as size, shape, and density have a major impact on particle transport, for non-spherical aerosols, their orientations also greatly influence particle trajectories due to modified lift and drag characteristics. Until present, however, most of our understanding of the dynamics of inhaled particles in the deep airways of the lungs has been limited to spherical particles only. In the present work, we seek to quantify through numerical simulations the transport of non-spherical airborne particles and their deposition under oscillatory shear flows at low Reynolds numbers, characteristic of acinar airways. Here, the Euler–Lagrangian model is used to solve the translational movement of a fiber, whereas the Eulerian rotational equations are introduced and solved to predict detailed unsteady fiber orientations. Overall, our efforts provide new insight into realistic dynamics of inhaled non-spherical aerosols under characteristic breathing motions. 12:01PM H32.00008 Relative motion between rigid fibers and fluid in turbulent channel flow1 , CRISTIAN MARCHIOLI, ALFREDO SOLDATI, University of Udine — We examine how rigid fibers with different length and inertia translate and rotate relative to the surrounding fluid in presence of non-linear mean shear and flow anisotropy. Two observables will be investigated: the fiber-to-fluid translational velocity (slip velocity) and angular velocity (slip spin). Slip velocity and slip spin statistics are extracted from DNS of turbulence at Reτ = 150 coupled with Lagrangian tracking of prolate ellipsoids with Stokes number 1 < St < 100, and aspect ratio 1 < λ < 50. We find that elongation has quantitative effects on the statistics, particularly for fibers with small St. As St increases, differences due to the aspect ration vanish and the relative motion is controlled by fiber inertia through preferential concentration. Inertial effects show up in the different distribution of slip velocities observed when fibers are trapped in sweeps/ejections or segregated in near-wall fluid streaks. The corresponding conditional PDFs are found to be non-Gaussian, suggesting that relative motions cannot be modeled as a standard diffusion process at steady state. This is evident in the strong shear region, where fiber anisotropy adds to flow anisotropy to induce strong deviations on fiber dynamics with respect to spherical particles. 1 COST Action FP1005 “Fiber suspension flow modelling” is gratefully acknowledged. 12:14PM H32.00009 Measuring the orientation and rotation rate of 3D printed particles in turbulent flow1 , GREG VOTH, GUY G. MARCUS, SHIMA PARSA, STEFAN KRAMEL, RUI NI, BRENDAN COLE, Wesleyan University — The orientation distribution and rotations of anisotropic particles plays a key role in many applications ranging from icy clouds to papermaking and drag reduction in pipe flow. Experimental access to time resolved orientations of anisotropic particles has not been easy to achieve. We have found that 3D printing technology can be used to fabricate a wide range of particle shapes with smallest dimension down to 300 µm. So far we have studied rods, crosses, jacks, tetrads, and helical shapes. We extract the particle orientations from stereoscopic video images using a method of least squares optimization in Euler angle space. We find that in turbulence the orientation and rotation rate of many particles can be understood using a simple picture of alignment of both the vorticity and a long axis of the particle with the Lagrangian stretching direction of the flow. 1 This research is supported by NSF grant DMR-1208990 12:27PM H32.00010 Spin, slip, and settle: effects of shape on motion for Taylor-scale particles in homogeneous isotropic turbulence , MARGARET BYRON, YIHENG TAO, ISABEL HOUGHTON, EVAN VARIANO, University of California Berkeley — We fabricate hydrogel cylinders of varying aspect ratios and suspend them in homogeneous isotropic turbulence at high Reynolds number. Cylinders are nearly neutrally buoyant and refractive-index-matched to water, with characteristic lengthscales that are close to the Taylor microscale. We simultaneously image these cylinders and the surrounding fluid for stereoscopic PIV measurement, permitting calculation of instantaneous particle slip velocity. We measure the particles’ settling velocity in quiescent flow and compare this to both the calculated slip velocities and empirically-predicted settling velocities. Particle rotation is determined via the solid-body rotation equation and compared with fluid-phase properties (vorticity, shear, et al). We find that the aspect ratio of the cylinder has only a weak effect on its expected value of angular velocity magnitude, and further examine the influence of aspect ratio on slip and settling velocities. Lastly, we discuss applications of our results to problems of underwater navigation in aquatic organisms. Monday, November 24, 2014 10:30AM - 12:27PM Session H33 Turbulent Multiphase Flows — 2022 - Antonio Ferrante, University of Washington 10:30AM H33.00001 Exploration of turbulence/surface tension interaction through direct numerical simulation , JEREMY MCCASLIN, OLIVIER DESJARDINS, Cornell University — Two canonical multiphase flows are constructed that provide a platform for a statistical description of surface tension effects on a surrounding turbulent flow field. In the first flow, short-time behavior is studied by inserting an initially flat interface into a triply periodic box of decaying homogeneous isotropic turbulence (HIT). Long-time behavior is studied in the second flow by inserting a randomly distributed interface into forced HIT. Simulations are performed for a variety of turbulent Reynolds and Weber numbers, including an infinite Weber number (no surface tension), on mesh sizes ranging from 2563 to 10243 . The interaction between fluid inertia and the surface tension force is isolated by utilizing unity density and viscosity ratios. The probability density function of principal curvature and global interface statistics are presented and discussed, highlighting the importance of the Kolmogorov critical radius on the spatial scales of interfacial corrugations that form. A spectral analysis of energy transfer is conducted, shedding light on the role played by surface tension in this process. 10:43AM H33.00002 Direct numerical simulation of interfacial wave generation in turbulent gas-liquid flows in horizontal channels , BRYCE CAMPBELL, KELLI HENDRICKSON, YUMING LIU, Massachusetts Inst of Tech-MIT, HARIPRASAD SUBRAMANI, Chevron Energy Technology Company — For gas-liquid flows through pipes and channels, a flow regime (referred to as slug flow) may occur when waves form at the interface of a stratified flow and grow until they bridge the pipe diameter trapping large elongated gas bubbles within the liquid. Slug formation is often accompanied by strong nonlinear wave-wave interactions, wave breaking, and gas entrainment. This work numerically investigates the fully nonlinear interfacial evolution of a two-phase density/viscosity stratified flow through a horizontal channel. A Navier-Stokes flow solver coupled with a conservative volume-of-fluid algorithm is use to carry out high resolution three-dimensional simulations of a turbulent gas flowing over laminar (or turbulent) liquid layers. The analysis of such flows over a range of gas and liquid Reynolds numbers permits the characterization of the interfacial stresses and turbulent flow statistics allowing for the development of physics-based models that approximate the coupled interfacial-turbulent interactions and supplement the heuristic models built into existing industrial slug simulators. 10:56AM H33.00003 Phase segregation in multiphase turbulent channel flow1 , FEDERICO BIANCO, ALFREDO SOLDATI, Dep. Elect., Manag. and Mechanical Engineering, University of Udine — The phase segregation of a rapidly quenched mixture (namely spinodal decomposition) is numerically investigated. A phase field approach is considered. Direct numerical simulation of the coupled Navier-Stokes and Cahn-Hilliard equations is performed with spectral accuracy and focus has been put on domain growth scaling laws, in a wide range of regimes. The numerical method has been first validated against well known results of literature, then spinodal decomposition in a turbulent bounded flow (channel flow) has been considered. As for homogeneous isotropic case, turbulent fluctuations suppress the segregation process when surface tension at the interfaces is relatively low (namely low Weber number regimes). For these regimes, segregated domains size reaches a statistically steady state due to mixing and break-up phenomena. In contrast with homogenous and isotropic turbulence, the presence of mean shear, leads to a typical domain size that show a wall-distance dependence. Finally, preliminary results on the effects to the drag forces at the wall, due to phase segregation, have been discussed. 1 Regione FVG, program PAR-FSC 11:09AM H33.00004 Destabilization of a liquid-gas interface at supercritical pressure1 , GUILHEM LACAZE, ANTHONY RUIZ, JOSEPH OEFELEIN, Sandia National Laboratories — To improve efficiency, advanced propulsion systems are operated at high pressure. In many cases the pressure exceeds the critical pressure of the fuel and oxidizer, which leads to radical changes in mixing. Even though this transition is understood theoretically, many important questions remain. One is the impact of the strong interfacial density-gradient on destabilization of the shear layer. At these conditions, experimental imaging techniques fail to provide the resolution required for detailed analysis of the flow structures. In this work, we use Large Eddy Simulation to study these structures in a three-dimensional turbulent mixing layer at a Reynolds number of 500,000. A splitter separates streams of liquid oxygen and gaseous hydrogen. In the last decade, similar conditions have been studied using two-dimensional computational domains. This work is the first attempt to simulate a three-dimensional flow at these conditions with this level of resolution. Simulation results provide new insights on the destabilization processes of the liquid interface. Dynamic instabilities leading to turbulence are enhanced by inhomogeneities in density through baroclinic effects and high shear in the interfacial region. 1 The U. S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences supported this work. 11:22AM H33.00005 Modulation of isotropic turbulence by deformable droplets of Taylor lengthscale size1 , MICHAEL DODD, ANTONINO FERRANTE, University of Washington — We investigate the effects of finite-size deformable droplets on decaying isotropic turbulence via direct numerical simulation (DNS). DNS is performed using the two-fluid pressure-correction method by Dodd and Ferrante [J. Comput. Phys. 273 (2014) 416–434] coupled with the volume of fluid method by Baraldi et al. [Comput. & Fluids 96 (2014) 322–337]. We fully-resolve the flow around and inside 3130 droplets of Taylor lengthscale size, resulting in a droplet volume fraction of 0.05. The initial Taylor lengthscale Reynolds number is Reλ0 = 75, and the computational mesh has 10243 grid points. We analyze the effects on turbulence modulation of varying the droplet- to carrier-fluid viscosity ratio (1 ≤ µd /µc ≤ 100) and the droplet Weber number based on the r.m.s velocity of turbulence (0.1 ≤ Werms ≤ 5). We discuss how varying these parameters affects the turbulence kinetic energy budget, and explain the physical mechanisms for such modulation. 1 This work was supported by the National Science Foundation CAREER Award, grant number OCI-1054591. 11:35AM H33.00006 Coalescence and break-up of large droplets in turbulent channel flow1 , LUCA SCARBOLO, FEDERICO BIANCO, ALFREDO SOLDATI, Dep. Elect., Manag. and Mechanical Engineering, University of Udine — The behaviour of large, deformable and coalescing droplets, released in a turbulent channel flow, has been numerically investigated with a Phase Field approach; focus has been put on droplet-droplet interactions and droplet fragmentation, enhanced by turbulent fluctuations. Two different dynamics of the dispersed phase may be observed depending on the Weber number (W e). For small W e surface tension balances turbulent shear; slightly deformed droplets, transported by the carrying fluid, may only coalesce. The analysis of the droplet pair distance shows that the geometrical separation of the droplets is a leading factor for the coalescence regime determination. On the contrary, if W e is larger than a critical value, a dynamic equilibrium between coalescences and breakups is shown. In this regime, in line with seminal work of Hinze (1955) and recent numerical simulations of Perlekar et al. (2012), W e controls the critical stable diameter of droplets as well as the average distance between droplet pairs. 1 Regione FVG, program PAR-FSC 11:48AM H33.00007 Multiphase turbulence modeling of the flow in the wake of a transom stern , KELLI HENDRICKSON, Massachusetts Inst of Tech-MIT, SANKHA BANERJEE, None, DICK YUE, Massachusetts Inst of Tech-MIT — The objective of this effort is to develop and assess multiphase turbulence closure models for incompressible highly variable density turbulent (IHVDT) flows such as the two-phase flow in the wake of a transom stern. These flows, which have an Atwood number At = (ρ2 − ρ1 )/(ρ2 + ρ1 ) ≈ 1, are characterized by significant turbulent mass flux for which there is little guidance in turbulence closure modeling for both the momentum and the scalar transport. In this work, high-resolution numerical simulations are performed on the wake of a canonical transom stern at large scales using conservative Volume-of-Fluid (cVOF) and implicit Large Eddy Simulation (iLES). Boundary Data Immersion Method (BDIM) is used to simulate the dry transom stern wake region at three different Froude numbers and two different effective viscosities. Analysis of the simulation results for the turbulent anisotropy, turbulent kinetic energy and turbulent mass flux budget, as well as a priori closure model testing will be presented. 12:01PM H33.00008 The Effect of Nose Shape on Water-Entry Cavity Formation , JEREMY ELLIS, TADD TRUSCOTT, Brigham Young University — We examine the effect of nose shape and wetting angle on the threshold velocity at which an underwater cavity will form in the wake of a slender axisymmetric rigid body. The study covers a range of Reynolds numbers (1E4 < Re < 1.5E5), wetting surface conditions (hydrophilic, hydrophobic and super-hydrophobic), and impacting nose shapes (cone, ogive, flat, and two concave profiles: cusp and hemisphere). Cavity formation is visualized using high-speed cameras and impact forces were determined using an embedded inertial measurement unit (IMU). More streamlined nose shapes require higher impact velocities in order to form a cavity. However, the concave profiles generate a uniquely different cavity due to a strong vortex ring formed in front of the nose at impact. Additional results of this experiment and variable dependence on threshold velocity will be presented during the water entry of rigid slender bodies. 12:14PM H33.00009 Growth of gravity-capillary waves in countercurrent air/water turbulent flow1 , FRANCESCO ZONTA, ALFREDO SOLDATI, University of Udine, MIGUEL ONORATO, University of Torino — Mass, momentum and energy transport phenomena through a deformable air-water interface are important in many geophysical processes and industrial applications. In this study we use Direct Numerical Simulations (DNS) to explore the dynamics of countercurrent air/water flow. The motion of the air/water interface is computed by solving an advection equation for the interface vertical elevation (boundary fitted method). At each time step, the physical domain is mapped into a rectangular domain using a nonorthogonal transformation. Continuity and Navier-Stokes equations are first solved separately in each domain, then coupled (velocity/stress) at the interface. DNS are performed in the Weber, Froude and Reynolds number (We,Fr,Re) parameter space. Regardless of Re, Fr and We, the process of wave generation is driven by surface tension and follows a universal scaling (t2/5 ). Later in time, the waves growth rate differs depending on the value of Fr,We: for small capillary waves, we don not observe substantial changes from t2/5 law; for larger and longer waves (gravity waves) we observe a faster growth rate. We also derive simple phenomenological models to explain our results. 1 Financial support by PAR-FSC 2007/2013-UBE Monday, November 24, 2014 10:30AM - 12:40PM Session H34 LES and Modeling of Turbulent Combustion — 2024 - Joe Ofelein, Sandia National Laboratories 10:30AM H34.00001 Analysis of operator splitting errors for DNS of low Mach number turbulent reacting flows , JONATHAN MACART, MICHAEL E. MUELLER, Princeton University — A formally second-order accurate Strang splitting approach has been developed and applied to the solution of scalar transport/reaction equations for Direct Numerical Simulation (DNS) of low Mach number turbulent reacting flows. The temporal discretization errors of the scheme are analyzed in both the asymptotic and non-asymptotic regimes of convergence and compared with a formally first-order accurate Lie splitting approach in a series of one-dimensional test problems with real combustion chemistry. The Strang splitting scheme is demonstrated to achieve its theoretical accuracy when all relevant chemical time scales are resolved; however, with larger time steps representative of those utilized in practice for low Mach number DNS of turbulent reacting flows, a reduction in order is observed. Nonetheless, the Strang splitting approach exhibits a higher order of accuracy and smaller errors than Lie splitting for all time steps. Preliminary DNS results for a turbulent planar jet computed with this scheme will also be discussed. 10:43AM H34.00002 Optimal Numerical Schemes for Compressible Large Eddy Simulations1 , AYABOE EDOH, ANN KARAGOZIAN, UCLA, VENKATESWARAN SANKARAN, Air Force Research Laboratory, CHARLES MERKLE, Purdue University — The design of optimal numerical schemes for subgrid scale (SGS) models in LES of reactive flows remains an area of continuing challenge. It has been shown that significant differences in solution can arise due to the choice of the SGS model’s numerical scheme and its inherent dissipation properties, which can be exacerbated in combustion computations.2 This presentation considers the individual roles of artificial dissipation, filtering,3 secondary conservation4 (Kinetic Energy Preservation), and collocated versus staggered grid arrangements with respect to the dissipation and dispersion characteristics and their overall impact on the robustness and accuracy for time-dependent simulations of relevance to reacting and non-reacting LES. We utilize von Neumann stability analysis in order to quantify these effects and to determine the relative strengths and weaknesses of the different approaches. 1 Distribution A: Approved for public release, distribution unlimited. Supported by AFOSR (PM: Dr. F. Fahroo) Sankaran and Soterioiu, AIAA 2013-0170 3 Kennedy and Carpenter, App. Num. Math.,14, 397-433, 1994 4 Subbareddy and Candler, J. Comp. Phys., 228,1347-1364, 2009 2 Cocks, 10:56AM H34.00003 A Trust-Region Constrained Fidelity Adaptive Combustion Model , YEE CHEE SEE, HAO WU, QING WANG, MATTHIAS IHME, Stanford University — A general framework is developed to dynamically adapt the local fidelity of the combustion model for reacting flows. This framework combines a hierarchy of combustion models with different fidelity, and the adaptation is achieved by dynamically assigning a combustion model under consideration of their accuracy and computational cost. The usage of each model is confined to the trust region whose size is specified by the user. The applicability of a certain combustion model is determined by the compatibility between its manifold and the local flow field. By doing so, it becomes possible to conduct a reacting flow simulation, in which fidelity and cost are subject to user-specific requirements, and prior knowledge about the combustion regime is not necessary. This fidelity-adaptive model is applied to a triple flame to demonstrate its capability and the model performance is assessed through direct comparisons against a detailed numerical simulation. 11:09AM H34.00004 Consideration of Turbulence Effects in One-Dimensional Laminar Flamelet Equations1 , WAI LEE CHAN, Univ of Michigan - Ann Arbor, MATTHIAS IHME, Stanford University — The laminar flamelet formulation has been used as a fundamental building block for the construction of turbulent combustion closures. By assuming that turbulence only leads to a deformation and straining of the local flame structure, the turbulence/chemistry interaction is then considered through a presumed shape probability density function (PDF) approach. However, the consistency of this approach remains unclear in the context of large-eddy simulations (LES), and the objective of this study is to examine the representation of turbulent scalar fluxes and turbulence/chemistry coupling on the flame structure. To this end, a detailed numerical simulation of a turbulent counterflow diffusion flame is performed, and the simulation results are used to analyze the limitations of the classic laminar flamelet formulation and explore a possible alternative approach. 1 Financial support through the Air Force Office of Scientific Research under Award No. FA9550-11-1-0031 is gratefully acknowledged. 11:22AM H34.00005 Chemical Source Term Closure in Turbulent Combustion using Approximate Deconvolution Methods1 , QING WANG, Stanford Univ — A closure model for the chemical source term in Large Eddy Simulation (LES) using the Approximate Deconvolution Method (ADM) is proposed. The model recovers the scalar field that is discarded by the LES filter from the information retained in the large-scale structures using an approximate deconvolution operator. The nonlinear chemical source term is then evaluated based on the de-convoluted scalar field. Since this formulation makes no presumptions on the combustion regime, it is applicable to complex combustion configurations and detailed chemistry. The capability of this sub-grid closure model is examined in an a priori study, and the performance, accuracy, and computational cost are characterized through a posteriori simulations. 1 This material is based upon work supported under a Stanford Graduate Fellowship. 11:35AM H34.00006 A priori DNS evaluation of the shadow-position mixing model in turbulent reactive flows , XINYU ZHAO, Combustion Energy Frontier Research Center, Princeton University, ANKIT BHAGATWALA, JACQUELINE CHEN, Combustion Research Facility, Sandia National Laboratories, DANIEL HAWORTH, Department of Mechanical and Nuclear Engineering, Pennsylvania State University, STEPHEN POPE, Sibley School of Mechanical and Aerospace Engineering, Cornell University — The modeling of molecular diffusion of chemical species is an important aspect of modeling turbulent reactive flows, especially for transported probability density function based methods. In this work, shadow-position mixing model (SPMM) is examined, using the DNS database of a temporally-evolving di-methyl ether jet flame undergoing local extinction and re-ignition. SPMM is similar to the conventional interaction by exchange with the mean (IEM) model, with the exception that there is an additional conditioning variable, the so-called “shadow displacement.” Turbulent statistics and the shadow displacement are first extracted from the DNS database. Based on the position, time and shadow displacement, the conditional species diffusion from DNS and from SPMM are calculated and compared for several major and minor species. Possible values of model constants are then derived from the comparison of the conditional diffusion. Finally, the relation of SPMM with the IEM model, and the relation of SPMM with interaction by exchange with the conditional mean model, are explored and discussed. 11:48AM H34.00007 Large Eddy Simulations of Colorless Distributed Combustion Systems , HUSAM F. ABDULRAHMAN, FARHAD JABERI, Michigan State University, ASHWANI GUPTA, University of Maryland — Development of efficient and low-emission colorless distributed combustion (CDC) systems for gas turbine applications require careful examination of the role of various flow and combustion parameters. Numerical simulations of CDC in a laboratory-scale combustor have been conducted to carefully examine the effects of these parameters on the CDC. The computational model is based on a hybrid modeling approach combining large eddy simulation (LES) with the filtered mass density function (FMDF) equations, solved with high order numerical methods and complex chemical kinetics. The simulated combustor operates based on the principle of high temperature air combustion (HiTAC) and has shown to significantly reduce the NOx, and CO emissions while improving the reaction pattern factor and stability without using any flame stabilizer and with low pressure drop and noise. The focus of the current work is to investigate the mixing of air and hydrocarbon fuels and the non-premixed and premixed reactions within the combustor by the LES/FMDF with the reduced chemical kinetic mechanisms for the same flow conditions and configurations investigated experimentally. The main goal is to develop better CDC with higher mixing and efficiency, ultra-low emission levels and optimum residence time. The computational results establish the consistency and the reliability of LES/FMDF and its Lagrangian-Eulerian numerical methodology. 12:01PM H34.00008 DNS and LES/FMDF of turbulent jet ignition and combustion , ABDOULAHAD VALIDI, FARHAD JABERI, Michigan State University — The ignition and combustion of lean fuel-air mixtures by a turbulent jet flow of hot combustion products injected into various geometries are studied by high fidelity numerical models. Turbulent jet ignition (TJI) is an efficient method for starting and controlling the combustion in complex propulsion systems and engines. The TJI and combustion of hydrogen and propane in various flow configurations are simulated with the direct numerical simulation (DNS) and the hybrid large eddy simulation/filtered mass density function (LES/FMDF) models. In the LES/FMDF model, the filtered form of the compressible Navier-Stokes equations are solved with a high-order finite difference scheme for the turbulent velocity and the FMDF transport equation is solved with a Lagrangian stochastic method to obtain the scalar field. The DNS and LES/FMDF data are used to study the physics of TJI and combustion for different turbulent jet igniter and gas mixture conditions. The results show the very complex and different behavior of the turbulence and the flame structure at different jet equivalence ratios. 12:14PM H34.00009 Study of Entropy Generation in Turbulent Jet Flames Using Large Eddy Simulation , MEHDI SAFARI, Miami University, REZA H. SHEIKHI, Northeastern University — Analysis of local entropy generation is an effective means to investigate sources of irreversibility in turbulent combustion. Large eddy simulation (LES) is employed to describe transport of entropy in turbulent reacting flows. The filtered form of this equation includes entropy production due to viscous dissipation, heat conduction, mass diffusion and chemical reaction, all of which appear as unclosed terms. The SGS effects are taken into account using a methodology based on the filtered density function (FDF). This methodology, entitled entropy FDF (En-FDF), is developed and utilized in the form of scalar-entropy FDF transport equation. This equation is modeled by a set of stochastic differential equations. The modeled En-FDF transport equation is solved by a Lagrangian Monte Carlo method. The methodology is employed for LES of a turbulent nonpremixed jet flame at several flow parameters. The main advantage of the En-FDF is that it provides closure for all individual entropy generation effects. It also includes the effect of chemical reaction in a closed form. Predictions show good agreements with the experimental data. Entropy generation effects are predicted by the En-FDF and analyzed. The sensitivity of entropy generation to flow parameters are investigated. 12:27PM H34.00010 CFD Modeling of a Laser-Induced Ethane Pyrolysis in a Wall-less Reactor1 , OLGA STADNICHENKO, VALERIY SNYTNIKOV, Borsekov Institute of Catalysis, JUNFENG YANG, OMAR MATAR, Imperial College London — Ethylene, as the most important feedstock, is widely used in chemical industry to produce various rubbers, plastics and synthetics. A recent study found the IR-laser irradiation induced ethane pyrolysis yields 25% higher ethylene production rates compared to the conventional steam cracking method. Laser induced pyrolysis is initiated by the generation of radicals upon heating of the ethane, then, followed by ethane/ethylene autocatalytic reaction in which ethane is converted into ethylene and other light hydrocarbons. This procedure is governed by micro-mixing of reactants and the feedstock residence time in reactor. Under mild turbulent conditions, the turbulence enhances the micro-mixing process and allows a high yield of ethylene. On the other hand, the high flow rate only allows a short residence time in the reactor which causes incomplete pyrolysis. This work attempts to investigate the interaction between turbulence and ethane pyrolysis process using large eddy simulation method. The modelling results could be applied to optimize the reactor design and operating conditions. 1 Skolkovo Foundation through the UNIHEAT Project Monday, November 24, 2014 10:30AM - 12:27PM Session H35 Detonation and Explosions — 2001A - Elaine Oran, University of Maryland 10:30AM H35.00001 Theory of weakly nonlinear self sustained detonations , LUIZ FARIA, ASLAN KASIMOV, KAUST, RODOLFO ROSALES, MIT — We derive a new weakly non-linear asymptotic model of detonation waves capable of capturing the rich dynamics observed in solutions of the reactive Euler equations, both in one and multiple space dimensions. We then investigate the travelling wave solutions of the asymptotic model, together with their linear stability. Finally, we study the non-linear dynamics through numerical simulations, and present a quantitative comparison between the asymptotic equations and the full system they are expected to approximate. 10:43AM H35.00002 Acoustic timescale characterization of asymmetric hot spot detonation initiation , JONATHAN D. REGELE, MICHAEL D. KURTZ, Iowa State University — Hot spots and temperature gradients are often used to model detonation initiation processes. Traditionally the focus of the analysis is on the critical gradient conditions necessary to facilitate detonation formation. However, hot spots usually have a local maximum of some finite size at the center. In previous work, acoustic timescale analysis has been used to characterize the behavior of a one-dimensional hot spot where a linear temperature gradient is joined with a constant temperature plateau. In the present work, the effects of multiple dimensions are analyzed by considering hot spots whose plateau and gradient regions are modeled as circles and ellipses. Even with clear differences in behavior between one and two dimensions, the a priori prescribed hot spot acoustic timescale ratio is shown to characterize the 2-D gasdynamic response. In asymmetric hot spots, it is shown that the behavior along the semi-minor axis is similar to the one-dimensional model over a limited period of time. 10:56AM H35.00003 Linear Theory for the Interaction of Small-Scale Turbulence with Overdriven Detonations , CESAR HUETE RUIZ DE LIRA, ANTONIO L. SANCHEZ, FORMAN A. WILLIAMS, Univ of California - San Diego — To complement our previous analysis of interactions of large-scale turbulence with strong detonations, the corresponding theory of interactions of small-scale turbulence is presented here. Focusing most directly on the results of greatest interest, the ultimate long-time effects of high-frequency vortical and entropic disturbances on the burnt-gas flow, a normal-mode analysis is selected. The interaction of the planar detonation with a monochromatic pattern of perturbations is addressed first, and then a Fourier superposition for two-dimensional and three-dimensional isotropic turbulent fields is employed to provide integral formulae for the amplification of the kinetic energy, enstrophy, and density fluctuations. Effects of the propagation Mach number and of the chemical heat release and the chemical reaction rate are identified, as well as the similarities and differences from the previous result for the thin-detonation (fast-reaction) limit. 11:09AM H35.00004 Nonlinear dynamics of hydrogen-air detonations with detailed kinetics and diffusion , JOSEPH POWERS, CHRISTOPHER ROMICK, University of Notre Dame, TARIQ ASLAM, Los Alamos National Laboratory — We consider the calculation of unsteady detonation in a mixture of calorically imperfect ideal gases with detailed kinetics. The use of detailed kinetics introduces multiple reaction length scales, and their interaction gives rise to complex dynamics. These are predicted using a wavelet-based adaptive mesh refinement technique and includes multi-component species, momentum, and energy diffusion, as well as DuFour and Soret effects. In the one-dimensional limit, we predict a transition from stability to unstable limit cycles as a driving piston velocity is lowered. At low overdrive, energy is partitioned into a variety of high frequency oscillatory modes. For weak low frequency instabilities, the dynamics are largely explained by a competition between advection and reaction time scales, with diffusion serving to perturb the dynamics. For higher frequency instabilities, the influence of diffusion is larger. We present new extensions to two-dimensional dynamics. 11:22AM H35.00005 Numerical study of detonation ignition via converging shock waves , CHRISTIAN SCHMITZ, MILTIADIS PAPALEXANDRIS, Univ Catholique de Louvain — In this talk we present results of a numerical study on gaseous detonation ignition via converging shock waves in reflectors. In our study, chemical kinetics is modelled by a three-step chain-branching mechanism possessing an explosion limit. According to our simulations, as soon as the shock reflects from the apex of the domain, the temperature and pressure behind it can exceed the explosion limit, thus initiating rapid burning. However, the subsequent expansion of the reflected shock might eventually inhibit detonation ignition. To explore further the interplay between these mechanisms, we discuss results of parametric studies with respect to confinement geometries and present estimates for the minimum shock strength required for detonation ignition. 11:35AM H35.00006 Mechanisms of strong pressure wave generations during knocking combustion: compressible reactive flow simulations with detailed chemical kinetics , HIROSHI TERASHIMA, The University of Tokyo, MITSUO KOSHI, Yokohama National University — Knocking is a very severe pressure oscillation caused by interactions between flame propagation and end-gas autoignition in spark-assisted engines. In this study, knocking combustion modeled in one-dimensional space is simulated using a highly efficient compressible flow solver with detailed chemical kinetics for clarifying the process of knocking occurrence. Especially, mechanisms of strong pressure wave generation are addressed. A robust and fast explicit integration method is used to efficiently handle stiff chemistry, and species bundling for effectively estimating the diffusion coefficients. The detailed mechanisms such as n-butane of 113 species and n-heptane of 373 species are directly applied. Results demonstrate that the negative temperature coefficient (NTC) region of n-heptane significantly influence the knocking timing and intensity. In the NTC region, stronger pressure wave is generated due to rapid heat release of a very small portion in the end-gas, which is attributed to low temperature oxidation and inhomogeneous temperature distributions in the end-gas. The knocking intensity is thus amplified in the NTC region, taking a maximum value. In the case of n-butane with no NTC region, relatively weak knocking intensity is observed in all conditions with no clear peak. 11:48AM H35.00007 Chemical Kinetics in the expansion flow field of a rotating detonationwave engine , KAZHIKATHRA KAILASANATH, DOUGLAS SCHWER, U.S. Naval Research Laboratory — Rotating detonation-wave engines (RDE) are a form of continuous detonation-wave engines. They potentially provide further gains in performance than an intermittent or pulsed detonation–wave engine (PDE). The overall flow field in an idealized RDE, primarily consisting of two concentric cylinders, has been discussed in previous meetings. Because of the high pressures involved and the lack of adequate reaction mechanisms for this regime, previous simulations have typically used simplified chemistry models. However, understanding the exhaust species concentrations in propulsion devices is important for both performance considerations as well as estimating pollutant emissions. A key step towards addressing this need will be discussed in this talk. In this approach, an induction parameter model is used for simulating the detonation but a more detailed finite-chemistry model is used in the expansion flow region, where the pressures are lower and the uncertainties in the chemistry model are greatly reduced. Results show that overall radical concentrations in the exhaust flow are substantially lower than from earlier predictions with simplified models. The performance of a baseline hydrogen/air RDE increased from 4940 s to 5000 s with the expansion flow chemistry, due to recombination of radicals and more production of H2O, resulting in additional heat release. 12:01PM H35.00008 A Common Initiation Criterion for CL-20 EBW Detonators , COLE VALANCIUS, CHRISTOPHER GARASI, PATRICK O’MALLEY, Sandia National Laboratories — In an effort to better understand the initiation mechanisms of hexanitrohexaazaisowurtzitane (CL-20) based Exploding Bridgewire (EBW) detonators, a series of studies were performed comparing electrical input parameters and detonator performance. Traditional methods of analysis, such as burst current and action, do not allow performance to be compared across multiple firesets. A new metric, electrical burst energy density (Eρ ), allows an explosive train to be characterized across all possible electrical configurations (different firesets, different sized gold bridges, different cables and cable lengths); by testing one electrical configuration, performance across all others is understood. This discovery has implications for design and surveillance, and for the first time, presents a link between modeling of electrical circuits (such as in ALEGRA) and explosive performance. 12:14PM H35.00009 Plasma Sensor Measurements in Pulse Detonation Engines , ERIC MATLIS, CURTIS MARSHALL, THOMAS CORKE, University of Notre Dame, SIVARAM GOGINENI, Spectral Energies, LLC — Measurements have been conducted in a pulse detonation and rotating detonation engine using a newly developed plasma sensor. This sensor relies on the novel approach of using an ac-driven, weakly-ionized electrical discharge as the main sensing element. The advantages of this approach include a native high bandwidth of 1 MHz without the need for electronic frequency compensation, a dual-mode capability that provides sensitivity to multiple flow parameters, including velocity, pressure, temperature, and gas-species, and a simple and robust design making it very cost effective. The sensor design is installation-compatible with conventional sensors commonly used in gas-turbine research such as the Kulite dynamic pressure sensor while providing much better longevity. Developmental work was performed in high temperature facilities that are relevant to the propulsion and high-speed research community. This includes tests performed in a J85 augmentor at full afterburner and pulse-detonation engines at the University of Cincinnati (UC) at temperatures approaching 2760◦ C (5000◦ F). Monday, November 24, 2014 10:30AM - 12:40PM Session H36 Jets II — Alcove A - Leonardo Chamorro, University of Illinois at Urbana-Champaign 10:30AM H36.00001 Bouncing and Merging of Liquid Jets , ABHISHEK SAHA, MINGLEI LI, CHUNG K. LAW, Princeton University — Collision of two fluid jets is a technique that is utilized in many industrial applications, such as in rocket engines, to achieve controlled mixing, atomization and sometimes liquid phase reactions. Thus, the dynamics of colliding jets have direct impact on the performance, efficiency and reliability of such applications. In analogy with the dynamics of droplet-droplet collision, in this work we have experimentally demonstrated, for n-alkane hydrocarbons as well as water, that with increasing impact inertia obliquely colliding jets also exhibit the same nonmonotonic responses of merging, bouncing, merging again, and merging followed by disintegration; and that the continuous entrainment of the boundary layer air over the jet surface into the colliding interfacial region leads to two distinguishing features of jet collision, namely: there exists a maximum impact angle beyond which merging is always possible, and that merging is inhibited and then promoted with increasing pressure. These distinct response regimes were mapped and explained on the bases of impact inertia, deformation of the jet surface, viscous loss within the jet interior, and the thickness and pressure build-up within the interfacial region in order to activate the attractive surface van der Waals force to effect merging. 10:43AM H36.00002 The Dynamics of Coherent Structures in Under-expanded Supersonic Impinging Jets1 , PAUL STEGEMAN, Monash University, ANDREW OOI, The University of Melbourne, JULIO SORIA, Monash University — This study looks at the spatio-temporal dynamics of the coherent structures found in under-expanded supersonic impinging jets from a circular nozzle at a pressure ratio of 3.4 and standoff distances of {1d, 2d, 3d}. In these jets the development of coherent structures within the shear layer and their interaction with a standoff-shock are the principle components of a fundamental non-linear acoustic feedback mechanism. Temporally resolved and phase averaged data for each case was generated from a three dimensional hybrid large-eddy simulation on a non-uniform structured cylindrical grid with computational domains consisting of approximately 8.6, 10.6 and 12.8 million nodes for each of the cases respectively. From these datasets we investigate the development of the energy distribution and topology of the coherent structures as a function of their distance traveled along the shear-layer from their initial growth period to their interaction with the standoff-shock wave. Furthering this analysis the various terms in the kinetic and internal energy transport equations are examined to gain insight into the physical mechanisms for the transfer of energy to/from the coherent structures. 1 The support of the Australian Research Council via a Discovery grant and NCI is gratefully acknowledged. 10:56AM H36.00003 Flow field features of impinging jets with fractal grids , GIOACCHINO CAFIERO, Università degli Studi di Napoli “Federico II”, STEFANO DISCETTI, Aerospace Engineering Group, Universidad Carlos III de Madrid, TOMMASO ASTARITA, Università degli Studi di Napoli “Federico II” — An experimental investigation of the flow field features of impinging jets equipped with a fractal turbulence promoter at short nozzle to plate distances is carried out by means of 2D-3C Particle Image Velocimetry (Stereo-PIV). The test Reynolds number based on the nozzle diameter is set to 10,000. Both the instantaneous and the time averaged features of such flow field are discussed. The comparison with the well-known features of a circular jet without any turbulator (JWT) reveals how some of the peculiar features of this flow field are suppressed by the presence of the grid. The typical ring-vortex that arises as a consequence of the shear layer instability is perturbed and suppressed by the high frequency disturbance introduced by the grid. As a consequence, there is no vortex separation in correspondence of the impinging plate, then leading to the absence of the characteristic “double peak” in the Nusselt number profile for JWT. 11:09AM H36.00004 Experimental studies on circular and AR4 elliptic vortex-ring impingement upon inclined surfaces , SHENGXIAN SHI, Shanghai Jiao Tong Univ, TZE HOW NEW, Nanyang Technological University, JIAN CHEN, Shanghai Jiao Tong Univ — PLIF flow visualisation and TR-PIV measurements were performed on the impingement of circular and AR4 elliptic vortex-rings upon flat surface with different inclination angles at Re=4000. This is aimed to investigate the effects of nozzle geometry, surface inclination angle and exit-surface separation distance on the vortex-ring impingement behaviour. Separation distance between nozzle exit and flat surface were adjusted for the elliptic vortex-ring so as to examine the flow structures for impingement prior, at and posterior the axis-switching point. Current results on circular vortex-ring show that at low inclination angle, vortex-ring underwent severe stretching during the impingement and vortex-ring core closer to the flat surface was observed to induce secondary vortex-ring and pair with it before its pinch-off. Meanwhile, vortex-ring core further away from the flat surface produced secondary and tertiary vortex-rings before transit into turbulence. At high inclination angles, vortex-ring core closer to the flat surface was quickly entrained by the primary vortex-ring after the impingement. Experiments on elliptic vortex-ring are undergoing at the moment, more findings will be presented in the conference. 11:22AM H36.00005 Experimental investigation of a confined developing axisymmetric wall jet , TIANQI GUO, MATTHEW RAU, PAVLOS VLACHOS, SURESH GARIMELLA, Purdue Univ — The present work reports an experimental study of an axisymmetric, confined wall jet using planar particle image velocimetry (PIV) and the dielectric liquid HFE-7100. The wall jet is formed downstream from a confined and submerged impinging round jet, 3.75 mm in diameter. Both the developing region and self-similar region of the wall jet are investigated. The experiments are conducted for Reynolds numbers (Re = Ud/υ) ranging from 1500 to 38000 and at a nozzle-to-plate spacing of four jet diameters. Imagepreprocessing techniques are used to eliminate background noise and an ensemble correlation scheme is applied to improve the resolution of the measurement of the mean velocity field near the wall. A maximum measurement resolution of 36 µm is achieved. Profiles of the mean velocity, turbulent intensities, decay rate, spread rate and wall shear stress are used to characterize the influence of confinement on the wall jet development and inner layer scaling. 11:35AM H36.00006 Jet Velocity Profile Effects on Spray Characteristics of Impinging Jets at High Reynolds and Weber Numbers1 , NEIL S. RODRIGUES, VARUN KULKARNI, PAUL E. SOJKA, Purdue University — While like-on-like doublet impinging jet atomization has been extensively studied in the literature, there is poor agreement between experimentally observed spray characteristics and theoretical predictions (Ryan et al. 1995, Anderson et al. 2006). Recent works (Bremond and Villermaux 2006, Choo and Kang 2007) have introduced a non-uniform jet velocity profile, which lead to a deviation from the standard assumptions for the sheet velocity and the sheet thickness parameter. These works have assumed a parabolic profile to serve as another limit to the traditional uniform jet velocity profile assumption. Incorporating a non-uniform jet velocity profile results in the sheet velocity and the sheet thickness parameter depending on the sheet azimuthal angle. In this work, the 1/7th power-law turbulent velocity profile is assumed to provide a closer match to the flow behavior of jets at high Reynolds and Weber numbers, which correspond to the impact wave regime. Predictions for the maximum wavelength, sheet breakup length, ligament diameter, and drop diameter are compared with experimental observations. The results demonstrate better agreement between experimentally measured values and predictions, compared to previous models. 1 U.S. Army Research Office under the Multi-University Research Initiative grant number W911NF-08-1-0171. 11:48AM H36.00007 Mixing Characteristics for Flush and Elevated Jets in Crossflow1 , LEVON GEVORKYAN, TAKESHI SHOJI, WEN YU PENG, DANIEL GETSINGER, OWEN SMITH, ANN KARAGOZIAN, UCLA — The present experiments explore the mixing and structural characteristics of equidensity and variable density gas-phase transverse jets using acetone PLIF as well as stereo PIV. Flush and elevated nozzles as well as a flush pipe geometry are explored in these studies, for a range of jet-to-crossflow momentum flux ratios J and density ratios S, spanning previously-determined conditions creating upstream shear layers which are either convectively unstable or absolutely unstable. The present studies quantify a range of mixing and flow metrics for the jet in crossflow, including conditional unmixedness, conditional probability density function, and scalar dissipation rates associated with both the jet cross-section and the centerplane longitudinal imaging. Correlations between mixing parameters and the structural symmetry/asymmetry in the JICF are observed, as are connections with the state of the shear layer and vorticity evolution. 1 Supported by NSF grant CBET-1437014 & AFOSR grant FA9550-11-1-0128 (A001768901) 12:01PM H36.00008 Pressure Modulated Sonic Jet in Supersonic Crossflow , TOBIAS ROSSMANN, Lafayette College — Sonic transverse jets in supersonic crossflow are modulated using high-amplitude variations in jet stagnation pressure to enhance jet penetration and mixing. An injection/modulation apparatus combining a powered resonance tube and acoustic resonator is used to create low momentum ratio jets (J = 1, 2) in a supersonic cross-stream (M = 3.5). The injector has the capability to modulate the jet supply pressure at sufficiently high frequency ( > 15 kHz) and amplitude (up to 190 dB) to access relevant Strouhal numbers (St = 0 − 0.3) and amplitudes (up to 10% of the jet stagnation pressure) related to mixing enhancement. Planar laser Mie scattering in both side and end views allows for instantaneous imaging of the jet fluid to quantify jet trajectory, spread, and mixing behavior. For modulated J = 2 transverse jets, the recirculation zone directly downstream of the injection location is eliminated and significantly faster centerline signal decay rates are seen. For the J = 1 modulated jets, substantial increases in centerline penetration, jet spread, and centerline signal decay rate are shown. Additionally, PDF analysis of the instantaneous jet fluid signal values is performed to compare local mixing efficiencies between the modulated and un-modulated cases. 12:14PM H36.00009 Rapid 3D Printing of Multifunctional Calcium Alginate Gel Pipes using Coaxial Jet Extruder1 , KONRAD RYKACZEWSKI, VIRAJ DAMLE, Arizona State University — Calcium alginate (CA) forms when solution containing sodium alginate (SA) comes in contact with a CaCl2 solution. The resulting gel is biocompatible as well as edible and is used in production of bio-scaffolds, artificial plant seeds, and edible substances. In the latter application, referred to in the culinary world as “spherification,” flavored liquids are mixed with the SA and dripped into CaCl2 solution to form gel encapsulated flavored “marbles.” Previously, crude 3D printing of CA structures has been achieved by stacking of such flavored liquid filled marbles [1]. In turn, solid CA rods have been fabricated by properly mixing flow of the two solutions using a microfluidic device [2]. Here we show that by using two circular cross-section coaxial nozzles to produce coaxial jets of the SA and CaCl2 solutions, liquid filled CA micro-to-mili scale gel pipes can be produced at speeds around ∼ 150 mm/s. Such extrusion rate is compatible with most commercially available 3D printers, facilitating adoption of the CA pipe coaxial jet extruder. Here, the impact of inner and outer liquid properties and flow speeds on the gel pipe extrusion process is discussed. [1] www.dovetailed.co [2] Beyer et al. Solid-State Sens., Act.nMicrosys. 2013, 1206-1209. 1 KR acknowledges startup funding from ASU 12:27PM H36.00010 Simulation of direct contact condensation of steam jets based on interfacial instability theories1 , DAVID HEINZE, Karlsruhe Institute of Technology / Kernkraftwerk Gundremmingen GmbH, THOMAS SCHULEN- BERG, ANDREAS CLASS, Karlsruhe Institute of Technology, LARS BEHNKE, Kernkraftwerk Gundremmingen GmbH — A simulation model2 for the direct contact condensation of steam in subcooled water is presented that allows to determine major parameters of the process such as the jet penetration length. Entrainment of water by the steam jet is modeled based on the Kelvin-Helmholtz and Rayleigh-Taylor instability theories. Primary atomization due to acceleration of interfacial waves and secondary atomization due to aerodynamic forces account for the initial size of entrained droplets. The resulting steam-water two-phase flow is simulated based on a one-dimensional two-fluid model. An interfacial area transport equation is used to track changes of the interfacial area density due to droplet entrainment and steam condensation. Interfacial heat and mass transfer rates during condensation are calculated using the two-resistance model. The resulting two-phase flow equations constitute a system of ordinary differential equations which is discretized by means of an explicit Runge-Kutta method. The simulation results are in good agreement with published experimental data over a wide range of pool temperatures and mass flow rates. 1 funded by RWE Power AG et al. A Physically Based, One-Dimensional Simulation Model of Direct Contact Condensation of Steam Jets, submitted to ASME Journal of Nuclear Engineering 2 Heinze Monday, November 24, 2014 2:00PM - 2:35PM — Session J8 Invited Session: Life on a Surface in a Low-Reynolds Number Flow 3001/3003 - John Hinch, University of Cambridge 2:00PM J8.00001 Life on a Surface in a Low-Reynolds-Number Flow , HOWARD A. STONE, Department of Mechanical and Aerospace Engineering, Princeton University — There are many studies of the dynamics of swimming microorganisms. There are far fewer studies of how bacteria that attach to surfaces respond to motions of the surrounding fluid. Such motions can influence bacterial orientation, which can, in turn, impact bacterial colonization and motility on surfaces. Moreover, as biofilms develop, stresses from the flow on the surface can cause three-dimensional rearrangements of the soft biofilm in the form of “streamers” that can bridge the sides of a porous system; such suspended filaments trigger rapid clogging. Some of the underlying fluid dynamics will be discussed in the spirit of how the flow couples to the spatial and temporal evolution of these bacterial systems. Joint work with N. Autrusson, B. Bassler, K. Drescher, Z. Gitai, L. Guglielmini, F. Ingremeau, M.K. Kim, S. Lecuyer, O. Pak, A. Persat, R. Rusconi, Y. Shen, A. Siryaporn, N. Wingreen Monday, November 24, 2014 2:00PM - 2:35PM — Session J14 Invited Session: Low Mach Number Modeling of Stratified Flows 3009/3011 - Philip Marcus, University of California, Berkeley 2:00PM J14.00001 Low Mach Number Modeling of Stratified Flows , ANN ALMGREN, Lawrence Berkeley National Laboratory — Low Mach number equation sets approximate the equations of motion of a compressible fluid by filtering out the sound waves, which allows the system to evolve on the advective rather than the acoustic time scale. Depending on the degree of approximation, low Mach number models retain some subset of possible compressible effects. In this talk I present an overview of low Mach number methods for modeling stratified flows arising in astrophysics and atmospheric science, and discuss the algorithmic components necessary to model such flows. Monday, November 24, 2014 2:40PM - 3:15PM — Session K8 Invited Session: Cavitation in Ultrasound and Shockwave Therapy 3001/3003 - Jonathan Freund, University of Illiniois 2:40PM K8.00001 Cavitation in ultrasound and shockwave therapy1 , TIM COLONIUS, California Institute of Technology — Acoustic waves, especially high-intensity ultrasound and shock waves, are used for medical imaging and intra- and extra-corporeal manipulation of cells, tissue, and urinary calculi. Waves are currently used to treat kidney stone disease, plantar fasciitis, and bone nonunion, and they are being investigated as a technique to ablate cancer tumors and mediate drug delivery. In many applications, acoustic waves induce the expansion and collapse of preexisting or newly cavitating bubbles whose presence can either mediate the generation of localized stresses or lead to collateral damage, depending on how effectively they can be controlled. We describe efforts aimed at simulating the collapse of bubbles, both individually and in clusters, with the aim to characterize the induced mechanical stresses and strains. To simulate collapse of one or a few bubbles, compressible Euler and Navier-Stokes simulations of multi-component materials are performed with WENO-based shock and interface capturing schemes. Repetitive insonification generates numerous bubbles that are difficult to resolve numerically. Such clouds are also important in traditional engineering applications such as caveating hydrofoils. Models that incorporate the dynamics of an unresolved dispersed phase consisting of the bubble cloud are also developed. The results of several model problems including bubble collapse near rigid surfaces, bubble collapse near compliant surfaces and in small capillaries are analyzed. The results are processed to determine the potential for micron-sized preexisting gas bubbles to damage capillaries. The translation of the fundamental fluid dynamics into improvements in the design and clinical application of shockwave lithotripters will be discussed. 1 NIH Grant PO1-DK043881 Monday, November 24, 2014 2:40PM - 3:15PM Session K14 Invited Session: In Pursuit of Internal Waves — 3009/3011 - Jeffrey Koseff, Stanford University 2:40PM K14.00001 In Pursuit of Internal Waves1 , THOMAS PEACOCK, Massachusetts Institute of Technology — Orders of magnitude larger than surface waves, and so powerful that their generation impacts the lunar orbit, internal waves, propagating disturbances of a densitystratified fluid, are ubiquitous throughout the ocean and atmosphere. Following the discovery of the phenomenon of “dead water” by early Arctic explorers and the classic laboratory visualizations of the curious St. Andrew’s Cross internal wave pattern, there has been a resurgence of interest in internal waves, inspired by their pivotal roles in local environmental and global climate processes, and their profound impact on ocean and aerospace engineering. We detail our widespread pursuit of internal waves through theoretical modeling, laboratory experiments and field studies, from the Pacific Ocean one thousand miles north and south of Hawaii, to the South China Sea, and on to the Arctic Ocean. We also describe our recent expedition to surf the most striking internal wave phenomenon of them all: the Morning Glory cloud in remote Northwest Australia. 1 This work was supported by the National Science Foundation through a CAREER grant OCE-064559 and through grants OCE-1129757 and OCE1357434, and by the Office of Naval Research through grants N00014-09-1-0282, N00014-08-1-0390 and N00014-05-1-0575. Monday, November 24, 2014 3:35PM - 5:58PM Session L1 General Fluid Dynamics III — 3000 - Steven Ceccio, University of Michigan 3:35PM L1.00001 Slow gravity-driven migration and interaction of a bubble and a solid particle near a free surface , FRANCK PIGEONNEAU, Surface du Verre et Interfaces, UMR 125 CNRS/Saint-Gobain, MARINE GUÉMAS, ANTOINE SELLIER, Laboratoire d’hydrodynamique de l’école polytechnique, UMR 7646, SURFACE DU VERRE ET INTERFACES, UMR 125 CNRS/SAINT-GOBAIN TEAM, LABORATOIRE D’HYDRODYNAMIQUE DE L’ÉCOLE POLYTECHNIQUE, UMR 7646 COLLABORATION — The interaction between a bubble and a free surface occurs in many industrial processes. This is the case for the glass melting process involving a high viscous fluid. Moreover, the bubble migration towards the free surface interacts with unmelted solid particles (sand grain). With this application in mind, the axi-symmetric gravity-driven migration of interacting bubble and solid particles near a free surface is examined. The solid particle locations and the bubble and free surface shapes are numerically tracked in time by solving the steady Stokes equations written under the boundary integral formulation. The theoretical material and the relevant numerical implementation, valid for bubbles and a free surface with either equal or unequal uniform surface tensions, are briefly described. Numerical results for a nearlyneutrally buoyant solid spherical particles interacting with a bubble and a free surface are investigated. We show that the solid particles decrease the bubble drainage dynamics. The effect of bubble size is also studied by changing the Bond number, ratio of buoyancy force to surface tension force. 3:48PM L1.00002 Dynamics of viscous jet under electric field , TIANTIAN KONG, ZHOU LIU, LIQIU WANG, HO CHEUNG SHUM, The University of Hong Kong — In this work, we study the folding and unfolding of viscous jets under an electric field. We show that the geometry of the jet responds sensitively to electric field and changes the jet dynamics accordingly. We demonstrate that a stable viscous straight jet can be induced to fold, and the folding morphology can be precisely tuned by varying the electric field. Under a controlled electric field configuration, a folded viscous jet can also be induced to unfold. We further confirm that viscous folding occurs only when a jet is sufficiently compressed and has an aspect ratio above a critical value, often known as the critical slenderness. The precise control of folding morphology is potentially useful for fabrication of nano-scaled features. Moreover, the underlying mechanisms have important implications for applications such as 3-dimensional printing and polymer processing, in which dispensing and manipulating of flowing viscous jets are of great importance. 4:01PM L1.00003 Experimental Investigation of Spatially-periodic Scalar Patterns in an Inline Mixer1 , OZGE BASKAN, MICHEL SPEETJENS, HERMAN CLERCX, Eindhoven University of Technology — Spatially persisting patterns with exponentially decaying intensities form during the downstream evolution of passive scalars in three-dimensional (3D) spatially periodic flows due to the coupled effect of the chaotic nature of the flow and the diffusivity of the material. This has been investigated in many computational and theoretical studies on 2D time-periodic and 3D spatially-periodic flow fields, however, experimental studies, to date, have mainly focused on flow visualization with streaks of dye rather than fully 3D scalar field measurements. Our study employs the-state-of-the-art experimental methods to analyze the evolution of 3D scalar fields and the correlation between the coherent flow/scalar field structures in a representative inline mixer, called Quatro static mixer. The experimental setup consists of an optically accessible test section with transparent internal elements, accommodating a pressure-driven pipe flow. The 3D scalar fields are measured by 3D Laser-Induced Fluorescence. The preliminary results are consistent with the literature and we discuss the comparative analysis between our experimental observations and the numerical simulations from the previous studies. 1 The authors gratefully acknowledge the support by Dutch Technology Foundation STW. 4:14PM L1.00004 Back to a classical problem: A fresh look at the asymptotics of the moving contact line , DAVID SIBLEY, ANDREAS NOLD, SERAFIM KALLIADASIS, Imperial College London — For contact line motion where the full Stokes flow equations hold, full matched asymptotic solutions using slip models have been obtained for droplet spreading and more general geometries, e.g. in the well-known results of Hocking and Cox. These solutions to the singular perturbation problem in the slip length, however, all involve matching through an intermediate region that is taken to be separate from the outer and inner regions and on the basis that the two do not match directly. Here, we show that not only is direct matching possible but the intermediate region is in fact an overlap region representing extensions of both the outer and the inner regions. In particular, we investigate in detail how a previously seen result of the matching of the cubes of the free surface slope is justified in the lubrication setting. We also extend this two-region direct matching to the more general Stokes flow case, offering a new perspective on the asymptotics of the moving contact line problem. 4:27PM L1.00005 Time-varying creeping flow in an elastic shell enveloping a slender rigid center-body1 , SHAI ELBAZ, AMIR GAT, Technion - Israel Institute of Technology — Flows in contact with elastic structures apply stress at the fluid-solid interface and thus create deformation fields in the solid. We study the time-varying interaction between elastic structures, subject to external forces, and an internal viscous liquid. We neglect inertia in the liquid and solid and focus on axi-symmetric annular flow enclosed by a thin-walled slender elastic shell and internally bounded by a variable cross-section rigid center-body. We employ elastic shell theory and the lubrication approximation to show that the problem is governed by the forced porous medium equation with regard to fluid pressure. We present several solutions of the flow-field and solid-deformation for various time-varying inlet pressure and external forces. The presented interaction between viscosity and elasticity may be applied to fields such as soft-robotics and micro-swimmers. 1 Israel Science Foundation 818/13 4:40PM L1.00006 Drawing of microstuctured optical fibres with pressurisation of the internal channels , MICHAEL CHEN, YVONNE STOKES, University of Adelaide, PETER BUCHAK, DARREN CROWDY, Imperial College London, HEIKE EBENDORFF-HEIDEPRIEM, University of Adelaide — Microstructured optical fibres are distinguished from solid optical fibres by the large number of internal air channels running along their length. These fibres are manufactured by heating and stretching a preform, which has some cross-sectional pattern of holes. In stretching the preform with a diameter of 1-3cm to a fibre with a diameter of the order of 100 micrometers, the cross-sectional hole pattern changes in scale but is also deformed due to surface tension. A practical way of countering this deformation is to introduce pressurisation in the internal channels. This pressure acts against surface tension and potentially provides an extra degree of control over the shape of the internal channel geometry. We generalise an existing model of fibre drawing to include channel pressurisation and present examples of pressurised fibre drawing for several cross-sectional geometries of practical importance. 4:53PM L1.00007 Response time of a gravity current to a change in flux , HERBERT HUPPERT, THOMASINA BALL, JEROME NEUFELD, University of Cambridge, HERBERT HUPPERT COLLABORATION — Many different fluid flows are in response to an input flux, often held constant. Gravity currents as a result of fluid input of one density into fluid of a different density are a prime example. Often the current settles down to a well-described law of propagation, quantitatively dependent on the input flux Q and other physical parameters. The presentation will discuss how long the fluid response takes to adjust to a change in flux, to a larger or smaller flux, including zero. We will demonstrate that for many flows, at low Reynolds numbers, high Reynolds numbers, through porous media, in either axisymmetric or one-dimensional situations, the response time is proportional to the time of persistence of the initial flux (before it is altered), independent of the value of other physical parameters, such as viscosity, permeability or the initial flux. The presentation will include analytical, numerical and experimental confirmation of the result. Application of the result to the storage of carbon dioxide in geological layers deep within the Earth will be described. 5:06PM L1.00008 The inception of eddy formation in the flow around a circular cylinder , NIKOLAOS MALAMATARIS, George Mason Univ / ATEI Thessaloniki, ANASTASIA GARATIDOU, GEORGE SAKKOS, KONSTANTINE MATSIANIKAS, TEI W Macedonia — The inception of eddy formation in the flow around a circular cylinder is studied with a home made Galerkin finite element code. The results show that the eddy begins at Re=6.2 which is the lowest Reynolds number reported so far in the literature. The code is validated with well known results for this flow. In addition, it is calculated for the first time how the pressure distribution around the cylinder surface varies for Reynolds numbers lower than 6.2 where experimental results exist from 1936. Finally, a new formula is given for the variation of the drag coefficient with respect to the Reynolds numbers under creeping flow conditions which deviates significantly from the well known result that exists in the literature. 5:19PM L1.00009 Entrainment effects in the long residence times of solid spheres settling in stratified fluids1 , CLAUDA FALCON, ROBERTO CAMASSA, RICHARD MCLAUGHLIN, University of North Carolina, UNC JOINT FLUIDS LAB TEAM — This talk will present results of a study, by a combination of experimental, analytical and numerical tools, of the effects of sharp density variations in the dynamics of a settling sphere in viscous dominated regimes. In particular, long residence times rivaling the ones observed for porous spheres in similar configurations will be demonstrated and discussed. Asymptotic approaches and exact solutions for the sphere exterior problem of Stokes equations will be compared in a parametric study of relevance for experiments. 1 Supported by NSF and ONR. 5:32PM L1.00010 On Gyroviscous Fluids1 , PHILIP J. MORRISON, MANASVI LINGAM, Institute for Fusion Studies, The University of Texas at Austin — Fluid models involving gyroviscous effects, whereby momentum is transported while conserving energy, are of interest for plasma, astrophysical, and condensed matter systems. Such fluids can be viewed as possessing intrinsic angular momentum. We present a systematic method for constructing such models from an action principle formalism [1,2] that allows for an unambiguous means for introducing these effects, instead of ad-hoc phenomenological prescriptions. We also apply Noether’s theorem to obtain the appropriate conserved quantities for these models. [1] M. Lingam and P.J. Morrison, “The action principle for generalized fluid motion including gyroviscosity” (to be submitted). [2] P.J. Morrison, M. Lingam and R. Acevedo, Phys. Plasmas 21, 082102 (2014). 1 Supported by U.S. Department of Energy Contract No. DE-FG05-80ET-53088 5:45PM L1.00011 Experimental Characterization of Inter-channel Mixing Through a Narrow Gap , JACK BUCHANAN1 , Bechtel Marine Propulsion Corp., SIMO MAKIHARJU, University of Michigan, ALEXANDER MYCHKOVSKY, KEVIN HOGAN, KIRK LOWE, Bechtel Marine Propulsion Corp., STEVEN CECCIO, University of Michigan — Mixing trough narrow gaps that connect primary flow paths is an important flow process for many thermal-hydraulic applications, such as flow through nuclear reactor rod bundles or heat exchangers. The flow in a narrow gap can exhibit periodic flow structures due to travelling vortices. These flow structures in the gap, as well as any pressure gradient across the gap, have a significant effect on the rate of mixing between the primary flow paths. To investigate such flows in detail, and to develop validation quality data sets for comparison with CFD, we have conducted a canonical inter-channel mixing experiment between two channels, with a (127 mm)2 cross-section. The channels were connected by a gap 914.4 mm long in the stream-wise direction and 50.8mm wide in the cross-stream direction. The gap height could be varied from 0 to 50.8 mm. The flow speed in both channels could be independently varied to have Re = (40 to 100) x 103 . The integral mixing rates were determined by injecting fluorescent dye into one of the channels well upstream of the test section and by measuring the dye concentration at the channel inlets and outlets. Additionally, the flow fields in the gap and channels were measured with LDV and PIV. 1 Bettis Laboratory, West Mifflin, PA 15122 Monday, November 24, 2014 3:35PM - 5:58PM Session L2 Surface Tension Effects: Interfacial Phenomena — 3002 - Valeria Garbin, Imperial College London 3:35PM L2.00001 Modeling Interfacial Adsorption of Polymer-Grafted Nanoparticles , XIN YONG, Binghamton University — Numerous natural and industrial processes demand advances in our fundamental understanding of colloidal adsorption at liquid interfaces. Using dissipative particle dynamics (DPD), we model the interfacial adsorption of core-shell nanoparticles at the water-oil interface. The solid core of the nanoparticle encompasses beads arranged in an fcc lattice structure and its surface is uniformly grafted with polymer chains. The nanoparticles bind to the interface from either phase to minimize total surface energy. With a single nanoparticle, we demonstrate detailed kinetics of different stages in the adsorption process. Prominent effect of grafted polymer chains is characterized by varying molecular weight and polydispersity of the chains. We also preload nanoparticles straddling the interface to reveal the influence of nanoparticle surface density on further adsorption. Importantly, these studies show how surface-grafted polymer chains can alter the interfacial behavior of colloidal particles and provide guidelines for designing on-demand Pickering emulsion. 3:48PM L2.00002 Particle-laden microbubbles: forced oscillations, surface modes, jetting and particle ejection , VINCENT POULICHET, VALERIA GARBIN, Imperial College London — Self-assembly of microparticles at fluid-fluid interfaces is exploited for emulsion stabilization, tunable nanomaterials, and nanocomposites with complex morphologies. Disassembly of interfacial particle monolayers is equally important, for instance for green catalytic processes, but has been far less explored. We demonstrate controlled disassembly of monolayers of microparticles trapped at the interface of microbubbles. The bubbles are driven into oscillations by an applied ultrasound wave, triggering particle ejection. We visualize forced ejection events at the single particle level using high-speed video microscopy. Measurements of the local area density of particles and of the acceleration of the bubble interface reveal that the interplay of several mechanisms is responsible for particle expulsion: tangential stresses due to the fast compression of the bubble interface, interparticle interactions at the interface, and the body force acting on the particles due to the acceleration of the interface. Non-linear bubble dynamics can also be exploited to design complex particle expulsion scenarios, such as surface modes and jetting, with relevance to directed particle delivery in microreactors. 4:01PM L2.00003 Viscous Marangoni migration of a drop in a Poiseuille flow at low surface Peclet numbers , ON SHUN PAK, JIE FENG, HOWARD STONE, Princeton University — The motion of a spherical drop with a bulk-insoluble surfactant immersed in a background flow in the low surface Peclet number and low Reynolds number limits is investigated. We develop a reciprocal theorem that applies to any prescribed background flow, and provide a specific example of an unbounded Poiseuille flow. Analytical formulas for the migration velocity of the drop are obtained perturbatively in powers of the surface Peclet number. We show that the redistribution of surfactant due to the background flow acts to retard the motion of the drop, with the magnitude of this slip velocity independent of the drop’s position in the Poiseuille flow. Moreover, a surfactant-induced cross-streamline migration of the drop occurs towards the center of the Poiseuille flow, with its magnitude depending linearly with the distance of the drop from the center of the Poiseuille flow. 4:14PM L2.00004 Volume of fluid simulations of liquefied metal nanofilms with Marangoni effects1 , IVANA SERIC, KYLE MAHADY, SHAHRIAR AFKHAMI, LOU KONDIC, New Jersey Institute of Technology — We present a method for including temperature dependent surface tension in a volume of fluid based Navier Stokes solver. The tangential gradient of the surface tension is implemented using an extension of the classical continuum surface force model that has been previously used for constant surface tension simulations. We apply the developed method to consider metal films liquefied by a pulse laser and discuss the effects of the resulting Marangoni stresses on the film evolution. 1 This work is partially supported by grants NSF DMS-1320037 and CBET-1235710 4:27PM L2.00005 Film deposition on a partially wetting plate withdrawn from a liquid reservoir1 , PENG GAO, LEI LI, University of Science and Technology of China — A partially wetting plate withdrawn from a liquid reservoir causes the deposition of liquid films, which are characterized by trapezoidal or triangular shapes. Interesting issues include the critical condition of the film deposition, the film structures and the dependence on the plate speed of the contact-line inclination angle. In the first part of this work, we performed numerical simulations of the problem with a diffuse-interface method, and reproduced the coexistence of the thick and thin films observed in recent experiments (Phys. Rev. Lett, 2006, 96, 174504, and Phys. Rev. Lett, 2008, 100, 244502). We demonstrated that the apparent contact angle vanishes at the onset of wetting transition, consistent with the lubrication theory. The critical condition for the onset of thin films was also quantified. In the second part of this work, we presented a lubrication analysis of films with inclined contact lines. It is shown that the traditional model of constant normal speed of the contact line is only a leading-order approximation; the normal speed actually exhibits a weak decrease with the inclination angle. In addition, the inclination of the contact line results in a tangential flux of the liquid. Simple scaling relations are provided for both the normal velocity and the flux. 1 This work is support by NSFC (No. 11102203) and the Chinese Academy of Sciences (KJZD-EW-J01). 4:40PM L2.00006 Dynamics of a solid sphere bouncing on or penetrating through a liquid-air interface , SEONG JIN KIM, SUNGHWAN JUNG, Virginia Tech, SUNGYON LEE, Texas A&M University — In this study, we investigate the dynamics of a solid particle moving from liquid to air through a liquid-air interface. The experimental setup consists of an air-piston system that shoots a solid particle into water towards the free surface from below. Experimental results indicate that the particle either penetrates or bounces back depending on the particle size, impact speed, and surface tension. In particular, the particle needs to overcome the resistive interfacial forces in order to penetrate through the liquid-air interface. This transition from bouncing to penetration regimes is captured theoretically by conducting a simple force balance and is further compared with experiments. 4:53PM L2.00007 Marangoni-buoyancy convection in binary fluids under varying noncondensable concentrations1 , YAOFA LI, MINAMI YODA, Georgia Institute of Technology — Marangoni-buoyancy convection in binary fluids in the presence of phase change is a complex and poorly understood problem. Nevertheless, this flow is of interest in evaporative cooling because solutocapillary stresses could reduce film dryout. Convection was therefore studied in methanol-water (MeOH-H2 O) layers of depth h ≈ 1 − 3 mm confined in a sealed rectangular cell driven by horizontal temperature differences of ∼ 6◦ C applied over ∼ 5 cm. Particle-image velocimetry (PIV) was used to study how varying the fraction of noncondensables (i.e., air) ca from ∼ 7 mol% to ambient conditions in the vapor space affects soluto- and thermocapillary stresses in this flow. Although solutocapillary stresses can be used to drive the flow towards hot regions, solutocapillarity appears to have the greatest effect on the flow at small ca , because noncondensables suppress phase change and hence the gradient in the liquid-phase composition at the interface. Surprisingly, convection at ca ≈ 50% leads to a very weak flow and significant condensation in the central portion of the layer i.e., away from the heated and cooled walls). 1 Supported by ONR 5:06PM L2.00008 The Capillary Fluidics of Espresso1 , NATHAN OTT2 , School of Science and Technology, DREW WOLLMAN, Portland State University, JOHN GRAF, NASA Johnson Space Center, MARK WEISLOGEL, Portland State University — Espresso is enjoyed by tens of millions of people daily. The coffee is distinguished by a complex low density colloid of emulsified oils. Due to gravity, these oils rise to the surface forming a foam lid called the crema. In this work we present a variety of large length scale capillary fluidic effects for espresso in a gravity-free environment. Drop tower tests are performed to establish brief microgravity conditions under which spontaneous capillarity-driven behavior is observed. Because the variety of espresso drinks is extensive, specific property measurements are made to assess the effects of wetting and surface tension for ‘Italian’ espresso, caffe latte, and caffe Americano. To some, the texture and aromatics of the crema play a critical role in the overall espresso experience. We show how in the low-g environment this may not be possible. We also suggest alternate methods for enjoying espresso aboard spacecraft. 1 NASA 2 High NNX09AP66A, Glenn Research Center School 5:19PM L2.00009 Relaxation of an elastic filament on a viscous interface , JOEL MARTHELOT, S. GANGA PRASATH, NARAYANAN MENON, TCIS, TIFR, Hyderabad — What is the shape of an elastic filament floating on a fluid interface? We observe the timedependent relaxation of a bent filament reopening towards its straight, stress-free, configuration. We study a regime in which the dynamics are overdamped, but with the rod initially bent into a geometrically nonlinear regime. The dynamics of reopening are governed by a competition between the viscous drag of the liquid and the bending elastic force of the rod. We study the relaxation of shape as a function of the length, diameter and elasticity of the rod, and the viscosity of the fluid interface. The opening dynamics are governed by a time scale that is much smaller than a time constructed from these quantities, but scales in the same way. This simple system could provide an easy method to characterize interfacial properties of fluid interfaces. 5:32PM L2.00010 Revised Capillary Breakup Rheometer Method , LOUISE LU, WILLIAM SCHULTZ, MICHAEL SOLOMON, University of Michigan — Rather than integrate the one-dimensional equation of motion for a capillary breakup rheometer, we take the axial derivative of that equation. This avoids the determination of the axial force with all of its complications and correction factors. The resulting evolutionary equation that involves either two or four derivatives of the capillary radius as a function of the axial coordinate determines the ratio of elongational viscosity to surface tension coefficient. We examine several silicone and olive oils to show the accuracy of the method for Newtonian fluids. We will discuss our surface tension measurement techniques and briefly describe measurements of viscoelastic materials, including saliva. 5:45PM L2.00011 Partial coalescence of sessile drops with different liquids1 , RODICA BORCIA, MICHAEL BESTEHORN, Brandenburgische Technische Universität Cottbus-Senftenberg — We examine numerically the interaction between two deformable drops consisting of two perfectly miscible liquids sitting on a solid substrate under a given contact angle. Driven by solutal Marangoni forces, several distinct coalescence regimes are achieved after the droplets collision [1]. Phase diagrams for different control parameters are emphasized, which give predictions about drop behavior along the solid substrates, control of various interfacial effects, manipulations of tiny droplets in micro- and nano-fluidic devices without power supply, design of droplets or cleaning surfaces. [1] R. Borcia, M. Bestehorn, Langmuir 29 (2013) 4426; Fluid Dynamics Research 46 (2014) 041405. 1 This work was partially supported by Deutsche Forschungsgemeinschaft (DFG) under the project “Dynamics of interfaces between drops with miscible liquids.” Monday, November 24, 2014 3:35PM - 6:11PM — Session L3 Electrokinetics: Concentration Polarization, Nanoscale and Porous Media 3004 - Carlos Hidrovo, Northeastern University 3:35PM L3.00001 Anomalous ion concentration distribution inside ion concentration polarization (ICP) layer by electrodeless measurement , INHEE CHO, WONSEOK KIM, Seoul National University, Korea, HYOMIN LEE, Pohang University of Science and Technology, JUNSUK KIM, Seoul National University, Korea, GUN YONG SUNG, Hallym University, SUNG JAE KIM, Seoul National University, Korea — An ion concentration profile inside the cathodic side of cation-selective membrane with dc bias has been reported to be a flat with aid of numerical simulation. While rigorous experimental evidences with microelectrode array have supported the flat profile, undesirable effects such as electrode reactions have hindered an accurate measurement. In this work, a microchannel with micro-grooves inside an ion depletion zone (or ICP layer) is employed to capture a vortical electrokinetic flow in the groove. By measuring the speed of the flow, one can convert it into the local ionic concentration, since the local speed of electrokinetic flow is proportional to the local electric field which is inversely proportional to the local ionic concentration. As a result, we can indirectly measure the full ion concentration profile inside ICP layer without any undesirable disturbance and find that the profile is neither flat nor monotonic. Instead, there are peaks and, more importantly, the locations of the peaks strongly depend on the mobility of majority carrier (Li+, Na+ and K+). The samples inside the ICP layer are analyzed by mass spectrometry to confirm the dependency. 3:48PM L3.00002 Capillary Ion Concentration Polarization for Power-Free Salt Purification1 , SUNGMIN PARK, ECE, SNU, YEONSU JUNG, MAE, SNU, INHEE CHO, ECE, SNU, HO-YOUNG KIM, MAE, SNU, SUNG JAE KIM, ECE, SNU — In this presentation, we experimentally and theoretically demonstrated the capillary based ion concentration polarization for power-free salt purification system. Traditional ion concentration polarization phenomenon has been studied for a decade for both fundamental nanoscale fluid dynamics and novel engineering applications such as desalination, preconcentration and energy harvesting devices. While the conventional system utilizes an external power source, the system based on capillary ion concentration polarization is capable of perm-selective ion transportation only by capillarity so that the same ion depletion zone can be formed without any external power sources. An ion concentration profile near the nanostructure was tracked using fluorescent probes and analyzed by solving the modified Nernst-Planck equation. As a result, the concentration in the vicinity of the nanostructure was at least 10 times lower than that of bulk electrolyte and thus, the liquid absorbed into the nanostructure had the low concentration. This mechanism can be used for the power free salt purification system which would be significantly useful in underdeveloped and remote area. 1 This work was supported by Samsung Research Funding Center of Samsung Electronics under Project Number SRFC-MA1301-02. 4:01PM L3.00003 Numerical and analytical models of concentration polarization in a microchannel , CHRISTOFFER P. NIELSEN, HENRIK BRUUS, Department of Physics, Technical University of Denmark — We present a comprehensive analysis of salt transport in microchannels during concentration polarization. We have carried out full numerical simulations of the coupled Poisson–Nernst– Planck–Stokes problem governing the transport and rationalized the behaviour of the system. A surprising discovery is that bulk advection relies heavily on the surface currents, even when these surface currents do not contribute much to the overlimiting current themselves. The numerical simulations are supplemented by analytical results valid in the long channel limit as well as in the limit of negligible surface charge. Notably, by including the effects of diffusion and advection in the diffuse double layers we extend a recently published analytical model of overlimiting current due to surface conduction. 4:14PM L3.00004 Concentration-Polarization, Electro-Convection and Colloid Dynamics in Microchannel-Nanochannel Interface Devices , GILAD YOSSIFON, NETA LEIBOWITZ, YOAV GREEN, URI LIEL, JARROD SCHIFFBAUER, SINWOOK PARK, Technion - Israel Institute of Technology — Understanding concentration-polarization (CP) and electroconvection processes along with colloid dynamics in microchannel-nanochannel/membrane interface devices are of particular interest in the field of micro- and nano-fluidics. Our design consists of a nano-slot/permselective membrane bounded by two micro-chambers, wherein we introduce dispersed colloids. Here we report various curios phenomena occurring in these systems. Among them: dielectrophoretic trapping of colloids at the nanoslot entrance in conjunction with the formation of electro-convective instability induced vortices; accumulation of colloids due field-focusing gradient effects within the diffusion layers; depression of the slope in the Warburg branch of the electrochemical impedance spectrum with increasing dc bias voltage as a result of nanochannel net electro-osmotic flow; suppression of the diffusion layer length via AC electrokinetics and its effect on ion transport; anomalous resistance minimum and unique chronopotentiometric signatures due to non-ideal nanochannel permselectivity. All of these stand as examples that highlight the essential differences between fabricated straight nanoslot and permselective membrane systems. 4:27PM L3.00005 Concentration Polarization and Electroconvection in a Nanochannel Array System , YOAV GREEN, SINWOOK PARK, GILAD YOSSIFON, Technion - Israel Institute of Technology — The passage of an electric current through a permselective medium (membranes/nanochannels) under an applied electric field is characterized by the formation of ionic concentration gradients which result in regions of depleted and enriched ionic concentration at opposite ends of the medium, i.e. concentration polarization (CP). Here we study experimentally the effects of 3D geometric field focusing on CP in realistic three dimensional and three layers system (i.e. microchannel-permselective medium-microchannel device). Previous analytical solutions were derived under the simplifying assumptions of local-electroneutrality, ideal permselectivity and negligible convection. In particular, we studied the effect of the interchannel spacing of an array of such permselective regions/channels, on the resulting current-voltage curves and concentration profiles, wherein an increased interchannel spacing corresponds to an increased geometric heterogeneity of the permselective medium. Good qualitative agreement is obtained between these theoretical predictions and experimental data obtained for a nanoslot array system with varying interchannel spacing. These results highlight the importance of geometric field focusing, interchannel communication and electro-convection effects on the ion transport in heterogeneous permselective systems. 4:40PM L3.00006 Computational modeling of electrokinetic transport in random networks of micro-pores and nano-pores , SHIMA ALIZADEH1 , ALI MANI, Stanford University — A reduced order model has been developed to study the nonlinear electrokinetic behaviors emerging in the transport of ionic species through micro-scale and nano-scale porous media. In this approach a porous structure is modeled as a network of long and thin pores. By assuming transport equilibrium in the thin dimensions for each pore, a 1D transport equation is developed in the longitudinal direction covering a wide range of conditions including extreme limits of thick and thin electric double layers. This 1D model includes transport via diffusion, electromigration and wide range of advection mechanisms including pressure driven flow, electroosmosis, and diffusion osmosis. The area-averaged equations governing the axial transport from different pores are coupled at the pore intersections using the proper conservation laws. Moreover, an asymptotic treatment has been included in order to remove singularities in the limit of small concentration. The proposed method provides an efficient framework for insightful simulations of porous electrokinetic systems with applications in water desalination and energy storage. 1 PhD student in Mechanical Engineering, Stanford University. She received her Master’s degree in Mechanical Engineering from Stanford at 2013. Her research interests include CFD, high performance computing, and optimization. 4:53PM L3.00007 Electroosmotic access resistance of a nanopore1 , SANDIP GHOSAL, Northwestern University, JOHN D. SHERWOOD, Cambridge University, MAO MAO, Northwestern University — Electroosmotic flow through a nanopore that traverses a dielectric membrane with a fixed surface charge density is considered. In the limit where the surface charge is small and the applied electric field weak, the reciprocal theorem is used to derive an expression for the electroosmotic flux through the pore up to quadratures over the fluid volume. Thus, an “electroosmotic conductance” (the fluid flux per unit applied voltage) may be defined in analogy to the corresponding electrical conductance of a hole in an insulating membrane immersed in a uniform conductor. In the limit when the membrane is thick compared to the pore diameter, the usual result for the electroosmotic conductance through long cylindrical channels (which varies inversely as the membrane thickness) is recovered. The electroosmotic conductance is shown to approach a finite value for an infinitely thin membrane: this residual electroosmotic resistance (inverse of conductance) is analogous to the concept of “access resistance of a pore” in the corresponding electrical problem. The dependence of the electroosmotic conductance on pore radius, Debye length and membrane thickness is investigated. Reference: JFM (2014) 749, 167; Langmuir (in press) 1 Supported by the NIH under grant 4R01HG004842. SG acknowledges a visiting professorship at Cambridge University funded by the Leverhulme Trust, UK. JDS thanks DAMTP (Cambridge University) and Institut de Mecanique des Fluides de Toulouse for hospitality 5:06PM L3.00008 Electroosmotic flow through a cylindrical nanopore in a charged membrane of finite thickness1 , MAO MAO, SANDIP GHOSAL, Northwestern University, JOHN D. SHERWOOD, University of Cambridge — We present numerical solutions to the coupled Nernst-Planck-Poisson-Stokes equation for electroosmotic flow through a cylindrical nanopore. The pore traverses a dielectric membrane with uniform surface charge. A multi-physics solver that incorporates electrostatics, ionic transport and electroosmotic flow is developed using the OpenFOAM CFD library. In the limit of small surface charge and weak applied electric field, the numerical results of fluid flux agree with theory when the thickness of the pore h is either very small or very large compared to the pore radius a. For intermediate h/a, our simulation agrees with the composite model of electroosmotic conductance [Sherwood et al. Langmuir (in press)]. When the finite permittivity of the dieletric membrane is taken into account, pairs of toroidal counter rotating eddies appear at the corner of the nanopore that expand to fill the entire pore as the pore radius is decreased. We discuss how the topology of the eddies/stagnation points varies as the aspect ratio of the pore increases. 1 Supported by the NIH under grant 4R01HG004842. 5:19PM L3.00009 Evaluating surface electrodes and hydrophobic patches for generating vortices in nanoconfined electroosmotic flows , HARVEY ZAMBRANO, Universidad de Concepcion, MARIE FUEST, Ohio State University, NICOLAS VASQUEZ, Universidad de Concepcion, A.T. CONLISK, SHAURYA PRAKASH, Ohio State University — As a silica surface is exposed to an electrolyte, a net charge arise on the solid-liquid interface. In a confined electrolyte, a consequence of the net charged interface is the development of an imbalance of ions near the confining walls. The net charged region near the walls is called the Electrical Double Layer (EDL). A critical technology for the next generation of nanodevices, such as lab on a chip and electroosmotic pumps is controlling the EDL structure. Furthermore, important technical processes such as desalination using membranes could be improved by mitigating the concentration polarization, a phenomenon directly related to the EDL. Here, we study the generation of interfacial vortices in nanoconfined electroosmotic flows. We conduct molecular dynamics simulations of a multivalent electrolyte solution in a slit silica nanochannel. We apply axial electric fields and evaluate the response of the system as a counter charged patch is placed on the channel wall. Moreover, we study an alternative method for generating vorticity by employing hydrophobic surface patches. Charge, density and flow velocity profiles are computed. The profiles reveal that both types of patches are able to generate counter flow in electroosmotic devices. We compared the results against experiments. 5:32PM L3.00010 Electrokinetic Transport in Polyelectrolyte-Grafted Nanochannels , SIDDHARTHA DAS, Univ of Maryland-College Park — We discuss here an analytical framework for describing the streaming potential generation in charged soft nanochannels in presence of a pure pressure-driven transport. Soft nanochannels are described by considering a charged polyelectrolyte layer (or PEL) resident on the nanochannel walls. We pinpoint the key dimensionless parameters that dictate the problem. We further identify that such electrokinetic transport leads to extremely efficient electrochemomechanical energy conversion associated with the generation of streaming potential. We also demonstrate that the PEL thickness as well as the depth-dependent distribution of the polymer density within the PEL need to be quantified from the balance of the elastic, volume-exclusion and electrostatic interactions - such PEL behaviour exerts non-trivial influences in the overall electrokinetic nanochannel transport. 5:45PM L3.00011 Transport and electrochemistry based characterization of porous electrodes for CDI applications and comparison with desalination performance , CARLOS RIOS PEREZ, Northeastern University, ELLEN WILKES, LUIS GUITIERREZ, The University of Texas at Austin, CARLOS HIDROVO, Northeastern University — Development of carbonbased materials with high specific surface area at the end of last century has made researchers to look back at capacitive deionization as a potential desalination technique for brackish water. Several publications evaluate the adsorption capacity of electrode materials under different conditions. Many others present the development/characterization of new electrode materials using electrochemical analysis and other techniques. Although some work has been done to model the electro-adsorption process at the macro and micro-scale, there is still a gap to tie the characterization of the electrodes to their performance. Here a simplified one-dimensional model is used to estimate the characteristic net electro-adsorption velocities for fully-developed or developing regimes in a flow-by capacitive deionization system. This methodology is applied to three commercially available materials with very distinct structure topology to estimate electromigration velocities at a specific solution flow rate. The calculated electro-adsorption rates and other characterization parameters obtained using traditional electrochemical techniques were compared against important desalination performance parameters such as amount of salt adsorbed and desalination proficiency (amount of salt adsorbed per unit of energy). The results obtained show interesting correlations and sometimes-unexpected behavior under constant current and constant voltage operation. 5:58PM L3.00012 ABSTRACT WITHDRAWN — Monday, November 24, 2014 3:35PM - 6:11PM Session L4 Bubbles: Nucleation, Growth, Heat Transfer and Boiling — 3006 - Michel Versluis, University of Twente 3:35PM L4.00001 A high-fidelity approach towards heat transfer prediction of pool boiling , MIAD YAZDANI, ABBAS ALAHYARI, THOMAS RADCLIFF, United Technologies Research Center — A novel numerical approach is developed to simulate the multiscale problem of pool-boiling phase change with an unprecedented fidelity and cost. The particular focus is to predict the heat transfer coefficient of pool-boiling regime and its transition to critical heat flux on surfaces of arbitrary shape and roughness distribution. The large-scale of the phase change and bubble dynamics is addressed through employing off-the-shelf methods for interface tracking and interphase mass and energy transfer. The small-scale of the microlayer which forms at early stage of bubble nucleation is resolved through asymptotic approximation of the thin-film theory which provides a closed-form solution for the distribution of the micro-layer and its influence on the evaporation process. In addition, the surface roughness and its role in bubble nucleation and growth is represented based on thermodynamics of nucleation process which allows the simulation of pool boiling on any surface with known roughness and enhancement characteristics. The numerical model is validated for dynamics and hydrothermal characteristics of a single nucleated bubble on a flat surface against available literature data. In addition, the model’s prediction of pool-boiling heat transfer coefficient is verified against reputable correlations for various roughness distributions and different surface alignment. Finally, the model is employed to demonstrate pool-boiling phenomenon on enhanced structures with reentrance cavities and to explore the effect of enhancement features on thermal and hydrodynamic characteristics of these surfaces. 3:48PM L4.00002 Effect of surface micro-texture on bubble dynamics and boiling critical heat flux1 , NAVDEEP DHILLON, JACOPO BUONGIORNO, KRIPA VARANASI, Massachusetts Inst of Tech-MIT — We present results of an experimental study on the effect of surface texture on the dynamics of bubble growth and departure in pool boiling of water and correlate them to the measured values of critical heat flux (CHF) on these surfaces. Although it is well known that surface roughness or micro-texture has a significant impact on macroscale boiling parameters such as boiling heat transfer coefficient (HTC) and CHF, the physics underlying these processes is not well understood. Using high speed optical and infrared (IR) imaging, we explored the mechanism of single bubble growth and departure on micro-textured surfaces fabricated using photolithography techniques. Interestingly, we observed that the introduction of the micro-texture not only completely changed bubble dynamics and boiling surface thermal characteristics but there was a clear correlation between the micro-texture parameters and the salient bubble characteristics such as the departure diameter and frequency. To explain these results, we propose a physical model based on micro-texture-induced surface microflows supplementing the conventional bubble growth and departure theory based on buoyancy and capillary pinning forces, and verify it using CHF measurements. 1 Funding for this project is provided by Chevron Corp. 4:01PM L4.00003 Kinetics-based phase change approach for VOF method applied to boiling flow , PAOLO CIFANI1 , BERNARD GEURTS2 , University of Twente, HANS KUERTEN3 , Eindhoven University of Technology — Direct numerical simulations of boiling flows are performed to better understand the interaction of boiling phenomena with turbulence. The multiphase flow is simulated by solving a single set of equations for the whole flow field according to the one-fluid formulation, using a VOF interface capturing method. Interface terms, related to surface tension, interphase mass transfer and latent heat, are added at the phase boundary. The mass transfer rate across the interface is derived from kinetic theory and subsequently coupled with the continuum representation of the flow field. The numerical model was implemented in OpenFOAM and validated against 3 cases: evaporation of a spherical uniformly heated droplet, growth of a spherical bubble in a superheated liquid and two dimensional film boiling. The computational model will be used to investigate the change in turbulence intensity in a fully developed channel flow due to interaction with boiling heat and mass transfer. In particular, we will focus on the influence of the vapor bubble volume fraction on enhancing heat and mass transfer. Furthermore, we will investigate kinetic energy spectra in order to identify the dynamics associated with the wakes of vapor bubbles. 1 Department of Applied Mathematics, 7500 AE Enschede, NL. of Applied Mathematics, 7500 AE Enschede, NL. 3 Department of Mechanical Engineering, 5600 MB Eindhoven, NL. 2 Department 4:14PM L4.00004 Growth and collapse of a single nucleated bubble in a subcooled flow , MARYAM MEDGHALCHI, NASSER ASHGRIZ, University of Toronto — Subcooled flow on heated surfaces may result in the condensation and collapse of nucleated bubbles. A numerical study accounting for the heat and mass transfer at the gas-liquid interfaces of a nucleated bubble is performed. The model considers (i) the microlayer between the bubble and the heated surfaces, where the liquid is trapped and its temperature rises above the saturation temperature; and (ii) bubble evaporation and condensation using Hertz-Knudsen equation. The results show that the thermal boundary layer inside the bubble grows faster than that inside the liquid. This is mainly due to the buoyancy induced circulating flows inside the bubble, and different interface heat and mass transfer rates at the top and the bottom of the bubble. The calculated microlayer thicknesses are found to be less than those provided by the existing correlations. 4:27PM L4.00005 Asymptotic approach in the limit of small contact angles to sessile vapor bubble growth in a superheated environment1 , ALEXEY REDNIKOV, NICOLAS HOLLANDER, MARTA HERNANDO REVILLA, PIERRE COLINET, Université Libre de Bruxelles - TIPs — A model of nucleate pool boiling is considered, and more concretely the growth dynamics of a single spherical-cap vapor bubble on a flat superheated substrate in a large volume of an equally superheated liquid. An asymptotic scheme is developed valid in the limit of small contact angles. These are basically supposed to be the evaporation-induced ones and hence finite even in the case of a perfectly wetting liquid implied here. The consideration generally involves four regions: i) microregion, where the contact line singularities are resolved and the evaporation-induced contact angles are established, ii) Cox-Voinov region, iii) foot of the bubble, and iv) macroregion. It is only in the latter region, which remarkably appears to leading order in the form of the exterior of a sphere touching a planar surface in one point (hence a fixed geometry even for variable contact angles), that the full Navier-Stokes and heat equations are to be (numerically) resolved. 1 ESA & BELSPO PRODEX, F.R.S.-FNRS 4:40PM L4.00006 Visualization of bubble formation induced by femtosecond laser pulses in water/acetone on a time scale from sub-picosecond to microseconds , YUKI MIZUSHIMA, Graduate school of Science and Techonology, Shizuoka University, TAKAYUKI SAITO, Research Institute of Green Science and Technology, Shizuoka University — Laser induced bubble formation is usually understood as a trigger pulled by a plasma formation in a bulk media. During the plasma growth, normally, bright light emission due to excitation of the energy state of the electrons in the molecules can be observed. However, femtosecond laser pulses (fs pulses) generate bubbles through a process without bright light emission. The fs pulse leads extraordinary phenomena due to their extremely higher energy density than usual laser pulses (nanoor pico-second). We think the bubble formation by fs pulses must be different from the ordinary laser-induced cavitation. In this study, a single fs pulse was focused on water and acetone in a glass cell through several types of lens. We visualized bubble formation processes from sub-picosecond to microsecond order through time-resolved visualization. We found out a strange time-series process of refraction index changes of the media irradiated by the fs pulse: the bubble nucleation, rapid growth of bubble nucleation and interesting bubble properties. Based on these results, we will discuss a relationship between those and fs pulse peak intensity, and differences in bubble formation in water and acetone. 4:53PM L4.00007 Nucleation and ultrafast vaporization dynamics of laser-activated polymeric microcapsules1 , GUILAUME LAJOINIE, ERIK GELDERBLOM, Physics of Fluids group, University of Twente, CECIEL CHLON, MARCEL BOEHMER, Philips Research Laboratories Europe, WIENDELT STEENBERGEN, Biomedical Photonic Imaging, University of Twente, NICO DE JONG, Biomedical Engineering, Erasmus MC Rotterdam, SRIRANG MANOHAR, Biomedical Photonic Imaging, University of Twente, MICHEL VERSLUIS, Physics of Fluids group, University of Twente — Precision control of vaporization, both in space and time, has many potential applications; however, the physical mechanisms underlying controlled boiling are not well understood. The reason is the combined microscopic length scales and ultra-short timescales associated with the initiation and subsequent dynamical behavior of the vapor bubbles formed. Here we study the nanoseconds vapor bubble dynamics of laser-heated single oil-filled microcapsules using coupled optical and acoustic detection. Pulsed laser excitation leads to vapor formation and collapse, and a simple physical model captures the observed radial dynamics and resulting acoustic pressures. Continuous wave laser excitation leads to a sequence of vaporization and condensation cycles, the result of absorbing microcapsule fragments moving in and out of the laser beam. A model incorporating thermal diffusion from the capsule shell into the oil core and surrounding water reveals the mechanisms behind the onset of vaporization. Excellent agreement is observed between the modeled dynamics and experiment. 1 This work is supported by NanoNextNL, a micro and nanotechnology consortium of the Government of the Netherlands and 130 partners. 5:06PM L4.00008 Surface wetting and bubble dynamics of dielectric fluids boiling in high electric fields1 , NAVDEEP DHILLON, CHRISTOPHER LOVE, SEYED REZA MAHMOUDI, KRIPA VARANASI, Massachusetts Inst of Tech-MIT — We present results of an experimental study on the effect of high electric fields on the nature of bubble formation and departure in nucleate pool boiling of dielectric fluids. Despite some past studies looking at the application of electric fields to enhance boiling performance, the exact mechanism of interaction of these fields with the fluid/surface is not well understood. In this study, we employed high-speed optical and infrared (IR) imaging to observe changes in wetting behavior of the fluid on the solid surface and the mode of bubble formation and departure under applied electric fields. The experimental results point towards a liquid film stabilization effect of the applied electric field on the boiling surface. Both the bubble departure size and surface dry spot dynamics is visibly altered under the effect of the electric field. These effects can be attributed to the development of surface charges on the bubble microlayer adjacent to the liquid-vapor contact line, which affect the liquid receding and surface rewetting mechanisms. 1 Funding for this project is provided by Chevron Corp. 5:19PM L4.00009 Numerical study of heat transfer in bubbly flows in channels1 , SAUL PIEDRA, Universidad Nacional Autonoma de Mexico, JIACAI LU, University of Notre Dame, EDUARDO RAMOS, Universidad Nacional Autonoma de Mexico, GRETAR TRYGGVASON, University of Notre Dame — The effects of bubbles on the heat transfer in channel flows is examined by direct numerical simulations (DNS), where every continuum length and time scale is resolved. Earlier simulations of bubbles in turbulent flow in vertical channels have shown that the presence of bubbles increases the Nusselt number, compared to flow without bubbles. This is the case for both nearly spherical as well as deformable bubbles, even though the flow structure is very different. Here we examine how bubbles modify the heat transfer in horizontal and sloping channels, for both laminar and turbulent flows. The results show that the bubbles generally increase the heat transfer, but the exact amount depends of the degree that the bubbles modify the structure of the flow. Preliminary efforts to use the results to aid in the development of models for the average flow are discussed and early results for more complex transient flows with bubbles of different sizes are shown. 1 Research supported by CONACyT 5:32PM L4.00010 GPU-based multi-resolution direct numerical simulation of multiphase flows with phase change , CHRISTOPHER J. FORSTER, MARC K. SMITH, Georgia Institute of Technology — Nucleate pool boiling heat transfer can be enhanced in several ways to increase the critical heat flux (CHF) and delay the transition to film boiling. Changes to the heated surface geometry using open microchannels and direct forcing of the vapor bubbles using acoustic interfacial excitation are being investigated for their effects on the CHF. The numerical simulation of boiling with these effects lends itself to multi-resolution techniques due to the multiple length and time scales present during evolution of the bubbles from initial nucleation in the microchannels to forming a bubble cloud above the heated surface. To this end, a wavelet multi-resolution boiling simulation based on a parallel GPU architecture is being developed to solve the compressible Navier-Stokes equations using a dual time stepping method with preconditioning to alleviate the stiffness problems associated with the liquid phase. Interface tracking is handled by the level-set method with a prescribed interface thickness based on the maximum amount of local grid refinement desired, which can approach the physical interface thickness. Initial cases to validate the simulation will be demonstrated, including the rising bubble test problem. 5:45PM L4.00011 Vapor explosions during the impact of molten tin droplets into a liquid pool , NADIA KOURAYTEM, ER QIANG LI, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology — High-speed video imaging is used to study the impact of a molten tin droplet into a liquid pool. Three different regimes have been identified as nucleation boiling, film boiling or vapor explosion. The latter generally comprises two stages; during the first stage, vapor gets entrapped into the molten tin drop and then, at a second stage, the vapor is superheated by the tin material, creating a violent expansion (explosion). It was observed that the addition of surfactant to the fluid pool could promote the explosion and make it occur at a lower temperature. Furthermore, other parameters such as the pool liquid surface tension, boiling temperature, viscosity and molten tin temperature have been varied to examine the explosion dynamics. 5:58PM L4.00012 A generation method of single bubbles of various sizes using a slitting elastic tube and acoustic pressure wave , TOSHIYUKI SANADA, KIMIHIKO ABE, Shizuoka University — A bubble generation method (Sanada and Abe, Rev. Sci. Instrum. 2013) using a slitting elastic tube and acoustic pressure wave in the gas phase can produce single bubbles of various sizes. In this study, we experimentally investigated the bubble generation mechanism in a slitting elastic tube. We used high-speed photography to observe the bubble generation process and slit motion in different liquids with different surface tensions. The results indicated that there was no significant difference in the slit opening time even if the amplitude of the acoustic pressure wave was changed, and the generated bubble radius was determined by the opening displacement of the slit, which was governed by the surface tension. In addition, the shape oscillation of a bubble due to surface tension promoted its detachment from an elastic tube with poor wettability. Monday, November 24, 2014 3:35PM - 6:11PM Session L5 Biofluids: Cellular and Molecular Biophysics — 3008 - Manu Prakash, Stanford University 3:35PM L5.00001 Coupling a mechanosensitive channel with a vesicle under shear flow , ON SHUN PAK, Princeton University, YUAN NAN YOUNG, New Jersey Institute of Technology, SHRAVAN VEERAPANENI, University of Michigan, HOWARD STONE, Princeton University — Mechanosensitive channels enable cells to respond to their local environment. Continuum mechanical models have been proposed to describe how bilayer deformation induced by the transmembrane protein and the membrane tension influence the free energy of channel gating under static conditions. The dynamics of mechanosensitive channels under flow conditions however remains largely unexplored. Cells under flow display interesting features not observed under static environments. Here we present a model coupling a mechanosensitive channel with the dynamics of a vesicle under shear flow to investigate how the channel gating responds to hydrodynamic stress. The model could be used to investigate the release of signaling molecules, transport of ions or drugs across cell membranes under flow in biological systems, as well as the design and control of channel gating in synthetic cells. 3:48PM L5.00002 Formation and organization of protein domains in the immunological synapse , ANDREAS CARLSON, Harvard University, School of Engineering and Applied Sciences and the Wyss Institute, L. MAHADEVAN, Harvard University, School of Engineering and Applied Sciences and Department of Physics — The cellular basis for the adaptive immune response during antigen recognition relies on a specialized protein interface known as the immunological synapse. Here, we propose a minimal mathematical model for the dynamics of the IS that encompass membrane mechanics, hydrodynamics and protein kinetics. Simple scaling laws describe the dynamics of protein clusters as a function of membrane stiffness, rigidity of the adhesive proteins, and fluid flow in the synaptic cleft. Numerical simulations complement the scaling laws by quantifying the nucleation, growth and stabilization of proteins domains on the size of the cell. Direct comparison with experiment suggests that passive dynamics suffices to describe the short-time formation and organization of protein clusters, while the stabilization and long time dynamics of the synapse is likely determined by active cytoskeleton processes triggered by receptor binding. Our study reveals that the fluid flow generated by the interplay between membrane deformation and protein binding kinetics can assist immune cells in regulating protein sorting. 4:01PM L5.00003 Could Life Originate between Mica Sheets? , HELEN HANSMA, University of California at Santa Barbara — Muscovite mica has many advantages as a site for the origins of life. Some of these advantages are: A. Spaces between mica sheets serve as cell-like compartments. B. K+ ions bridge Muscovite mica sheets, providing a high K+ environment, as found in all living cells. C. Mica’s hexagonal 0.5-nm clay crystal lattice is comparable to the length of the amino acids, sugars, and nucleotides that polymerize to form life’s major biological macromolecules. D. Mechanical energy from mica sheets, moving in response to water flows and temperature changes, provide an endless energy source for forming chemical bonds, rearranging polymers, and blebbing off protocells in a primitive form of cell division. [1-3] How might fluid dynamics in the planar nanometer- to micron-high spaces between mica sheets affect the processes involved in the origins of life? [1] Hansma, H G (2009) In Probing Mechanics at Nanoscale Dimensions. N. Tamura, A. Minor, C. Murray and L. Friedman.Warrendale, PA, Materials Research Society. 1185: II03-15. [2] Hansma, H G (2010) Journal of Theoretical Biology 266(1): 175. [3] Hansma, H G (2013) J. Biol. Struct. Dynamics 31(8): 888. 4:14PM L5.00004 A mathematical model of stress generation in microtubule pair interactions , FANG FANG, Courant Inst, MEREDITH BETTERTON, Physics Department, University of Colorado at Boulder, MICHAEL SHELLEY, Courant Institute of Mathematical Sciences — Microtubules and motor proteins are basic ingredients in many cellular structures and of new biosynthetic “active” suspensions. The interaction of microtubules with their surrounding fluid medium depends fundamentally upon the force generation afforded them through cross-linking motile motor proteins. Here we develop a simple mathematical model, based on the statistical mechanics, motor proteins binding and unbinding, to study the generation of active fluid stresses. We study the role and contributions of “polarity sorting” and “tether” relaxation on the generation of intrinsic, destabilizing stresses. 4:27PM L5.00005 Mechanoregulation of molecular motors in flagella , HERMES GADELHA, University of Oxford — Molecular motors are nano-biological machines responsible for exerting forces that drive movement in living organisms, from cargo transport to cell division and motility. Interestingly, despite the inherent complexity of many interacting motors, order and structure may arise naturally, as exemplified by the harmonic, self-organized undulatory motion of the flagellum. The real mechanisms behind this collective spontaneous oscillation are still unknown, and it is challenging task to measure experimentally the molecular motor dynamics within the flagellar structure in real time. In this talk we will explore different competing hypotheses that are capable of generating flagellar bending waves that “resemble” in-vitro observations, emphasizing the need for further mathematical analysis and model validation. It also highlight that this is a fertile and challenging area of inter-disciplinary research for applied mathematicians and demonstrates the importance of future observational and theoretical studies in understanding the underlying mechanics of these motile cell appendages. 4:40PM L5.00006 Hydrodynamics of pronuclear migration , EHSSAN NAZOCKDAST, Courant Institute, NYU, DANIEL NEEDLEMAN, Harvard School of Engineering and Applied Sciences, MICHAEL SHELLEY, Courant Institute, NYU — Microtubule (MT) filaments play a key role in many processes involved in cell devision including spindle formation, chromosome segregation, and pronuclear positioning. We present a direct numerical technique to simulate MT dynamics in such processes. Our method includes hydrodynamically mediated interactions between MTs and other cytoskeletal objects, using singularity methods for Stokes flow. Long-ranged many-body hydrodynamic interactions are computed using a highly efficient and scalable fast multipole method, enabling the simulation of thousands of MTs. Our simulation method also takes into account the flexibility of MTs using Euler-Bernoulli beam theory as well as their dynamic instability. Using this technique, we simulate pronuclear migration in single-celled Caenorhabditis elegans embryos. Two different positioning mechanisms, based on the interactions of MTs with the motor proteins and the cell cortex, are explored: cytoplasmic pulling and cortical pushing. We find that although the pronuclear complex migrates towards the center of the cell in both models, the generated cytoplasmic flows are fundamentally different. This suggest that cytoplasmic flow visualization during pronuclear migration can be utilized to differentiate between the two mechanisms. 4:53PM L5.00007 Characterization of Intracellular Streaming and Traction Forces in Migrating Physarum Plasmodia , SHUN ZHANG, RUEDI MEILI, Univ of California - San Diego, ROBERT GUY, Univ of California - Davis, JUAN LASHERAS, JUAN C. DEL ALAMO, Univ of California - San Diego — Physarum plasmodium is a model organism for cell migration that exhibits fast intracellular streaming. Single amoebae were seeded and allowed to move on polyacrilamide gels that contained 0.5-micron fluorescent beads. Joint time-lapse sequences of intracellular streaming and gel deformation were acquired respectively in the bright and fluorescent fields under microscope. These images were analyzed using particle image velocimetry (PIV) algorithms, and the traction stresses applied by the amoebae on the surface were computed by solving the elastostatic equation for the gel using the measured bead displacements as boundary conditions. These measurements provide, for the first time, a joint characterization of intracellular mass transport and the forces applied on the substrate of motile amoeboid cells with high resolution in both time and space, enables a through study about the locomotive mechanism and the relation between intracellular flow and traction stress, shedding light on related biomimetic research. The results reveal a pronounced auto-oscillation character in intracellular flow, contact area, centroid speed and strain energy, all with the same periodicity about 100 seconds. Locomotion modes that were distinct in flow/ traction stress pattern as well as migration speed have been discovered and studied. 5:06PM L5.00008 Coordination of Flow and Traction in Migration of Amoeboid Physarum polycephalum: Model and Measurement , OWEN LEWIS, University of Utah, ROBERT GUY, UC Davis, SHUN ZHANG, JUAN CARLOS DEL ALAMO, UC San Diego — In this research, we develop a computational model of crawling Physarum based on the Immersed Boundary Method. Our model incorporates the effects of cell cytoplasm, the internal cytoskeleton and adhesions to the substrate. Cytoplasmic flows and traction stresses predicted by the model are compared to experimentally measured values obtained using simultaneous Traction Force Microscopy (TFM) and Particle Image Velocimetry (PIV). Of particular interest are stresses generated by flow and how transmission of stresses to the substrate is coordinated. We identify methods of adhesion-flow coordination which are consistent with experiments. Certain consisten coordinations are seen to be “optimal” with regards to crawling speed, and robust to perturbations in the extracellular environment. 5:19PM L5.00009 Viral video: Live imaging of virus-host encounters , KWANGMIN SON, MIT, JEFFREY S. GUASTO, Tufts university, ANDRES CUBILLOS-RUIZ, SALLIE W. CHISHOLM, MIT, MATTHEW B. SULLIVAN, University of Arizona, ROMAN STOCKER, MIT — Viruses are non-motile infectious agents that rely on Brownian motion to encounter and subsequently adsorb to their hosts. Paradoxically, the viral adsorption rate is often reported to be larger than the theoretical limit imposed by the virus-host encounter rate, highlighting a major gap in the experimental quantification of virus-host interactions. Here we present the first direct quantification of the viral adsorption rate, obtained using live imaging of individual host cells and viruses for thousands of encounter events. The host-virus pair consisted of Prochlorococcus MED4, a 800 nm small non-motile bacterium that dominates photosynthesis in the oceans, and its virus PHM-2, a myovirus that has a 80 nm icosahedral capsid and a 200 nm long rigid tail. We simultaneously imaged hosts and viruses moving by Brownian motion using two-channel epifluorescent microscopy in a microfluidic device. This detailed quantification of viral transport yielded a 20-fold smaller adsorption efficiency than previously reported, indicating the need for a major revision in infection models for marine and likely other ecosystems. 5:32PM L5.00010 Self-assembly of protein fibrils in stable circular Couette flow1 , SAMANTHA MCBRIDE, CHRISTOPHER TILGER, AMIR HIRSA, Rensselaer Polytechnic Institute, JUAN LOPEZ, Arizona State University — Fluid flows are known to contribute to the chemical dynamics of self-assembling protein fibrils yet the roles of mixing and shear have not been elucidated. These long, crystalline structures are ubiquitous in-vivo and strongly associated with many neurodegenerative disorders. Understanding the mechanism of formation is a significant challenge because of the variety of gradients proteins are exposed to in biological fluid channels. A stable circular Couette flow device was constructed in order to conduct comprehensive tests on the effects of pure shear on a protein solution initially free of any pre-existing aggregates. The protein insulin was sheared at various Reynolds numbers at normothermia (37◦ C). Changes in fluid properties are observed at the onset of fibril precipitation, as the elongated structures generate complex particle-laden fluid dynamics. Measurements include fibrillization lag times, images of protein fibrils induced by shear, and changes to viscosity after exposure to shear. Discussion will cover biological implications and the role of fluid mechanics in pathogenesis of neurodegenerative disorders. 1 NASA grant NNX13AQ22G, NSF grants CBET-1064644 and CBET-1064498 5:45PM L5.00011 A Fast Multipole Method and a Metropolis Method for Coarse-grained Brownian Dynamics Simulations of a DNA with Hydrodynamic Interactions1 , SZU-PEI FU, YUAN-NAN YOUNG, SHIDONG JIANG, New Jersey Inst of Tech — The coarse-grained molecular dynamics (MD) or Brownian dynamics (BD) simulation is a particle-based approach that has been applied to a wide range of biological problems that involve interaction with water molecules. The simulations are often numerically expensive for exploring long-time dynamics over meso-scales due to the amount of water molecules needed for capturing the non-local hydrodynamic interactions (HIs). In this paper a fast multipole method for computing the HIs and a metropolis method for molecular dynamics are validated by comparing against both experiments and simulations of a single DNA molecule in linear flow. In addition, it is shown that the Metropolis integration scheme for self–adjoint diffusions can be used to expedite the time it takes to prepare the initial configuration of the macromolecule for the BD simulations. Further numerical tests show that 3 the fast multipole method scales linearly to the total number N of beads for the long-chain molecule when N > ∼ O(10 ) while other numerical algorithms scale 2 to O(N ) (at least). 1 Y.-N. Young acknowledges support from NSF under grant DMS-1222550. 5:58PM L5.00012 Modeling and design of light powered biomimicry micropump utilizing transporter proteins1 , JIN LIU, TSUN-KAY JACKIE SZE, PRASHANTA DUTTA, Washington State University, Pullman WA — The creation of compact micropumps to provide steady flow has been an on-going challenge in the field of microfluidics. We present a mathematical model for a micropump utilizing Bacteriorhodopsin and sugar transporter proteins. This micropump utilizes transporter proteins as method to drive fluid flow by converting light energy into chemical potential. The fluid flow through a microchannel is simulated using the Nernst-Planck, Navier-Stokes, and continuity equations. Numerical results show that the micropump is capable of generating usable pressure. Designing parameters influencing the performance of the micropump are investigated including membrane fraction, lipid proton permeability, illumination, and channel height. The results show that there is a substantial membrane fraction region at which fluid flow is maximized. The use of lipids with low membrane proton permeability allows illumination to be used as a method to turn the pump on and off. This capability allows the micropump to be activated and shut off remotely without bulky support equipment. This modeling work provides new insights on mechanisms potentially useful for fluidic pumping in self-sustained bio-mimic microfluidic pumps. 1 This work is supported in part by the National Science Fundation grant CBET-1250107 Monday, November 24, 2014 3:35PM - 6:11PM Session L6 Biofluids: From Birds and Bats to Insects — 3010 - David Lentink, Stanford University 3:35PM L6.00001 Near and far wake structures behind freely flying bats1 , COSIMA SCHUNK, SHARON M. SWARTZ, KENNETH S. BREUER, Brown University — While pseudo-volumetric reconstructions of the wakes of flying animals, based on transverse (Trefftz) wake measurements, have become a well-established tool in the study of animal aerodynamics in recent years, there are a number of concerns that persist regarding their use in estimating drag and flight efficiency. Here we report on stereo particle image velocimetry (PIV) measurements behind freely flying bats (Eptesicus fuscus) in both the transverse and streamwise planes. The streamwise plane measurements are taken on the wing as well as in the near and far wake region up to eight chord lengths behind the bat. By organizing the data according to the flight speed, wingbeat phase and the spanwise position of the laser sheet on the wing we are able to connect specific features of the wing and body geometry with observed wake structures and thereby construct a detailed time-space map of the wake. Furthermore, we can quantitatively assess wake distortion and assess the validity of lift and drag estimates based on transverse wake measurements. 1 supported by AFOSR 3:48PM L6.00002 Aerodynamic role of dynamic wing morphing in hummingbird maneuvering flight1 , YAN REN, GREGORY SHALLCROSS, HAIBO DONG, University of Virginia, XINYAN DENG, Purdue University, BRET TOBALSKE, Montana State University, FLOW SIMULATION RESEARCH GROUP TEAM, BIO-ROBOTICS LAB COLLABORATION, UNIVERSITY OF MONTANA FLIGHT LABORATORY COLLABORATION — The flexibility and deformation of hummingbird wing gives hummingbird a great degree of control over fluid forces in flapping flight. Unlike insect wing’s passive deformation, hummingbird wing employs a more complicated wing morphing mechanism through both active muscle control and passive feather-air interaction, which results in highly complex 3D wing topology variations during the unsteady flight. Three camera high speed (1000 fps) high resolution digital video was taken and digitized to measure 3D wing conformation in all its complexity during steady flying and maneuvering. Results have shown that the dynamic wing morphing is more prominent in maneuvering flight. Complicated cambering and twisting patterns are observed along the wing pitching axis. A newly developed immersed boundary method which realistically models wing-joint-body of the hummingbird is then employed to simulate the flow associated with dynamic morphing. The simulations provide a first of its kind glimpse of the fluid and vortex dynamics associated with dynamic wing morphing and aerodynamic force computations allow us to gain a better understanding of force producing mechanisms in hummingbird maneuvering flight. 1 This work is supported by AFOSR FA9550-12-1-007 and NSF CEBT-1313217 4:01PM L6.00003 Short revolving wings enable hovering animals to avoid stall and reduce drag , DAVID LENTINK, JAN W. KRUYT, Department of Mechanical Engineering, Stanford University, GERTJAN F. HEIJST, Department of Applied Physics, Eindhoven University of Technology, DOUGLAS L. ALTSHULER, Department of Zoology, University of British Columbia — Long and slender wings reduce the drag of airplanes, helicopters, and gliding animals, which operate at low angle of attack (incidence). Remarkably, there is no evidence for such influence of wing aspect ratio on the energetics of hovering animals that operate their wings at much higher incidence. High incidence causes aircraft wings to stall, hovering animals avoid stall by generating an attached vortex along the leading edge of their wings that elevates lift. Hypotheses that explain this capability include the necessity for a short radial distance between the shoulder joint and wing tip, measured in chord lengths, instead of the long tip-to-tip distance that elevates aircraft performance. This stems from how hovering animals revolve their wings around a joint, a condition for which the precise effect of aspect ratio on stall performance is unknown. Here we show that the attachment of the leading edge vortex is determined by wing aspect ratio with respect to the center of rotation–for a suite of aspect ratios that represent both animal and aircraft wings. The vortex remains attached when the local radius is shorter than 4 chord lengths, and separates outboard on more slender wings. Like most other hovering animals, hummingbirds have wing aspect ratios between 3 and 4, much stubbier than helicopters. Our results show this makes their wings robust against flow separation, which reduces drag below values obtained with more slender wings. This revises our understanding of how aspect ratio improves performance at low Reynolds numbers. 4:14PM L6.00004 Do hummingbirds use a different mechanism than insects to flip and twist their wings?1 , JIALEI SONG, HAOXIANG LUO, Vanderbilt University, TYSON HEDRICK, The University of North Carolina at Chapel Hill — Hovering hummingbirds flap their wings in an almost horizontal stroke plane and flip the wings to invert the angle of attack after stroke reversal, a strategy also utilized by many hovering insects such as fruit flies. However, unlike insects whose wing actuation mechanism is only located at the base, hummingbirds have a vertebrate musculoskeletal system and their wings contain bones and muscles and thus, they may be capable of both actively flipping and twisting their wings. To investigate this issue, we constructed a hummingbird wing model and study its pitching dynamics. The wing kinematics are reconstructed from high-speed imaging data, and the inertial torques are calculated in a rotating frame of reference using mass distribution data measured from dissections of hummingbird wings. Pressure data from a previous CFD study of the same wing kinematics are used to calculate the aerodynamic torque. The results show that like insect wings, the hummingbird wing pitching is driven by its own inertia during reversal, and the aerodynamic torque is responsible for wing twist during mid-stroke. In conclusion, our study suggests that their wing dynamics are very similar even though their actuation systems are entirely different. 1 This research was supported by the NSF 4:27PM L6.00005 Flight testing of live Monarch butterflies to determine the aerodynamic benefit of butterfly scales1 , AMY LANG, University of Alabama, JACOB CRANFORD, University of Alabama Huntsville, JASMINE CONWAY, Tennessee State University, NATHAN SLEGERS, George Fox University, NICOLE DECHELLO, Smith College, JACOB WILROY, University of Alabama — Evolutionary adaptations in the morphological structure of butterfly scales (0.1 mm in size) to develop a unique micro-patterning resulting in a surface drag alteration, stem from a probable aerodynamic benefit of minimizing the energy requirement to fly a very lightweight body with comparably large surface area in a low Re flow regime. Live Monarch butterflies were tested at UAHuntsville’s Autonomous Tracking and Optical Measurement (ATOM) Laboratory, which uses 22 Vicon T40 cameras that allow for millimeter level tracking of reflective markers at 515 fps over a 4 m x 6 m x 7 m volume. Data recorded included the flight path as well as the wing flapping angle and wing-beat frequency. Insects were first tested with their scales intact, and then again with the scales carefully removed. Differences in flapping frequency and/or energy obtained during flight due to the removal of the scales will be discussed. Initial data analysis indicates that scale removal in some specimens leads to increased flapping frequencies for similar energetic flight or reduced flight speed for similar flapping frequencies. Both results point to the scales providing an aerodynamic benefit, which is hypothesized to be linked to leading-edge vortex formation and induced drag. 1 Funding from the National Science Foundation (CBET and REU) is gratefully acknowledged. 4:40PM L6.00006 The Effects of Scales on Autorotation of Monarch Butterfly Forewings1 , NICOLE DECHELLO, Smith College, AMY LANG, University of Alabama — The wings of Monarch butterflies (Danus plexippus) have scales of approximately 100 micrometers that cover their wings in a roof-shingle pattern, and these scales are hypothesized to help improve flight efficiency for their long migration. The aerodynamic effects of the scales, particularly involving the leading edge vortex formation and resulting lift, were investigated by observing the natural autorotation of forewing specimen when dropped in quiescent air. A high-speed camera recorded drop tests of 32 forewings both with scales and after removal of the scales. It was found that the scales, on average, comprised 17% of the forewing mass. Tracking software was used to analyze the videos for several parameters, including descent speed and radius of rotation. 1 NSF ECE Grant #1358991 supported the first author as an research experience for undergraduate (REU) student. 4:53PM L6.00007 Effect of shape on wing kinematics control in dragonfly maneuvering flight1 , AYODEJI BODE-OKE, SAMANE ZEYGHAMI, HAIBO DONG, University of Virginia, FSRG TEAM — Flying insects execute aerial maneuvers through fine modulations in their wing kinematics. It’s yet not known that to what extend the wing kinematics can be controlled and altered by the insect. To investigate the question, we recorded a yaw turn maneuver of a dragonfly in free flight. Our measurements show that this flight consists of two kinematically and dynamically distinct phases; acceleration and deceleration. In a systematic study, we first clipped the left forewing and then the right forewing of the same dragonfly and recorded its yaw turn maneuver. The signatures (in kinematics and dynamics) of the two identified phases stay unchanged by wing damage but the duration of both phases extends. The rotational velocity of the body drops dramatically by wing damage which implies the dragonfly is incapable of controlling the wing kinematics to achieve similar performance as in the intact wing. Our results suggest that the wing kinematics control is tightly influenced by the wing shapes and the aerodynamics of flapping flight. 1 This work was supported by NSF grant number CEBT-1313217 and REU program. 5:06PM L6.00008 How do dragonflies recover from falling upside down? , Z. JANE WANG, JAMES MELFI JR, Cornell University, ANTHONY LEONARDO, Janelia Farm Research Campus, HHMI — We release dragonflies from a magnetic tether so that they fall from an initially upside down orientation. To recover, the dragonflies roll their body 180 degrees every time. This set up offers an effective method for eliciting a stereotypical turn so that we can collect a large amount of data on the same turn. From the wing and body kinematics, we can tease out the strategy dragonflies use to roll their body. We record these flights with three zoomed in high-speed video cameras. By filming at 4000 to 8000fps, we measure the wing twist along each of the four wings as a part of the 3D wing kinematics. The shape of the wing twist depends on the interaction between the aerodynamic torque and the torque exerted by muscles, therefore providing clues on which of their four wings actively participate in creating the turn. By applying dynamic calculations to the measured kinematics, we further deduce the amount of torques dragonflies exert in order to turn. 5:19PM L6.00009 Pitch-Perfect: How Do Flies Control Their Pitch Angle During Aerial Stumbles? , SAMUEL WHITEHEAD, Cornell Univ, LUCA CANALE, École Polytechnique, TSEVI BEATUS, ITAI COHEN, Cornell Univ — The successful flight of flapping-wing insects is contingent upon a complex and beautiful relationship between sensory input, neural response, and muscular actuation. In particular, the inherent instabilities of flapping-wing flight require insects like D. melanogaster to constantly sense, process, and adjust for in-flight stumbles. Here we present an analysis of the mechanisms for pitch control in D. melanogaster. By gluing small ferromagnetic pins to the backs of the flies and applying an external magnetic field, we induce torques along the flies’ pitch axis during free flight. Using an automated hull reconstruction technique developed in the lab, we analyze these torque events and the flies’ subsequent recoveries in order to characterize the flies’ response to external perturbations. Ultimately, we aim to develop a reduced-order controller model that will capture the salient aspects of the flies’ recovery mechanism. 5:32PM L6.00010 Free flight simulations of a dragonfly-like flapping wing-body model by the immersed boundary-lattice Boltzmann method1 , TAKAJI INAMURO, KEISUKE MINAMI, KOSUKE SUZUKI, Dept. Aeronautics and Astronautics, Kyoto University — Free flights of the dragonfly-like flapping wing-body model are numerically investigated by using the immersed boundary-lattice Boltzmann method (IB-LBM). First, we simulate free flights of the model without the pitching rotation for various values of the phase lag angle φ between the forewing and the hindwing motions. We find that the wing-body model goes forward in spite of φ, and the model with φ =0◦ and 90◦ goes upward against gravity. The model with φ = 180◦ goes almost horizontally, and the model with φ = 270◦ goes downward. Secondly, we simulate free flights with the pitching rotation for various values of the phase lag angle φ. It is found that in spite of φ the wing-body model turns gradually in the nose-up direction and goes back and down as the pitching angle Θ c increases. That is, the wing-body model cannot make a stable forward flight without control. Finally, we show a way to control the pitching motion by changing the lead-lag angle γ(t). We propose a simple proportional controller of γ(t) which makes stable flights within Θc = ±5◦ and works well even for a large disturbance. 1 The authors acknowledge the HPCI System Research Project (Project ID: hp120112) 5:45PM L6.00011 High Order Large Eddy Simulation (LES) of Gliding Snake Aerodynamics: Effect of 3D Flow on Gliding Performance , YANN DELORME, Technion Israel Institute of Technology, SYED HARRIS HASSAN, Purdue University, JAKE SOCHA, VirginiaTech, PAVLOS VLACHOS, Purdue University, STEVEN FRANKEL, Technion Israel Institute of Technology — Chrysopelea paradisi are snakes that are able to glide over long distances by morphing the cross section of their bodies from circular to a triangular airfoil, and undulating through the air. Snake glide is characterized by relatively low Reynolds number and high angle of attack as well as three dimensional and unsteady flow. Here we study the 3D dynamics of the flow using an in-house high-order large eddy simulation code. The code features a novel multi block immersed boundary method to accurately and efficiently represent the complex snake geometry. We investigate the steady state 3-dimensionality of the flow, especially the wake flow induced by the presence of the snake’s body, as well as the vortex-body interaction thought to be responsible for part of the lift enhancement. Numerical predictions of global lift and drag will be compared to experimental measurements, as well as the lift distribution along the body of the snake due to cross sectional variations. Comparisons with previously published 2D results are made to highlight the importance of 3-dimensional effects. Additional efforts are made to quantify properties of the vortex shedding and Dynamic Mode Decomposition (DMD) is used to analyse the main modes responsible for the lift and drag forces. 5:58PM L6.00012 Pitching motion control of a butterfly-like 3D flapping wing-body model , KOSUKE SUZUKI, Institute of Engineering, Shinshu University, KEISUKE MINAMI, TAKAJI INAMURO, Dept. Aeronautics and Astronautics, Kyoto University — Free flights and a pitching motion control of a butterfly-like flapping wing-body model are numerically investigated by using an immersed boundary-lattice Boltzmann method. The model flaps downward for generating the lift force and backward for generating the thrust force. Although the model can go upward against the gravity by the generated lift force, the model generates the nose-up torque, consequently gets off-balance. In this study, we discuss a way to control the pitching motion by flexing the body of the wing-body model like an actual butterfly. The body of the model is composed of two straight rigid rod connected by a rotary actuator. It is found that the pitching angle is suppressed in the range of ±5◦ by using the proportional-plus-integral-plus-derivative (PID) control for the input torque of the rotary actuator. Monday, November 24, 2014 3:35PM - 6:11PM Session L7 Biofluids: Cardiovascular Fluid Mechanics II — 3012 - Shawn Shadden, University of California, Berkeley 3:35PM L7.00001 Transition to turbulence in pulsating pipe flow1 , DUO XU, Friedrich-Alexander-Universitaet Erlangen-Nuernberg, SASCHA WARNECKE, BJOERN HOF, Institute of Science and Technology Austria, MARC AVILA, Friedrich-Alexander-Universitaet Erlangen-Nuernberg — We report an experimental investigation of the transition to turbulence in a pulsating pipe flow. This flow is a prototype of various pulsating flows in both nature and engineering, such as in the cardiovascular system where the onset of turbulence is often possibly related to various diseases (e.g., the formation of aneurysms). The experiments are carried out in a straight rigid pipe using water with a sinusoidal modulation of the flow rate. The governing parameters, Reynolds number, Womersley number α (dimensionless pulsating frequency) and the pulsating amplitude A, cover a wide range 3 < α < 23 and 0 < A < 1. To characterize the transition to turbulence, we determine how the characteristic lifetime of turbulent spots (/puffs) are affected by the pulsation. While at high pulsation frequencies (α > 12) lifetimes of turbulent spots are entirely unaffected by the pulsation, at lower frequencies they are substantially affected. With decreasing frequency much larger Reynolds numbers are needed to obtain spots of the same characteristic lifetime. Hence at low frequency transition is delayed significantly. In addition the effect of the pulsation amplitude on the transition delay is quantified. 1 Duo Xu would like to acknowledge the support from Humboldt Foundation 3:48PM L7.00002 Quantification of disturbed wall shear stress patterns in complex cardiovascular flows , AMIRHOSSEIN ARZANI, SHAWN C. SHADDEN, University of California Berkeley — Wall shear stress (WSS) affects the cardiovascular system in numerous ways, and is thought to play an important role in the pathology of many cardiovascular diseases. The (endothelial) cells lining the inner wall of blood vessels, and perhaps the cells inside the vessel wall, can actively sense WSS and respond both chemically and mechanically. The complexity of WSS in cardiovascular flows extends both spatially and temporally. Furthermore, WSS has magnitude and direction. These facets make simple quantification of WSS in cardiovascular applications difficult. In this study we propose a framework to quantify measures such as WSS angle gradient, WSS magnitude gradient, WSS angle time derivative and WSS magnitude time derivative. We will explain the relation of these parameters to the tensorial WSS gradient and WSS vector time derivative, and propose a new methodology to unify these concepts into a single measure. The correlation between these metrics and more common WSS metrics used in the literature will be demonstrated. For demonstration, these methods will be used for the quantification of complex blood flow inside abdominal aortic aneurysms. 4:01PM L7.00003 Uncertainty quantification in virtual surgery predictions for single ventricle palliation , DANIELE SCHIAVAZZI, ALISON MARSDEN, Univ of California - San Diego — Hemodynamic results from numerical simulations of physiology in patients are invariably presented as deterministic quantities without assessment of associated confidence. Recent advances in cardiovascular simulation and Uncertainty Analysis can be leveraged to challenge this paradigm and to quantify the variability of output quantities of interest, of paramount importance to complement clinical decision making. Physiological variability and errors are responsible for the uncertainty typically associated with measurements in the clinic; starting from a characterization of these quantities in probability, we present applications in the context of estimating the distributions of lumped parameters in 0D models of single-ventricle circulation. We also present results in virtual Fontan palliation surgery, where the variability of both local and systemic hemodynamic indicators is inferred from the uncertainty in pre-operative clinical measurements. Efficient numerical algorithms are required to mitigate the computational cost of propagating the uncertainty through multiscale coupled 0D-3D models of pulsatile flow at the cavopulmonary connection. This work constitutes a first step towards systematic application of robust numerical simulations to virtual surgery predictions. 4:14PM L7.00004 The Direct Effect of Flexible Walls on Fontan Connection Fluid Dynamics , MIKE TREE, KILEY FAGAN, AJIT YOGANATHAN, Georgia Inst of Tech — The current standard treatment for sufferers of congenital heart defects is the palliative Fontan procedure. The Fontan procedure results in an anastomosis of major veins directly to the branched pulmonary arteries bypassing the dysfunctional ventricle. This total cavopulmonary connection (TCPC) extends life past birth, but Fontan patients still suffer long-term complications like decreased exercise capacity, protein-losing enteropathy, and pulmonary arteriovenous malformations (PAVM). These complications have direct ties to fluid dynamics within the connection. Previous experimental and computation studies of Fontan connection fluid dynamics employed rigid vessel models. More recent studies utilize flexible models, but a direct comparison of the fundamental fluid dynamics between rigid and flexible vessels only exists for a computational model, without a direct experimental validation. Thus, this study was a direct comparison of fluid dynamics within a rigid and two compliant idealized TCPCs. 2D particle image velocimetry measurements were collected at the connection center plane. Results include power loss, hepatic flow distribution, fluid shear stress, and flow structure recognition. The effect of flexible walls on these values and clinical impact will be discussed. 4:27PM L7.00005 GPU-accelerated Lattice Boltzmann method for anatomical extraction in patient-specific computational hemodynamics , H. YU, Mechanical Engineering, IUPUI, IN, Z. WANG, Computer Science, Kent State University, OH, C. ZHANG, Resources and Environmental Science, Wuhan University, China, N. CHEN, Mechanical Engineering, IUPUI, Indianapolis, IN, Y. ZHAO, Computer Science, Kent State University, OH, A.P. SAWCHUK, M.C. DALSING, Vascular Surgery, School of Medicine, Indiana University, IN, S.D. TEAGUE, Radiology and Imaging Sciences, School of Medicine, Indiana University, IN, Y. CHENG, Resources and Environmental Science, Wuhan University, China — Existing research of patient-specific computational hemodynamics (PSCH) heavily relies on software for anatomical extraction of blood arteries. Data reconstruction and mesh generation have to be done using existing commercial software due to the gap between medical image processing and CFD, which increases computation burden and introduces inaccuracy during data transformation thus limits the medical applications of PSCH. We use lattice Boltzmann method (LBM) to solve the level-set equation over an Eulerian distance field and implicitly and dynamically segment the artery surfaces from radiological CT/MRI imaging data. The segments seamlessly feed to the LBM based CFD computation of PSCH thus explicit mesh construction and extra data management are avoided. The LBM is ideally suited for GPU (graphic processing unit)-based parallel computing. The parallel acceleration over GPU achieves excellent performance in PSCH computation. An application study will be presented which segments an aortic artery from a chest CT dataset and models PSCH of the segmented artery. 4:40PM L7.00006 One dimensional modeling of blood flow in large networks , XIAOFEI WANG, Institut Jean le Rond d’Alembert UPMC (Univ. Paris 6), PIERRE-YVES LAGREE, Institut Jean le Rond d’Alembert CNRS, JOSE-MARIA FULLANA, Institut Jean le Rond d’Alembert UPMC (Univ. Paris 6), SYLVIE LORTHOIS, Institut de Mecanique des Fluides de Toulouse, CNRS, INSTITUT DE MECANIQUE DES FLUIDES DE TOULOUSE COLLABORATION — A fast and valid simulation of blood flow in large networks of vessels can be achieved with a one-dimensional viscoelastic model. In this paper, we developed a parallel code with this model and computed several networks: a circle of arteries, a human systemic network with 55 arteries and a vascular network of mouse kidney with more than one thousand segments. The numerical results were verified and the speedup of parallel computing was tested on multi-core computers. The evolution of pressure distributions in all the networks were visualized and we can see clearly the propagation patterns of the waves. This provides us a convenient tool to simulate blood flow in networks. 4:53PM L7.00007 Quantitative Assessment of Wall Shear Stress in an Aortic Coarctation – Impact of Virtual Interventions , MATTS KARLSSON, MAGNUS ANDERSSON, Depertment of Management and Engineering, Linkoping University, JONAS LANTZ, Department of Science and Technology, Linkoping University — Turbulent and wall impinging blood flow causes abnormal shear forces onto the lumen and may play an important role in the pathogenesis of numerous cardiovascular diseases. In the present study, wall shear stress (WSS) and related flow parameters were studied in a pre-treated aortic coarctation (CoA) as well as after several virtual interventions using computational fluid dynamics (CFD). Patient-specific geometry and flow conditions were derived from magnetic resonance imaging (MRI) data. Finite element analysis was performed to acquire six different dilated CoAs. The unsteady pulsatile flow was resolved by large eddy simulation (LES) including non-Newtonian blood rheology. Preintervention, the presence of jet flow wall impingement caused an elevated WSS zone, with a distal region of low and oscillatory WSS. After intervention, cases with a more favorable centralized jet showed reduced high WSS values at the opposed wall. Despite significant turbulence reduction post-treatment, enhanced regions of low and oscillatory WSS were observed for all cases. This numerical method has demonstrated the morphological impact on WSS distribution in an CoA. With the predictability and validation capabilities of a combined CFD/MRI approach, a step towards patient-specific intervention planning is taken. 5:06PM L7.00008 An effective fractal-tree closure model for simulating blood flow in large arterial networks , PARIS PERDIKARIS, Brown University, LEOPOLD GRINBERG, IBM T.J Watson Research Center, GEORGE KARNIADAKIS, Brown University — The aim of the present work is to address the closure problem for hemodynamics simulations by developing a flexible and effective model that accurately distributes flow in the downstream vasculature and can stably provide a physiological pressure outflow boundary condition. We model blood flow in the sub-pixel vasculature by using a nonlinear 1D model in self-similar networks of compliant arteries that mimic the structure and hierarchy of vessels in the meso-vascular regime. The proposed model accounts for wall viscoelasticity and non-Newtonian flow effects in arterioles, overcomes cut-off radius sensitivity issues by introducing a monotonically decreasing artery length to radius ratio across different generations of the fractal tree, and convergences to a periodic state in just two cardiac cycles. The resulting fractal trees typically consist of thousands to millions of arteries, posing the need for efficient parallel algorithms. To this end, we have developed a scalable hybrid MPI/OpenMP solver that is capable of computing near real-time solutions. The proposed model is tested on a large patient-specific cranial network returning physiological flow and pressure wave predictions without requiring any parameter estimation or calibration procedures. 5:19PM L7.00009 A Computational Model for Thrombus Formation in Response to Cardiovascular Implantable Devices , JOHN HORN, Lawrence Livermore National Laboratory, Texas A&M University, JASON ORTEGA, Lawrence Livermore National Laboratory, DUNCAN MAITLAND, Texas A&M University — Cardiovascular implantable devices elicit complex physiological responses within blood. Notably, alterations in blood flow dynamics and interactions between blood proteins and biomaterial surface chemistry may lead to the formation of thrombus. For some devices, such as stents and heart valves, this is an adverse outcome. For other devices, such as embolic aneurysm treatments, efficient blood clot formation is desired. Thus a method to study how biomedical devices induce thrombosis is paramount to device development and optimization. A multiscale, multiphysics computational model is developed to predict thrombus formation within the vasculature. The model consists of a set of convectiondiffusion-reaction partial differential equations for blood protein constituents involved in the progression of the clotting cascades. This model is used to study thrombus production from endovascular devices with the goal of optimizing the device design to generate the desired clotting response. This work was performed in part under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. 5:32PM L7.00010 In Vitro MRV-based Hemodynamic Study of Complex Helical Flow in a Patient-specific Jugular Model , SARAH KEFAYATI, GABRIEL ACEVEDO-BOLTON, HENRIK HARALDSSON, DAVID SALONER, Uni- versity of California, San Francisco, Radiology & Biomedical Imaging — Neurointerventional Radiologists are frequently requested to evaluate the venous side of the intracranial circulation for a variety of conditions including: Chronic Cerebrospinal Venous Insufficiency thought to play a role in the development of multiple sclerosis; sigmoid sinus diverticulum which has been linked to the presence of pulsatile tinnitus; and jugular vein distension which is related to cardiac dysfunction. Most approaches to evaluating these conditions rely on structural assessment or two dimensional flow analyses. This study was designed to investigate the highly complex jugular flow conditions using magnetic resonance velocimetry (MRV). A jugular phantom was fabricated based on the geometry of the dominant jugular in a tinnitus patient. Volumetric three-component time-resolved velocity fields were obtained using 4D PC-MRI –with the protocol enabling turbulence acquisition– and the patient-specific pulsatile waveform. Flow was highly complex exhibiting regions of jet, high swirling strength, and strong helical pattern with the core originating from the focal point of the jugular bulb. Specifically, flow was analyzed for helicity and the level of turbulence kinetic energy elevated in the core of helix and distally, in the post-narrowing region. 5:45PM L7.00011 Velocimetry modalities for secondary flows in a curved artery test section1 , KARTIK V. BULUSU, George Washington University, CHRISTOPHER J. ELKINS, ANDREW J. BANKO, Stanford University, MICHAEL W. PLESNIAK, George Washington University, JOHN K. EATON, Stanford University — Secondary flow structures arise due to curvature-related centrifugal forces and pressure imbalances. These flow structures influence wall shear stress and alter blood particle residence times. Magnetic resonance velocimetry (MRV) and particle image velocimetry (PIV) techniques were implemented independently, under the same physiological inflow conditions (Womersley number = 4.2). A 180-degree curved artery test section with curvature ratio (1/7) was used as an idealized geometry for curved arteries. Newtonian blood analog fluids were used for both MRV and PIV experiments. The MRV-technique offers the advantage of three-dimensional velocity field acquisition without requiring optical access or flow markers. Phase-averaged, two-dimensional, PIV-data at certain cross-sectional planes and inflow phases were compared to phase-averaged MRV-data to facilitate the characterization of large-scale, Dean-type vortices. Coherent structures detection methods that included a novel wavelet decomposition-based approach to characterize these flow structures was applied to both PIV- and MRV-data. The overarching goal of this study is the detection of motific, three-dimensional shapes of secondary flow structures using MRV techniques with guidance obtained from high fidelity, 2D-PIV measurements. 1 This material is based in part upon work supported by the National Science Foundation under Grant Number CBET-0828903, and GW Center for Biomimetics and Bioinspired Engineering (COBRE). 5:58PM L7.00012 Flow Structures in a Healthy and Plaqued Artificial Artery using Fully Index Matched Vascular Flow Facility , FARAZ MEHDI, AKASH JAIN, JIAN SHENG, Texas Tech University — Particle Image Velocimetry measurements are made in a closed loop fully index matched flow facility to study the flow structures and flow wall interactions in healthy and diseased model arteries. The test section is 0.63 m long and the facility is capable of emulating both steady and pulsatile flows under physiologically relevant conditions. The model arteries are in-house developed compliant polymer (PDMS) tubes with 1 cm diameter and 1 mm wall thickness. The Reynolds numbers of flows vary up to 20,000. The plaque is simulated by introducing a radially asymmetric bump that can be varied in shape, size and compliancy. The overall compliancy of the model can be also controlled by varying ratio between the elastomer and the curing agent. The tubes are doped with particles allowing the simultaneous measurements of wall deformation and flows over it. The working fluid in the facility is NaI and is refractive index matched to the PDMS model. This allows flow measurement very close to the wall and measurement of wall shear stress. The aim of this study is to characterize the changes in flow as the compliancy and geometry of blood vessels change due to age or disease. These differences can be used to develop a diagnostic tool to detect early onset of vascular diseases. Monday, November 24, 2014 3:35PM - 5:45PM — Session L9 Minisymposium I: Frontiers of Computational Science in Transport Phenomena 3014/3016 - Gianluca Iaccarino, Stanford University 3:35PM L9.00001 Petascale Flow Simulations Using Particles and Grids1 , PETROS KOUMOUTSAKOS, ETH Zurich — How to chose the discretization of flow models in order to harness the power of available computer architectures? Our group explores this question for particle (vortex methods, molecular and dissipative particle dynamics) and grid based (finite difference, finite volume) discretisations for flow simulations across scales. I will discuss methodologies to transition between these methods and their implementation in massively parallel computer architectures. I will present simulations ranging from flows of cells in microfluidic channels to cloud cavitation collapse at 14.5 PFLOP/s. 1 This research was supported by the European Research Council, the Swiss National Science Foundation and the Swiss National Supercomputing Center. 4:01PM L9.00002 DNS/LES of Complex Turbulent Flows beyond Petascale1 , PAUL FISCHER, University of Illinois — Petascale computing platforms currently feature million-way parallelism and it is anticipated that exascale computers with billion-way concurrency will be deployed by 2020. In this talk, we explore the potential of computing at these scales with a focus on turbulent fluid flow and heat transfer in a variety of applications including nuclear energy, combustion, oceanography, vascular flows, and astrophysics. Following Kreiss and Oliger ’72, we argue that high-order methods are essential for scalable simulation of transport phenomena. We demonstrate that these methods can be realized at costs equivalent to those of low-order methods having the same number of gridpoints. We further show that, with care, efficient multilevel solvers having bounded iteration counts will scale to billion-way concurrency. Using data from leading-edge platforms over the past 25 years, we analyze the scalability of state-of-the-art solvers to predict parallel performance on exascale architectures. The analysis sheds light on the expected scope of exascale physics simulations and provides insight to design requirements for future algorithms, codes, and architectures. 1 Supported by DOE Applied Mathematics Research Program 4:27PM L9.00003 High-Fidelity Simulations of Multiphysics Systems , FRANK HAM, Cascade Technologies Inc — A pacing theme in the high-fidelity simulations of multi-physics flows is the continual push towards constitutive models that reflect the underlying physics more closely than ever before. At the same time, to impact the design and understanding of real fluidic devices, these models must ultimately be developed in the setting of a highly flexible computational infrastructure capable of both massive parallelism and geometric flexibility. This theme is illustrated using two multi-physics simulations that provide new incite into the behavior of complex fluidic devices. In the first, a novel unstructured Volume-of-Fluid (VoF) method is applied to simulate the liquid fuel atomization processes in a complex high shear nozzle typical of realistic gas turbine injectors. The simulation make aggressive use of directional grid adaptation to support the local resolution of critical instability mechanisms associated with the atomization process. In a companion example, the prediction of flow field and noise in a subsonic jet is linked critically to modeling and resolution of the nozzle boundary layers. 4:53PM L9.00004 Computational aeroacoustics of turbulent high-speed jets1 , JOSEPH W. NICHOLS, Aerospace Engineering and Mechanics Dept., University of Minnesota — Despite significant scientific investigation, jet noise remains a large component of the overall noise generated by supersonic aircraft. Experiments show that alterations to nozzle geometry, such as the addition of chevrons to the nozzle lip, can significantly reduce jet noise. In this talk, we assess unstructured large eddy simulation as a tool for predicting and understanding the aeroacoustic effects of complex geometry upon supersonic jets. Body-fitted, adaptive meshes are used to simulate the flow inside, around and through complicated nozzles, and results are validated against experimental measurements. High-fidelity simulations utilizing as many as one million processors simultaneously will be discussed, allowing for a detailed description of interactions between turbulence, shocks, and acoustics. This includes observations of the phenomenon of “crackle” noise in heated supersonic jets. We will briefly discuss challenges met and overcome along this frontier of com putational science, and describe how information extracted from the high-fidelity simulations can be used to construct accurate reduced-order models useful for aeroacoustic design. 1 Computational resources were provided by the Argonne Leadership Computing Facility at Argonne National Laboratory and the ERDC and AFRL supercomputing centers. 5:19PM L9.00005 Multiscale modeling of brain blow flow , GEORGE KARNIADAKIS, Brown University — Cardiovascular pathologies, such as brain aneurysms, are affected by the global blood circulation as well as by the local microrheology. Hence, developing computational models for such cases requires the coupling of disparate spatial and temporal scales often governed by diverse mathematical descriptions, e.g., by partial differential equations (continuum, 3D or 1D) and ordinary differential equations for discrete particles (atomistic). However, interfacing atomistic-based with continuum-based domain discretizations is a challenging problem that requires both mathematical and computational advances. We will present a physical model of the brain vasculature consisting at the macro level of all major arteries (about 200 down to 0.5 mm), at the mesoscale the fractal arteriolar tree (more than 10 millions down to 20 nm) and at the microscale the capillary bed. Correspondingly, we employ three different methods to model the total brain vasculature by developing proper interface conditions at each level. We will present examples from aneurysms and other hematological diseases, where red blood cell rheology is modeled explicitly. Monday, November 24, 2014 3:35PM - 6:11PM Session L10 Microscale Flows: Particles — 3005 - Haim Bau, University of Pennsylvania 3:35PM L10.00001 Towards mico-ThFFF for polymer analysis: Lattice-Boltzmann based simulations , MICHAEL ANTONELLI, Department of Biology and Marine Biology, Roger Williams University, JENNIFER KREFT PEARCE, Department of Physics, Roger Williams University — Thermophoresis describes a behavior, observed at micro-scales, in which particles migrate due to a temperature gradient. The purpose of this project is to study the parameters that have the greatest effect on thermophoresis and to use these properties to design a device for separating biological macromolecules using extremely small samples. A Lattice-Boltzmann based computer simulation of a microfluidic cell was used to determine the conditions under which DNA molecules, in a buffered salt solution, will exhibit this phenomenon. The simulation monitored particle positions within the cell, beginning from random initial conditions. Particle-solvent and particle-particle interactions were examined. Particle-particle interactions were modeled using the Lennard-Jones potential. By modifying the distance at which potential is minimized as well as the magnitude of the potential, conditions that increase the response of the molecule to the temperature gradient were observed. Once satisfactory conditions had been determined, separation of particles in a theoretical microfluidic device was simulated. The periodic boundary conditions were changed and a more dynamic channel was modeled. Unidirectional flow fields as well as particles with differing thermophoretic properties were simulated in the micro-channel and their concentrations across the channel measured. 3:48PM L10.00002 Glancing, reversing, tumbling, and sliding: sedimentation near walls in viscous fluids , WILLIAM MITCHELL, SAVERIO SPAGNOLIE, University of Wisconsin - Madison — The sedimentation of ellipsoidal particles near a wall in a viscous fluid has been studied from a numerical perspective by a number of authors, but analytical solutions have been given only in special cases, such as for spherical particles. As an application of the method of images, the dynamics of ellipsoids of arbitrary aspect ratio in a wall-bounded Stokes flow may be reduced to a system of ordinary differential equations. In many cases the system leads to analytical descriptions of the particle motion which agree very well with full numerical simulations. As an application, we investigate the conditions under which the “glancing” and “reversing” trajectories first observed by Russel et al. prevail, and we identify two new possibilities: a periodic “tumbling” trajectory for nearly spherical bodies and a “sliding” trajectory which occurs when the wall is inclined at a small angle from the vertical. The sliding trajectory is an attracting fixed point for the dynamics, and thus may have applications in sorting processes for heterogeneous dilute suspensions. 4:01PM L10.00003 Model colloid system for interfacial sorption kinetics , PAUL SALIPANTE, STEVEN HUDSON, National Institute of Standards and Technology — Adsorption kinetics of nanometer scale molecules, such as proteins at interfaces, is usually determined through measurements of surface coverage. Their small size limits the ability to directly observe individual molecule behavior. To better understand the behavior of nanometer size molecules and the effect on interfacial kinetics, we use micron size colloids with a weak interfacial interaction potential as a model system. Thus, the interaction strength is comparable to many nanoscale systems (less than 10 kB T). The colloid-interface interaction potential is tuned using a combination of depletion, electrostatic, and gravitational forces. The colloids transition between an entropically trapped adsorbed state and a desorbed state through Brownian motion. Observations are made using an LED-based Total Internal Reflection Microscopy (TIRM) setup. The observed adsorption and desorption rates are compared theoretical predictions based on the measured interaction potential and near wall particle diffusivity. This experimental system also allows for the study of more complex dynamics such as nonspherical colloids and collective effects at higher concentrations. 4:14PM L10.00004 Theory and simulation of acoustic interaction forces between small particles in an ideal fluid , HENRIK BRUUS, Department of Physics, Technical University of Denmark, Denmark, GLAUBER T. SILVA, Instituto de Fı́sica, Universidade Federal de Alagoas, Brazil — We present a theoretical expression for the acoustic interaction force between small spherical particles suspended in an ideal fluid exposed to an external acoustic wave as used in, say, microchannel acoustophoresis. The acoustic interaction force is the part of the acoustic radiation force on one given particle involving the scattered waves from the other particles. The particles, either compressible liquid droplets or elastic microspheres, are considered to be much smaller than the acoustic wavelength. In this so-called Rayleigh limit, the acoustic interaction forces between the particles are well approximated by gradients of pair-interaction potentials with no restriction on the inter-particle distance. The theory is applied to studies of the acoustic interaction force on a particle suspension in either standing or traveling plane waves. The results show aggregation regions along the wave propagation direction, while particles may attract or repel each other in the transverse direction. In addition, a mean-field approximation is developed to describe the acoustic interaction force in an oil-in-water emulsion. 4:27PM L10.00005 Clogging in a microfluidic hourglass , ALVARO MARIN, MASSIMILIANO ROSSI, CHRISTIAN J. KÄHLER, Bundeswehr University Munich — One of the main disadvantages of microfluidic devices is their tendency to clog when a high density of particles or droplets is forced through them. The same problem is often encountered in classical granular flows in silos and hourglasses. It is well-known that hourglasses work optimally when the particle-to-neck ratio is within certain ratio without interruption (Zuriguel et al., Phys. Rev. E, 2003), while arching occurs for particle-to-neck ratios above d/D ≈ 2. Microfluidic devices normally work in geometries in which d/D > 10, in which the arching probability is negligible. Clogging is nonetheless possible, but mainly due to the accumulation of particles at the walls (Wyss et al, Phys. Rev. E, 2006). On the other hand, clogging by arching in systems with d/D ∼ O(1) are expected to have radically different physics and statistics, due to collective behavior and hydrodynamic interactions. To study these regimes, we study microfluidic devices with a bottleneck of squared crossed section and side length D through which we force polystyrene particles with diameters from d/D ≈ 1 to 0.25 at packing fractions ranging from 10% up to 50%. Our results show that clogging of such systems have more in common with granular flows in hourglasses than expected. 4:40PM L10.00006 Dynamics of flexible fibers in shear flow , AGNIESZKA SLOWICKA, ELIGIUSZ WAJNRYB, MARIA EKIEL-JEZEWSKA, Institute of Fundamental Technological Research Polish Academy of Sciences — We consider dynamics and shape evolution of a flexible non-Brownian fiber in steady shear flow under low-Reynolds-number. Fibers are described by the bead-spring model. Their evolution is determined by solving the Stokes equations with the use of the multipole method, corrected for lubrication within the accurate numerical code HYDROMULTIPOLE. The fibers are initially aligned with the ambient flow. Owing to symmetry, their motion takes place in the plane perpendicular to vorticity direction. We investigate migration of fibers across the flow and quantify their shape evolution. Depending on the ratio of the fiber bending energy to its hydrodynamic energy, we find out different modes of the dynamics. Distinction between these modes is based on values of the fiber migration velocity, its tumbling frequency, curvature and length of the end-to-end distance. 4:53PM L10.00007 Nanohybrid particle-particle interaction in Dissipative Particle Dynamics (DPD) simulations1 , MINH VO, The University of Oklahoma, DIMITRIOS PAPAVASSILIOU, The University of Oklahoma & NSF — Carbon nanotube (CNT) hybrid particles have recently received attention in hydrocarbon reservoir technology due to their ability to stabilize water/oil interface. CNTs tend to agglomerate in solution, so polymers are used to prevent this phenomenon forming nanohybrid (NH) particles (i.e., CNT-polymer particles). In the presence of PVP polymer, CNTs can be dispersed and stabilized successfully. In this work, the coarse graining DPD method is utilized to explore NH particle interactions in water. The NH particles are created after the equilibrium of the system with cylindrical CNTs and polymers is reached. To compute the interaction force, one NH particle is stationary and another is moving around it. Then, the effect of distance and angle between the two main axes of the particles on the interaction force is determined. Based on these data, a general equation to describe this interaction is obtained. Besides, different sizes of particles are considered in order to find out the effect of the CNT aspect ratio on the interaction force. Additionally, the steric effect of polymer on particle-particle interaction is studied. 1 Acknowledgements: Advanced Energy Consortium (AEC BEG08-022) 5:06PM L10.00008 Hydrodynamic interactions for complex-shaped nanocarriers in targeted drug delivery , YAOHONG WANG, DAVID ECKMANN, RAVI RADHAKRISHNAN, PORTONOVO AYYASWAMY, University of Pennsylvania — Nanocarrier motion in a blood vessel involves hydrodynamic and Brownian interactions, which collectively dictate the efficacy in targeted drug delivery. The shape of nanocarriers plays a crucial role in drug delivery. In order to quantify the flow and association properties of elliptical nanoparticles, we have developed an arbitrary Lagrangian-Eulerian framework with capabilities to simulate the hydrodynamic motion of nanoparticles of arbitrary shapes. We introduce the quaternions for rotational motion, and two collision models, namely, (a) an impulse-based model for wall–particle collision, and (b) the short-range repulsive Gay-Berne potential for particle-particle collision. We also study the red blood cell and nanocarrier (such as ellipsoid) interactions. We compare our results with those obtained for a hard sphere model for both RBCs and nanocarriers. Supported by NIH through grant U01-EB016027. 5:19PM L10.00009 Analysis of angle effect on particle flocculation in branch flow , KARTHIK PRASAD, KATHRYN FINK, DORIAN LIEPMANN, Univ of California - Berkeley — Hollow point microneedle drug delivery systems are known to be highly susceptible to blockage, owing to their very small structures. This problem has been especially noted when delivering suspended particle solutions, such as vaccines. Attempts to reduce particle flocculation in such devices through surface treatments of the particles have been largely unsuccessful. Furthermore, the particle clog only forms at the mouths of the microneedle structures, leaving the downstream walls clear. This implies that the sudden change in length scales alter the hydrodynamic interactions, creating the conditions for particle flocculation. However, while it is known that particle flocculation occurs, the physics behind the event are obscure. We utilize micro-PIV to observe how the occurrence and formation of particle flocculation changes in relation to the angle encountered by particle laden flow into microfluidic branch structures. The results offer the ability to optimize particle flocculation in MEMS devices, increasing device efficacy and longevity. 5:32PM L10.00010 Capillary Interactions of Micro-particles on Curved Interfaces , NIMA SHARIFIMOOD, LU YAO, IRIS LIU, KATHLEEN STEBE, University of Pennsylvania — Microparticles trapped at fluid interfaces interact by capillarity to migrate, form structures and find preferred locations. These phenomena are exploited to organize colloids at fluid interfaces, impacting emulsion stabilization and forming the basis for advanced materials which exploit, e.g. the mechanics or optical properties of the structures which form. Interface curvature plays a strong role in microparticle behavior by acting as a field which directs microparticle migration. Here, we discuss the behavior of microparticles with pinned contact lines at the oil-water interface on interfaces with well-defined curvature fields. Once the particles attach to the interface, they migrate in deterministic paths towards sites of high curvature. These experiments are well described by our analysis. We theoretically determine the disturbance field imposed by particles via an asymptotic analysis, and have quantified the associated capillary energy and the capillary force. Forces can be understood simply in terms of slope variation of the disturbed interface along the contact line. Capillary energies are expressed in closed form as a function of mean and deviatoric curvatures of the interface prior to the particles deposition. Pair interactions between particles will be also discussed. 5:45PM L10.00011 Controlling Lateral Inertial Migration Rate of Particles in Microchannels , ARMIN KARIMI, SAMUEL BRAY, DINO DI CARLO, UCLA — It was previously demonstrated that particles in confined channels can migrate across streams due to the net inertial lift force acting on them. The initial location of particles within the channel cross-section is shown to effect the migration time as particles starting at different locations experience a different history of lift forces. This initial variation in distribution of focusing positions of particles upstream was a limiting factor in achieving precise control over the migration time in previous studies. In order to improve uniformity of the focusing position, a set of sequential cylindrical pillars is integrated to one side of the channel which is shown to aid particles in achieving a single stable equilibrium position, by inducing a net helical flow. The modified focusing positions are characterized as a function of pillar diameter and spacing for various channel Reynolds numbers. Using this initial focusing channel, a comprehensive numerical and experimental study is performed to characterize the range of lateral migration rate for particles as a function of particle position, and flow rates of each stream for a given finite Reynolds number and channel geometry. The tool developed in this study can be used to achieve precise migration characteristics for the microparticles crossing fluid streams in microchannels. 5:58PM L10.00012 Tuning particle focusing in inertial microfluidic devices , KAITLYN HOOD, SOROUSH KAHKESHANI, DINO DI CARLO, MARCUS ROPER, Univ of California - Los Angeles — Particles in microfluidic devices at finite Reynolds number are subject to two forces: (i) inertial focusing and (ii) particle-particle interactions. Although microfluidic chips exploit these forces to manipulate particles for particle/cell sorting and high throughput flow cytometry, the forces are not understood well enough to allow rational design of devices that can tune and attenuate particle focusing. We present a mathematical model addressing both inertial focusing and particle interactions, and we apply our model to various channel geometries to determine the balance of forces. In addition, we present experimental data that illustrate the accuracy of our model. We will address the following questions: Why do high aspect ratio channels favor two equilibrium positions? Why do particle chains form? Monday, November 24, 2014 3:35PM - 6:11PM — Session L11 Microscale Flows: Oscillations and Magnetic Manipulation 3007 - Brian Storey, Olin College of Engineering 3:35PM L11.00001 Theory and numerical analysis of thermoviscous effects in ultrasoundinduced acoustic streaming in microchannels , PETER BARKHOLT MULLER, HENRIK BRUUS, Technical University of Denmark, DTU Physics — We present a numerical study of the thermoviscous effects on the acoustic streaming flow generated by an ultrasound standing wave resonance in a long straight microfluidic channel. These effects enter through the temperature and density dependence of the fluid viscosity. The resulting magnitude of the streaming flow is calculated and characterized numerically, and remarkably, we find that even for thin acoustic boundary layers, the channel height affects the magnitude of the streaming flow. For the special case of a sufficiently large channel height we have successfully validated our numerics with analytical results from 2011 by Rednikov and Sadhal for a planar wall. Furthermore, the time-averaged energy transport in the system is analyzed, and the time-averaged second-order temperature perturbation of the fluid is calculated. 3:48PM L11.00002 Oscillations of free cylinders at low Reynolds numbers inside a Hele-Shaw cell1 , J.-P. HULIN, Univ Paris Sud, CNRS, Lab FAST, Bt 502, Campus Univ. Orsay, F-91405 (France), V. D’ANGELO, L. GIANORIO, M. CACHILE, GMP-FIUBA, Paseo Colon 850, 1063, Buenos Aires (Argentina), CONICET (Argentina), H. AURADOU, Univ Paris Sud, CNRS, Lab FAST, Bt 502, Campus Univ. Orsay, F-91405 (France), B. SEMIN, LPS Laboratory, Dept. Physique ENS, 24. rue Lhomond, 75231 Paris Cedex 05 (France) — We study two instabilities of a horizontal free cylinder in a vertical viscous Hele-Shaw flow: they are shown experimentally to depend critically on the transverse and lateral confinement of the flow characterized respectively by the ratios D/H (resp. L/W ) of the diameter (resp. the length) of the cylinder to the gap (resp. the width) of the cell. The onset of the instabilities depends essentially on D/H. For 0.4 ≤ D/H ≤ 0.6, we observe transverse horizontal oscillations of the cylinder perpendicular to the walls: their frequency is constant with D/H and L/W at a given vertical cylinder velocity Vc . This instability is locally 2D along the length of the cylinder and controlled by the local relative velocity Vrloc of the cylinder and the fluid: it occurs down to Reynolds numbers Reloc = Vrloc H/ν ≃ 15, i.e. below the vortex shedding threshold (150 − 250) for fixed cylinders between parallel planes. These results are compared to 2D numerical simulations. For D/H ≥ 0.55, we observe a fluttering motion with periodic oscillations of the tilt angle of the cylinder from the horizontal and of its horizontal position: their frequency decreases as L/W increases and is independent of D/H and Vc . 1 Research supported by the LIA: Physics and Mechanics of Fluids. 4:01PM L11.00003 Taylor Dispersion in Oscillatory Flow in Rectangular Channels , JINKEE LEE, Sungkyunkwan University, ANUBHAV TRIPATHI, Brown University, ANUJ CHAUHAN, University of Florida — This paper focuses on exploring the effect of the side walls on dispersion in oscillatory Poiseuille flows in rectangular channels. The method of multiple time scales with regular expansions is utilized to obtain D∗ 2 ∗ . The dispersion coefficient is of the form 3D = f (Ω ≡ ωh , Sc ≡ D , χ ≡ w ) where P e ≡ <u>h , analytical expressions for the effective dispersivity D3D D ν h D P e2 <u>is the root mean square of the cross-section averaged velocity, ω is the angular velocity, 2w and 2h are the width and the height of the cross-section, D is the solute diffusivity, ν is the fluid kinematic viscosity. The analytical results are compared with full numerical simulations and asymptotic expressions. Also effect of various parameters on dispersion coefficient is explored. For small oscillation frequency Ω, the dispersion coefficient approaches the time averaged ∗ scales as Pe 2 /Ω2 where Pe=<u>h/D. Due to its relative simplicity, the 2D model is frequently utilized dispersion of the Poiseuille flow and for large Ω, D3D for calculating dispersion in channels. However at small dimensionless frequencies, the 2D model can significantly underestimate the dispersion, particularly for channels with large χ. At large Ω, the dispersion coefficient predicted from the 2D model becomes reasonably accurate, particularly for channels with large χ. For a square channel, the 2D prediction is reasonably accurate for all frequencies. The results of this study will enhance our understanding of transport in microscale systems that are subjected to oscillating flows, and potentially aid technological advances in diverse areas relevant to microfluidic devices. 4:14PM L11.00004 Experimental and numerical analysis of the steady streaming around a cylinder pair , E. GALÁN-VICENTE, W. COENEN, Dept. Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, 28911 Leganés, Spain — The steady streaming motion that develops around a cylinder pair in small-amplitude oscillatory flow, is studied experimentally and numerically. The axes of the cylinders are perpendicular to the plane of motion, and the angle that the flow makes with the line connecting the cylinder centers, as well as the distance between them, is varied. We focus on the regime where the ratio ǫ of the amplitude of oscillation to a cylinder radius a is small. A theoretical analysis shows that the action of the Reynolds stresses in thin Stokes shear-wave layers close to the cylinder surfaces induces a steady streaming motion that persists at the edge of these layers with velocities of O(ǫU ), where U is the velocity amplitude of the basic oscillatory flow. This streaming velocity at its turn drives an outer flow, governed by the steady Navier-Stokes equations with streaming Reynolds number Rs = ǫU a/ν. We consider cases with Rs ≫ 1. The steady equations are solved numerically, imposing the streaming velocity obtained from the asymptotic analysis as a slip boundary condition at the cylinder surfaces. The resuling flow patterns show good agreement with experimental flow visualizations in the form of phase-averages over various oscillation cycles. 4:27PM L11.00005 The Temporal Resolution of Laser Induced Fluorescence Photobleaching Anemometer1 , WEI ZHAO, University of South Carolina, FANG YANG, Carnegie Mellon University, GUIREN WANG, University of South Carolina — Recently, in microfluidics, electrokinetic flows are widely used on micromixer designing. However, there is unfortunately no valid velocimeter today that can measure the random velocity fluctuation at high temporal and spatial resolution simultaneously in the complicated flow circumstance. We recently introduced laser induced fluorescence photobleaching anemometer (LIFPA), which has been successfully used in the measurement of velocity field in AC electrically driven microflow. Here, we theoretically study the temporal resolution (TR) of and experimentally verify, LIFPA can have simultaneously ultrahigh temporal (∼4 µs) and spatial (∼203 nm) resolution and can measure velocity fluctuation up to at least 2 kHz, whose corresponding wave number is about 6 × 106 1/m in an electrokinetically forced unsteady flow in microfluidics. The measurement of LIFPA is also compared with the widely used micro Particle Imaging Velocimetry (µPIV). We found, at the inlet, due to multiple uncertainties, the velocity fluctuations by µPIV exhibits apparently smaller values than that by LIFPA. But at downstreams, where velocity fluctuation is much lower than at the inlet and the uncertainties of complicated electric field on particles becomes smaller, LIFPA and µPIV indicate similar measurement. 1 The work was supported by NSF under grant no. CAREER CBET-0954977 and MRI CBET-1040227, respectively. 4:40PM L11.00006 Compact Two-Liquid Microfluidic Hyperelastic Capacitive Strain Sensors1 , SHANLIANGZI LIU, XIAODA SUN, KONRAD RYKACZEWSKI, Arizona State University — Applications of liquid metal microfluidic devices include flexible electronics, biomedical devices, and soft robotics. In addition to single channel resistive strain sensors, two channel capacitive sensors have also been developed. However, these capacitive strain sensors have low capacitance with a footprint of about a square centimeter, making strain-output correlation quite complex [1]. To address this issue, we developed a compact two liquid single straight channel capacitive strain sensor with a dielectric liquid sandwiched between two liquid metal electrodes. Formation of the capacitor with a liquid dielectric instead of PDMS enables capacitance increase through selection of high permittivity liquid. Using a custom experimental setup, we show that use of water and glycerol instead of silicone oil in-between the liquid metal electrodes can increase the device capacitance by fivefold. We discuss the effect of channel diameter, dielectric spacing, interfacial meniscus shape, and the liquid flow on device capacitance as well as response to strain. In addition, we discuss the effect of gallium oxide shell formation at the dielectric-liquid metal interface. [1] Fassler A. and Majidi C. Smart Mater. Struct. 22 (2013). 1 KR acknowledges startup funding from ASU. 4:53PM L11.00007 Sound-induced Interfacial Dynamics in a Microfluidic Two-phase Flow , SZE YI MAK, HO CHEUNG SHUM, The Univ of Hong Kong — Retrieving sound wave by a fluidic means is challenging due to the difficulty in visualizing the very minute sound-induced fluid motion. This work studies the interfacial response of multiphase systems towards fluctuation in the flow. We demonstrate a direct visualization of music in the form of ripples at a microfluidic aqueous-aqueous interface with an ultra-low interfacial tension. The interface shows a passive response to sound of different frequencies with sufficiently precise time resolution, enabling the recording of musical notes and even subsequent reconstruction with high fidelity. This suggests that sensing and transmitting vibrations as tiny as those induced by sound could be realized in low interfacial tension systems. The robust control of the interfacial dynamics could be adopted for droplet and complex-fiber generation. 5:06PM L11.00008 Conformal coating of non-spherical magnetic particles using microfluidics , BYEONG-UI MOON, NAVID HAKIMI, DAE KUN HWANG, SCOTT TSAI, Ryerson University, DEPARTMENT OF MECHANICAL AND INDUSTRIAL ENGINEERING TEAM, DEPARTMENT OF CHEMICAL ENGINEERING COLLABORATION — We present the conformal coating of non-spherical magnetic particles in a microfluidic channel. We first prepare three-dimensional (3D) bullet-shaped magnetic microparticles using stop-flow lithography. We then suspend the bullet-shaped microparticles in an aqueous solution, and flow the particle suspension with a co-flow of a non-aqueous mixture. A magnetic field gradient from a permanent magnet pulls the microparticles in the transverse direction to the fluid flow, until the particles reach the interface between the immiscible fluids. In a physical domain characterized by a low particle Reynolds number and a high magnetic Bond number, we observe that the microparticles cross the oil-water interface, and then become coated by a thin film of the aqueous fluid. When we increase the two-fluid interfacial tension by reducing the surfactant concentration, we observe that the particles become trapped at the interface. We use this observation to approximate the magnetic susceptibility of the manufactured non-spherical microparticles, which are not known a priori. Using fluorescence imaging, we confirm the uniformity of the thin film coating along the surface of the bullet-shaped particles. 5:19PM L11.00009 Particle Transport and Size Sorting in Bubble Microstreaming Flow , RAQEEB THAMEEM, BHARGAV RALLABANDI, CHENG WANG, SASCHA HILGENFELDT, University of Illinois at Urbana-Champaign — Ultrasonic driving of sessile semicylindrical bubbles results in powerful steady streaming flows that are robust over a wide range of driving frequencies. In a microchannel, this flow field pattern can be fine-tuned to achieve size-sensitive sorting and trapping of particles at scales much smaller than the bubble itself; the sorting mechanism has been successfully described based on simple geometrical considerations. We investigate the sorting process in more detail, both experimentally (using new parameter variations that allow greater control over the sorting) and theoretically (incorporating the device geometry as well as the superimposed channel flow into an asymptotic theory). This results in optimized criteria for size sorting and a theoretical description that closely matches the particle behavior close to the bubble, the crucial region for size sorting. 5:32PM L11.00010 Interfacial deformation and jetting of a magnetic fluid1 , SHAHRIAR AFKHAMI, LINDA CUMMINGS, New Jersey Institute of Technology — An attractive experimental technique, for forming and collecting aggregates of magnetic material at a liquid-air interface by an applied magnetic field, was recently addressed theoretically [Soft Matter, 2013, 9, 8600-8608]. These authors find that, when the magnetic field is weak, the deflection of the liquid-air interface is static, while for sufficiently strong fields, the interface destabilizes and forms a jet. Motivated by this work, here we develop a numerical model for the closely-related problem of solving two-phase Navier-Stokes equations coupled with the static Maxwell equations. We computationally model the magnetically induced interfacial deflection of a magnetic fluid (ferrofluid) and the transition into a jet by a magnetic field gradient from a permanent magnet. We analyze the shape of the liquid-air interface during the deformation stage and the critical magnet distance, for which the static interface transitions into a jet. We draw conclusions on the ability of our numerical model to predict the large interfacial deformation and the consequent jetting, free of any fitting parameter. 1 NSF DMS 1320037, 1211713, 1261596 5:45PM L11.00011 When does aggregation affect magnetic separation? , ALMUT EISENTRAEGER, DOMINIC VELLA, IAN GRIFFITHS, University of Oxford — Magnetic separation is an efficient way to remove magnetic and paramagnetic particles suspended in a carrier fluid, and can be used to remove heavy metals from drinking water. Particles are filtered by moving along the gradient of a strong outer magnetic field towards a collection site. Experimental evidence suggests that aggregation of particles to form chains or clusters plays a vital role in determining the efficiency of separation. In diffusion-dominated systems, aggregation may even be required to induce any collection at all. Modelling approaches so far largely consider aggregation in a uniform outer magnetic field, neglecting collective motion, and hydrodynamic interactions between particles and chains. However, long-range hydrodynamic interactions between particles, which gives rise to the concept of hydrodynamic diffusion, have been considered. Here we combine these ideas to investigate how the average velocity and the relative motion of chains and particles during collection influences chain aggregation rates. A one-dimensional model system provides insight into the relative importance of magnetic and hydrodynamic interactions during aggregation and collection, which may be validated by microfluidic experiments. 5:58PM L11.00012 Three-dimensional microbubble streaming flows , BHARGAV RALLABANDI, Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, ALVARO MARIN, MASSIMILIANO ROSSI, CHRISTIAN KAEHLER, Bundeswehr University Munich, SASCHA HILGENFELDT, Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — Streaming due to acoustically excited bubbles has been used successfully for applications such as size-sorting, trapping and focusing of particles, as well as fluid mixing. Many of these applications involve the precise control of particle trajectories, typically achieved using cylindrical bubbles, which establish planar flows. Using astigmatic particle tracking velocimetry (APTV), we show that, while this two-dimensional picture is a useful description of the flow over short times, a systematic three-dimensional flow structure is evident over long time scales. We demonstrate that this long-time three-dimensional fluid motion can be understood through asymptotic theory, superimposing secondary axial flows (induced by boundary conditions at the device walls) onto the two-dimensional description. This leads to a general framework that describes three-dimensional flows in confined microstreaming systems, guiding the design of applications that profit from minimizing or maximizing these effects. Monday, November 24, 2014 3:35PM - 6:11PM Session L12 Biofluids: Paddling and Jetting — 3018 - Laura Miller, University of North Carolina 3:35PM L12.00001 Hydrodynamics of Peristaltic Propulsion , ATHANASIOS ATHANASSIADIS, DOUGLAS HART, Massachusetts Inst of Tech-MIT — A curious class of animals called salps live in marine environments and self-propel by ejecting vortex rings much like jellyfish and squid. However, unlike other jetting creatures that siphon and eject water from one side of their body, salps produce vortex rings by pumping water through siphons on opposite ends of their hollow cylindrical bodies. In the simplest cases, it seems like some species of salp can successfully move by contracting just two siphons connected by an elastic body. When thought of as a chain of timed contractions, salp propulsion is reminiscent of peristaltic pumping applied to marine locomotion. Inspired by salps, we investigate the hydrodynamics of peristaltic propulsion, focusing on the scaling relationships that determine flow rate, thrust production, and energy usage in a model system. We discuss possible actuation methods for a model peristaltic vehicle, considering both the material and geometrical requirements for such a system. 3:48PM L12.00002 Fluid-Dynamics of Underwater Flight in Sea Butterflies: Analysis using Tomographic PIV , D. ADHIKARI, Georgia Tech, D.W. MURPHY, Johns Hopkins University, D.R. WEBSTER, J. YEN, Georgia Tech — Sea butterflies, Limacina helicina, swim in sea water with a pair of gelatinous “wings” (or parapodia). Their unique propulsion mechanism has been hypothesized to consist of a combination of drag-based propulsion (rowing) and lift-based propulsion (flapping). Drag-based propulsion utilizes maximum drag on the wings during power stroke, followed by minimum drag during recovery stroke. Lift-based propulsion, in contrast, utilizes a pressure difference between the top and bottom of the wings. We present the 3D kinematics of a free-swimming sea butterfly and its induced volumetric velocity field using tomographic PIV. Both upstroke and downstroke motions propel the animal (1 – 3 mm) upward in a sawtooth-like trajectory with average speed of 5 – 15 mm/s (Re = 5 – 45) and roll the calcareous shell forwards-and-backwards at 4 – 5 Hz. The rolling motion effectively positions the wings such that they stroke downward during both the power and recovery strokes, hence inducing upward motion during both phases. A clap-and-fling mechanism is observed at the beginning of the flapping cycle. As the wings come into contact, the velocity of the organism is 2 mm/s. During fling motion, high (unsteady) lift causes the organism velocity to reach 35 mm/s. Separation vortices are observed during the fling motion, and vortices with an opposite sense of rotation form closer to the base of the wing due to the upward translation of the organism. The separation vortices shed into the wake, as the organism translates upward, in the form of separate vortex pairs. 4:01PM L12.00003 Fluid Dynamics of Underwater Flight in Sea Butterflies: Insights from Computational Modeling1 , ZHUOYU ZHOU, RAJAT MITTAL, Johns Hopkins University, JEANNETTE YEN, DONALD WEBSTER, Georgia Institute of Technology — Sea butterflies such as Limacine helicina swim by flapping their wing-like parapodia, in a stroke that exhibits a clap-and-fling type kinematics as well as a strong interaction between the parapodia and the body of the animal at the end of downstroke. We used numerical simulations based on videogrammetric data to examine the fluid dynamics and force generation associated with this swimming motion. The unsteady lift-generating mechanism of clap-and-fling results in a sawtooth trajectory with a characteristic “wobble” in pitch. We employ coupled flow-body-dynamics simulations to model the free-swimming motion of the organism and explore the efficiency of propulsion as well the factors such as shell weight, that affect its sawtooth swimming trajectory. 1 This work is funded by NSF grant 1246317 from the Division of Polar Programs. 4:14PM L12.00004 Kinematics and Fluid Dynamics of Jellyfish Maneuvering , LAURA MILLER, ALEX HOOVER, University of North Carolina — Jellyfish propel themselves through the water through periodic contractions of their elastic bells. Some jellyfish, such as the moon jellyfish Aurelia aurita and the upside down jellyfish Cassiopea xamachana, can perform turns via asymmetric contractions of the bell. The fluid dynamics of jellyfish forward propulsion and turning is explored here by analyzing the contraction kinematics of several species and using flow visualization to quantify the resulting flow fields. The asymmetric contraction and structure of the jellyfish generates asymmetries in the starting and stopping vortices. This creates a diagonal jet and a net torque acting on the jellyfish. Results are compared to immersed boundary simulations 4:27PM L12.00005 Simulation of prolate swimming jellyfish with jet-based locomotion1 , SUNG GOON PARK, BOYOUNG KIM, JIN LEE, KAIST, WEI-XI HUANG, Tsinghua university, HYUNG JIN SUNG, KAIST — The hydrodynamic patterns in the wake of swimming jellyfish are based on the bell morphology. Jellyfish with a prolate bell morphology form a clear jet structure in the wake. A three-dimensional computational model was used to analyze the hydrodynamic patterns. The Froude propulsion efficiency, defined by the ratio of the value of the energy required to deform the elastic bell to the value of the average center velocity multiplied by the thrust, was compared with different forms of the elastic bell deformation. The immersed boundary method was adopted to consider the interaction between the swimming jellyfish and surrounding fluid. Due to the effect of the momentum transferred to the surrounding fluid by the bell deformation, the rotational fluid mass was formed, called vortices. The vortex structures in the wake of prolate swimming jellyfish were elucidated in detail in both quantitative and qualitative ways. A dimensionless temporal parameter was employed to investigate the vortex formation process quantitatively. The starting/stopping vortex structures were generated during the contraction/relaxation phase. During the early stage of the contraction, the vortex structures were mainly generated by the stroke, then the ejected fluid was entrained into the vortex structures. 1 This study was supported by the Creative Research Initiatives (No.2014-001493) program of the National Research Foundation of Korea (MSIP). 4:40PM L12.00006 Muscular Control of Turning and Maneuvering in Jellyfish Bells , ALEXANDER HOOVER, LAURA MILLER, BOYCE GRIFFITH, University of North Carolina at Chapel Hill — Jellyfish represent one of the earliest and simplest examples of swimming by a macroscopic organism. Contractions of an elastic bell that expels water are driven by coronal swimming muscles. The re-expansion of the bell is passively driven by stored elastic energy. A current question in jellyfish propulsion is how the underlying neuromuscular organization of their bell allows for maneuvering. Using an immersed boundary framework, we will examine the mechanics of swimming by incorporating material models that are informed by the musculature present in jellyfish into a model of the elastic jellyfish bell in three dimensions. The fully-coupled fluid structure interaction problem is solved using an adaptive and parallelized version of the immersed boundary method (IBAMR). We then use this model to understand how variability in the muscular activation patterns allows for complicated swimming behavior, such as steering. We will compare the results of the simulations with the actual turning maneuvers of several species of jellyfish. Numerical flow fields will also be compared to those produced by actual jellyfish using particle image velocimetry (PIV). 4:53PM L12.00007 A flow visualization study of single-arm sculling movement emulating cephalopod thrust generation1 , ASIMINA KAZAKIDI, Foundation for Research & Technology - Hellas, Greece, EBENEZER P. GNANAMAN- ICKAM, Embry-Riddle Aeronautical University, USA, DIMITRIS P. TSAKIRIS, Foundation for Research & Technology - Hellas, Greece, JOHN A. EKATERINARIS, Embry-Riddle Aeronautical University, USA — In addition to jet propulsion, octopuses use arm-swimming motion as an effective means of generating bursts of thrust, for hunting, defense, or escape. The individual role of their arms, acting as thrust generators during this motion, is still under investigation, in view of an increasing robotic interest for alternative modes of propulsion, inspired by the octopus. Computational studies have revealed that thrust generation is associated with complex vortical flow patterns in the wake of the moving arm, however further experimental validation is required. Using the hydrogen bubble technique, we studied the flow disturbance around a single octopus-like robotic arm, undergoing two-stroke sculling movements in quiescent fluid. Although simplified, sculling profiles have been found to adequately capture the fundamental kinematics of the octopus arm-swimming behavior. In fact, variation of the sculling parameters alters considerably the generation of forward thrust. Flow visualization revealed the generation of complex vortical structures around both rigid and compliant arms. Increased disturbance was evident near the tip, particularly at the transitional phase between recovery and power strokes. These results are in good qualitative agreement with computational and robotic studies. 1 Work funded by the ESF-GSRT HYDRO-ROB Project PE7(281). 5:06PM L12.00008 Optimality of Metachronal Paddling in Crustacean Swimming , ROBERT GUY, Department of Mathematics, University of California Davis, CALVIN ZHANG, Courant Institute of Mathematical Sciences, New York University, TIMOTHY LEWIS, Department of Mathematics, University of California Davis — Crayfish and other long-tailed crustaceans swim by rhythmically moving four or five pairs of limbs. Despite variations in limb size and stroke frequency, movements of ipsilateral limbs always maintain a tail-to-head metachronal rhythm with an approximate quarter-period inter-limb phase difference. Relatively few studies have examined the fluid dynamics of metachronal limb stroke for the range of Reynolds numbers at which crustaceans operate. Here, we use a computational fluid dynamics model to explore the performance of different paddling rhythms. We show that the natural tail-to-head metachronal rhythm with an approximate quarter-period phase difference is the most effective and efficient rhythm across a wide range of Reynolds numbers. 5:19PM L12.00009 Shrimp theorem: paddle swimming at low Reynolds number , DAISUKE TAKAGI, University of Hawaii at Manoa — A large variety of aquatic organisms, such as small planktonic crustaceans, use multiple legs as paddles; however the resultant dynamics and efficiency of locomotion are not yet clear. I will present a simple model of swimming with multiple pairs of stiff legs. The legs are assumed to oscillate in a metachronal pattern in a model based on slender-body theory for Stokes flow. The model predicts locomotion in the direction of the metachronal wave, as frequently observed in nature. Unlike scallops undergoing reciprocal motion, shrimp can swim at low Reynolds number. This study offers a possible explanation why crustaceans thrive in aquatic environments, and could inspire a new generation of powerful biomimetic robots. 5:32PM L12.00010 Interactions of Copepods with Fractal-Grid Generated Turbulence based on Tomo-PIV and 3D-PTV1 , ZHENGZHONG SUN, DANIEL KRIZAN, ELLEN LONGMIRE, University of Minnesota — A copepod escapes from predation by sensing fluid motion caused by the predator. It is thought that the escape reaction is elicited by a threshold value of the maximum principal strain rate (MPSR) in the flow. The present experimental work attempts to investigate and quantify the MPSR threshold value. In the experiment, copepods interact with turbulence generated by a fractal grid in a recirculating channel. The turbulent flow is measured by time-resolved Tomo-PIV, while the copepod motion is tracked simultaneously through 3D-PTV. Escape reactions are detected based on copepod trajectories and velocity vectors, while the surrounding hydrodynamic information is retrieved from the corresponding location in the 3D instantaneous flow field. Measurements are performed at three locations downstream of the fractal grid, such that various turbulence levels can be achieved. Preliminary results show that the number of escape reactions decreases at locations with reduced turbulence levels, where shorter jump distances and smaller change of swimming orientation are exhibited. Detailed quantitative results of MPSR threshold values and the dynamics of copepod escape will be presented. 1 Supported by NSF-IDBR Grant #0852875. 5:45PM L12.00011 Time-resolved Tomographic PIV Measurements of Water Flea Hopping: Body Size Comparison , A.N. SKIPPER, Georgia Tech, D.W. MURPHY, Johns Hopkins University, D.R. WEBSTER, J. YEN, Georgia Tech — The flow field of the freshwater crustacean Daphnia magna is quantified with time-resolved tomographic PIV. In the current work, we compare body kinematics and flow disturbance between organisms of small (body length = 1.8 mm) versus medium (2.3 mm) versus large (2.65 mm) size. These plankters are equipped with a pair of antennae that are biramous such that the protopodite splits or branches into an exopodite and an endopodite. They beat the antennae pair synchronously to impulsively propel themselves, or ‘hop,’ through the water. The stroke cycle of Daphnia magna is roughly 80 ms in duration and this period is evenly split between the power and recovery strokes. A typical hop carries the daphniid one body length forward and is followed by a period of sinking. Unlike copepod escape motion, no body vortex is observed in front of the animal. Rather, the flow induced by each antennae consists of a viscous vortex ring that demonstrates a slow decay. The time-record of velocity (peak of 40 mm/s for the medium specimen) and hop acceleration (1.8 m/s2 for the medium specimen) are compared, as well as the strength, size, and decay of the induced viscous vortex rings. The viscous vortex ring analysis will be presented in the context of a double Stokeslet model consisting of two impulsively applied point forces separated by the animal width. 5:58PM L12.00012 Fluid Flow Simulation and Energetic Analysis of Anomalocarididae Locomotion , MAXWELL MIKEL-STITES, ANNE STAPLES, Virginia Tech — While an abundance of animal locomotion simulations have been performed modeling the motions of living arthropods and aquatic animals, little quantitative simulation and reconstruction of gait parameters has been done to model the locomotion of extinct animals, many of which bear little physical resemblance to their modern descendants. To that end, this project seeks to analyze potential swimming patterns used by the anomalocaridid family, (specifically Anomalocaris canadensis, a Cambrian Era aquatic predator), and determine the most probable modes of movement. This will serve to either verify or cast into question the current assumed movement patterns and properties of these animals and create a bridge between similar flexible-bodied swimmers and their robotic counterparts. This will be accomplished by particle-based fluid flow simulations of the flow around the fins of the animal, as well as an energy analysis of a variety of sample gaits. The energy analysis will then be compared to the extant information regarding speed/energy use curves in an attempt to determine which modes of swimming were most energy efficient for a given range of speeds. These results will provide a better understanding of how these long-extinct animals moved, possibly allowing an improved understanding of their behavioral patterns, and may also lead to a novel potential platform for bio-inspired underwater autonomous vehicles (UAVs).. Monday, November 24, 2014 3:35PM - 6:11PM Session L13 Drops: Pinch-Off and Coalescence — 3020 - Osman A. Basaran, Purdue University 3:35PM L13.00001 Coalescence of surfactant covered drops , SUMEET THETE, KRISHNARAJ SAMBATH, OSMAN BASARAN, School of Chemical Engineering, Purdue University, West Lafayette, IN-47906 — Drop coalescence plays a central role in industry, e.g. emulsions, sintering, and inkjets, and in nature, e.g. raindrop growth. During coalescence, two drops touch and merge as a liquid neck connecting them grows from microscopic to macroscopic scales. In applications, the drops, while still Newtonian, may be covered with surfactant. Here, we use simulation to analyze the simplest of such problems: two identical drops covered with a monolayer of insoluble surfactant merging in air. The dynamics is analyzed by using as guide the recent work of Paulsen et. al. (2012) who revolutionized the understanding of the coalescence singularity for drops with clean interfaces by demonstrating that the asymptotic regime of coalescence must always involve a balance between inertial, viscous, and capillary forces. These authors summarized their findings by a phase diagram which showed that as coalescence proceeds, the dynamics transitions from this early inertially limited viscous regime to a late time inertial (Stokes) regime when drop viscosity is low (high). Among other things, the talk will highlight how the presence of surfactant modifies the phase diagram obtained by Paulsen et al. 3:48PM L13.00002 Coalescence of Drops of a Power-law Fluid , PRITISH KAMAT, SUMEET THETE, OSMAN BASARAN, Purdue University — Drop coalescence is crucial in a host of industrial, household, and natural processes that involve dispersions. Coalescence is a rate-controlling process in breaking emulsions and strongly influences drop-size-distributions in sprays. In a continuum approach, coalescence begins by the formation of a microscopic, non-slender bridge connecting the two drops. Indefinitely large axial curvature at the neck results in local lowering of pressure that drives fluid from the bulk of the drops toward the neck, thereby causing the bridge radius r(t) and height z(t) to increase in time t. The coalescence of Newtonian drops in air has heretofore been thoroughly studied. Here, we extend these earlier studies by analyzing the coalescence of drops of power-law fluids because many fluids encountered in real applications, including cosmetic creams, shampoos, grease, and paint, exhibit power-law (deformation-rate thinning) rheology. On account of the non-slender geometry of the liquid bridge connecting the two drops (z ≪ r), we analyze the resulting free surface flow problem by numerical simulation. Among other results, we present and discuss the nature of flows and scaling behaviors for r and z as functions of the initial viscosity and power-law index (0 < n ≤ 1). 4:01PM L13.00003 Damped coalescence cascade of a liquid drop , SUIN SHIM, HEEJAE JANG, HOWARD STONE, Princeton Univ — A drop of surfactant (sodium dodecyl sulfate) solution, falling on a bath of the same liquid, shows both normal and damped coalescence cascades while the effects of interfacial tension, viscosity, and gravity are strictly controlled. Unlike the normal coalescence cascade where stages of non-coalescence and partial coalescence alternate until the smallest drop totally coalesces, in the damped coalescence cascade regime, there is no apparent residence time between two successive partial coalescence events. We interpret the results by considering the natural vibration of the drop and the bath surface, and its effects on the drainage of the air film between the drop and the bath. 4:14PM L13.00004 An experimental study of liquid drop - interface coalescence in the presence of surfactants , PANAGIOTA ANGELI, MAXIME CHINAUD, Department of Chemical Engineering, University College London, Torrington Place, London WC1E 7JE, UK, KAI LI, WEI WANG, Beijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum, Beijing, PR China, UNIVERSITY COLLEGE LONDON TEAM, BEIJING KEY LABORATORY OF URBAN OIL AND GAS DISTRIBUTION TECHNOLOGY TEAM — Dropinterface coalescence has been the subject of many studies both theoretical and experimental. It is of particular interest for the oil industries particularly during the transportation of multiphase mixtures where coalescence rates can affect the stability and separation of dispersions. It is well-known that the presence of surfactants can significantly affect the coalescence rates. In this work a silicon oil -water system has been studied in a rectangular coalescence cell. Both rising oil drops and falling water drops coalescing with the water-oil interface have been investigated. A water soluble surfactant, SPAN 80, was used. High speed imaging has been performed to study the coalescence phenomenon and obtain the coalescence time of the drops with the interface with and without the presence of the surfactant. The velocity fields in the bulk fluid and in the liquid film forming between the drop and the interface were studied with shadowgraphy (bright field Particle Image Velocimetry). To increase the spatial resolution particularly in the liquid film microscope lenses were implemented. Results have been compared against existing literature. 4:27PM L13.00005 Simulations of the pinch-off and coalescence of conical droplets , CASEY BARTLETT, Boston University, GUILLAUME GENERO, Boston University/Ecole Polytechnique, JAMES BIRD, Boston University — In the presence of electric fields, pairs of liquid drops can be rapidly drawn together such that, at contact, the deformed interface resembles a double-cone. Following contact, these drop pairs are observed to either coalesce or recoil. Here we use high-resolution numerical simulations to highlight the impact of the initial double-cone angle on drop coalescence. The results demonstrate a self-similar behavior at intermediate scales for both coalescence and recoil that is independent of the other length-scales in the problem and is consistent with previous experiments. 4:40PM L13.00006 Coalescence of liquid drops is governed by surface tension driven capillary pressure in the inertial regime , MD MAHMUDUR RAHMAN, Univ of Nebraska - Lincoln — Droplet coalescence is a complex hydrodynamic phenomenon where it has been thought that, at the moment of contact, a singularity occurs due to the inversion of one of its two radii of curvature. However, the effect of this singularity cannot be observed experimentally through coalescence of liquid drops in three dimensions. A recent study examined coalescence mathematically in the inertial regime where no such singularity is assumed. After coalescence happens in the “viscous” regime, hydrodynamic scaling starts working with a certain bridge radius other than a singular point. In our experimental analysis, we studied coalescence in a confined geometry where drops were sandwiched in the Hele-Shaw cell. We observed very well defined mixing interface which signifies that early coalescence (‘viscous’ regime) is not a hydrodynamic phenomenon, rather its characteristics may be evaluated from molecular dynamics analysis. Our experiment will be helpful in studying coalescence of liquid drops in any given shape through mathematical modeling where initial bridge radius can be assumed or determined through other means. 4:53PM L13.00007 Coalescence of drops in the Hele-Shaw geometry , JOHN DAVIDSON, DONGHEE LEE, MD. MAHMUDUR RAHMAN, SANGJIN RYU, University of Nebraska-Lincoln — Coalescence of drops is an interfacial fluid dynamics phenomenon commonly occurring in engineering devices as well as in nature. The physics of such coalescence has been studied mainly for three-dimensional (spherical drops) and semi-infinite two-dimensional (hemispherical drops) geometries. In contrast, the coalescence of drops in the pseudo two-dimensional geometries has been less documented. To investigate the coalescence of two drops of the same liquid in the Hele-Shaw geometry, we observed with the high-speed videography the neck growth of the two coalescing drops squeezed between two parallel hydrophobic surfaces. We also studied the effects of differences in liquid properties on the scaling behavior of coalescence. 5:06PM L13.00008 Coalescence of Drops Due to a Constant Force Interaction in a Viscous Quiescent Fluid , JOHN FROSTAD, University of California, Santa Barbara, ALEXANDRA PAUL, Saarland University, GARY LEAL, University of California, Santa Barbara — A Cantilevered-Capillary Force Apparatus is used to study the time scale for the coalescence of two droplets compressed together with a constant force. Power-law trends for the coalescence time as a function of droplet radius and compression force are measured. The measurements are compared against several different scaling theories from the literature. One of the existing theories is found to correctly predict the dependence on the droplet radius, but all of the theories over-predict the dependence on the force. A transition is also measured in the coalescence process from a predominately deterministic to a predominately stochastic process. A qualitative explanation for this transition is provided via scaling arguments. 5:19PM L13.00009 Coalescence of Liquid Drops: Modelling, Computation and Scaling , JAMES SPRITTLES, University of Warwick, YULII SHIKHMURZAEV, University of Birmingham — The coalescence of two liquid drops surrounded by a viscous gas is simulated by a computational approach which resolves the unprecedentedly small spatio-temporal scales which have recently been accessed experimentally. A systematic study of the parameter space of practical interest allows the influence of the governing parameters in the system to be identified and the role of viscous gas to be determined. In particular, it is shown that the viscosity of the gas suppresses the formation of toroidal bubble predicted in some cases when the gas’ dynamics are neglected. Considering the entire parameter space allows us to examine the accuracy of various “scaling laws” proposed for different “regimes” and, in doing so, to (a) reveal certain inconsistencies in recent works and (b) develop a new scaling law for the inertial regime which captures experimental data from the literature remarkably well. Notably, the conventional model is shown to reproduce many qualitative features of the initial stages of coalescence observed experimentally, such as a collapse of calculations onto a “master curve” but, quantitatively, overpredicts the speed of coalescence. Finally, a phase diagram of parameter space, differing from previously published ones, is used to illustrate the key findings. 5:32PM L13.00010 Inverted Break-up Behaviour in Continuous Inkjet (CIJ) Printing1 , CLAIRE MCILROY, OLIVER HARLEN, NEIL MORRISON, Univ of Leeds — Although droplet creation during continuous jetting of Newtonian fluids has been widely studied, unsolved problems surrounding the break-up dynamics remain. Jetting through a nozzle creates a stream of liquid that is rendered unstable by surface tension. This instability creates a succession of main drops connected by thin filaments, with drop separation determined by the fastest growing wavelength. In order to control break-up and increase printing speeds, continuous inkjet (CIJ) printing exploits the effects of finite amplitude modulations in the jet velocity profile giving conditions where jet stability deviates from the usual Rayleigh behaviour. To explore these non-linear effects, we have developed a one-dimensional jetting model. In particular, we identify a modulation range for which pinching occurs upstream of the connecting filament, rather than downstream – a phenomenon we call “inverted” break-up. Furthermore, this behaviour can be controlled by the addition of harmonics to the initial driving signal. Our results are compared to full axisymmetric simulations in order to incorporate the effects of nozzle geometry. 1 EPSRC Innovation in Industrial Technology 5:45PM L13.00011 Dynamic behaviors of oppositely charged emulsion droplets , ZHOU LIU, The University of Hong Kong, HANS M. WYSS, Eindhoven University of Technology, ALBERTO FERNANDEZ-NIEVES, Georgia Institute of Technology, HO CHEUNG SHUM, The University of Hong Kong — In this work, we investigate the dynamic behaviors of two oppositely charged emulsion droplets. Using an electrocapillary number and separation distance between droplets, we have characterized three types of droplet behaviors in electric field. Besides the common seen coalescence, two qualitatively different dynamic behaviors are identified: fuse-and-split and periodic non-coalescence. In fuse-and-split regime, the droplets fuse into a jet, which subsequently breaks up into two droplets. In periodic non-coalescence regime, the droplets contact and bounce away periodically without coalescence. Further analysis indicates that the applied voltage always decreases dramatically upon droplets’ contact due to spikes of discharging current. Thus, the electric field strength drops and surface tension quickly dominates over electric stress upon droplet’s contact. By analyzing capillary instability, all the observed dynamic states can be attributed to the different initial shapes of dumbbell-like jet formed upon droplets’ contact. By controlling the initial separation distance between droplets, the shapes of the jet and thus the resultant dynamics can be precisely manipulated. 5:58PM L13.00012 Multiscale computations of thin films between colliding drops , BAHMAN ABOULHASANZADEH, University of Notre Dame, SADEGH DABIRI, Purdue University, GRETAR TRYGGVASON, University of Notre Dame — In multiphase flows thin films frequently appear between fluid blobs colliding with each other. These films can become very thin and be difficult to resolve accurately in numerical simulations, particularly in DNS of many co-flowing drops, requiring very fine resolution and resulting in excessive computational cost due to very fine uniform grids or time consuming adaptive mesh refinement. Here, we describe an algorithm for detecting thin films using a front tracking method. We propose a subscale model to describe the physics and the evolution of a thin film between two drops. We also demonstrate the importance of correctly reconstructing the viscosity field on getting a grid independent solution. Comparison between results for a fully resolved film on a fine grid and simulations using a much coarser grid plus the model for the description of the film, show good agreement. This study was funded by NSF Grant CBET-1132410. Monday, November 24, 2014 3:35PM - 5:45PM — Session L15 Minisymposium II: Honoring the Memory of Professor Howard Brenner 3022/3024 - Anthony Davis, University of Californa, San Diego 3:35PM L15.00001 Howard Brenner: Visionary researcher, profound scholar and close friend , ANDREAS ACRIVOS, CUNY-CCNY and Stanford University — During his long (60+ years) professional career, Howard Brenner made an astonishingly large number of seminal contributions in a variety of subjects such as, to name only a small fraction: particle motions in very viscous fluids, the mechanics of complex fluids, multiphase flow in porous media, emulsion rheology and many others. In my talk I shall focus on a few of his early publications in “low-Reynolds number fluid mechanics” which helped transform that subject from one that was originally viewed as being of only academic interest (and, therefore, “very dull and of no practical value whatsoever”) into the presently exciting and active field of “micro-fluidics.” 4:01PM L15.00002 Brenner’s bi-velocity fluid mechanics and gradient effects in general continua , JOE GODDARD, University of California, San Diego — The field of bi-velocity fluid mechanics represents a notable contribution to the impressive scientific legacy of Professor Howard Brenner. Dating approximately from 2004, Brenner authored or co-authored some thirty papers concerned with the possible breakdown of the Navier Stokes/Fourier models of momentum and heat transport in fluids, a body of work often cited in the literature on statistical and continuum mechanics. Central to the theory is the notion that the barycentric velocity, which represents inertial terms in the Navier Stokes equations, differs generally from the velocity that represents viscous stress, denoted variously by Brenner as “volume” or “work” velocity. The present paper, based heavily on a previous publication by the present author (Int. J. Eng. Sci. 48 1279-88, 2010), shows that, while the work of Brenner poses a challenge to certain continuum mechanical notions of material points and velocities, it is also subsumed in a more general framework of higher-gradient models of continuous media. Within this framework, linear constitutive models represent weak non-locality as an expansion in spatial wave number or Knudsen number. Some comments are also offered on Brenner’s concept of a non-material volume velocity. 4:27PM L15.00003 Getting Something For Nothing , HOWARD A. STONE, Department of Mechanical and Aerospace Engineering, Princeton University — The Reciprocal Theorem of low-Reynolds-number hydrodynamics is a useful tool for (i) understanding some of the theoretical structures that underlie the subject and (ii) for using perturbation expansions to solve various flow problems when only an integrated quantity such as a force or a pressure drop are required. The latter applications often give the impression that you are getting something for nothing. We will highlight several uses of these ideas by Howard Brenner and then give examples of recent applications of the Reciprocal Theorem, including two examples where Marangoni stresses are important. 4:53PM L15.00004 Howard Brenner’s Legacy for Biological Transport Processes , JOHANNES NITSCHE1 , Department of Chemical and Biological Engineering, University at Buffalo, the State University of New York — This talk discusses the manner in which Howard Brenner’s theoretical contributions have had, and long will have, strong and direct impact on the understanding of transport processes occurring in biological systems. His early work on low Reynolds number resistance/mobility coefficients of arbitrarily shaped particles, and particles near walls and in pores, is an essential component of models of hindered diffusion through many types of membranes and tissues, and convective transport in microfluidic diagnostic systems. His seminal contributions to macrotransport (coarse-graining, homogenization) theory presaged the growing discipline of multiscale modeling. For biological systems they represent the key to infusing diffusion models of a wide variety of tissues with a sound basis in their microscopic structure and properties, often over a hierarchy of scales. Both scientific currents are illustrated within the concrete context of diffusion models of drug/chemical diffusion through the skin. This area of theory, which is key to transdermal drug development and risk assessment of chemical exposure, has benefitted very directly from Brenner’s contributions. In this as in other areas, Brenner’s physicochemical insight, mathematical virtuosity, drive for fully justified analysis free of ad hoc assumptions, quest for generality, and impeccable exposition, have consistently elevated the level of theoretical understanding and presentation. We close with anecdotes showing how his personal qualities and warmth helped to impart high standards of rigor to generations of grateful research students. 1 Authors are Johannes M. Nitsche, Ludwig C. Nitsche and Gerald B. Kasting. 5:19PM L15.00005 Oscillatory Counter-Centrifugation: Effects of History and Lift Forces , ALI NADIM, Claremont Graduate University — This work is co-authored with my doctoral student Shujing Xu and is dedicated to the memory of my doctoral advisor Howard Brenner who enjoyed thought experiments related to rotating systems. Oscillatory Counter-Centrifugation refers to our theoretical discovery that within a liquid-filled container that rotates in an oscillatory manner about a fixed axis as a rigid body, a suspended particle can be made to migrate on average in the direction opposite to that of ordinary centrifugation. That is, a heavy (or light) particle can move toward (or away from) the rotation axis, when the frequency of oscillations is high enough. In this work we analyze the effects of the Basset history force and the Saffman lift force on particle trajectories and find that the counter-centrifugation phenomenon persists even when these forces are active. Monday, November 24, 2014 3:35PM - 6:11PM Session L16 Waves I: Surface Waves — 2000 - Francesco Fedele, Georgia Institute of Technology 3:35PM L16.00001 Quasi-periodic water wave dynamics , JON WILKENING, University of California, Berkeley — We present a framework for computing quasi-periodic solutions of the free-surface Euler equations with spectral accuracy. Some of the new solutions are hybrid traveling-standing waves that return to a spatial translation of their initial condition at a later time. Others are nonlinear superpositions of several standing waves with irrationally related periods. We also present a Floquet analysis of the stability of pure standing waves. When they are stable, generic perturbations appear to yield quasi-periodic solutions that remain close to nearby pure standing waves. 3:48PM L16.00002 Lensing of Oceanic Gravity Waves: Theory and Experiment , MOHAMMAD-REZA ALAM, RYAN BLAKE ELANDT, MOSTAFA SHAKERI, Univ of California - Berkeley — In this talk we show that small features embedded to the seafloor can result in a lensing effect for overpassing oceanic surface waves, similar to how glass lenses focus or defocus light. These seafloor features are typically in the shape of curved periodic sandbars, and the effect is a result of a nonlinear interaction between surface waves and seabed undulations which is known as “Bragg Resonance.” We further show that for a broadband incident wave spectrum (i.e. a wave group composed of multitude of different-frequency waves) a polychromatic topography (occupying no more than the area required for a monochromatic lens) can achieve a broadband lensing effect. Gravity wave lenses can be utilized to create localized high-energy wave zones (e.g. for wave energy harvesting or creating artificial surf zones) as well as to disperse waves in order to create protected areas (e.g. harbors or areas near important offshore facilities). In reverse, lensing of oceanic waves may be caused by natural seabed features and may explain the frequent appearance of very high amplitude waves at certain bodies of water. 4:01PM L16.00003 Perfect Broadband Cloaking of shallow water Waves via Nonlinear Medium Transformation , AHMAD ZAREEI, M.-REZA ALAM, UC Berkeley — The major obstacle in achieving a perfect cloaking for shallow water waves is that the linear transformation media scheme (aka transformation optics) requires variations of two independent medium properties. These two medium properties for the case of electromagnetic waves are permittivity and permeability. Designing a medium with a variable permittivity and permeability is difficult to achieve. For gravity waves, the two required spatially variable properties are the water depth and the gravity acceleration, but here changing of the gravity acceleration is simply impossible. Here we present a nonlinear transformation that only requires the change in one of the medium properties, i.e., in the case of shallow water waves just the water depth, and hence enables us to design a perfect cloak for long gravity waves. We show that with this nonlinear transformation an object can be cloaked for any wave satisfying merely the shallow water condition. The presented transformation can as well be applied for the design of non-magnetic optical cloak for electromagnetic waves. 4:14PM L16.00004 Ocean crest slowdown and geometric phases , FRANCESCO FEDELE, Georgia Institute of Technology — Several studies over the past two decades suggest that the initial speed of breaking crests of dominant open ocean wave groups, or breaker speeds, are typically 20% lower than expected from linear wave theory (Rapp & Melville, 1990). A recent multifaceted study in Banner et al. (PRL, 2014) explains the reduced breaker speed by means of the crest slowdown, a new fundamental property of non-breaking ocean waves as they occur naturally, not as uniform wavetrains, but within evolving groups. Before the focusing point, the crest of the largest wave in the group slows down as it advances leaning forward, and it becomes symmetrical as the maximum height is approached. As the wave decays after focus, the crest accelerates as it leans backward. In this talk, I will show that the crest slowdown and the associated forward/backward leaning are generic features of each crest of water wave groups. They are associated with the energy convergence in the neighborhood of the focal region, irrespective of whether the wave evolves to break or not (Fedele, JFM 2014). In particular, I will show that the crest slowdown is induced by the natural dispersion of unsteady wave groups. Drawing from quantum mechanics and differential geometry, it can be explained in terms of geometric phases associated with the wave motion with U(1) group symmetry (e.g. Berry 1984). The theoretical findings are in fair agreement with ocean field observations off the Venice coast, Italy, obtained by state-of-the-art stereo imaging techniques. 4:27PM L16.00005 Wave Impact on a Wall: Comparison of Experiments with Similarity Solutions1 , A. WANG, J.H. DUNCAN, D.P. LATHROP, University of Maryland — The impact of a steep water wave on a fixed partially submerged cube is studied with experiments and theory. The temporal evolution of the water surface profile upstream of the front face of the cube in its center plane is measured with a cinematic laser-induced fluorescence technique using frame rates up to 4,500 Hz. For a small range of cube positions, the surface profiles are found to form a nearly circular arc with upward curvature between the front face of the cube and a point just downstream of the wave crest. As the crest approaches the cube, the effective radius of this portion of the profile decreases rapidly. At the same time, the portion of the profile that is upstream of the crest approaches a straight line with a downward slope of about 15◦ . As the wave impact continues, the circular arc shrinks to zero radius with very high acceleration and a sudden transition to a high-speed vertical jet occurs. This flow singularity is modeled with a power-law scaling in time, which is used to create a time-independent system of equations of motion. The scaled governing equations are solved numerically and the similarly scaled measured free surface shapes, are favorably compared with the solutions. 1 The support of the Office of Naval Research is gratefully acknowledged. 4:40PM L16.00006 Surface tension effects in wave breaking , LUC DEIKE, W.K. MELVILLE, Scripps Insitution of Oceanography, University of California San Diego, STEPHANE POPINET, Institut d’Alembert, Univeriste Pierre et Marie Curie, Paris — We present a numerical study of wave breaking by solving the full Navier-Stokes equations for two-phase air-water flows using the solver Gerris [1]. We describe a parametric study of the influence of capillary effects on wave breaking using two-dimensional simulations. The onset of wave breaking as a function of the Bond number, Bo, and the initial wave steepness S is determined and a phase diagram in terms of (S,Bo) is presented that distinguishes between non-breaking gravity waves, parasitic capillaries on a gravity wave, spilling breakers and plunging breakers. The wave energy dissipation is computed for each wave regime and is found to be in good agreement with experimental results for breaking waves. Moreover, the enhanced dissipation just by parasitic capillaries is comparable to the dissipation due to breaking [2]. Extending the simulations to three dimensions permits studies of the generation and statistics of bubbles and spray during breaking. [1] Popinet, S. 2003. Journal of Computational Physics 190, 572–600. Popinet, S. 2009. Journal of Computational Physics 228, 5838–5866. [2] Deike, L., Popinet, S., and Melville, W.K. Submitted to Journal of Fluid Mechanics (June 2014). 4:53PM L16.00007 The virial theorem for water waves and its application to deep-water wave breaking , NICHOLAS PIZZO, W. KEN MELVILLE, Scripps Institution of Oceanography, University of California, San Diego — The connection between the geometry, kinematics and dynamics of steep and breaking waves is crucial for an improved understanding of air-sea interaction processes. In this study, we present a virial theorem for deep-water surface gravity waves, related to a conserved integral quantity originally derived by Benjamin and Olver (1982), and we apply this theorem to the study of properties of steep and breaking waves. Specifically, we relate the geometry and dynamics of these wave scenarios in an attempt to better understand the breakdown of equipartition between the kinetic and potential energy. The virial theorem will be studied both analytically and numerically, where in the latter case we make use of a variational description of water waves in a conformally mapped reference frame (Balk 1996) that we have developed for use in a numerical model. Particular attention will be given to the application of these findings to recent theoretical and laboratory studies in which it has been shown that the potential energy available to breaking waves plays a crucial role in setting the scales of post-breaking phenomena; for example, the breaking induced energy dissipation rate (Drazen et al. 2008) and the circulation generated by breaking (Pizzo and Melville 2013). 5:06PM L16.00008 Experimental study of breaking and energy dissipation in surface waves1 , GERARDO RUIZ CHAVARRIA, Facultad de Ciencias, Universidad Nacional Autonoma de Mexico, PATRICE LE GAL, MICHAEL LE BARS, IRPHE, UMR 7342, CNRS, Aix-Marseille University — We present an experimental study of the evolution of monochromatic waves produced by a parabolic wave maker. Because of the parabolic shape of the wave front, the waves exhibit spatial focusing and their amplitude dramatically increases over distances of a few wavelengths. Unlike linear waves, the amplitude of the free surface deformation cannot exceed a certain threshold and when this happens the waves break. In order to give a criterion for the appearance of breaking, we calculate the steepness defined as ε =H/λ (where H is the wave height and λ their wavelength) for waves of frequencies in the range 4-10 Hz. We found that wave breaking develops when ε attains approximately a value of 0.10. We also evaluate the lost of energy carried by the waves during their breaking by a detailed and accurate measurement of their amplitude using an optical Fourier transform profilometry. 1 G. Ruiz Chavarria acknowledges DGAPA-UNAM by support under project IN 116312 (Vorticidad y ondas no lineales en fluidos) 5:19PM L16.00009 Subharmonic waves produced by oscillating submerged solids , JOSE M. PEREZGRACIA, ETSI Aeronauticos, Universidad Politecnica de Madrid, FERNANDO VARAS, EI Telecomunicacion, Universidad de Vigo, JOSE M. VEGA, ETSI Aeronauticos, Universidad Politecnica de Madrid — Parametric excitation of subharmonic waves in a container due to the vertical oscillation of a (deeply) submerged solid is considered in this presentation. In general, two parametric forcing mechanisms will appear in this configuration, namely forcing from (directly excited) surface waves and forcing from an oscillatory flow in the bulk. Nevertheless, if the (oscillating) obstacle is submerged deeply enough (as it will be assumed) the second mechanism will dominate. This problem can then be seen as a generalization of the (classical) Faraday waves problem with a non-homogeneous forcing (associated to the oscillating flow generated near the cylinder). In fact, this problem corresponds (in the case of a cylinder with a proper symmetry) to the simplest case of symmetric non-homogeneous forcing of subharmonic waves, and it can be considered as the counterpart of horizontal vibration of containers (where an antisymmetric non-homogeneous parametric forcing is found). The analysis recently developed by the authors in the case of a horizontally vibrated container (Journal of Fluid Mechanics, vol. 739 pp. 196-228, 2014) is adapted here in order to obtain predictions of threshold vibration amplitudes, pattern orientation and periodic or quasi-periodic nature of subharmonic waves. 5:32PM L16.00010 Pressure Stagnation Line on a Planing Hull in Calm Water1 , CHRISTINE IKEDA2 , CAROLYN JUDGE, United States Naval Academy — High-speed planing boats are subjected to repeat impacts due to slamming, which can cause structural damage and discomfort or injury to passengers. An experimental study aimed at understanding and predicting the physics of a planing craft re-entering the water after becoming partially airborne was conducted. A subset of this experiment includes calm water analysis to gain an understanding of the pressure stagnation line and its correlation with the wetted surface on the planning craft in calm water conditions. A planing hull model was towed in a 116-m long, 8-m wide tow-tank with a water depth of 5 m. Hull models at 1/10 and 1/4 of full-scale were examined. These models, only free to move in heave and pitch, were instrumented to measure dynamic pressures with point-pressure sensors at 12 locations near the LCG (longitudinal center of gravity) and transom as well as a highly spatially resolved pressure mapping system. These pressure measurements were sampled at rates up to 20kHz. Using these pressure measurements along with underwater photos of the wetted surface allowed for the v-shaped wetted line and stagnation line to be measured. Preliminary results show that the peak pressures occur before the wetted line and that atmospheric pressure is reached at the transom. 1 Supported 2 Currently by the Office of Naval Research. at the University of New Orleans 5:45PM L16.00011 Kelvin ship waves: the effect of nonlinearity on the apparent wake angle1 , SCOTT MCCUE, RAVINDRA PETHIYAGODA, TIMOTHY MORONEY, Queensland University of Technology — We learn as undergraduates that the halfangle which encloses a Kelvin ship wave pattern is simply arcsin(1/3) ≈ 19.47◦ , provided the fluid is deep and the disturbance is small. But observations and calculations for sufficiently fast-moving ships suggest that the apparent wake angle decreases with ship speed. One explanation of this phenomenon is that the wave pattern that is observed in practice is defined by the location of the highest peaks; for wakes created by sufficiently fast-moving objects, these highest peaks no longer lie on the outermost divergent waves, resulting in a smaller apparent angle. We explore these ideas by analysing the linearised problems of flow past a submerged point source (semi-infinite Rankine body) or past a submerged source doublet (sphere). Then we consider the nonlinear versions of these problems. One result is that nonlinearity has the effect of increasing the apparent wake angle so that some highly nonlinear solutions have apparent wake angles that are greater than Kelvin’s angle. 1 We acknowledge support of the Australian Research Council via the Discovery Project DP140100933 5:58PM L16.00012 Reconstruction of a energy wave spectrum using a non-intrusive technique , DIANA VARGAS, ADOLFO LUGO, EDGAR MENDOZA, RODOLFO SILVA, Universidad Nacional Autonoma de Mexico — For studies taken in a wave flume, it is frequent to use wave gauges to measure directly the free surface fluctuations. Sometimes these gauges can interfere the measures because this probes act as obstacles to water. Therefore we designed a non intrusive technique using a bubble curtain. In this work we pretend to reconstruct the energy wave spectrum of regular and irregular waves, generated in a wave flume, assuming linear and non linear wave theory by analyzing the time series of the bubbles velocity field given with the aid of PIV. Monday, November 24, 2014 3:35PM - 6:11PM Session L17 Geophysical Fluid Dynamics: Gravity Currents — 2002 - Claudia Cenedese, Woods Hole Oceano- graphic Institution 3:35PM L17.00001 Entrainment in a density-driven current flowing down a rough slope in a rotating fluid1 , CLAUDIA CENEDESE, Woods Hole Oceanographic Institution, LUISA OTTOLENGHI, CLAUDIA ADDUCE, University of Roma Tre — Dense oceanic overflows mix with surrounding waters along the descent down the continental slope. The amount of entrainment dictates the final properties of these overflows, and thus is of fundamental importance to the understanding of the formation of deep water masses. We will discuss laboratory experiments investigating the influence of bottom roughness on entrainment in a dense current flowing down a sloping bottom in a rotating homogeneous fluid. The bottom roughness has been idealized by an array of cylinders. Both spacing (sparse vs. dense configuration) and height of the roughness elements compared with the height of the current have been varied. The presence of roughness elements has been observed to enhance entrainment for low values of the Froude number (Fr ). This suggests that if a dense current is vigorously entraining via shear-induced entrainment at the interface between the dense and ambient fluids (i.e. large Fr ) the additional entrainment occurring via the turbulence generated by roughness elements at the bottom boundary is negligible. However, for low Fr, when the entrainment at the interface between the dense and ambient fluids is low, the additional entrainment due to bottom roughness elements dominates. As in the case of a smooth bottom, we observed a strong dependence of the entrainment on the Reynolds number. Furthermore, density measurements indicate that stratification within the dense current is enhanced when the roughness elements occupy a large portion of the current, especially for the dense roughness configuration. 1 Support was given by NSF project OCE-1333174. 3:48PM L17.00002 Experimental and Numerical Studies of Oceanic Overflow1 , THOMAS GIBSON, Baylor University, FRED HOHMAN, University of Georgia, THERESA MORRISON, San Diego State University, SHANON RECKINGER, Fairfield University, SCOTT RECKINGER, Brown University — Oceanic overflows occur when dense water flows down a continental slope into less dense ambient water. The resulting density driven plumes occur naturally in various regions of the global ocean and affect the large-scale circulation. General circulation models currently rely on parameterizations for representing dense overflows due to resolution restrictions. The work presented here involves a direct qualitative and quantitative comparison between physical laboratory experiments and lab-scale numerical simulations. Laboratory experiments are conducted using a rotating square tank customized for idealized overflow and a high-resolution camera mounted on the table in the rotating reference frame for data collection. Corresponding numerical simulations are performed using the MIT general circulation model (MITgcm) run in the non-hydrostatic configuration. Resolution and numerical parameter studies are presented to ensure accuracy of the simulation. Laboratory and computational experiments are compared across a wide range of physical parameters, including Coriolis parameter, inflow density anomaly, and dense inflow volumetric flow rate. The results are analyzed using various calculated metrics, such as the plume velocity. 1 Funding for this project is provided by the National Science Foundation. 4:01PM L17.00003 Sensitivity of resolution and vertical grid types on 3D overflow simulations using mpas-ocean , SHANON RECKINGER, Fairfield University, MARK PETERSEN, Los Alamos National Laboratory, SCOTT RECKINGER, Brown University — The Model for Prediction Across Scales (MPAS) is a climate model framework that supports unstructured, variable resolution grids. Since a primary issue in ocean modeling is the treatment of the vertical coordinate, MPAS-Ocean has been developed to allow for a variety of options in the vertical coordinate choice. The representation of overflows has been shown to be difficult at horizontal resolutions coarser than a few kilometers. Therefore, the combination of the unstructured horizontal grid and the variety of vertical grid choices available with MPAS-Ocean provides a unique approach. MPAS-Ocean is used to simulate an idealized density driven overflow using the dynamics of overflow mixing and entrainment (DOME) setup. Numerical simulations are carried out at a variety of resolutions to compare the accuracy and computational cost of increasing the vertical versus the horizontal resolution. Additionally, various vertical grid types are studied including z-level, z-level with partial bottom cells, and sigma coordinates. Entrainment and transport metrics are calculated and analyzed in order to compare the results from the various grid setups. 4:14PM L17.00004 Mixing Induced by Slope and Valley Flow Collisions in Complex Terrain1 , H.J.S. FERNANDO, C. HOCUT, Q. ZHONG, Univ of Notre Dame, MATERHORN TEAM — Collision of slope and valley flows at night in complex terrain air basins lead to powerful, recurring turbulence generating events. The contributions of these collisions to turbulent mixing in complex terrain basins has been studied using data taken during the field experiments of Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) Program as well as laboratory measurements conducted under controlled conditions using counter flowing gravity currents in which detailed turbulence observations were made using LDV/PLIF. These collisions cause localized instabilities, which, together with turbulence generated by impingement of fronts on one another generate a turbulence field that decay rapidly under local stable stratification. Buoyancy fluxes measured during laboratory experiments are parameterized using suitable dimensionless parameters that characterize the nature of gravity currents. The laboratory results are compared with field measurments. 1 Funded by ONR grant # N00014-11-1-0709. 4:27PM L17.00005 Lateral spreading in a steady turbulent density current from an isolated source , ANDREW WELLS, JOSH VIVIAN, University of Oxford — Turbulent buoyancy-driven flows on slopes occur in a range of environmental settings, such as dense ocean overflows, atmospheric katabatic winds, meltwater flows under ice shelves, or discharge of industrial effluents. A convenient modelling approach for dense currents from isolated sources considers a so-called “streamtube approximation,” averaging over the cross-section of the current to yield an effectively one-dimensional model for the evolution of flow along a streamline. However, such modelling approaches typically parameterise any changes in current width, rather than directly predicting the dynamics of lateral spreading. To build insight into the relevant dynamics, we consider steady density currents flowing down a planar slope, supplied by a continuous buoyancy flux from an isolated source. A model is developed to describe the downslope evolution of flow averaged over the width and depth of the current, including a new dynamical treatment of lateral spreading. We analyse theoretical and numerical solutions, before comparing to laboratory experiments with a dense saline current flowing down a slope. 4:40PM L17.00006 Stratification established by peeling detrainment from gravity currents: laboratory experiments and models , CHARLIE HOGG, STUART DALZIEL, HERBERT HUPPERT, Cambridge University, JORG IMBERGER, Centre for Water Research, DEPARTMENT OF APPLIED MATHEMATICS AND THEORETICAL PHYSICS TEAM, CENTRE FOR WATER RESEARCH TEAM — Dense gravity currents feed fluid into confined basins in lakes, the oceans and many industrial applications. Existing models of the circulation and mixing in such basins are often based on the currents entraining ambient fluid. However, recent observations have suggested that uni-directional entrainment into a gravity current does not fully describe the mixing in such currents. Laboratory experiments were carried out which visualised peeling detrainment from the gravity current occurring when the ambient fluid was stratified. A theoretical model of the observed peeling detrainment was developed to predict the stratification in the basin. This new model gives a better approximation of the stratification observed in the experiments than the pre-existing entraining model. The model can now be developed such that it integrates into operational models of lakes. 4:53PM L17.00007 Gravity currents penetrating into a sheared and stratified ambient fluid , MOHAMAD M. NASR-AZADANI, AMIN KHODKAR, ECKART MEIBURG, University of California Santa Barbara — We have developed a circulation-based theoretical model to study gravity currents penetrating into a sheared and stratified ambient fluid. Unlike previous theories, our circulation model, which employs the vorticity equation, does not require any assumptions regarding headloss along a specific streamline in the flow. Our theoretical framework enables us to identify the existence of gravity currents penetrating into ambient environments having arbitrary velocity and/or density profiles across the channel height by means of a headloss analysis. First, we investigated a two-layered free stream configuration, where we observed excellent agreement between our theoretical model and DNS results. For various shear magnitudes, we demonstrated the existence of gravity currents of more than half of the channel height, which are not physically possible in the classical gravity current setup without any shear. Furthermore, we investigated the influence of stratification on the behavior of gravity currents. We identified the regions of physically feasible solutions. We also observed situations that produced internal waves and/or rarefaction waves. 5:06PM L17.00008 Gravity current flow over topography with a two-layer stratified ambient1 , MITCH NICHOLSON, MORRIS FLYNN, U.Alberta — We report upon a laboratory experimental study of dense lock-released gravity currents propagating through a two-layer ambient and over sinusoidal topography of amplitude A and wavelength λ or 2λ. Particular emphasis is placed on a Boussinesq flow regime and the initial (or “slumping”) stage of motion. Because of the presence of the topography, the height of both the lower layer and the channel varies in the downstream direction. In contrast to the flat-bottom case, the gravity current front therefore accelerates and decelerates as it respectively flows up- and downhill. Overall, the topography has a retarding effect on the average front speed, Uavg , whose variation with A, the layer densities and the interface height is described. The topography also alters the structure of the gravity current head by inducing large-scale vortices in regions characterized by a substantial shear flow. As in the flat-bottom case, the forward advance of the gravity current can excite a downstream-propagating interfacial wave. We identify the parametric region corresponding to wave generation. 1 funding by NSERC 5:19PM L17.00009 Gravity currents in non-rectangular cross-area channels with stratified ambient , MARIUS UNGARISH, Israel Inst of Tech — The propagation of a high-Reynolds-number gravity current (GC) in a horizontal channel along the horizontal coordinate x is considered. The current is of constant density, ρc , and the ambient has a linear stable stratification, from ρb at the bottom z = 0 to ρo at z = H. The cross-section of the channel is given by the general −f1 (z) ≤ y ≤ f2 (z) for 0 ≤ z ≤ H. A shallow-water model is developed for the solution of a GC of fixed volume released from a lock on the bottom (ρc ≥ ρb ). The dependent variables are the position of the interface, h(x, t), and the speed (area-averaged), u(x, t), where t is time. The cross-section geometry enters the formulation via the width of the channel f (z) = f1 (z) + f2 (z). For a given f (z), the free input parameters of are the height ratio H/h0 of ambient to lock and the stratification parameter S = (ρb − ρo )/(ρc − ρo ). The equations of motion are a hyperbolic PDE system. The initial motion displays a “slumping” stage with constant speed, calculated analytically. An analytical solution for the long-time self-similar propagation is also available for special cases. The model is a significant generalization of the rectangular-channel analysis. 5:32PM L17.00010 Gravity Currents Propagating Up a Slope in a Uniform Ambient and a Two-Layer Stratified Ambient Depth , LARISSA MARLEAU, MORRIS FLYNN, BRUCE SUTHERLAND, Univ of Alberta — Bottom propagating gravity currents resulting from full- and partial-depth lock-release experiments are investigated as they propagate up a slope within a uniform and a two-layer stratified ambient. For the former case we adapt the theory of Shin et al. (J. Fluid Mech., 521, 2004) and derive a relationship for the deceleration of the front as a function of the slope angle and gravity current density. Experimental results show that the shape of the gravity current is self-similar as it decelerates over relatively steep slopes. The evolution of a gravity current in a two-layer ambient is complicated by the excitation of and interaction with a solitary wave. If the initial gravity current speed is subcritical to the wave speed, the current eventually stops abruptly while the wave continues at constant speed. If supercritical, turbulence at the current head is suppressed as it approaches the interface at the leading edge of the wave and the is re-established as the head becomes in direct contact with the upper layer. 5:45PM L17.00011 Sediment-laden density currents propagating down slopes into stratified ambient , SENTHIL RADHAKRISHNAN, KEVIN SCHMIDMAYER, ECKART MEIBURG, UC Santa Barbara — Intrusions can form when sediment-laden gravity currents propagate down the continental slope into the density stratified ambient ocean. As the particles settle from the initially bottom propagating sediment-laden current, its bulk density decreases, and it eventually lifts off the ground to propagate as an intrusion current. Numerical simulations are performed to study such currents in the lock-exchange configuration. The flow characteristics of the currents, such as their front speed, their lift-off location and their deposit profiles are analyzed as functions of particle size, ambient strength and Reynolds number. As a general trend, currents with larger particles lift off earlier to form intrusions, and they propagate closer to the top surface as compared to currents with smaller particles. We furthermore compare our simulation results with laboratory experiments of Snow and Sutherland (2014). 5:58PM L17.00012 Fluid-Mud Gravity Currents through Vegetation , FIRAT TESTIK, NAZLI YILMAZ, Clemson University — This study was to investigate the effects of emergent stiff aquatic vegetation on the anatomy and propagation dynamics of fluid mud gravity currents. Fluid mud bottom gravity currents propagating through vegetated areas may form during coastal dredge disposal operations. Such currents have distinct anatomical and propagation characteristics. To study these non-Newtonian flows, a set of laboratory experiments were conducted with constantflux release fluid mud (Kaolinite clay mixed with tap water) gravity currents propagating through a vegetated section of a laboratory tank. Emergent aquatic vegetation was simulated using stiff plastic rods of selected patterns. In the experiments, wide ranges of vegetation densities and fluid mud mixture concentrations were used. The experimental gravity currents experienced a drag-dominated propagation phase that was different than the typical propagation phases observed in the absence of vegetation. In this propagation phase, the gravity current exhibited a well-defined triangular / wedge profile. In the talk, these distinct gravity current characteristics associated with the vegetation effects will be discussed along with the underlying physical explanations and developed parameterizations. Monday, November 24, 2014 3:35PM - 6:11PM Session L18 Vortex Dynamics: Vortex Rings — 2004 - Michael Triantafyllou, Massachusetts Institute of Technology 3:35PM L18.00001 On Axisymmetry of Vortex Rings , AHMAD FALAHATPISHEH, ARASH KHERADVAR, Univ of California - Irvine — The shape of vortex rings can be an indication of the axisymmetry of propulsion or transient jet flows. There are many conditions that vortex rings deviate from axisymmetry. For the first time, we introduce a metric, called axisymmetry index, ξ, that measures the axisymmetry of vortex rings with a single value. Axisymmetry index examines the spectrum of the impulse of the ring in azimuthal planes and reports the degree by which the ring deviates from axisymmetry. The index is systematically investigated in analytical and numerical cases. A perfect axisymmetric ring is associated with ξ = 1. It is validated in Gaussian vortex ring and Hill’s spherical vortex, where the computed indices are found to be in agreement with ξ = 1. A family of non-axisymmetric vortex rings are parametrically generated to study the relation of the index with the ratio of the maximum of circulation to the minimum of circulation of the ring. The results show that as the second moment of vorticity on one side increases compared to the other side, the deviation of the ring from axisymmetry increases, hence, a decrease in the index from unity. We also present results of numerically investigating the axisymmetry of a non-axisymmetric vortex ring that forms downstream of an oval-shaped nozzle. 3:48PM L18.00002 On the vortex ring state , RICHARD GREEN, E. GILLIES, M. GIUNI, J. HISLOP, Univ of Glasgow, OMER SAVAS, University of California at Berkeley — The investigation considers the vortex ring state, a phenomenon normally associated with the collapse of a trailing, helical vortex wake into a unstable vortex ring, and is a problem encountered when a helicopter rotor descends into its own wake. A series of wind tunnel and towing tank experiments on rotor systems have been performed, and a comparison is then made with the behaviour of a specially designed open core, annular jet system that generates a mean flow velocity profile similar to that observed below a rotor. In experimentally simulated descents the jet system forms flow patterns that are topologically similar to the vortex ring state of a rotor system. Furthermore the dynamic behaviour of the flow shares many of the important characteristics of the rotor flow. This result suggests that the phenomenon of the vortex ring state of a rotor wake is decoupled from the detailed vortex dynamics of the helical vortex filaments themselves. The presentation will describe the principle behind the investigation, the details of the annular jet system and the results gained using PIV and flow visualisation of the wake and jet systems. 4:01PM L18.00003 Pressure drag evolution of a pair of interacting vortex rings , ANGELIKI LASKARI, RAMMAH SHAMI, ROELAND DE KAT, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton — To determine whether there is a drag benefit for interacting vortex rings compared with a single one, we obtain time evolution of pressure drag coefficients of a pair of interacting vortex rings using 2D Particle Image Velocimetry data. Finite-Time Lyapunov Exponent fields are used to identify vortex boundaries. Streamwise pressure gradients are computed using the incompressible Navier Stokes equations and subsequently integrated across the boundary of the rings. The acceleration term is estimated by either an Eulerian, Lagrangian or Taylor’s Hypothesis approach. Preliminary results show that the latter appears to be the least sensitive to noise, resulting in smoother acceleration fields and fewer oscillations in the evolution of the pressure drag coefficient. Effects of different separation lengths are assessed for the interacting pair and results are compared with those for a single ring. 4:14PM L18.00004 A Pressure-Based Analysis of Vortex Ring Pinch-Off1 , KRISTY SCHLUETER, NOAH BRAUN, JOHN DABIRI, California Institute of Technology — This study investigated the development of vortex rings over a range of maximum stroke ratios, and analyzed vorticity and pressure data for clues to the physical mechanisms underlying vortex pinch-off. An impulsive piston velocity profile and Reynolds number of 3000 were used for all cases. The formation number was consistently found to be 3.6 +/- 0.3. A recently developed algorithm was used to generate pressure fields by integrating the pressure gradient along several paths through the velocity field and taking the median to get explicit values for pressure. The formation time at the occurrence of a local maximum in the pressure between the vortex ring and the lip of the nozzle, known as the trailing pressure maximum, was found to occur concurrently with the formation number for each case, within the error associated with the temporal resolution of the data. This suggests that the trailing pressure maximum is an indicator of vortex ring pinch-off. This is consistent with the results of Lawson and Dawson (2014), who found that the appearance of the trailing pressure maximum was coincident with the formation number. This pressure based approach to determining vortex ring pinch-off will be applied to a biological flow to examine the efficiency of such a flow. 1 This research was partially supported by the Department of Defense (DoD) through the National Defense Science & Engineering Graduate Fellowship (NDSEG) Program. 4:27PM L18.00005 Entrainment in interacting vortex rings , RAMMAH SHAMI, BHARATHRAM GANAPATHISUBRAMANI, Department of Aerodynamics and Flight Mechanics, University of Southampton — The efficiency of entrainment in single vortex rings has been examined by various studies in the literature. These studies have shown that this efficiency is greatly increased for smaller stroke-time to nozzle-diameter ratios, L/D. However, no clear consensus exists regarding the effect on the entrainment process for the sectioned delivery of the vortex forming impulse. In the present work the entrainment mechanism associated with the interaction between two co-axially separated vortex rings is explored. Planar, time-resolved particle image velocimetry (PIV) measurements are taken of a interacting vortex flow field. Lagrangian coherent structures (LCS) extracted from the finite-time Lyapunov exponent (FTLE) fields are employed to determine the vortex boundaries of the interacting rings and is then used to measure entrainment. Preliminary results indicate that whilst the most efficient entrainment of ambient fluid by the ring pairs occurs at larger separations, the rate and overall mass transport increase can be controlled by altering the spatial/temporal separation between successive rings and is higher at smaller ring spacing. Variation in mass transport behaviour for different ring strengths (L/D) and Reynolds numbers will also be discussed. 4:40PM L18.00006 Numerical Study of the Formation and Interaction of Concentric Vortex Rings1 , VAHID SADRI, PAUL S. KRUEGER, SMU — Transient flow between concentric cylinders produces concentric opposite signed vortex rings, which exhibit a range of interesting behavior. Concentric vortex–ring interaction was studied numerically to determine the effects of cylinder gap ratio (∆D/D) and jet stroke length-to-gap ratio (L/∆D) on the evolution of the vorticity and the trajectories of the resulting vortex pair. The flow was simulated at a jet Reynolds number of 2,000, L/∆D in the range 1–15, and ∆D/D in the range 0.05–0.25. The results showed that the position of the vortices relative to each other during the formation phase played an important role in the trajectories of the vorticity centroids at later time. In particular, the vortex pair did not separate during the simulation period when the gap size was less than 0.1 and L/∆D was larger than 5. In the case that ∆D/D was smaller than 0.1 and L/∆D was less than 5, the stopping vortices disturbed the orientation of the vortex pair and affected the evolution of the flow at later time. The general behavior of the vortex trajectories was categorized with respect to the generator parameters (L/∆D, ∆D/D). 1 This material is based upon work supported by the National Science Foundation under Grant No. 1133876 4:53PM L18.00007 Impact of a vortex ring on a conical wall1 , SERGIO HERNANDEZ ZAPATA, ERICK JAVIER LOPEZ SANCHEZ, GERARDO RUIZ CHAVARRIA, Facultad de Ciencias, Universidad Nacional Autonoma de Mexico — In this work we present a numerical and experimental research of a vortex ring impinging a cone. Both the vortex and the conical wall have the same axis of symmetry. For this study we solve the Navier-Stokes and continuity equations in cylindrical coordinates using a finite difference scheme for r, z and time, whereas a Fourier spectral method is used for the angular variable. As initial conditions we assume that velocity is given by the Biot-Savart law for a vorticity distribution of constant magnitude inside a torus. With respect the experiments, measurements of velocity were made with a hot wire anemometer. To have a mapping in space we use a traverse system to place the hot wire probe in points of a grid. Additionally, the measurements of velocity are synchronized with the production of the vortex. Unlike the case of the impact with a flat wall, in this case the diameter of the vortex ring cannot grow. We study the shape of the vortex before the impact, the instabilities and the production of secondary vorticity during the impact. Finally, we made a comparison between experiment and the numerical simulations. 1 Authors acknowledge DGAPA-UNAM by support under project IIN 116312 (Vorticidad y ondas no lineales en fluidos) 5:06PM L18.00008 Vortex Ring Induced Mixing in a “Step” Stratification , JASON OLSTHOORN, STUART DALZIEL, University of Cambridge — The dynamics of fully developed turbulence in a density stratified fluid is highly complex. The highly unstable nature of stratified turbulence and its large range of length scales impede the analysis of the mixing of the density field. In the present work, we consider the mixing induced by coherent vortex rings. Vortex rings provide a reproducible source of kinetic energy and vorticity and have a well defined length scale. By measuring the mixing induced by the isolated mixing events as a result of a vortex ring interacting with a density stratification, we hope to shed insight into stratified turbulent mixing. We initialize a stable density stratification of two different density salt-water layers with a sharp pycnocline between them. We generate vortex rings in the less dense upper layer, and allow these rings to propagate into the more dense lower layer. The result is a pycnocline disturbance which mixes the fluid. Analyzing the change in potential energy of the fluid over multiple vortex ring/pycnocline interactions, we determine that after an initial setup period, the Richardson number dependence of the mixing is balanced by the change in pycnocline height resulting in a constant mixing rate. We present the analysis of the experimental work and discuss its implications. 5:19PM L18.00009 A numerical study of a vortex ring impacting a permeable wall , JING LOU, MING CHENG, Institute of High Performance Computing, T.T. LIM, National University of Singapore — We numerically simulate a vortex ring impacting a permeable wall by using a lattice Boltzmann method. The study is on vortex ring/permeable wall interaction and to address some of the unanswered questions, including core vorticity, kinetic energy and enstrophy of the flow field. The simulation was conducted for a range of parameters such as wall open-area ratios, structure dimensions, wall-thicknesses ( and Reynolds numbers. Results show that with increasing φ or ReΓ enhances vorticity transport across the permeable wall, leading to the formation of a regenerated vortex ring, whilst increasing H impedes vorticity transportation and the formation of regenerated vortex ring. Moreover, higher A promotes vortex shedding from the wire grids and generates fine-scale structures in the wake. 5:32PM L18.00010 Buoyant Norbury’s vortex rings , MARK BLYTH, Univ. of East Anglia, JAVIER RODRIGUEZRODRIGUEZ, Carlos III Univ. Madrid (UC3M), HAYDER SALMAN, Univ. of East Anglia — Norbury’s vortices are a one-parameter family of axisymmetric vortex rings that are exact solutions to the Euler equations. Due to their relative simplicity, they are extensively used to model the behavior of real vortex rings found in experiments and in Nature. In this work, we extend the original formulation of the problem to include buoyancy effects for the case where the fluid that lies within the vortex has a different density to that of the ambient. In this modified formulation, buoyancy effects enter the problem through the baroclinic term of the vorticity equation. This permits an efficient numerical solution of the governing equation of motion in terms of a vortex contour method that tracks the evolution of the boundary of the vortex. Finally, we compare our numerical results with the theoretical analysis of the short-time evolution of a buoyant vortex. Funded by the Spanish Ministry of Economy and Competitiveness through grant DPI2011-28356-C03-02 and by the London Mathematical Society. 5:45PM L18.00011 Numerical investigation of vortex ring formation through a moving valve , LUCAS FARRAR, XUDONG ZHENG, QIAN XUE, None — Impulsively started, low-speed, incompressible jets observed in nature, are commonly found as starting flows through a moving valve. Similar flows are found the human heart where blood is transported from the left atrium, through the mitral valve, and into the left ventricle. During this process, a vortex is formed around the lip of the moving valve before propagating into the left ventricle. We use numerical simulations to investigate the vortex dynamics of starting flows through an axisymmetric nozzle with time varying exit geometry. Following the experimental work of Dabiri & Gharib (J. Fluid Mech., 2005, vol. 538, pp. 111-136), volumetric flow rate is held constant at the nozzle inlet, while the nozzle is treated as a rigid body with motion independent of fluid forces. We show that nozzle motion affects both vortex formation time and pinch-off time as well as the circulation and energy associated with the leading vortex ring. By parametrically ranging over a variety of prescribed flow rates and exit diameter frequencies, the independent contributions of the nozzle motion to the developing vortex structure are assessed. 5:58PM L18.00012 Vortex rings in non-Newtonian viscoelastic fluids play yo-yo , JULIE ALBAGNAC, DAVID LAUPSIEN, DOMINIQUE ANNE-ARCHARD, Institut de Mecanique des Fluides de Toulouse — Vortex rings are coherent vortical structures widely presents in geophysical flows and engineering applications. Numerous applications imply industrial processes including food processing, or petrol industry. Those applications are very often confronted with non-Newtonian fluids. Nevertheless, to the best of our knowledge, only few studies dealing with vortex dynamics in non-Newtonian shear-thinning fluids exist, and none with viscoelastic ones. The aim for the present study is to characterize experimentally the dynamics of vortex rings generated thanks to a piston-cylinder apparatus in various viscoelastic fluids as a function of the generalized Reynolds number, the piston stroke and the final piston position relative to the cylinder exit. In particular, the elastic property of the fluid will be highlighted by the furling-unfurling of vortex rings. Monday, November 24, 2014 3:35PM - 6:11PM — Session L19 Convection and Buoyancy-Driven Flows: Confined Spaces and Geometric Effects 2006 - Gregory Chini, University of New Hampshire 3:35PM L19.00001 2D stratified cavity flow under harmonic forcing1 , BRUNO WELFERT, JUAN LOPEZ, STEPHANIE TAYLOR, Arizona State Univ — Turbulence at the boundary of a stably stratified fluid region can penetrate deep into the region provided the turbulence contains sufficient energy at frequencies less than the buoyancy frequency. This phenomenon manifests itself in the form of internal waves at angles which depend on the perturbation frequencies. Here we consider a 2D lid-driven cavity flow with an imposed stable linear temperature gradient on the sidewalls and constant cold temperature on the bottom and constant hot temperature on the driven lid. In particular, we determine numerically the response to harmonic oscillations of the lid over a range of frequencies and identify resonances and the dynamics associated with their saturation. 1 Supported by NSF grant CBET-1336410 3:48PM L19.00002 Large-scale-circulation dynamics of turbulent Rayleigh-Benard convection in a cubic container , KUNLUN BAI, DANDAN JI, ERIC BROWN, Department of Mechanical Engineering and Materials Science, Yale University — We present measurements of the large-scale-circulation (LSC) in turbulent Rayleigh-Benard convection in a cubic container. The experiments cover the Rayleigh number ranging from 0.5 × 109 to 3 × 109 at Prandtl number 6.4. Using three rows of thermistors at different height of the container, the large-scale-circulation (LSC) can then be identified. It is found that the LSC prefers to lock at the corners of the container, and switches between them stochastically. The strength of LSC keeps as a constant during the switching which suggests those switching correspond to fluctuation driven crossings of a potential barrier in θ due to the side wall geometry as predicted by Brown and Ahlers (Phys. Fluids, 2008). The switching frequency is found to decrease as Ra increases. The measured LSC orientation and its switching will be compared to the model which predicts the effects of container geometry on large-scale coherent structure dynamics for arbitrary geometry. 4:01PM L19.00003 Influence of container shape on scaling of turbulent fluctuations in convection , NAJMEH FOROOZANI, University of Trieste, JOSEPH J. NIEMELA, International Centre for Theoretical Physics, VINCENZO ARMENIO, University of Trieste, KATEPALLI R. SREENIVASAN, New York University — We perform large-eddy simulations of turbulent convection in a cubic enclosure for Rayleigh numbers 1 × 106 ≤ Ra ≤ 1 × 1010 and molecular Prandtl number, P r = 0.7. The simulations were carried out using a second-order-accurate finite-difference method in which subgrid-scale fluxes of momentum and heat were parametrized using a Lagrangian dynamic Smagorinsky model. The scalings of root-meansquare fluctuations of density and velocity in the cell center with Ra differ significantly from those in cylindrical containers, and are in agreement with laboratory observations by Daya and Ecke [Phys. Rev. Lett. 87, 184501 (2001)], also using a cell with square cross-section. We find that the time-averaged spatial distributions of the local heat flux and temperature fluctuations are inhomogeneous in the horizontal plane, associated with the forced orientation of the mean wind along either one or the other diagonal. Larger values of the steady-state density (temperature) gradients occur at the mid-plane corners of the diagonal opposite to that of the mean wind, due to the presence of strong counter-rotating circulations. 4:14PM L19.00004 Experimental and numerical results for CO2 concentration and temperature profiles in an occupied room1 , ALINE COTEL, LARS JUNGHANS, XIAOXIANG WANG, Univ of Michigan - Ann Arbor — In recent years, a recognition of the scope of the negative environmental impact of existing buildings has spurred academic and industrial interest in transforming existing building design practices and disciplinary knowledge. For example, buildings alone consume 72% of the electricity produced annually in the United States; this share is expected to rise to 75% by 2025 (EPA, 2009). Significant reductions in overall building energy consumption can be achieved using green building methods such as natural ventilation. An office was instrumented on campus to acquire CO2 concentrations and temperature profiles at multiple locations while a single occupant was present. Using openFOAM, numerical calculations were performed to allow for comparisons of the CO2 concentration and temperature profiles for different ventilation strategies. Ultimately, these results will be the inputs into a real time feedback control system that can adjust actuators for indoor ventilation and utilize green design strategies. 1 Funded by UM Office of Vice President for Research. 4:27PM L19.00005 How time-varying heating of a wall changes the stratification in a room , RACHAEL BONNEBAIGT, DAMTP, University of Cambridge, C.P. CAULFIELD, BP Institute & DAMTP, University of Cambridge, PAUL LINDEN, DAMTP, University of Cambridge — Building interiors often experience time-dependent heating of vertical surfaces, for example, through sunlight falling on walls. How do these heated surfaces change the temperature stratification in a room? We consider a vertically distributed source of buoyancy, in a sealed insulated space, that provides a linearly-varying-in-time (with slope a) buoyancy flux. This source drives a time-dependent flow: a plume rising up the wall, and return flow in the ambient. We solve the governing equations numerically, using Germeles’s method (1975 J. Fluid Mech. 71 601-623), but we allow the plume to detrain. We find that at small times, the ambient stratification profiles for rates of decrease of source buoyancy flux that are slower than a critical rate, ac < a < 0, are qualitatively similar to those with a > 0, with the profiles getting steeper near the ceiling, while the profiles for a < ac < 0 are qualitatively different, with the profiles getting shallower near the ceiling. We compare these predictions with analogue laboratory experiments. 4:40PM L19.00006 Transient Flows and Stratification of an Enclosure Containing Both a Localised and Distributed Source of Buoyancy , JAMIE PARTRIDGE, PAUL LINDEN, University of Cambridge — We examine the transient flow and stratification in a naturally ventilated enclosure containing both a localised and distributed source of buoyancy. Both sources of buoyancy are located at the base of the enclosure to represent a building where there is a distributed heat flux from the floor, for example from a sun patch, that competes with a localised heat source within the space. The steady conditions of the space are controlled purely by the geometry of the enclosure and the ratio of the distributed and localised buoyancy fluxes Ψ and are independent of the order buoyancy fluxes are introduced into the space. However, the order sources are introduced into the space, such as delaying the introduction of a localised source, alter the transients significantly. To investigate this problem, small-scale experiments were conducted and compared to a ‘perfect-mixing’ model of the transients. How the stratification evolves in time, in particular how long it takes to reach steady conditions, is key to understanding what can be expected in real buildings. The transient evolution of the interior stratification is reported here and compared to the theoretical model. 4:53PM L19.00007 Small scale properties of Rayleigh Benard convection in confined space1 , KAI-LEONG CHONG, MATTHIAS KACZOROWSKI, KE-QING XIA, Department of Physics, The Chinese University of Hong Kong, Shatin — We report a direct numerical simulation (DNS) study on small scale properties of turbulent Rayleigh Benard convection (RBC) in highly confined configurations. Our simulations span a wide range of Rayleigh number (from 107 to 1010 ) at Pr=0.7 and Pr=4.38. It is found that the cell’s smallest dimension, characterized by the aspect ratio, introduces a cut off for the local Bolgiano length scale (evaluated in the bulk of the cell). This result may provide an opportunity for studying the cascade processes in the RBC system through a simply geometrical confinement. Another finding of the study is that the change in flow topology induced by confinement (decreasing aspect ratio) leads to more plumes entering the bulk, thus increasing the velocity and temperature fluctuations in the bulk until the merging of viscous and thermal boundary layers from sidewalls. 1 This work is supported by the NSFC-RGC Joint Research Scheme under Grant No. N CUHK462/11. 5:06PM L19.00008 Aspect-Ratio-Dependent Upper Bounds for Two-Dimensional Rayleigh– Bénard Convection between Stress-Free Isothermal Boundaries , GREGORY CHINI, BAOLE WEN, University of New Hampshire, CHARLES DOERING, University of Michigan — One of the central challenges in studies of Rayleigh–Bénard convection is the determination of the heat transport enhancement factor, i.e. the Nusselt number N u, as a function of the Rayleigh number Ra, Prandtl number P r, and domain aspect ratio L. Although the functional relation between N u, P r and Ra is usually presumed to be N u ∼ P rα Raβ in the “ultimate” high-Ra regime, experiments and simulations have yielded different scaling exponents. Here, we investigate this scaling relationship for two-dimensional Rayleigh–Bénard convection between stress-free isothermal boundaries by computing rigorous upper bounds on the heat transport in domains of varying aspect ratio. Using a novel two-step algorithm (Wen et al. PLA 2013), we numerically solve the full “background field” variational problem arising from the upper bound analysis of Whitehead & Doering √ (PRL 2011) to obtain the optimal bound for Ra ≤ 1010 as a function of L. Our results show that N u ≤ 0.106Ra5/12 at fixed L = 2 2 uniformly in P r, confirming that molecular transport cannot be neglected even at extreme values of Ra. Moreover, for large Ra, the aspect ratio has little impact on the bounds until the domain becomes sufficiently small. 5:19PM L19.00009 Experimental study of free convection in a slender cell using PIV , FERNANDO ARAGON, JUAN CASILLAS, IPN, SALOMON PERALTA, Instituto Mexicano del Petroleo, MARIO SANCHEZ, ABRAHAM MEDINA, IPN — An experimental study of the steady free convection flow induced by a cylindrical heat source immersed in a slender cell for several values of Richardson and Raleigh numbers is undertaken using PIV in order to determine the corresponding velocity fields. The flow is set in motion under the action of a temperature differential (dT) that is induced between the heat source and the surrounding fluid. In the case in which such differential is positive (i.e. hot source) vertical ascending flow occurs, while in the case of a negative value of dT vertical descending flow takes place. Stream or close path trajectories occur, depending on the value of Ri and Ra numbers. Velocity fields and streamlines are presented for several values of Ri and Ra. Said results are compared with numerical models. 5:32PM L19.00010 The effect of cell tilting on turbulent thermal convection in a rectangular cell1 , SHENG-QI ZHOU, SHUANG-XI GUO, XIAN-RONG CEN, LING QU, YUAN-ZHENG LU, South China Sea Institute of Oceanology, CAS, LIANG SUN, University of Science and Technology of China, XIAO-DONG SHANG, South China Sea Institute of Oceanology, CAS — In the study, the influence of cell tilting on flow dynamics is explored experimentally in a rectangular cell (aspect ratios Γx = 1 and Γy = 0.25). The measurements are carried out in a wide range of tilt angles (0 ≤ β ≤ π/2 rad) at Prandtl number (P r ≃ 6.3) and Rayleigh number (Ra ≃ 4.42 × 109 ). With the velocity measurements, the large-scale circulation (LSC) is found to be sensitive to the symmetry of the system. In the level case, the LSC is at about quarter width of the cell. As the cell is slightly tilted (β ≃ 0.04 rad), the LSC moves quickly towards the boundary. With increasing β, the LSC changes gradually from oblique ellipse-like to square-like, and to more complicated patterns. Oscillation has been found for almost all β and it is the strongest at around β ≃ 0.48 rad. With increasing β, the Reynolds number (Re) first increases till it reaches its maximum at the transition angle β = 0.15 rad, then it gradually decreases. A simple energy model is proposed to interpret the cell tilting on flow dynamics. It is predicted that the spatial distribution of the boundary layer affects the flow dynamics by varing the potential energy of system. 1 Supported by China NSF 41176027 and 11072253 and SPR Program of CAS (XDA11030302). 5:45PM L19.00011 Buoyancy and Pressure Induced Flow of Hot Gases in Vertical Shafts with Natural and Forced Ventilation , YOGESH JALURIA, Rutgers University, GUNNAR OLAVI TAMM, US Military Academy — An experimental investigation was conducted to study buoyancy and pressure induced flow of hot gases in vertical shafts to model smoke propagation in elevator and ventilation shafts of high rise building fires. Various configurations were tested with regard to natural and forced ventilation imposed at the upper and lower surfaces of the vertical shaft. The aspect ratio was taken at a typical value of 6. From a lower vent, the inlet conditions for smoke and hot gases were varied in terms of the Reynolds and Grashof numbers. The forced ventilation at the upper or lower boundary was of the same order as the bulk shaft flow. Measurements were taken within the shaft to allow a detailed study of the steady state flow and thermal fields established for various shaft configurations and inlet conditions, from which optimal means for smoke alleviation in high rise building fires may be developed. Results indicated a wall plume as the primary transport mechanism for smoke propagating from the inlet towards the exhaust region. Recirculation and entrainment dominated at high inlet Grashof number flows, while increased inlet Reynolds numbers allowed greater mixing in the shaft. The development and stability of these flow patterns and their effects on the smoke behavior were assessed for several shaft configurations with different inlet conditions. The comparisons indicated that the fastest smoke removal and lowest overall shaft temperatures occur for a configuration with natural ventilation at the top surface and forced ventilation up from the shaft bottom. 5:58PM L19.00012 Instability Of An Advective Flow In An Inclined Fluid Layer With Perfectly Heat-Conducting Boundaries1 , ALBERT SHARIFULIN, Associate prof of Perm National Polytechnic University, RAFIL SAGITOV, Assistant prof of Perm Pharmaceutical Academy — The interest to studying flows in infinite layers induced by the longitudinal temperature gradient is inspired by numerous geophysical and engineering applications (horizontal advective flows in the atmosphere and ocean, convection in vertical and inclined mines and oil wells, etc.). The first investigations of this kind has been done by Ostroumov [1]. He formulated the problem about plane parallel convectional flow in an inclined plane parallel layer with perfectly heat-conducting boundaries caused by the presence of a transverse temperature gradient and the longitudinal temperature gradient. Ostroumov obtained exact analytical solution for the case where the axis of inclination was horizontal and the longitudinal and transverse temperature gradients were perpendicular to this axis. Stability of various limiting cases of the problem (plane layer heated from below, vertical layer heated from the side with and without the longitudinal temperature gradient, and inclined layer between isothermal plates) were studied by many authors. The present paper describes the results of numerical studying of short-wave instability of a plane parallel convective flow in an inclined plane layer with perfectly heat-conducting boundaries under the action of the longitudinal temperature gradient (limited case of Ostroumov problem). [1] G.A. Ostroumov Svobodnaia konvektsiia v usloviakh vnutrennei zadachi. Gostekhizdat (1952). English translation: GA Ostroumov Free convection under conditions of the internal problem. NACA Tech. Memo. 1407(1958). 1 Internal grant of PNIPU Monday, November 24, 2014 3:35PM - 6:11PM Session L20 Flow Control: Plasmas and Actuators — 2008 - Michael Amitay, Rensselaer Polytechnic Institute 3:35PM L20.00001 On the pulsating electric wind of a Single Dielectric Barrier Discharge (SDBD) plasma actuator1 , JULIE VERNET, RAMIS ÖRLÜ, P. HENRIK ALFREDSSON, Linné FLOW Centre, KTH Mechanics, Stockholm, Sweden — An experimental study is conducted on the electric wind produced by a Single Dielectric Barrier Discharge (SDBD) plasma actuator placed at the top of a half cylinder. Laser Doppler Velocimetry (LDV) measurements were performed and results show that increasing the driving voltage (6-16 kV peak-to-peak) and frequency (0.5-2 kHz) of the actuator increases the induced jet velocity (up to 4 m/s) and thus the momentum added by the actuator. The focus of the present study is on the phase-resolved behavior of the electric wind, in particular, its two strokes. Phase-averaged LDV data reveals that while the velocity during both strokes remains positive, there is nearly a factor of two in amplitude. The difference of behavior between the two strokes and its downstream and wall-normal evolution are mapped for various driving voltages. Results indicate that this difference is restricted to the vicinity of the actuator, thereby justifying the assumption of a steady force in simulations to model the induced force. The study is part of a larger investigation aiming at separation control on the A-pillar of a truck cabin. 1 The support of the Swedish Energy Agency and SCANIA CV of the project Flow Research on Active and Novel Control Efficiency (FRANCE) is greatly acknowledged. 3:48PM L20.00002 DBD Actuated Flow Control of Wall-Jet and Cross-Flow Interaction for Film Cooling Applications1 , RAKSHIT TIRUMALA, NICOLAS BENARD, ERIC MOREAU, MATTHIEU FENOT, GILDAS LALIZEL, EVA DORIGNAC, Institut PPRIME, Université de Poitiers (CNRS UPR 3346, ISAE-ENSMA), Boulevard Marie et Pierre Curie, BP30179, 86962 Futuroscope, France — In this work, we use surface DBD actuators to control the interaction between a wall jet and mainstream flow in film cooling applications. The intention of the study is to improve the contact of the jet with the wall and enhance the convective heat transfer coefficient downstream of the jet exit. A 2D wall jet (10 mm height) is injected into the mainstream flow at an angle of 30◦ . With an injected jet velocity (Ui ) of 5 m/s, two blowing ratios M (= ρi Ui / ρ∞ U∞ ) of 1.0 and 0.5 are studied corresponding to the mainstream flow velocity (U∞ ) of 5 m/s and 10 m/s respectively. Different configurations of the DBD actuator are studied, positioned both inside the jet and on the downstream side. PIV measurements are conducted to investigate the flow field of the interaction between the jet and cross flow. Streamwise velocity profiles at different downstream locations are compared to analyze the efficacy of the plasma actuator in improving the contact between the injected jet stream and the wall surface. Reynolds shear stress measurements are also conducted to study the mixing regions in the plasma-jet-mainstream flow interaction. 1 Work was partially funded by the French government program “Investissements d’avenir” (LABEX INTERACTIFS, reference ANR-11-LABX-0017-01). 4:01PM L20.00003 In-Flight Infrared Measurements for Quantification of Transition Delay with DBD Plasma Actuators1 , BERNHARD SIMON, SVEN GRUNDMANN, CSI, TU Darmstadt — Active flow control with a single DBD plasma actuator is performed in flight on wing of a motorized in order to delay laminar-turbulent transition at Rec = 3 · 106 . While earlier experiments measured transition delay with point wise sensors such as microphones or surface hot wires, these dynamic sensors are now simultaneously applied with the infrared measurement technique. This allows a more accurate spatial quantification of the flow control impact. The miniature high resolution IR camera is mounted below the wing as the experiments are conducted on the pressure side. Two control strategies, boundary layer stabilization and active wave cancelation of Tollmien Schlichting (TS) waves, are performed in flight experiments, showing significant advantages of the IR measurement technique. Spanwise and streamwise effects on the transition delay are measured and evaluated with novel post processing strategies. This allows a detailed view on the correlation of TS wave damping and transition delay for different plasma actuator operation modes and flight conditions. 1 This project is founded by the German Research Foundation DFG (GR 3524/4-1) 4:14PM L20.00004 An Out-of-Plane Velocity Component in Dielectric Barrier Discharge Actuator Flow1 , JILLIAN R. KISER, KENNETH S. BREUER, Brown University — The performance of an array of two, two-dimensional dielectric barrier discharge actuators was studied, including both power dissipation measurements and flow visualization using stereo particle image velocimetry. The power dissipation over a range of operating conditions was characterized, showing a relationship between power dissipation, frequency, and voltage such that Pdiss ∝ f V 3.5 . Additionally, the flow induced by plasma generation was measured in quiescent air using PIV, with the driving voltage and frequency being varied. Kinetic energy within a control volume was calculated to quantify the effect of each driving condition, both during the transient start-up flow (lasting up to 600 ms) and at steady state. The induced flow was found to have a non-negligible velocity component in the out-of-plane direction. This component of the kinetic energy, as compared with in-plane kinetic energy, is studied as a function of voltage, frequency, PIV particle size, and actuator design. Potential causes of this velocity component are discussed, considering both electrophoretic forces on the PIV particles, as well as the possibility that it is inherent to the plasma induced flow. 1 This research is funded by the Office of Naval Research (ONR) 4:27PM L20.00005 Electro-Fluid Dynamic Jets , NICHOLAS CAMPBELL, University of Florida — The success of dielectric barrier discharge (DBD) plasma actuators as flow control devices in transducing electrical energy directly into near instantaneous fluid motion has been limited due to momentum loss near the wall. To increase the feasibility of these devices, they have been used to drive a channel flow, creating a jet under quiescent conditions. Electrostatic Fluid Accelerators (EFA) have also been shown to drive internal gas flows. The present work draws on the success of the DBD driven plasma channels, while exploring a new electrode configuration that stems from EFA designs, in order to actuate more of the bulk fluid. Major parameters, applied voltage and operating frequency as well as electrode gap and choice of electrode (material, shape, size); were experimentally investigated using Particle Image Velocimetry to obtain time averaged, 2D velocity fields. Results indicate significant variation of performance with these parameters and suggest that in comparison to surface DBD actuators an order magnitude improvement in efficiency is possible. Furthermore, the qualitative aspect of an electro-fluid dynamic jet shows greater versatility in application for use as both boundary layer flow control and driving internal gas flows. 4:40PM L20.00006 Numerical analysis of ion wind flow using space charge for optimal design1 , HAN SEO KO, DONG HO SHIN, SOO HONG BAEK, Sungkyunkwan University — Ion wind flow has been widly studied for its advantages of a micro fluidic device. However, it is very difficult to predict the performance of the ion wind flow for various conditions because of its complicated electrohydrodynamic phenomena. Thus, a reliable numerical modeling is required to design an otimal ion wind generator and calculate velocity of the ion wind for the proper performance. In this study, the numerical modeling of the ion wind has been modified and newly defined to calculate the veloctiy of the ion wind flow by combining three basic models such as electrostatics, electrodynamics and fluid dynamics. The model has included presence of initial space charges to calculate transfer energy between space charges and air gas molecules using a developed space charge correlation. The simulation has been performed for a geometry of a pin to parallel plate electrode. Finally, the results of the simulation have been compared with the experimental data for the ion wind velocity to confirm the accuracy of the modified numerical modeling and to obtain the optimal design of the ion wind generator. 1 This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Korean government (MEST) (No. 2013R1A2A2A01068653). 4:53PM L20.00007 Influence of Cavity Shape on Synthetic Jet Performance1 , MARK FEERO, PHILIPPE LAVOIE, PIERRE SULLIVAN, University of Toronto — A synthetic jet is a fluidic actuator that transfers linear momentum to the surroundings by alternately ingesting and expelling fluid from a cavity containing an oscillating diaphragm. This work presents the first experimental effort to validate the limited number of numerical investigations that have postulated synthetic jets are insensitive to cavity shape. Three axisymmetric synthetic jets with different cavity shapes were used to examine jet performance while keeping other parameters constant such as cavity volume, nozzle length and orifice diameter. Cylindrical, conical and contraction shaped cavities were considered. The cavity pressure and velocity at the orifice exit plane were measured using a microphone and hot-wire, respectively. The results demonstrated that for several operating conditions near Helmholtz resonance of the cavity, noticeable differences were observed in the radial velocity profiles between the three geometries. The Reynolds number decreased sequentially from the cylindrical to conical to contraction cavity. The momentum flux, which is relevant in flow control applications, followed the same trend. In general, the experimental results showed that synthetic jet performance is, to some degree, dependent on cavity shape. 1 The authors gratefully acknowledge support from the Natural Sciences and Engineering Research Council of Canada. 5:06PM L20.00008 Interactions of a Cross-flow with Dynamic Discrete Roughness Elements , SAMANTHA GILDERSLEEVE, DANIEL SPATCHER, CHIA MIN LEONG, RPI, DAN CLINGMAN, The Boeing Company, MICHAEL AMITAY, RPI, RPI COLLABORATION, THE BOEING COMPANY COLLABORATION — Flow over passive discrete roughness elements was studied in the past to understand their role in controlling transition from laminar to turbulent boundary layers. In the present study, the use of dynamic roughness elements (i.e., low aspect ratio circular cylinders) was explored. These roughness elements were actuated using piezoelectric strips and were designed to have a frequency range of up to 300 Hz. The interaction of a laminar boundary layer over a flat plate with the dynamic roughness elements was studied with both a single and a spanwise array of roughness elements. The discrete elements were 4 mm in diameter, and with a free-stream velocity of 10 m/s, the Reynolds number based on diameter was about 2500. Simultaneous measurements of the performance of the surface roughness elements and the flow field around them were accomplished using a laser displacement sensor and Stereoscopic Particle Image Velocimetry, respectively. Data were collected along spanwise planes at multiple streamwise locations downstream of the roughness elements for the static and dynamic cases at different protrusions into the boundary layer. For the dynamic cases, the effects of amplitude and frequencies were also explored. Flow structures in the wake of these roughness elements will be discussed. 5:19PM L20.00009 Synthetic Jet Actuator Performance Enhancement , LUCIA PIKCILINGIS, KEVIN HOUSLEY, Rensselaer Polytechnic Institute, ED WHALEN, The Boeing Company, MICHAEL AMITAY, Rensselaer Polytechnic Institute, RENSSELAER POLYTECHNIC INSTITUTE COLLABORATION, THE BOEING COMPANY COLLABORATION — Over the last 20 years synthetic jets have been studied as a means for aerodynamic flow control. Specifically, synthetic jets provide momentum transfer with zero-net mass flux, which has been proven to be effective for controlling flow fields. A synthetic jet is created by the periodic formation of vortex rings at its orifice due to the periodic motion of a piezoelectric disk(s). The present study seeks to optimize the performance of a synthetic jet actuator by utilizing different geometrical parameters such as disk thickness, orifice width and length, cavity height and cavity diameter, and different input parameters such as voltage and frequency. Experiments were conducted using a synthetic jet apparatus designed for various geometrical parameters utilizing a dual disk configuration. Velocity and temperature measurements were acquired at the center of the synthetic jet orifice using a temperature compensated hotwire and thermocouple probe. The disk displacement was measured at the center of the disk with a laser displacement sensor. It was shown that the synthetic jet actuators are capable of exceeding peak velocities of 200 m/s with a relatively large orifice. Data suggests that jet velocities greater than 200 m/s are attainable. 5:32PM L20.00010 Interaction of a Dynamic Vortex Generator with a laminar Boundary Layer , ERICA CRUZ, WILFRED CHAN, SHELBY HAYOSTEK, CHIA MIN LEONG, Rensselaer Polytechnic Institute, DAN CLINGMAN, The Boeing Company, MICHAEL AMITAY, Rensselaer Polytechnic Institute — The effectiveness in delaying boundary layer separation by vortex generators (VGs) is well established. However, there could be a drag penalty when the flow it attached. Therefore, in this study, a piezo-based dynamic vortex generator was developed with the goal of mitigating any additional drag that might occur when not in use. The dynamic VG (DVG) was driven by bimorph piezoelectric motor and was designed to operate at frequencies up to 300 Hz. Experimental studies were performed on the interaction of the laminar boundary layer over a flat plate with a DVG placed at a skew angle of 18◦ with respect to the free-stream direction. The experiments were conducted for different height of the DVG, where the Reynolds number based on the local boundary layer thickness was about 2000. In addition, the DVG was oscillating at different frequencies and amplitudes and its effect of the flow field was compared to a steady VG. Simultaneous measurements of the DVG performance and the flow field behind it were accomplished using a laser displacement sensor and Stereoscopic Particle Image Velocimetry (SPIV), respectively. The SPIV data were taken at multiple downstream locations and the flow structures formed in the wake of the DVG will be discussed. 5:45PM L20.00011 Experiments on the Thrust of a Synthetic Jet in Crossflow1 , BRADLEY AYERS, California State University, Northridge, CHARLES HENOCH, Naval Undersea Warfare Center, Newport, HAMID JOHARI, California State University, Northridge — A set of water tunnel experiments were conducted to investigate the effect of crossflow on the thrust of a synthetic jet. This research was motivated by the desire to generate significant turning moments on a fully-submerged, supercavitating vehicle without using control fins or canards. The water tunnel model was a sting-mounted, 3-inch diameter cylindrical body interfaced to a 6-axis waterproof load cell. The synthetic jet actuator was contained within the model and the jet orifice located near the aft end of the model was oriented perpendicular to the mean flow. The actuator consisted of an externally controlled solenoid driving a piston into the cavity. The jet thrust was measured over a broad range of synthetic jet operating parameters, including the actuation frequency and duty cycle, as well as the jet-to-crossflow velocity ratios. Previous work which is based on the slug flow model of an individual vortex ring predicts the time-averaged thrust scales with the square of actuation frequency and the stroke length. The measurements will be compared with the theoretical predictions, and the results will be used to assess the effect of crossflow on the thrust of synthetic jet. 1 Sponsored by the ONR-ULI program 5:58PM L20.00012 High-magnification velocity field measurements on high-frequency, supersonic microactuators , PHIL KRETH, ERIK FERNANDEZ, MOHD ALI, FARRUKH ALVI, Florida Center for Advanced Aero-Propulsion at Florida State University — The Resonance-Enhanced Microjet (REM) actuator developed at our laboratory produces pulsed, supersonic microjets by utilizing a number of microscale, flow-acoustic resonance phenomena. The microactuator used in this study consists of an underexpanded source jet flowing into a cylindrical cavity with a single orifice through which an unsteady, supersonic jet issues at a resonant frequency of 7 kHz. The flowfields of a 1 mm underexpanded free jet and the microactuator are studied in detail using high-magnification, phase-locked flow visualizations (microschlieren) and 2-component particle image velocimetry. The challenges of these measurements at such small scales and supersonic velocities are discussed. The results clearly show that the microactuator produces supersonic pulsed jets with velocities exceeding 400 m/s. This is the first direct measurement of the velocity field and its temporal evolution produced by such actuators. Comparisons are made between the flow visualizations, velocity field measurements, and simulations using Implicit LES for a similar microactuator. With high, unsteady momentum output, this type of microactuator has potential in a range of flow control applications. Monday, November 24, 2014 3:35PM - 6:11PM Session L21 Instability: Richtmyer-Meshkov II — 2010 - Kathy Prestridge, Los Alamos National Laboratory 3:35PM L21.00001 DSMC Studies of the Richtmyer-Meshkov Instability , M.A. GALLIS, T.P. KOEHLER, J.R. TORCZYNSKI, Sandia National Laboratories — A new exascale-capable Direct Simulation Monte Carlo (DSMC) code, SPARTA, developed to be highly efficient on massively parallel computers, has extended the applicability of DSMC to challenging, transient three-dimensional problems in the continuum regime. Because DSMC inherently accounts for compressibility, viscosity, and diffusivity, it has the potential to improve the understanding of the mechanisms responsible for hydrodynamic instabilities. Here, the Richtmyer-Meshkov instability at the interface between two gases was studied parametrically using SPARTA. Simulations performed on Sequoia, an IBM Blue Gene/Q supercomputer at Lawrence Livermore National Laboratory, are used to investigate various Atwood numbers (0.330.94) and Mach numbers (1.2-12.0) for two-dimensional and three-dimensional perturbations. Comparisons with theoretical predictions demonstrate that DSMC accurately predicts the early-time growth of the instability. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. 3:48PM L21.00002 Effects of Initial Conditions on the Mixing Transition of RichtmyerMeshkov Instabilities , RICARDO MEJIA-ALVAREZ, BRANDON WILSON, KATHY PRESTRIDGE, Los Alamos National Laboratory, EXTREME FLUIDS TEAM — A Richtmyer-Meshkov instability (RMI) might experience a mixing transition given the necessary Reynolds number and evolution time (Zhou et al., Phys. Rev. E, 2003). Some studies over broadband initial conditions suggest that the emergence of a classical Kolmogorov κ−5/3 inertial range, in an RMI that has experienced a mixing transition, is independent on the initial conditions. Since observations of this kind have not been replicated for singleor multi-mode perturbed interfaces, it is still premature to consider that the emergence of a Kolmogorov inertial range in RMI after the mixing transition is universal. To shed light on this subject, we are conducting high-resolution simultaneous PIV/PLIF measurements on a multi-mode perturbed interface between air and SF6 . Since our data are also intended for code validation, we used statistically stationary initial conditions, measuring the velocity and density fields both instantaneously and in an averaged sense. Based on our experimental data, we estimate a number of relevant turbulence statistics for different stages of the evolution of the shocked interface. 4:01PM L21.00003 Oscillations of a standing shock in the Richtmyer-Meshkov instability1 , KARNIG MIKAELIAN, Lawrence Livermore Natl Lab — Using the Richtmyer-Meshkov instability we present a method to study the damped oscillations of a standing shock, with and without viscosity. 1 This work was performed under the auspices of the U. S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344. 4:14PM L21.00004 Large-Eddy and Unsteady RANS Simulations of a Shock-Accelerated Heavy Gas Cylinder1 , BRANDON MORGAN, JEFFREY GREENOUGH, Lawrence Livermore National Laboratory — Two-dimensional numerical simulations of the so-called “shock-jet” test problem for Richtmyer-Meshkov instability (RMI) are conducted using both large-eddy simulation (LES) and unsteady Reynolds-averaged Navier-Stokes (URANS) approaches in an arbitrary Lagrangian/Eulerian (ALE) hydrodynamics code. Turbulence statistics are extracted from LES by running an ensemble of simulations with multi-mode perturbations to the initial conditions. Detailed grid convergence studies are conducted, and LES results are found to agree well with both experiment and high-order simulations conducted by Shankar, Kawai, and Lele (Phys. Fluids, 2011). URANS results using a k-L approach are found to be highly sensitive to the initialization of L and to the time at which L becomes resolved on the computational mesh. It is observed that a gradient diffusion closure for turbulent species flux is a poor approximation at early time, and a new closure based on the mass-flux velocity is proposed for low-Reynolds-number mixing. 1 This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344. LLNL-ABS-654292. 4:27PM L21.00005 Numerical investigation of 3D effects on a 2D dominated flow1 , DANIEL REESE, University of Wisconsin, CHRISTOPHER WEBER, Lawrence Livermore National Laboratory — A nominally two-dimensional interface, unstable to the RayleighTaylor or Richtmyer-Meshkov instability, will become three-dimensional at high Reynolds numbers due to the growth of background noise and 3D effects like vortex stretching. This three-dimensionality changes macroscopic features, such as the perturbation growth rate and mixing, as it enhances turbulent dissipation. In this study a 2D perturbation with small-scale, 3D fluctuations is modeled using the hydrodynamics code Miranda. A Mach 1.95 shockwave accelerates a helium/SF6 interface, similar to the experiments of Motl et al. [1], to explore the regime where a 2D dominated flow will experience 3D turbulent effects. We report on the structure and growth of the post-shocked interface, as well as mixing measurements and energy spectra. These metrics are compared against 2D simulations to probe the influence of three-dimensionality on the evolution of the RMI. [1] Motl et al., “Experimental Validation of a Richtmyer-Meshkov Scaling Law Over Large Density Ratio and Shock Strength Ranges” Phys. Fluids (2009) 1 Part of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. 4:40PM L21.00006 A Numerical Investigation of Richtmyer-Meshkov Induced Vorticity in the Multi-Phase Interstellar Medium , GANDHARI WATTAL, SEBASTIAN HEINZ, RICCARDO BONAZZA, JASON OAKLEY, UW-Madison, WISCONSIN SHOCK TUBE LABORATORY COLLABORATION — The interstellar medium (ISM) is inherently a multi-phase fluid, with density contrast of many orders of magnitude. Interstellar turbulence is one of the critical, poorly understood phenomena that regulate processes ranging from star formation to the formation of galactic structure. Baroclinic vorticity generated by the propagation of shocks through the multi-phase ISM is one of the potential drivers of turbulence and an important process in the distribution of momentum and energy into the ISM. We present hydrodynamic and magneto-hydrodynamic models of shock-bubble interactions that investigate the efficiency of vorticity generation in the ISM. The simulations are designed to complement laboratory experiments performed at the Wisconsin Shock Tube, with the goal to (a) calibrate the numerical method, and (b) to extend the investigation to regimes relevant for astrophysics but so far not reproducible in the lab (large magnetization, high Mach numbers). 4:53PM L21.00007 Mach number effects on velocity and density measurements in RichtmyerMeshkov mixing , BRANDON WILSON, RICARDO MEJIA-ALVAREZ, KATHY PRESTRIDGE, Los Alamos Natl Lab, LIUYANG DING, Arizona State University, Laboratory for Energetic Flow and Turbulence (LEFT) — Richtmyer-Meshkov (RM) mixing is sensitive to many parameters: incident Mach number, Atwood number, and initial interface perturbations. The correlation between turbulence and mixing quantities and these parameters is not well-understood. The Vertical Shock Tube (VST) at Los Alamos National Lab is designed to measure the spectrum of scales existing in RM mixing growth and transition to turbulence. We use density (PLIF) and velocity (PIV) diagnostics to understand the effects of Mach number on an air-SF6 interface with multimode perturbations. We quantify Ma effects on the evolution of RM growth at large scales. We then statistically characterize the effect of shock strength on small scale turbulence and mixing (e.g. Favre-averaged Reynolds stresses, instantaneous dissipation rate, and vorticity) at specific times after first shock. First shock mixing appears to transition to turbulence, and we examine the conditions of this transition. We also use first measurements of tomographic PIV in the vertical shock tube to investigate RM mixing anisotropy. 5:06PM L21.00008 Experiments and simulations of single shock Richtmeyer-Meshkov Instability with measured, volumetric initial conditions , EVEREST SEWELL, KEVIN FERGUSON, The University of Arizona, JEFFREY GREENOUGH, Lawrence Livermore National Laboratory, JEFFREY JACOBS, The University of Arizona — We describe new experiments of single shock Richtmeyer-Meshkov Instability (RMI) performed on the shock tube apparatus at the University of Arizona in which the initial conditions are volumetrically imaged prior to shock wave arrival. Initial perturbation plays a major role in the evolution of RMI, and previous experimental efforts only capture a narrow slice of the initial condition. The method presented uses a rastered laser sheet to capture additional images in the depth of the initial condition shortly before the experimental start time. These images are then used to reconstruct a volumetric approximation of the experimental perturbation, which is simulated using the hydrodynamics code ARES, developed at Lawrence Livermore National Laboratory (LLNL). Comparison is made between the time evolution of the interface width and the mixedness ratio measured from the experiments against the predictions from the numerical simulations. 5:19PM L21.00009 Shock-acceleration of a pair of gas inhomogeneities , JOSE ALONSO NAVARRO NUNEZ, DANIEL REESE, JASON OAKLEY, DAVID ROTHAMER, RICCARDO BONAZZA, University of Wisconsin-Madison — A shock wave moving through the interstellar medium distorts density inhomogeneities through the deposition of baroclinic vorticity. This process is modeled experimentally in a shock tube for a two-bubble interaction. A planar shock wave in nitrogen traverses two soap-film bubbles filled with argon. The two bubbles share an axis that is orthogonal to the shock wave and are separated from one another by a distance of approximately one bubble diameter. Atomization of the soap-film by the shock wave results in dispersal of droplets that are imaged using Mie scattering with a laser sheet through the bubble axis. Initial condition images of the bubbles in free-fall (no holder) are taken using a high-speed camera and then two post-shock images are obtained with two laser pulses and two cameras. The first post-shock image is of the early time compression stage when the sphere has become ellipsoidal, and the second image shows the emergence of vortex rings which have evolved due to vorticity depostion by the shock wave. Bubble morphology is characterized with length scale measurements. 5:32PM L21.00010 Shock-Driven Variable-Density Turbulence: New Insights1 , DAVID REILLY, Georgia Institute of Technology, JACOB MCFARLAND, University of Missouri, DEVESH RANJAN, Georgia Institute of Technology — Results are presented from a newly-constructed inclined shock tube facility which was used to study the coupled Richtmyer-Meshkov instability and Kelvin-Helmholtz instability before and after reshock. This study focuses on the effect of multiple initial conditions, which include two Atwood numbers (0.23 and 0.67), two Mach numbers (1.55 and 2.01), and two inclination angles (60◦ and 80◦ ). Mie scattering images of the interface development were acquired to track mixing width. Particle image velocimetry measurements were ensemble averaged over ten instantaneous realizations, which were used to determine circulation deposition as well as turbulent stresses and the cross correlation (u′v′) across the mixing width. Furthermore, energy spectra were obtained for three stages of development before and after reshock. The most developed case exhibited the beginning of an inertial subrange after reshock, which may indicate a turbulent state has been reached. High-resolution planar laser-induced fluorescence was employed to obtain full-field density statistics. The density field was quantified with the density p.d.f. across the mixing width. 1 This 0185. research was funded by the Air Force Office of Scientific Research Young Investigator Research Program (AFOSR-YIP) Grant No. FA9550-13-1- 5:45PM L21.00011 Computational Study of the Richtmyer-Meshkov Instability with a Complex Initial Condition , JACOB MCFARLAND, University of Missouri, DAVID REILLY, Georgia Institute of Technology, JEFFREY GREE- NOUGH, Lawrence Livermore National Laboratory, DEVESH RANJAN, Georgia Institute of Technology — Results are presented for a computational study of the Richtmyer-Meshkov instability with a complex initial condition. This study covers experiments which will be conducted at the newly-built inclined shock tube facility at the Georgia Institute of Technology. The complex initial condition employed consists of an underlying inclined interface perturbation with a broadband spectrum of modes superimposed. A three-dimensional staggered mesh arbitrary Lagrange Eulerian (ALE) hydrodynamics code developed at Lawerence Livermore National Laboratory called ARES was used to obtain both qualitative and quantitative results. Qualitative results are discussed using time series of density plots from which mixing width may be extracted. Quantitative results are also discussed using vorticity fields, circulation components, and energy spectra. The inclined interface case is compared to the complex interface case in order to study the effect of initial conditions on shocked, variable-density flows. 5:58PM L21.00012 Multicomponent Reynolds-Averaged Navier–Stokes Simulations of Reshocked Richtmyer–Meshkov Instability and Turbulent Mixing: Mach Number and Atwood Number Effects1 , TIBERIUS MORAN-LOPEZ, Department of Energy, National Nuclear Security Administration, OLEG SCHILLING, Lawrence Livermore National Laboratory — Reshocked Richtmyer–Meshkov turbulent mixing for various gas pairs and large shock Mach numbers is simulated using a third-order weighted essentially nonoscillatory (WENO) implementation of a new K–ǫ multicomponent Reynolds-averaged Navier–Stokes model. Experiments previously performed at the University of Provence with gas pairs CO2 /He, CO2 /Ar, and CO2 /Kr (with At = −0.73, −0.05, and 0.3, respectively) and incident shock Mach numbers Ma = 2.4, 3.1, 3.7, 4.2, and 4.5 are considered. The evolution of the mixing layer widths is shown to be in good agreement with the experimental data. Budgets of the turbulent transport equations are used to elucidate the mechanisms contributing to turbulent mixing in large Mach number reshocked Richtmyer–Meshkov instability. These results are contrasted with those from previous modeling of smaller Mach number experiments to identify the physical effects which require accurate modeling, including mean and turbulent enthalpy diffusion, pressure–dilatation, and dilatation dissipation. 1 This work was performed under the auspices of the US Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344. Monday, November 24, 2014 3:35PM - 6:11PM Session L22 Instability: Interfacial and Thin Films — 2012 - Patrick Bunton, William Jewell College 3:35PM L22.00001 Forced spreading of films and droplets of colloidal suspensions , LEONARDO ESPIN, SATISH KUMAR, University of Minnesota — When a thin film of a colloidal suspension flows over a substrate, uneven distribution of the suspended particles can lead to an uneven coating. Motivated by this phenomenon, we analyse the flow of perfectly wetting films and droplets of colloidal suspensions down an inclined plane. Lubrication theory and the rapid-vertical-diffusion approximation are used to derive a coupled pair of one-dimensional partial differential equations describing the evolution of the interface height and particle concentration. Precursor films are assumed to be present, the colloidal particles are taken to be hard spheres, and particle and liquid dynamics are coupled through a concentration-dependent viscosity and diffusivity. We find that for sufficiently high Péclet numbers, even small initial concentration inhomogeneities produce viscosity gradients that cause the film or droplet front to evolve continuously in time instead of travelling without changing shape as happens in the absence of colloidal particles. Our results suggest that particle concentration gradients can have a dramatic influence on interface evolution in flowing films and droplets, a finding which may be relevant for understanding the onset of patterns that are observed experimentally. 3:48PM L22.00002 Controlling Viscous Fingering Using Time-dependent Strategies , ZHONG ZHENG, HYOUNGSOO KIM, HOWARD STONE, Department of Mechanical and Aerospace Engineering, Princeton University — Control and stabilization of viscous fingering of immiscible fluids impacts a wide variety of pressure-driven multiphase flows. We report theoretical and experimental results on time-dependent control strategy by manipulating the gap thickness b(t) in a lifting Hele-Shaw cell in the power-law form b(t) = b1 t1/7 . Experimental results show good quantitative agreement with the predictions of linear stability analysis. By choosing the value of a single time-independent control parameter we can either totally suppress the viscous fingering instability or maintain a series of non-splitting viscous fingers during the fluid displacement process. Besides the gap thickness of a Hele-Shaw cell, in principle, time-dependent control strategies can also be placed on the injection rate, viscosity of the displaced fluid, and interfacial tension between the two fluids. 4:01PM L22.00003 Subglacial ice sheet lubrication , KATARZYNA N. KOWAL, M. GRAE WORSTER, Institute of Theoretical Geophysics, DAMTP, University of Cambridge — Large-scale ice-sheet dynamics can be greatly affected by glacial slip, enhanced by subglacial meltwater and water-saturated sediment that acts as a lubricant at the ice-bed contact. Ice streams, for example, are generally lubricated by a layer of water and till at their base and slide up to two orders of magnitude faster than the surrounding ice, making them a major source of discharge of ice into the oceans despite them occupying a relatively small fraction of present-day ice sheets. We present a theoretical and experimental study in which we model the ice and the lubricant as two layers of fluid spreading under their own weight over a smooth, rigid, horizontal surface. The resulting flows are driven by buoyancy and viscous coupling between the layers. Although we are primarily interested in the case in which the underlying fluid has a much smaller viscosity than that of the overlying fluid, the applicability of our model extends to two-layer gravity currents with general viscosity ratios. There is excellent quantitative agreement between our theory and a series of laboratory experiments that we have conducted using simple, Newtonian fluids. A novel fingering instability develops at later stages of our experiments. 4:14PM L22.00004 A Solutal Fingering Instability during Capillary Imbibition in Fibrous Media , CHRISTOPHER GUIDO, NICOLAS YOUNG, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — We report the existence of a solute-driven, humidity-dependent fingering instability that occurs during capillary imbibition into cellulosic fibrous media (e.g., paper). For sufficiently low solute concentrations and sufficiently high ambient humidities, the imbibition front moves forward smoothly; for higher concentrations and lower humidities, however, the imbibition front develops spatially periodic fluctuations that grow with time. We derive and experimentally corroborate a stability criterion based on solute-induced changes in the air/liquid interfacial tension, which are magnified by liquid infiltration into a humiditydependent precursor film. The results have broad implications for any process involving motion of liquids through fibrous media, including chromatographic separations, paper-based diagnostic assays, and conservation efforts involving aged manuscripts or artwork. 4:27PM L22.00005 Non-modal disturbances growth of miscible viscous fingering in porous media1 , TAPAN KUMAR HOTA, MANORANJAN MISHRA, Indian Institute of Technology Ropar, India — The transient amplification of disturbances in a pressure driven rectilinear flow of two miscible fluids with varying viscosity in porous media are examined. The system has been studied by coupling the continuity and Darcy equations with a convection-diffusion equation for the evolution of solute concentration. Since the base state is time dependent, the common techniques used in the literatures for studying the linear stability are either quasi-steady state approach or initial value approach with random initial disturbance or both. To overcome difficulties in these approaches, the non-modal analysis (NMA) has been employed to study the amplification of disturbances. The Runge-Kutta method has been used to solve the matrix differential equation obtained by NMA from the linearized equations. The optimum amplification and structures of the disturbances are found by singular value decomposition. Initial disturbances that lead to the optimum amplifications are found to be localized within the diffusive layers, unlike the random disturbances used in the initial value technique. It has also been observed that the optimum growth obtained by NMA decays at early time due to the diffusion before it starts amplifying, unlike the results of modal analysis. 1 Financial support from DST, Govt. of India has been gratefully acknowledged 4:40PM L22.00006 Elastic viscous fingering , JOHN LISTER, GUNNAR PENG, University of Cambridge — The Saffman–Taylor viscous-fingering instability in a circular Hele-Shaw cell can be suppressed by replacing one of the rigid walls with an elastic sheet (Pihler-Puzovic, Illien, Heil, Juel, 2012). We successfully reproduce these results numerically by considering linear non-axisymmetric perturbations to an axisymmetric evolving base state. Our calculations show that, in the relevant parameter regime, the non-axisymmetric perturbations to the elastic sheet are negligible. Instead, the elastic suppression of the fingering instability is due to changes to the axisymmetric base state. We identify four physical mechanisms that affect the stability of the system, and find that the contribution from each one is significant. 4:53PM L22.00007 What the geometry of a river network says about its growth , OLIVIER DEVAUCHELLE, Institut de Physique du Globe de Paris, YOSSI COHEN, HANSJOERG F. SEYBOLD, ROBERT S. YI, Lorenz Center, Department of Earth Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, PIOTR SZYMCZAK, Institute of Theoretical Physics, Faculty of Physics, Warsaw University, DANIEL H. ROTHMAN, Lorenz Center, Department of Earth Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology — The growth of a river network is governed by the flow of rainwater towards it. When the streams drain groundwater, this flow conforms to a harmonic field, thus turning the network growth into an analogue of Saffman-Taylor fingering and diffusion-limited aggregation. A theoretical description of this process should specify (i) how fast a river grows, (ii) in which direction and (iii) when it bifurcates. Simple physical reasoning suggests that a river grows along the groundwater flow lines (geodesic growth). In a harmonic field, this hypothesis sets the branching angle of the network to 72◦ , regardless of the other growth rules. This geometrical property appears unambiguously in nature. Inspired by fracture mechanics, we reformulate the geodesic growth in terms of local symmetry: as it cuts into the landscape, a river maintains a symmetric groundwater flow around its tip. Based on this principle, we reconstruct the history of the network by growing it backwards from its present geometry. We then use this history to infer the network’s dynamics. 5:06PM L22.00008 Schlieren Imaging of Viscous-Fingering in a Horizontal Hele-Shaw Cell1 , PATRICK BUNTON, GABRIELLE BROOKS, SIMONE STEWART, Department of Physics, William Jewell College, Liberty, MO USA, ANNE DE WIT, Nonlinear Physical Chemistry Unit, Université Libre de Bruxelles, Brussels, Belgium — Viscous fingering (VF) occurs when a fluid of high mobility displaces a fluid of lower mobility. Recent increased interest is motivated by applications to enhanced petroleum recover, pollutant dispersal, and climatological issues along with increased computational capability. Most often VF is observed in a Hele-Shaw (HS) cell consisting of two transparent plates separated by a narrow gap. For the typical case of transparent fluids, dyes are used for observation. Chemical indicators are used for reactive studies. Other techniques have been used such as interferometry, Schlieren, shadowgraph, fluorescence, and MRI. Here is reported a modification of Schlieren for use in imaging horizontal flows in a HS cell. The technique requires no dyes or chemical indicators that might complicate interpretation or even alter the dynamics. It is exquisitely sensitive, readily yielding information about 3D flows in gaps under a mm. Schlieren imaging is particularly useful in that it allows one to image flows within the fingers, rather than merely imaging the boundary. Following a description of the technique, data for water-glycerol systems are presented revealing previously unobserved internal detail. This detail is interpreted in terms of recently published 3D models of VF. 1 Supported by the National Science Foundation CBET 1335739 5:19PM L22.00009 A new Saffman-Taylor growth rate formula , PRABIR DARIPA, Texas A&M University — In this talk, we discuss modification of the classical Saffman-Taylor growth rate formula when the dynamic Laplace law including viscous stress tensor on the interface is included in the linear stability analysis for the displacement of a Newtonian fluid by air. In particular, we derive a new formula for the growth rate and show that the problem is linearly well-posed for all values of surface tension. This is a joint work with Gelu Pasa. 5:32PM L22.00010 Experimental study on effects of effective interfacial tension on miscible viscous fingering , FU WEI QUAH, YU QI, YUICHIRO NAGATSU, Department of Chemical Engineering, Tokyo University of Agriculture and Technology — We experimentally investigate effects of effective interfacial tension (EIT) on miscible viscous fingering (VF). To do so, we prepare two miscible liquid systems in which the viscosity contrast between the more- and less viscous liquids is the same but the EIT between the two liquids is different. We confirm that the viscosity is the same in both the systems but EIT is different by means of the measurements of viscosity and EIT. We perform VF experiment by using a Hele-Shaw cell. We find that the typical width of the fingers is larger in the system involving larger EIT. This experimental result has a good agreement with recent numerical studies of the related issue. 5:45PM L22.00011 Characteristics of proportionate growth observed in instability patterns of miscible fluids , IRMGARD BISCHOFBERGER, RADHA RAMACHANDRAN, SIDNEY R. NAGEL, University of Chicago, NAGEL LAB TEAM — As a baby mammal grows, different parts of its body develop at the nearly the same rate and thus to a good approximation in direct proportion to one another. This type of growth is called proportionate growth. As familiar as it appears to us, it is very rarely found in physical systems outside of the biological world. We here show an example of proportionate growth that occurs in the instability formed when a less viscous liquid, of viscosity ηin displaces a more viscous miscible one, of viscosity ηout . We investigate the growth of these patterns in a quasi-two-dimensional geometry. Within a range of viscosity ratios 0.1 <ηin /ηout <0.3, we observe the formation of small blunt structures that form at the edges of an inner circular region devoid of fingers. As the pattern grows, the size of these structures increases in proportion to the size of the inner circle, such that even small details in the shape of the pattern remain essentially unchanged during growth. These characteristics of proportionate growth are reflected in the shape of the interface in the third dimension as well. 5:58PM L22.00012 Shear-induced morphology in mixed phospholipid films , AMIR HIRSA, JAMES YOUNG, DAVID POSADA, Rensselaer Polytechnic Institute, JUAN LOPEZ, Arizona State University — Flow of mixed phospholipid films on liquid surfaces plays a significant role in biological processes ranging from lipid bilayer fluidity and the associated behavior of cellular membranes, to flow on the liquid lining in the lungs. Phospholipid films are also central to the process of two-dimensional protein crystallization below a ligand-bearing film. Here, we study a binary mixture of phospholipids that form an insoluble monolayer on the air-water interface. Brewster angle microscopy reveals that a shearing flow induces a phase separation in the binary film, resulting in the appearance of 10 micron-scale dark domains. Hydrodynamic response of the binary film is quantified at the macro-scale by measurements of the surface shear viscosity, via a deep-channel surface viscometer. Reynolds number was shown to be a state variable, along with surface pressure, controlling the surface shear viscosity of a biotinylated lipid film. Monday, November 24, 2014 3:35PM - 6:11PM Session L23 Geophysical Fluid Dynamics: Boundary Layers — 2001 - Eric Arobone, Stanford University 3:35PM L23.00001 The near-wall structure of the vorticity field in atmospheric flows , CURTIS HAMMAN, PARVIZ MOIN, Center for Turbulence Research, Stanford University — Prompted by suggestions that long-lived, coherent structures are regions of high helicity (~ ω·~ u) and low dissipation, Rogers & Moin (1987, PoF) examined the helicity field in turbulent channel flow, and did not find evidence to support this helicity conjecture. They concluded, however, that buoyancy forces may preferentially and chronically concentrate cork-screw eddies in wall turbulence. We examine this hypothesis by studying numerical simulation databases of thermal convection with a mean flow in large-aspect ratio channels. Roll cells generated by buoyancy forces in the bulk are contrasted with near-wall, hairpin-like vortices sustained by mean shear. At moderate bulk Richardson numbers, near-wall helicity fluctuations increase showing strong peaks in relative helicity density pdfs but less so in regions of low dissipation. Transverse strain imposed by erupting and impinging thermal plumes embedded in the streamwise-aligned, large-scale circulation is found to tilt these hairpin packets in a herringbone pattern reducing the local turbulence production but increasing the local turbulent dissipation as in 3D turbulent boundary layers. Recent simulations and implications for understanding large-eddy structures in PBLs using LES are discussed. 3:48PM L23.00002 Thermal transport processes in stable boundary layers , WALTER GUTIERREZ, GUILLERMO ARAYA, National Wind Resources Center, Texas Tech University, Lubbock, TX, USA, PRAJU KILIYANPILAKKIL, SUKANTA BASU, Department of Marine, Earth, and Atmospheric Sciences; North Carolina State University, Raleigh, NC, USA, ARQUIMEDES RUIZ-COLUMBIE, National Wind Institute, Texas Tech University, Lubbock, TX, USA, LUCIANO CASTILLO, National Wind Resources Center, Texas Tech University, Lubbock, TX, USA — Using the 200-m tower data (Reese, Texas), profiler and Mesonet data, and WRF runs, a 4-dim model is introduced which summarizes the main features of the Low Level Jet (LLJ) in stable boundary conditions over the aforementioned region and shows its patterns along the year. We also demonstrate the importance of LLJs for wind energy production. It has been observed that during a LLJ event the level of turbulence intensities and TKE are significantly much lower than those during unstable conditions. The major salient results from this study include: the vertical shears in the LLJ are very large at the current wind turbine heights, causing higher static and cyclical aerodynamic loads. The WRF model has accurately captured the beginning and end of the LLJ event; however, the local maximum wind speed at the LLJ “nose” has been under-predicted by approximately 15%, which highlights the difficulties WRF still faces in predicting this phenomenon. Furthermore, power spectra and time-autocorrelations of thermal fluctuations will help us in the understanding of the thermal coherent structures involved in moderate and strong LLJ. 4:01PM L23.00003 Non-equilibrium model of spray-stratified atmospheric boundary layer under high wind conditions1 , YEVGENII RASTIGEJEV, North Carolina A&T State University, SERGEY SUSLOV, Swinburne University of Technology, Australia — Tropical cyclone is a complex meteorological phenomenon which dynamics is defined by a wide variety of factors including exchange of momentum, heat and moisture between the atmosphere and the ocean. Ocean spray plays an important role in this air-sea interaction. Here we developed a two-temperature non-equilibrium variable density (non-Bousinessq) turbulence closure model to describe the ocean spray-stratified hurricane boundary layer structure and dynamics. The model consistently describes a two-way coupling between mechanical and thermodynamic influences of the ocean spray. The obtained results confirm that the impact of non-equilibrium effects is significant over the complete range of possible spray concentration values, therefore has to be included into a consistent parameterization of moisture, heat and momentum transfer process over the ocean under high wind condition of a hurricane. 1 NSF HRD-1036563 4:14PM L23.00004 Evaluation of the sphere anemometer for atmospheric wind measurements1 , HENDRIK HEISSELMANN, JOACHIM PEINKE, MICHAEL HOELLING, ForWind - University of Oldenburg — Our contribution will compare the sphere anemometer and two standard sensors for wind energy and meteorology based on data from a near-shore measurement campaign. We will introduce the characteristics of the sphere anemometer - a drag-based sensor for simultaneous wind speed and direction measurements, which makes use of the highly resolving light pointer principle to detect the velocity-dependent deflection of sphere mounted on a flexible tube. Sphere anemometer, cup anemometer and 3D sonic anemometer were installed at near-shore site in the German Wadden Sea. A comparison of the anemometers was carried out based on several month of high frequency data obtained from this campaign. The measured wind speed and direction data were analyzed to evaluate the capability of the sphere anemometer under real operating conditions, while the sensor characteristics obtained from previous wind tunnel experiments under turbulent conditions served as a reference to assess the durability and to identify challenges of the new anemometer. A characterization of the atmospheric wind conditions at the test site is performed based on the recorded wind data. Wind speed and wind direction averages and turbulence intensities are analyzed as well as power spectra and probability density functions. 1 Supported by the German Ministry of Economics and Energy. 4:27PM L23.00005 Effects of stratification on an ocean surface Ekman layer , HIEU PHAM, SUTANU SARKAR, UC San Diego — Large-eddy simulations are used to investigate the effects of stratification on structural and turbulent dynamics of an upper-ocean Ekman layer that is driven by a constant wind stress (friction velocity u∗ ) at low latitude with Coriolis parameter f . The surface layer evolves in the presence of interior stratification whose buoyancy frequency varies among cases, taking three values: N/f = 19, 60 and 192. At quasi-steady state, a stratified turbulent Ekman layer forms with a surface current veering to the right of the wind direction. The thickness of the Ekman layer decreases with increasing N and is found to scale with u∗ , f , and N , similar to the neutral atmospheric boundary layer of Zilitinkevich & Esau (2002) that is capped by a stratified layer with buoyancy frequency, N . As N increases, the speed of the Ekman current increases but the Ekman transport is invariant. The surface veering angle also increases with larger N . The shear rate and buoyancy frequency are elevated at the base of the Ekman layer. The peak of down-wind Reynolds stress occurs near the surface and scales with u∗2 in all cases while the peak of cross-wind Reynolds stress occurs in the middle of the Ekman layer and decreases with increasing N . 4:40PM L23.00006 A mass-spring-damper model for unsteady Ekman boundary layers1 , MOSTAFA MOMEN, ELIE BOU-ZEID, Department of Civil and Environmental Engineering, Princeton University, Princeton, New Jersey, USA — The Ekman boundary layer is a central problem in geophysical fluid dynamics that emerges in atmospheric and oceanic boundary layers when pressure gradient forces, Coriolis forces, and molecular or turbulent friction forces interact in a flow. These boundary layers are dynamical systems; however, due to their inherent complexity most studies of these wall-bounded flows have focused on steady state conditions. The transient version of the problem, which occurs when these forces are not in equilibrium such as when the pressure gradients are changing in time, is solvable analytically only for a limited set of forcing variability modes, and the resulting solutions are intricate and difficult to interpret. In this study, we derive a simple physical model that reduces Navier-Stokes equations into a second-order ordinary differential equation that is very similar to the dynamical equation of a mass-spring-damper system. The validation of the proposed model is performed by comparing it to results from a suite of large-eddy simulations. The reduced model can be solved for a wider range of variable forcing conditions and serves to elucidate the physical origin of the inertia (mass), energy storage (spring), and energy dissipation (damper) attributes of the Ekman layer. 1 The authors acknowledge support from the CICS of Princeton University and the NOAA’s under grant number 344-6127, and the Physical and Dynamic Meteorology Program of the National Science Foundation under AGS-1026636. 4:53PM L23.00007 Effects of three-dimensionality on frontal instability , ERIC AROBONE, Stanford University, SUTANU SARKAR, University of California, San Diego — The pure symmetric instability (SI) is a frontal instability that is independent of the along-front coordinate. Observational evidence suggests that along-front variability in vertical velocity and SST anomaly is not negligible (D’Asaro et al. Science. 2011 and Thomas et al. DSR2. 2013). We examine the three-dimensional evolution of frontal shear instabilities from both linear and non-linear perspectives. Linear stability suggests that significant growth rates are possible when along-front and across-front variabilities are comparable. Additionally, along-front variability results in misalignment of perturbations with respect to isopycnals. A suite of three-dimensional Direct Numerical Simulations (DNS) are performed exploring a horizontally homogeneous front with differing domain lengths in the along-front direction. For sufficiently large along-front domain lengths, the front develops along-front variation and a pure symmetric instability is not found in the DNS. The consequence of asymmetry in the instability on frontal evolution will be discussed. The effect of three-dimensionality of initial conditions will also be explored. 5:06PM L23.00008 Lagrangian Coherent Structures in an Unstable Bottom Boundary Layer Under a Solitary Wave , DANIEL NELSON, GUSTAAF JACOBS, San Diego State University, MAHMOUD SADEK, PETER DIAMESSIS, Cornell University — The role of Lagrangian Coherent Structures (LCS) in fluid mixing is investigated in the unstable bottom boundary layer (BBL) under a solitary surface wave mimicked by a soliton-like pressure gradient driven flow in an oscillating water tunnel. The finite-time Lyapunov exponent (FTLE) field, both backward in time and forward in time, is determined for a two-dimensional direct numerical simulation (DNS) of the unstable BBL from the development of the instability through the growth of the large scale transport structures. Attracting LCS are identified trailing the primary vortices that form moving separation surfaces which pick up material from the boundary and transport it into the primary vortices. Weaker, secondary separation surfaces form beneath smaller, secondary vortices. At a later time, the secondary vortices are absorbed by the primary vortices and the separation surfaces from the smaller vortices merge with the separation surfaces from the larger vortices. The primary vortices are the most significant sources of mixing between the near wall and outside the boundary layer, implying that the primary vortices are the physical mechanism for particle resuspension. 5:19PM L23.00009 Instability map and transition characteristics of the bottom boundary layer under solitary wave , MAHMOUD SADEK, Cornell University, LUIS PARRAS, University of Malaga, PETER DIAMESSIS, PHILIP LIU, Cornell University — Transition prediction in the bottom boundary layer (BBL) flow driven by a soliton-like pressure gradient in an oscillating water tunnel (an approximation for the BBL under solitary waves) is investigated using hydrodynamic linear stability theory. The study of transition in such a flow is divided into two approaches. The first approach is associated with the classical transition resulting from the exponential growth of TS waves. The findings for this approach can be summarized in a map for the temporal instability of the base flow (BF). In this map, the connections between experimental observations, classical stability analysis and fully non-linear 2D numerical simulations have been established. The second approach deals with an alternative transition scenario for this BF due to the algebraic growth of the disturbance leading to the formation of turbulent spots (TS) as reported in laboratory experiments. In this regard, the stability analysis is reformulated in the non-modal framework for the purpose of finding the optimum disturbance characteristics leading to the formation of the observed TS waves. The results of the non-modal analysis are used as an input for 3D DNS of the BBL which aims to mimic the experimental observations and understand the different possible transition scenarios within the BBL under solitary waves. 5:32PM L23.00010 Flow Structure and Turbulence Characteristics downstream of a Spanwise Suspended Linear Canopy through Laboratory Experiments , JUNDONG QIAO, SARAH DELAVAN, State Univ of NY - Buffalo — Laboratory experiments were conducted to explore the mean flow structure and turbulence properties downstream of a spanwise suspended linear canopy in a 2-D open channel flow using the Particle Tracking Velocimetry technique. This canopy simulated the effect of one long-line structure of a mussel farm. Four experimental scenarios with the approach velocities 50, 80, 110, and 140 mm s−1 were under investigation. Three sub-layers formed downstream of the canopy. An internal canopy layer, where the time-averaged velocity decreases linearly with increasing distance downstream, a canopy mixing layer increasing in vertical extent with increasing distance downstream of the canopy, and an external canopy layer with higher velocity under the canopy, which may bring nutrients from the local ambient environment into this layer. The canopy turbulence results in upward momentum transport downstream of the canopy within a distance of 0.60 of the canopy depth and downward momentum transport beyond 1.20 of it. In the scenarios with relatively lower approach velocities 50 and 80 mm s−1 , the wake turbulence results in upward momentum transport. The broader goal of this study is to offer guidelines for the design and site selection of more productive mussel farms. The results suggest that distance interval between the parallel long-lines in a mussel farm should be less than 0.6 times the height of a long-line dropper. Also, potential farm locations that are characterized with current velocity from 50 to 80 mm s−1 are suggested. 5:45PM L23.00011 Dynamic Exact Solutions For Stratified Wall Shear Flows1 , GEORGE HARABIN, ROBERTO CAMASSA, TYLER KRESS, GRACE MCLAUGHLIN, RICHARD MCLAUGHLIN, UNC Chapel Hill, UNC JOINT FLUIDS LAB TEAM — An exact time dependent shear flow solution to the full Navier-Stokes equations under the Boussinesq approximation coupled to the advection-diffusion equation for density is investigated in semi-infinite domains with sloped wall boundaries. This solution extends the static solution found by O.M. Phillips in 1969 to include oscillatory time evolution. Long time asymptotics based on the analysis of the branch cut structures in the transform domain are derived and analyzed. Comparisons with preliminary experiments will be discussed. 1 NSF and ONR 5:58PM L23.00012 Prandtl effects on mixing in nonlinear spinup1 , MELINE BAGHDASARIAN, ARTURO PACHECO-VEGA, Cal State Univ- Los Angeles, J. RAFAEL PACHECO, SAP Americas, ROBERTO VERZICCO, Universita di Roma “Tor Vergata” — Stratified spin-up experiments in enclosed cylinders have reported the presence of small pockets of well-mixed fluids; however, there have been shortfalls in terms of quantitative accounts of the mixedness of the fluid. Previous numerical studies reported in the literature have not been able to quantify these measurements either. Here we present a series of three-dimensional numerical simulations that address how the combined effect of spin-up and thermal boundary conditions for various Prandtl numbers enhances or hinders mixing of a fluid in a cylinder. Measurements of efficiency of mixing are based on the variance of temperature and explained in terms of the potential energy available. The numerical simulations of the Navier–Stokes equations for the problem with different sets of thermal boundary conditions at the horizontal walls and varying Prandtl number reported here have helped shed some light on the physical mechanisms of mixing, for which a clear explanation was lacking. 1 This project has been partially supported by NSF grants HRD-0932421 and ARA-R2-0963539. Monday, November 24, 2014 3:35PM - 5:58PM Session L24 Granular Flows: Fluctuations and Instabilities — 2003 - Prabhu Nott, Indian Institute of Science 3:35PM L24.00001 Stability analysis of Couette flows of dry granular materials , CHRISTOS VARSAKELIS, MILTIADIS PAPALEXANDRIS, Univ Catholique de Louvain — In this talk, we investigate the stability of a unidirectional Couette flow of a dry granular material, as predicted by a continuum flow model, via a linear stability analysis. A classical normal-mode analysis is employed which results in a fourth-order polynomial eigenvalue equation for the modes of disturbance. The eigenvalue problem is solved numerically via a Chebyshev polynomial method and extensive parametric studies are performed. The results of this study suggest that the flow of interest is linearly unstable for all values of Froude, Reynolds and Galilei numbers of practical interest. Additionally, we discuss the relation between the magnitude of the predicted growth rates and the observability of this instability, as well as the connection between the shape of the predicted eigenfunctions and the formation of particle-clusters. Finally, we compare the results of the present study with those of earlier studies based on different flow models. 3:48PM L24.00002 Taylor-Couette like vortices in slow shear of dense granular materials , PRABHU NOTT, KRISHNARAJ K P, Indian Institute of Science — In two recent papers, we had reported a remarkable anomaly in the stress in a dense granular column sheared in cylindrical Couette device: the magnitudes of all components of the stress increase nearly exponentially with depth, and the vertical shear stress changes sign upon shearing. While it was clear that gravity and the confining walls play important roles, the precise mechanism responsible for the anomaly was then unclear. Here we report the results of subsequent experiments and DEM simulations that partially reveal the origin of the anomaly. It appears that shearing the granular material results in Taylor-Couette like vortices, as in a fluid. However, the analogy with fluids does not carry very far: (i) despite the velocity of the secondary flow being very small, its contribution to the stress is very large, (ii) there is always a single vortex that spans the entire column, regardless of its height, and (iii) the vortex does not appear to be caused by a centrifugal instability. We shall discuss these features, and comment on the ability of plasticity theories in predicting the observations. 4:01PM L24.00003 Collapse of a brittle granular column: implications for rock fragmentation in a landslide , VINCENT LANGLOIS, Universite Claude Bernard Lyon 1, AMELIE QUIQUEREZ, Universite de Bourgogne, PASCAL ALLEMAND, Universite Claude Bernard Lyon 1 — We investigate numerically the failure, collapse and flow of a brittle granular column over a horizontal surface. In our discrete element simulations, spherical particles are initially held together by tensile bonds, which can be irreversibly broken during the collapse. This leads to dynamic fragmentation within the material during the flow. Compared to what happens in the case of a non-cohesive granular column, the deposit is much rougher, and the stratigraphy of the column is not preserved during the collapse. As has been observed in natural rockslides, we find that the deposit consists of large blocks laying on a basal layer of fine fragments. The influence of the aspect ratio of the column on the run-out distance is roughly the same as in the granular case. Finally, we show that for a given aspect ratio of the column, the run-out distance is higher when the deposit is highly fragmented, which confirms previous hypotheses by Davies et al. (1999). 4:14PM L24.00004 From episodic avalanching to continuous flow in a granular drum , NEIL BALMFORTH, University of British Columbia, JIM MCELWAINE, University of Durhm — Experiments are performed with a rotating cylindrical drum half full of granular material in order to study the transition from episodic avalanching to continuous flow (slumping to rolling). To examine the effect of drum and particle geometry, drums with different radii and widths are used, and different granular materials, ranging from glass spheres with different radii to irregularly shaped sand. For the drums and materials used, it is found that the transition mostly takes the form of a blend of the characteristics of episodic avalanching and continuous flow, that gradually switches from slumping to rolling as rotation rate increases. Only for the sand in the narrower drums is there a hysteretic transition in which one can observe prolonged episodic avalanching or continuous flow at the same rotation rate, over a window of rotation speeds. The transition takes the form of intermittent switching driven by noisy fluctuations (a “bifurcation by intermittency”) for sand in the widest drums and for the smallest ballotini (1mm diameter). 4:27PM L24.00005 Numerical study of axisymmetric collapses of submarine granular > columns , DAVIDE MONSORNO, CHRISTOS VARSAKELIS, Univ Catholique de Louvain — In this talk, we report on the results of a numerical study of the axisymmetric collapse of subaqueous granular columns. Our study is based on a 2-pressure, 2-velocity continuum flow model for fluid-saturated granular materials. This model is integrated via a multi-phase projection method that incorporates a regularization method for the treatment of material interfaces. In our simulations, a dense column of a granular material immersed in water is placed on a horizontal plane and is allowed to collapse and spread due to its weight. Emphasis is placed on the run-out distance and the termination height and their correlation with the aspect ratio, the volume fraction and the diameter of the grains. Comparisons against experimental measurements and previous numerical predictions are also performed. Finally, in order to examine and quantify the role of the interstitial fluid, we compare our numerical predictions against experimental results from column collapses of dry granular materials. 4:40PM L24.00006 Computer simulations of the axisymmetric collapse of a granular column made of mixed grains , HORACIO TAPIA-MCCLUNG, LANIA Mexico — Measures of the final height and final run-out distance of deposits formed by collapses of granular columns formed by mixed grains, seem to follow a power law on the initial aspect ratio parameter, similar to the homogeneous case. We investigate this matter using numerical simulations in 3D, and present preliminary results of the collapse of a granular column formed by mixtures of grains with different shapes, considering the cases of spherical grains mixed with sticks and mixtures of sticks, keeping all other parameters fixed. 4:53PM L24.00007 Low-frequency oscillations in vibrated granular media , NICOLAS RIVAS, ANTHONY THORNTON, STEFAN LUDING, University of Twente, KIT WINDOWS-YULE, DAVID PARKER, University of Birmingham — We present simulations and a theoretical treatment of vertically vibrated granular media. The systems considered are confined in narrow quasi-two-dimensional and quasi-one-dimensional (column) geometries, where the vertical extension of the container is much larger than both horizontal lengths. The additional geometric constraint present in the column setup frustrates the convection state that is normally observed in wider geometries. This makes it possible to study collective oscillations of the grains with a characteristic frequency that is much lower than the frequency of energy injection. We observe that, in the quasi-two-dimensional setup, low-frequency oscillations are present even in the convective regime. This suggests that they may play a significant role in the transition from a density inverted state to convection. Our hydrodynamic model shows that a sufficient condition for the existence of the low-frequency mode is an inverted density profile with distinct low and high density regions, a condition that may apply to other systems. Lastly, we also present experimental results that confirm the presence of the oscillations in a vast region of phase-space. Theory, experiments and simulations are seen to be in high agreement, specially for high energy inputs. 5:06PM L24.00008 Hopping dynamics of granular kinks1 , CLAUDIO FALCON, JUAN MACIAS, Univ. de Chile — We report on the experimental observation and theoretical characterization of the bifurcation diagram, dynamical properties and fluctuations of spatially modulated kinks in a shallow one-dimensional fluidized granular layer subjected to a periodic air flow. We show the appearance of these solutions as the layer undergoes a parametric instability. Due to the inherent fluctuations of the granular layer, the kink profile exhibits an effective wavelength, termed precursor, which modulates spatially the homogeneous states and drastically modifies the kink dynamics. We characterize the average and fluctuating properties of this solution. The long term evolution of these kinks is dominated by a hopping dynamics, related directly to the underlying spatial structure and inherent fluctuation. The properties of this motion can be described by a brownian particle in a symmetric periodic potential. Both the noise intensity of the brownian fluctuations and the amplitude and periodicity of the potential arise from the underlying precursor structure. 1 FONDECYT 1130354 5:19PM L24.00009 Connections between the Boson peak and the Van Hove Singularity– insights from the normal modes analysis of granular experiments , LING ZHANG, JIE ZHENG, JIE ZHANG, Shanghai Jiaotong University — We have experimentally measured the density of states (DOS) from the hexagonal lattice to the disordered structures in 2D packing of granular materials, which are made of photo-elastic disks allowing a precise measurement of contact forces between disks to determine the dynamical matrix of the system. Two different analyses have been performed with and without the inclusion of the rotational degree of freedom. By varying the pressure of the disordered crystal, we find the strong evidence that the first Van Hove singularity gradually evolves into the Boson peak. In geometrically disordered packing, the position of the Boson peak is influenced by the degree of the geometric disorder. Incorporating the rotational degree of freedom, two peaks would appear at the vicinity of the original first Van Hove singularity in the hexagonal lattice and similarly at the vicinity of the original Boson peak in a disordered crystal; in a contrast, the two peaks are nearly merged in a geometrically disordered system. Moreover, further analysis shows that the first peak is only related to the rotational degree of freedom, whereas the second peak is due to the coupling between the rotational and translational degrees of freedom. 5:32PM L24.00010 Transient behavior of granular materials with symmetric conditions for tumbler shapes and fill fractions , NICHOLAS POHLMAN, YUN SI, Northern Illinois University — The typical granular motion in circular tumblers is considered steady-state since there are no features to disrupt the top surface layer dimension. In polygon tumblers, however, the flowing layer is perpetually changing length, which creates unsteady conditions with corresponding change in the flow behavior. Prior work showed the minimization of free surface energy is independent of tumbler dimension, particle size, and rotation rate. This subsequent research reports on experiments where dimensional symmetry of the free surface in triangular and square tumblers with varying fill fractions do not necessarily produce the symmetric flow behaviors. Results of the quasi-2D tumbler experiment show that other dimensions aligned with gravity and the instantaneous free surface influence the phase when extrema for angle of repose and other flow features occur. The conclusion is that 50% fill fraction may produce geometric symmetry of dimensions, but the symmetry point of flow likely occurs at a lower fill fraction. 5:45PM L24.00011 Simulations of secondary currents in rapid granular chute flow , JOSEPH CALANTONI, Naval Research Lab, JUSTIN FINN, University of Liverpool, JULIAN SIMEONOV, SAMUEL BATEMAN, Naval Research Lab — The desire to understand granular flow as a fluid mechanical phenomena has long been the focus of theoreticians and experimentalists alike. Several analogies can be drawn with complex hydrodynamic behaviors at the continuum scale including Leidenfrost states, Rayleigh-Benard (R-B) convection, and Rayleigh-Taylor instability that allow for deeper understanding of collective granular motions. Here, we consider the case of longitudinal vortices in rapid granular flow down an inclined chute, and draw an analogy to hydraulic open channel flow. Previous experiments of rapid granular flow down inclined chutes have uncovered a unique regime where the flow exhibits stripes of slower and faster moving grains. We present results of molecular dynamics simulations that allowed us to study the full scale of the phenomena including smooth sidewalls and the rough bumpy bottom. We found that the secondary currents were intensified near the lateral walls. Monday, November 24, 2014 3:35PM - 5:58PM Session L25 Turbulence: Planetary Boundary Layers — 2005 - Catherine Gorle, Stanford University 3:35PM L25.00001 Coherency and Large Scale Motions in Turbulent Ekman Flow1 , CEDRICK ANSORGE, JUAN PEDRO MELLADO, Max Planck Institute for Meteorology — We study turbulence in the planetary boundary layer using direct numerical simulations of neutrally and stably stratified Ekman flow. The Reynolds number is varied in the range 500 < δ + < 1500 where δ + is the boundary layer thickness δ expressed in wall units. We vary the stratification, expressed in terms of a bulk Richardson number from very weak stability, where turbulence acts as a passive scalar, to very strong stability, where the flow relaminarizes partly. When the aspect ratio is sufficiently large, i.e. at a horizontal extent of about (20δ)2 , large-scale modes are present in the flow. These large-scale modes govern the spatio-temporal structure of external and global intermittency in the flow. We use a dual approach to investigate the large-scale motions and coherency in the flow: The analysis of spatially resolved fields of the turbulent flow is complemented by temporally fully resolved data at vertical intersections through the domain. From this data we quantify the expected error of local measurements carried out over a finite period of time with respect to the ensemble average. Moreover, we analyze the coherency in this complex flow, in particular of the large-scale structures occurring when the flow is exposed to a stable stratification. 1 Computing resources provided by Jülich Supercomputing Centre 3:48PM L25.00002 Grid-dependent Convection in WRF-LES , JASON SIMON, Department of Civil and Environmental Engineering, University of California, Berkeley, BOWEN ZHOU, Key Laboratory for Mesoscale Severe Weather/MOE, and School of Atmospheric Science, Nanjing University, FOTINI CHOW, Department of Civil and Environmental Engineering, University of California, Berkeley — Traditional numerical weather prediction (NWP) models parameterize the boundary layer with planetary boundary layer (PBL) schemes, which assume a coarse resolution so that energy-containing eddies are nearly exclusively sub-grid scale (SGS). Newer NWP models can also be used as large-eddy simulation (LES) models, which use a grid resolution that is sufficiently fine to resolve energy-containing eddies. For atmospheric flows the energy-containing eddies are typically on the scale of the PBL depth [O(1 km)]. The range of resolutions between the maximum appropriate resolution for LES and the minimum for PBL schemes is the turbulent gray zone, or terra incognita. The resolution limit for atmospheric LES is largely unexamined despite its dynamical significance. Here we examine the Weather Research and Forecasting model in LES mode (WRF-LES). We attempt to identify the symptoms of the turbulent gray zone with WRF-LES under primarily convective conditions using the Wangara Day 33 case. Grid-dependence, a signal of the gray zone, is evaluated by considering the stability profile, resolved convection, higher-order statistical profiles, and turbulence spectra. Also considered are the effects of isotropic mixing length-scales, domain extent and spatially heterogeneous surface fluxes. 4:01PM L25.00003 Coherent vorticity extraction in turbulent channel flow using anisotropic wavelets , KATSUNORI YOSHIMATSU, TELUO SAKURAI, Nagoya University, KAI SCHNEIDER, M2P2-CNRS and CMI, Aix-Marseille Universite, MARIE FARGE, LMD-IPSL-CNRS Ecole Normale Superieure, KOJI MORISHITA, Kobe University, TAKASHI ISHIHARA, Nagoya University — We examine the role of coherent vorticity in a turbulent channel flow. DNS data computed at friction-velocity based Reynolds number 320 is analyzed. The vorticity is decomposed using three-dimensional anisotropic orthogonal wavelets. Thresholding of the wavelet coefficients allows to extract the coherent vorticity, corresponding to few strong wavelet coefficients. It retains the vortex tubes of the turbulent flow. Turbulent statistics, e.g., energy, enstrophy and energy spectra, are close to those of the total flow. The nonlinear energy budgets are also found to be well preserved. The remaining incoherent part, represented by the large majority of the weak coefficients, corresponds to a structureless, i.e., a noise-like background flow. 4:14PM L25.00004 ABSTRACT WITHDRAWN — 4:27PM L25.00005 LES of turbulent boundary layer flow over urban-like roughness elements , TETSURO TAMURA, Tokyo Institute of Technology, MAKOTO TSUBOKURA, RIKEN, TSUYOSHI NOZU, Shimizu Corporation, KEIJI ONISHI, RIKEN — LES of turbulent boundary layer flow over urban-like roughness elements has been performed. Final goal of this paper is to elucidate the availability of LES on the wind flow within the canopy among buildings in cities. Firstly rectangular blocks, definitely larger than those on conventional rough wall such as grain or sand, are homogeneously arrayed and above-region equilibrium profiles of mean velocity and turbulent statistics are investigated. Also, in order to predict the fluctuating velocity characteristics of urban boundary layer, actual complicated-shaped buildings are used for reproducing the surface shape in cities. For numerical modeling, this study employs the unstructured-grid system where grid lines correctly fit to the building shape and BCM (Building Cube Method) which is formulated on very fine Cartesian mesh system. Based on the GIS data, BCM employs the external forcing technique named IBM (Immersed Boundary Method). Also, in BCM, computational process is so simple that the parallel algorithm and the memory access obtain the perfect efficiency. Using both the LES results, turbulence structures in the urban canopy are discussed. Appropriate 3D vortical structures can be recognized at inflow, along the street and among a pack of tall buildings. 4:40PM L25.00006 Turbulent boundary layer flow over distributions of cubes and evaluation of transient dynamics , WILLIAM ANDERSON, University of Texas at Dallas, QI LI, ELIE BOU-ZEID, Princeton University — We have used large-eddy simulation with an immersed boundary method to study turbulent flows over a distribution of uniform height, staggered cubes. The computational domain was designed such that both the roughness sublayer and a region of the aloft inertial layer was resolved. With this, we record vertical profiles of time series of fluctuations of streamwise velocity and vertical velocity (where fluctuation is computed as a quantity’s deviation from its time-averaged value during a time period over which the simulation exhibits statistical stationarity). Contour images of fluctuating velocity component shown relative to vertical position and time reveals an advective-lag between the passage of a high- or low-momentum region in the aloft inertial layer and excitation or relaxation of cube-scale coherent vortices in the sublayer. We quantify this advective lag and demonstrate how these events precede elevated Reynolds stresses associated with turbulent sweeps at the cube height. We propose that coherent, low and high momentum regions in the inertial layer are responsible for the reported advective lag. Vortex identification techniques are used to illustrate the presence of hairpin packets encapsulating low momentum regions, thereby supporting our hypothesis. Based on this, a simple, semi-empirical model for prediction of advective lag with height is developed. In spite of its simplicity, the model manages to capture the advective lag profiles reasonably well. 4:53PM L25.00007 Identifying turbulent coherent structures during LLJ events , VELAYUDHAN PRAJU KILIYANPILAKKIL, North Carolina State University, Raleigh, NC, GUILLERMO ARAYA, Texas Tech University, Lubbock, TX, SUKANTA BASU, North Carolina State University, Raleigh, NC, ARQUIMEDES RUIZ-COLUMBIÉ, WALTER GUTIERREZ, LUCIANO CASTILLO, Texas Tech University, Lubbock, TX, TEXAS TECH UNIVERSITY TEAM, NORTH CAROLINA STATE UNIVERSITY TEAM — Turbulent structures in the unstable atmospheric boundary layer have been extensively studied in the past. However our recent research show that the state-of-the art Weather Research & Forecasting model (WRF) model needs improvement in the simulation of nocturnal low level jet (LLJ) characteristics. Under these scenarios, the nocturnal stable boundary layer offers some gray areas to explore, particularly when conditions of high stability and strong vertical wind shear occur. Furthermore, the interactions of nighttime intermittent turbulence (high frequency) with coherent structures play an essential role in transport processes. In the present study, using wavelet analysis techniques, the WRF large-eddy simulation data are evaluated for coherent structure features during LLJ occurrences over the West Texas region. Those structural attributes will be compared to those observed by the high frequency (50 Hz) of the 200-meter meteorological tower (Reese, West Texas Mesonet). Additionally, the meteorological tower data are used to evaluate the influence of data acquisition frequency on small turbulent scale detection. 5:06PM L25.00008 Numerical prediction of pollutant dispersion and transport in an atmospheric boundary layer , STÉPHANIE ZEOLI, LAURENT BRICTEUX, University of Mons (UMONS) Belgium, MECH. ENG. DPT. TEAM — The ability to accurately predict concentration levels of air pollutant released from point sources is required in order to determine their environmental impact. A wall modeled large-eddy simulation (WMLES) of the ABL is performed using the OpenFoam based solver SOWFA (Churchfield and Lee, NREL). It uses Boussinesq approximation for buoyancy effects and takes into account Coriolis forces. A synthetic eddy method is proposed to properly model turbulence inlet velocity boundary conditions. This method will be compared with the standard pressure gradient forcing. WMLES are usually performed using a standard Smagorinsky model or its dynamic version. It is proposed here to investigate a subgrid scale (SGS) model with a better spectral behavior. To this end, a regularized variational multiscale (RVMs) model (Jeanmart and Winckelmans, 2007) is implemented together with standard wall function in order to preserve the dynamics of the large scales within the Ekman layer. The influence of the improved SGS model on the wind simulation and scalar transport will be discussed based on turbulence diagnostics. 5:19PM L25.00009 The vertical structure of eddy diffusivity in pure slope flows over smooth surfaces , MARCO GIOMETTO, JIANNONG FANG, School of Architecture, Civil and Environmental Engineering, École Polytechnique Fédérale de Lausanne, MARC B. PARLANGE, Faculty of Applied Sciences, University of British Columbia, EPFL / UBC COLLABORATION — Thermally driven slope flows are ubiquitous in nature and play a major role in regulating local microclimates in valleys, glaciers and ice-sheets. They control in large part surface momentum, heat and moisture fluxes, and their effects must be accounted for in weather prediction and climate models. Due to the interplay between shear and buoyancy in generating and destroying turbulence, the thermal and hydrodynamic boundary layers that characterize slope flows are very shallow, with turbulent motions of reduced size, when compared to those populating neutral and convective boundary layers. This poses serious difficulties in terms of computational resolution and modeling requirements for thermally driven slope flows. Not surprisingly, therefore, considerable effort has been devoted in designing parameterizations, for use in larger scale models. In this presentation we explore new parameterizations guided by direct numerical simulations (DNS) to determine the vertical profiles of momentum and buoyancy eddy diffusivities for pure slope flows over smooth surfaces. Mean flow profiles from one-dimensional models are then compared against DNS and results will be discussed. 5:32PM L25.00010 The evolution of large scale dense gas clouds at Jack Rabbit , PABLO HUQ, University of Delaware, TOM SPICER, University of Arkansas — Typically ammonia and chlorine are stored or transported as pressurized liquefied gas. There have been many accidents involving storage tanks and also accidents during transport. There is a need for accurate evaluation of the hazards associated with accidental releases of ammonia and chlorine which typically result in denser than air clouds which are toxic. The dense gas cloud slumps under the action of gravity into a thin layer with stable density gradients which suppress ambient atmospheric turbulence, and so complicating the physics of mixing. We present similarity analyses of one and two ton experimental releases of ammonia and chlorine at Jack Rabbit. Similarity analysis discriminates inertia-buoyancy and viscous-buoyancy regimes. Sequences of visualizations are used to determine propagation speeds of dense clouds. There is good agreement between observed speeds and the predictions of similarity analysis of the propagation of radial, dense gas clouds. Finally, comparison of one ton with two ton releases for both ammonia and chlorine lead to insights on scaling which are likely to be useful in the design of even larger scale experiments on dense gas clouds arising from similar configurations. 5:45PM L25.00011 Nonequilibrium Behavior of the Daytime Atmospheric Boundary Layer, from LES1 , BALAJI JAYARAMAN, JAMES BRASSEUR, Penn State U, TYLER MCCANDLESS, SUE HAUPT, NCAR — LES of the daytime atmospheric boundary layer (ABL) over flat topography is universally developed as an equilibrium ABL with steady surface heat flux Q0 and steady unidirectional “geostrophic” wind vector Vg above a capping inversion, where Vg also defines a spatially uniform transverse mean pressure gradient. The LES approaches a quasiequilibrium state characterized statistically by the ratio of boundary layer depth to Obukhov length scale. In contrast, the true daytime ABL is driven by surface heat flux increases to peak mid-day and drops in the afternoon, and by mesoscale wind vectors Ug that change in magnitude and direction during the day. We study the consequences of mesoscale weather on ABL dynamics by forcing ABL LES with a WRF simulation of the Midwest during 3 days of frontal passage over Kansas. Assuming horizontal homogeneity, we derive the relationship between Ug and Vg and study ABL response with systematic variation in Q0 and the magnitude and direction of Ug . Interesting results include: (1) asymmetry nonequilibrium diurnal response of the ABL; (2) directional changes in surface layer winds relevant to wind turbine function; and (3) changes in ABL stability state arising solely from changes in the direction of Ug . 1 Supported by DOE. Computer resources by NSF/XSEDE. Monday, November 24, 2014 3:35PM - 6:11PM Session L26 Thermal Transport in Boundary Layers — 2007 - Johan Larsson, University of Maryland 3:35PM L26.00001 DNS of transcritical turbulent boundary layers at supercritical pressures under abrupt variations in thermodynamic properties1 , SOSHI KAWAI, Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency — In this talk, we first propose a numerical strategy that is robust and high-order accurate for enabling to simulate transcritical flows at supercritical pressures under abrupt variations in thermodynamic properties due to the real fluid effects. The method is based on introducing artificial density diffusion in a physically-consistent manner in order to capture the steep variation of thermodynamic properties in transcritical conditions robustly, while solving a pressure evolution equation to achieve pressure equilibrium at the transcritical interfaces. We then discuss the direct numerical simulation (DNS) of transcritical heated turbulent boundary layers on a zero-pressure-gradient flat plate at supercritical pressures. To the best of my knowledge, the present DNS is the first DNS of zero-pressure-gradient flat-plate transcritical turbulent boundary layer. The turbulent kinetic budget indicates that the compressibility effects (especially, pressure-dilatation correlation) are not negligible at the transcritical conditions even if the flow is subsonic. The unique and interesting interactions between the real fluid effects and wall turbulence, and their turbulence statistics, which have never been seen in the ideal-fluid turbulent boundary layers, are also discussed. 1 This work was supported in part by Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Young Scientists (A) KAKENHI 26709066 and the JAXA International Top Young Fellowship Program. 3:48PM L26.00002 Design and Validation of a Constant Wall Temperature Plate , DRUMMOND BILES, ALIREZA EBADI, ALLEN MA, CHRIS WHITE, Univ of New Hampshire — A thermally conductive constant temperature wall-plate has been constructed and wind tunnel validation tests of the wall-plate design have been performed. The wall-plate is a sectioned wall design, where each section is independently heated and controlled. Each section consists of an aluminum 6061 plate, an array of resistive heaters affixed to the bottom of the aluminum plate, and a calcium silicate holder used for thermal isolation. A 3 × 3 grid of embedded thermocouples in each aluminum plate are used to monitor wall temperature and for feedback control of wall heating. The streamwise (flow direction) length of each section increases with downstream position since the wall heat flux decreases with downstream position.The section components sit in a Delrin (acetal) frame, chosen for its low thermal conductivity and machinability. The wall-plate will be used to investigate thermal transport in non-equilibrium boundary layer flows. In this talk, we report on the validation tests performed to-date to investigate the aerodynamic and thermal performance of the wall-plate, and the capability of the controller to maintain the wall-plate at a pre-selected fixed temperature in steady and unsteady laminar boundary layer flow. 4:01PM L26.00003 Log-Law scaling of a convective boundary layer in an unstably stratified turbulent channel flow , ANDREA SCAGLIARINI, Department of Physics, University of Rome “Tor Vergata”, HALLDOR EINARSSON, ARMANN GYLFASON, School of Science and Engineering, Reykjavik University, FEDERICO TOSCHI, Department of Applied Physics, Eindhoven University of Technology — Turbulent convection is ubiquitous in a variety of natural and industrial flows. In particular, convective motions may play a role in sheared flows. In this work, we are concerned with the interplay of buoyancy and shear in the dynamical boundary layer structure. The lattice Boltzmann Method (LBM) is applied to study numerically an unstably-stratified, fully developed, turbulent channel flow, driven by a longitudinal pressure gradient and with an imposed transverse wall temperature difference along the direction of gravity. Spanning the friction Reynolds (Retau ≤ 205) and Rayleigh numbers (Ra ≤ 1.3 × 107 ) we could systematically study the influence of the convection on the boundary layer structure and mean profiles of flow quantities in the channel. Our focus is on providing physical understanding of the deviations observed from the logarithmic law of the wall due to the buoyant motions as well as providing a model of this behavior, and link with fundamental quantities of heat transfer in the convective channel flow. Our findings show that the introduction of an unstably stratified thermal field results in an effective drag increase in the channel flow, quantified in the logarithmic region by a modified log-law, with model parameters dependent on Ra, Retau . 4:14PM L26.00004 Characteristics of Spatiotemporally Homogenized Boundary Layers at Atmospheric Reentry-like Conditions1 , RHYS ULERICH, ROBERT MOSER, Univ of Texas, Austin — Turbulent boundary layers approxi- mating those found on the NASA Orion Multi-Purpose Crew Vehicle thermal protection system during atmospheric reentry from the International Space Station have been studied by direct numerical simulation using a “slow growth” spatiotemporal homogenization approach recently developed by Topalian et al. The two data sets generated were Mae ≈ 0.9 and 1.15 homogenized boundary layers possessing Reθ ≈ 382 and 531, respectively. Edge-to-wall temperature ratios were + approximately 4.15 and wall blowing velocities, vw = vw /uτ , were roughly 8 × 10−3 . The favorable pressure gradients had Pohlhausen parameters between 25 and 42. Nusselt numbers under 22 were observed. Small or negative displacement effects are evident. Near-wall vorticity fluctuations show qualitatively different profiles than observed by Spalart [J. Fluid Mech. 187 (1988)] or Guarini et al. [J. Fluid Mech. 414 (2000)] suggesting that the simulations have atypical structures perhaps as a consequence of wall blowing or the homogenization. 1 This material is based in part upon work supported by the Department of Energy [National Nuclear Security Administration] under Award Number [DE-FC52-08NA28615]. 4:27PM L26.00005 Effect of cooling on compressible wall-turbulence , ANDREW TRETTEL, JOHAN LARSSON, University of Maryland — The modifications of the inner layer turbulence due to significant wall-cooling in a perfect gas is studied through a sequence of direct numerical simulations of compressible channel flow. The thickening of the buffer layer due to the imposed viscosity-gradient is quantified, as is the shift in the log-law intercept of the van Driest transformed velocity. The modification of the near-wall turbulent length scales and how these change with wall distance is examined. Finally, alternatives to the classic van Driest (1951) transformation of the mean velocity profile are considered. 4:40PM L26.00006 Multilayer scaling of mean velocity and thermal fields of compressible turbulent boundary layer , WEITAO BI, BIN WU, Peking Univ, YOUSHENG ZHANG, Institute of Applied Physics and Computational Mathematics, FAZLE HUSSAIN, Department of Mechanical Engineering, Texas Tech University, ZHEN-SU SHE, Peking Univ — Recently, a symmetry based structural ensemble dynamics (SED) theory was proposed by She et al. for canonical wall bounded turbulent flows, yielding prediction of the mean velocity profile at an unprecedented accuracy (99%). Here, we extend the theory to compressible turbulent boundary layers (TBL) at supersonic and hypersonic Mach numbers. The flows are acquired by spatially evolving direct numerical simulations (DNS). A momentum mixing length displays a four layer structure and quantitatively obeys the dilation group invariance as for the incompressible TBL. In addition, a temperature mixing length behaves very similarly to the momentum mixing length when the wall is adiabatic, with a small difference in the scaling exponents in the buffer layer - consistent with the strong Reynolds analogy. The Lie group based formulization of the two mixing lengths yields a multilayer model for the turbulent Prandtl number, along with predictions to the mean thermal and velocity profiles, both in good agreement with the DNS. Thus, we assert that the compressible TBLs are governed by the same symmetry principle as that in the canonical wall bounded turbulent flows, and its mean fields can be accurately described by the SED theory. 4:53PM L26.00007 Turbulent heat flux measurements in thermally stable boundary layers1 , OWEN J. WILLIAMS, TYLER VAN BUREN, Princeton University, ALEXANDER J. SMITS, Princeton University, Monash University — Thermally stable turbulent boundary layers are prevalent in the polar regions and nocturnal atmospheric surface layer but heat and momentum flux measurements in such flow are often difficult. Here, a new method is employed using a nanoscale cold-wire (T-NSTAP) adjacent to a 2D PIV light sheet to measure these fluxes within rough-wall turbulent boundary layer. This method combines the advantages of fast thermal frequency response with measurement of the spatial variation of the velocity field. Resolution is limited solely by the separation of the probe and the light sheet. The new technique is used to examine the applicability of Monin-Obukhov similarity over a range of Richardson numbers from weak to strongly stable. In addition, the velocity fields are conditionally averaged subject to strong deviations of temperature above and below the local average in an effort to determine the relationship between the coherent turbulent motions and the fluctuating temperature field. 1 This work was supported by the Princeton University Cooperative Institute for Climate Science 5:06PM L26.00008 Thermal Transport Phenomena in the Quasi-laminarization Process of Turbulent Boundary Layers , LUCIANO CASTILLO, GUILLERMO ARAYA, FAZLE HUSSAIN, Texas Tech University — Direct Numerical Simulation of a spatially evolving turbulent boundary layer subject to strong favorable pressure gradient (SFPG) with eventual quasi-laminarization has been performed to include the thermal field. In this talk, the following questions will be addressed: i) to which extend the Reynolds analogy is satisfied during flow laminarization? ii) can the thermal boundary layer under quasi-laminarization be described as two quasi-independent inner/outer regions? To the best of our knowledge, documented investigation regarding heat transfer phenomena associated with quasi-laminarization process of spatially developing boundary layers is rather scarce. In order to introduce realistic thermal inlet fluctuations in a SFPG we employed the multi-scale technique for thermal boundary layers devised by Araya and Castillo PoF (2013). Such methodology enable us to more effective capture the pressure gradient and thermal field than using a single scaling approach. It is shown that the Reynolds normal stresses for the streamwise component remains frozen in space while the wall-normal and spanwise components continue to decrease as the flow moves downstream and never becomes laminar due to the survival of the upstream turbulence during dissipation on viscous time scales. 5:19PM L26.00009 Effect of Small Roughness Elements on Thermal Statistics of Turbulent Boundary Layer at Moderate Reynolds Number , ALI DOOSTTALAB, GUILLERMO ARAYA, Texas Tech University, RONALD ADRIAN, Arizona State University, LUCIANO CASTILLO, Texas Tech University — DNS simulations of the zero pressure gradient turbulent boundary layer subject to forced convection over a transitionally rough surface with k+ ≈ 11 and Reynolds numbers based on momentum thickness of 2400, are presented for the first time. Prescribing realistic turbulent inlet conditions for rough surface and spatial evolving flows is extremely challenging. In order to solve the computationally intensive simulations, a dynamic method proposed by G. Araya et al. (JFM, vol. 670, pp. 581-605, 2011) for prescribing realistic inflow boundary conditions is used for simulations of spatially developing thermal turbulent boundary layers. Preliminary results showed how a transitionally rough surface alters thermal statistics in the inner and outer layers. Based on variations of Cf and St, the validity of Reynolds analogy was tested and confirmed in the rough case and an increase to isotropy was obsereved. Furthermore, it was found that roughness enhances wall-normal heat flux transport in the inner layer and reduce it in the outer region of boundary layer. In addition. The rough surface decreased the ratio of Reynolds stress to turbulent heat flux in the near wall region, leading to a decreased turbulent Prandtl number. 5:32PM L26.00010 An Integral Method to Evaluate Wall Heat Flux in Oscillatory WallBounded Flow , ALIREZA EBADI, CHRISTOPHER WHITE, University of New Hampshire, IAN POND, YVES DUBIEF, University of Vermont — An integral method to evaluate wall heat flux in oscillatory wall-bounded flow is presented. The method is mathematically exact and has the advantage of having no explicit streamwise gradient terms. Importantly, the mathematical exactness of the method allows for a direct measurement of the wall heat flux without any a priori assumptions regarding the flow field or invoking flow transport analogies. It is useful in cases when measurements at multiple streamwise locations are not available or feasible, for flows with ill-defined outer boundary conditions, or when the measurement grid does not extend over the whole boundary layer thickness. The method is validated using DNS datasets of reciprocating turbulent channel flow with heat transfer for which independent estimates of wall heat flux were known, and the different results compare favorably. Complications owing to experimental limitations and measurement error in determining wall heat flux from the proposed method are presented, and mitigating strategies are described. 5:45PM L26.00011 Connecting the classical limits: the Graetz-Nusselt problem for partial, homogeneous slip1 , ROB LAMMERTINK, SANDER HAASE, University of Twente, JON CHAPMAN, University of Oxford, PEICHUN TSAI, DETLEF LOHSE, University of Twente — The classical Graetz-Nusselt problem concerns the transport of heat between a hydrodynamically fully developed flow and the wall of a cylindrical pipe at constant temperature. In the thermally developing regime, the Nusselt number scales as Nu ∝ Gz−β , where Gz = RePrD/L is the Graetz number. In case of a non-slippery wall β = 1/3, whereas for no-shear surfaces β = 1/2. The generally assumed no-slip boundary condition does not always hold. Intrinsic slip lengths in micro- and nanofluidic systems vary from nearly zero to almost infinity. Here we studied the Graetz-Nusselt problem for partial slip. We present a solution for the Graetz-Nusselt problem for partial slip, connecting the two classical solutions. We show numerically and analytically that for surfaces displaying partial slip, β gradually changes from 1/3 to 1/2. Also the developed Nusselt number Nu∞ slowly changes value from 3.66 to 5.78. We provide a mathematical and physical explanation for these two transitions points, which are separated more than one decade apart for β and Nu∞ . 1 Funding from ERC (starting grant R.G.H. Lammertink) is greatly acknowledged 5:58PM L26.00012 Thermo-fluid-dynamics of turbulent boundary layer over a moving continuous flat sheet in a parallel free stream , BUSHRA AFZAL, Corning Incorporated, NOOR AFZAL TEAM1 , BUSHRA AFZAL TEAM2 — The momentum and thermal turbulent boundary layers over a continuous moving sheet subjected to a free stream have been analyzed in two layers (inner wall and outer wake) theory at large Reynolds number. The present work is based on open Reynolds equations of momentum and heat transfer without any closure model say, like eddy viscosity or mixing length etc. The matching of inner and outer layers has been carried out by Izakson-Millikan-Kolmogorov hypothesis. The matching for velocity and temperature profiles yields the logarithmic laws and power laws in overlap region of inner and outer layers, along with friction factor and heat transfer laws. The uniformly valid solution for velocity, Reynolds shear stress, temperature and thermal Reynolds heat flux have been proposed by introducing the outer wake functions due to momentum and thermal boundary layers. The comparison with experimental data for velocity profile, temperature profile, skin friction and heat transfer are presented. In outer non-linear layers, the lowest order momentum and thermal boundary layer equations have also been analyses by using eddy viscosity closure model, and results are compared with experimental data. 1 Retired 2 Corning Professor, Embassy Hotel, Rasal Ganj, Aligarh 202001 India. Incorporated Monday, November 24, 2014 3:35PM - 6:11PM Session L27 Turbulence: RANS & Hybrid Modeling — 2009 - Svetlana Poroseva, University of New Mexico 3:35PM L27.00001 The determination of turbulence-model statistics from the velocityacceleration correlation , STEPHEN POPE, Cornell University — In Reynolds-stress models, a primary unknown is the pressure–rate-of-strain; and, in velocity probability density function (PDF) models, a primary unknown is the conditional mean pressure gradient (conditional on velocity). Except from direct numerical simulations (DNS) of simple canonical flows, there is little information about these statistics. Currently, it is not possible to measure pressure with the necessary resolution, so there are no measurements of these important quantities. It is shown that essentially the same information can be obtained from the velocity-acceleration correlation and the Reynolds stresses. Since these correlations arise predominantly from the larger, energy-containing motions, they can be obtained experimentally (without Kolomorov-scale resolution), and from DNS, and from well-resolved large-eddy simulations (LES). In terms of the second moments of velocity and acceleration, expressions are given for the redistribution term in the Reynolds-stress equation, and for the drift term in the generalized Langevin model for the PDF. 3:48PM L27.00002 New Models for Velocity/Pressure-Gradient Correlations in Turbulent Boundary Layers1 , SVETLANA POROSEVA, University of New Mexico, SCOTT MURMAN, NASA Ames Research Center — To improve the performance of Reynolds-Averaged Navier-Stokes (RANS) turbulence models, one has to improve the accuracy of models for three physical processes: turbulent diffusion, interaction of turbulent pressure and velocity fluctuation fields, and dissipative processes. The accuracy of modeling the turbulent diffusion depends on the order of a statistical closure chosen as a basis for a RANS model. When the Gram-Charlier series expansions for the velocity correlations are used to close the set of RANS equations, no assumption on Gaussian turbulence is invoked and no unknown model coefficients are introduced into the modeled equations. In such a way, this closure procedure reduces the modeling uncertainty of fourth-order RANS (FORANS) closures. Experimental and direct numerical simulation data confirmed the validity of using the Gram-Charlier series expansions in various flows including boundary layers. We will address modeling the velocity/pressure-gradient correlations. New linear models will be introduced for the second- and higher-order correlations applicable to two-dimensional incompressible wall-bounded flows. Results of models’ validation with DNS data in a channel flow and in a zero-pressure gradient boundary layer over a flat plate will be demonstrated. 1A part of the material is based upon work supported by NASA under award NNX12AJ61A. 4:01PM L27.00003 Investigation of the pressure-strain-rate correlation using high-resolution LES of the atmospheric boundary layer1 , KHUONG NGUYEN, Clemson University, MARTIN OTTE, US Environmental Protection Agency, EDWARD PATTON, PETER SULLIVAN, National Center for Atmospheric Research, CHENNING TONG, Clemson University — We analyze the pressure-strain term in the Reynolds stress transport equation using large-eddy simulations of the atmospheric boundary layer (ABL). The simulations are implemented on computational meshes varying from 2563 to 10243 grid points and employ several different SGS closures (Smagorinsky 1963; Sullivan et al. 1994; Kosovic 1997). The results highlight the influence of both shear and buoyancy on the pressure-strain-rate correlation. In the neutral (shear dominated) ABL, the behavior of the pressure-strain-rate correlation predicted by the Smagorinsky and Kosovic SGS models are consistent with the log-layer scaling and DNS results. In the strongly convective ABL, all three models predict behaviors for the pressure-strain-rate correlation that are consistent with the mixed- (outer-) layer scaling and field measurements. In cases where both shear and buoyancy are important, the highest-resolution runs are able to predict a combination of the log-layer scaling (near the wall) and the mixed-layer scaling (away from the wall), whereas the coarser-resolution runs are unable to capture this transition. The results are potentially useful for both Reynolds stress models and transport-equation-based SGS models for the convective atmospheric boundary layer. 1 Funded by NSF 4:14PM L27.00004 Formulation and calibration of a stochastic model form error representation for RANS , TODD OLIVER, BRYAN REUTER, ROBERT MOSER, The University of Texas at Austin — It is well-known that RANS turbulence models fail to accurately represent the effects of turbulence on the mean flow for many important flows. We consider probabilistic representations of this model inadequacy for wall-bounded flows. The particular probabilistic representations considered here take the form of stochastic differential equations that are loosely based on the Reynolds stress transport equations, but include random forcing to represent uncertainty due to the closure problem. This model is disretized using finite elements and a priori uncertainty quantification studies are conducted using Monte Carlo sampling. The results demonstrate that the resulting uncertainties in the mean velocity scale as desired with Reynolds number. In addition to the random forcing, the model contains a number of uncertain parameters. We demonstrate that these can be calibrated using available DNS data. The model is further tested via comparison against additional DNS data outside of the orignal calibration set. 4:27PM L27.00005 Low-Order Models For Assessing RANS Closures , DANIEL ISRAEL, Los Alamos National Laboratory — Historically, most coefficients in RANS models have been calibrated to match the growth rates of certain canonical self-similar flows. However, for many of these flows, the growth rates observed in experiments and DNS vary widely. In fact, George (1986) argues that a universal self-similar solution does not exist. This would imply that RANS model calibration is specific to a particular experiment. Using classical integral methods to reduce RANS models to ODEs, it is possible to obtain a low-order dynamical system which can be used to study the approach to self-similarity for the model. Comparing the trajectory maps for such low-order models to data suggests that most, if not all, of the discrepancy between different experiments can be explained by transient deviations from self-similarity, and that there is indeed a universal self-similar behavior. Furthermore, such trajectory maps can be used to assess how well transient behavior due to the initial conditions in RANS calculations captures the experimentally observed flow physics. 4:40PM L27.00006 An intermittency model for predicting roughness induced transition1 , XUAN GE, PAUL DURBIN, Iowa State University — An extended model for roughness-induced transition is proposed based on an intermittency transport equation for RANS modeling formulated in local variables. To predict roughness effects in the fully turbulent boundary layer, published boundary conditions for k and ω are used, which depend on the equivalent sand grain roughness height, and account for the effective displacement of wall distance origin. Similarly in our approach, wall distance in the transition model for smooth surfaces is modified by an effective origin, which depends on roughness. Flat plate test cases are computed to show that the proposed model is able to predict the transition onset in agreement with a data correlation of transition location versus roughness height, Reynolds number, and inlet turbulence intensity. Experimental data for a turbine cascade are compared with the predicted results to validate the applicability of the proposed model. 1 Supported by NSF Award Number 1228195 4:53PM L27.00007 Near Wall Treatment of the Variable Resolution Partially Averaged NavierStokes Model , POOYAN RAZI, Graduate Student, SHARATH GIRIMAJI, Professor — The objective of this work is to develop appropriate turbulence closures for bridging between different resolutions in the near-wall region. The development is made in the context of partially-averaged Navier-Stokes (PANS) method. Seamless transition from region of low-resolution near the wall to high-resolution away from the wall is controlled using the PANS filter parameter. The resolution variation introduces commutation effects which are modeled using additional terms in the turbulent kinetic energy equation. In addition, to conserve the total turbulent energy due to the interaction of unresolved and resolved flow fields, innovative strategies are evaluated for channel flow as well as flat plate boundary layer. This study identifies some important challenges regarding the numerical stability and appropriate implementation of the energy conservation principles. The preliminary results are shown to be encouraging. 5:06PM L27.00008 Dynamic DDES On DES Type Grid , ZIFEI YIN, PAUL DURBIN, None — A dynamic procedure allows a DES formulation that we developed to adjust CDES for different flow configurations. Similarly to the dynamic Smagorinsky model, the grid is required to be fine enough to resolve a significant portion of the inertial range. In some cases, that requirement conflicts with the goal of DES to cut down computing cost. The current effort is therefore to determine a proper CDES value by approximately recovering some unresolved small scales from primary, filtered solution. Repeated test filtering is adopted here to compute the approximation of the unfiltered solution. The formulation is based on the dynamic l2 w DDES model and different geometries with varies grid resolution are tested to determine the applicability of proposed formultion on DES type grids. 5:19PM L27.00009 New hybrid turbulence modelling approach, with application to dynamic stall control , SIGFRIED HAERING, ROBERT MOSER, University of Texas at Austin — We present numerical studies of a stalled airfoil experiencing transitory flow control using a new hybrid RANS/LES modeling approach developed specifically for such challenging flow scenarios. Traditional hybrid approaches exhibit deficiencies when used for fluctuating smooth-wall separation and reattachment necessitating ad-hoc delaying functions and model tuning making them no longer useful as a predictive tool. Additionally, complex geometries and flows often require high cell aspect-ratios and large grid gradients as a compromise between resolution and cost. Such transitions and inconsistencies in resolution detrimentally effect the fidelity of the simulation. Our approach more naturally transitions between RANS to LES obviating the need for tuning and directly accounts for anisotropy and inhomogeneity in the flow and grid. The results of these simulations not only provide fundamental insight into experimentally observed stall control mechanisms but also display the versatility and accuracy of the new modeling method in simulating complex flow phenomena. 5:32PM L27.00010 Scale dependence of Reynolds stress transport in wall-bounded turbulence at Reτ = 52001 , MYOUNGKYU LEE, ROBERT D. MOSER, University of Texas at Austin — A direct numerical simulation (DNS) of turbulent channel flow has been performed to study high Reynolds number wall-bounded turbulence. In particular, in this talk we will focus on the characteristics of the terms in the Reynolds stress transport equations in two recent channel flow DNS at Reτ = 1000 and 5200. The Reτ = 5200 case is at sufficiently high Reynolds number for there to be a significant scale separation between the near-wall and outer layer turbulence. A spectral analysis of the Reynolds stress transport terms shows how the inner- and outer-layer turbulence interact across scale. One striking result of this analysis is that over a broad range of y, the turbulent transport of turbulent kinetic energy occurs at scales that are proportional to y. There is also a weak direct interaction between the outer-layer and near-wall turbulence at large scales, presumably resulting in the large-scale modulation of near-wall turbulence. Further results from this spectral Reynolds stress transport analysis will be presented to explore the characteristics of turbulent, viscous and pressure effects. 1 This work was supported by NSF (OCI-0749223 and PRAC Grant 0832634), and computation resources were provided by the Argonne Leadership Computing Facility through the Early Science, INCITE 2013 and Directors Discretionary Programs. 5:45PM L27.00011 A revisit of the equilibrium assumption for prediction of near-wall turbulence1 , FARID KARIMPOUR, SUBHAS VENAYAGAMOORTHY, Colorado State University — Assuming equilibrium between the rates of pro- duction (P ) and dissipation (ǫ) of the turbulent kinetic energy (k) is widely employed for prediction and modeling of turbulent flows. In this study, we revisit the consequence of using equilibrium assumption for prediction of near-wall turbulence. To this end, the relevant scales inherent in the turbulent viscosity (νt ) formulation of the standard k-ǫ model is derived. We show that such turbulent viscosity formulations are not suitable for modeling near-wall turbulence. Furthermore, by using the turbulent viscosity (νt ) formulation suggested by Durbin, we also show that the anisotropic Reynolds stress is correlated with the wall-normal, isotropic Reynolds stress. ‘A priori’ tests are performed to assess the validity of the propositions using the direct numerical simulation (DNS) data of unstratified channel flow. The comparisons with the data are excellent and confirm our findings. 1 Funded by the National Science Foundation 5:58PM L27.00012 Evaluation of turbulence models for prediction of separated turbulent boundary layer under unsteady adverse pressure gradients1 , JUNSHIN PARK, DONGHYUN YOU, POSTECH — Predicitive capabilites of Reynolds-averaged Navier-Stokes (RANS) techniques for separated flow under unsteady adverse pressure gradients have been assessed using SST k − ω model and Spalart-Allmaras model by comparing their results with direct numerical simulation (DNS) results. Both DNS and RANS have been conducted with a zero pressure gradient, a steady adverse pressure gradient, and an unsteady adverse pressure gradient, respectively. Comparative studies show that both RANS models predict earlier separation and fuller velocity profiles at the reattachment zone than DNS in the unsteady case, while reasonable agreements with DNS are observed for steady counterparts. Causes for differences in the predictive capability of RANS for steady and unsteady cases, are explained by examining the Reynolds stress term and eddy viscosity term in detail. The Reynolds stress and eddy viscosity are under-predicted by both RANS models in the unsteady case. The origin of the under-prediction of the Reynolds stress with both RANS models is revealed by investigating Reynolds stress budget terms obtained from DNS. 1 Supported by the National Research Foundation of Korea Grant NRF- 2012R1A1A2003699 and the Brain Korea 21+ program Monday, November 24, 2014 3:35PM - 6:11PM Session L28 Turbulence: Jets and Wakes — 2011 - Roberto Paoli, CERFACS 3:35PM L28.00001 Large-eddy simulations of turbulent plane and radial wall-jets , RAYHANEH BANYASSADY, UGO PIOMELLI, Queen’s University — Large-eddy simulations of turbulent plane and radial wall-jets were conducted at different Reynolds numbers. The results were validated with the available experimental data. The radial wall-jets decay faster compared to the plane ones, due to the extra expansion in the azimuthal direction. This causes the pressure-gradient distributions to be different in radial and plane wall-jets (e.g. the inner layer in the plane case is under a favorable pressure-gradient, while in the radial case it subjected to an adverse pressure-gradient). However, these pressure gradients are not strong enough to cause any structural difference between plane and radial wall-jets. In both cases, the local Reynolds number (based on the local maximum velocity and local boundary-layer thickness) is an important determining factor in characterization of the flow. The joint probability-density function analysis shows that the local Reynolds number determines the level of intrusion of the outer layer into the inner layer: the lower the local Reynolds number the stronger is the interaction of inner and outer layers. These results were used to clarify some of the observations reported in literature; as an example the scatter of the reported log-law constants can be explained using the above-mentioned results. 3:48PM L28.00002 Large eddy simulations of a Mach 0.9 jet with fully-turbulent nozzle-exit boundary layer1 , GUILLAUME BRES, FRANK HAM, Cascade Technologies Inc., PETER JORDAN, Institut Pprime — From past studies, it is well known that the state of the nozzle-exit boundary layer is a key parameter for the flow development and noise characteristics of a jet. However, because of the computational cost of simulating high Reynolds number wall-driven turbulence, the nozzle boundary layer is typically assumed to be laminar or weakly disturbed in most jet simulations. This approach often leads to enhanced laminar to turbulent shear-layer transition and increased noise due to vortex pairing. In the present work, large eddy simulations of an isothermal Mach 0.9 jet (Re = 1E6) issued from a convergent-straight nozzle are performed using the compressible flow solver “Charles” developed at Cascade Technologies. Localized adaptive mesh refinement, synthetic turbulence and wall modeling are used inside the nozzle to ensure fully turbulent profiles at the nozzle exit. This resulted in significant improvements for the flowfield and sound predictions, compared to the typical approach based on laminar flow assumption in the nozzle. The far-field noise spectra now remarkably match the measurements from the companion experiment conducted at Pprime Institute, within 0.5 dB for most angles and relevant frequencies. As a next step toward better understanding of jet noise, the large transient database collected during the simulation is currently being mined using reduced order modeling and wavepacket analysis. 1 Work supported in part by NAVAIR. Computer allocation provided by DoD HPC centers at ERDC and AFRL. 4:01PM L28.00003 Numerical study of high-speed turbulent jets in crossflow , PRAHLADH IYER, XIAOCHUAN CHAI, KRISHNAN MAHESH, University of Minnesota — Large-eddy Simulation (LES) is used to study (i) a sonic jet injected into a supersonic crossflow, and (ii) a supersonic jet injected into a subsonic crossflow, whose conditions are based on experiments by Santiago et al. (1997) and Beresh et al. (2005) respectively. An unstructured, finite volume compressible solver (Park & Mahesh 2007) along with the Dynamic Smagorinsky Model (DSM) (Moin et al. 1991) is used in the simulations. Qualitative and quantitative comparison with experiment show good agreement for both flows. Dynamic Mode Decomposition (DMD) of the three-dimensional flow field is performed to identify dominant frequencies and their corresponding flow features. 4:14PM L28.00004 Assessing grid resolution effects in large-eddy simulations of a jet-in-crossflow1 , ANTHONY RUIZ, GUILHEM LACAZE, JOSEPH OEFELEIN, Sandia National Laboratories — Calculations using the Large Eddy Simulation technique are conducted over a range of resolutions in a Jet In Cross Flow. The configuration corresponds to the experiment of Su and Mungal (2004). A turbulent jet at a Reynolds Number of 5,000 is injected into a laminar cross flow, with a jet to cross-flow velocity ratio of 5.7. Resolutions are varied from typical engineering spatial and temporal resolutions to near-DNS resolution. The near-DNS resolution has been extensively validated against experimental results in a previous study [Ruiz et al., Phys. Fluid (2014)], which also focused on the evolution of the major scales of turbulence. The grid resolution is coarsened to investigate the impact on the topology of coherent structures and mixing. Fourier analysis is conducted at all resolutions to observe the impact of filtering on the turbulent energy cascade, identify the main hydrodynamic frequencies, and determine the degree to which these are grid-dependent. Once hydrodynamics modes are known, phase-locked analysis of the flow field shows the spatial structure of these modes as a function of resolution. This enables a clear understanding of the impact of resolution on the flow. 1 The U. S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences supported this work. 4:27PM L28.00005 The Far Field Structure of a Jet in Cross-Flow , NICOLAS LANITIS, Department of Engineering, University of Cambridge, Cambridge CB2 1PZ, UK, JAMES DAWSON , Department of Energy and Process Engineering, Norwegian University of Science and Technology, 7491 Trondheim, Norway — Stereoscopic PIV measurements were performed in the far field of a cross-flow jet. Measurements were taken in a water channel in the spanwise-wall normal plane (y-z) containing the Counter-Rotating vortex pair (CVP). The jet’s Reynolds number was Rejet = 2 × 104 and had an exit diameter of dj = 4mm. Measurements were taken for a jet to cross-flow velocity ratio of Vr = 10 at three downstream positions of x/dj = 30, 55, 85 and for a Vr = 15, 20 at x/dj = 85. Two point spatial correlations hint at the presence of arch shaped structures titled in the streamwise x-direction on the windward side of the CVP as well as straight vortex tubes extending into the wake. The arched shaped structure is compounded by PDFs of the location of streamwise vorticity peaks (vortex tubes) in the instantaneous field indicating the presence of a vortex structure aligned in the spanwise direction. This information together with the use of High Speed Stereoscopic PIV and Taylor’s Hypothesis, which allowed for the extraction of 3D structures, led to the development of an eddy model comprised of hairpin, roller and wake structures to predict turbulence statistics of a jet in cross-flow. 4:40PM L28.00006 Amplitude and frequency modulation in a turbulent jet at high Reynolds number , DANIELE FISCALETTI, Delft Univ of Tech, BHARATHRAM GANAPATHISUBRAMANI, University of Southampton, GERRIT ELSINGA, Delft Univ of Tech — In this work, the amplitude and frequency modulation of the small scales of turbulence is investigated experimentally in a jet at high Reynolds number. Hot-wire anemometry (HWA) and long-range µPIV measurements are performed in the fully developed region of the jet. From HWA, time series are converted into space series by applying the Taylor hypothesis. Using spectral filters, two signals representative of the large, and the small scales are constructed. It was found that for positive large-scale fluctuations, the associated small-scale signal is stronger in amplitude (amplitude modulation), and presents locally a higher number of local maxima and minima (frequency modulation). Moreover, the local standard deviation of the small-scale signal (representative of the amplitude of the small-scale signal) increases with the local strength of the large-scale fluctuations. A further investigation with PIV allowed to resolve the small scales of turbulence, without the need for Taylor hypothesis. From this analysis, the amount of amplitude modulation was found to be only 25% of the value obtained with HWA. This difference can be explained considering that the structures of intense vorticity travel, on average, at velocities higher than the mean velocity of the flow. 4:53PM L28.00007 Pressure ratio effects on self-similar scalar mixing of high-pressure turbulent jets in a pressurized volume , ADAM RUGGLES, LYLE PICKETT, JONATHAN FRANK, Sandia National Laboratories — Many real world combustion devices model fuel scalar mixing by assuming the self-similar argument established in atmospheric free jets. This allows simple prediction of the mean and rms fuel scalar fields to describe the mixing. This approach has been adopted in super critical liquid injections found in diesel engines where the liquid behaves as a dense fluid. The effect of pressure ratio (injection to ambient) when the ambient is greater than atmospheric pressure, upon the self-similar collapse has not been well characterized, particularly the effect upon mixing constants, jet spreading rates, and virtual origins. Changes in these self-similar parameters control the reproduction of the scalar mixing statistics. This experiment investigates the steady state mixing of high pressure ethylene jets in a pressurized pure nitrogen environment for various pressure ratios and jet orifice diameters. Quantitative laser Rayleigh scattering imaging was performed utilizing a calibration procedure to account for the pressure effects upon scattering interference within the high-pressure vessel. 5:06PM L28.00008 Numerical simulation of the flow field from a radially lobed nozzle and validation via HWA , NOUSHIN AMINI, Texas A&M University, AARTHI SEKARAN, Jawaharlal Nehru Centre for Advanced Scientific Research — With a constant need for higher performance and efficiency in engineering (particularly aerospace) applications, lobed nozzles have experienced a regained interest in the recent past, owing to their superior mixing capabilities. Although previous experimental studies (Hu et al 1999, Hu et al 2008) have analyzed the flow field from lobed nozzles and made conjectures about the physics and flow mechanisms involved, the absence of a “complete” 3D dataset elicits unanswered questions. The present numerical study is intended as a complement to an existing experimental (single component hot wire anemometry) investigation, involving the analysis of the flow field downstream of a six lobed nozzle (N. Amini et al, 2012). A full 3D URANS simulation of the lobed nozzle is carried out, initially validated with experimental data, and then used to examine the stream-wise vortices and obtain a visual corroboration of the structure formation and breakup mechanism as described earlier (Hu et al, 2008). Further, the study takes a close look at the nature of the instabilities which trigger and enhance the mixing process in lobed nozzles in order to determine the precise role of the lobes and eventually obtain more effective mixing in industrial applications. 5:19PM L28.00009 Large-eddy simulations of a solid-rocket booster jet , ROBERTO PAOLI, ADELE POUBEAU, DANIEL CARIOLLE, CERFACS — Emissions from solid-rocket boosters are responsible for a severe decrease in ozone concentration in the rocket plume during the first hours after a launch. The main source of ozone depletion is due to hydrogen chloride that is converted into chlorine in the high temperature regions of the jet (afterburning). The objective of this study is to evaluate the active chlorine concentration in the plume of a solid-rocket booster using large-eddy simulations. The gas is injected through the entire nozzle of the booster and a local time-stepping method based on coupling multi-instances of a fluid solver is used to extend the computational domain up to 600 nozzle exit diameters. The methodology is validated for a non-reactive case by analyzing the flow characteristics of supersonic co-flowing under expanded jets. Then, the chemistry of chlorine is studied offline using a complex chemistry solver and the LES data extracted from the mean trajectories of sample fluid particles. Finally, the online chemistry is analyzed by means of the multispecies version of the LES solver using a reduced chemistry scheme. The LES are able to capture the mixing of the exhaust with ambient air and the species concentrations, which is also useful to initialize atmospheric simulations on larger domains. 5:32PM L28.00010 Direct numerical simulations of turbulent wakes with non-equilibrium similarity scalings , THIBAULT DAIRAY, JOHN CHRISTOS VASSILICOS, Imperial College London — Recently, turbulent flow regions with dissipation scalings incompatible with equilibrium Richardson-Kolmogorov phenomenology have been discovered in the lee of regular and fractal grids. Considering the non-equilibrium dissipation law with a similarity analysis, new scaling laws have recently been obtained for the streamwise evolution of the centreline wake mean profiles (PRL 111, 144503 (2013)). In the present study, DNS of spatially evolving wakes generated by bluff plates with both simple square and irregular edge peripheries (the latter allowing the formation of jet-wake flows) have been carried out using the in-house code Incompact3d. The Reynolds number based on the plate length L, equal to the square-root of the plate area, and the freestream velocity is 5000. The self-similarity of the mean flow, Reynolds-stresses and the dissipation rate of turbulent kinetic energy have been analysed as well as the scaling laws. In the region where the flow is found to be axisymmetric and self-similar, the viscous term in the momentum equation is two orders of magnitude smaller than the other terms and the mean flow profile evolves in accordance with the non-equilibrium law up to 100L. Furthermore, the non-equilibrium dissipation law is observed for both regular and irregular plates. 5:45PM L28.00011 Asymmetries in the high Reynolds number wake of a submarine model in pitch , ANAND ASHOK, TYLER VAN BUREN, Princeton University, ALEXANDER SMITS, Princeton University and Monash University — Experiments are reported in the wake of a submarine model (DARPA SUBOFF) over a wide range of Reynolds numbers based on the length ReL between 105 and 30 x 106 at a pitch angle of 8 degrees. Two-component velocity measurements were taken at five cross-stream planes, downstream of the stern of the model (2<x/D<14), using hot wire anemometry. The wake is distinguished by two principal vortex structures, but the strength of the two vortices are not equal, leading to an asymmetric wake that slowly rotates. The asymmetry appears to be endemic, and is not affected by freestream turbulence, changes in tripping, surface roughness, and small angles of yaw. They persist across all the Reynolds numbers measured, and the effects of Reynolds number are only important at low Reynolds number in that the wake becomes independent of Reynolds number when ReL ≥ 4.8 x 106 . This work was supported under ONR Grant N00014-13-1-0174 (Ron Joslin). 5:58PM L28.00012 Steady imperfect bifurcation with generic 3D bluff bodies at large Reynolds numbers , OLIVIER CADOT, ENSTA-ParisTech, LUC PASTUR, LIMSI, ANTOINE EVRARD, GUILLAUME SOYER, ENSTA-ParisTech — The turbulent wake of parallelepiped bodies exhibits a strong bi-modal behavior. The wake randomly undergoes symmetry breaking reversals, between two mirror asymmetric steady modes (RSB modes). The characteristic time for reversals is about 2 or three orders of magnitudes larger than the natural time for vortex shedding. Such a dynamics has been recently observed on real car which points out its importance about industrial applications. Both the viscosity and the proximity of a wall in the vicinity of the parallelepiped body (similarly to the road with a car model), stabilize the RSB modes on a single symmetric mode. It is shown that these stabilizations occur through imperfect fork bifurcations at large Reynolds numbers. The extra drag due to the presence of the RSB modes is evidenced. Monday, November 24, 2014 3:35PM - 5:45PM Session L29 Experimental Techniques: PIV I — 2014 - Pavlos Vlachos, Purdue University 3:35PM L29.00001 Triple Pulse Particle Image Velocimeter/Accelerometer Measurements of Flow-Structure Interaction1 , SIVARAM GOGINENI, Spectral Energies, LLC, LIUYANG DING, RONALD ADRIAN, Arizona State University — A PIV-based instrument has been developed to measure position, velocity and acceleration of moving in fluids and the velocity and acceleration fields of the fluid motion simultaneously. The instrument extends conventional PIV by adding a third and sometimes fourth pulse, thereby increasing spatial resolution, velocity accuracy and enabling acceleration measurement. Images of the moving solid body are segmented from the fluid field and displacements are measured by cross-correlation, as in the fluid. To test the capabilities of this approach, a cylinder supported by elastic rods is oscillated sinusoidally in water to produce shed vortices that interact with the cylinder non-linearly Phase averaged fields are obtained in the fluid, and accuracy of the measurements is assessed by comparing the measurements of fluid velocity and acceleration to solid their known counterparts at the solid-fluid interface, 1 This work is supported by ONR under STTR program 3:48PM L29.00002 Optimization and Application of Surface Segmentation Technique for Tomographic PIV , LIUYANG DING, RONALD ADRIAN, Arizona State University, BRANDON WILSON, KATHY PRESTRIDGE, Los Alamos National Laboratory, LABORATORY FOR ENERGETIC FLOW AND TURBULENCE, ARIZONA STATE UNIVERSITY TEAM, EXTREME FLUID TEAM, P-23, PHYSICS DIVISION, LOS ALAMOS NATIONAL LABORATORY TEAM — Tomographic PIV is a widely used 3D flow measurement technique. It utilizes images recorded by multiple cameras to reconstruct the intensity distribution of a measured volume. The 3D3C velocity field is then computed by 3D crosscorrelation. Surface segmentation [1] aims to reduce computational cost. It extracts from a cloud of particles an image of those particles that lie on a mathematically prescribed surface. 2D2C velocity fields are computed on stacks of orthogonal surfaces, then assembled to construct the full 3D3C velocity field. We investigate the reconstruction of adaptive surfaces aligned with the main flow direction minimizing the out-of-plane motion. Numerical assessment is performed on curved-surface reconstruction for Taylor-Couette flow. An optimizing 2D interrogation scheme involving volumetric deformation is proposed to improve the accuracy of the 3D3C velocity field. The numerical test is performed on a synthetic vortex ring showing good measurement accuracy. Experimental results measuring the shock-driven turbulent mixing will also be presented. References [1] Ziskin, I.B., R.J. Adrian, and K. Prestridge. “Volume segmentation tomographic particle image velocimetry.” Proceedings of 9th international symposium on particle image velocimetry, Kobe, Japan. 2011. 4:01PM L29.00003 A Green’s function approach to PIV Pressure estimates1 , OLEG GOUSHCHA, PETER GANATOS, NIELL ELVIN, YIANNIS ANDREOPOULOS, The City College of New York — Spatial resolution of PIV data limits the ability to calculate the pressure along a solid boundary of a body immersed in a fluid and hence to accurately estimate the force exerted. Current methodologies solve numerically Navier-Stokes equations to calculate the pressure field from velocity data. An analytical approach has the potential of more accurate estimation of pressure in comparison to existing methods. A methodology has been developed to calculate the pressure distribution on the body in the flow by analytically solving the pressure Poisson Equation using a Green’s function approach. The pressure is then extrapolated to the solid boundary resulting in an accurate pressure distribution and total net force on the boundary. This technique has been applied to the case of a flexible cantilever beam vibrating after interacting with a traveling vortex in an experimental setup to harvest energy from an air-flow. Time-resolved PIV has been used to acquire a two-dimensional velocity field which has been used to obtain a time-dependent pressure distribution acting on the surface of the beam and resultant forces. The analytical solution is compared to the force measured directly by a force sensor placed at the base of the beam as well as the power harvested. 1 Sponsored by NSF Grant: CBET #1033117. 4:14PM L29.00004 Effects of 3D PIV post-processing on impulse and force analysis in vortical flows , LEAH MENDELSON, ALEXANDRA TECHET, MIT — Vortical flows measured using 3D PIV techniques are fundamentally filtered versions of physical phenomena, with velocity information lost below the length and time scales of the measurement system. In the context of propulsive vortices, such as those generated during biological locomotion, these factors, combined with experimental noise and error, can lead to inaccuracies in analysis of the vortex momentum and net thrust. As a result, while 3D velocity measurements remove many of the assumptions required to analyze planar PIV data, they should not be considered absolute physical quantities. Our work focuses on post-processing for 3D PIV data sets to enable the extraction of accurate, quantitative 3D force measurements for unsteady vortical propulsion. In this study, we compare utilizing measurement signal processing techniques, orthogonal decomposition, and identification of coherent structures to measure the impulse of a canonical vortex ring generated by a mechanical piston. In particular, we consider the ability of these methods to confront the influences of limited spatial resolution and arbitrary geometries, and make recommendations for a general procedure for propulsion analysis from 3D PIV data, regardless of which PIV technique is used to obtain the velocity fields. 4:27PM L29.00005 Method for fast and non iterative synthetic aperture reconstruction for 3D PIV and PTV , ABHISHEK BAJPAYEE, ALEXANDRA TECHET, MIT — Three dimensional particle image velocimetry (PIV) is becoming a widely used measurement technique since the introduction of tomographic PIV by Elsinga et al. (2006). New methods such as synthetic aperture PIV have recently been demonstrated as a viable alternative to tomographic PIV, and extended for accurate 3D particle tracking velocimetry (PTV) by Bajpayee et al. in 2013. Presented here is an improvement to the synthetic aperture reconstruction technique, using a homography fit (HF) method for projecting points into a camera through refractive interfaces, that allows non iterative, accurate and significantly faster reconstruction. The underlying algorithm is computationally cheap and can be easily and massively parallelized thereby allowing further improvement in speed depending on the hardware available. 4:40PM L29.00006 Application of Plenoptic PIV for 3D Velocity Measurements Over Roughness Elements in a Refractive Index Matched Facility1 , BRIAN THUROW, KYLE JOHNSON, Auburn University, TAEHOON KIM, GIANLUCA BLOIS, JIM BEST, University of Illinois, KEN CHRISTENSEN, University of Notre Dame — The application of Plenoptic PIV in a Refractive Index Matched (RIM) facility housed at Illinois is presented. Plenoptic PIV is an emerging 3D diagnostic that exploits the light-field imaging capabilities of a plenoptic camera. Plenoptic cameras utilize a microlens array to measure the position and angle of light rays captured by the camera. 3D/3C velocity fields are determined through application of the MART algorithm for volume reconstruction and a conventional 3D cross-correlation PIV algorithm. The RIM facility is a recirculating tunnel with a 62.5% aqueous solution of sodium iodide used as the working fluid. Its resulting index of 1.49 is equal to that of acrylic. Plenoptic PIV was used to measure the 3D velocity field of a turbulent boundary layer flow over a smooth wall, a single wall-mounted hemisphere and a full array of hemispheres (i.e. a rough wall) with a k/δ ≈ 4.6. Preliminary time averaged and instantaneous 3D velocity fields will be presented. 1 This material is based upon work supported by the National Science Foundation under grant no. 1235726 4:53PM L29.00007 Development and Characterization of Temperature-Sensitive Microbeads for Simultaneous Thermometry and Velocimetry , TREY COTTINGHAM, LILLIAN PRYOR, WEI-HSIN TIEN, GAMAL KHALIL, DANA DABIRI, University of Washington — The use of luminescent temperature sensitive paint (TSP) is well established as a means of measuring heat transfer on a surface. Particle image velocimetry (PIV) is a well-established method of measuring velocity fields in a fluid flow. By coating micro-spherical particles with temperature sensitive and reference (temperature insensitive) dyes, it will be possible to measure simultaneously both temperature and velocity. Temperature is measured using the intensity ratio method with the temperature and reference dye images obtained from EMCCD cameras, and the velocity will be obtained with PIV techniques. Response times of these dual-dye microsphere will also be investigated. Further results regarding technique development will be discussed. 5:06PM L29.00008 New experimental opportunities using refraction matched hydrogel: invisible objects, arrays, and features that obstruct flow but not light , JOEL WEITZMAN, LIANNA SAMUEL, ANNA CRAIG, ROBERT ZELLER, STEPHEN MONISMITH, JEFFREY KOSEFF, Stanford University — Water flow in and around immersed bodies, roughness arrays, and major bathymetric features is characterized by a large amount of spatial complexity. In both natural and designed settings, the associated hydrodynamic intricacies have influence on energy dissipation, thermal transfer, and mass exchange. However, the same surfaces that disrupt and redirect fluid motion also greatly restrict observation and measurement options. Solid boundaries tend to limit instrument access and block optical lines of sight. This presentation introduces a new technique expressly designed to overcome these hurdles. High-complexity solid models have been manufactured using a unique super-absorbent copolymer hydrogel. This material is wholly transparent, with an index of refraction nearly identical to that of water. When hydrogel object are submerged, light passes through them just as it passes through the fluid itself. Consequently, these objects and all their features become indistinguishable from their surroundings - effectively invisible. This opens up the entire internal flow field to direct observation and high-resolution quantitative measurement, a feat accomplished without reliance on unconventional fluids or specialized flow facilities. 5:19PM L29.00009 On the Application of Compressed Sensing to Non-Time-Resolved PIV Measurements1 , ERIC DEEM, TIMOTHY DAVIS, LOUIS CATTAFESTA, FARRUKH ALVI, Florida State University — Temporally resolved, full-field flow measurements are still impractical for most flows of interest. Fortunately, the spectral content of many flows can be described in a low dimensional space. This sparsity has inspired the recent adaptation of compressed sensing into the fluid mechanics community as a method for reconstructing spectral content of sub-Nyquist sampled data (arXiv:1401.7047). We apply this method to the analysis of several example fluid flow data sets, varying in spectral content. These data sets include the measured flow about a high-lift airfoil, an impinging jet, and a zero-net mass-flux (ZNMF) actuator. The Proper Orthogonal Decomposition (POD) is applied to the random PIV snapshots and we apply Orthogonal Matching Pursuit (OMP) to approximate the discrete Fourier transform of the POD coefficients. Additionally, reconstruction parameters are varied for to examine criteria regarding the probability of a successful reconstruction versus the degree of spectral sparsity. The advantages and restrictions of this method are discussed. 1 This work is supported by the Air Force Office of Scientic Research (grant FA9550-09-1-0257), monitored by Dr. Doug Smith, and the Florida Center for Advanced Aero-Propulsion 5:32PM L29.00010 Wave and flow field phenomena in planar falling films by simultaneous Laser-Induced Fluorescence and Particle Image/Tracking Velocimetry1 , ALEXANDROS CHAROGIANNIS, IVAN ZADRAZIL, CHRISTOS MARKIDES, Imperial College London — Falling films along an inclined flat plane test section were investigated using simultaneous Laser-Induced Fluorescence and Particle Image/Tracking Velocimetry techniques. The investigated conditions covered a range of Reynolds (2.2 – 8.2) and Kapitza numbers (28.6 – 41.4). The main challenge of the research is the development of routines that allow for simultaneous detailed measurements of liquid film topology as well as the instantaneous velocity fields within the liquid film while correcting for the refractive index discrepancy at the solid-liquid and gas-liquid interfaces. The uncertainties of the laser-based measurement techniques used to determine the local film thickness were compared with a micrometer based measurements as well as with the solution to the Navier-Stokes equations based on the assumptions for 1-D steady and fully developed flow. The results presented consist of in-detail characterisation of the aforementioned conditions as well as of flows with inlet pulsation frequencies in the range 1 – 8 Hz. 1 This work was supported by the Engineering and Physical Sciences Research Council (Grant Number EP/K008595/1). Monday, November 24, 2014 3:35PM - 6:11PM Session L30 Wind Turbines: Blade Designs — 2016 - Luciano Castillo, Texas Technical University 3:35PM L30.00001 Passive control of a dynamically pitching wind turbine airfoil under aeroelastic conditions using a Gurney flap1 , POURYA NIKOUEEYAN, ANDREW MAGSTADT, JOHN STRIKE, MICHAEL HIND, JONATHAN NAUGHTON, University of Wyoming — To reduce the cost of energy, wind turbine design has moved towards larger blades that are heavier and have lower relative structural stiffness compared to shorter blades. To address the lower blade stiffness, different flow control techniques have been considered. The Gurney flap, a small, low-cost and effective control method, is a promising control actuator. Wind tunnel testing has been performed on a DU97-W-300 10% flatback airfoil undergoing dynamic pitching relevant to flow conditions encountered by wind turbine blades. To mimic blade compliance, the airfoil is actively driven through a torsionally elastic element. Time-resolved surface pressure measurements have been acquired from which lift Cl and moment Cm coefficients were calculated. Changes in Cl and Cm in moderate and deep dynamic stall regimes for different Gurney flap heights were studied for different pitch drive conditions (amplitude and frequency). The results show the significant impact of compliance on the angle of attack (α) range experienced by the airfoil. Shifts in α range result in different hysteresis behavior in both Cl and Cm and demonstrate the effectiveness of the Gurney flap in modifying the aerodynamics of wind turbine blades experiencing dynamic pitching. 1 This work supported by DOE and a gift from BP 3:48PM L30.00002 Sub-scale Inverse Wind Turbine Blade Design Using Bound Circulation1 , CHRISTOPHER KELLEY, JONATHAN BERG, Sandia National Laboratories — A goal of the National Rotor Testbed project at Sandia is to design a sub-scale wind turbine blade that has similitude to a modern, commercial size blade. However, a smaller diameter wind turbine operating at the same tip-speed-ratio exhibits a different range of operating Reynolds numbers across the blade span, thus changing the local lift and drag coefficients. Differences to load distribution also affect the wake dynamics and stability. An inverse wind turbine blade design tool has been implemented which uses a target, dimensionless circulation distribution from a full-scale blade to find the chord and twist along a sub-scale blade. In addition, airfoil polar data are interpolated from a few specified span stations leading to a smooth, manufacturable blade. The iterative process perturbs chord and twist, after running a blade element momentum theory code, to reduce the residual sum of the squares between the modeled sub-scale circulation and the target full-scale circulation. It is shown that the converged sub-scale design also leads to performance similarity in thrust and power coefficients. 1 Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy under contract DE-AC04-94AL85000. 4:01PM L30.00003 Two Key Discoveries on Atmospheric Turbulent Wind Forcing of Nonsteady Wind Turbine Loadings, from HPC1 , JAMES BRASSEUR, GANESH VIJAYAKUMAR, ADAM LAVELY, BALAJI JAYARA- MAN, Penn State U, ERIC PATERSON, VA Tech, PETER SULLIVAN, NCAR — Loading transients on wind turbine blades underlie premature component failure. We research the underlying causes of nonsteady blade loadings from interactions with atmospheric eddies in the atmospheric boundary layer (ABL) using combinations of blade-boundary-layer-resolving HPC simulation and lower-order blade models (ALM, BEMT). A daytime ABL simulated with a 760 760 256 pseudo-spectral LES interacts with a 62 m rotating wind turbine blade, simulated with advanced finite-volume-based algorithms in two complex multi-grid/scale domains in relative motion. We focus on two key discoveries: (1) Whereas nonsteady blade loadings are generally interpreted as in response to nonsteadiness in wind speed, time changes in wind vector direction are a much greater contributor to load transients, and strongly impact boundary layer dynamics; (2) Large temporal variations in loadings occur within two disparate time scales, an advective time scale associated with atmospheric eddy passage, and a sub blade-rotation time scale associated with turbulence-induced forcings as the blades traverse internal atmospheric eddy structure. The latter generates the strongest transients; the former modulates the response. 1 Supported by DOE & NSF. Computer resources by XSEDE, OLCF, NREL. 4:14PM L30.00004 Direct Numerical Simulations of a Full Stationary Wind-Turbine Blade1 , ADNAN QAMAR, WEI ZHANG, WEI GAO, RAVI SAMTANEY, Division of Physical Sciences and Engineering, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, KSA — Direct numerical simulation of flow past a full stationary wind-turbine blade is carried out at Reynolds number, Re=10,000 placed at 0 and 5 (degree) angle of attack. The study is targeted to create a DNS database for verification of solvers and turbulent models that are utilized in wind-turbine modeling applications. The full blade comprises of a circular cylinder base that is attached to a spanwise varying airfoil cross-section profile (without twist). An overlapping composite grid technique is utilized to perform these DNS computations, which permits block structure in the mapped computational space. Different flow shedding regimes are observed along the blade length. Von-Karman shedding is observed in the cylinder shaft region of the turbine blade. Along the airfoil cross-section of the blade, near body shear layer breakdown is observed. A long tip vortex originates from the blade tip region, which exits the computational plane without being perturbed. Laminar to turbulent flow transition is observed along the blade length. The turbulent fluctuations amplitude decreases along the blade length and the flow remains laminar regime in the vicinity of the blade tip. The Strouhal number is found to decrease monotonously along the blade length. Average lift and drag coefficients are also reported for the cases investigated. 1 Supported by funding under a KAUST OCRF-CRG grant. 4:27PM L30.00005 Bio-inspired turbine blades offer new perspectives for wind energy , BENJAMIN THIRIA, VINCENT COGNET, PMMH-ESPCI, SYLVAIN COURRECH DU PONT, MSC-UPD, PMMH TEAM, MSC TEAM — The efficiency of wind turbines is especially poor if the wind speed is too low for the working range of the rotor, or if the oncoming wind has a too large incident angle with respect to the rotor axis. The consequence is that a large amount of potential available wind energy is not converted by the turbines, leading to heavy energetic and economic losses. The present work introduces a solution to overcome this technological limitation, using new types of blades connected to the rotors. This new type of blades is inspired by recent studies showing how insects improve flight performance by taking benefit from the flexibility of their wings (Ramananarivo et al. PNAS, 2011). Here, we show that, by bending along the chord under the action of the wind, the deformable blade plays the role of a shape factor to reorientate the torque in the direction of the rotation of the rotor, an especially helpful feature for critical wind conditions. The flexibility of the wing can significantly extend the performance range of wind turbines to low wind speeds and high azimuthal incoming wind directions, solving the technological barrier specific to this type of machines. The consequences of the presented results are outstanding for renewable solutions. Our estimation based on real wind data predicts a large increase in energy production, which is drawn using passive, non-consuming mechanisms, from the reservoir of energy available at critical wind conditions. 4:40PM L30.00006 The Role of Free Stream Turbulence on the Aerodynamic Performance of a Wind Turbine Blade , VICTOR MALDONADO, University of Texas at San Antonio, ADRIEN THORMANN, CHARLES MENEVEAU, Johns Hopkins University, LUCIANO CASTILLO, Texas Tech University — Effects of free stream turbulence with large integral scale on the aerodynamic performance of an S809 airfoil-based wind turbine blade at low Reynolds number are studied using wind tunnel experiments. A constant chord (2-D) S809 airfoil wind turbine blade model with an operating Reynolds number of 208,000 based on chord length was tested for a range of angles of attack representative of fully attached and stalled flow as encountered in typical wind turbine operation. The smooth-surface blade was subjected to a quasi-laminar free stream with very low free-stream turbulence as well as to elevated free-stream turbulence generated by an active grid. This turbulence contained large-scale eddies with levels of free-stream turbulence intensity of up to 6.14% and an integral length scale of about 60% of chord-length. The pressure distribution was acquired using static pressure taps and the lift was subsequently computed by numerical integration. The wake velocity deficit was measured utilizing hot-wire anemometry to compute the drag coefficient also via integration. In addition, the mean flow was quantified using 2-D particle image velocimetry (PIV) over the suction surface of the blade. Results indicate that turbulence, even with very large-scale eddies comparable in size to the chord-length, significantly improves the aerodynamic performance of the blade by increasing the lift coefficient and overall lift-to-drag ratio, L/D for all angles tested except zero degrees. 4:53PM L30.00007 Stall behavior of a scaled three-dimensional wind turbine blade , KAREN MULLENERS, Leibniz Universität Hannover, MATTHEW MELIUS, RAUL BAYOAN CAL, Portland State University — The power generation of a wind turbine is influenced by many factors including the unsteady incoming flow characteristics, pitch regulation, and the geometry of the various turbine components. Within the framework of maximizing energy extraction, it is important to understand and tailor the aerodynamics of a wind turbine. In the interest of seeking further understanding into the complex flow over wind turbine blades, a three-dimensional scaled blade model has been designed and manufactured to be dynamically similar to a rotating full-scale NREL 5MW wind turbine blade. A wind tunnel experiment has been carried out in the 2.2m x 1.8m cross-section closed loop wind tunnel at DLR in Göttingen by means of time-resolved stereoscopic PIV. An extensive coherent structure analysis of the time-resolved velocity field over the suction side of the blade was performed to study stall characteristics under a geometrically induced pressure gradient. In particular, the radial extent and propagation of stalled flow regions were characterized for various static angles of attack. 5:06PM L30.00008 Dynamic stall development in the near-root region of a model wind turbine blade , MATTHEW MELIUS, RAUL BAYOAN CAL, Portland State University, KAREN MULLENERS, Leibniz Universität Hannover — The dynamic behavior of atmospheric flows create highly variable operational conditions which affect the life expectancy of the turbine components and the power output of the turbine. To gain insight into the unsteady aerodynamics of wind turbine blades, wind tunnel experiments were conducted with a scaled three-dimensional NREL 5MW wind turbine blade model in the 2.2m x 1.8m cross-section closed loop wind tunnel DLR in Göttingen. The development of dynamic stall in response to a sudden change in the blades angle of attack are studied by means of time-resolved stereoscopic PIV in span-wisely distributed planes capturing the suction side of the blade. The change in angle of attack was obtained by varying the blade pitch angle to simulate a sudden change in wind speed or pitch angle regulation. Resulting time scales associated with flow separation and reattachment are determined at different radial positions ranging from r/R = 0.19 to r/R = 0.38. The influence of the three-dimensionality of the blade geometry on the corresponding aerodynamic effects is captured by analyzing the radial flow component in neighboring measurement fields during stall development. 5:19PM L30.00009 Dynamic Stall Patterns1 , PHILLIP DAVIDSON, ASHLI BABBITT, ANDREW MAGSTADT, POURYA NIKOUEEYAN, JONATHAN NAUGHTON, University of Wyoming, JONATHAN NAUGHTON TEAM — The performance of helicopter and wind turbine blades is affected by dynamic stall. Dynamic stall has received considerable attention, but it is still difficult to simulate and not fully understood. Over the past seven years, many airfoils for helicopter and wind turbine use ranging from 9.5 to 30% thick have been experimentally tested and simulated while dynamically pitching to further characterize dynamic stall. Tests have been run at chord Reynolds number between 225,000-440,000 for various reduced frequencies, mean angles of attack, and oscillation amplitudes. Characterization of stall has been accomplished using data from previous studies as well as the unsteady pressure and flow-field data available from our own work. Where available, combined surface and flow-field data allow for clear identification of the types of stall observed and the flow structure associated with them. The results indicate that thin airfoil stall, leading edge stall, and trailing edge stall are observed in the oscillating airfoil experiments and simulations. These three main stall types are further divided into subcategories. By improving our understanding of the features of dynamic stall, it is expected that physics-based simulations can be improved. 1 Work supported by DOE and a gift from BP 5:32PM L30.00010 Mechanisms of Vortex Evolution in Unsteady Stalled Flows1 , JAMES BUCHHOLZ, KEVIN WABICK, JAMES AKKALA, AZAR ESLAM PANAH, University of Iowa — Formation of a leading-edge vortex is considered on plunging and rotating flat plates at a chord-based Reynolds number of 104 . In all cases, a concentrated leading-edge vortex is formed. The physical mechanisms of vorticity transport governing the growth and evolution of the vortex are investigated within selected spanwise regions. It is demonstrated that the net flux magnitude of (oppositesign) secondary vorticity is often significant during formation of the leading-edge vortex, in comparison to that of the leading-edge shear layer, suggesting that the secondary flux plays a substantial role in regulating the growth and evolution of leading-edge vortex circulation. Other mechanisms of vorticity transport will also be discussed, including the importance of spanwise flow to vortex circulation, and the roles of vortex tilting and stretching on the evolution of the vorticity field. 1 This work was supported by the Air Force Office of Scientific Research through grant FA9550-11-1-0019 and the National Science Foundation Iowa EPSCoR program through grant EPS1101284. 5:45PM L30.00011 Implementation of a Forth-Order Aeroelastic Coupling into a ViscousInviscid Flow Solver with Experimental Validation (for One Degree of Freedom) , SIRKO BARTHOLOMAY, NÉSTOR RAMOS-GARCÍA, ROBERT FLEMMING MIKKELSEN, DTU, TECHNICAL UNIVERSITY OF DENMARK (DTU) - WIND ENERGY TEAM — The viscous-inviscid flow solver Q3 UIC for 2D aerodynamics has recently been developed at the Technical University of Denmark [1]. The Q3 UIC solver takes viscous and unsteady effects into account by coupling an unsteady inviscid panel method with the integral boundary layer equations by means of a strong coupling between the viscous and inviscid parts, and in this respect differs from other classic panel codes e.g. Xfoil. In the current work a Runge-Kutta-Nyström scheme was employed to couple inertial, elastic and aerodynamical forces and moments calculated by Q3 UIC for a two-dimensional blade section in the time-domain. Numerical simulations are validated by a three step experimental verification process carried out in the low-turbulence wind tunnel at DTU. First, a comparison against steady experiments for a NACA 64418 profile and a flexible trailing edge flap is presented for different fixed flap angles, and second, the measured aerodynamic characteristics considering prescribed motion of the airfoil with a moving flap are compared to the Q3 UIC predictions. Finally, an aeroelastic experiment for one degree of freedom –airfoil pitching- is used to evaluate the accuracy of aeroelastic coupling. [1] A strong viscous–inviscid interaction model for rotating airfoils. Ramos-Garcı́a, Néstor; Sørensen, Jens Nørkær; Shen, Wen Zhong. Wind Energy, 2013. 5:58PM L30.00012 Control of wing-tip vortex using winglets at low Reynolds number1 , SEUNGHYUN CHO, HAECHEON CHOI, Seoul National University — Winglets are considered as one of the effective devices for reducing induced drag, and thus many studies have been conducted, but mainly at high Reynolds numbers (Re ≈ 106 ∼ 107 ) for commercial airplanes. However, small-size unmanned air vehicles (UAV), operating at low Reynolds numbers (Re <105 ), become an important transportation system for different purposes. Therefore, in the present study, we experimentally investigate the effect of winglets on the aerodynamic performance of an UAV by varying the cant angle. The WASP UAV model is used and the Reynolds numbers considered are 110,000∼140,000 based on the free stream velocity and mean chord length of the WASP wing. The lift and drag forces on UAV are measured, and PIV measurements are conducted at several cross-flow planes for a few different angles of attack (α). At high angles of attack (7◦ ∼ 13◦ ), the winglets with the cant angle of 70◦ increase the aerodynamic performance, whereas at low angles of attack (2◦ ∼ 6◦ ), the wing-tip extension (cant angle of 0◦ ) shows better performances. The velocity fields measured from PIV indicate that, with the winglet, the wing-tip vortex moves away from the wing surface at α = 12◦ , and the downwash motion in the wake behind the trailing edge is decreased, reducing the magnitude of the induced drag. A concept of changing the cant angle during flight is also suggested at this talk. 1 Supported by 2011-0028032 Monday, November 24, 2014 3:35PM - 6:11PM Session L31 CFD: Lattice Boltzmann Methods — 2018 - Mouhamadou Aziz Diop 3:35PM L31.00001 LBM study of flow-induced cavitation , GIUSEPPE GONNELLA, GOETZ KAEHLER, FRANCESCO BONELLI, Università di Bari, Via Amendola 173, 70126 Bari, Italy, ANTONIO LAMURA, Istituto Applicazioni Calcolo, CNR, Via Amendola 122/D, 70126 Bari, Italy — This work deals with the investigation of homogeneous cavitation, induced by a fast flow past a sack wall, by using the lattice Boltzmann method (LBM). Cavitation occurs, in a liquid, because of a pressure drop, which falls below a certain threshold, with the consequent formation of vapor bubbles [1]. The aim is to study the inception of cavitation by using LBM without any “ad hoc” cavitation model [2]. A LBM with a body force term and redefined equilibrium distribution functions is employed for describing the continuity and Navier-Stokes equations for a fluid locally satisfying the van der Waals equation of state [3]. In such a way, cavitation is directly described by the solution of the LB equation. The numerical study shows the formation of a depletion zone just under the obstacle, near its left edge, where the pressure reaches a minimum value. Cavitation occurs only when the pressure of this depletion zone reaches a value lower than the spinodal of the liquid branch, thus not confirming the Joseph’s maximum tension criterion [4]. A detailed study of the flow field, of the Reynolds number effects, and of the developed cavitation regime are presented. [1] C. E. Brennen, Cavitation and Bubble Dynamics (Oxford University Press, 1995) [2] See for example M. Darbandi and H. Sadeghi, Numer. Heat Transfer A 58, 505 (2010). [3] A. Coclite, G. Gonnella, and A. Lamura, Phys. Rev. E 89, 063303 (2014). [4] G. Falcucci, E. Jannelli, S. Ubertini, and S. Succi, J. Fluid Mech. 728, 362 (2013). 3:48PM L31.00002 Numerical Simulation of Capillary Channels Growth in Heterogeneous Porous Anode in Aluminum Electrolysis Cells by Lattice Boltzmann Method , MOUHAMADOU DIOP, MORAN WANG, Dept of Engineering Mechanics, School of Aerospace Center for Nano and Micro Mechanics Tsinghua University, Beijing 100084, China — This paper presents results obtained from three-dimensional numerical simulations of multiphase reactive flows in porous anode block in aluminum cells controlling a great extent of mass, heat and chemical balance in the anode-cathode region. A lattice Boltzmann method based on thermal reactive multiphase flows, is developed to simulate the spatial and temporal distribution of fluids, the effects of gas rate and capillary instabilities in the cryolite. A new model, which involves eighteen lattice particles for the first and second derivative, is proposed to achieve accurate simulations at high fluid density ratio. The effects of the dissolution of gas and the capillary number on the flow field induced by gas bubbles evolution are investigated. It is found that capillary channels in the limit of small Stefan, the radial transport of reactant out of the capillary channel decay exponentially with the height of penetration in the porous anode. Several examples are solved by the proposed method to demonstrate the accuracy and robustness of the method. 4:01PM L31.00003 Deformation and breakup of viscoelastic droplets in confined shear flow1 , MAURO SBRAGAGLIA, ANUPAM GUPTA, University of Rome “Tor Vergata” — The deformation and breakup of Newtonian/viscoelastic droplets in systems with a Newtonian matrix are studied in confined shear flow. Our numerical approach is based on a combination of Lattice-Boltzmann models (LBM) and Finite Difference (FD) schemes, the former used to model two immiscible fluids with variable viscous ratio, and the latter used to model the polymer dynamics. The kinetics of the polymers is introduced using constitutive equations for viscoelastic fluids with finitely extensible non-linear elastic dumbbells with Peterlin’s closure (FENE-P). We quantify the droplet response by changing the polymer relaxation time, the maximum extensibility of the polymers, and the degree of confinement, i.e. the ratio of droplet diameter to gap spacing. In bulk shear flow, the effects of droplet viscoelasticity on the critical capillary number for breakup are moderate in all cases studied. However, in confined conditions a different behaviour is observed: the critical capillary number of a viscoelastic droplet increases or decreases, depending on the maximum elongation of the polymers, the latter affecting the extensional viscosity of the polymeric solution. Force balance is monitored in the numerical simulations to validate the physical picture. 1 ERC Grant Agreement n.279004 4:14PM L31.00004 A new multi-block-LBM scheme for turbulent flow simulations , YUSUKE KUWATA, KAZUHIKO SUGA, Osaka Prefecture University — A new lattice Boltzmann multi-block scheme based on the D3Q27 multiple relaxation time method is developed for turbulent flow simulations. In the streaming step, the distribution functions in the interface of each block are transferred by considering the continuity of the macroscopic variables. The mass and momentum continuity is achieved by keeping the consistency between the equilibrium distribution functions of the finer and coarse grids, whilst the non-equilibrium part is scaled for the continuity of the stress tensor. The 3rd order Lagrangian and Helmite interpolations are applied to temporally and spatially discretized variables in the interface region of the blocks. In order to relax the numerical errors occurring at the interface, which may affect the mass and momentum conservation, new distribution functions which are defined by the combination of the two distribution functions from the finer and coarse grids are streamed. The turbulent quantities such as the Reynolds stresses, budget terms of the Reynolds stress equation and power spectrum distributions are compared with those of DNS data by the pseudo spectrum method with good agreement. Moreover, the results show seamless profiles even at the interface of the blocks. 4:27PM L31.00005 Collisional Lattice Boltzmann Method for simulation of continuum through free molecular flow regimes , PRAKASH VEDULA, University of Oklahoma, ABDELAZIZ ALIAT, None — The collisional Lattice Boltzmann Method (cLBM) involves a lattice based numerical solution of Boltzmann equation including the full collision operator (Green & Vedula, J. Stat. Mech., 2013). Owing to accurate representation of important symmetries of the full collision operator (beyond collision invariants) and the lack of restrictive equilibrium based assumptions, this method could be particularly useful for efficient and accurate simulation of nonequilibrium flows. In the talk, we will discuss a generalization of cLBM using arbitrary lattices for description of two-dimensional flows. We will also discuss some physical and mathematical constraints that need to be considered for selection of lattices. Based on these considerations, we will demonstrate significant improvement in accuracy of simulations of selected flows in the continuum through free molecular flow regimes (up to Knudsen number, Kn O(100)). We will compare results (including the variation of velocity profiles, wall shear stress and mass flow rate with Kn) obtained from traditional LBM and DSMC with those obtained from cLBM using non-standard lattices. Using insights from these studies, we will also present techniques for significant improvement in accuracy of conventional LBM (based on BGK collision model). 4:40PM L31.00006 A Moving Boundary Condition based on Chapman-Enskog Expansion for the Lattice Boltzmann Method , LINA XU, LAURA SCHAEFER, University of Pittsburgh — The lattice Boltzmann method (LBM) has been shown to be an effective numerical method to model various fluid flows, by dealing boundary conditions from the mesoscopic level with straightforward and easy-to-implement approaches, the fundamental understanding of the hydrodynamic interactions between the solid and fluid in the particulate suspensions systems can be further improved. However, most of the previous boundary conditions used for the moving complex boundaries are based on the half way bounce-back boundary condition, where the geometric integrity of the body cannot be maintained. In this presentation, a moving boundary condition based on the Chapman-Enskog expansion is proposed and applied for the moving complex surfaces, where the precise shape of the solid can be preserved. Based on the numerical experiments for modelling the particulate suspensions system, the new moving boundary condition exhibits improved numerical accuracy and stability, stronger capability to preserve the geometry integrity, and better Galilean invariance character. Moreover, this presentation provides a novel concept to construct a boundary condition for the LBM without the limitation of being based on the information from the already existing lattice nodes. 4:53PM L31.00007 An Immersed Boundary-Lattice Boltzmann Approach to the Direct Numerical Simulation of Complex Particulate Flows , BAILI ZHANG, MING CHENG, ZHI SHANG, JING LOU, Institute of High Performance Computing — A three-dimensional momentum exchange-based immersed boundary-lattice Boltzmann method has been developed for solving fluid-particles interaction problems. This method combines the most desirable features of the lattice Boltzmann method and the immersed boundary method by using a regular Eulerian mesh for the flow domain and a Lagrangian mesh for the moving particles in the flow field. The non-slip boundary conditions for the fluid and the particles are enforced by adding a force density term into the lattice Boltzmann equation, and the forcing term is simply calculated by the momentum exchange of the boundary particle density distribution functions, which are interpolated by the Lagrangian polynomials from the underlying Eulerian mesh. This method preserves the advantages of lattice Boltzmann method in tracking a group of particles and, at the same time, provides an alternative approach to treat solid-fluid boundary conditions. Numerical validations show that the present method is very accurate and efficient. The code developed using this approach has been parallelized and allows the direct numerical simulation of fairly complicated phenomena such as three-dimensional particulate flow with very large numbers of particles. 5:06PM L31.00008 A multi-mesh lattice Boltzmann scheme for modeling solidification microstructure , MOHAMMAD HASHEMI, AMIRREZA HASHEMI, University of Akron, Akron, MOHSEN ESHRAGHI, California State University, Los Angeles, SERGIO FELICELLI, University of Akron, Akron — A multi-mesh lattice Boltzmann (LB) scheme is developed for modelling dendritic growth during solidification of binary alloys. Different physical phenomena including: mass transport, fluid flow, and heat transfer are involved in solidification, which are solved using the lattice Boltzmann method. Considering the difference in the length scales, a separate grid is introduced for each physical model to enhance the stability and computational performance of the method. Since the solutal boundary layer is very thin, a finer mesh is required near the interface to accurately simulate the transport phenomena. To address this problem, a non-uniform mesh is considered within each model. A conservative treatment was employed between neighbouring mesh blocks to ensure the continuity of mass, energy, and momentum. The multi-mesh model developed in this work is several times faster than the conventional unigrid LB models and offers a much better stability. Considering the high computational demands of the micro-scale simulations, the model can be employed as an efficient tool for simulating microstructural evolution during solidification. 5:19PM L31.00009 A conservative Dirichlet boundary treatment for the finite volume lattice Boltzmann method , LEITAO CHEN, LAURA SCHAEFER, University of Pittsburgh — The finite volume lattice Boltzmann method (FVLBM) enables the model to use the exact body-fitting mesh in the flow problems that involve the complex boundaries. However, the development of proper boundary treatment for the FVLBM has been outpaced. The boundary treatments designed for the conventional lattice Boltzmann method (LBM) framework are still heavily applied to the FVLBM. The largest defect of using the old boundary treatment is that, on the Dirichlet boundaries, the macroscopic variables cannot be conserved. In another word, there exist nontrivial discrepancies between the macroscopic variables defined by the boundary conditions and those yield by the numerical solutions. The errors on the boundaries will contaminate the internal solutions and even cause instability, especially on the complex boundaries. To overcome such a shortcoming, a conservative boundary treatment for the Dirichlet hydrodynamic boundary conditions is developed for the FVLBM. Through the benchmark tests, it is shown that the macroscopic conservations on the Direchlet boundaries are up to machine accuracy and completely independent of the size of relaxation time, the type of lattice model, the level of mesh resolution, the shape of boundaries and the type of internal scheme. 5:32PM L31.00010 Thermal lattice Boltzmann simulations with non-space-filling lattices , PARTHIB RAO, LAURA SCHAEFER, University of Pittsburgh — Thermal lattice Boltzmann (LB) models that ensure energy conservation have been less than satisfactory, compared to the athermal LB models due to variety of reasons. However, in the last few years, there has been a renewed effort in developing kinetically-consistent, stable, and accurate thermal models based on a new theoretical interpretation of the LB method-the Gauss-Hermite (GH) expansion technique. These, so-called higher-order models have been theorized to be able to model Navier-Stokes level thermo-hydrodynamics. Pursuant to this approach, we propose to use a third-order GH expansion of the equilibrium distribution along with a non-space-filling lattice (D2Q12), to model low-speed thermal flows, where temperature evolves according to an advection-diffusion equation. Additionally, this model is also compared for accuracy and computational efficiency, with an equivalent space-filling lattice (D2Q17) and the passive-scalar model. On benchmark thermal simulations such as Rayleigh-Benard convection and thermal Couette flows, our preliminary results indicate that the D2Q12 lattice is not only as as accurate as the D2Q17 lattice, but also computationally more efficient. Therefore, the D2Q12 lattice can be used modelling flows where temperature plays the role of a passive-scalar. 5:45PM L31.00011 Chapman-Enskog analyses on gray lattice Boltzmann schemes for fluid flow in porous media , CHEN CHEN, LIKE LI, RENWEI MEI, JAMES KLAUSNER, Univ of Florida - Gainesville — Gray lattice Boltzmann (GLB) schemes have recently been used to simulate fluid flow in porous media. It employs a partial bounce-back of populations (through a fractional reflection coefficient θ, which represents the fraction of populations being reflected by the solid phase) in the evolution equation to account for linear drag of the medium. These schemes are very easy to implement; but there exists uncertainty about the need for redefining macroscopic velocity as there has been no systematic analysis to recover the Brinkman equation from various GLB schemes. In this work, Chapman-Enskog analyses are performed to show that the momentum equation recovered from these schemes can satisfy Brinkman equation to second order in ε only if θ =O(ε) in which ε is the ratio of the lattice spacing to the characteristic length of physical dimension. The need for redefining macroscopic velocity is shown to be scheme-dependent. When gravitational force is considered or a body force is used to represent pressure gradient, the forcing term requires a modification factor that accounts for the effect of θ. The modification factor is derived for each combination of the forcing implementation method and the GLB scheme. The theoretical findings are verified by numerical results. 5:58PM L31.00012 Disc-shaped colloids interacting in a nematic liquid crystal , ALENA ANTIPOVA, COLIN DENNISTON, Univ of Western Ontario — We studied the behavior of pairs of disc-shaped colloidal particles in a nematic liquid crystal using Lattice Boltzmann algorithm. Without any external forces the position of the disc with respect to the liquid crystal director minimizes the free energy of the system and no distortion of the director field is observed. When the rotating magnetic field is present, the torque on the disc with homeotropic surface anchoring should change with analogy to electrostatic energy, which implies the disc continues turning following the field. However, when the disc reaches some critical position and the director field around it is highly distorted, the disc suddenly flips to minimize the free energy. Position and motion of pairs of such discs under similar conditions can be controlled by the angular velocity of magnetic field, it’s magnitude and initial configuration of the system. As a result of analysis of discs’ dynamics, a new way to control self-organization of disc particles was produced. Monday, November 24, 2014 3:35PM - 6:11PM Session L32 Turbulence: Internal Flows — 2020 - Roberto Verzicco, Universita degli Studi di Roma Tor Vergata 3:35PM L32.00001 Direct numerical simulations of Taylor-Couette up to Re=400,000 , RODOLFO OSTILLA MONICO, Physics of Fluids, Mesa+ Institute, University of Twente, P.O. Box 217, 7500 AE Enschede, ROBERTO VERZICCO, Dipartimento di Ingegneria Meccanica, University of Rome “Tor Vergata,” Via del Politecnico 1, Roma 00133, Italy, SIEGFRIED GROSSMANN, Department of Physics, University of Marburg, Renthof 6, D-35032 Marburg, Germany, DETLEF LOHSE, Physics of Fluids, Mesa+ Institute, University of Twente, P.O. Box 217, 7500 AE Enschede — Taylor-Couette, the flow between two coaxial, independently rotating cylinders, is simulated up to shear Reynolds numbers of Res ∼ 4 · 105 , corresponding to frictional Reynolds numbers of Reτ ∼ 4000. The radius ratio is set to η = ri /ro = 0.909, to reduce curvature effects and axially periodic boundary conditions are used. One-point statistics and spectra are calculated. Using these, the flow is divided into boundary layer and bulk. The boundary layer, containing a logarithmic sublayer is found to extend up to a tenth of the gap width from the wall. This log-layer shows comparable one-point statistics and spectra to log-layers in other DNS of channels, i.e. κ ≈ 0.4 and a k−1 energy spectra in the streamwise (azimuthal) velocity. Further away from the wall, the flow is modified by the large scale Taylor rolls, and a bulk region appears, with very different behaviour. Finally, the effect of the computational domain size on the flow is quantified, and larger computational boxes are compared to the ones used at high Re. 3:48PM L32.00002 Direct numerical simulation of Taylor-Couette flows between two corotating cylinders with radial heating1 , HAO TENG, NANSHENG LIU, XIYUN LU, Department of Modern Mechanics, University of Science and Technology of China, MULTI-SCALE COMPLEX FLOW LAB. TEAM — In the present work, Taylor-Couette (TC) flows subjected to radial heating have been investigated by using of direct numerical simulation (DNS). For our simulations, the base flow is driven between two co-rotating concentric cylinders; radial heating is modeled by a radial temperature gradient between the hot inner and cold outer cylinders, which introduces a coupling effect characterized by a ratio (σ Gr/Re2 ) between the buoyancy force driving fluid elements moving axially and the centrifugal force driving a radial fluid motion. Here, the Grashof (Gr) and Reynolds (Re) numbers represent the non-dimensionalized buoyancy and centrifugal forces, respectively. It is demonstrated that increasing σ from 0 to 0.4 leads to a flow state transition from the wavy Taylor vortex (WTV) TC flows driven by the centrifugal force to the distorted Taylor vortex (DTV) TC flows arising as a result of the buoyancy and centrifugal force competition, and eventually to the buoyancy dominated turbulent (BDT) TC flows identified as the vanishing of Taylor vortices that are replaced by turbulent vertical structures of small scales. These three flow states are distinguished by their flow structures, dynamical properties, and energy spectra. 1 This work is supported by NSFC No. 11272306. 4:01PM L32.00003 The phase space of turbulent Taylor-Couette flow , DETLEF LOHSE, RODOLFO OSTILLA MONICO, ERWIN VAN DER POEL, ROBERTO VERZICCO, University of Twente, SIEGFRIED GROSSMANN, University of Marburg — Direct numerical simulations of Taylor-Couette flow, i.e. the flow between two coaxial and independently rotating cylinders were performed. Shear Reynolds numbers of up to 3 · 105 , corresponding to T a = 4.6 · 1010 , were reached. The transition to the ultimate regime, in which asymptotic scaling laws for the torque are expected to hold up to arbitrarily high driving, is analysed for different radius ratios, different aspect ratios and different rotation ratios. We also calculate the local angular velocity profiles and visualize different flow regimes that depend both on the shearing of the flow and the Coriolis force originating from the outer cylinder rotation. Two main regimes are distinguished, based on the magnitude of the Coriolis force, namely the co-rotating and weakly counter-rotating regime dominated by Rayleigh-unstable regions, and the strongly counter-rotating regime where a mixture of Rayleigh-stable and Rayleigh-unstable regions exist. The work culminates in phase spaces in the inner vs outer Reynolds number parameter space and in the Taylor vs inverse Rossby number parameter space, which can be seen as the extension of the Andereck et al. (J. Fluid Mech. 164, 155-183, 1986) phase space towards the ultimate regime. 4:14PM L32.00004 Direct numerical simulation of turbulent plane Couette flow at Rew=6000 , JIE GAI, ZHENHUA XIA, QINGDONG CAI, State Key Laboratory for Turbulence and Complex Systems, College of Engineering, Peking University, Beijing 100871, P. R. China — The large-scale counter-rotating streamwise vortices (secondary vortices) in fully-developed plane Couette flow have been reported by both experimental and numerical communities. However, the number of vortex pairs does not increase linearly with the spanwise width of the domain, which is the same as the reported results of the turbulent Taylor-Couette flow. In present work, a series of direct numerical simulations at Reynolds number of 6000 (based on the relative wall speed and half the channel height δ) with different streamwise and spanwise lengths were conducted to investigate the effect of the box size on the large-scale structures. Our results showed that the correlation of secondary structures in the box with smaller streamwise length is much stronger than that with larger streamwise length and the rate of the turbulent kinetic energy contributed by the secondary structures is influenced by the mean spanwise scale of the secondary eddies and the streamwise length of the computational domain. In addition, the spanwise length scale of each secondary vortex pairs was found in a domain of 4(1 ± 0.3)δ. 4:27PM L32.00005 Coherent structures in turbulent rectangular duct flows , HASSAN NAGIB, IIT, Chicago, RICARDO VINUESA, KTH, Stockholm and IIT, Chicago, CEZARY PRUS, PHILIPP SCHLATTER, KTH, Stockholm — Turbulent duct flows computed by means of DNS with spectral element code Nek5000 are analyzed to characterize coherent structures present in this flow. A number of aspect ratio (defined as duct width over height) cases ranging from 1 to 18 at two different Reynolds numbers (Reτ ≃ 180 and 330) constitute the data set under study. Common methods for coherent vortex identification (λ2 , Q, λci and ∆), together with a less widely used approach by Kida and Miura, are used to characterize the structures in the various duct cases. All methods yield very similar results, and identify the occurrence of buffer layer vortices along the horizontal and side walls, with the well-documented spacing between streaks of ≃ 100+ . Secondary vortices in the duct corners are only found from two-dimensional fields when averaged over time and streamwise computational domain. The results indicate that the corner vortices may exhibit much slower time-scales than buffer layer vortices. The concept of turbulent net force is also applied to the 2D fields to assess impact of the corner vortices on the flow. The main features of these structures are compared with the ones found in spanwise-periodic turbulent channel flows at the same Reynolds numbers. 4:40PM L32.00006 The effect of large property fluctuations on turbulent heat transfer to supercritical pressure fluids in pipes , RENE PECNIK, HASSAN NEMATI, ASHISH PATEL, BENDIKS JAN BOERSMA, Delft University of Technology — When a fluid slightly above the thermodynamic critical pressure is heated, such that the fluid’s state crosses the pseudo-critical line, no distinct liquid to gas phase transition occur. However, the fluid properties change abruptly. If these property variations occur in a turbulent flow the conventional behavior of turbulence is strongly altered. We study the influence of these large property fluctuations in forced convection heat transfer to supercritical carbon dioxide in a pipe, with DNS at a Karman number of Re=180 (based on the pipe inlet conditions). At the inlet the temperature is slightly below the pseudo-critical point, such that during the heating process the developing thermal boundary layer crosses the pseudo-critical line. We show that the occurring property fluctuations have a strong effect on the averaged wall enthalpy if a constant wall heat flux boundary condition (infinite thermal effusivity ratio of fluid to solid) is used. By changing the boundary conditions to constant wall temperature (vanishing thermal effusivity ratio) these fluctuations are eliminated at the wall and the heat transfer coefficient is decreased. 4:53PM L32.00007 Sparse energetically dominant frequencies in direct numerical simulation of turbulent pipe flow: origin and application to reduced-order models1 , FRANCISCO GOMEZ, HUGH M. BLACKBURN, MURRAY RUDMAN, Monash University, BEVERLEY J. MCKEON, MITUL LUHAR, RASHAD MOARREF, California Institute of Technology, ATI S. SHARMA, University of Southampton — The idea of constructing reduced-order models for canonical wall-bounded turbulent flows based on exploiting the sparse energetically dominant frequencies observed in direct numerical simulation of pipe flow by Bourguignon et al. (2013, Phys. Fluids) is examined. The resolvent analysis of a pipe flow is extended in order to consider the influence of finite domain length on the flow dynamics, which restricts the possible wavespeeds in the flow. This analysis shows that large sparse amplifications take place when one of the allowable wavespeeds is equal to the local wavespeed via the critical layer mechanism. A connection between amplification and energy is presented through the similar features displayed by the most energetically relevant flow structures, emerging from a dynamic mode decomposition of direct numerical simulation data, and the resolvent modes associated with the most amplified sparse frequencies. These findings support the viability of reduced-order models based on the selection of the most amplified modes arising from the resolvent model, with the potential to drastically decrease the computational costs required to represent turbulent flows. 1 This work was supported by ARC grant DP130103103 and AFOSR grant FA9550- 09-1-0701 (ML, RM, BJM). 5:06PM L32.00008 Experimental study of a turbulent flow under transient conditions , SHUISHENG HE, SAM GORJI, MEHDI SEDDIGHI, University of Sheffield, TOM O’DONOGHUE, DUBRAVKA POKRAJAC, University of Aberdeen, ALAN VARDY, University of Dundee — Particle Image Velocimetry (PIV) is applied to investigate the behaviour of transient turbulent channel flows. During the experiments, the flowrate is accelerated from a lower Re turbulent flow to one at a higher Re. The investigations reveal novel insights into turbulence behaviour in the transient process. It is shown that the unsteady flows behave strikingly similar to the so-called boundary layer bypass transition due to free-stream-turbulence. Consistent with the DNS of He and Seddighi (J. Fluid Mech., 715: 60-102), the process begins with the elongation of streaks much similar to the Klebanoff modes in the buffeted laminar boundary layer in a bypass transition. During the second stage, the formation and propagation of isolated turbulent spots eventually lead to a complete breakdown of the organised streaky structures resulting in a new turbulent flow corresponding to the final Reynolds number. The present investigation covers a range of initial and final Reynolds numbers to elucidate the underlying mechanisms involved in transient flows. 5:19PM L32.00009 Transition to Turbulence in curved pipe , AMIRREZA HASHEMI, FRANCIS LOTH, University of Akron, Akron, Ohio — Studies have shown that transitional turbulence in a curved pipe is delayed significantly compared with straight pipes. These analytical, numerical and experimental studies employed a helical geometry that is infinitely long such that the effect of the inlet and outlet can be neglected. The present study examined transition to turbulence in a finite curved pipe with a straight inlet/outlet and a 180 degrees curved pipe with a constant radius of curvature and diameter (D). We have employed the large scale direct numerical simulation (DNS) by using the spectral element method, nek5000, to simulate the flow field within curved pipe geometry with different curvature radii and Reynolds numbers to determine the point of the transition to turbulence. Long extensions for the inlet (5D) and outlet (20D) were used to diminish the effect of the boundary conditions. Our numerical results for radius of curvatures of 1.5D and 5D show transition turbulence is near Re=3000. This is delayed compared with a straight pipe (Re=2200) but still less that observed for helical geometries (Reynolds number less than 5000). Our research aims to describe the critical Reynolds number for transition to turbulence for a finite curved pipe at various curvature radii. 5:32PM L32.00010 Direct numerical simulation of turbulence and heat transfer in a hexagonal shaped duct , OANA MARIN, ALEKS OBABKO, MCS, Argonne National Laboratories, USA, PHILIPP SCHLATTER, KTH Mechanics, Sweden — Flows in hexagonal shapes frequently occur in nuclear reactor applications, and are also present in honeycomb-shaped settling chambers for e.g. wind tunnels. Whereas wall-bounded turbulence has been studied comprehensively in two-dimensional channels, and to a lesser degree also in square and rectangular ducts and triangles, only very limited data for hexagonal ducts is available, including resistance correlations and mean profiles. Here, we use resolved spectral-element simulations to compute velocity and temperature in fully-developed (periodic) hexagonal duct flow. The Reynolds number, based on the fixed flow rate and the hydraulic diameter, ranges between 2000 and 20000. The temperature assumes constant wall flux or constant wall temperature. First DNS results are focused on the mean characteristics such a head loss, Nusselt number, and critical Reynolds number for sustained turbulence. Profiles, both for mean and fluctuating quantities, are extracted and discussed in the context of square ducts and pipes. Comparisons to existing experiments, RANS and empirical correlations are supplied as well. The results show a complicated and fine-scale pattern of the in-plane secondary flow, which clearly affects the momentum and temperature distribution throughout the cross section. 5:45PM L32.00011 Osborne Reynolds pipe flow: direct numerical simulation from laminar to fully-developed turbulence , R.J. ADRIAN, Arizona State Univ., X. WU, Canadian Royal Military College, P. MOIN, Stanford Univ., J.R. BALTZER, Los Alamos National Lab. — Osborne Reynolds’ pipe experiment marked the onset of modern viscous flow research, yet the detailed mechanism carrying the laminar state to fully-developed turbulence has been quite elusive, despite notable progress related to dynamic edge-state theory. Here, we continue our direct numerical simulation study on this problem using a 250R long, spatially-developing pipe configuration with various Reynolds numbers, inflow disturbances, and inlet base flow states. For the inlet base flow, both fully-developed laminar profile and the uniform plug profile are considered. Inlet disturbances consist of rings of turbulence of different width and radial location. In all the six cases examined so far, energy norms show exponential growth with axial distance until transition after an initial decay near the inlet. Skin-friction overshoots the Moody’s correlation in most, but not all, the cases. Another common theme is that lambda vortices amplified out of susceptible elements in the inlet disturbances trigger rapidly growing hairpin packets at random locations and times, after which infant turbulent spots appear. Mature turbulent spots in the pipe transition are actually tight concentrations of hairpin packets looking like a hairpin forest. The plug flow inlet profile requires much stronger disturbances to transition than the parabolic profile. 5:58PM L32.00012 A Visualization Study of Wall Layer of Swirling Turbulent Pipe Flow , MERIAM MALEK1 , RACHAEL HAGER, OMER SAVAS, University of California-Berkeley — The streaky vortical structure of the viscous sublayer of a turbulent boundary layer is well known. Turbulent flows in pipes also exhibit similar structures. The effect of swirl on that structure is the subject matter of this study. The experiments are conducted in water in a 5-cm diameter clear cast-acrylic pipe at Reynolds numbers up to 80,000. Initial geometric swirl angles up to 60◦ at the wall are generated by placing 3D printed inserts at the inlet of the pipe. Flows are visualized using reflective flakes of size distribution 10-80 µm under diffuse illumination. Flows are recorded at high framing rates. After preprocessing, the streaky structure is quantified by using autocorrelation of the images. Lateral spacing and longitudinal length scales are extracted. Also studied is the decay of the swirl angle and its influence of the wall structure. 1 Undergraduate Researcher Monday, November 24, 2014 3:35PM - 6:11PM — Session L33 Computational Methods and Modeling of Multiphase Flows I 2022 - Marcus Herrmann, Arizona State University 3:35PM L33.00001 Numerical Modelling of Three-Fluid Flow Using The Level-set Method1 , HONGYING LI, JING LOU, ZHI SHANG, Institute of High Performance Computing, Agency for Science, Technology and Research — This work presents a numerical model for simulation of three-fluid flow involving two different moving interfaces. These interfaces are captured using the level-set method via two different level-set functions. A combined formulation with only one set of conservation equations for the whole physical domain, consisting of the three different immiscible fluids, is employed. Numerical solution is performed on a fixed mesh using the finite volume method. Surface tension effect is incorporated using the Continuum Surface Force model. Validation of the present model is made against available results for stratified flow and rising bubble in a container with a free surface. Applications of the present model are demonstrated by a variety of three-fluid flow systems including (1) three-fluid stratified flow, (2) two-fluid stratified flow carrying the third fluid in the form of drops and (3) simultaneous rising and settling of two drops in a stationary third fluid. 1 The work is supported by a Thematic and Strategic Research from A*STAR, Singapore (Ref. #: 1021640075). 3:48PM L33.00002 Modified level set equation: extension to the case of hyperbolic tangent level set function , ANDREY OVSYANNIKOV, Stanford University, USA, VLADIMIR SABELNIKOV, ONERA, France , MIKHAEL GOROKHOVSKI, Ecole Centrale de Lyon, France — The well-known problem of the level set method, applied to the computation of interfacial two-phase flow, is this: if the flow velocity is not constant, the level set scalar field may become strongly distorted. Usually, this problem is remedied by the reinitialization procedure. In our previous work, we proposed an alternative approach. We modified directly the level set equation by embedding a source term. The exact expression of this term is such that the eikonal equation is automatically satisfied, similar to the extension velocity method. The exact expression of the source term makes also possible the derivation of its local approximate forms, of zero-, first- and higher-order accuracy. Application of approximate forms reduced significantly the number of reinitializations. In this study, we extend our modification of the level set equation to the conservative level set method (CLS). Despite a good mass conservation, CLS suffers from inaccurate prediction of the in interface shape due to numerous reinitializations. We introduce a new source term in the CLS equation which preserves the hyperbolic tangent profile. The proposed method is assessed for different test cases including interface stretching by vortex flow and Rayleigh-Taylor instability. 4:01PM L33.00003 DNS and modeling of bubbly flows in vertical channels1 , MING MA, JIACAI LU, GRETAR TRYGGVASON, University of Notre Dame — The transient motion of bubbly flow, in a vertical channel is studied, using direct numerical simulations (DNS) where every continuum length and time scale is resolved. Nearly spherical bubbles of the same size, injected into laminar upflow, are quickly pushed to the walls due to lift. The velocity then slows down, eventually resulting in some of the bubbles returning to the core forming a mixture where the weight matches the imposed pressure gradient and the void fraction is easily predicted. Unlike the statistically steady state, where the flow structure is relatively simple and in some cases depends only on the sign of the lift coefficient, the transient evolution is more sensitive to the governing parameters. The DNS results are used to provide values for the unresolved closure terms in a simple average model for the flow, found by mining the data, using various techniques such as regression and neural networks. Results for a large number of bubbles of several different sizes in turbulent upflow are also presented and the prospects of using a similar approach for LES-like simulations of more complex flows are discussed, including the simplification of the interface structure resulting from filtering. 1 Research supported by DOE (CASL) and NSF Grant CBET 1335913. 4:14PM L33.00004 Multiscale computations of mass accumulation effect on mass transfer in bubbly flow , BAHMAN ABOULHASANZADEH, GRETAR TRYGGVASON, University of Notre Dame — Mass transfer in bubbly flow generally takes place on a much smaller length and time scale than the length and time scale of the momentum flow, resulting in a thin mass boundary layer around the bubbles. We developed a multiscale model to solve a boundary layer equation for the mass boundary layer next to the bubble interface, assuming zero mass concentration in the far field, which couples with the rest of domain using a source/sink term. Here, we extend our model to account for non-zero concentration next to the mass boundary layer. Comparison of simple case studies in 1D and 2D problems show good agreement between the fully resolved solution and the solution on a much coarser grid using our model. We study the effect of mass accumulation in a domain and also the effect of bubble moving into the wake of another bubble on the mass transfer. This study was funded by NSF Grant CBET-1132410. 4:27PM L33.00005 Comparisons and Limitations of Gradient Augmented Level Set and Algebraic Volume of Fluid Methods , LAKSHMAN ANUMOLU, DOUGLAS RYDDNER, MARIO TRUJILLO, University of Wisconsin Madison — Recent numerical methods for implicit interface transport are generally presented as enjoying higher order of spatial-temporal convergence when compared to classical methods or less sophisticated approaches. However, when applied to test cases, which are designed to simulate practical industrial conditions, significant reduction in convergence is observed in higher-order methods, whereas for the less sophisticated approaches same convergence is achieved but a growth in the error norms occurs. This provides an opportunity to understand the underlying issues which causes this decrease in accuracy in both types of methods. As an example we consider the Gradient Augmented Level Set method (GALS) and a variant of the Volume of Fluid (VoF) method in our study. Results show that while both methods do suffer from a loss of accuracy, it is the higher order method that suffers more. The implication is a significant reduction in the performance advantage of the GALS method over the VoF scheme. Reasons for this lie in the behavior of the higher order derivatives, particular in situations where the level set field is highly distorted. For the VoF approach, serious spurious deformations of the interface are observed, albeit with a deceptive zero loss of mass. 4:40PM L33.00006 Development of high order numerical methods for particle-laden flows on unstructured grids: A realizability-preserving Discontinuous Galerkin method for moderate Stokes number flows , ADAM LARAT, MACOLE SABAT, EM2C-Ecole Centrale Paris, AYMERIC VIÉ, Center for Turbulence Research, CHRISTOPHE CHALONS, Université de Versailles Saint-Quentin, MARC MASSOT, EM2C-Ecole Centrale Paris — The simulation of particle-laden flows is of primary importance for several industrial applications, like sprays in aeronautical combustors or particles in fluidized beds. Our focus is on Moment methods that describes the disperse phase as a continuum. The accuracy and performance of such approaches highly depends on the number of controlled moments for correctly describing the physics of the flow, but also on the numerics that are used to solve the continuous system of equations at a discrete level. In the present work, we investigate the use of Discontinuous Galerkin methods to solve for the convective part of the moment equations. By deriving realizability conditions on the moment system that are associated to a convex space, a projection strategy is used to maintain the solution in the realizable space. This method is applied to the resolution of the pressure less gas dynamics and the Anisotropic Gaussian moment approach, the former solving for low Stokes number flows where no Particle Trajectory Crossing occurs, while the latter is solving for moderate Stokes number flows and can handle PTC through a pressure tensor in the convective term. The strategy is assessed on turbulent flows through comparisons with Lagrangian results. 4:53PM L33.00007 Horizontal annular flow modelling using a compositional based interface capturing approach1 , DIMITRIOS PAVLIDIS, ZHIZHUA XIE, JAMES PERCIVAL, Imperial College London, JEFFERSON GOMES, University of Aberdeen, CHRIS PAIN, OMAR MATAR, Imperial College London — Progress on a consistent approach for interface-capturing in which each component represents a different phase/fluid is described. The aim is to develop a general multi-phase modelling approach based on fully-unstructured meshes that can exploit the latest mesh adaptivity methods, and in which each fluid phase may have a number of components. The method is compared against experimental results for a collapsing water column test case and a convergence study is performed. A number of numerical test cases are undertaken to demonstrate the method’s ability to model arbitrary numbers of phases with arbitrary equations of state. The method is then used to simulate horizontal annular flows. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 5:06PM L33.00008 Accurate VoF based curvature evaluation method for low-resolution interface geometries , MARK OWKES, Montana State University, MARCUS HERRMANN, Arizona State University, OLIVIER DESJARDINS, Cornell University — The height function method is a common approach to compute the curvature of a gas-liquid interface in the context of the volume-of-fluid method. While the approach has been shown to produce second-order curvature estimates for many interfaces, the height function method deteriorates when the curvature becomes large and the interface becomes under-resolved by the computational mesh. In this work, we propose a modification to the height function method that improves the curvature calculation for under-resolved structures. The proposed scheme computes heights within columns that are not aligned with the underlying computational mesh but rather the interface normal vector which are found to be more robust for under-resolved interfaces. A computational geometry toolbox is used to compute the heights in the complex geometry that is formed at the intersection of the computational mesh and the columns. The resulting scheme has significantly reduced curvature errors for under-resolved interfaces and recovers the second-order convergence of the standard height function method for well-resolved interfaces. 5:19PM L33.00009 Accurate curvature estimates from volume fractions on unstructured meshes using embedded height functions1 , CHRISTOPHER IVEY, PARVIZ MOIN, Stanford University, Center for Turbulence Research — A novel methodology for extracting curvatures from volume fractions on non-convex, unstructured grids is presented. Estimating curvature in volume of fluid methods is difficult due the discontinuous nature of the volume fraction field. On simple structured meshes, height functions can be used to map the volume fraction field to a surface height field that is smooth along the surface. Our algorithm utilizes a local cartesian mesh and a suitable interpolation strategy to harness the height function technique developed for uniform meshes. Accuracy of the algorithm is demonstrated through comparison with the with the reconstructed distance function method on unstructured meshes and with the traditional height function method on uniform meshes of similar grid density. 1 Supported by the DOE CSGF (grant number DE-FG02-97ER25308) 5:32PM L33.00010 Development of Numerical Method for Two-phase Flows on Threedimensional Arbitrarily-shaped Polyhedral Meshes , KOHEI SUZUKI, TAKESI OMORI, TAKEO KAJISHIMA, Department of Mechanical Engineering, Osaka University — Although the advantage of using arbitrarily-shaped polyhedral meshes for the industrial flow applications is clear, their employment to two-phase flows is rather limited due to the poor prediction accuracy of the existing numerical methods on such meshes. We present a numerical method based on VOF (Volume of Fluid) method which works on arbitrarily-shaped three-dimensional polyhedral meshes with little volume/shape error for the interface advection and with little curvature estimation error. To make the implementation in three-dimensional geometry feasible, we extend THINC (Tangent of Hyperbola Interface Capturing) method for polyhedral meshes which does not require laborious geometric arithmetics. In the oral presentation we will also show that the combination of RDF (Reconstructed Distance Function) algorithm and the carefully selected discretization procedure gives good performance in the interface curvature estimation. 5:45PM L33.00011 A Numerical Study of Mesh Adaptivity in Multiphase Flows with NonNewtonian Fluids1 , JAMES PERCIVAL, DIMITRIOS PAVLIDIS, ZHIHUA XIE, Imperial College London, FEDERICO ALBERINI, MARK SIMMONS, University of Birmingham, CHRISTOPHER PAIN, OMAR MATAR, Imperial College London — We present an investigation into the computational efficiency benefits of dynamic mesh adaptivity in the numerical simulation of transient multiphase fluid flow problems involving Non-Newtonian fluids. Such fluids appear in a range of industrial applications, from printing inks to toothpastes and introduce new challenges for mesh adaptivity due to the additional “memory” of viscoelastic fluids. Nevertheless, the multiscale nature of these flows implies huge potential benefits for a successful implementation. The study is performed using the open source package Fluidity, which couples an unstructured mesh control volume finite element solver for the multiphase Navier-Stokes equations to a dynamic anisotropic mesh adaptivity algorithm, based on estimated solution interpolation error criteria, and conservative mesh-to-mesh interpolation routine. The code is applied to problems involving rheologies ranging from simple Newtonian to shear-thinning to viscoelastic materials and verified against experimental data for various industrial and microfluidic flows. 1 This work was undertaken as part of the EPSRC MEMPHIS programme grant EP/K003976/1. 5:58PM L33.00012 A Multiscale FSI Analysis of Flow past a Cylinder , JAMES CHEN, MAURIN LOPEZ, Penn State Altoona — Micropolar fluid theory generalizes classical continuum mechanics by incorporating the microscale spinning effects of fluid molecules. It has been seen that Micropolar fluid theory shows strong promises of predicting the microscale fluid behaviours in the continuum level. With the two-level motion in Micropolar theory, the interaction between macromotion and micromotion in the fluid flows can be utilized to interpret flow phenomena. It is understood that the classical fluid theory is not fully capable of explaining fluids phenomena involving energy dissipation across multiple length scales from theoretical perspective. Such phenomena include vortex formation, boundary layer development and etc. Flow past a cylinder is studied as an example. An in-house developed solver based in a high order spectral difference method to solve the Micropolar equations with moving and deformable grids for fluid-solid interaction (FSI) is used. By studying how the translational velocity (macromotion) dissipates into gyration (micromotion) it is possible to understand how the energy cascade into smaller scales for vortex formation, this mechanism explains how vortices form and how the coherent structures of vortices and eddies construct. Monday, November 24, 2014 3:35PM - 6:11PM Session L34 CFD: Turbulence Modeling I — 2024 - Dale Pullin, California Institute of Technology 3:35PM L34.00001 RANS study of flow Characteristics Over flight deck of Simplified frigate Ship , SHRISH SHUKLA1 , SIDH NATH SINGH2 , BALAJI SRINIVASAN3 , Indian Institute of Technology Delhi — The combined operation of a ship and helicopter is ubiquitous in every naval organization. The operation of ship with the landing and takeoff of a helicopter over sea results in very complex flow phenomena due to presence of ship air wakes, strong velocity gradients and widely varying turbulence length scales. This complexity of flow is increased with the addition of helicopter downwash during landing and takeoff. The resultant flow is therefore very complicated and accurate prediction represents a computational challenge. We present Reynolds-averaged-Navier-Stokes (RANS) of turbulent flow over a simple frigate ship to gain insight into the flow phenomena over a flight deck. Flow conditions analysis is carried out numerically over the generic simplified frigate ship. Profiles of mean velocity across longitudinal and transverse plane have been analyzed along the ship. Further, we propose some design modifications in order to reduce pilot load and increase the ship helicopter operation limit (SHOL). Computational results for these modified designs are also presented and their efficacy in reducing the turbulence levels and recirculation zone in the ship air wakes is discussed. 1 Graduate student 2 Professor 3 Associate Professor 3:48PM L34.00002 DNS of Supersonic Turbulent Flows in a DLR Scramjet Intake , XINLIANG LI, CHANGPING YU, LHD, Institute of Mechanics, Chinese Academy of Science (CAS) — Direct numerical simulation (DNS) of supersonic/hypersonic flow through a DLR scramjet intake GK01 is performed. The free stream Mach numbers are 3, 5 and 7, and the angle of attack is zero degree. The DNS cases are performed by using an optimized MP scheme with adaptive dissipation (OMP-AD) developed by the authors, and the blow-and-suction perturbations near the leading edge are used to trigger the transition. To stabilize the simulation, locally non-linear flittering is used in high-Mach-number case. The transition, separation, and shock-turbulent boundary layer interaction are studied by using both flow visualization and statistical analysis. A new method, OMP-AD, is also addressed in this paper. The OMP-AD scheme is developed by using joint MP method and optimized technique, and the coefficients in the scheme are flexible to show low dissipation in the smoothing region, and to show high robust (but high dissipation) in the large gradient region. Numerical tests show that the OMP-AD is more robust than the original MP schemes, and the numerical dissipation of OMP-AD is very low. 4:01PM L34.00003 Assessing the numerical dissipation rate and viscosity in CFD simulations of fluid flows1 , F.S. SCHRANNER, Technical University Munich, J.A. DOMARADZKI, University of Southern California, S. HICKEL, N.A. ADAMS, Technical University Munich — We describe a method for quantifying the effective numerical dissipation rate and the effective numerical viscosity in Computational Fluid Dynamics simulations. Differently from the previous approach that was formulated in spectral space, the proposed method is developed in a physical-space representation and allows for determining numerical dissipation rates and viscosities locally, i.e., at the individual cell level or for arbitrary subdomains of the computational domain. The method is self-contained using only results produced by the Navier-Stokes solver being investigated. Since no extraneous information is required, the method is suitable for a straightforward quantification of the numerical dissipation as a post-processing step. We demonstrate the method’s capabilities on the example of implicit large-eddy simulations of three-dimensional Taylor-Green vortex flows that exhibit laminar, transitional, and turbulent flow behavior at different stages of time evolution. For validation, we compare the numerical dissipation rate obtained using this method with exact reference data obtained with an accurate, spectral-space approach. 1 Supported by Deutsche Forschungsgemeinschaft and Alexander von Humboldt Foundation. 4:14PM L34.00004 Wall boundary condition for LES using particle method , AKIHIKO NAKAYAMA, Kobe University, NOBUYUKI HISASUE, Kansai Electric Power Company — In large eddy simulation that does not resolve the near-wall viscous sub-layer, the no-slip boundary condition cannot be applied and a wall model is needed. A method is proposed in which a special wall model is applied to the motion of particles that lie within a thin layer next to solid wall to accomplish the wall modeling in a SPH formulation. It is verified in fully developed open channel flows and is applied to dam break flow with reasonable results. 4:27PM L34.00005 Quantifying numerical dissipation rate in a commercial CFD code , GIACOMO CASTIGLIONI, J. ANDRZEJ DOMARADZKI, University of Southern California — Recently it has become increasingly clear that the role of a numerical dissipation, originating from the discretization of Navier-Stokes equations, rarely can be ignored regardless of the formal order of accuracy of a numerical scheme used in explicit or implicit Large Eddy Simulations (LES). The numerical dissipation inhibits the predictive capabilities of LES whenever it is of the same order of magnitude or larger than the subgrid-scale dissipation. The need to estimate the numerical dissipation is most pressing for lower order methods employed by commercial CFD codes. Following the recent work of Schranner et al. the equations and procedure for estimating the numerical dissipation rate and the numerical viscosity in a commercial code will be presented. The method allows to compute the numerical dissipation rate and numerical viscosity in the physical space for arbitrary sub-domains in a self-consistent way, using only information provided the code in question. It is the first time this analysis has been applied to low order solvers. Two equivalent ways to compute the numerical dissipation rate are described and compared. The procedure is tested for a 3D Taylor-Green vortex flow and compared with benchmark results obtained using an accurate, incompressible spectral solver. 4:40PM L34.00006 Large-eddy simulations of impinging jet with embedded vortices on rough surface , WEN WU, RAYHANEH BANYASSADY, UGO PIOMELLI, Queens Univ — Large-eddy simulations (LES) are used to study round jets impinging on rough surface at nozzle-to-plate distance H/D = 1 (D is the nozzle exit diameter) and Reynolds numbers Re = Uo D/ν = 6.6 × 104 (Uo is the mean jet velocity). Our aim is to explore the effect of roughness on the evolution of vortices in the analysis of impinging jet. Two cases, one with turbulent, the other with laminar inflow, are performed. Roughness is represented by uniformly distributed but randomly oriented ellipsoids of equivalent sand-grain height ks /D = 0.02, modelled by immersed boundary method. Results are compared to our previous LES simulations of jets impinging on a smooth surface. A wider and weaker wall jet is observed in the rough surface turbulent jet, compared to the reference turbulent one with smooth surface. The vortices and the peak of wall jet velocity shift away from the surface. Secondary vorticity is formed and lifted up, as in the smooth-surface case.The wall shear stress increases significantly; the separated vorticity, however, has the same strength as the one in the smooth case. The roughness causes higher turbulent fluctuations, and leads to the transition to turbulent wall jet even when the inflow is laminar, changing the vortex dynamics during vortex interaction. 4:53PM L34.00007 High-Order Velocity and Pressure Statistics from Direct Numerical Simulations of a Zero-Pressure-Gradient Turbulent Boundary Layer1 , BRYAN KAISER, SVETLANA POROSEVA, University of New Mexico — High-order turbulence statistics in a zero-pressure-gradient turbulent boundary layer are important for developing turbulence models. They also provide an insight into the physics of turbulent flows. A complete database of one-point statistics extracted from the dataset of direct numerical simulations (DNS) of a zero-pressure-gradient turbulent boundary layer by the Universidad Politecnica de Madrid Fluid Dynamics Group was collected. Third-, fourth-, and fifth-order velocity central moments and second-order pressure-velocity correlations at two Reynolds numbers of 4101 and 5200 based on the momentum thickness will be presented. DNS data are in a good agreement with experimental data. Results of the validation of Millionshtchikov’s hypothesis of quasi-normality and the Gram-Charlier series expansions for representing higher-order velocity moments in terms of lower-order moments will be reported. The two approaches are used as closing procedures in third- and fourth-order statistical closures, respectively. 1A part of the material is based upon work supported by NASA under award NNX12AJ61A. 5:06PM L34.00008 Compressible DNS study of separation bubbles for flow past a low pressure turbine blade1 , RAJESH RANJAN, SURESH DESHPANDE, RODDAM NARASIMHA, JNCASR — A representative low pressure turbine blade T106A is subjected to a direct numerical simulation (DNS) study for low Reynolds Number (Re = 51831 based on inflow velocity and axial chord) and angle of incidence (45.5 deg from the axial chord). The DNS code used here solves the compressible Navier-Stokes equations and uses a semi-kinetic energy preserving scheme. A hybrid grid is used for the computational domain, with a very fine wall-bounded boundary layer grid near the surface of the blade and an unstructured grid for rest of the domain. Total grid size for the current simulation is around 160 million. In the mean flow, a long but shallow separation bubble is found near the trailing edge. However, the instantaneous flow reveals a train of bubbles at this location. These instantaneous bubbles continually break and merge in time. The presence of these separation bubbles make the flow very complicated, as the bubbles are responsible for tripping the otherwise laminar flow to a transitional state. Skin friction and heat transfer co-efficient are also computed over the blade to understand the effect of these bubbles on parameters of engineering importance. 1 Supported by a GATET funded project on DNS of turbomachinery blading. The Param Yuva-II at CDAC was utilized for the simulations. 5:19PM L34.00009 Improving large-eddy simulation on adaptive mesh refinement grids using the turbulence closure , LAUREN GOODFRIEND, FOTINI CHOW, University of California, Berkeley, MARCOS VANELLA, ELIAS BALARAS, George Washington University — Large-eddy simulation (LES) and adaptive mesh refinement (AMR) reduce the computational cost of turbulence modeling by restricting resolved length scales, but combining these techniques generates additional errors. The grid refinement interfaces in AMR grids can create interpolation errors and reflect resolved energy. This talk will explore using the turbulence closure to mitigate grid interface errors in LES. Specifically, explicit filtering of the advection term and the mixed model are compared to implicit filtering and the eddy viscosity model. I will present a half-channel case study in which the domain is split into two structured static grids, one fine and one coarse. This simple test case allows observation of the effects of the grid interfaces. It is found that explicitly filtering the advection term allows both mass and momentum to be conserved across grid refinement interfaces by reducing interpolation errors. The mixed model decreases unphysical energy accumulation generated by wave reflection. These results inform the use of LES on block-structured non-uniform grids, including dynamic AMR grids. 5:32PM L34.00010 Mean wall shear stress boundary condition for large eddy simulation using near-wall streamwise momentum equation1 , MINJEONG CHO, HAECHEON CHOI, Seoul National University, JUNGIL LEE, Ajou University — The mean wall shear stress boundary condition based on the log law has been proven as an appropriate boundary condition for large eddy simulations (LES) of turbulent channel and boundary layer flows without resolving near-wall region (Lee, Cho & Choi, PoF, 2013). In the present study, we use near-wall streamwise momentum equation following Chung & Pullin (JFM, 2009), to determine the mean shear stress at the wall. In this procedure, the near-wall streamwise momentum equation is averaged over a few off-wall grid points, in which the velocity at the first grid point is approximated with the Taylor series expansion. We test this wall boundary condition for turbulent channel and boundary layer flows, showing good prediction capability at high Reynolds numbers. The result of applying this boundary condition to a separating flow will be also shown at the presentation. 1 Supported by NRF-2011-0028032, 2013055323. 5:45PM L34.00011 Tensor Diffusivity and Lagrangian Particle Methods , A. LEONARD, California Institute of Technology — We consider the tensor diffusivity model for large-eddy simulation and its connection with lagrangian particle methods for the vorticity transport equation and for the scalar advection equation. If φ is a scalar field being advected by an incompressible velocity field u then application of the tensor diffusivity 2 model results in the term − σ2 S¯ij ∂ 2 φ̄/∂xi ∂xj on the RHS of the filtered advection equation, where S¯ij is the filtered strain rate tensor and σ is the width of the gaussian filter. Previous investigators have shown, in a priori tests, that this model yields results that are relatively well correlated with the actual subgrid scale terms. Also previously noted is the fact that one has negative diffusion along the the principal axes of S¯ij that correspond to positive eigenvalues. Examples are given where the negative diffusivity produces meaningful backscatter of energy into the resolved scales. Also shown is that lagrangian particle methods have a truncation error that produces exactly the tensor diffusivity model. For 3D vorticity transport and scalar advection, the particle velocity must be slightly modified from the local velocity. 5:58PM L34.00012 A 3D Computational fluid dynamics model validation for candidate molybdenum-99 target geometry1 , LIN ZHENG, The University of New Mexico, GREG DALE, Los Alamos National Laboratory, PETER VOROBIEFF, The University of New Mexico — Molybdenum-99 (99 Mo) is the parent product of technetium-99m (99m Tc), a radioisotope used in approximately 50,000 medical diagnostic tests per day in the U.S. The primary uses of this product include detection of heart disease, cancer, study of organ structure and function, and other applications. The US Department of Energy seeks new methods for generating 99 Mo without the use of highly enriched uranium, to eliminate proliferation issues and provide a domestic supply of 99 mTc for medical imaging. For this project, electron accelerating technology is used by sending an electron beam through a series of 100 Mo targets. During this process a large amount of heat is created, which directly affects the operating temperature dictated by the tensile stress limit of the wall material. To maintain the required temperature range, helium gas is used as a cooling agent that flows through narrow channels between the target disks. In our numerical study, we investigate the cooling performance on a series of new geometry designs of the cooling channel. 1 This research is supported by Los Alamos National Laboratory. Monday, November 24, 2014 3:35PM - 6:11PM Session L35 Wind Turbines: General — 2001A - Raul Bayoan Cal, Oregon State University 3:35PM L35.00001 An adaptive lattice Boltzmann method for predicting turbulent wake fields in wind parks , RALF DEITERDING, German Aerospace Center (DLR) - Institute for Aerodynamics and Flow Technology, STEPHEN L. WOOD, University of Tennessee - Knoxville, The Bredesen Center — Wind turbines create large-scale wake structures that can affect downstream turbines considerably. Numerical simulation of the turbulent flow field is a viable approach in order to obtain a better understanding of these interactions and to optimize the turbine placement in wind parks. Yet, the development of effective computational methods for predictive wind farm simulation is challenging. As an alternative approach to presently employed vortex and actuator-based methods, we are currently developing a parallel adaptive lattice Boltzmann method for large eddy simulation of turbulent weakly compressible flows with embedded moving structures that shows good potential for effective wind turbine wake prediction. Since the method is formulated in an Eulerian frame of reference and on a dynamically changing nonuniform Cartesian grid, even moving boundaries can be considered rather easily. The presentation will describe all crucial components of the numerical method and discuss first verification computations. Among other configurations, simulations of the wake fields created by multiple Vesta V27 turbines will be shown. 3:48PM L35.00002 Effects of Offshore Wind Turbines on Ocean Waves , NICHOLAS WIMER, University of Colorado at Boulder, MATTHEW CHURCHFIELD, National Renewable Energy Laboratory, PETER HAMLINGTON, University of Colorado at Boulder — Wakes from horizontal axis wind turbines create large downstream velocity deficits, thus reducing the available energy for downstream turbines while simultaneously increasing turbulent loading. Along with this deficit, however, comes a local increase in the velocity around the turbine rotor, resulting in increased surface wind speeds. For offshore turbines, these increased speeds can result in changes to the properties of wind-induced waves at the ocean surface. In this study, the characteristics and implications of such waves are explored by coupling a wave simulation code to the Simulator for Offshore Wind Farm Applications (SOWFA) developed by the National Renewable Energy Laboratory. The wave simulator and SOWFA are bi-directionally coupled using the surface wind field produced by an offshore wind farm to drive an ocean wave field, which is used to calculate a wave-dependent surface roughness that is fed back into SOWFA. The details of this combined framework are outlined. The potential for using the wave field created at offshore wind farms as an additional energy resource through the installation of on-site wave converters is discussed. Potential negative impacts of the turbine-induced wave field are also discussed, including increased oscillation of floating turbines. 4:01PM L35.00003 Influence of pitch motion on the turbulent mixing in the wake of floating wind turbine models , STANISLAV ROCKEL, JOACHIM PEINKE, MICHAEL HOELLING, ForWind - Institute of Physics, University of Oldenburg, Germany, RAÚL BAYOÁN CAL, Portland State University — Offshore wind turbines use fixed foundations, which are economical in shallow water up to a depth of 50m. For deeper water areas floating support structures are feasible alternatives. The added degrees of freedom of a floating platform introduce additional oscillations to the wind turbine and therefore influence the aerodynamics at the rotor and its wake, respectively. The influence of platform pitch motion on the wake of an upstream wind turbine and a turbine positioned in the wake is investigated. Wind tunnel experiments were performed using classical bottom fixed wind turbine models and turbines in free pitch motion. Using 2D-3C particle image elocimetry (SPIV), wakes of both turbines were measured. In both cases - fixed and pitching - the inflow conditions were kept constant. The differences in the turbulent quantities of the wake of the upwind turbine for the fixed and oscillating case are investigated and their influence the wake of the downwind turbine. Our results show that platform pitch and oscillatory motions of the wind turbine have a strong impact on the shape of the fluctuating components of the wake. Also the turbulent mixing is changed by the oscillations, which is transferred to statistical quantities of higher order in the wake of the downwind turbine. 4:14PM L35.00004 Effects of wave induced motion on power generation of offshore floating wind farms , KOUROSH SHOELE, johns Hopkins University — Wind power has been the world’s fastest growing energy source for more than a decade. There is a continuous effort to study the potentials of offshore floating wind farms in producing electricity. One of the major technical challenges in studying the performance of offshore floating wind farms is the hydrodynamic and aerodynamic interactions between individual turbines. In this study, a novel approach is presented to study the hydrodynamic interaction between group of floating wind turbines and determine how wave induced motion of the platforms modifies the power generation of the farm. In particular, exact analytical models are presented to solve the hydrodynamic diffraction and radiation problem of a group of floating wind turbine platforms, to model the aerodynamic interaction between turbines, and to quantify the nonlinear dynamic of the mooring lines used to stabilize the floating platforms through connecting them to the seabed. The overall performance of the farm with different configuration and at different wind and wave conditions are investigated and the effects of the sea state condition as well as the distance between the turbines in the farm on the low frequency temporal variation of the power output are discussed. 4:27PM L35.00005 A concurrent precursor inflow method for LES of atmospheric boundary layer flows with variable inflow direction for coupling with meso-scale models1 , WIM MUNTERS, KU Leuven, Mechanical Engineering, B3001 Leuven, Belgium, CHARLES MENEVEAU, Johhns Hopkins University, Baltimore MD 21218, USA, JOHAN MEYERS, KU Leuven, Belgium — In order to incorporate multiple scales of meteorological phenomena in atmospheric simulations, subsequent nesting of meso-scale models is often used. However, the spatial and temporal resolution in such models is too coarse to resolve the three-dimensional turbulent eddies that are characteristic for atmospheric boundary layer flows. This motivates the development of tools to couple meso-scale models to Large-Eddy Simulations (LES), in which turbulent fluctuations are explicitly resolved. A major challenge in this area is the spin-up region near the inlet of the LES in which the flow has to evolve from a RANS-like inflow, originating from the meso-scale model, to a fully turbulent velocity field. We propose a generalized concurrent precursor inflow method capable of imposing boundary conditions for time-varying inflow directions. The method is based on a periodic fully-developed precursor boundary-layer simulation that is dynamically rotated with the wind direction that drives the main LES. In this way realistic turbulent inflow conditions are applied while still retaining flexibility to dynamically adapt to meso-scale variations in wind directions. Applications to wind simulations with varying inflow directions, and comparisons to conventional coupling methods are shown. 1 Work supported by ERC (ActiveWindFarms, grant no: 306471). CM is supported by NSF (grant no. 1243482). 4:40PM L35.00006 Towards an Experimental Investigation of Wind Turbine Aerodynamics at Full Dynamic Similarity , MARK A. MILLER, MARCUS HULTMARK, Princeton University — As horizontal axis wind turbines continue to increase in size (with the largest approaching 200 meters in diameter) it becomes progressively more difficult to test new designs without high computational power or extensive experimental effort using conventional tools. Therefore, compromises are often made between the important non-dimensional parameters (Reynolds number and Strouhal number, or tip speed ratio) so that reasonable engineering insight can be gained. Using the unique facilities available at Princeton University, we aim to match both non-dimensional parameters and thus achieve full dynamic similarity at realistic conditions. This is accomplished by using the High Reynolds number Test Facility (or HRTF), which is a high pressure (200 atmospheres) wind tunnel. We present the design, manufacture, and testing of an apparatus suited to the unique environment of a high-pressure facility as well as future plans for investigating the underlying aerodynamics of large-scale wind turbines. 4:53PM L35.00007 Volumetric characterization of the flow over miniature wind farms: An experimental study , LAI WING, DAN TROOLIN, TSI, JIN KIM HYUN, NICOLAS TOBIN, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, CARLO ZUNIGA ZAMALLOA, University of Illinois at Urbana-Champaign, LEONARDO P. CHAMORRO, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — An internal boundary layer is known to develop from the interaction between wind farms and the atmospheric boundary layer. It possesses characteristic features able to modulate the turbulence dynamics over large regions and eventually modify the micro climate in the vicinity of the wind farm. In this study, we examine the structure of the turbulence above various miniature wind farm configurations using 3D Particle Image velocimetry (PIV). Each miniature wind farm is placed in the boundary-layer wind tunnel at the Mechanical Science Engineering, UIUC. The turbines are fabricated using 3D printing and have a loading system that controls their tip-speed ratio and allows for characterizing the loads. Volumetric PIV is performed at various locations over and downstream a series of wind farm layouts. High-order turbulence statistics, turbulence structure and characteristic coherent motions are obtained and discussed in terms of the wind farm layout. 5:06PM L35.00008 Experimental Study of Fully Developed Wind Turbine Array Boundary Layer , JOHN TURNER V, MARTIN WOSNIK, Univ of New Hampshire — Results from an experimental study of an array of up to 100 model wind turbines with 0.25 m diameter, conducted in the turbulent boundary layer of the 6.0 m wide x 2.7 m tall x 72.0 m long test section of the UNH Flow Physics Facility, are reported. The study aims to address two questions. First, for a given configuration (turbine spacing, initial conditions, etc.), when will the model wind farm reach a “fully developed” condition, in which turbulence statistics remain the same from one row to the next within and above the wind turbine array. Second, how is kinetic energy transported in the wind turbine array boundary layer (WTABL). Measurements in the fully developed WTABL can provide valuable insight to the optimization of wind farm energy production. Previous experimental studies with smaller model wind farms were unable to reach the fully developed condition. Due to the size of the UNH facility and the current model array, the fully developed WTABL condition can be achieved. The wind turbine array was simulated by a combination of drag-matched porous disks, used in the upstream part of the array, and by a smaller array of realistic, scaled 3-bladed wind turbines immediately upstream of the measurement location. 5:19PM L35.00009 Boundary layer development over a large array of porous-disk-modeled wind turbines via stereo particle image velocimetry , ELIZABETH CAMP, VASANT VUPPULURI, RAÚL CAL, Portland State University — The increasing size of wind turbine arrays in service highlights the importance of understanding the flow physics within such large turbine arrays. Thus, the development of a wind turbine array boundary layer (WTBL) was investigated experimentally for an 8x5 array of model wind turbines. Model wind turbines were on a 1:2000 scale and turbine rotors were represented by porous disks. Stereoscopic Particle Image Velocimetry (SPIV) measurements were done along the centerline of the wind turbine array at several streamwise positions both within and above the canopy. Measurements and analysis of the mean and streamwise-averaged statistics of the SPIV fields focus on the rotors in the furthest downstream positions. Statistics will be used to determine if a fully developed WTBL has been achieved. 5:32PM L35.00010 Kite propulsion , EMMANUEL DU PONTAVICE, CHRISTOPHE CLANET, DAVID QUÉRÉ, Ladhyx, Ecole Polytechnique/PMMH ESPCI — Kite propulsion is one way to harvest wind energy. The typical force is 1 kilo Newton per square meter, which means that with kites in the range 100 to 1000 square meters, one is able to propel ships from the trawler to the tanker. Several scientific issues arise when trying to design kites of these sizes. They first need to take off and land autonomously. This leads to the use of kites with an inflatable structure that can be compact when stored but very rigid and light once in the air. For that matter, we studied the behavior of large inflatable structures under static and dynamic load. Then, the kite needs to stay in the air. However, it appears that under certain conditions, kites without active control tend to engage into large oscillations and eventually crash. Through wind tunnel experiments, we try to understand this flight behavior to find the conditions of stability. 5:45PM L35.00011 Design and Construction of a Hydroturbine Test Facility , ECE AYLI, BERAT KAVURMACI, HUSEYIN CETINTURK, ALPER KAPLAN, KUTAY CELEBIOGLU, SELIN ARADAG, YIGIT TASCIOGLU, TOBB University of Economics and Technology, ETU HYDRO RESEARCH CENTER TEAM — Hydropower is one of the clean, renewable, flexible and efficient energy resources. Most of the developing countries invest on this cost-effective energy source. Hydroturbines for hydroelectric power plants are tailor-made. Each turbine is designed and constructed according to the properties, namely the head and flow rate values of the specific water source. Therefore, a center (ETU Hydro-Center for Hydro Energy Research) for the design, manufacturing and performance tests of hydraulic turbines is established at TOBB University of Economics and Technology to promote research in this area. CFD aided hydraulic and structural design, geometry optimization, manufacturing and performance tests of hydraulic turbines are the areas of expertise of this center. In this paper, technical details of the design and construction of this one of a kind test facility in Turkey, is explained. All the necessary standards of IEC (International Electrotechnical Commission) are met since the test facility will act as a certificated test center for hydraulic turbines. 5:58PM L35.00012 Design of an Adaptive Power Regulation Mechanism and a Nozzle for a Hydroelectric Power Plant Turbine Test Rig1 , BURAK MERT, ZEYNEP AYTAC, YIGIT TASCIOGLU, KUTAY CELEBIOGLU, SELIN ARADAG, TOBB University of Economics and Technology, ETU HYDRO RESEARCH CENTER TEAM — This study deals with the design of a power regulation mechanism for a Hydroelectric Power Plant (HEPP) model turbine test system which is designed to test Francis type hydroturbines up to 2 MW power with varying head and flow(discharge) values. Unlike the tailor made regulation mechanisms of full-sized, functional HEPPs; the design for the test system must be easily adapted to various turbines that are to be tested. In order to achieve this adaptability, a dynamic simulation model is constructed in MATLAB/Simulink SimMechanics. This model acquires geometric data and hydraulic loading data of the regulation system from Autodesk Inventor CAD models and Computational Fluid Dynamics (CFD) analysis respectively. The dynamic model is explained and case studies of two different HEPPs are performed for validation. CFD aided design of the turbine guide vanes, which is used as input for the dynamic model, is also presented. 1 This research is financially supported by Turkish Ministry of Development. Monday, November 24, 2014 3:35PM - 6:11PM Session L36 Nano Flows I — Alcove A - Ali Beskok, Southern Methodist University 3:35PM L36.00001 Ultra-sensitive flow measurement in nanopores through pressure-driven particle translocation , ALESSANDRO SIRIA, CNRS-Univ Lyon1, ALESSANDRO GADALETA, University Lyon 1, ANNE-LAURE BIANCE, CNRS-Univ Lyon1, LYDERIC BOCQUET, University Lyon 1, INSTITUT LUMIERE ET MATIERE TEAM — The field of nanofluidics is of growing interest, both for applications and fundamental research. Nevertheless, this discipline still lacks a fundamental tool, i.e. the ability of measure the extremely small liquid flows in nanometric systems. This is especially aggravating, considering that one of the most interesting open problems in the field is the deviation of hydraulic permeability, in some systems from the values predicted by classical fluid mechanics. We propose a novel method for the measurement of pressure-driven flows in nanometric systems to characterize the translocation rate and dwell time of nanoparticles contained in a colloidal suspension. We are able to detect the passage of each nanoparticle across a nanopore by observing the sudden change in ionic current, and by analyzing the statistics of translocation events we can measure the permeability of the pore with high sensitivity and good accuracy. 3:48PM L36.00002 Experimental investigation of flow and slip transition in nanochannels1 , ZHIGANG LI, LONG LI, JINGWEN MO, Hong Kong Univ of Sci & Tech — Flow slip in nanochannels is sought in many applications, such as sea water desalination and molecular separation, because it can enhance fluid transport, which is essential in nanofluidic systems. Previous findings about the slip length for simple fluids at the nanoscale appear to be controversial. Some experiments and simulations showed that the slip length is independent of shear rate, which agrees with the prediction of classic slip theories. However, there is increasing work showing that slip length is shear rate dependent. In this work, we experimentally investigate the Poiseuille flows in nanochannels. It is found that the flow rate undergoes a transition between two linear regimes as the shear rate is varied. The transition indicates that the non-slip boundary condition is valid at low shear rate. When the shear rate is larger than a critical value, slip takes place and the slip length increases linearly with increasing shear rate before approaching a constant value. The results reported in this work can help advance the understanding of flow slip in nanochannels. 1 This work was supported by the Research Grants Council of the Hong Kong Special Administrative Region under Grant Nos. 615710 and 615312. J. Mo was partially supported by the Postgraduate Scholarship through the Energy Program at HKUST. 4:01PM L36.00003 Droplets and the three-phase contact line at the nano-scale. Statics and dynamics , PETR YATSYSHIN, DAVID SIBLEY, Imperial College London, NIKOS SAVVA, Imperial College London; Cardiff University, SERAFIM KALLIADASIS, Imperial College London — Understanding the behaviour of the solid-liquid-vapour contact line at the scale of several tens of molecular diameters is important in wetting hydrodynamics with applications in micro- and nano-fluidics, including the design of lab-on-a-chip devices and surfaces with specific wetting properties. Due to the fluid inhomogeneity at the nano-scale, the application of continuum-mechanical approaches is limited, and a natural way to remedy this is to seek descriptions accounting for the non-local molecular-level interactions. Density Functional Theory (DFT) for fluids offers a statisticalmechanical framework based on expressing the free energy of the fluid-solid pair as a functional of the spatially varying fluid density. DFT allows us to investigate small drops deposited on planar substrates whilst keeping track of the microscopic structural details of the fluid. Starting from a model of intermolecular forces, we systematically obtain interfaces, surface tensions, and the microscopic contact angle. Using a dynamic extension of equilibrium DFT, we investigate the diffusion-driven evolution of the three-phase contact line to gain insight into the dynamic behaviour of the microscopic contact angle, which is still under debate. 4:14PM L36.00004 Two phase flow of helium in single nanopipes , ANGEL VELASCO, CRYSTAL YANG, ZUZANNA SIWY, University of California, Irvine, M.E. TOIMIL-MOLARES, GSI Helmholtzzentrum fur Schwerionenforschung GmbH, PETER TABOREK, University of California, Irvine — We report measurements of pressure driven flow of liquid helium entering vacuum through a single pipe of nanometer scale diameter. Nanopores were fabricated by etching a single ion track in either PET or mica. A calibrated mass spectrometer was used to measure the flow rates of liquid helium through pipes with diameter ranging from 80 nm to 31 nm. The liquid evaporates inside or near the exit of the nanopipe. The flow of helium was studied from 0.5 K to the lambda point (2.18 K) and from the lambda point to above the critical point (5.2 K). Flow rates were controlled by changing the pressure drop across the pipe in the range 0-5 Atm. When the pressure in the pipe reached the saturated vapor pressure, an abrupt flow transition was observed. For normal helium a viscous flow model accounting for interfacial forces is used to determine its position inside the pipe [1]. The observed mass flow rates are consistent with no slip boundary conditions. In superfluid the flow is essentially independent of the pressure drop with a maximum critical velocity of 11 m/s. The critical velocity has temperature dependence consistent with the homogeneous nucleation of vortices. [1] A. E. Velasco, C. Yang, Z. S. Siwy, M. E. Toimil-Molares, and P. Taborek, Applied Physics Letters 105 (2014) 4:27PM L36.00005 Gas Flows in Nano-Scale Confinements , ALI BESKOK, Southern Methodist University, MURAT BARISIK, Izmir Institute of Technology — Most studies on gas transport in nano-scale confinements assume dynamic similarity with rarefied gas flows, and employ kinetic theory based models. This approach is incomplete, since it neglects the van der Waals forces imposed on gas molecules by the surfaces. Using 3D molecular dynamics (MD) simulations of force driven gas flows, we show the significance of the wall force field in nano-scale confinements by defining a new dimensionless parameter (B) as the ratio of the wall force-penetration length to the channel height. Investigation of gas transport in different nano-channels at various Knudsen numbers show the importance of wall force field for finite B values, where the dynamic similarity between the rarefied and nano-scale gas flows break down. Molecularly structured walls determine the bulk flow physics by setting a proper tangential momentum accommodation coefficient, and also determine transport in the near wall region. Gas nano-flows with finite B exhibit significant differences in the local density and velocity profiles, affecting the mass flow rate and the behavior of Knudsen’s minimum in nano-channels. 4:40PM L36.00006 Predicting the Anomalous Density of a Dense Fluid Confined within a Carbon Nanotube , GERALD WANG, NICOLAS HADJICONSTANTINOU, MIT — The equilibrium density of fluids under nanoconfinement can be substantially smaller than their bulk density. Understanding the physical basis for and magnitude of these anomalous densities is key to many nanoengineering applications, such as constructing a sub-continuum model of nanoscale fluid flow. We provide here a theoretical description of this phenomenon in the most frequently, perhaps, studied system - a dense fluid confined within a carbon nanotube (CNT). We show that the reduced density is primarily due to repulsive interactions between the fluid and the CNT, which modify the fluid structure near the fluid-CNT interface and lead to a “stand-off” distance between the two materials. Using a mean-field approach to describe the energetic landscape near the CNT wall, we obtain closed-form analytical results describing the length scales associated with the layered fluid. Combined with empirical knowledge of the layered-fluid density, these results allow us to derive a prediction for the equilibrium fluid density as a function of the CNT radius that is in excellent agreement with molecular dynamics simulations. We also show how aspects of this theory can be extended to describe water confined within CNTs and find good agreement with results from the literature. 4:53PM L36.00007 Molecular dynamics simulation of oxygen flows in graphene nanochannels , HARUKA YASUOKA, RYO TAKAHAMA, MASAYUKI KANEDA, KAZUHIKO SUGA, Osaka Prefecture Univ — MD simulations are performed to investigate the flow characteristics of oxygen flows in graphene nanochannels. For comparison, flows of argon molecules which have relatively similar values of mass and diameter are also simulated. The L-J potential is used for the fluid-fluid interaction and the wall-fluid interaction. For the bond of the carbon molecules for the channel walls, the Brenner potential is used. For all the cases, the normalized number density, pressure and temperature are set as ρ = 0.2, P = 0.4 T = 2.0, Two channel height cases H = 20σ and H = 50σ, where σ is the argon molecule diameter, are considered. In those conditions, Knudsen numbers are estimated to be about 0.056 and 0.023. In both channel height cases, it is found that the oxygen flow rates are larger than those of the argon flows even though acceleration acting on fluid molecules is constant. This is because the wall-fluid interaction between oxygen and carbon molecules is weaker than that of argon flow cases. It is found that the normalized velocity profiles are indifferent of the fluid molecules. Therefore, it can be said that the diatomic molecular structure of the fluid molecules does not have significant effects on the flow characteristics in the graphene nanochannels. 5:06PM L36.00008 Shear Viscous Response of Molecularly Thin Liquid Films1 , CHARLES TSCHIRHART, SANDRA TROIAN, Calfornia Institute of Technology, MC 128-95, Pasadena, CA 91125 — Fluids that exhibit Newtonian response at the macroscale can display interesting deviations at the nanoscale caused by internal fluid microstructure or conformational entropy reduction near an adjacent solid boundary. Such deviations signal the breakdown of the continuum and isotropic fluid approximations at molecular length scales. These effects are particularly pronounced near the interface separating a liquid film from a supporting solid substrate where molecular layering in the fluid can result in inhomogeneity in the shear viscosity. Here we describe ellipsometric measurements of the surface deformation of non-volatile liquid nanofilms subject to a constant interfacial shear stress. For simple Newtonian response, the slope of the deformation can be used to extract the value of the shear viscosity in ultrathin films, which in our experiments range from 2 - 200 nm in thickness. For complex films, we observe deviations from linear deformation which require augmentation of the analytic model normally used to describe the viscous response. These findings may be helpful for improved parametrization of the shear response of supported free surface films as well as course grained models for nanofluidic applications. 1 Support from the Fred and Jean Felberg and Winifred and Robert Gardner Summer Undergraduate Research Fellowships is gratefully acknowledged. 5:19PM L36.00009 Measurement of velocity distribution of fluid flows in nanochannel using evanescent wave-based particle velocimetry , YUTAKA KAZOE, YOJIROU HIRAMATSU, KAZUMA MAWATARI, TAKEHIKO KITAMORI, The University of Tokyo, KITAMORI TEAM — The field of nanofluidics for single molecule analysis, ultra filtration and energy conversion has been expanded with recent micro- and nanotechnology. Since liquids in nanospace with dominant surface effects are in a transitional regime from single molecules to continuum, specific fluid properties different from bulk can be expected. Previously, our group has revealed unique properties in size-regulated 10-1000 nm spaces such as higher viscosity, lower dielectric constant and higher proton mobility. However, fluid flows in the nanochannel are still unknown owing to lack of measurement method because nanochannel is smaller than light wavelength. For breaking through the limitation, evanescent wave light, which exponentially penetrates from the surface within 100 nm-order distance, is a key optical phenomenon. In this study, we developed evanescent wave-based particle tracking method for measuring flow profile in nanochannel. 10 nm-order fluorescent tracer materials were used in the measurements, and the position of tracer in the nanochannel was estimated from the brightness. The method was demonstrated in measurements of pressure driven flows in a nanochannel. 5:32PM L36.00010 Continuum Simulations of Water Flow in Carbon Nanotube Membranes , J.H. WALTHER, Technical University of Denmark, Denmark, A. POPADIC, National Institute of Chemistry, Ljubljana, Slovenia, P. KOUMOUTSAKOS, ETH Zurich, Switzerland, M. PRAPROTNIK, National Institute of Chemistry, Ljubljana, Slovenia — We propose the use of the Navier-Stokes equations subject to partial-slip boundary conditions to simulate water flows in Carbon NanoTube (CNT) membranes. The finite volume discretisations of the Navier-Stokes equations are combined with slip lengths extracted from Molecular Dynamics (MD) simulations to predict the pressure losses at the CNT entrance as well as the enhancement of the flow rate in the CNT. The flow quantities calculated from the present hybrid approach are in excellent agreement with pure MD results while they are obtained at a fraction of the computational cost. The method enables simulations of system sizes and times well beyond the present capabilities of MD simulations. Our simulations provide an asymptotic flow rate enhancement and indicate that the pressure losses at the CNT ends can be reduced by reducing their curvature. More importantly, our results suggest that flows at nanoscale channels can be described by continuum solvers with proper boundary conditions that reflect the molecular interactions of the liquid with the walls of the nanochannel. 5:45PM L36.00011 Early regimes of water imbibition in nanoslit silica channels1 , ELTON OYARZUA, HARVEY ZAMBRANO, Universidad de Concepcion, JENS HONORE WALTHER2 , Technical University of Denmark, ANDRES MEJIA, Universidad de Concepcion — Capillarity is currently subject to a significant research interest. Attention is mainly paid to the late stage of the imbibition when a developed flow is reached and the Laplace pressure is balanced by the viscosity. Nevertheless, as the miniaturization of devices is reaching the nanoscale a thorough understanding of fluid flow in nanoconfinement is required. In nanofluidics, short timescales and surface characteristics dominate the flows. In this study, molecular simulations are conducted to investigate the early stage of water imbibition in silica nanochannels with heights of 4 to 10 nm. Results indicate that nanoscale imbibition is divided in three regimes. An initial regime with imbibition linearly dependent of time, where the capillary force is mainly balanced by inertia. Thereafter, a period, in which, the balance has contributions from both inertia and viscosity and, subsequently, a final regime, wherein, viscosity dominates the capillary force balance. Velocity profiles confirm the passage from an inviscid flow to a developed Poisseuille flow. The meniscus position as a function of time and air accumulation in front of the advancing meniscus are computed for different air pressures, the results reveal a systematic retarding effect of gas pressurization on the imbibition. 1 We acknowledge support from Fondecyt project No 11130559 at Chair of Computational Science ETH Zurich 2 Also 5:58PM L36.00012 ABSTRACT WITHDRAWN — Tuesday, November 25, 2014 8:00AM - 9:44AM Session M1 Granular Flows: Impact and Force Transmission — 3000 - Parisa Mirbod, CUNY - Queens College 8:00AM M1.00001 High-speed granular flows around a cylindrical obstacle , JOSHUA S. CAPLAN, STUART B. DALZIEL, University of Cambridge, JIM N. MCELWAINE, University of Durham, NATHALIE M. VRIEND, University of Cambridge — Geophysical granular flows are extremely destructive, but their behavior when they impact on obstacles in their path is still poorly understood. In this talk we will present the results of a series of experiments where we consider the granular flow around a cylindrical obstacle, extending previous work by Cui and Gray (2013). By using a unique recirculating chute, we are able to consider flows of up to 20 kg s-1 at speeds of several meters per second. This gives us access to flow regimes that could not be previously considered. As has been previously observed, we find a large bow shock ahead of the obstacle and a granular vacuum behind it. We, however, find that the shock shapes are significantly different to previous observations. We will also be presenting PIV measurements of the surface velocity, height profiles of the flow, and measurements of the forces on the obstacle. 8:13AM M1.00002 Granular Impact at High Mach Number1 , ABE CLARK2 , YUE ZHANG, Duke University, LOU KONDIC, New Jersey Institute of Technology, R.P. BEHRINGER, Duke University — How do dynamic stresses propagate in granular material after a high-speed impact? This occurs often in natural and industrial processes. Stress propagation in a granular material is controlled by the inter-particle force law, f , in terms of particle deformation, δ, often given by f ∝ δ α , with α > 1. This means that a linear wave description is invalid when dynamic stresses are large compared to the original confining pressure. With high-speed video and photoelastic grains with varying stiffness, we experimentally study how forces propagate following an impact and explain the results in terms of the nonlinear force law (we measure α ≈ 1.4). The spatial structure of the forces and the propagation speed, vf , depend on a dimensionless parameter, M ′ = tc v0 /d, where v0 is the intruder speed at impact, d is the grain diameter, and tc is a binary collision time between α−1 grains with relative speed v0 . For M ′ ≪ 1, propagating forces are chain-like, and the measured vf ∝ d/tc ∝ vb (v0 /vb ) α+1 , where vb is the bulk sound speed. For larger M ′ , the force response has a 2D character, and forces propagate faster than predicted by d/tc due to collective stiffening of a packing. 1 Funded by DTRA, under Grant No. HDTRA1-10-0021. at Yale University 2 Currently 8:26AM M1.00003 Forces and flows during high speed impacts on a non-Newtonian suspension , MELODY LIM, JONATHAN BARES, ROBERT BEHRINGER, Duke University — A suspension made of starch particles dispersed in water displays significant non-Newtonian behavior for high enough particulate concentration. In order to shed light on the possible micro-structural basis of this behavior, we perform collisions on a quasi-2D suspension, using a high speed camera to gain access to the dynamics of the suspension. We suspend small dark particles (charcoal) in the cornstarch suspension. From these, we can carry out particle tracking to determine the velocity field during impact. We observe a shock-like propagation in the cornstarch suspension. Although the dynamics of this shockfront are strongly correlated to the dynamics of the intruder, we find that a simple process of momentum transfer to the suspension is insufficient to account for the force experienced by the impactor. We use boundaries made from a photoelastic material which then registers the arrival of strong forces at the boundaries. By linking the forces observed at the boundaries with the dynamics of the suspension, we assess the role of interactions with the boundaries of the suspension. 8:39AM M1.00004 Head-on collisions of Newtonian and granular jets , JAKE ELLOWITZ, WENDY W. ZHANG, Department of Physics and the James Franck Institute at the University of Chicago — When a wide fluid jet collides head-on with a narrow jet, incompressibility, together with energy and momentum conservation requires that the excess forward momentum flux be transported away from the impact zone by two identical symmetrically-angled ejecta streams. The central impact zone itself remains fixed. Introducing any kind of dissipation breaks time-reversal symmetry, thus allowing the excess forward momentum flux to be partitioned between the ejecta streams and the impact zone. Such a partition would cause the impact zone to drift steadily over time rather than remaining fixed. Motivated by the potential relevance of this mechanism to splash formation and microreactors from impinging jets, we simulate head-on collisions of two Newtonian jets and compare it against collisions of two densely packed granular jets. In both cases, the steady-state solutions display impact zones moving with finite drift speeds. Increasing the dissipation, either by increasing the viscosity of the Newtonian liquid or by increasing the coefficient of friction in the granular system, increases the drift speed. To our surprise, plotting the drift speeds against the total dissipation rates collapses results for Newtonian and granular jets. 8:52AM M1.00005 Does the Fluid Matter? Impact Into Wet Granular Materials , KERSTIN NORDSTROM, Mount Holyoke College, DYLAN POWERS, SAM ARRINGTON, WOLFGANG LOSERT, University of Maryland — We study the impact of a projectile onto a bed of 3 mm grains immersed in a fluid. We vary the viscosity of the fluid, and see how the impact depth vs impact energy scaling changes. We find only an appreciable change when the viscosity is quite large. We also study the trajectory of the intruder for different viscosities and impact energies. We find these trajectories are well-described by a modified version of the Poncelet-type stopping force model. 9:05AM M1.00006 Granular impact cratering by liquid drops: Understanding raindrop imprints through an analogy of asteroid strikes , XIANG CHENG, RUNCHEN ZHAO, QIANYUN ZHANG, HENDRO TJUGITO, University of Minnesota — When a granular material is impacted by a sphere, its surface deforms like a liquid yet it preserves a circular crater like a solid. Although the mechanism of granular impact cratering by solid spheres is well understood, our knowledge on granular impact cratering by liquid drops is still very limited. Using high-speed photography, we investigate liquid-drop impact dynamics on granular media. Surprisingly, we find that granular impact cratering by liquid drops follows the same energy scaling as that of asteroid impact cratering. Inspired by this similarity, we develop a simple model that quantitatively describes the observed crater morphologies. Our study sheds light on the mechanisms governing raindrop impacts on granular surfaces and reveals an interesting analogy between familiar phenomena of raining and catastrophic asteroid strikes. 9:18AM M1.00007 Influence of particle shape on properties of force networks in particulate systems1 , LOU KONDIC, NJIT, LUIS PUGNALONI, Universidad Tecnológica Nacional, La Plata, Argentina, MANUEL CARLEVARO, Instituto de Fı́sica de Lı́quidos y Sistemas Biológicos, La Plata, Argentina, MIROSLAV KRAMAR, KONSTANTIN MISCHAIKOW, Rutgers University — Simulations of particulate systems usually consider circular or spherical particles due to computational simplicity. Realistic particles however are often not circular or spherical, posing an important question: to what degree particles’ shape influences mechanical properties of the corresponding systems? To start answering this question, we carry out MD simulations of circular and polygonal frictional particles exposed to tapping and analyze the resulting force networks. In addition to using classical measures, we carry out topological analysis that allows us to describe and quantify structural properties of the considered networks. Perhaps surprisingly, topological analysis allows us to identify the differences between systems that appear undistinguishable based on the classical measures. 1 Supported by NSF Grant No. DMS-0835611. 9:31AM M1.00008 Force transmitted to a subsurface due to particle-laden liquids , ERIK WORDEN, REZA GHEISARI, PARISA MIRBOD, Department of Mechanical and Aeronautical Engineering, Clarkson University, Potsdam, New York, 13699 — In this study, we investigate force transmission due to a layer of neutrally buoyant suspension on a substrate. By applying a constant force on a solid body and pushing it through the suspension, some indentations were produced. The profile of the indentations and its relation to the size and squeezing speed of the solid body for three different volume fractions was determined. The dependence of the indentation depth on the compression height was examined and the variation of the indentations depth with strain rate was also investigated. Finally, by characterizing the response of the substrate to deformation, the force transmitted through the suspension was examined and compared to the applied force. We also studied the effect of the substrate material, solid body shape and squeezing speed, concentration of suspension, and the onset of plastic deformation. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M2 Surface Tension Effects: Textured Substrates — 3002 - Jonathan P. Rothstein, University of Massachusetts 8:00AM M2.00001 Capillary trapping on a rough surface1 , JASON WEXLER, IAN JACOBI, MELISSA CHOW, HOWARD STONE, Princeton University — Recent research has shown that rough or patterned surfaces infused with a lubricating liquid can display superhydrophobic properties. However, if such a surface is exposed to external flow, the shear induced by the outer fluid can drain the lubricating layer, causing regions of the surface to transition to a hydrophilic Wenzel state. In addition, the high specific gravity of lubricating liquids means that this loss can be driven by gravity alone, in the absence of flow. We examine the shear- and gravity-driven failure modes of liquid-infused patterned surfaces experimentally, and develop a unified model to predict the dynamics of drainage via these two types of forcing. We find that the dynamic evolution of the two drainage mechanisms takes on a single functional form. Under the influence of gravity, we show that a finite length of the surface will remain filled indefinitely; this is a variant of the familiar capillary rise height. Under the influence of external shear, the steady-state liquid retention depends on the outer flow velocity field. 1 This work was supported under ONR MURI Grants N00014-12-1-0875 and N00014-12-1-0962 (Program Manager Dr. Ki-Han Kim). 8:13AM M2.00002 In situ observations of wetting transition on submerged microstructured hydrophobic surfaces , PENGYU LV, YAHUI XUE, Peking University, HAO LIN, Rutgers, The State University of New Jersey, HUILING DUAN, Peking University — Superhydrophobicity of microstructured surfaces has a promising application in drag reduction. The air pockets trapped in the microstructures play the key role. However, the wetting transition from Cassie to Wenzel state will spontaneously take place due to air diffusion into the water around under pressurization, leading to the loss of air pockets and the failure of superhydrophobicity. The current work examines in situ liquid-air interfaces on a submerged surface patterned with cylindrical micropores using confocal microscopy. The dynamic process of wetting transition are directly observed and measured quantitatively, and the data are in good agreement with a diffusion-based model prediction. A similarity law along with a characteristic time scale is derived, which governs the lifetime of the air pockets. Moreover, two kinds of collapses of the menisci in the final stage of transition which refer to the symmetric and asymmetric collapses are also captured. A strategy of hierarchical structures is proposed to avoid the loss of stability of the liquid-air interfaces in advance due to asymmetric collapse. The present work enables a better prediction of underwater superhydrophobicity, and benefits design explorations to enhance its longevity. 8:26AM M2.00003 Force of Adhesion upon loss of Contact Angle Hysteresis: when a Liquid behaves like a Solid , JUAN ESCOBAR, Physics Department, Universidad Autónoma Metropolitana-Iztapalapa, México City, 09340, México, ROLANDO CASTILLO, Physics Department, Universidad Nacional Autonoma de México, UNAM — Liquids and solids are in general expected to behave very differently in their contact with a solid surface. While that the mechanical deformation of an elastic solid sphere is perfectly reversible, a liquid drop normally deforms in an irreversible way. Nevertheless, a liquid drop in contact with a perfectly-solvophobic surface should also deform reversibly, giving rise to loss of contact angle hysteresis. In this work,1 the theoretically predicted vanishment of the macroscopic contact angle hysteresis is found experimentally along with a small but finite force of adhesion 0.55 µN that, unexpectedly, is independent of the history of the preload. These results are obtained with a novel Capillary Force Microscope. Our results agree with the prediction of a model in which the surface tension of the liquid provides the counterpart of the restoring force of an elastic solid, evidencing that the dewetting of a liquid in the absence of strong pinning points is equivalent to the detachment of an elastic solid. 1 Escobar J.V. and Castillo R., Phys. Rev. Lett., 111, 22, 226102, (2013) 8:39AM M2.00004 Thermocapillary Driven Droplet Motion on Lubricant Impregnated Textured Surfaces , NADA BJELOBRK1 , Massachusetts Institute of Technology, HENRI-LOUIS GIRARD2 , Ecole Polytechnique, HYUK-MIN KWON, Massachusetts Institute of Technology, DAVID QUERE, ESPCI Paris Tech, KRIPA K. VARANASI, Massachusetts Institute of Technology — Here, we show that lubricant impregnated surfaces (LIS) promote thermocapillary induced motion of liquid droplets. The contact angle of a droplet on LIS is low, which increases the effective temperature gradient over the droplet. At the same time, the contact angle hysteresis is significantly decreased and thus the pinning is suppressed. The shear forces due to the thermocapillary effect at the free interface between the droplet and the lubricant can enhance the propulsion velocity by an order of magnitude. Hereby, two aspects are are supporting the propulsion: A lubricant/droplet pair should have a large and positive interfacial tension gradient. The lubricant should also fully encapsulate the substrate. We compare various lubricant viscosities and droplet sizes to examine the thermocapillary effect and propose a model to predict the velocity of droplets moving on LIS on a temperature gradient. 1 equal 2 equal contribution contribution 8:52AM M2.00005 Nanostructured surfaces and the dynamics of colloidal particles, droplets, and slugs , CARLOS COLOSQUI, Stony Brook University, JOEL KOPLIK, Levich Institute and Department of Physics, City College of CUNY, JEFFREY MORRIS, Levich Institute and Department of Chemical Engineering, City College of CUNY, ANTONIO CHECCO, Soft Matter Group, Brookhaven National Laboratory — Nanoscale heterogeneities in physical and/or chemical surface properties can have major consequences in the dynamics of colloidal particles at a liquid-fluid interface or femto/picoliter droplets and slugs on a solid substrate. For example, nanoscale heterogeneities can lead to crossovers from fast exponential to slow logarithmic adsorption of colloidal particles at interfaces or membranes, as well as self-propelled motion of microdroplets and slugs on nanopatterned substrates or capillaries. Theoretical models based on continuum thermodynamics can be extended to describe these phenomena at the colloidal scale, while molecular dynamics simulations can assist by providing critical insights into the coupling between thermal fluctuations, interfacial forces, and hydrodynamics at the nanoscale. This talk will present recent theoretical, numerical, and experimental results that (i) document transitions between different dynamic regimes, and (ii) establish relations between physical parameters and characteristic scales in the dynamics of colloidal particles, microdroplets, and slugs induced by nanoscale surface heterogeneities. 9:05AM M2.00006 Elastocapillary assembly of silver nanotube forest , XIN YANG, MIN PACK, YING SUN, Drexel University — Nanorods/nanotubes have large surface areas making them promising for applications such as high-performance battery and capacitor electrodes, photovoltaics, and interconnects. In this study, we demonstrate the formation of 3D microarchitectures via elastocapillary self-assembly of silver nanotube forests. Patterned silver nanotube forests of different lengths and diameters are made by inkjet printing of silver nanoinks into nanoporous anodic aluminum oxide membranes. These silver nanotube forests are then self-assembled into ordered microstructures via capillary forces induced by liquid condensation, which is compared with immersing nanotubes directly into a liquid. The effects of length, diameter, and footprint of the nanotube forest on self-assembled patterns are systematically studied. By decreasing the footprint and/or increasing the length of nanotube forest, the stiffness of the nanotube forest decreases, bringing the nanotubes together to form closely packed microstructures. 9:18AM M2.00007 Programming self assembly by designing the 3D shape of floating objects , MARTIN POTY, GUILLAUME LAGUBEAU, GEOFFROY LUMAY, NICOLAS VANDEWALLE, GRASP, Physics Department B5a, University of Liège, B-4000 Liège, Belgium — Self-assembly of floating particles driven by capillary forces at some liquid-air interface leads to the formation of two-dimensionnal structures. Using a 3d printer, milimeter scale objets are produced. Their 3d shape is chosen in order to create capillary multipoles. The capillary interactions between these components can be either attractive or repulsive depending on the interface local deformations along the liquid-air interface. In order to understand how the shape of an object deforms the interface, we developed an original profilometry method. The measurements show that specific structures can be programmed by selecting the 3d branched shapes. 9:31AM M2.00008 Discrete and continuum dynamics of elastocapillary coalescence of plates and pillars1 , ZHIYAN WEI, School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA, T. SCHNIEDER, Max Planck Institute, Am Fassberg 17, D-37077 Goettingen, Germany, J. KIM, H.-Y. KIM, Department of Mechanical and Aerospace Engineering, Seoul National University, Seoul 151-744, Korea, J. AIZENBERG, School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA, L. MAHADEVAN, School of Engineering and Applied Sciences, Department of Physics, Harvard University, Cambridge, MA 02138, USA — When a fluid-immersed array of lamellae or filaments that is attached to a substrate is dried, evaporation leads to the formation of menisci on the tips of the plates or pillars that bring them together. Building on prior experimental observations, we use a combination of theory and computation to understand the nature of this instability and its evolution in both the two-dimensional and three-dimensional setting of the problem. For the case of lamellae, we derive the interaction torques based on the relevant physical parameters, predict the critical volume of the liquid and the 2-plate-collapse eigenmode at the onset of instability, and use numerical simulations to explain the hierarchical cluster formation and characterize the sensitive dependence of the final structures on the initial perturbations. We also characterize these at a continuum level by partial differential equations. We then generalize our analysis to treat the problem of pillar collapse in 3D, where the fluid domain is completely connected. 1 We thank the Harvard-MRSEC DMR -0820484 and the MacArthur Foundation (LM) for support. 9:44AM M2.00009 Interfacial transport and orientation of Janus nanoparticles under shear flow , HOSSEIN REZVANTALAB, SHAHAB SHOJAEI-ZADEH, Rutgers University — We investigate the configuration of Janus nanoparticles adsorbed at liquid-fluid interfaces in presence of shear flow. The interfacial behavior of nanoparticles with different size and shape is followed using Molecular Dynamics simulations. We create Janus nanoparticles by tuning the affinity of the atoms on each side of the particle with the two fluids, and model the shear as a symmetric Couette flow. We demonstrate the existence of a steady-state orientation for particles at the interface, which is governed by the balance between the shear-induced torque and the resistance due to capillary forces. There is a threshold shear rate above which the nanoparticle starts to rotate out of its energetically favorable upright orientation. This threshold is found to to be higher for more amphiphilic particles due to the stronger interfacial tension resisting against the shear-induced disturbance. Moreover, the geometry plays a significant role in defining the range of attainable orientations. Janus cylinders with high aspect ratio show a sudden shift in orientation which does not drastically change with increasing shear rate, while thin discs require a larger threshold shear but can achieve a wider range of orientations. Our analysis of a fluid flow that reorients suspended nanoparticles can also give insight into the hemodynamics of blood flow and the interaction of anisotropic drug carriers with cell membranes. 9:57AM M2.00010 The dynamics and breakup of water streams flowing down an inclined superhydrophobic surface1 , JONATHAN ROTHSTEIN, ELIZABETH BAUMHOFF, University of Massachusetts - Amherst — In this talk, we present a series of experiments investigating the flow of water streams down a series of hydrophobic and superhydrophobic surfaces. To create the superhydrophobic surfaces, random texture was imparted onto a Teflon surface by sanding it with sand papers with a range of grit sizes. Our previous work has showed that there exists an optimal sand paper grit (240 grit) for eliminating contact angle hysteresis and reducing drag. The effect of advancing contact angle, contact angle hysteresis, plate inclination and flow rate on the shape of the meandering streams of water will be presented. We will show that the dynamics and breakup of water streams flowing down superhydrophobic surfaces is strongly dependent on contact angle hysteresis. We will show that decreasing the contact angle hysteresis makes the rivulets less stable resulting in an increased number of bends, more side-to-side motion of the stream and a reduction in the length of the stream at the moment it breaks up into drops. Additionally, decreasing hysteresis also results in a reduction in the radius of curvature of the bends observed along the meandering stream. Finally, we will show that at high flow rates, ejection of an intact liquid stream from the superhydrophobic surface can be observed. 1 NSF CBET-1334962 Tuesday, November 25, 2014 8:00AM - 10:10AM Session M3 Electrokinetics: Instability and Chaos — 3004 - Ali Mani, Stanford University 8:00AM M3.00001 Electroconvective Instability in Flow-shear-induced Transport Barrier: Threshold for Stable Vortices and Chaotic Eddies1 , RHOKYUN KWAK, KIST / Massachusetts Institute of Technology, VAN SANG PHAM, JONGYOON HAN, Massachusetts Institute of Technology — Suppression of turbulence and transport by shear flow is a common process in plasma fluid dynamics, while it has been rarely observed in nonionized fluids. Here, we visualize this effect in microfluidic nonionized system with electroconvective instability (EC) initiated by ion concentration polarization on ion selective membrane. The membranes act as the source of both instability and flow shear (wall shear of Hagen-Poiseuille (HP) flow) simultaneously, fitting the requisite for this shear suppression effect; turbulence in the domain of flow shear. To the best of our knowledge, this is the first characterization of flow-shear-induced transport barrier in microfluidics, captured by scaling analysis, experiment, and numerical modeling. Selected by balancing flow shear and velocity fluctuation, which generated by HP flow and vortical EC, the threshold for shear suppression scales by EC thickness dec /w <0.618. Stable unidirectional EC occurs under the threshold, while chaotic EC occurs over the threshold by overcoming flow shear. It also has significant implications on the energy saving of electrochemical systems (e.g. electrodialysis) to prevent chaotic turbulences and corresponding energy dissipations. 1 This work was supported by the Advanced Research Projects Agency-Energy grant (DE-AR0000294). 8:13AM M3.00002 Chaotic electroconvection near ion-selective membranes: investigation of transport dynamics from a 3D DNS , CLARA DRUZGALSKI, ALI MANI, Stanford University — We have investigated the transport dynamics of an electrokinetic instability that occurs when ions are driven from bulk fluids to ion-selective membranes due to externally applied electric fields. This phenomenon is relevant to a wide range of electrochemical applications including electrodialysis for fresh water production. Using data from our 3D DNS, we show how electroconvective instability, arising from concentration polarization, results in a chaotic flow that significantly alters the net ion transport rate across the membrane surface. The 3D DNS results, which fully resolve the spatiotemporal scales including the electric double layers, enable visualization of instantaneous snapshots of current density directly on the membrane surface, as well as analysis of transport statistics such as concentration variance and fluctuating advective fluxes. Furthermore, we present a full spectral analysis revealing broadband spectra in both concentration and flow fields and deduce the key parameter controlling the range of contributing scales. 8:26AM M3.00003 Electrokinetic Instability, Geometric Confinement, and Overlimiting Conductance , JARROD SCHIFFBAUER, Techinion, Israel Institute of Technology, MATHIAS BAEKBO ANDERSON, ALI MANI, Stanford University, GILAD YOSSIFON, Techinion, Israel Institute of Technology — For systems containing ion-selective membranes or nanochannels, concentration polarization (CP) under DC voltage beyond the classical Levich limit leads to the loss of local electroneutrality over micron or larger scales at the salt-depleted interface. This manifests itself in the appearance of an extended space charge (ESC) region, which is rendered unstable above a critical voltage drop. The instability drives the the formation of a fast-flowing vortex system with complex, often chaotic, dynamics. In unconfined systems, i.e. large electrolytic cells, this contributes strongly to the overlimiting conductance (OLC) of the system. However, both the role of the instability in OLC as well as its origin and onset become more complicated in highly confined systems such as microchannel devices. The problem of instability under geometric confinement has been studied both analytically and numerically using two different approaches. We compare the two approaches, and discuss relevant experimental evidence. 8:39AM M3.00004 Suitability of commercial software for direct numerical simulations of chaotic electrokinetic transport1 , ELIF KARATAY, ALI MANI, Stanford Univ — Many microfluidic and electrochemical applications involve chaotic transport phenomena that arise due to instabilities stemming from coupling of hydrodynamics with ion transport and electrostatic forces. Recent investigations have revealed contribution of a wide range of spatio-temporal scales in such chaotic systems similar to those observed in turbulent flows. Given that these scales can span several orders of magnitude, significant numerical resolution is needed for accurate prediction of these phenomena. The objective of this work is to assess efficiency of commercial software for prediction of such phenomena. To this end we have considered Comsol Multiphysics as a generalpurpose commercial CFD/transport solver, and have compared its performance against a custom-made DNS code tailored to the specific physics of chaotic electrokinetic phenomena [1]. We present comparison for small systems, which can be simulated on a single core, and show detailed statistics including velocity and concentration spectra over a wide range of frequencies. Our results indicate that while accuracy can be guaranteed with proper mesh resolution, commercial solvers are generally at least an order of magnitude slower than custom-made DNS codes. [1] Druzgalski, Andersen, and Mani, Phys. Fluids 25, 1 1 Supported by NWO, Rubicon Grant 8:52AM M3.00005 Analysis of chaotic electroconvection near electrodes1 , SCOTT DAVIDSON, ALI MANI, Stanford University — Electroconvective instability has recently been shown computationally to occur near electrode surfaces in induced-charge electro-osmosis (ICEO) in addition to its well-known occurrence near ion-selective membranes under large applied fields. This instability occurs due to the interaction of the extended space charge region of nonequilibrium electrical double layers with the applied field. The presence of the instability causes chaotic flow leading to order one changes to mean flow rates in ICEO as well as leading to flow between parallel electrodes where the fluid would otherwise remain stationary. We present direct numerical simulations (DNS) of the coupled Poisson-Nernst-Planck and Navier-Stokes equations analyzing both flow and transport effects in various regimes of the governing nondimensional parameters. 1 S. D. Supported by NSF GRFP, SGF 9:05AM M3.00006 Scaling law of velocity and conductivity in EK turbulence1 , WEI ZHAO, Department of Mechanical Engineering, University of South Carolina, FANG YANG, Carnegie Mellon University, GUIREN WANG, Department of Mechanical Engineering & Biomedical Engineering Program, University of South Carolina — In microfluidics, when electrokinetic (EK) flow is applied with sufficiently high electric Rayleigh number (Rae ), turbulence can be achieved, and there can even be an universal equilibrium range of conductivity field. In this flow, a new scaling law region of velocity and conductivity structures where the energy cascade is dominated by electric body force (EBF) can be found. This is similar to the Bolgiano-Obukhov scaling law (BO59) in Rayleigh-Bénard (RB) convection. By both directly analyzing Navier-Stokes (N-S) equation and dimensional analysis, the scaling exponent of the second order moment of velocity structure function is 2/5, while that of conductivity structures is 4/5. Compared to the buoyancy in RB convection which decreases with decreasing length scale, EBF actually increases with decreasing spatial scales. This leads to two different microscales depending on the strength of EBF. The scaling law of velocity fluctuation is verified experimentally in a micro-EK turbulent flow. Although due to the restriction of geometry of our microchannel, the bandwidth of the EBF dominant subrange is narrow. By adjusting Rae and other parameters, a wider EBF dominant subrange is predicable. 1 The work was supported by NSF under grant no. CAREER CBET-0954977 and MRI CBET-1040227, respectively. 9:18AM M3.00007 Electrokinetic flow characteristics of two fluids with different electrical conductivities in cross-shaped microchannels by the lattice-Boltzmann Method , AMADOR GUZMAN, Pontificia Universidad Catolica de Chile, ALVARO SOCIAS, DIEGO OYARZUN, Universidad de Santiago de Chile — Electrokinetic inestabilities (EKI) in microchannels flow are important to determine and characterize when either suppressing or enhancing flow features for injection and separation or mixing of multiple species are desired features. Convective and absolute electrokinetic instabilities (EKI) can be triggered or suppressed by active means such as externally applied AC or DC on the channel inlet, outlet and walls, and passively by building geometrical patterns on the wall channels such as grooves or waves. EKI are caused when a strong conductivity gradient between two fluids with different conductivities under an externally applied electric field becomes unstable. We model and simulate electrokinetic flow in a cross-shaped microchannel of two fluids with different electrical conductivity under an applied electrical field among the microchannel wells. We use the lattice-Boltzmann method (LBM) for solving the discretized Boltzmann Transport Equations (BTE) describing the coupled processes of hydrodynamics, electrodynamic and concentration of species of three fluids having different electrical conductivities under an external voltage in a cross-shaped microchannel with grooves in the outlet channel. Our numerical simulations predict well the conductivity gradient across the interface among the fluids and the unstable behavior of this interface when the local Rayleigh electrical number achieved, setting up EKIs. 9:31AM M3.00008 Electrohydrodynamic Instability of a Capacitive Elastic Incompressible Membrane1 , YUAN-NAN YOUNG, Department of Mathematical Sciences, New Jersey Institute of Technology, MICHAEL MIKSIS, ESAM, Northwestern University — The electrohydrodynamic instability of a leaky capacitive membrane in a direct current (DC) electric field, both perpendicular and parallel to the membrane in a micro-fluidic channel, is investigated theoretically. Under a parallel electric field, the membrane can be driven unstable with a vanishing membrane conductance. On the other hand a non-conducting capacitive membrane is always stable under a perpendicular electric field, and membrane conductance is essential for membrane instability due to a perpendicular electric field. The effects of membrane conductance, bending modulus, and charge relaxation time on the membrane instability are elucidated for several combinations of conductivity ratio and permittivity ratio in the bulk fluids. The tangential electric field acts similarly to the membrane tension in terms of its damping effects at small length scales (high wave number), while either bending or membrane tension is needed to damp out the small-scale perturbations under a perpendicular electric field. 1 YNY is supported by NSF DMS-1222550 and MM is supported by NSF DMS-1312935. 9:44AM M3.00009 Two layer flow of thin leaky dielectric films between electrodes , ELIZAVETA DUBROVINA, RICHARD CRASTER, DEMETRIOS PAPAGEORGIOU, Imperial College London — The flow of two viscous conducting fluids between two electrodes is investigated. The fluids are assumed to be leaky dielectrics and two nonlinear coupled evolution equations are derived for the moving interface and the interfacial charge. These are solved numerically for three different cases in which the magnitude of the ratios of electric conductivities and permittivities is varied. A linear stability analysis indicates that electrical forces destabilize the system. These predictions are confirmed by numerical results which show that increasing the ratios of conductivities and permittivities leads to traveling waves that grow in amplitude. 9:57AM M3.00010 Electro-hydrodynamic Stability of Electrified Jet , - DHARMANSH, PARESH CHOKSHI, Indian Institute of Technology Delhi — The axisymmetric stability of the straight jet in electrospinning process is examined for both Newtonian and polymeric fluids using leaky dielectric model. Contrary to previous studies which consider cylindrical jet as the base-state, in the present study the thinning jet profile obtained as steady-state solution of the 1D model is considered as the base-state. The linear stability of the thinning jet is analyzed for axisymmetric disturbances, which are believed to be responsible for the bead formation. The growth rate eigen-specturm is constructed using Chebyshev collocation method. Two different types of axisymmetric instability modes are observed, the Rayleigh mode and the conducting mode. Competition between these two modes is revealed for the thinning jet. The most unstable growth rate for thinning jet is found to be significantly different from that for the uniform jet. The role of various material and process parameters is also investigated. For the viscoelastic fluids, the thinning jet with non-uniform extension rate captures the role of nonlinear rheology of fluid in the stability behavior. The viscoelastic jet profile obtained from steady-state 1D model is analyzed for stability. The role of fluid elasticity on various instability modes is studied. Interestingly, the strain hardening behavior in polymer solution tends to suppress the instability producing smooth fibers. Also, increasing the polymer concentration exhibits stabilizing effect on the axisymmetric instability modes. Tuesday, November 25, 2014 8:00AM - 9:31AM Session M4 Bubbles: Collapse and Coalescence — 3006 - Gretar Tryggvason, University of Notre Dama 8:00AM M4.00001 Investigation of bubble-bubble interaction effect during the collapse of multi-bubble system1 , XUEMING SHAO, LINGXIN ZHANG, WENFENG WANG, Department of Mechanics, Zhejiang University — Bubble collapse is not only an important subject among bubble dynamics, but also a key consequence of cavitation. It has been demonstrated that the structural damage is associated with the rapid change in flow fields during bubble collapse. How to model and simulate the behavior of the bubble collapse is now of great interest. In the present study, both theoretical analysis and a direct numerical simulation on the basis of VOF are performed to investigate the collapses of single bubble and bubble cluster. The effect of bubble-bubble interaction on the collapse of multi-bubble system is presented. 1 The work was supported by the National Natural Science Foundation of China (11272284, 11332009) 8:13AM M4.00002 Temperature considerations in numerical simulations of collapsing bubbles1 , ERIC JOHNSEN, SHAHABODDIN ALAHYARI BEIG, University of Michigan — In naval and biomedical engineering applications, the inertial collapse of cavitation bubbles is known to damage its surroundings. While significant attention has been dedicated to investigating the pressures produced by this process, less is known about heating of the surrounding medium, which may be important when collapse occurs near objects whose properties strongly depend on temperature (e.g., polymers). Euler simulations are capable of predicting the high pressures thereby generated. However, numerical errors can occur when solving the Navier-Stokes equations for compressible interface problems. Using a newly developed computational approach that prevents such errors, we investigate the dynamics of shock-induced and Rayleigh collapse of individual and collections of gas bubbles, in a free field and near rigid surfaces. We characterize the temperature rises based on the relevant non-dimensional parameters entering the problem. In particular, we show that the temperature of a neighboring object rises due to two mechanisms: the shock produced at collapse and heat diffusion from the hot bubble as it moves toward the object. 1 This work was supported by ONR grant N00014-12-1-0751. 8:26AM M4.00003 Simulations of bubble collapse in viscous and viscoelastic media near a second viscoelastic medium , MAURO RODRIGUEZ, ERIC JOHNSEN, University of Michigan, Ann Arbor — Understanding the dynamics of cavitation bubbles and the shock waves emitted by their collapse in a viscoelastic medium is important for various naval and medical applications. Two examples are histotripsy, which utilizes this phenomenon for the ablation of pathogenic tissue, and erosion to elastomeric coatings on propellers. To study these problems in a general sense, a canonical problem is considered, which involves the shock-induced collapse of a gaseous bubble in a viscous or viscoelastic medium next to a second viscoelastic or elastic medium of a certain thickness. A novel Eulerian approach, which incorporates nonlinear elasticity, is used to simulate this problem. The stresses, strains and temperatures produced during this process will be presented for different initial stand-off distances, thicknesses of the second medium and shear moduli. Additionally, studies using relevant waveforms that induce the bubble collapse will be presented. 8:39AM M4.00004 Implosion of Cylindrical Cavities via Short Duration Impulsive Loading , JUSTIN HUNEAULT, ANDREW HIGGINS, McGill University — An apparatus has been developed to study the collapse of a cylindrical cavity in gelatin subjected to a symmetric impact-driven impulsive loading. A gas-driven annular projectile is accelerated to approximately 50 m/s, at which point it impacts a gelatin casting confined by curved steel surfaces that allow a transition from an annular geometry to a cylindrically imploding motion. The implosion is visualized by a high-speed camera through a window which forms the top confining wall of the implosion cavity. The initial size of the cavity is such that the gelatin wall is two to five times thicker than the impacting projectile. Thus, during impact the compression wave which travels towards the cavity is closely followed by a rarefaction resulting from the free surface reflection of the compression wave in the projectile. As the compression wave in the gelatin reaches the inner surface, it will also reflect as a rarefaction wave. The interaction between the rarefaction waves from the gelatin and projectile free surfaces leads to large tensile stresses resulting in the spallation of a relatively thin shell. The study focuses on the effect of impact parameters on the thickness and uniformity of the imploding shell formed by the cavitation in the imploding gelatin cylinder. 8:52AM M4.00005 A unified physical model to explain Supercavity closure1 , ROGER ARNDT, Retired, ASHISH KARN, JIARONG HONG, University of Minnesota — An insight into underlying physics behind supercavity closure is an important issue for the operation of underwater vehicles for a number of reasons viz. associated gas flow requirement with each closure regime, effect of cavity closure on the overall cavity behavior and collapse, differences between natural and ventilated supercavity closure etc. There have been several reports on supercavity closure since the 1950s and many empirical relationships governing different closure modes have been proposed by different authors. Yet, there is no universal agreement between results obtained at different experimental facilities. In some cases, contradictory observations have been made. In this talk, systematic investigations conducted into supercavity closure across a wide range of experimental conditions at the Saint Anthony Falls Laboratory (SAFL) are presented. A variety of closure mechanisms were observed including the ones widely reported in the literature, viz. twin vortex, re-entrant jet; new stable closure modes viz. quad vortex and interacting vortex and a host of transition closure modes. A hypothesis on the physical mechanism based on the pressure gradient across the cavity that determines the closure modes is proposed. Using this hypothesis and the control volume analysis at supercavity closure, we explain the observations from SAFL experiments as well as reconcile the observations reported by different researchers. The hypothesis explains the supercavity closure across different experimental facilities, at different blockage ratios and at different flow conditions. Thus, a unified understanding into supercavity closure from the viewpoint of fundamental physics is attempted. 1 Supported by the Office Of Naval Research. 9:05AM M4.00006 Coalescence of Bubbles , CHRISTOPHER ANTHONY, SUMEET THETE, KRISHNARAJ SAMBATH, OSMAN BASARAN, Purdue University — Drop and bubble coalescence plays a central role in industry and nature. During drop coalescence, two drops touch and merge as a liquid neck connecting them grows from microscopic to macroscopic scales. The hydrodynamic singularity that arises as two drops begin coalescing in a dynamically passive outer fluid (air) has been studied thoroughly in recent years. As a preliminary to developing a similar level of understanding when two drops coalesce in an outer fluid of non-negligible density and viscosity, we use simulation to analyze the coalescence of two identical gas bubbles (idealized as two passive spherical voids) in a liquid. This problem has recently been studied experimentally by Nagel and coworkers (2014). The simulations allow probing of the dynamics for neck radii much smaller than what is possible in experiments. At times earlier than those accessible in experiments, simulations reveal a new type of scaling response than those reported by Nagel et al. However, at larger times, the dynamics is shown to transition to regimes that have been proposed by Nagel and coworkers. Unlike in the experiments, it is shown that the observed scaling regimes can be readily rationalized by judicious interrogation of computed flow fields. 9:18AM M4.00007 Bubble coalescence in channels flows1 , JIACAI LU, GRETAR TRYGGVASON, University of Notre Dame — Direct numerical simulations (DNS) of bubbly flows in vertical channels have shown that the steady state flow structure is particularly simple and can be described by relatively elementary considerations. Similarly, DNS have been used to examine the transient evolution of both laminar and turbulent channel flows, also leading to considerable increase of the understanding of such flows. However, as the void fraction increases the assumption of bubbly flows becomes unrealistic and it is necessary to account for topological changes through coalescence and breakup and flow regime transitions. Here the transition of high void fraction laminar bubbly flows to slug flow is examined by DNS, using a front tracking method where the exact coalescence criteria can be changed and its effect studied. To quantify the transition we monitor the flow rate and the wall shear, as well as the interface structure, including the different components of the area concentration tensor, which gives the projection of the bubble surface area in different directions. The effect of the precise representation of when and how coalescence takes place is also studied. Preliminary results for turbulent flows, where both coalescence and breakup take place are also shown and the use the results to aid in the developm 1 Research supported by DOE (CASL). Tuesday, November 25, 2014 8:00AM - 10:10AM Session M5 Waves II: Internal and Interfacial Waves — 3008 - Bruce Sutherland, University of Alberta 8:00AM M5.00001 Generation of Long Internal Waves by Vertically Propagating Compact Internal Wavepackets , BRUCE SUTHERLAND, University of Alberta, TON VAN DEN BREMER, University of Oxford — The divergence of the horizontal flux of horizontal momentum associated with surface gravity wavepackets results in a horizontal flow that turns out to be identical to the Stokes drift. This “divergent-flux induced flow” is itself divergent and so induces a deep response flow whose momentum is equal and opposite to the momentum associated with the Stokes drift. Thus the total momentum is zero. By contrast there is momentum associated with internal wavepackets. Like surface gravity wavepackets, the divergent-flux induced flow of horizontally localized internal waves is itself divergent. However, because the ambient is stratified, and so inhibits vertical motion, there can be no deep return flow. Different from the approach of Bretherton (1969), we follow a physically intuitive but mathematically rigorous quasi-monochromatic wavepacket analysis complemented by fully nonlinear numerical simulations to show that the dominant response is an induced horizontally long internal wave that extends laterally well to either side of the wavepacket. This suggests a new mechanism for efficient energy and momentum transfer from local to long and slow time-scale disturbances that does not involve irreversible deposition through wave breaking. Weakly nonlinear effects are discussed. 8:13AM M5.00002 On 3D internal wave beams and induced large-scale mean flows1 , T.R. AKYLAS, MIT, TAKESHI KATAOKA, Kobe University, Japan — A theoretical model is developed for the 3D propagation of internal gravity wave beams in a uniformly stratified Boussinesq fluid, assuming that variations in the along-beam and transverse directions are of long lengthscale in comparison with the beam width. This situation applies, for instance, to the far-field behavior of a wave beam generated by a horizontal line source with weak transverse dependence. In the 2D case, where only along-beam variations are present, it is known that nonlinear effects are minor, even for beams with finite steepness. By contrast, in 3D, nonlinear interactions can cause transfer of energy to a circulating horizontal mean flow far from the vicinity of the beam. For a small-steepness beam, this process is described by two coupled equations, which govern the 3D beam evolution along with the induced mean flow. This asymptotic model is applied to the experimental setup of Bordes et al. (2012) and qualitative agreement with their observations is found. 1 Supported by NSF and KTC. 8:26AM M5.00003 Can internal waves descend a double-diffusive staircase? , SASAN GHAEMSAIDI, MIT, HAYLEY DOSSER, LUC RAINVILLE, University of Washington, THOMAS PEACOCK, MIT — Due to the rapid loss of ice cover, internal waves are expected to play an increasingly important role in the Arctic Ocean. As such, we present the results of a theoretical study investigating the role of double-diffusive layering, characteristic of the Arctic Ocean, on the fate of internal waves. We begin by considering the transmission properties of a single double-diffusive layer, from which we progress to consider multiple layers, and conclude with a realistic stratification. We investigate the possibility that double-diffusive layer structures can be efficient internal wave inhibitors, shielding the deep ocean from the transmission of momentum and energy flux associated with near inertial waves generated by passing storms. 8:39AM M5.00004 Assessing the importance of internal tide scattering in the deep ocean , MAHA HAJI, MIT-WHOI, THOMAS PEACOCK, Massachusetts Institute of Technology, GLENN CARTER, University of Hawai’i at Manoa, T.M. SHAUN JOHNSTON, Scripps Institution of Oceanography — Tides are one of the main sources of energy input to the deep ocean, and the pathways of energy transfer from barotropic tides to turbulent mixing scales via internal tides are not well understood. Large-scale (low-mode) internal tides account for the bulk of energy extracted from barotropic tides and have been observed to propagate over 1000 km from their generation sites. We seek to examine the fate of these large-scale internal tides and the processes by which their energy is transferred, or “scattered,” to small-scale (high-mode) internal tides, which dissipate locally and are responsible for internal tide driven mixing. The EXperiment on Internal Tide Scattering (EXITS) field study conducted in 2010-2011 sought to examine the role of topographic scattering at the Line Islands Ridge. The scattering process was examined via data from three moorings equipped with moored profilers, spanning total depths of 3000-5000 m. The results of our field data analysis are rationalized via comparison to data from two- and three-dimensional numerical models and a two-dimensional analytical model based on Green function theory. 8:52AM M5.00005 Internal wave penetration into an evanescent layer via parametric subharmonic instability , SASAN GHAEMSAIDI, MIT, THIERRY DAUXOIS, SYLVAIN JOUBAUD, PHILIPPE ODIER, ENS Lyon, THOMAS PEACOCK, MIT — The effect of parametric subharmonic instability (PSI) on the transmission properties of a boundary forced, two-layer density stratification is experimentally studied. In regimes where linear theory simply predicts evanescent decay in the lower layer, PSI creates two daughter waves that are capable of penetrating deep into the stratification in opposing horizontal directions. PSI is shown to be a reasonable mechanism for the injection of energy flux and momentum into an otherwise forbidden lower layer by means of the creation of “burrowing” daughter waves of the primary, forced mother wave. 9:05AM M5.00006 Sensitivity of Rogue Waves Predictions to the Oceanic Stratification , QIUCHEN GUO, MOHAMMAD-REZA ALAM, None — Oceanic rogue waves are short-lived very large amplitude waves (a giant crest typically followed or preceded by a deep trough) that appear and disappear suddenly in the ocean causing damages to ships and offshore structures. Assuming that the state of the ocean at the present time is perfectly known, then the upcoming rogue waves can be predicted via numerically solving the equations that govern the evolution of the waves. The state of the art radar technology can now provide accurate wave height measurement over large spatial domains and when combined with advanced wave-field reconstruction techniques together render deterministic details of the current state of the ocean (i.e. surface elevation and velocity field) at any given moment of the time with a very high accuracy. The ocean water density is, however, stratified (mainly due to the salinity and temperature differences). This density stratification, with today’s technology, is very difficult to be measured accurately. As a result in most predictive schemes these density variations are neglected. While the overall effect of the stratification on the average state of the ocean may not be significant, here we show that these density variations can strongly affect the prediction of oceanic rogue waves. Specifically, we consider a broadband oceanic spectrum in a two-layer density stratified fluid, and study via extensive statistical analysis the effects of strength of the stratification (difference between densities) and the depth of the thermocline on the prediction of upcoming rogue waves. 9:18AM M5.00007 Internal waves incident upon an interface , JOHN MCHUGH, University of New Hampshire — Recent results have shown that a vertical packet of internal waves that are horizontally periodic will develop a discontinuous mean flow at an interface, depending on wave reflection. Here we consider a similar configuration where the waves are not horizontally periodic, but instead exist within a wave packet that is limited both horizontally and vertically. The basic state has constant stability N in two layers without a shear flow. The horizontal limit of the wavepacket results in a much different wave-induced mean flow than the periodic case, as the mean flow is confined to the wavepacket and therefore must have approximately a zero net flow across any vertical surface. The net effect is that gradients of the mean flow at the interface are stronger than the periodic case. Waves are treated with the nonlinear Schrodinger equations that are solved numerically. 9:31AM M5.00008 Shoaling Large Amplitude Internal Solitary Waves in a Laboratory Tank1 , MICHAEL ALLSHOUSE, CONNER LARUE, HARRY SWINNEY, University of Texas at Austin — The shoaling of internal solitary waves onto the continental shelf can change both the wave dynamics and the state of the environment. Previous observations have demonstrated that these waves can trap fluid and transport it over long distances. Through the use of a camshaft-based wavemaker, we produce large amplitude shoaling waves in a stratified fluid in a laboratory tank. Simulations of solitary waves are used to guide the tuning of the wave generator to approximate solitary waves; thus nonlinear waves can be produced within the 4m long tank. PIV and synthetic schlieren measurements are made to study the transport of fluid by the wave as it moves up a sloping boundary. The results are then compared to numerical simulations and analyzed using finite time Lyapunov exponent calculations. This Lagrangian analysis provides an objective measure of barriers surrounding trapped regions in the flow. 1 Supported by ONR MURI Grant N000141110701 (WHOI) 9:44AM M5.00009 Sound propagation through internal gravity wave fields in a laboratory tank1 , LIKUN ZHANG, HARRY L. SWINNEY, University of Texas at Austin, YING-TSING LIN, Woods Hole Oceanographic Institution — We conduct laboratory experiments and numerical simulations for sound propagation through an internal gravity wave field. The goal is to improve the understanding of the effect of internal gravity waves on acoustic propagation in the oceans. The laboratory tank is filled with a fluid whose density decreases linearly from the bottom to the top of the tank; the resultant buoyancy frequency is 0.15 Hz. A 1 MHz sound wave is generated and received by 12.5 mm diameter transducers, which are positioned 0.2 m apart on a horizontal acoustic axis that is perpendicular to the internal wave beam. The fluid velocity field, measured by Particle Image Velocimetry (PIV), agrees well with results from simulations made using a Navier-Stokes spectral code. The sound intensity at the receiver is computed numerically for different measured and simulated frozen density fields. Fluctuations in the sound speed and intensity are determined as a function of the location of the receiver and the frequency and phase of the internal waves. 1 Supported by ONR MURI Grant N000141110701 (WHOI). Also, LZ is supported by the 2013-14 ASA F. V. Hunt Postdoctoral Research Fellowship. 9:57AM M5.00010 Internal waves patterns in the wake of a 3D body towed in a two-layer fluid , LAURENT LACAZE, CNRS / Institut de Mécanique des Fluides de Toulouse, MATTHIEU MERCIER, OLIVIER THUAL, Institut de Mécanique des Fluides de Toulouse, ALEXANDRE PACI, CNRM / GAME Météo France — Stratified flows over obstacles are important features in meteorology and oceanography. The characterization of these flows is crucial in order to propose models of geophysical processes such as mixing and ocean circulation or orographic drag in the atmosphere. For some specific stratification profiles, the energy of internal waves generated by the obstacle can be trapped at a given depth, at the base of the oceanic mixing layer or at the top of the atmospheric boundary layer for instance. This scenario can be modelled by a two-layer stratified fluid for which gravity waves spread at the interface between the two layers. The work presented here focuses on a two-layer flow over a 3D obstacle, or equivalently, an obstacle towed in a fluid at rest. Experiments performed both in the large-scale flume of CNRM-GAME Toulouse (METEO-FRANCE & CNRS) and in a smaller tank apparatus, are presented with a specific attention on the measurement of the 3D wave patterns. A non-hydrostatic linear analysis is used to describe the observed wave patterns. The experiments highlight the strong influence of the Froude number on the generated waves. More specifically, we investigate the nature of the wake angle obtained from the wave pattern, and discuss a transition from Kelvin to Mach angle. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M6 Biofluids: Microswimmers III — 3010 - Roman Stocker, Massachusetts Institute of Technology 8:00AM M6.00001 The curved shape of the bacterium Caulobacter crescentus enhances colonization of surfaces in flow , ALEXANDRE PERSAT, ZEMER GITAI, HOWARD STONE, Princeton University — Bacteria thrive in all types of fluid environments; flow is thus a ubiquitous aspect of their lives. Bacteria have evolved a variety of cellular components contributing to their growth in specific environments. However, cellular features that help them survive and develop in flow have been rarely characterized. Here, we show that Caulobacter crescentus may have evolved its curved shape to enhance the colonization of surfaces in flow. C. crescentus curvature is preserved in the wild but straight mutants have no known growth disadvantage in standard laboratory conditions. Leveraging microfluidics and single-cell imaging, we demonstrate that curvature enhances surface colonization in flow, promoting the formation of larger microcolonies. Cells attach to a surface from a single pole, so that flow affects their orientation. In flow, viscous forces generate a torque on the curved cell body, which reorients the cell in the direction of the flow. The curved cell appears to arc above the surface, optimally orienting its unattached pole towards the surface. This reduces the distance between the surface and the pole, thereby enhancing attachment of its progeny. Additionally, we show that curved shape enhances colony spreading across the direction of the flow, generating more robust biofilm compared to straight mutants. 8:13AM M6.00002 Chemotactic decision making in swimming microorganisms , M. MEHDI SALEK, MIT, JEFFREY S. GUASTO, Tufts University, ROMAN STOCKER, MIT — Swimming cells are often guided by chemical gradients (“chemotaxis”) to search for nutrients, hosts, and mates, and to avoid predators and noxious substances. It remains unclear, however, how variable the chemotactic abilities of cells are among cells of one species, and whether there are better “decision makers” within a population. Inspired by studies in macro-organism ecology, we fabricated a microfluidic “T-maze” in which marine bacteria are subjected to a chemical attractant gradient at each of a series of consecutive T-junctions. We used video microscopy to capture the motion of thousands of bacteria as they migrate up or down the gradient at each subsequent junction. This approach provides detailed statistics at both the single-cell and population levels, while simultaneously sorting the cells by chemotactic ability. Using a range of bacteria, we demonstrate how the microfluidic T-maze allows us to sort the better decision-making cells in the population, opening the door for improved efficiency of a range of microbial processes in nature and industry. 8:26AM M6.00003 The deadly swimming of Cercariae: an unusual Stokesian swimmer , MANU PRAKASH, DEEPAK KRISHNAMURTHY, Stanford University — Schistosomiasis, also known as Bilharzia, is a Neglected Tropical Disease (NTD) caused by a parasitic Trematode blood fluke worm. In terms of socio-economic and public health impact, Schistosomiasis is second only to Malaria as the most devastating parasitic disease in tropical countries; with roughly 200 million people infected at any time world-wide and up to 200,000 deaths every year. The infectious form of the parasite, known as Cercariae, emerge from snails into freshwater and infect humans by directly burrowing into the skin. Thus, anyone in contact with infected waters is at risk, which mostly includes children. By establishing a safe experimental means of studying the Cercariae in our lab, we report here their unusual swimming dynamics which include both head-first and tail-first swimming modes. These swimming modes are crucial for the chemotactic activity of Cercariae which allows them to seek out and burrow into human skin. By experimental and analytical means, we demonstrate that Cercariae break symmetry and achieve locomotion at small Reynolds number differently when compared to well-known methods involving traveling waves in the flagellum or chiral beating. Although they utilize the well-known drag anisotropy of a slender body in Stokes flow, the geometry and kinematics of their propulsion mechanism is novel. Based on these results, we propose a new kind of simple Stokesian swimmer (T-joint swimmer) in an attempt to explain the evolutionary advantages of this novel swimming mechanism. Using the above physical insights from a biological and global-health standpoint, we explore ways to hinder the chemotactic capabilities of this parasite. 8:39AM M6.00004 Shape and shear guide sperm cells spiraling upstream , VASILY KANTSLER, Skolkovo Institute of Science and Technology, Russia; Department of Physics, University of Warwick, UK, JORN DUNKEL, Department of Mathematics, Massachusetts Institute of Technology, RAYMOND E. GOLDSTEIN, DAMTP, University of Cambridge, UK — A major puzzle in biology is how mammalian sperm determine and maintain the correct swimming direction during the various phases of the sexual reproduction process. Currently debated mechanisms for sperm long range travel vary from peristaltic pumping to temperature sensing (thermotaxis) and direct response to fluid flow (rheotaxis), but little is known quantitatively about their relative importance. Here, we report the first quantitative experimental study of mammalian sperm rheotaxis. Using microfluidic devices, we investigate systematically the swimming behavior of human and bull sperm over a wide range of physiologically relevant shear rates and viscosities. Our measurements show that the interplay of fluid shear, steric surface-interactions and chirality of the flagellar beat leads to a stable upstream spiraling motion of sperm cells, thus providing a generic and robust rectification mechanism to support mammalian fertilization. To rationalize these findings, we identify a minimal mathematical model that is capable of describing quantitatively the experimental observations. 8:52AM M6.00005 The unique low-Reynolds-number spinning hydrodynamics of release of a giant multinucleate multiflagellate zoospore , JAVIER URZAY, Stanford University, DONALD OTT, University of Akron, MANU PRAKASH, Stanford University — Asexual reproduction in aquatic algal species of Vaucheria occurs by the formation of large multinucleate zoospores formed within elongated club-shaped zoosporangia at the tips of young branches. During development, the zoosporangia are separated from the rest of the thallus by membranes, resulting in multiple chambers hosting zoospores which will be released and dispersed in the surrounding aqueous environment. The apical gelatinization of the zoosporangial tip, together with the turgor pressure in the segregated portion of the filament, lead to a narrow aperture through which the zoospore escapes. However ordinary this may seem, Vaucheria zoospores have a unique multiflagellated patterned surface that warrants helicoidal flow entrainment at relatively high speeds, and which enables them to undergo a spinning motion that elastohydrodynamically assists the rather unfavorable escape maneuver. Experimental observations of this phenomenon, together with quantitative interpretations, are provided in this talk. 9:05AM M6.00006 Random walk of microswimmers: puller and pusher cases , SALIMA RAFAI, PHILIPPE PEYLA, LIPhy - CNRS/Univ. Grenoble, DYFCOM TEAM — Swimming at a micrometer scale demands particular strategies. Indeed when inertia is negligible as compared to viscous forces (i.e. Reynolds number Re is lower than unity), hydrodynamics equations are reversible in time. To achieve propulsion a low Reynolds number, swimmers must then deform in a way that is not invariant under time reversal. Here we investigate the dispersal properties of self propelled organisms by means of microscopy and cell tracking. Our systems of interest are, on the one hand, the microalga Chlamydomonas Reinhardtii, a puller-type swimmer and on the other hand, Lingulodinium polyedrum, a pusher. Both are quasi-spherical single celled alga. In the case of dilute suspensions, we show that tracked trajectories are well modelled by a correlated random walk. This process is based on short time correlations in the direction of movement called persistence. At longer times, correlations are lost and a standard random walk characterizes the trajectories. Finally we show how drag forces modify the characteristics of this particular random walk. 9:18AM M6.00007 How to be invisible as a microscopic swimmer , NAVISH WADHWA, THOMAS KIØRBOE, ANDERS ANDERSEN, Technical University of Denmark — Microscopic plankton live a difficult life in open waters, having to continuously scan large amounts of water for food and mates, and hide from predators at the same time. In the absence of vision at these small scales, all interactions are dominated by chemical and hydromechanical cues. Thus, there is an evolutionary pressure to minimize the hydromechanical disturbance generated during processes such as feeding and locomotion. We report experimental observations that breast stroke swimming plankton generate a fluid disturbance that decays faster with distance than what is predicted from the commonly used stresslet model of a self-propelled organism. We rationalize these observations by using a three-Stokeslet model of a breast stroke swimmer, and show that it is possible for a swimmer to dramatically reduce its fluid disturbance by appropriately positioning the propulsive apparatus. A generalization of this concept may be used in understanding the large diversity of shapes and swimming modes found in the plankton world. 9:31AM M6.00008 Viscous constraints on predator:food size ratios in microscale feeding , MEHDI JABBARZADEH, HENRY FU, University of Nevada, Reno — Small organisms such as protists or copepods may try to capture food by manipulating food with cilia, limbs, or feeding appendages. At these small scales, viscous flow may complicate the ability of a feeding appendage to closely approach a food particle. As a simplified but tractable model of such feeding approach, we consider the problem of two spheres approaching in a Stokes fluid. The first “feeding” sphere, which represents a body part or feeding appendage, is pushed with a constant force towards a force-free “food” sphere. When the feeding sphere reaches within a cutoff distance of the food sphere we assume that nonhydrodynamic interactions lead to capture. We examine approach for a range of size ratios between the feeding and food sphere. To investigate the approach efficiency, we examine the time required for the feeding sphere to capture the food sphere, as well as how far the feeding sphere must move before it captures the food sphere. We also examine the effect of varying the cutoff distance for capture. We find that hydrodynamic interactions strongly affect the results when the size of the spheres is comparable. We describe what relative sizes between feeding sphere and food particles may be most effective for food capture. 9:44AM M6.00009 Mass transfer effect of the stalk contraction-relaxation cycle of Vorticella convallaria 1 , JIAZHONG ZHOU, DAVID ADMIRAAL, SANGJIN RYU, University of Nebraska-Lincoln — Vorticella convallaria is a genus of protozoa living in freshwater. Its stalk contracts and coil pulling the cell body towards the substrate at a remarkable speed, and then relaxes to its extended state much more slowly than the contraction. However, the reason for Vorticella’s stalk contraction is still unknown. It is presumed that water flow induced by the stalk contraction-relaxation cycle may augment mass transfer near the substrate. We investigated this hypothesis using an experimental model with particle tracking velocimetry and a computational fluid dynamics model. In both approaches, Vorticella was modeled as a solid sphere translating perpendicular to a solid surface in water. After having been validated by the experimental model and verified by grid convergence index test, the computational model simulated water flow during the cycle based on the measured time course of stalk length changes of Vorticella. Based on the simulated flow field, we calculated trajectories of particles near the model Vorticella, and then evaluated the mass transfer effect of Vorticella’s stalk contraction based on the particles’ motion. 1 We acknowlege support from Laymann Seed Grant of the University of Nebraska-Lincoln 9:57AM M6.00010 Direct measurement of the forces generated by an undulatory microswimmer , RAFAEL SCHULMAN, MATILDA BACKHOLM, McMaster University, WILLIAM RYU, University of Toronto, KARI DALNOKI-VERESS, McMaster University — C. elegans is a millimeter-sized nematode which has served as a model organism in biology for several decades, primarily due to its simple anatomy. Employing an undulatory form of locomotion, this worm is capable of propelling itself through various media. Using a micropipette deflection technique, in conjunction with high speed imaging, we directly measure the time-varying forces generated by C. elegans. We observe excellent agreement between our measured forces and the predictions of resistive force theory, through which we determine the drag coefficients of the worm. We also perform the direct force measurements at controlled distances from a single solid boundary as well as between two solid boundaries. We extract the drag coefficients of the worm to quantify the influence of the boundary on the swimming and the hydrodynamic forces involved. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M7 Biofluids: Respiratory Flows — 3012 - Joseph Bull, University of Michigan 8:00AM M7.00001 Experimental investigation of particle deposition mechanisms in the lung acinus using microfluidic models. , RAMI FISHLER, MOLLY MULLIGAN, YAEL DUBOWSKI, JOSUE SZNITMAN, Technion- Israel Institute of Technology, SZNITMAN LAB- DEPARTMENT OF BIOMEDICAL ENGINEERING TEAM, DUBOWSKI LAB- FACULTY OF CIVIL AND ENVIRONMENTAL ENGINEERING TEAM — In order to experimentally investigate particle deposition mechanisms in the deep alveolated regions of the lungs, we have developed a novel microfluidic device mimicking breathing acinar flow conditions directly at the physiological scale. The model features an anatomicallyinspired acinar geometry with five dichotomously branching airway generations lined with periodically expanding and contracting alveoli. Deposition patterns of airborne polystyrene microspheres (spanning 0.1 µm to 2 µm in diameter) inside the airway tree network compare well with CFD simulations and reveal the roles of gravity and Brownian motion on particle deposition sites. Furthermore, measured trajectories of incense particles (0.1-1 µm) inside the breathing device show a critical role for Brownian diffusion in determining the fate of inhaled sub-micron particles by enabling particles to cross from the acinar ducts into alveolar cavities, especially during the short time lag between inhalation and exhalation phases. 8:13AM M7.00002 Unsteady Oxygen Transfer in Space-Filling Models of the Pulmonary Acinus , PHILIPP HOFEMEIER, LIHI SHACHAR-BERMAN, Department of Biomedical Engineering, Technion - Israel Institute of Technology, MARCEL FILOCHE, French National Centre for Scientific Research - Institut de physique (INP), PMC, Ecole Polytechnique, JOSUE SZNITMAN, Department of Biomedical Engineering, Technion - Israel Institute of Technology — Diffusional screening in the pulmonary acinus is a well-known physical phenomenon that results from the depletion of fresh oxygen in proximal acinar generations diffusing through the alveolar wall membranes and effectively creating a gradient in the oxygen partial pressure along the acinar airways. Until present, most studies have focused on steady-state oxygen diffusion in generic sub-acinar structures and discarded convective oxygen transport due to low Peclet numbers in this region. Such studies, however, fall typically short in capturing the complex morphology of acinar airways as well as the oscillatory nature of convecive acinar breathing. Here, we revisit this problem and solve the convective-diffusive transport equations in breathing 3D acinar structures, underlining the significance of convective flows in proximal acinar generations as well as recirculating alveolar flow patterns. In particular, to assess diffusional screening, we monitor time-dependent efficiencies of the acinus under cyclic breathing motion. Our study emphasizes the necessity of capturing both a dynamically breathing and anatomically-realistic model of the sub-acinus to characterize unsteady oxygen transport across the acinar walls. 8:26AM M7.00003 Turbulent dispersivity under conditions relevant to airborne disease transmission between laboratory animals , SIOBHAN HALLORAN, Dept. Chemical Engineering and Materials Science, University of California Davis, ANTHONY WEXLER, Dept. Mech. and Aerospace Eng; Air Quality Research Center; Dept Civil and Environ. Eng.; and Dept Land Air Water Res. University California Davis, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — Virologists and other researchers who test pathogens for airborne disease transmissibility often place a test animal downstream from an inoculated animal and later determine whether the test animal became infected. Despite the crucial role of the airflow in modulating the pathogen transmission, to date the infectious disease community has paid little attention to the effect of airspeed or turbulence intensity on the probability of transmission. Here we present measurements of the turbulent dispersivity under conditions relevant to experimental tests of airborne disease transmissibility between laboratory animals. We used time lapse photography to visualize the downstream transport and turbulent dispersion of smoke particulates released from a point source downstream of a standard axial fan, thus mimicking the release and transport of expiratory aerosols exhaled by an inoculated animal. We demonstrate that the fan speed counterintuitively has no effect on the downstream plume width, a result replicated with a variety of different fan types and configurations. The results point toward a useful simplification in modeling of airborne disease transmission via fan-generated flows. 8:39AM M7.00004 An immersed-boundary framework for patient-specific optimization of inhaled drug delivery , LAURA NICOLAOU, Imperial College London, TAMER ZAKI, Johns Hopkins University, Imperial College London — Predictive numerical simulations have the potential to significantly enhance therapies for lung disease by providing a valuable clinical aid and a platform to optimize drug delivery. A difficult challenge, however, is the influence of inter-subject variations of the airway geometries and their impact on the airflow and aerosol deposition. A personalized approach to the treatment of respiratory diseases is therefore required. An in silico framework for patient-specific predictions of the flow and aerosol deposition in the respiratory airways is presented. The approach efficiently accommodates geometric variation and airway motion in order to optimize pulmonary drug delivery. A non-rigid registration method is adopted to construct dynamic airway models conforming to the patient’s breathing. Accurate predictions of the flow in realistic airway geometries are computed using direct numerical simulations (DNS) with boundary conditions enforced using a robust, implicit immersed boundary (IB) method for curvilinear meshes. A Lagrangian particle-tracking scheme is adopted to model the transport and deposition of the aerosol particles in the airways. Examples of flow and aerosol deposition in realistic extrathoracic airways and of a patient-specific dynamic lung model are presented. 8:52AM M7.00005 DNS and PIV Measurements of the Flow in a Model of the Human Upper Airway1 , YONG WANG, Univ of California - Irvine, LIRAN OREN, EPHARIM GUTMARK, Univ. of Cincinnati, Cincinnati, OH 45267, SAID ELGHOBASHI, Univ of California - Irvine, UNIVERSITY OF CALIFORNIA, IRVINE COLLABORATION, UNIV. OF CINCINNATI, CINCINNATI COLLABORATION — The flow in the human upper airway (HUA) is 3D, unsteady, undergoes transition from laminar to turbulent, and reverses its main direction about every two seconds. In order to enhance the understanding of the flow properties, both numerical and experimental studies are needed. In the present study, DNS results of the flow in a patient-specific model of HUA are compared with experimental data. The DNS solver uses the lattice Boltzmann method which was validated [1] for some canonical laminar and turbulent flows The experimental model was constructed from transparent silicone using a mold prepared by 3D printing. Velocity measurements were performed via high speed particle image velocimetry (HSPIV). The refractive index of the fluid used in the HUA experimental model matched the refractive index of the silicone. Both inspiration and expiration cases with several flow rates in the HUA are studied. The DNS velocity fields at several cross section planes are compared with the HSPIV measurements. The computed pressure and vorticity distributions will be also presented. [1] Y. Wang & S. Elghobashi, (2014). Respir Physiol Neurobiol., 193, 1–10. 1 NIH Heart Lung and Blood Inst.-Grant HL105215 9:05AM M7.00006 Effect of cartilaginous rings on the fluid structures in a bifurcating tube , HUMBERTO BOCANEGRA EVANS, LUCIANO CASTILLO, Texas Tech University — Fluid dynamical models of the respiratory system typically represent the bronchial tree as a collection of smooth tube bifurcations, ignoring the presence of cartilaginous rings in the first few generations, i.e. trachea and main bronchi. While accurate in certain instances, this simplification may considerably affect the results when the issue at hand is the dispersion and deposition of particles within the respiratory tract. In this study, we use a refractive index-matched particle image velocimetry facility to obtain velocity data in a scaled model of a bifurcating corrugated tube. We will present data on the fluid characteristics and how these are affected by cartilaginous rings. 9:18AM M7.00007 A Missing Puzzle Piece in Murray’s Law: the Optimal Angle of Junctions , RUO-QIAN WANG, KATHERINE TAYLOR, AMOS G. WINTER, Massachusetts Inst of Tech-MIT, GLOBAL ENGINEERING AND RESEARCH LAB TEAM — Branching flows are common in biological systems, such as the circulatory and respiratory systems of animals. The optimal radii of parent and daughter branches can be explained with Murray’s law, which dictates that the sum of metabolic and pumping costs is minimized. Murray’s Law can be used to determine the diameter of cascading channels but misses an important parameter: the angles of the branches. Past hydraulic studies have investigated the angle effect, but have not focused on whether this geometry follows Murray’s Law; while a simple network optimization is able to show that at low Reynolds numbers a branch with a parent channel connecting to n equally distant channels obeying Murray’s Law has a minimum total head loss with a branching angle θ, such 2 that cos θ = n− 3 , but it’s not valid for high Reynolds number flows, which may experience separation and turbulence at the branches. The present study is focused on determining the optimal branch angle that complies with Murray’s Law for moderate Reynolds numbers. Computational studies using Open FOAM and experiments using 3D printed branched channels will be presented. These results will be used to quantify the effect of Reynolds number on optimal branch geometry. 9:31AM M7.00008 Physical principle of airway design in human lungs , KEUNHWAN PARK, TAEHO SON, Department of Mechanical & Aerospace Engineering, Seoul National University, Seoul, Korea, WONJUNG KIM, Department of Mechanical Engineering, Sogang University, Seoul, Korea, HO-YOUNG KIM, Department of Mechanical & Aerospace Engineering, Seoul National University, Seoul, Korea — From an engineering perspective, lungs are natural microfluidic devices that extract oxygen from air. In the bronchial tree, airways branch by dichotomy with a systematic reduction of their diameters. It is generally accepted that in conducting airways, which air passes on the way to the acinar airways from the atmosphere, the reduction ratio of diameter is closely related to the minimization of viscous dissipation. Such a principle is formulated as the Hess-Murray law. However, in acinar airways, where oxygen transfer to alveolae occurs, the diameter reduction with progressive generations is more moderate than in conducting airways. Noting that the dominant transfer mechanism in acinar airways is diffusion rather than advection, unlike conducting airways, we construct a mathematical model for oxygen transfer through a series of acinar airways. Our model allows us to predict the optimal airway reduction ratio that maximizes the oxygen transfer in a finite airway volume, thereby rationalizing the observed airway reduction ratio in acinar airways. 9:44AM M7.00009 Diurnal respiration of a termite mound1 , HUNTER KING, SAMUEL OCKO, L. MAHADEVAN, Harvard Univ — Many species of fungus-harvesting termites build largely empty, massive mound structures which protrude from the ground above their subterranean nests. It has been long proposed that the function of these mounds is to facilitate exchange of heat, humidity, and respiratory gases; this would give the colony a controlled climate in which to raise fungus and brood. However, the specific mechanism by which the mound achieves ventilation has remained a topic of debate, as direct measurement of internal air flows has remained difficult. By directly measuring these elusive, tiny flows with a custom sensor, we find that the mound architecture of the species Odontotermes obesus takes advantage of daily oscillations in ambient temperature to drive convection and gas transport. This contradicts previous theories, which point to internal metabolic heating and external wind as driving forces. Our result, a novel example of deriving useful work from a fluctuating scalar parameter, should contribute to better understanding insect swarm construction and possible development in passive human architecture, both of which have been spurred by previous research on termites. 1 We acknowledge support from HFSP. 9:57AM M7.00010 Assessment of regional effects in pulmonary aerosol delivery using Direct Numerical Simulation (DNS)1 , STAVROS KASSINOS, FOTOS STYLIANOU2 , PANTELIS KOULLAPIS, Univ of Cyprus, UCY-COMPSCI TEAM3 — Recent computational studies have shown that the airflow in the upper human airways is turbulent during much of the respiratory cycle. One of the features of respiratory airflow that poses a challenge to computations based on Reynolds-Averaged Navier-Stokes (RANS) closures is the laminar-turbulentlaminar transition as the flow moves from the mouth through the glottis and down to the lower conducting airways. Turbulence and unsteadiness are expected at least through the first few bifurcations of the airways. In the case of inhaled medicines, and depending on the size of the particles in the formulation, airway bifurcations are areas of preferential deposition. Here, we use Direct Numerical Simulations (DNS) to examine aerosol deposition in the case of turbulent flow through a realistic representation of the tracheal bifurcation. We examine the flow characteristics in detail, including the turbulent structures and how they affect the deposition of particles of different sizes. DNS results are compared with RANS computations. 1 Supported by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 315760 HEXACOMM. by the Cyprus Research Promotion Foundation through the Framework Programme for Research, Technological Development and Innovation 2009-2010 under Grant KOYLTOYRA/BENS/0510/01. 3 Computational Science Laboratory (UCY-CompSci) 2 Supported Tuesday, November 25, 2014 8:00AM - 10:10AM Session M9 Biofluids: Plant Physiology — 3014/3016 - Tomas Bohr, Technical University of Denmark 8:00AM M9.00001 Morphological characteristics of motile plants for dynamic motion1 , KAHYE SONG, EUNSEOP YEOM, KIWOONG KIM, SANG JOON LEE, Pohang Univ of Sci & Tech — Most plants have been considered as non-motile organisms. However, plants move in response to environmental changes for survival. In addition, some species drive dynamic motions in a short period of time. Mimosa pudica is a plant that rapidly shrinks its body in response to external stimuli. It has specialized organs that are omnidirectionally activated due to morphological features. In addition, scales of pinecone open or close up depending on humidity for efficient seed release. A number of previous studies on the dynamic motion of plants have been investigated in a biochemical point of view. In this study, the morphological characteristics of those motile organs were investigated by using X-ray CT and micro-imaging techniques. The results show that the dynamic motions of motile plants are supported by structural features related with water transport. These studies would provide new insight for better understanding the moving mechanism of motile plant in morphological point of view. 1 This research was financially supported by the Creative Research Initiative of the Ministry of Science, ICT and Future Planning (MSIP) and the National Research Foundation (NRF) of Korea (grant number: 2008-0061991). 8:13AM M9.00002 Dynamic analysis on cavitation and embolization in vascular plants under tension1 , JEONGEUN RYU, BAE GEUN HWANG, YANGMIN KIM, SANG JOON LEE, Department of Mechanical Engineering, POSTECH — Plants can transport sap water from the soil to the tip of their leaves using the tensile forces created by leaf transpiration without any mechanical pumps. However, the high tension adversely induces a thermodynamically metastable state in sap water with negative pressure and gas bubbles are prone to be formed in xylem vessels. Cavitation easily breaks down continuous water columns and grows into embolization, which limits water transport through xylem vessels. Meanwhile, the repair process of embolization is closely related to water management and regulation of sap flow in plants. In this study, the cavitation and embolization phenomena of liquid water in vascular plants and a physical model system are experimentally and theoretically investigated in detail under in vivo and in vitro conditions. This study will not only shed light on the understanding of these multiphase flows under tension but also provide a clue to solve cavitation problems in micro-scale conduits and microfluidic network systems. 1 This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2008-0061991). 8:26AM M9.00003 Taking the Pulse of Plants , KAARE H. JENSEN, Technical University of Denmark, SIERRA BEECHER, Washington State University, N. MICHELE HOLBROOK, Harvard University, MICHAEL KNOBLAUCH, Washington State University — Many biological systems use complex networks of vascular conduits to distribute energy over great distances. Examples include sugar transport in the phloem tissue of vascular plants and cytoplasmic streaming in some slime molds. Detailed knowledge of transport patterns in these systems is important for our fundamental understanding of energy distribution during development and for engineering of more efficient crops. Current techniques for quantifying transport in these microfluidic systems, however, only allow for the determination of either the flow speed or the concentration of material. Here we demonstrate a new method, based on confocal microscopy, which allows us to simultaneously determine velocity and solute concentration by tracking the dispersion of a tracer dye. We attempt to rationalize the observed transport patterns through consideration of constrained optimization problems. 8:39AM M9.00004 Hydraulic pulse induced by bending in synthetic and natural branches: role in plant mechano-perception1 , JEAN-FRANCOIS LOUF, GEOFFROY GUENA, YOEL FORTERRE, Aix Marseille Université, CNRS, IUSTI UMR 7343, 13453, Marseille, France, ERIC BADEL, INRA, UMR547 PIAF, Université Blaise Pascal, F-63100, Clermont-Ferrand, France — Plants can detect mechanical stimuli such as wind or touch and respond to these stimuli by modifying their development and growth. A fascinating feature of this mechanical-induced-growth response is that it is not only local, but also non-local: bending locally a stem or a branch can induce a very rapid (∼ min) modification of the growth far away from the stimulated area. The nature and mechanism of this long distance signal is not well understood, but it has been suggested that it could result from a purely hydraulic pressure signal, in response to the mechanical bending of the hydrated wood tissue. To address this issue, we investigate the poroelastic response to sudden bending of both natural tree branches and synthetic branches made of PDMS elastomer perforated with longitudinal micro-channels and filled with a viscous fluid. In both systems, we observe that the bending of the branch generates a sudden increase of the mean pore pressure, which scales with the beam elasticity and increases quadratically with the bending amplitude. We propose a simple non-linear model to explain the generation of this hydraulic pulse and discuss our results in the context of plant mechano-perception. 1 This work is supported by the French National Agency (ANR) through the program ANR-13-JS09-0011 ARTIS 8:52AM M9.00005 Tree-on-a-chip: a microfluidic osmotic pump mimicking passive phloem loaders , JEAN COMTET, Massachusetts Institute of Technology, KAARE H. JENSEN, Technical University of Denmark, ABRAHAM D. STROOCK, Cornell University, ANETTE (PEKO) HOSOI, Massachusetts Institute of Technology — According to the Münch mechanism, vascular plants rely on osmotic pressure gradients to export sugars from regions of synthesis (mature leaves) to regions of consumption (roots, fruits). A crucial step in this process is the loading of sugars from photosynthetic cells (known as mesophylls) to the export conduit (the phloem). In some plants, known as passive loaders, sugars are thought to simply diffuse from mesophylls to the phloem. In this case, we show that a single nondimensional “flushing number,” characterizing the relative balance of diffusive sugar loading and convective phloem transport, accurately describes the state of the system (phloem hydrostatic pressure, sugar export rates...). We build a synthetic microfluidic osmotic pump mimicking this biological transport mechanism. In particular, our pump can work in a diffusion-limited regime, for which the flow rate scales weakly with the resistance of the hydraulic circuit. This bio-inspired device provides insight into the biophysical mechanism of passive phloem loading, and could be relevant for microfluidic or micro-robotic applications, where high actuation pressures and steady-state flow rates are necessary. 9:05AM M9.00006 Fluid dynamics of hydrophilous pollination in Ruppia (widgeon grass)1 , NAGA MUSUNURI, DANIEL BUNKER, NJIT, SUSAN PELL, Brooklyn Botanic Garden, IAN FISCHER, PUSHPENDRA SINGH, NJIT — The aim of this work is to understand the physics underlying the mechanisms of two-dimensional aquatic pollen dispersal, known as hydrophily, that have evolved in several genera of aquatic plants, including Halodule, Halophila, Lepilaena, and Ruppia. We selected Ruppia, which grows in the wetlands of the New Jersey/New York metropolitan area, for this study. Our experiments show that the pollen grains from an anther suddenly disperse and form a monolayer when they come in contact with a water surface. This is a crucial first step in the formation of floating porous pollen structures called “pollen rafts,” which often contain pollen grains from several anthers. The formation of porous pollen rafts increases the probability of pollination by increasing the two-dimensional reach of the pollen from each individual anther. 1 The work was supported by National Science Foundation 9:18AM M9.00007 Experimental investigation of the flow field and pollen trajectories/deposition around ovulate pine cones , NETA-LEE JACOBSON, RENÉ VAN HOUT, Faculty of Mechanical Engineering, Technion - Israel Institute of Technology — Particle deposition on bluff bodies is important both in industrial applications as well as in furthering our understanding of ecological networks. It has been hypothesized that plant structural morphology manipulates the flow field in order to enhance capturing of species-specific pollen and thereby increase fertilization chances. Here, the deposition mechanism of different pine pollen on freshly harvested ovulate pine cones (Pinus Halepensis/Brutia) was investigated using high speed, planar particle image velocimetry and holographic 3D technique enabling measurement of both Lagrangian particle tracks and instantaneous flow fields. Measurements were performed in a small blow–through windtunnel at Reynolds numbers ranging from Re = 174 to 767. The roughness on a pine cone is characterized by “scales” organized as Fibonacci spirals. Effects of this roughness on the flow field are compared to results for a smooth sphere at similar Re. Particle deposition results indicate that inertial deposition on the windward side of the cone is the main mechanism. However, at the lowest Reynolds numbers pollen with Stokes numbers less than one were entrained into the cone’s near wake and advected towards the leeward side of the cone. 9:31AM M9.00008 Modeling of wind-initiated liberation of fungal propagules from host plant leaves , TREVOR GONZALINAJEC, Univ of California - Berkeley — Successful airborne propagule dispersal must begin with liberation into the air. The physical shedding mechanism of airborne propagules in the 100-250µm size range are not well understood. Many fungal plant pathogens have propagules in this size range that are shed from the bottom of infected leaves. If turbulent air flow is sufficient to liberate the sporocarps of fungi from leaves then the aerodynamic forces exerted must be sufficient to overcome adhesive forces. In this study I have sought to quantify the magnitude and direction of these aerodynamic forces and their causal flow fields with dynamically scaled physical models. I chose a genus of powdery mildew because maturation of the sporocarp entails morphological changes that lever the sporocarp further away from the leaf surface and out of the viscous boundary layer. Consequently I varied the sporocarp morphology, the boundary layer thickness, and the flow velocity as forces on models were measured with a transducer. Additionally I analyzed the fluid velocity around the models using PIV, which allowed for quantification of the relative importance of shear forces and pressure-gradient forces. The results suggest that forces from steady and unsteady wind alike are insufficient to explain liberation. 9:44AM M9.00009 The efficient flight of Ruellia ciliatiflora seeds , DWIGHT WHITAKER, FRANKLIN MARSH, PETER CHEN, DAVID VEJAR, Pomona College, PATRICK BABB, JOSUE CASTILLO, SABRINA CORDERO, MAHARANI LUMBAN-GAOL, IRLANDA MORA, TANIA PARTIDA, JULIAN PINEDA, AARON RODRIGUEZ, Pomona College Academy for Youth Success (PAYS) — The seeds of Ruellia ciliatiflora are small disks measuring approximately 3 mm in diameter and 0.3 mm in height, which are launched from exploding fruits at speeds exceeding 10 m/s. The seeds fly with backspin such that the axis of symmetry is parallel to the ground. With rotation rates that exceed 1 kHz they keep an aerodynamic profile and move through the air with a extremely low drag. Using high-speed video we have learned that the drag coefficients for these flying seeds can measure less than 0.01 for those launched with the least wobble. To understand the role of seed morphology and rotation rate on the flight of the seeds, we will also present work using 3D printed models of the seeds for studies in wind tunnels. Three-dimensional models are created by photographing seeds from many angles and inferring a shape using commercial software, which also creates a printable model. These studies should help guide work that compares explosions from fruits within the Acanthaceae family to which R. ciliatiflora belongs. This family consists of over 2000 species with exploding fruit with diverse habitats and morphologies. 9:57AM M9.00010 Make a wish: coins falling in water , LIONEL VINCENT, LUKE HEISINGER, EVA KANSO, University of Southern California — Accurate prediction of the flight range and landing site of an object descending under the influence of gravitational and aerodynamic forces is relevant to many engineering and science applications. Examples range from forecasting the touchdown locations of re-entry space vehicles to understanding the settlement patterns of seeds. While the descent motion follows the laws of classical mechanics, the delicate interplay between the fluid medium and the physical properties of the descending object makes the exact landing site difficult to predict a priori and thus best treated probabilistically. Indeed, objects falling in a fluid medium rarely descend in a straight line. The descent motion is generally complex, even for regularly shaped objects such as coins and cards. For such objects, four types of descent regimes have been identified: steady, fluttering, chaotic, or tumbling. Here, we assess the dependence of landing sites on the type of descent motion through controlled experiments of coins falling in water where we quantify the spread in the landing sites and probability of landing heads up. Interestingly, we find that, in certain descent regimes, the fluid medium acts as a randomization device, while other regimes only cause small uncertainties in the landing sites. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M10 Microscale Flows: General — 3005 - Carl Meinhart, University of California, Santa Barbara 8:00AM M10.00001 Control of microparticles packing density in a microfluidic channel for bead based immunoassays applications1 , GABRIEL CABALLERO-ROBLEDO, PABLO GUEVARA-PANTOJA, CINVESTAV-Monterrey, PIIT 66600, Nuevo Leon, Mexico — Bead based immunoassays in microfluidic devices have shown to greatly outperform conventional methods. But if functional point-of-care devices are to be developed, precise and reproducible control over the granulate packings inside microchannels is needed. In this work we study the efficiency of a nanoparticles magnetic trap previously developed by B. Teste et al. [Lab Chip 11, 4207 (2011)] when we vary the compaction of micrometric iron beads packed against a restriction inside a microfluidic channel. The packing density of the beads is finely and reproducibly changed by applying a vibrational protocol originally developed for macroscopic, dry granular systems. We find, counterintuitively, that the most compact and stable packings are up to four times less efficient in trapping nano particles than the loosest packings. 1 This work has been supported by Conacyt, Mexico, under Grant No. 180873 8:13AM M10.00002 Salmonella capture using orbiting magnetic microbeads1 , DREW OWEN, MATTHEW BALLARD, ZACHARY MILLS, SRINIVAS HANASOGE, PETER HESKETH, ALEXANDER ALEXEEV, Georgia Institute of Technology — Using threedimensional simulations and experiments, we examine capture of salmonella from a complex fluid sample flowing through a microfluidic channel. Capture is performed using orbiting magnetic microbeads, which can easily be extracted from the system for analysis after salmonella capture. Numerical simulations are used to model the dynamics of the system, which consists of a microchannel filled with a viscous fluid, model salmonella, magnetic microbeads and a series of angled parallel ridges lining the top of the microchannel. Simulations provide a statistical measure of the ability of the system to capture target salmonella. Our modeling findings guide the design of a lab-on-a-chip experimental device to be used for the detection of salmonella from complex food samples, allowing for the detection of the bacteria at the food source and preventing the consumption of contaminated food. Such a device can be used as a generic platform for the detection of a variety of biomaterials from complex fluids. 1 This work is supported by a grant from the United States Department of Agriculture. 8:26AM M10.00003 Programmable Electro Osmotic Lab on a Chip , ANDREAS G. CLASS, Karlsruhe Institute of Technology — We propose to use a 2D check-board patterned surface with alternating zeta potential made of semiconductors and individually controllable electrodes surrounding each field to drive by electro osmosis an arbitrary flow along the surface within the cavity of a lab-on-a-chip. In contrast to other fluid mechanic devices the flow is not driven by pressure gradients but rather by a controllable fluid velocity within the Debay boundary layer. Thus fluid is transported like a parcel on a conveyor belt. The use of alternating zeta potential fields and alternating electrode polarities allows to transport flow along multiple fields without the need to increase voltage. Basic functionality of the chip is accomplished by appropriate programming: fluid transport along straight and curved path, merging and splitting flow paths, flow crossing by red light traffic control, and mixing. Implementing sensors for electric resistance on the Lab-On-A-Chip allows to program a diagnosis application using electrophoresis for detection. Transport within the Lab-On-A-Chip can be described by Stokes-flow subject to the boundary conditions given by asymptotic theory in the thin-Debay-layer-limit describing field driven electro kinetic effects. 8:39AM M10.00004 Transport Mechanisms of Circulating Tumor Cells in Microfluidic Devices1 , KAUSHIK RANGHARAJAN, A.T. CONLISK, SHAURYA PRAKASH, The Ohio State University — Lab-on-a-chip (LoC) devices are becoming an essential tool for several emerging point-of-care healthcare needs and applications. Among the plethora of challenging problems in the personalized healthcare domain, early detection of cancer continues to be a challenge. For instance, identification of most tumors occurs by the time the tumor comprises approximately 1 billion cells, with poor prognosis for metastatic disease. The key obstacle in identifying and subsequent capture of circulating tumor cells (CTCs) is that the amount of CTCs in the blood stream is ∼ 1 in 109 cells. The fundamental challenge in design and fabrication of microfluidic devices arises due to lack of information on suitable sorting needed for sample preparation before any labeling or capture scheme can be employed. Moreover, the ability to study these low concentration cells relies on knowledge of their physical and chemical properties, of which the physical properties are poorly understood. Also, nearly all existing microfluidic mixers were developed for aqueous electrolyte solutions to enhance mixing in traditional low Re flows. However, no systematic studies have developed design rules for particle mixing. Therefore, we present a numerical model to discuss design rules for microscale mixers and sorters for particle sorting for high efficiency antibody labeling of CTCs along with presenting a pathway for a device to capture CTCs without the need for labeling based on particle electrical properties. 1 NSF Nanoscale Science and Engineering Center (NSEC) for the Affordable Nanoengineering of Polymeric Biomedical Devices EEC-0914790. 8:52AM M10.00005 Flow against the grain: a Newtonian microfluidic diode , JOSÉ ALVARADO, JEAN COMTET, ANETTE PEKO HOSOI, MIT — Many biological structures are coated with a dense layer of fine hairs immersed in fluid, such as ciliary beds and microvilli brushes. These elastically deformable hairs are mechanically coupled to the fluid and can thus transduce mechanical forces. When hairs are slanted (forming a sub-normal angle to the surface) they deform differentially depending on whether fluid flows with or against the grain. Using theory and experiment, we show that a channel coated with slanted hairs leads to an anisotropic pressure drop. Surprisingly, this anisotropy holds for Newtonian fluids at arbitrarily low Reynolds numbers. We suggest that beds of slanted hairs could be used to build a Newtonian microfluidic diode. 9:05AM M10.00006 Inertial instability of viscosity-stratified flows in microchannels1 , XIAOYI HU, THOMAS CUBAUD, Stony Brook University — The hydrodynamic stability of stratifications made between miscible fluids having large differences in viscosity is experimentally investigated in square microchannels. Parallel fluid layers with a fast central stream and a slow sheath flow are produced by focusing a low-viscosity fluid into a high-viscosity fluid in a straight microchannel. Although such fluid arrangements are typically governed with the flow rate ratio and the viscosity contrast at low Reynolds numbers Re, the formation of periodic wave trains at each fluid interface is observed for moderate Re. Several functional relationships are developed for the propagating velocity, size, and frequency of the generated waves over a range of viscosities and flow rates. In particular, we demonstrate the wave phase locking for small central streams and show the production of high-viscosity fluid ligaments at the wave crests. In this regime, minute amount of high-viscosity fluid is entrained and blended into the low-viscosity fluid recirculating plumes formed by the traveling interfacial waves. 1 This work is supported by NSF (CBET-1150389) 9:18AM M10.00007 Modeling and Simulation of Molecular Couette Flow in a Wide Range of Knudsen Number , LI-SHI LUO, Old Dominion University, WEI LI, Dept. of Math. & Stat., Old Dominion University, Norfolk, USA, ZHAOLI GUO, Huazhong University of Science and Technology, Wuhan, China, JIE SHEN, Dept. of Math., Purdue University, West Lafayette, USA, SHIDONG JIANG, Dept. of Math., New Jersey Institute of Technology, Newark, USA — We consider the planer Couette flow in a wide range of Knudsen number 0.003 ≤ k ≤ 10.0. We first solve the integral equation derived from the linearized BGK equation. We then use the molecular dynamics (MD) to simulation the Couette flow with van der Waals interactions between channel walls and molecules. The kinetic solution and the MD are used to construct macroscopic model to simulation the flow. The macroscopic model is based on the lattice Boltzmann equation (LBE) with multiple-relaxation-time (MRT) model for collisions. The proposed MRT-LBE approach is shown to be effective and efficient to simulate molecular gaseous Couette flow in the entire range of Knudsen number. 9:31AM M10.00008 Closing the gap: Exploring the limits of lubrication theory via molecular dynamics , AMIR M. RAHMANI, MEHLAM JUPITERWALA, YANG SHAO, CARLOS E. COLOSQUI, Department of Mechanical Engineering, Stony Brook University — Advances in nanofabrication allow the engineering of nano-electromechanical systems and nanofluidic devices with dimensions on the order of 1 to 100 nanometers. Lubrication flows with characteristic lengths approaching the molecular scale have become ubiquitous in a wide spectrum of applications ranging from biomass sensing and atomic force microscopy to drug delivery and synthesis of nanomaterials. At nanometer scales, the effects of thermal fluctuations, disjoining pressure, and finite atomic size produce various phenomena beyond the reach of classical continuum hydrodynamics. We will present a theoretical and computational study on nanoscale lubrication flows where bodies immersed in dense fluid media either are kept under static conditions or are allowed to undergo thermal fluctuations about a prescribed position. We find that under static conditions, continuum-based lubrication models can accurately predict hydrodynamic forces computed via molecular dynamics simulations at surprisingly small scales (i.e., for flows having sub-nanometer-sized lubrication gaps). Thermal vibration, however, can induce drag reduction and other dynamic effects that can potentially be described by extending conventional lubrication models. 9:44AM M10.00009 Flow rate and slip length measurements of water in single micrometer pipes , PETER TABOREK, ANERUDH KANNAN, DAVID MALLIN, ANGEL VELASCO, University of California, Irvine — Measurements of pressure driven water flows in hydrophobic and hydrophilic fused quartz capillaries of 1.8 um diameter are compared. Typical flow rates of 1 picoliter/s and pressure drops up to 25 Atm were used. Water exited the capillaries into an oil reservoir where the volume of the pendant drop could be monitored using time lapse photography. The typical growth rate for the drop diameter was ∼ 300 µm per day. The drop size saturates due to diffusion at the interface. For the untreated quartz capillary the results are consistent with a slip of zero. The hydrophilic capillaries are chemically treated with octadecyltrichlorosilane (OTS) to form hydrophobic surfaces. Successful surface preparation is confirmed with the absence of capillary rise. Our technique can detect slip lengths above 20 nm. 9:57AM M10.00010 Particle orientation during thermophoretic transport in a gas phase , STEFFEN HARDT, TOBIAS BAIER, Center of Smart Interfaces, TU Darmstadt, SAMIR SHRESTHA, SUDARSHAN TIWARI, AXEL KLAR, Fachbereich Mathematik, TU Kaiserslautern — Using numerical and (semi)analytical techniques it is shown that during thermophoretic transport in a gas small particles may take a preferred orientation. For that purpose spherical model particles are considered, consisting of two hemispheres with diffuse and specular reflection boundary conditions, respectively. A Monte-Carlo method is used to simulate the translational and rotational dynamics of a particle colliding with surrounding gas molecules. The simulations show that a particle exposed to a temperature gradient takes a preferred orientation, moving through the gas with its diffusely reflecting hemisphere pointing towards the direction where the temperature is lower. In addition to that, in the free molecular flow regime the Langevin equation is used to study the rotational dynamics. In this regime it is possible to derive analytical expressions for the torque acting on the particle and for the dissipation force related to rotational motion. The stationary solution of the Langevin equation gives the probability density function for the particle orientation. The results could be of relevance for a number of processes in which nanoparticles are synthesized in a gas phase and deposited on a cold surface. Tuesday, November 25, 2014 8:00AM - 9:57AM Session M11 Instability: General — 3007 - Malcolm Andrews, Texas A&M University 8:00AM M11.00001 Encapsulated formulation of the selective frequency damping method1 , BASTIEN JORDI, COLIN COTTER, SPENCER SHERWIN, Imperial College London — We present an alternative “encapsulated” formulation of the selective frequency damping (SFD) method. This (time-discrete) formulation makes use of splitting methods, which means that it can be wrapped around an existing time-stepping code as a “black box.” Hence the implementation of a steady-state solver is very easy because the existing unsteady solver does not need to be modified. It is simply called each time-step and a linear operator (modelling a feedback control and a low-pass time filter) is applied to its outcome. The method is first applied to a scalar problem in order to analyse its stability and highlight the roles of the control coefficient and the filter width in the convergence (or not) towards the steady-state. Then we show that by knowing the most unstable eigenmode of a fluid flow, we can guarantee convergence of the SFD method towards the steady-state solution. Finally, we discuss the possibility of coupling the SFD method with an Arnoldi method. The goal is to approximate the eigenmodes of an unstable flow and then to adjust the parameters of the SFD method to ensure convergence towards the steady-state. We are currently using this approach to obtain a steady-state solution of co-rotating Batchelor vortices and we present our latest results. 1 Seventh Framework Programme of the European Commission - ANADE project under grant contract PITN-GA-289428 8:13AM M11.00002 Stability Results on Multi-Layer Radial Hele-Shaw Flows with Variable Viscosity , CRAIG GIN, PRABIR DARIPA, Texas A&M Univ — Saffman-Taylor instability, which occurs when a less viscous fluid drives a more viscous fluid, has been studied for many years and has a wide range of applications. In particular, an understanding of this phenomenon is helpful in the attempt to maximize the effectiveness of chemically enhanced oil recovery techniques. We study this instability through linear stability analysis of three-layer radial Hele-Shaw flows of immiscible fluids in which the middle layer consists of a variable viscosity fluid. We study the growth rate of instabilities both numerically and analytically, including the derivation of upper bounds. We also connect this problem to the related cases of variable viscosity rectilinear flows and constant viscosity radial flows. We attempt to extend this work to an arbitrary number of fluid layers. 8:26AM M11.00003 Unsteady regimes in a T-mixer , MARIA VITTORIA SALVETTI, SIMONE CAMARRI, DICI, University of Pisa, ANDREA FANI, EPFL, Lausanne — Micro T-mixers are devices aimed at providing efficient mixing. Most of the studies in the literature focused on the steady engulfment regime, characterized by a loss of the flow symmetries in the outflow channel which leads to a considerable increase of the mixing efficiency. Unsteady regimes were recently observed for Reynolds numbers (Re) larger than the steady engulfment critical value. We investigated these regimes for a given T-mixer configuration through direct numerical simulations. A first unsteady regime appears, in which the flow remains asymmetric in the mean but becomes periodic in time. As Re is further increased, the flow remains time-periodic but it continuously switches between a symmetric configuration and an asymmetric one. Three-dimensional linear stability and sensitivity analyses are also used to characterize the instability leading to the unsteady asymmetric regime (UAR), which is interesting for applications due to its high mixing efficiency. The largest sensitivity was observed to base-flow modifications introduced close to the 3D vortical structures forming at the confluence between the inlet channels. Finally, it is found that for a flat inlet velocity profile the UAR onset is delayed at larger Re than for a fully developed profile. 8:39AM M11.00004 Stability theory for the synchronized waving of marine grass , RAVI SINGH, Brown University, AMALA MAHADEVAN, Woods Hole Oceanographic Institution, SHREYAS MANDRE, Brown University, L.M. MAHADEVAN, Harvard — Synchronized waving of grass blades in the presence of fluid flow has been observed in cases such as wheat field in wind, marine grass in tidal currents. The synchronous motion can have important environmental and ecological impact via mixing of fluid due to waving. When the hydrodynamic and elastic time scales are well separated, this waving is thought to be due to Kelvin-Helmholtz instability resulting from an inflection point in the flow profile. We find that the inflection point is located near the tip of grass canopy. We extend the Orr-Sommerfeld equation for the stability of a shear flow to include a continuum mean-field approximation for the vegetation, thus capturing the essential ingredients for flow instability leading to coherent waving. Our linear stability analysis shows that the flow in presence of grass become unstable not only through a mechanism of Kelvin-Helmholtz instability but also through shear instability of flow above grass. We also find that flow with low submergence ratio of grass becomes unstable due to Kelvin-Helmholtz instability whereas flow high submergence ratio becomes unstable due to shear instability of flow above the grass. Numerical results demonstrating these instability mechanism will also be presented. 8:52AM M11.00005 Hydrodynamic Stability Analysis on Sheared Stratified Flow in a Convective Flow Environment , YUAN XIAO, WENXIAN LIN, College of Sccience, Technology & Engineering, James Cook University, STEVEN ARMFILED, MICHAEL KIRKPATRICK, School of Aerospace, Mechanical and Mechatronic Engineering, The University of Sydney, YINGHE HE, College of Sccience, Technology & Engineering, James Cook University, FLUID DYNAMICS RESEARCH GROUP, JAMES COOK UNIVERSITY TEAM, FLUID DYNAMICS RESEARCH GROUP, UNIVERSITY OF SYDNEY TEAM — A hydrodynamic stability analysis on the convective sheared boundary layer (SCBL) flow, where a sheared stratified flow and a thermally convective flow coexist, is carried out in this study. The linear unstable stratifications representing the convective flow are included in the TaylorGoldstein equations as an unstable factor Jb. A new unstable region corresponding to the convective instability, which is not present in pure sheared stratified flows, is found with the analysis. It is also found that the boundaries of the convective instability regions expand with increasing Jb and interact with the sheared stratified instability region. More results will be presented at the conference 9:05AM M11.00006 Study of spatial growth of disturbances in an Incompressible Double Shear Layer flow configuration , HARESHRAM NATARAJAN, CSRC-SanDiego, GUSTAAF JACOBS, SanDiego State University — The spatial growth of disturbance within the linear instability regime in an incompressible 2D double shear layer flow configuration is studied by performing a Direct Numerical Simulation. The motivation of this study is to characterize the effect of the presence of an additional shear layer on the spatial growth of a shear layer instability. Initially, a DNS of an incompressible single shear layer is performed and the spatial growth rate of various disturbance frequency modes are validated with Linear Stability Analysis. The addtional shear layer is found to impact the spatial growth rates of the different disturbances and the frequency of the mode with the maximum growth rate is found to be shifted. 9:18AM M11.00007 Stability of a rolling fluid filled cylinder , ROHIT SUPEKAR, MAHESH PANCHAGNULA, Indian Institute of Technology Madras — We present an analytical solution to the problem of a fluid filled hollow cylindrical shell rolling on an inclined plane and then investigate the temporal stability of the system using linear stability analysis. We study the motion in two dimensions by analyzing the interaction between the fluid and the hollow cylinder. We show that the terminal state is associated with a constant acceleration, similar to a rigid body motion. Surprisingly, it is independent of the liquid viscosity and only depends on the ratio of the mass of the shell to the mass of the fluid contained (say, πm ). We analyze this base flow for its stability behavior using the frozen-time approximation. In this approach, we treat time as a parameter, the evolution of which causes the flow to transition from a stable to an unstable state. The point of neutral stability is noted and the spatial modes that show the maximum growth rate are analyzed. It was observed that instability sets in due to long wavelength axial waves, which are transverse to the flow direction. We find a critical Reynolds number based on the time to instability, above which the flow becomes unstable. Again, this Reynolds number appears to be only a function of πm . 9:31AM M11.00008 Stability study of flows around an airfoil based on energy gradient method , JADE JUNQUA1 , HUA-SHU DOU2 , Zhejiang Sci-Tech University, FLUID MECHANICS RESEARCH TEAM — Numerical simulation is carried out to study the turbulent flow around an airfoil and the energy gradient theory is used to analyze the stability of the flow. The governing equations are the Reynolds averaged Navier-Stokes equations for compressible flow and the k-epsilon turbulent model is used to close the system. The finite volume method and the time marching scheme are used to solve the unsteady governing equations. The simulation and calculation have been completed for various attack angle of the airfoil, from 0 and 8 degree. The Reynolds number is about 3.5X10**6 for all situations, and the Mach number is about 0.15. The flow is considered as shear driven flow and the distribution of the energy gradient function K around the airfoil is calculated with the simulation data. The results shows good agreement between the distribution of the energy gradient function and the experimental observations in regard of the turbulent intensity, while there is little relation between the distribution of the vorticity and the turbulent intensity. It is concluded that energy gradient function dominates the flow stability and the sustenance of turbulence rather than the magnitude of vorticity. 1 Graduate 2 Professor Student in Fluid Mechanics in Fluid Mechanics; AIAA Associate Fellow 9:44AM M11.00009 A straightforward characterization of non-modal effects from the evolution of linear dynamical systems1 , CRISTOBAL ARRATIA, Univ de Chile — A simple construction will be shown, which reveals a general property satisfied by the evolution in time of a state vector composed by a superposition of orthogonal eigenmodes of a linear dynamical system. This property results from the conservation of the inner product between such state vectors evolving forward and backwards in time, and it can be simply evaluated from the state vector and its first and second time derivatives. This provides an efficient way to characterize, instantaneously along any specific phase-space trajectory of the linear system, the relevance of the non-normality of the linearized Navier-Stokes operator on the energy (or any other norm) gain or decay of small perturbations. Examples of this characterization applied to stationary or time dependent base flows will be shown. 1 CONICYT, Concurso de Apoyo al Retorno de Investigadores del Extranjero, folio 821320055. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M12 Drops: Wetting and Spreading I — 3018 - Martin Brinkmann, Max Planck Gesellschaft 8:00AM M12.00001 Hydrodynamics of micro-scale surface flows induced by triangulated droplet stream impingement array , TAOLUE ZHANG, JORGE ALVARADO, Texas A&M Univ, ANOOP KANJIRAKAT, REZA SADR, Texas A&M University at Qatar, TAMU-TAMUQ TEAM — A study of surface flow hydrodynamics caused by triple stream of impinging droplets arranged in a triangular array is presented. Triple streams of mono-dispersed droplets were produced using a piezoelectric droplet generator with the ability to adjust parameters such as droplet impingement frequency, droplet diameter, droplet velocity and spacing between adjacent impinging droplet streams. A translucent Zinc Selenide (ZnSe) substrate was used for characterizing the hydrodynamic phenomena of the droplet impingement zone using a high speed imaging technique. Surface jet-like fluid flows were observed among impact craters during the high-frequency droplet impingement process. A transition from laminar-like to turbulent-like surface jet flow was observed by increasing droplet Weber number or decreasing droplet impingement spacing. A correlation based on visual observations has been postulated by taking into account the droplets’ Weber number (W e) and non-dimensional droplet impingement spacing (S ∗ ). The correlation has a mathematical form of S ∗ · W en = K, where K is a constant. One major result from the study is the relative accuracy of the postulated model in predicting the laminar-turbulent like transition in terms of W e and S ∗ . 8:13AM M12.00002 Droplet Spreading with Sol-Gel Transition , MAZIYAR JALAAL, BORIS STOEBER, Department of Mechanical Engineering, The University of British Columbia, Vancouver, BC, Canada, NEIL J. BALMFORTH, Department of Mathematics, The University of British Columbia, Vancouver, BC, Canada — The impact and spreading of liquid droplets on a smooth solid substrate is a classical subject with several industrial applications such as ink-jet printing, spray cooling, coating, and many others. For many of these deposition processes, controlling the final shape of the drop is critical. In the current research, a new technique for controlling the spreading of droplets impacting a substrate is presented. This technique exploits the rheology of a thermo-responsive polymer solution that undergoes a reversible sol/gel transition above a critical temperature. Experiments are conducted using a combination of shadowgraphy and micro-PIV to observe spreading drops. It is shown that the final diameter of a droplet can be controlled through the temperature of the substrate and the tunable sol/gel transition temperature of the fluid.A mathematical model is provided to further elucidate the flow dynamics. 8:26AM M12.00003 Advancing contact angles on large structured surfaces , YUMIKO YOSHITAKE, YOSHINORI ITAKURA, JUNICHI GOBO, TSUTOMU TAKAHASHI, Nagaoka University of Technology — To understand wetting phenomena on complex surfaces, simple modeling experiments in two-dimension system would be one of the most efficient approaches. We develop a new experimental method for wetting dynamics using a large pseudo two- dimensional droplet. This method is useful to examine theoretical studies developed in two dimensional systems. In this study, we examine a pinning and depinning phenomena on millimeter-size structured surface to explain the origin of contact angle hysteresis. Contact lines of the droplet are pinned and deppined at the edge of surface texture. The contact lines can move when the contact angle is equal to the Young’s contact angle which are determined by the balance of the surface and interfacial tension immediate vicinity of the contact lines, which is different from the Wenzel’s low. Our approach enables to realize a macroscopic modelling experiment of wetting on complex surfaces, which opens a path to design functional surfaces with chemical and physical structure. 8:39AM M12.00004 Sessile drops and condensation on chemically patterned micropillars , OREST SHARDT, PRASHANT WAGHMARE, Department of Mechanical Engineering, University of Alberta, DANIEL OREJON, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, NAGA GUNDA, Department of Mechanical Engineering, University of Alberta, YASUYUKI TAKATA, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, SUSHANTA MITRA, Department of Mechanical Engineering, University of Alberta — We examine the state of sessile drops on silicon micropillars with patterned wettability as well as condensation of water onto such surfaces. These patterned micropillar arrays were created by treatment with a perfluoroalkylsilane to create a hydrophobic surface and subsequent patterning with sodium hydroxide solution to create hydrophilic regions. The surfaces were characterized by measuring the contact angles and observing the states of sessile drops, and the results are compared with those of uniformly hydrophobic and hydrophilic pillars. The nature of condensation onto patterned pillars has been examined with environmental scanning electron microscopy (ESEM). The results show the initial dropwise condensation on the different types of pillars and the transition to a film. Surfaces that combine texturing with chemical patterning could be useful for enhanced control of condensation and droplet motion. 8:52AM M12.00005 Electrostatically-driven precursor films , SEYED REZA MAHMOUDI, KRIPA K. VARANASI, MIT — Here, we report a new class of electrostatically assisted precursors containing microscopic charged particles. This precursor manifests itself as the late stage of forced-spreading of a macroscopic dielectric film subjected to a unipolar ionic bombardment in a gas containing particulates. We put a model forward to predict dynamic behaviour of this electrostatic precursor dynamics. The spreading of the precursor film is predicted to be proportional to the square root of exposure time, which is consistent with the ellipsometric measurements. 9:05AM M12.00006 Detachment of Sessile Droplets in Immiscible Fluids Using Electrowetting , JIWOO HONG, SANG JOON LEE, Pohang University of Science and Technology (POSTECH) — The detachment (or removal) of droplets from a solid surface is an indispensable process in numerous practical applications. Here we firstly detach sessile droplets in immiscible fluids from a hydrophobic surface by electrowetting. The critical conditions for droplet detachment are determined by exploring the retracting dynamics for a wide range of driving voltages and physical properties of fluids. The relationships between physical parameters and dynamic characteristics of retracting and jumping droplets, such as contact time and jumping height, are also established. The threshold voltage for droplet detachment in oil with high viscosity is largely reduced by electrowetting actuations with a square pulse. Finally, by using DC and AC electrowetting actuations, we demonstrate the detachment of oil droplets with very low contact angle on a hydrophobic surface in water. 9:18AM M12.00007 Inclusion of fluid-solid interaction in Volume of Fluid to simulate spreading and dewetting for large contact angles1 , KYLE MAHADY, SHAHRIAR AFKHAMI, LOU KONDIC, New Jersey Institute of Technology — The van der Waals (vdW) interaction between molecules is of fundamental importance in determining the behavior of three phase systems in fluid mechanics. This interaction gives rise to interfacial energies, and thus the contact angle for a droplet on a solid surface, and additionally leads to instability of very thin liquid films. We develop a hybrid method for including a Lennard-Jones type vdW interaction in a finite volume, Volume of Fluid (VoF) based solver for the full two-phase Navier-Stokes equations. This method includes the full interaction between each fluid phase and the solid substrate via a finite-volume approximation of the vdW body force. Our work is distinguished from conventional VoF based implementations in that the contact angle arises from simulation of the underlying physics, as well as successfully treating vdW induced film rupture. At the same time, it avoids the simplifications of calculations based on disjoining-pressure, where the vdW interaction is included as a pressure jump across the interface which is derived under the assumption of a flat film. This is especially relevant in the simulation of nanoscale film ruptures involving large contact angles, which have been studied recently in the context of bottom-up nanoparticle fabrication. 1 This work is partially supported by the grants NSF DMS-1320037 and CBET-1235710. 9:31AM M12.00008 Exact solutions for contact lines on a soft substrate with uniform surface tension , LAURENT LIMAT, Laboratoire MSC, UMR 7057 of CNRS and University Paris Diderot, France, JULIEN DERVAUX, Laboratoire LIED, UMR 8236 of CNRS and University Paris Diderot, France — We have found an analytical solution describing the deformations of a soft substrate with uniform surface tension loaded by a straight contact line. Starting from the exact solution of this Flamant-Cerruti problem, we have extended our solution to contact lines with finite microscopic width, and to the case of a liquid strip between two straight contact lines (a 2D rivulet). These solutions are close to the approximate logarithmic ones proposed by one of us [1], except when two length scales of the problem are not separated. We discuss the liquid/solid force transmission, and the double transition of Lubbers et al [3], when two contact lines at finite distance are deforming a material with increasing softness. We provide analytical expressions for substrate distortion, as a function of the three length scales involved: distance between contact lines, microscopic scale, and elastocapillary length. Finally we discuss the selection of apparent contact angle, the possible extension to circular contact lines and the comparison with available experiments. [1] L. Limat, EPJ-E Soft Matter 35, 134 (2012). [2] L. A. Lubbers, J.H. Weijs, L. Botto, S. Das, B. Andreotti and J.H. Snoeijer, J. Fluid Mech. 747, R1 (2014). 9:44AM M12.00009 Role of contact line evaporation on the spreading of viscous droplet , WASSIM BOU-ZEID, DAVID BRUTIN, Aix-Marseille University - IUSTI UMR 7343 — The effect of relative humidity and viscosity on the spreading dynamics of waterglycerol mixtures was analyzed for a range of relative humidities from 20% to 80%. Droplets of identical volume were deposited on ultra-clean microscope glass substrates. We demonstrated that, in addition to the competition between viscous forces, capillary forces and disjoining pressure, droplet spreading was also affected by the evaporation that occurred at the triple line. We provide an updated Tanner’s law, which was modified to take into account the evaporative contribution. The same mechanism can be applied to adjust any fluid to Tanner’s coefficient of 1/10. 9:57AM M12.00010 Tailoring concentration gradients in microfluidic networks , CYPRIEN GUERMONPREZ, CHARLES BAROUD, SEBASTIEN MICHELIN, LadHyX and Department of Mechanics, Ecole Polytechnique, CNRS 91128 Palaiseau — We aim to produce precise concentration gradients in a microfluidic device in order to generate large number of droplets with a wide range of solute concentration. A method to study the distribution of a diffusing species in a highly parallel microfluidic network is proposed exploiting the spatial distribution of hydrodynamic resistances within the network. Starting from two co-flowing streams, the main channel supplies 10 to 64 side branches connected to a single outlet channel. Experiments and theoretical analysis are carried out for low Reynolds numbers and moderate to high Péclet numbers. The distribution of flow rates within the network is determined from its geometry and the distribution of hydrodynamic resistances and show maximum flow rates through the first and the last branches. Once the velocity distribution is known, a finite-difference method is used to predict the diffusion of a dichlorophenolindophenol solution in pure water. Both experimental and numerical results yield a variety of concentration distribution profiles that range from nearly uniform (small Pe ) to linear and sigmoidal profiles (larger Pe ). A good understanding of the underlying hydrodynamics enables the design of devices that generate rapidly precise chemical gradients. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M13 Drops: Heat Transfer and Evaporation II — 3020 - David B. Thiessen, Washington State University 8:00AM M13.00001 Marangoni Effect on the Shape of Freely Receding Evaporating Sessile Droplets of Perfectly Wetting Liquids1 , YANNIS TSOUMPAS, SAM DEHAECK, ALEXEY REDNIKOV, PIERRE COLINET, Université Libre de Bruxelles — Freely receding evaporating sessile droplets of perfectly wetting liquids (HFE-7100, 7200 and 7500), with small finite contact angles induced by evaporation, are studied with a Mach-Zehnder interferometer. Surprisingly, the experimentally obtained profiles turn out to deviate from the classical macroscopic static shape of a sessile droplet (as determined by gravity and capillarity), often used when modeling evaporating droplets. These deviations can be seen in two ways. Namely, either the droplet appears to be inflated as compared to the classical static shape assuming the same contact angle and contact radius, or the apparent contact angle appears lower than the classical static one assuming the same volume and contact radius. In reality, the experimental profiles exhibit a local decrease of the slope near the contact line, which we attribute to the Marangoni effect in an evaporating sessile droplet. In this case, the radially inward (along the liquid-air interface) direction of the flow delivers more liquid to the center of the droplet making it appear inflated. When the Marangoni effect is weak, as in the case of the poorly volatile HFE-7500, no significant influence is noticed on the drop shape. The experimental results are compared with the predictions of a lubrication-type theoretical model that incorporates the evaporation-induced Marangoni flow. 1 Financial support of FP7 Marie Curie MULTIFLOW Network (PITN-GA-2008-214919), ESA/BELSPO-PRODEX, BELSPO-µMAST (IAP 7/38) & FRS-FNRS is gratefully acknowledged. 8:13AM M13.00002 How long does it take for sessile droplets to evaporate? , STEPHEN WILSON1 , JUTTA STAUBER, BRIAN DUFFY, University of Strathclyde, KHELLIL SEFIANE, University of Edinburgh — The evaporation of sessile droplets plays a crucial part in many practical applications, and in many of these applications it is important to be able to understand and/or control the lifetimes of droplets. The lifetime of an evaporating droplet depends on the manner in which it evaporates. There are various qualitatively different modes of droplet evaporation, of which the most extreme are the constant radius mode (in which the contact line is always pinned) and the constant angle mode (in which the contact angle θ always takes its initial value θ = θ0 ), and probably the most commonly occurring is the stick-slide mode (in which the drop initially evaporates in a constant radius phase until θ reaches a critical transition angle θ ∗ , and thereafter evaporates in a constant angle phase with θ = θ ∗ ). In this talk we describe a theoretical model for the stick-slide mode and discuss the relationship between θ0 and θ ∗ and its implications. Theoretical predictions for the lifetimes of droplets are compared with previously published experimental results. Further details of the theoretical model are given in the recent paper by Stauber, Wilson, Duffy and Sefiane [J. Fluid Mech. 744, R2 (2014)]. 1 Currently a Leverhulme Trust Research Fellow supported by award RF-2013-355. 8:26AM M13.00003 Wettability Patterning for Enhanced Dropwise Condensation1 , ARITRA GHOSH, RANJAN GANGULY2 , CONSTANTINE MEGARIDIS, University of Illinois at Chicago — Dropwise condensation (DwC), in order to be sustainable, requires removal of the condensate droplets. This removal is frequently facilitated by gravity. The rate of DwC heat transfer depends strongly on the maximum departing droplet diameter. Based on wettability patterning, we present a facile technique designed to control the maximum droplet size in DwC within vapor/air atmospheres, and demonstrate how this approach can be used to enhance the corresponding heat transfer rate. We examine various hydrophilic-superhydrophilic patterns, which, respectively sustain DwC and filmwise (FwC) condensation on the substrate. The fabrication method does not employ any hydrophobizing agent. By juxtaposing parallel lines of hydrophilic (CA ∼ 78◦ ) and superhydrophilic (CA ∼ 0◦ ) regions on the condensing surface, we create alternating domains of DwC and FwC. The average droplet size on the DwC domain is reduced by ∼ 60% compared to the theoretical maximum, which corresponds to the line width. We compare heat transfer rate between unpatternend DwC surfaces and patterned DwC surfaces. Even after sacrificing 40% of condensing area, we achieve up to 20% improvement in condensate collection rate using an interdigitated superhydrophilic pattern, inspired by the vein network of plant leaves. The bioinspired interdigitated pattern is found to outperform the straight hydrophilic-superhydrophilic pattern, particularly under higher vapor loadings in an air/vapor ambient atmosphere. 1 NSF 2 On STTR Grant 1331817 via NBD Nano sabbatical from Jadavpur University, Kolkata, India 8:39AM M13.00004 Simulation Prediction of Transient Dropwise Condensation1 , ASHLEY MACNER, SUSAN DANIEL, PAUL STEEN, Cornell University — In order to design effective surfaces for large-scale dropwise condensation, an understanding of how surface functionalization affects drop growth and coalescence is needed. The long term technological goal is a set of design conditions to help NASA achieve maximum heat transfer rates of waste heat generated from electronics and habitable environments under microgravity conditions. Prediction of condenser surface heat transfer performance requires accurate simulation and modeling of the evolution of populations of drops in time. At shorter times, drops are primarily isolated and grow mainly by condensation onto the liquid-gas interface. At longer times, drops grow mainly by coalescence with neighbors. Simulation of dropwise condensation on a neutrally wetting surface and comparison with our previous experimental results is reported. A steady-state single drop conduction model is empirically fitted to determine a temperature profile that captures the drop size evolution. The simulation accurately predicts the continuous time evolution of number-density of drops, drop-size distributions, total condensate volume, fractional coverage, and median drop-size for both transient and steady states, all with no free parameters. 1 This work was supported by a NASA Office of the Chief Technologist’s Space Technology Research Fellowship. 8:52AM M13.00005 Universality of Tip Singularity Formation in Freezing Water Drops , OSCAR ENRIQUEZ, University of Twente, ALVARO MARIN, Bundeswehr University Munich, PHILIPPE BRUNET, CNRS/Université Paris 7, PIERRE COLINET, Université Libre de Bruxelles, JACCO SNOEIJER, University of Twente — A drop of water on a cold plate freezes from the bottom up and forms a pointy tip in the last moments of the process. Although this phenomenon is known to be caused by the expansion of water upon freezing, a quantitative description of the tip singularity has remained elusive. Our systematic measurements of the angles of the conical tip, for a wide range of temperatures and wetting angles, suggest a universal, self-similar mechanism that does not depend on the rate of solidification. Furthermore, using a Hele-Shaw geometry, we have observed the dynamics of the solidification front. Here we demonstrate how the geometry of the freezing front, determined by heat transfer considerations, is crucial for the tip formation. We propose a geometrical model for the tip formation and derive resulting tip angles analytically, in good agreement with the experiments. 9:05AM M13.00006 Edge effects on water droplet condensation , LAURENT ROYON, M.S.C. University Paris Diderot, ANNE MONTGRUEL, P.M.M.H. ESPCI, MARIE GABRIELLE MEDICI, L.PS.M. University of Nice, DANIEL BEYSENS, P.M.M.H. ESPCI — The effect of geometrical or thermal discontinuities on the growth of water droplets condensing on a cooled substrate is investigated. Edges, corners, cooled/non cooled boundaries can have a strong effect on the vapor concentration profile and mass diffusion around the drops. In comparison to growth in a pattern where droplets have to compete to catch vapor, which results in a linear water concentration profile directed perpendicular to the substrate, droplets near discontinuities can get more vapor (outer edges, corners), resulting in faster growth or less vapor (inner edges), giving lower growth. When the cooling heat flux limits growth instead of mass diffusion (substrate with low thermal conductivity, strong heat exchange with air), edges effects can be canceled. In certain cases, the growth enhancement can reach nearly 500% on edges or corners which, on an inclined substrate, make droplets near the edges detach sooner than in the middle of the substrate. This effect is frequently observed with dew condensing on windows or car windshields. Such droplets, acting as wipers, can thus appreciably increase dew collection on a substrate. 9:18AM M13.00007 Star-shaped oscillations of Leidenfrost droplets on a curved surface , XIAOLEI MA, JUAN-JOSÉ LIÉTOR-SANTOS, JUSTIN BURTON, Department of Physics, Emory University — We investigate the spontaneous oscillations of a Leidenfrost droplet, which is levitated by a cushion of evaporated vapor on a hot surface. The oscillations exhibit a star-shaped pattern determined by a standing wave along the droplet periphery, and obey a quasi-2D dispersion relation. The bowl-shaped curvature of the surface suppresses the buoyancy-driven Rayleigh-Taylor instability in the vapor layer, allowing for very large droplets with up to 13 lobes. Although droplets of a given size can theoretically contain various oscillatory modes, we observe only one mode of oscillation, so that all star-shaped droplets have nearly the same frequency regardless of size. We suspect that the origin of this mode selection is due to a parametric coupling between vertical and azimuthal oscillations of the droplet, similar to experiments of droplets on hydrophobic, vibrated surfaces [1]. In order to investigate the phenomenon further, we also measure the pressure variations beneath the droplet during quiescent and oscillatory phases. [1] P. Brunet and J. H. Snoeijer, Eur. Phys. J. Spec. Top. 192, 207 (2011). 9:31AM M13.00008 Leidenfrost effect: accurate drop shape modeling and new scaling laws , BENJAMIN SOBAC, ALEXEY REDNIKOV, Université Libre de Bruxelles, TIPs - Fluid Physics, STÉPHANE DORBOLO, Université de Liège, GRASP, Physics Departement, PIERRE COLINET, Université Libre de Bruxelles, TIPs - Fluid Physics — In this study, we theoretically investigate the shape of a drop in a Leidenfrost state, focusing on the geometry of the vapor layer. The drop geometry is modeled by numerically matching the solution of the hydrostatic shape of a superhydrophobic drop (for the upper part) with the solution of the lubrication equation of the vapor flow underlying the drop (for the bottom part). The results highlight that the vapor layer, fed by evaporation, forms a concave depression in the drop interface that becomes increasingly marked with the drop size. The vapor layer then consists of a gas pocket in the center and a thin annular neck surrounding it. The film thickness increases with the size of the drop, and the thickness at the neck appears to be of the order of 10-100 µm in the case of water. The model is compared to recent experimental results [Burton et al., Phys. Rev. Lett., 074301 (2012)] and shows an excellent agreement, without any fitting parameter. New scaling laws also emerge from this model. The geometry of the vapor pocket is only weakly dependent on the superheat (and thus on the evaporation rate), this weak dependence being more pronounced in the neck region. In turn, the vapor layer characteristics strongly depend on the drop size. 9:44AM M13.00009 Tracking liquid in drying colloidal fluids with polarized light microscopy1 , KUN CHO, School of Advanced Materials Science and Engineering, Sungkyunkwan University, JUNG SOO PARK, JOON HEON KIM, Advanced Photonics Research Institute, Gwangju Institute of Science and Technology, BYUNG MOOK WEON, School of Advanced Materials Science and Engineering, SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University — When colloidal fluids dry, tracking liquid surfaces around colloids is difficult with conventional imaging techniques. Here we show that polarized light microscopy (PM) is very useful in tracking liquid surfaces during drying processes of colloidal fluids. In particular, the PM mode is not a new or difficult way but is able to visualize liquid films above colloids in real time. We demonstrate that when liquid films above colloidal particles are broken, the PM patterns appear clearly: this feature is useful to identify the moment of liquid film rupture above colloids in drying colloidal fluids. This result is helpful to improve relevant processes such as inkjet printing, painting, and nanoparticle patterning (K.C. and J.S.P. equally contributed). 1 This work (NRF-2013R1A22A04008115) was supported by Mid-career Researcher Program through NRF grant funded by the MEST. 9:57AM M13.00010 Secondary atomization pathways in burning functional droplets subjected to travelling pressure wave , ANKUR MIGLANI, SAPTARSHI BASU, Department of Mechanical Engineering, Indian Institute of Science — Self-induced internal boiling in burning multicomponent droplets and the resulting pressure upsurge is observed to initiate characteristic bubble ejection/droplet disruption events. These bubble ejections (also termed as secondary atomization events) corrugate the droplet surface and induce bulk shape deformation in the droplet. In this study, first, we identify the entire spectrum of secondary break-up modes that occur at distinct stages of droplet lifetime and at different temporal scales. Based on the increasing magnitude of their droplet-shape deformation inducing potential they range from high aspect ratio, high momentum needle type ligament break-up to low momentum, thick ligament break-up. Needle-type ejections are dominant at initial stages of droplet lifecycle and are primarily responsible for triggering only small-scale, localized surface wrinkling. In contrast, latter modes of atomization occur at later stages and initiate large-length scale droplet deformation. Second, we show that by exciting the droplet flame in its critical responsive frequency range (80 Hz ≤ f P ≤ 120 Hz) the latter high intensity modes can be suppressed. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M15 Drops: Impact on Surfaces — 3022/3024 - J.R. Saylor, Clemson University 8:00AM M15.00001 Drop impact on solid surface: Short time self-similarity , JULIEN PHILIPPI, PIERREYVES LAGRÉE, ARNAUD ANTKOWIAK, UPMC Univ Paris 06, CNRS, UMR 7190 Institut Jean Le Rond d’Alembert, Paris, France — Drop impact on a solid surface is a problem with many industrial or environmental applications. Many studies focused on the last stages of this phenomenon as spreading or splashing. In this study we are interested in the early stages of drop impact on solid surface. Inspired by Wagner theory developed by water entry community we shown the self-similar structure of the velocity field and the pressure field. The latter is shown to exhibit a maximum not near the impact point, but rather at the contact line. The motion of the contact line is furthermore shown to exhibit a transition from “tank treading” motion to pure sweeping when the lamella appears. We performed numerical simulations with the open-cource code Gerris which √ are in good agreement with theoretical predictions. Interestingly the inviscid self-similar impact pressure and velocities depend on the self-similar variable r/ t. This allows to construct a seamless uniform analytical solution encompassing both impact and viscous effects. We predict quantitatively observables of interest, such as the evolution of total and maximum √ viscous shear stresses and net total force. We finally demonstrate that the structure of the flow resembles a stagnation point flow unexpectedly involving r/ t. 8:13AM M15.00002 Free radially expanding liquid sheet in air: time- and space-resolved measurement of the thickness field1 , CHRISTIAN LIGOURE, CLARA VERNAY, LAURENCE RAMOS, Laboratoire Charles Coulomb, UMR no5221 CNRS & University Montpellier 2 — The collision of a liquid drop against a small target results in the formation of a thin liquid sheet that extends radially until it reaches a maximum diameter. We have developed an original time- and space-resolved technique to measure the thickness field of this class of liquid sheet, based on the grey level measurement of the image of a dyed liquid sheet recorded using a fast camera. This method enables a precise measurement of the thickness in the range (10 − 450) µm, with a temporal resolution equals to that of the camera. Two asymptotic regimes for the expansion of the sheet are evidenced. The scalings of the thickness with t and r measured in the two regimes are those that were predicted but never experimentally measured before. Interestingly, our experimental data also evidence the existence of a maximum of the film thickness hmax (r) at a radial position rhmax (t) corresponding to the crossover of these two asymptotic regimes. The maximum moves with a constant velocity of the order of the impact velocity, Hence, our data has allowed one to reconcile the two apparently inconsistent theoretical predictions found in the literature Thanks to our visualization technique, we also evidence an azimuthal thickness modulation. 1 Financial support from Solvay is acknowledged. 8:26AM M15.00003 The Leidenfrost temperature increase for impacting droplets on carbonnanofiber surfaces , HENDRIK STAAT, HRUDYA NAIR, University of Twente, TUAN TRAN, Nanyang Technological University, ARIE VAN HOUSELT, University of Twente, ANDREA PROSPERETTI, Johns Hopkins University, DETLEF LOHSE, CHAO SUN, University of Twente — When a droplet impacts a smooth solid plate that is heated to a temperature above the boiling point of the liquid, the droplet will evaporate upon impact. Above a certain threshold plate temperature, a vapor film between the plate and the droplet prevents direct contact during impact due to the dynamic Leidenfrost effect. This state is unwanted in applications like spray cooling, as vapor limits the heat transfer from the solid to the liquid. We show that the dynamic Leidenfrost temperature for droplets that impact on a surface covered with carbon-nanofibers is higher than on the surface without these nanofibers. This is attributed to the cooling effect that vapor has on the superheated nanofibers. Because of the small scale of the carbon fibers, they are cooled by the vapor flow just before the liquid impact, resulting in a higher dynamic Leidenfrost temperature than on smooth surfaces. 8:39AM M15.00004 Role of cavitation in high-speed droplet impact problems , TOMOKI KONDO, KEITA ANDO, Keio University — High-speed droplet impact is found in physical cleaning using liquid jets, but its mechanisms for particle removal from target surfaces are yet unclear. In this study, we explore the possibility of having cavitation inside the droplet. The pressure evolution within a droplet colliding with a flat surface of deformable materials is determined by multicomponent Euler equations. Dynamics of cavitation bubbles heterogeneously nucleated from preexisting nuclei are determined from Rayleigh-Plesset calculations according to the pressure evolution within the droplet in one-way-coupling manner. The simulation shows that cavitation indeed occurs due to tension that arises from the water hammer shock reflection at the droplet interface. The role of cavitation including pressure emission from its collapse is to be discussed based on the one-way-coupling computations. 8:52AM M15.00005 Liquid droplet impact and penetration on 3D structured pore network , SAMAN HOSSEINI, NASSER ASHGRIZ, SANJEEV CHANDRA, University of Toronto — Impact of a water droplet on 3D structured permeable geometry is studied numerically as model for the impact and penetration of a droplet with a porous substrate. The 3D structured permeable geometry is characterized by its hole diameter, its stem length and hole spaces. Droplets with different diameters, various impact velocities and viscosity are impacted on the network of the holes and the infiltration of liquid through the hole is investigated. The outcome of such impact is characterized based on the conditions for liquid penetration volume in compare with initial droplet volume. Porosity of the substrate has been studied to evaluate the effect of this value in penetration regime. At the end, dynamic of bubble formation at range of Reynolds has been observed. 9:05AM M15.00006 Spray impact on a smooth, unheated surface: drop impact cavity diameter vs time1 , JOHN KUHLMAN, JONATHAN TAYLOR, NICHOLAS HILLEN, CHRISTOPHER SOMMERS, West Virginia University, Mechanical and Aerospace Engineering Dept., WVU SPRAY COOLING LABORATORY TEAM — A dense water spray impacting a smooth, unheated glass surface is studied. Average drop diameter is 90-60 microns; average axial velocity is 7.5-12 m/s for nozzle pressures of 1.4-4.2 bar gage, respectively, from PDPA data. Average spray Weber numbers are 70-130. Half the spray mass flux is due to larger drops with Weber numbers of 200-800, causing the splashed secondary drops and large impact cavities with longer lifetimes. Liquid film thickness is about 160 microns at all radii over the present range of nozzle pressures, with transient fluctuations of 30-40 microns. This thickness increases vs drop radial impact location, and decreases vs nozzle pressure. Drop impact cavity diameter from video images is 0.5 mm-1.5 mm, giving drop diameters of 100-300 microns, consistent with the PDPA data. Spray drop impact cavity growth vs time is fit approximately by (t)0.2 as seen in the literature. These results will be used to improve correlations in an existing preliminary Monte Carlo model of the complex spray impact process. It is believed that the transient thin liquid films formed beneath the droplet impact cavities are an important source of heat transfer augmentation via transient conduction. 1 Supported under NASA Cooperative Agreement NNX10AN0YA. 9:18AM M15.00007 Drop Impingement Induced Dispersal of Microorganisms and Contaminants Within Porous Media , YOUNG SOO JOUNG, ZHIFEI GE, CULLEN BUIE, Massachusetts Institute of Technology — We investigate migration of chemicals and microbes with aerosol generated by drop impingement on porous media. In our previous work we found that aerosol generation from droplets hitting porous media within a specific range of the Weber number (We) and a modified Pelect number (Pe). We and Pe reflect the impact condition of droplets and the wetting properties of porous media, respectively. The relationship between We and Pe can be expressed by a third dimensionless group, the Washburn Reynolds number (ReW =We/Pe). In a specific range of ReW , hundreds of aerosol particles can be generated within milliseconds of drop impingement. In this work we investigate if microbes such as Corynebacterium glutamicum, a soil bacterium, and chemicals such as Rhodamine B can be dispersed by aerosols generated from droplet impact. Experimentally, C. glutamicum and Rhodamine B are permeated into porous media. Then drop impingements are conducted on the porous media with different We and Pe in an airflow tunnel. We quantitatively investigate the volume and speed of aerosol migration as a function of ReW of the drop impingement and Re of the airflow. Results of this study will shed light upon the dispersal of elemental compounds and microbes within soils due to aerosol generated by rainfall. 9:31AM M15.00008 Experimental characterization and numerical simulation of crown propagation induced by impingement of droplet train , TAOLUE ZHANG, JORGE ALVARADO, Texas A&M Univ, ANOOP KANJIRAKAT, REZA SADR, Texas A&M University at Qatar, TAMU-TAMUQ TEAM — In this combined experimental and numerical study, hydrodynamics of single stream of HFE-7100 droplets striking a pre-wetted solid surface was investigated. ANSYS Fluent CFD software was employed to simulate this process numerically. Experimentally, single stream of mono-dispersed droplets were produced using a piezoelectric droplet generator with the ability to adjust parameters such as droplet impingement frequency, droplet diameter and droplet velocity. A high speed camera system was used to capture the liquid crown propagation process given the high frequency of droplet impingement. Low-Weber number droplet impingements resulted in smooth spreading of the liquid crown while splashing (i.e. the emergence of secondary droplets from the rim of the crown) was observed at high Weber number cases. The dynamics of the crown propagation was analyzed and a correlation that takes into account non-dimensional crown diameter (d∗ ) and non-dimensional time (t∗ ) has been postulated. The correlation has a mathematical form of d∗ = K · (t∗ )1/2 , where K is a constant. Comparison of the dynamics of crown propagation between experiments and numerical simulations yielded reasonable agreement. 9:44AM M15.00009 Drop Impact on to Moving Liquid Pools1 , BEATRIZ NATIVIDAD MUÑOZ-SÁNCHEZ, Xaar plc, United Kingdom, JOSÉ RAFAEL CASTREJÓN-PITA, University of Cambridge, ALFONSO ARTURO CASTREJÓN-PITA, University of Oxford, IAN M. HUTCHINGS, University of Cambridge — The deposition of droplets on to moving liquid substrates is an omnipresent situation both in nature and industry. A diverse spectrum of phenomena emerges from this simple process. In this work we present a parametric experimental study that discerns the dynamics of the impact in terms of the physical properties of the fluid and the relative velocity between the impacting drop and the moving liquid pool. The behaviour ranges from smooth coalescence (characterized by little mixing) to violent splashing (generation of multiple satellite droplets and interfacial vorticity). In addition, transitional regimes such as bouncing and surfing are also found. We classify the system dynamics and show a parametric diagram for the conditions of each regime. 1 This work was supported by the EPSRC (Grant EP/H018913/1), the Royal Society, Becas Santander Universidades and the International Relationships Office of the University of Extremadura. 9:57AM M15.00010 Impact of Small Raindrops on Crude Oil Slicks1 , DAVID MORRA, NOURAH ALMASHAN, DAVID MURPHY, JOSEPH KATZ, Johns Hopkins University Department of Mechanical Engineering — The impact of millimeter size water droplets falling near terminal velocity (e.g. rainfall) on a pool is known to produce air bubbles at the bottom of the splash cavity. These bubbles produce noise and contribute to marine aerosol production. Layers of crude oil resulting from oil spills alter air-sea interfacial properties. Our high speed observations examine the effect of oil layer thickness on the entrainment of air and oil as small raindrops impact the surface. They reveal that layers in the 10-400 µm range suppress bubble entrainment, likely due to the reduction of air-liquid surface tension (from 72 to 28 mN/m). For “low energy” impacts (droplets <2 mm and speed <2.5 m/s) and <200 µm layers, rupture of the film in less than 1 ms causes rapid retraction of the oil layer across the subsurface cavity and formation of oil droplets on the cavity side. Subsequently, as the cavity collapses, a vortex ring develops at the bottom of this cavity and forces these droplets downward. Impact on thicker oil layers results initially in accumulation of the drop fluid at the cavity base. When the drop subsequently penetrates the layer, it creates multiphase vesicles, i.e. drops of freshwater coated by a thin oil film, which migrate down into the bulk seawater. 1 Sponsored by Gulf of Mexico Research Initiative (GoMRI). Tuesday, November 25, 2014 8:00AM - 10:10AM Session M16 Convection and Buoyancy-Driven Flows: General II Wisconsin-Madison — 2000 - David Sondak, University of 8:00AM M16.00001 An application of the unifying theory of thermal convection in vertical natural convection , CHONG SHEN NG, ANDREW OOI, The University of Melbourne, DETLEF LOHSE, University of Twente, DANIEL CHUNG, The University of Melbourne — Using direct numerical simulations of vertical natural convection (VNC) at Rayleigh numbers 1.0 × 105 –1.0 × 109 and Prandtl number 0.709, we provide support for a generalised applicability of the Grossmann–Lohse (GL) theory, originally developed for horizontal natural (Rayleigh– Bénard) convection. In accordance with the theory, the boundary-layer thicknesses of the velocity and temperature fields in VNC obey laminar-like scaling, whereas away from the walls, the dissipation of the turbulent fluctuations obey the scaling for fully developed turbulence. In contrast to Rayleigh–Bénard convection, the direction of gravity in VNC is parallel to the mean flow. Thus, there no longer exists an exact relation linking the normalised global dissipations to the Nusselt, Rayleigh and Prandtl numbers. Nevertheless, we show that the unclosed term, namely the global-averaged buoyancy flux, also exhibits laminar and turbulent scaling, consistent with the GL theory. The findings suggest that, similar to Rayleigh–Bénard convection, a pure power-law relationship between the Nusselt, Rayleigh and Prandtl numbers is not the best description for VNC and existing empirical power-law relationships should be recalibrated to better reflect the underlying physics. 8:13AM M16.00002 Maximal transport in the Lorenz equations1 , CHARLES R. DOERING, ANDRE N. SOUZA, University of Michigan — We derive rigorous upper bounds on the transport hXY i where h·i indicates time average, for solutions of the Lorenz equations without assuming statistical stationarity. The bounds are saturated by nontrivial steady (albeit often unstable) states, and hence they are sharp. Moreover, using an optimal control formulation we prove that no other flow protocol of the same strength, i.e., no other function of time X(t) driving the Y (t) and Z(t) variables while satisfying the basic balance hX 2 i = hXY i, produces higher transport. 1 Supported by NSF Mathematical Physics award PHY-1205219 with an Alliances for Graduate Education and the Professoriate (AGEP) Graduate Research Supplement. 8:26AM M16.00003 Optimal transport in truncated models of Rayleigh-Bénard convection1 , ANDRE N. SOUZA, CHARLES R. DOERING, University of Michigan — We investigate absolute limits on heat transport in a truncated model of RayleighBénard convection. Two complementary analyses are used to derive upper bounds in an eight model: a background method analysis and an optimal control approach. In the optimal control formulation the flow no longer obeys an equation of motion, but is instead a control variable. The background method and the optimal control approach produce the same estimate. However, in contrast to a simpler system (i.e., the Lorenz equations) the optimizing flow field—which is observed to be time independent—does not correspond to an exact solution of the equations of motion. 1 Supported by NSF Mathematical Physics award PHY-1205219 with an Alliances for Graduate Education and the Professoriate (AGEP) Graduate Research Supplement. 8:39AM M16.00004 The oscillation modes of large-scale circulation in turbulent RayleighBénard convection , DANDAN JI, KUNLUN BAI, ERIC BROWN, Department of Mechanical Engineering and Materials Science, Yale University — We present measurement of the large-scale circulation (LSC) of turbulent Rayleigh-Bénard convection of cubic cell. We found the reorientation events by rotation through the LSC orientation θ0 with a multi-peaked probability distribution p(θ0 ), as predicted by the model presented by Brown and Ahlers (Phys. Fluids, 2008). In contrast to the results of oscillation modes in cylindrical cell, when the LSC was confined into one corner, the flow didn’t exhibit the twisting and sloshing oscillation with a well-defined periodicity. The phase relation of θ0 at different heights in the cell was not fixed, so LSC was not in a plane. The sloshing displacement of the LSC from a center plane exhibited random switching between two states. 8:52AM M16.00005 Path instability of a buoyancy-driven body: a sensitivity analysis to measure the fluid-object coupling , JOEL TCHOUFAG, Universite de Toulouse- IMFT, OLIVIER MARQUET, ONERA Meudon, DAVID FABRE, Universite de Toulouse- IMFT, JACQUES MAGNAUDET, CNRS - IMFT — The dynamical path of buoyancy-driven bodies in a viscous fluid is investigated in a linear stability framework. The departure of falling/rising objects from a straight vertical path can be understood by examining the unstable linear global modes of the fully coupled fluid-solid system linearized around the falling/rising steady state. Although this approach offers a quantitative prediction of the various possible trajectories, it raises new questions about the physical interpretation of fully coupled fluid/solid modes. Are the observed trajectories driven by the fluid dynamics, the solid dynamics, or by their coupling? In which flow regions are those dynamics most active? To answer these questions, we present a straightforward adjoint-based-method that can be used to measure the coupling in any problem where reciprocal interactions between two sub-parts of a system take place. This method is exemplified on the case of a two-dimensional falling ellipse. In the particular case of large body-to-fluid inertia ratios, a clear distinction between body-related and wake-induced modes is observed, in line with results predicted by a quasi-static approach. 9:05AM M16.00006 Bénard-Marangoni instability driven by moisture absorption , SANGWOO SHIN, IAN JACOBI, JASON WEXLER, HOWARD STONE, Department of Mechanical and Aerospace Engineering, Princeton University — We describe experiments that exhibit Bénard-Marangoni convection cells in hygroscopic fluids without external heating. Bénard-Marangoni convection cells are classically driven by a heat source beneath a thin layer of fluid with a free-surface. External heating provides a reservoir of hot fluid to amplify the free-surface temperature perturbations which drive Marangoni flow; without the heat source, the system naturally damps the temperature fluctuations and stabilizes itself. By drawing water vapor from ambient air, certain hygroscopic fluids can generate their own internal heat source by exploiting an exothermic enthalpy of solution with water. We verify the origin of the instability by using different hygroscopic fluids. The dynamics of this unusual instability are measured as a function of the fluid and air properties of the system, and a mathematical model is developed to rationalize the results quantitatively. 9:18AM M16.00007 Oscillatory magnetoconvective instability in a ferrofluid layer placed in an oblique external magnetic field , SERGEY A. SUSLOV, HABIBUR RAHMAN, Swinburne University of Technology, Australia, ALEKSANDRA A. BOZHKO, Perm State National Research University, Russia — Magnetite-based ferrofluids are manufactured magneto-polarisable nanofluids that magnetize in an external magnetic field in a similar way to natural paramagnetic fluids(e.g. oxygen), however to a much higher degree. Paramagnetic and ferrofluid flows are described by similar equations and it is expected that they would exhibit a similar behaviour. Indeed we show that in both type of fluids the most prominent instability structures align with the in-layer field component and the onset of magnetoconvection is delayed by the field inclination. However we find that in contrast to paramagnetic fluids the instabilities arising in differentially heated ferrofluids placed in a uniform external oblique magnetic field are oscillatory. This is traced back to the nonlinearity of the magnetic field distribution induced inside the ferrofluid layer that arises whenever the direction of the applied magnetic field is not normal. Given that the magnetic field inclination with respect to the plane of the layer is inevitable near its edges the obtained stability results shed light on the possible reasons for the existnce of unsteady patterns that have been detected in the normal field experiments we reported previously. 9:31AM M16.00008 Stratified shear flow in an inclined square duct1 , COLIN MEYER, Harvard Univ, PAUL LINDEN, Cambridge Univ — We present results of experiments on stratified shear flow in an inclined duct. The duct connects two reservoirs of fluid with different densities, which drives a counterflow with a dense layer flowing beneath a less-dense layer moving in the opposite direction. Depending on the dimensionless Atwood number A and duct angle θ, we identify four flow states: a laminar L state, a Holmboe wavemode H state, a spatio-temporally intermittent I state, and a fully developed turbulent T state. We map a state diagram of these flows in the Atwood number – θ plane and examine the force balances that determine each of these states. We find the L and H states to be hydraulically controlled at the ends of the duct and the flow is determined by the pressure difference associated with the density difference between the reservoirs. The I and T states are associated with increasing dissipation within the duct. We replot the state-space in the Grashof number – θ phase plane and find the transition to the T-state is governed by a critical Grashof number. We then evaluate the level of turbulence by examining scalings for the thickness of interfacial region between the two layers. 1 NSF GRF No. DGE1144152 9:44AM M16.00009 Stratified shear flow in an inclined duct: equations and scalings , SIMON VINCENT, PIERRE AUGIER, Univ of Cambridge, COLIN MEYER, Harvard University, PAUL LINDEN, Univ of Cambridge — We present a theoretical approach to model the behaviour of a stratified shear flow in an inclined duct, and relate the scalings emerging from these equations to the experimental work realized on this problem. We consider a system composed of two reservoirs, filled with fluids of different densities, connected by a square duct inclined from the horizontal. We observe from the experiments that a counterflow is established inside the duct with the denser fluid flowing beneath the less dense fluid, exhibiting a wide range of different regimes as the density difference and the inclination angle are increased. Our model shows that the velocities of the flow scale differently depending on the type of regime the system is in. We compare those scalings to the experimental data and show that the transition from the laminar regimes to the more turbulent ones can be described by different non dimensional numbers depending on the inclination angle and the Reynolds number. 9:57AM M16.00010 Stratified shear flow in an inclined duct: measurements of velocity and scalar fields1 , PAUL LINDEN, SIMON VINCENT, STUART DALZIEL, Cambridge University, GKB LAB TEAM — The effect of stable stratification on turbulent shear flow is a fundamental problem in turbulence. We present quantitative experimental results on the flow and density fields in a duct, inclined slightly from the horizontal, connecting two reservoirs containing fluids of different densities and. A counterflow is established in the duct with the denser fluid flowing beneath the less dense fluid. This flow exhibits a range of different flow regimes, from wavelike to intermittent to turbulent, depending on the angle of inclination of the duct, and the relative density difference between the two reservoir fluids. We use two-dimensional PIV and PLIF to measure and compare the velocity and density fields for each of the different regimes. We examine the mean signals to determine governing features such as the average gradient Richardson numbers for each regime. We also determine the characteristic features of the fluctuating fields in the different flow regimes and relate these to the structures observed in visualisations of the flow. 1 This research is supported by EPSRC Programme Grant EP/K034529/1 Tuesday, November 25, 2014 8:00AM - 10:10AM Session M17 Geophysical Fluid Dynamics: Rotating Flows — 2002 - Philip Marcus, University of California, Berkeley 8:00AM M17.00001 On the Surprising Longevity of Jupiter’s Centuries-Old Great Red Spot , PHILIP MARCUS, UC Berkeley, PEDRAM HASSANZADEH, Harvard — Jupiter’s Great Red Spot (GRS) has been observed continuously for 100 years and is possibly older than 350 years. However, the area of its cloud cover is quickly shrinking. Although the areas of the clouds and of the potential vorticity of the GRS might not be well correlated, it motivates us to examine the physics that determines the GRS lifetime. When the GRS is in quasi-equilibrium, the ratio of its potential (i.e. thermal) energy to its kinetic energy is ∼ 2/Ro ≃ 6, where Ro is the Rossby number. Because the atmospheric radiative decay time is 4-5 years, the overall energy and structure of the GRS would be expected to decay in 4-5 years, as it does in our 2D simulations of the GRS (or with an faster decay rate in the low-resolution 3D simulations by others). We show that in high-resolution, 3D calculations, meridional circulations (consisting of vertical and radial velocities) develop spontaneously in the GRS. The vertical velocity sustains the GRS by drawing energy from the ambient atmosphere: this circulation transports mass downward in the ambient atmosphere, thereby decreasing its potential energy. This released energy, along with kinetic energy from the ambient zonal jets, is carried to the GRS by the meridional circulation, sustaining the GRS for centuries. 8:13AM M17.00002 Linear Stability and Nonlinear Evolution of 3D Vortices in Rotating Stratified Flows , MANI MAHDINIA, UC Berkeley, PEDRAM HASSANZADEH, Harvard, PHILIP MARCUS, UC Berkeley — Axisymmetric Gaussian vortices are widely-used to model oceanic vortices. We study their stability in rotating, stratified flows by using the full Boussinesq equations. We created a stability map as a function of the Burger and Rossby numbers of the vortices. We computed the linear growth rates of the most-unstable eigenmodes and their corresponding eigenmodes. Our map shows a significant cyclone/anti-cyclone asymmetry. The vortices are linearly unstable in most of the parameter space that we studied. However, the anticyclonic vortices, over most of the parameter space, have eigenmodes with only very weak growth rates – longer than 50 vortex turn-around times. For oceanic vortices, that time corresponds to several months, so we argue that this slow growth rate means that the oceanic anticyclones lifetimes are not determined by linear stability, but by other processes. We also use our full, nonlinear simulations to show an example of an unstable cyclone with a very fast growing linear eigenmodes. However, we show that cyclone quickly redistributes its vorticity and becomes a stable tripole with a large core that is nearly axisymmetric. 8:26AM M17.00003 Dynamics of SQG Vortices , CECILY KEPPEL, STEFAN LLEWELLYN SMITH, UCSD — The surface quasi-geostrophic (SQG) equations are a model for low-Rossby number geophysical flows in which the dynamics are governed by potential temperature dynamics on the boundary. The model can be used to explore the transition from two-dimensional to three-dimensional mesoscale geophysical flows. We examine the dynamics of SQG vortices and the resulting flow in the entire fluid including at first order in Rossby number (O(Ro)). This requires solving an extension to the usual QG equation to compute the velocity corrections, and we demonstrate this mathematical procedure. As we show, it is simple to obtain the vertical velocity, but difficult to find the O(Ro) horizontal corrections. We then consider the specific case of an exact SQG vortex solution developed by Dritschel (2011). We examine the interaction of two such vortices in both the infinite and doubly periodic domain. 8:39AM M17.00004 The baroclinic instability of an initially stratified fluid layer1 , PATRICE LE GAL, IRPHE - Aix Marseille University - CNRS, MIKLOS VINCZE, UWE HARLANDER, Department of Aerodynamics and Fluid Mechanics Brandenburg University of Technology, Cottbus — Our project aims to describe the baroclinic instability that destabilizes an initially stratified layer of fluid. Classically, this instability is studied using pure fluid. Here, the originality of the project comes from the use of a layer of water initially stratified with salt. Before rotation is started, double convection sets in within the stratified layer with a strongly non-homogeneous pattern consisting of a double diffusive staircase at the bottom of the container in the very dense water layers and a shallow convective cell in the top surface layer. As radial motions take place due to the presence of these convective cells, the action of the Coriolis force generates strong zonal flows as soon as rotation is started. Thus, above a rotation rate threshold, the baroclinic instability destabilizes the flow in a shallow layer, generating a ring of pancake vortices. Infrared camera images measure the temperature distributions at the water surface and PIV velocity maps describe the wavy flow pattern and the pancake vortices. Note finally that if we prepare a stratification profile with an inner shallow non-stratified zone, it is possible to confine the baroclinic instability within this confined zone immersed inside the stratified fluid. 1 PLG acknowledges the financial support of EuHIT 8:52AM M17.00005 An experimental study of the spread of buoyant water into a rotating environment , THOMAS CRAWFORD, PAUL LINDEN, University of Cambridge — We present an experimental study that aims to investigate the spread of buoyant water, released from a finite potential vorticity source, into a rotating environment. The source structure is designed to simulate the discharge of a river into the ocean and as a result the freshwater enters the salt water ambient horizontally and with considerable momentum flux. The finite depth of the source gives rise to a non-zero potential vorticity as seen in the natural environment. We perform a parametric study in which we vary the rotation rate, freshwater volume flux and density difference between the incoming buoyant fluid and the stationary ambient. The parameter values are chosen to match the regimes seen in the River Rhine and River Elbe when entering the North Sea. Persistent features of an anticyclonic outflow vortex and a propagating boundary current can be identified in each experimental run and their properties are quantified. The flow is seen to become unstable for small values of the deformation radius, suggesting it has an important role to play in determining the behaviour of the flow. We also present a finite potential vorticity, geostrophic model that provides theoretical predictions for the current height, width and velocity. These are compared with the experimental data. 9:05AM M17.00006 Experimental observation of steady inertial wave turbulence in deep rotating flows1 , EHUD YAROM, ERAN SHARON, Hebrew Univ of Jerusalem — The theoretical framework that should be used for describing rotating turbulence is the subject of an active debate. It was shown experimentally and numerically that the formalism of 2D turbulence is useful in the description of many aspects of rotating turbulence. On the other hand, theoretical and numerical work suggests that the formalism of wave turbulence should provide a reliable description of the entire 3D flow field. The waves that are suggested as the basis for this turbulence are Coriolis-force-driven inertial waves. Here we present experimental results that suggest the existence of inertial wave turbulence in deep steady rotating turbulence. Our measurements show energy transfer from the injection scale to larger scales, although the energy spectra are concentrated along the dispersion relation of inertial waves. The turbulent fields are, therefore, well described as ensembles of 3D interacting inertial waves. 1 This work was supported by the Israel Science Foundation, Grant No. 81/12. 9:18AM M17.00007 Inertial wave excitation in a rotating annulus with partially librating boundaries1 , ION DAN BORCIA, UWE HARLANDER, CHRISTOPH EGBERS, ABOUZAR GHASEMI V., MARTEN KLEIN, EBERHARD SCHALLER, TORSTEN SEELIG, ANDREAS WILL, Brandenburg University of Technology Cottbus-Senftenberg, MICHAEL V. KURGANSKY, Russian Academy of Sciences — Inertial waves are excited in a fluid filled rotating annulus by modulating the rotation rate of parts of the vessel boundary. This forcing leads to inertial wave beams emitted from the corner regions of the annulus due to periodic motions in the boundary layers. Firstly we use a meridional symmetrical geometry. When the forcing frequency matches with the eigenfrequency of the rotating annulus the beam pattern amplitude is increasing, the beams broaden and mode structures can be observed. The eigenmodes are compared with analytical solutions of the corresponding inviscid problem. In particular for the pressure field a good agreement can be found. However, shear layers related to the excited wave beams are present for all frequencies. Then, the meridional symmetry is broken by replacing the inner cylinder with a truncated cone (frustum). The geometry is non-separable and exhibits wave focusing and wave attractors. Under the assumption that the inertial waves do not essentially affect the boundary-layer structure, we use classical boundary-layer analysis to study oscillating Ekman layers over a librating wall that is at a non-zero angle to the axis of rotation. 1 This work is part of the project “Mischung und Grundstromanregung durch propagierende Trägheitswellen: Theorie, Experiment und Simulation,” financed by the German Science Foundation (DFG). 9:31AM M17.00008 Nonlinear flows driven by libration in a rotating half cone , MICHAEL PATTERSON, ROSEN RACHEV, Department of Mechanical Engineering, University of Bristol, BS8 1TR, UK, LIGANG LI, KEKE ZHANG, Department of Mathematical Sciences, University of Exeter, Exeter EX4 4QF, UK — We investigate the problem of nonlinear oscillatory flow of homogeneous fluid with viscosity ν confined in a half cone that rotates rapidly about a fixed axis with angular velocity Ω0 and that undergoes weak longitudinal libration with amplitude ǫΩ0 and frequency ω̂Ω0 , where ǫ is the Poincaré number and ω̂ is dimensionless frequency with 0 < ω̂ < 2. Two different methods are employed in this investigation: experimental studies and direct numerical simulation using a finite element method. 9:44AM M17.00009 What is the energy dissipation rate in rotating turbulence? , FREDERIC MOISY, ANTOINE CAMPAGNE, PIERRE-PHILIPPE CORTET, Laboratoire FAST, CNRS, Universite Paris-Sud, BASILE GALLET, Laboratoire SPHYNX, Service de Physique de l’Etat Condensee, DSM, CEA Saclay, CNRS — The scaling of the energy dissipation rate ǫ is one of the most fundamental open issues for rapidly rotating turbulence. For non-rotating 3D turbulence at large Reynolds number, it takes the classical form ǫ3D ≃ U 3 /L, with U and L the characteristic velocity and length scales. Here, we propose a simple experiment aiming to probe directly the influence of the background rotation on ǫ: we measure the torque Γ acting on a propeller rotating at constant rate ω in a large volume of fluid rotating at Ω (the torque measurement being performed in the rotating frame). The normalized torque Kp = Γ/(ρR4 Hω 2 ) (where R and H are the propeller radius and height) provides a direct measure of the normalized dissipation ǫ/ǫ3D as a function of the Rossby number Ro = ω/Ω. For cyclonic propeller rotation (Ro > 0) we find a transition between Kp =constant at large Ro (no rotation) and Kp ≃ Ro at small Ro (large rotation), in agreement with weakly nonlinear rotating turbulence prediction. The situation is more intricate for anticyclonic rotation (Ro < 0), showing a peak dissipation at intermediate Ro, and a decrease at small Ro but with a different scaling. 9:57AM M17.00010 Comparison of Two-Dimensional Turbulence on the Surface of a Sphere with Two-Dimensional Turbulence on a Plane , LEILA AZADANI, ANNE STAPLES, Virginia Tech — Although there are not two-dimensional turbulent flows in nature, there are many applications in which the fluid motion can be described by two-dimensional models. For example largescale geophysical flows in the atmosphere and ocean can be accurately represented by two-dimensional turbulence models. The combined effects of geometry, stratification and rotation restricts the flow motion in the vertical direction and makes these flows almost two-dimensional. While Cartesian coordinates are usually used to perform these computations, spherical coordinates are more natural and account for the Earth’s curvature. Computations of two-dimensional turbulent flows in Cartesian and spherical geometries yield different results. The energy transfer mechanism, the rate of enstrophy transfer to higher wave numbers and the behavior of coherent structures in spherical coordinates are different from those in Cartesian coordinates. Here, we compare two-dimensional turbulence on a plane with two-dimensional turbulence on the surface of a sphere is spectral space and explain the differences in Cartesian and spherical geometries. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M18 Vortex Dynamics: Applications — 2004 - Kamran Moseni, University of Florida 8:00AM M18.00001 Vortex shedding from cylinders with circular fins , JEFF MCCLURE, SERHIY YARUSEVYCH, University of Waterloo — The flow development around uniform cylinders with circular fins is investigated experimentally. Finned cylinders with a diameter ratio (D/d) of two, a fin thickness ratio (t/d) of 0.027, and a range of fin pitch ratios 0.083 ≤ c/d ≤ 1.0 are studied at a Reynolds number (Re D ) of 3150, which pertains to the separated shear layer transition regime. All experiments are performed in a water flume facility using time-resolved, two-component, planar Particle Image Velocimetry measurements in spanwise and transverse planes, as well as Laser Doppler Velocimetry. The independent PIV measurements in two different planes capture spatio-temporal development of the main vortical structures in the cylinder wake. A comparative analysis of the results obtained for uniform and finned cylinder models is performed to investigate the effect of circular fins and their spacing on turbulent wake development. The experimental data is used to characterize the near-wake development, vortex formation, and the evolution of coherent structures. The results show that vortex shedding characteristics exhibit strong dependence on the fin pitch ratio and, for a given pitch ratio, differ significantly from those observed for a uniform cylinder of equivalent diameter. 8:13AM M18.00002 Dynamics of Isolated Tip Vortex Cavitation1 , PEPIJN PENNINGS, Delft University of Technology, JOHAN BOSSCHERS, Maritime Research Institute Netherlands (MARIN), TOM VAN TERWISGA, Delft University of Technology — Performance of ship propellers and comfort levels in the surroundings are limited by various forms of cavitation. Amongst these forms tip vortex cavitation is one of the first appearing forms and is expected to be mainly responsible for the emission of broadband pressure fluctuations typically occurring between the 4th to the 7th blade passing frequency (approx. 40-70 Hz). These radiated pressure pulses are likely to excite parts of the hull structure resulting in a design compromise between efficiency and comfort. Insight is needed in the mechanism of acoustic emission from the oscillations by a tip vortex cavity. In the current experimental study the tip vortex cavity from a blade with an elliptic planform and sections based on N ACA 662 − 415 with meanline a = 0.8 is observed using high speed shadowgraphy in combination with blade force and acoustic measurements. An analytic model describing three main cavity deformation modes is verified and used to explain the origin of a cavity eigenfrequency or “vortex singing” phenomenon observed by Maines and Arndt (1997) on the tip vortex cavity originating from the same blade. As no hydrodynamic sound originating from the tip vortex cavity was observed it is posed that a tip flow instability is essential for “vortex singing.” 1 This research was funded by the Lloyd’s Register Foundation as part of the International Institute for Cavitation Research 8:26AM M18.00003 Circulation based analysis of an axisymmetric, deformable, jetting-cavity body1 , MICHAEL KRIEG, KAMRAN MOHSENI, University of Florida — Here a methodology for calculating pressure distribution internal to a generic, deformable, axisymmetric, body with an internal cavity region is presented. The pressure distribution is derived by integrating the momentum equation along the axis of symmetry, and then along the cavity boundaries where the velocities are prescribed. Unknown velocity integrals are extracted from the total circulation of characteristic regions, and basic modeling is provided to relate the circulation in these regions to deformation parameters. From the pressure distribution, the total instantaneous jetting force is calculated along with total work required to drive the fluid motion. A prototype jet actuator was designed and tested to determine the circulation in and around the device as well as the actual forces. The total instantaneous forces acting on the actuator are observed to be well modeled by the pressure analysis during both expulsion and refilling phases of the jetting cycle. The functional dependence of total forcing on both jet velocity and acceleration is presented. It was observed that for all phases of the jetting cycle total required work is lower for impulsive velocity programs with fast accelerations than sinusoidal velocity programs with smoother gradual accelerations. Sinusoidal programs result in a peak in pressure (force) at the same instant when the manipulator driving fluid motion is at its maximum velocity; for the impulsive programs these peaks are out of phase and overall energy consumption is reduced. 1 This was supported by the ONR 8:39AM M18.00004 Wake dynamics behind a harbor seal vibrissa: a comparative view by PIV measurements , YINGZHENG LIU, SHAOFEI WANG, HANPING CHEN, Shanghai Jiao Tong University — A comprehensive study was performed of wake dynamics behind a scaled-up model of harbor seal vibrissa, and the baseline configurations of circular cylinder, wavy cylinder and the elliptical cylinder were provided for comparison. A low-speed water channel and wind tunnel were employed for the model tests at the Reynolds number 102 ∼ 104 based on diameter of the cylinder. A load cell and Particle Image Velocimetry were synchronized to measure the fluctuating lift/drag forces and the instantaneous flow field, respectively. By means of the comparative study, the unique three-dimensional wake characteristics in response to contour variations of the harbor seal vibrissa was elucidated through the Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) analyses of the measured flow field, demonstrating the ability of the vibrissa to suppress the vortex-induced vibration. 8:52AM M18.00005 Harbor seal whiskers synchronize with upstream wake over a range of distances , HEATHER BEEM, MICHAEL TRIANTAFYLLOU, MIT — Harbor seal whiskers have been shown to exhibit unique vibration properties as they encounter vortex wakes [1]. Seals may use this information to detect hydrodynamic trails left by fish prey. A scaled model, which captures the undulatory morphology of the harbor seal whisker and is designed to freely vibrate, is tested here to explore these properties in more detail. This model is towed downstream of a larger cylinder, which generates a vortex wake. Effects of downstream distance, lateral distance, and diameter ratio between the two objects are explored. Frequency measurements are collected simultaneously through use of a pressure sensor placed in the wake. Cross-correlation of the whisker motion and cylinder wake pressure provides evidence that frequency synchronization holds for a range of separation distances and wake generator sizes. [1] Beem, H., Triantafyllou, M. (2013). “Harbor seal whiskers synchronize with frequency of upstream wake,” Bulletin of the American Physical Society 58. 9:05AM M18.00006 Rotor Vortex Wake in Close Proximity of Walls in Hover , MEHMET FATIH KONUS, OMER SAVAS, University of California at Berkeley — Expanding flight envelopes of rotorcraft raise concerns about their behavior in very close proximity of walls or corners where the separation between the wall and the rotor disk can almost vanish. A series of experiments are conducted in a water tank to study the hover behavior of the wake of a 25-cm diameter three-bladed rotor at 8 rev/s. Particle image velocimetry, strain gage force balance measurements and flow visualization are employed. The vortex wake, which is axisymmetric on the average in an unbounded surrounding, is distorted increasingly with decreasing separation from a wall or corner. The vortex wake bends toward the wall and into the corner. The individual helical filaments off the rotor tips are distorted and closely follow the wall. Intermittent reversed vortical flow regions appear upstream of the rotor disk in the proximity of the wall. The mean streamlines indicate that the wake is bent toward the wall or into the corner. The component of the thrust vector along the axis of the rotor decreases. These observation suggest that the thrust vector progressively deviates from the geometric axis of the rotor. 9:18AM M18.00007 High Re wall-modeled LES of aircraft wake vortices in ground effect , OLIVIER THIRY1 , GREGOIRE WINCKELMANS, MATTHIEU DUPONCHEEL, Universite catholique de Louvain (UCL) - Institute of Mechanics, Materials and Civil Engineering (iMMC) — We have been able to perform wall-resolved LES, using a fourth order code, to simulate (aircraft) wake vortices interacting with the ground, also with cross or head winds, up to Reynolds numbers of the order of Re = Γ/ν = 2 × 104 . The present work aims at providing higher Re simulations, and also simulations with rough walls (e.g., grass), through the use of LES with near wall modeling. Various types of models are compared: point-wise and averaged algebraic models, and two-layers models. When using averaged models, the averaging methodology is of importance, since there is essentially no homogeneous direction in the case of wake vortices in ground effects. Uni- and multi-directional averaging strategies, with and without additional time averaging will be considered. When two-layer models are used, a RANS sub-layer will be compared to a simpler approach based on simplified turbulent boundary layer equations. The approaches are first validated on simpler flows, channel flow or wake flow, for which reference wall-resolved LES or DNS results are available. 1 Research fellow (Ph.D. student) at the F.R.S. - FNRS (Belgium) 9:31AM M18.00008 Effect of Vorticity Amplification on Flow Separation from Landing Gear Wheels , GRAHAM FELTHAM, PHILIP MCCARTHY, ALIS EKMEKCI, University of Toronto — The flow near the stagnation point of landing gear wheels has been previously shown to support a mechanism for inbound streams of weak vorticity to collect, growth, and amplify into large-scale discrete vortex structures. The current experimental study is an extension to investigate the effects of these vortex structures on the separation characteristics of the flow around the outboard sides of the wheels. Experiments were performed in a water channel with qualitative understanding of the flow topology achieved by employing the hydrogen bubble visualization technique and quantitative measurements performed using Particle Image Velocimetry (PIV). The upstream vorticity source is a platinum wire (d = 100 µm) placed 30 mm upstream of the model wheels. The Reynolds number based on wire diameter is 21 and based on wheel diameter (D = 152 mm) is 32,500. The inbound pair of vorticity streams impinged at the wheel surface where maximum vortex growth and amplification occurs as identified by previous experiments. The growth and shedding of the resulting vortical structures is shown to alter the shape and size of the separation bubbles on the outboard sides of the wheels. A vortex identification and tracking method is applied to map the growth and movement of the observed structures. 9:44AM M18.00009 Vorticity amplification near the stagnation point of landing gear wheels: effect of the orientation of the impinging vorticity , MINGYAO GU, GRAHAM FELTHAM, ALIS EKMEKCI, University of Toronto Institute for Aerospace Studies — When oncoming streams of weak vorticity aligned with the axle axis of a two-wheel landing gear impinge near the forward stagnation point of the wheels, a mechanism for vorticity collection, growth, amplification into discrete large-scale vortices, and shedding was formerly shown to exist. In the current study, the impinging vorticity streams are perpendicular to the axle axis, i.e. in a vertical orientation as opposed to the horizontal orientation before. Experiments are conducted in a recirculating water channel using hydrogen bubble visualization and particle image velocimetry at a Reynolds number of 32,500 (based on the wheel diameter). As with the horizontal orientation, vorticity collection and amplification are observed, but the large-scale vortices thus formed are stretched around the wheel circumference in contrast to being stretched around the wheel sides, as observed for the horizontal orientation. This flow behavior varies with the impingement location of the vorticity streams across the wheel width. Maximum vorticity amplification occurs at a critical impingement location and drastically alters the flow separation along the wheel circumference. In addition, the instantaneous vortical structures are identified and tracked using a Galilean-invariant criterion. 9:57AM M18.00010 Development and validation of a 2-D compressible vortex particle-mesh method , PHILIPPE PARMENTIER1 , GREGOIRE WINCKELMANS, PHILIPPE CHATELAIN, Universite catholique de Louvain (UCL) - Institute of Mechanics, Materials and Civil Engineering (iMMC) — A compressible hybrid Vortex Particle-Mesh (VPM) method is being developed to study unsteady and fully compressible flows, either confined or unconfined. The particles are advected by the local velocity field and carry the vorticity, dilatation, density and enthalpy fields. They also change volume so as to conserve mass. The velocity is expressed into solenoidal and irrotational components using the Helmholtz decomposition. In the present approach, a Fourier-based method is used to efficiently solve the corresponding Poisson problems; it can handle bounded and unbounded problems. The underlying grid is also used to perform the spatial differential operations (except the lagrangian advection) as well as the redistribution of particles and the particle-mesh operations. The no-slip condition is enforced at solid walls while a nonreflecting boundary condition is used at the far field boundaries. The methodology is validated on prototypical unbounded vortical flows and on a driven cavity flow. 1 Supported by the Fund for Research Training in Industry and Agriculture (FRIA) Tuesday, November 25, 2014 8:00AM - 10:10AM — Session M19 Convection and Buoyancy-Driven Flows: Plumes and Gravity Waves/Currents 2006 - Stefan Llewellyn Smith, University of California, San Diego 8:00AM M19.00001 Stressed Horizontal Convection , KATARZYNA MATUSIK, STEFAN LLEWELLYN SMITH, UCSD Mechanical & Aerospace Engineering Dept. — We present experiments aimed at elucidating the interaction between wind-induced surface shear and the meridional overturning circulation. The effect of a shear stress on convection driven by a maintained dense source entering a homogeneous environment is explored. A saline plume enters a confined fresh-water environment at a boundary along with which a constant shear stress is being applied simultaneously. The system is driven to a statistically steady state, and the resulting density and velocity fields are obtained by Synthetic Schlieren and PIV techniques, respectively. The magnitude and direction of the shear stress is varied between experiments, as well as the density of the plume. Results indicate that there exists a competitive regime between the buoyancy and mechanical forcing, resulting in marked variations in flow features such as the interior stratification and boundary layer thickness, among others. 8:13AM M19.00002 Instabilities of plumes driven by localized heating in a stably stratified ambient1 , FRANCISCO MARQUES, Univ Politecnica de Catalunya, JUAN LOPEZ, Arizona State Univ — Plumes due to localized buoyancy sources are of wide interest due to their prevalence in many geophysical situations. This study investigates the transition from laminar to turbulent dynamics. Several experiments have reported that this transition is sensitive to external perturbations. As such, a well-controlled set-up has been chosen for our numerical study, consisting of a localized heat source at the bottom of an enclosed cylinder whose sidewall is maintained at a fixed temperature which varies linearly up the wall, and there is a localized heat source on the bottom. Restrincting the dynamics to the axisymmetric subspace, the first instability is to a puffing state. However, for smaller Grashof numbers, the plume becomes unstable to 3D perturbations and a swirling plume spontaneously appear. Further bifurcations observed in the rotating frame where the plume is stationary also exibits puffing, suggesting a connection between the unstable axisymmetric solution and the swirling plume. Further bifurcations result in quasiperiodic states with a very low frequency modulation, that eventually become turbulent. 1 Spanish Ministry of Education and Science grant (with FEDER funds) FIS2013- 40880 and U.S. National Science Foundation grant CBET-1336410 8:26AM M19.00003 Plume emission statistics in turbulent Rayleigh-Bénard convection , ERWIN VAN DER POEL, University of Twente, ROBERTO VERZICCO, University of Rome Tor Vergata, SIEGFRIED GROSSMANN, Philipps-Universität Marburg, DETLEF LOHSE, University of Twente — Rayleigh-Benard convection features ubiquitous coherent structures, which continue to survive in strong turbulence. The most prevalent are the thermal plumes and the large scale circulation (LSC). The thermal plumes and the LSC are intrinsically coupled, as thermal plumes cluster to form a LSC. We report statistics of the area, width and location of plumes extracted from high Rayleigh number (Ra ≤ 1012 ) direct numerical simulations in a cylindrical domain of aspect-ratio 0.33. While the area of the plume is unimodally distributed close to the plates, far from the plates plume clustering results in a bimodal distribution. In addition, the analysis reveals that more plumes are emitted from areas with low shear as compared to areas with high shear. 8:39AM M19.00004 Impingement of a plume on a non-horizontal rigid boundary , ALAN JAMIESON, STUART DALZIEL, University Of Cambridge — Buoyancy driven flows created by density differences, plumes are a phenomenon observed in many situations in both nature and industry. Instances of plume impingement on a rigid boundary are also common. Whether this smoke from a candle impacting on a ceiling or, for a much larger scale example, plumes in the ocean descending onto the continental shelf, such as in dense water formation in the Weddell Sea. In both these cases, and many others, the boundary is rarely a horizontal plane and so motivates the study for a plume impacting on a non-horizontal geometry. After reviewing previous work of a plume impinging on a horizontal, we introduce the problem of a plume impinging on an incline by presenting experiments varying the angle of inclination and the distance between the boundary and plume source. In an attempt to understand dynamics of large scale plumes in ocean, we also present the same experiment in a rotating system. 8:52AM M19.00005 Bent-over plume models for large-area highly-buoyant turbulent plumes , NIGEL KAYE, ALI TOHIDI, Clemson University — The problem of large-area turbulent plumes driven laterally by wind has numerous applications in environmental fluid mechanics. For example, one of the primary mechanisms of wildfire spread is through the creation of spot fires that result from embers being lofted into the atmosphere by a fire plume and then transported ahead of the fire by wind. We review existing entrainment models for bent over plumes and investigate the modeling approach most appropriate for large-area highly-buoyant plumes. We present analytic solutions for the far-field behavior of a bent-over plume in the presence of both a uniform and power-law velocity profile. The plume trajectory in a power-law velocity profile is flatter and the volume and momentum fluxes are larger compared to a plume in a uniform velocity field. Comparison with experimental measurements shows that modeling the boundary layer velocity profile is important to accurate prediction of plume trajectory. The results of a sensitivity analysis show that the choice of entrainment model has little influence on plumes with flatter trajectories but has a large effect on more vertical trajectories that are typical of large-area highly-buoyant plumes under low wind conditions. Further, the choice of boundary layer velocity profile function influences the trajectory of more vertical plumes. However, model predictions are insensitive to any eccentricity in the plume cross-section. 9:05AM M19.00006 Local stability of axisymmetric plumes1 , CHAKRAVARTHY R.V.K., LUTZ LESSHAFFT, PATRICK HUERRE, Ladhyx, CNRS, Ecole polytechnique, 91128 Palaiseau Cedex, France — A linear stability analysis of a forced plume with non-zero momentum at the inlet is performed for P r = 1, Re = 100 and Ri near 1. The steady base flow is obtained as a laminar solution of the steady Navier Stokes equations. The base flow asymptotes to a self-similar solution as it evolves downstream. In the non-self-similar regime close to the inlet, both axisymmetric mode (m = 0) and the helical mode (m = 1) are convectively unstable at sufficiently low Richardson number. In the self-similar regime, only the helical mode is absolutely unstable and the axisymmetric mode is stable. Higher helical modes (m ≥ 2) are seen to be convectively unstable very close to the inlet and become stable as the flow evolves downstream. The transition from convective to absolute instability makes the flow a good candidate for observing steep nonlinear global modes associated with buoyancy. 1 This work is supported by a PhD scholarship from Ecole polytechnique. 9:18AM M19.00007 Experimental study of lock-exchange gravity currents: Coupling between particle distributions and flow structures1 , ZHUANG SU, Peking Univ, MING PENG, Guangdong Yudean Group Co Ltd, HUIJING YUAN, CUNBIAO LEE2 , Peking Univ — This work presents detailed experimental investigations of the interactions between the particles and flows of the lock-exchange particle-laden gravity currents. A phase Doppler particle analyzer provided non-intrusive and synchronous measurements of the velocities and grain sizes of the particles. High-speed particle image velocimetry was used to measure the flow fields. The measurements showed that the particle behavior and the currents were intricately coupled. Particle distributions at different parts of the current are given, showing that the particles’ behaviors are highly related to the flow fields. The influences of the grain size to the flow fields are also investigated by comparing flow fields of currents carrying different particles to each other, as well as the un-laden currents. The presence of particles seems to postpone the evolving of the flow structures, it weakens the vorticity of the shear layer in the head but strengthens the voriticty in the body or tail of the currents. The influences to the flow fields increases with the grain size. 1 This work was supported by the National Natural Science Foundation of China under Grant No. 109103010062. This work was also supported by the National Climb- B Plan under Grant No. 2009CB724100. 2 Corresponding author 9:31AM M19.00008 Experimental study of lock-exchange gravity currents: Flow structures and instabilities1 , ZHUANG SU, Peking Univ, MING PENG, Guangdong Yudean Group Co Ltd, HUIJING YUAN, CUNBIAO LEE2 , Peking Univ — This work describes experimental investigations of lock-exchange gravity currents. High-speed particle image velocimetry and laser-induced fluorescence were used to study the spatial and temporal evolving of various flow structures and instabilities. Gravity currents develop a head-body-tail structure with different characteristics. The flow details inside the head are investigated. Kelvin-Helmholtz and gravitational instabilities are the two dominant instability modes here. These instabilities and their interactions strongly affect the head shape and the formation of the vertical structures in the currents. Two sets of inclined vertical structures are observed far behind the head. One set locates at the upper shear layer, tilting opposite to the flow direction. This set of structures evolves form the broken down K-H billows resulting from the interaction of the Kelvin-Helmholtz instability and gravitational instability in the current head. The other set locates at the bottom of the flow, tilting along the flow direction. This set is the result of the gravitational instability at the bottom. 1 This work was supported by the National Natural Science Foundation of China under Grant No. 109103010062. This work was also supported by the National Climb- B Plan under Grant No. 2009CB724100. 2 Corresponding author 9:44AM M19.00009 Three dimensional simulations of internal solitary waves1 , GUOTU LI, Duke University, FRANCESCO RIZZI, Sandia National Laboratories, OMAR KNIO, Duke University — This study focuses on mass transport and mixing induced by mode-2 internal solitary waves (ISWs) propagating along a pycnocline between two continuously stratified fluid layers. A direct numerical simulation (DNS) model is developed for the incompressible three-dimensional Navier-Stokes equations in the Boussinesq limit. By using high order schemes in both space and time, the model is able to accurately capture the convection-dominated flow at high Reynolds and Schmidt numbers. Simulations both with and without background shear are conducted. The spatial frequency analysis of both density and vorticity fields reveals that no long range spanwise structures are present during the propagation of ISWs, which makes a relatively short spanwise depth sufficient to characterize the evolution of the flow. The growth of 3D structures during the propagation of ISWs is quantified using a spanwise roughness measure. The flow energy budget, dye transport, density mixing and vortex circulations are also analyzed. 1 Work supported by the Office of Naval Research, Physical Oceanography Program. 9:57AM M19.00010 Simulations of Convective Excitation of Internal Waves in Water , DANIEL LECOANET, UC-Berkeley, GEOFF VASIL, University of Sydney, ELIOT QUATAERT, UC-Berkeley, KEATON BURNS, MIT, BEN BROWN, CU-Boulder, JEFF OISHI, AMNH & Farmingdale State College — We will present a series of simulations of convective excitation of internal gravity waves (IGWs) in water. We mimic the experimental set-up of Perrard et al. (2013), where water is cooled to zero degrees at the bottom of a tank, and keep at approximately room temperature at the top of a tank. The density maximum of water at 4 degrees renders the fluid convectively unstable between zero and four degrees, and stably stratified above four degrees. Our 2D simulations of the experiment show qualitatively similar IGW excitation spectrum. We then investigate two commonly discussed excitation mechanisms: interface forcing, and deep excitation. We run simplified simulations testing these two excitation mechanisms using the data from the full simulation, and compare the wave fields. We find that the interface forcing simulations overestimate the excitation of high frequency waves because high frequency interface motions are associated with nonlinear convection and not linear IGWs. On the other hand, the deep excitation of IGWs by Reynolds stresses accurately reproduces excitation spectrum. The correlation between the full simulation wave field and the deep excitation wave field is 0.95. This suggests that deep excitation is the dominant excitation mechanism for this system. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M20 Experimental Techniques: PIV II — 2008 - Alexandra Techet, Massachusetts Institute of Technology 8:00AM M20.00001 An experimental database for evaluating PIV uncertainty quantification methods , SCOTT WARNER, Utah State University, DOUGLAS NEAL, LaVision Inc., ANDREA SCIACCHITANO, a.sciacchitano@tudelft.nl — Uncertainty quantification for particle image velocimetry (PIV) data has recently become a topic of great interest as shown by the publishing of several different methods within the past few years. A unique experiment has been designed to test the efficacy of PIV uncertainty methods, using a rectangular jet as the flow field. The novel aspect of the experimental setup consists of simultaneous measurements by means of two different time-resolved PIV systems and a hot-wire anemometer (HWA). The first PIV system, called the “PIV-Measurement” system, collects the data for which uncertainty is to be evaluated. It is based on a single camera and features a dynamic velocity range (DVR) representative of many PIV experiments. The second PIV system, called the “PIV-HDR” (high dynamic range) system, has a significantly higher DVR obtained with a higher digital imaging resolution. The hot-wire was placed in close proximity to the PIV measurement domain. All three of the measurement systems were carefully set to simultaneously collect time-resolved data on a point-by-point basis. The HWA validates the PIV-HDR system as the reference velocity so that it can be used to evaluate the instantaneous error in the PIV-measurement system. 8:13AM M20.00002 New in-situ, non-intrusive calibration , HEATHER ZUNINO, RONALD ADRIAN, LIUYANG DING, Arizona State University, KATHY PRESTRIDGE, Los Alamos National Laboratory — Tomographic particle image velocimetry (PIV) experiments require precise and accurate camera calibration. Standard techniques make assumptions about hard-to-measure camera parameters (i.e. optical axis angle, distortions, etc.)–reducing the calibration accuracy. Additionally, vibrations and slight movements after calibration may cause significant errors–particularly for tomographic PIV. These problems are exacerbated when a calibration target cannot be placed within the test section. A new PIV camera calibration method has been developed to permit precise calibration without placing a calibration target inside the test section or scanning the target over a volume. The method is capable of correcting for dynamic calibration changes occurring between PIV laser pulses. A transparent calibration plate with fine marks on both sides is positioned on the test section window. Dual-plane mapping makes it possible to determine a mapping function containing both position and angular direction of central rays from particles. From this information, central rays can be traced into the test section with high accuracy. Image distortion by the lens and refraction at various air-glass-liquid interfaces are accounted for, and no information about the position or angle of the camera(s) is required. 8:26AM M20.00003 Iterative Particle Image Velocimetry Algorithm for Rotating Flows , MATTHEW GIARRA, Virginia Tech, JOHN CHARONKO, Los Alamos National Laboratory, PAVLOS VLACHOS, Purdue University — Particle image velocimetry (PIV) can fail to reliably estimate fluid velocities in flows with large spatial velocity gradients because the shearing, stretching, and rotation of particle image patterns decreases the signal-to-noise ratio of cross correlations (CCs). We present a new PIV correlation algorithm called the Fourier-Mellin correlation (FMC) that accurately measures particle pattern displacements in flow regions with large rotation (like vortex cores) compared to traditional correlations by measuring rotation and then aligning particle patterns before performing Cartesian CCs. FMC reliably measures particle displacements between interrogation regions with up to 180 degrees of angular misalignment compared to 6-8 degrees for traditional correlations. We combined our FMC algorithm with iterative discrete window offset (DWO) to measure velocity and vorticity fields in synthetic PIV images of counter-rotating vortex cores and an experimental vortex ring in water. FMC with DWO reduced the errors in velocity and vorticity estimates by an order of magnitude compared to traditional correlations with DWO, increased the correlation peak height ratios in synthetic and experimental images, and accelerated the convergence of iterative image deformation algorithms. 8:39AM M20.00004 Uncertainty Estimation in Stereoscopic Particle Image Velocimetry , SAYANTAN BHATTACHARYA, PAVLOS VLACHOS, Purdue University — In Stereoscopic Particle Image Velocimetry (Stereo-PIV) particle images are recorded using two viewing directions and the projected velocity components obtained in each view are combined to predict the three component velocity vector in the plane of measurement. The accuracy of the method depends on precise determination of viewing angles, measurement plane location and estimation of projected velocity components. However, the complex measurement chain with non-linear combination of errors make uncertainty estimation in Stereo-PIV challenging. Here we consider the overall uncertainty stemming from various error sources involved in the measurement process. The uncertainty in the absolute particle locations due to mismatch in the overlapping camera views are combined with the uncertainty in individual camera velocity components to predict the uncertainty in the reconstructed velocity field. The mapping function uncertainty and viewing angle uncertainty are also considered in the propagation equation. Present framework is tested with both simulated random field and experimental vortex ring image set. The RMS error and predicted uncertainties are compared for different viewing angle camera pairs. A sensitivity analysis of the individual uncertainty contributions to the overall uncertainty coverage is also presented. 8:52AM M20.00005 Comparative assessment of four a-posteriori uncertainty quantification methods for PIV data , PAVLOS VLACHOS, Purdue University, ANDREA SCIACCHITANO, Delft University of Technology, DOUGLAS NEAL, LaVision Inc., BARTON SMITH, SCOTT WARNER, Utah State University — Particle Image Velocimetry (PIV) is a well-established technique for the measurement of the flow velocity in a two or three dimensional domain. As in any other technique, PIV data are affected by measurement errors, defined as the difference between the measured velocity and its actual value, which is unknown. The objective of uncertainty quantification is estimating an interval that contains the (unknown) actual error magnitude with a certain probability. In the present work, four methods for the a-posteriori uncertainty quantification of PIV data are assessed. The methods are: the uncertainty surface method (Timmins et al., 2012), the particle disparity approach (Sciacchitano et al., 2013; the peak ratio approach (Charonko and Vlachos, 2013) and the correlation statistics method (Wieneke 2014). For the assessment, a dedicated experimental database of a rectangular jet flow has been produced (Neal et al. 2014) where a reference velocity is known with a high degree of confidence. The comparative assessment has shown strengths and weaknesses of the four uncertainty quantification methods under different flow fields and imaging conditions. 9:05AM M20.00006 Direct Estimation of Particle Image Velocimetry Measurement Uncertainty from Cross-Correlation Plane Moments , ZHENYU XUE, Mechanical Engineering, Virginia Tech, USA, SAYANTAN BHATTACHARYA, Mechanical Engineering, Purdue University, USA, JOHN CHARONKO, Physics Division, Los Alamos National Laboratory, USA, PAVLOS VLACHOS, Mechanical Engineering, Purdue University, USA — Particle Image Velocimetry is a non-invasive measurement technique in which images of flow tracers are correlated to estimate flow velocity. The coupled effect of error sources including particle image size, velocity gradient, out of plane motion, and seeding density poses a challenge in quantifying the uncertainty. Here we establish a method to quantify PIV uncertainty by extracting the Probability Density Function (PDF) of all possible displacements from the cross-correlation plane. The PDF is obtained by deconvolving particle image size from the correlation plane, and approximating its shape and standard deviation by an elliptic Gaussian least squares fit. The PDF variance is then scaled by a normalized estimate of the number of correlated particles between the image pairs to obtain the standard uncertainty. The method takes into account the peak stretching due to velocity gradients and also includes an estimate of bias error. The calculated uncertainty is compared with the RMS error for synthetic and experimental images, including a vortex ring and the recent uncertainty benchmark jet flow cases. Results show reasonable uncertainty coverage. Thus, the current framework provides a direct approach to quantify PIV uncertainty from the correlation plane. 9:18AM M20.00007 Benchmark measurements for evaluation of PIV uncertainty method , STAMATIOS POTHOS, TSI Inc., SAYANTAN BHATTACHARYA, PAVLOS VLACHOS, Purdue Univ, DAN TROOLIN, WING LAI, TSI Inc. — PIV combines a series of instruments, algorithms and user inputs in order to quantify the displacement of flow tracer patterns in complex flows. Each of these components is bound to introduce uncertainty in the resulting measurement, and often these uncertainties are coupled or difficult to estimate. Recent developments have now presented a series of methods for quantification of uncertainty in planar PIV measurements, however each of these methods appears to offer different advantages or disadvantages and their strengths and weaknesses are not well understood. Moreover, there is a need for extensive testing of these methods against a variety of real experimental data and flow conditions. In this work we execute a benchmark experiment of a flow over a cylinder using time resolved PIV with simultaneous LDV measurements to serve as a comparison benchmark, and we use these data to compare the different uncertainty quantification methods and assess their reliability. The presented comparisons will include signal to noise ratio methods, image disparity methods and correlation plane statistics and the estimated uncertainties will be assessed using error probability distributions, time series analysis, and coverage factors. 9:31AM M20.00008 Measurement uncertainty of mean velocity fields acquired by PIV , SVEN SCHARNOWSKI, CHRISTIAN J. KÄHLER, Bundeswehr University Munich — Particle Image Velocimetry (PIV) has become a standard tool for the investigation of various flow fields. In order to compare the mean velocity distributions or higher order statistics from experiments and numerical predictions, it is essential to know the uncertainty of the estimated values. However, due to the complex evaluation procedure of PIV the error cannot be estimated with standard methods. Many parameters, including particle image size, particle image density, turbulence level, noise level, velocity gradients, number if PIV image pairs . . . affect the accuracy. This work systematically analyzes the effect of several parameters on the random and bias errors of the estimated mean velocity by using single-pixel ensemble-correlation as well as window-correlation based PIV. To have full control of all parameters, synthetic PIV images are generated and analyzed, while identifying the most relevant error sources. The different parameters can be determined from the raw data by generating a multidimensional uncertainty hyper-surface that allows for determining the random error of the shift vectors. Furthermore, the knowledge about the dependency on the different parameters enables to identify the bottleneck and thus, to optimize the measurement setup and evaluation procedure to improve the accuracy. 9:44AM M20.00009 Image preprocessing method for particle image velocimetry (PIV) image interrogation near a fluid-solid surface , YIDING ZHU, LICHAO JIA, YE BAI, HUIJING YUAN, CUNBIAO LEE, Peking Univ — Accurate particle image velocimetry (PIV) measurements near the moving wall are a great challenge. The problem is compounded by the very large in-plane displacement on PIV images commonly encountered in measurements of the high speed flow. An improved image preprocessing method is presented in this paper. A wall detection technique is used first to qualify the wall position and the movement of the solid body. Virtual particle images are imposed in the solid region, of which the displacements are evaluated by the body movement. The estimation near the wall is then smoothed by data from both sides of the shear layer to reduce the large random uncertainties. Interrogations in the following iterative steps then converge to the correct results to provide accurate predictions for particle tracking velocimetries(PTV). Significant improvement is seen in Monte Carlo simulations and experimental tests such as measurements near a flapping flag or compressor plates. The algorithm also successfully extracted the small flow structures of the 2nd mode wave in the hypersonic boundary layer from PIV images with low signal-noise-ratios(SNR) when the traditional method was not successful. 9:57AM M20.00010 Two-colour micro-PIV and high speed shadowgraphy measurements for liquid-liquid plug flows1 , MAXIME CHINAUD, DIMITRIOS TSAOULIDIS, PANAGIOTA ANGELI, Department of Chemical Engineering, University College London, Torrington Place, London WC1E 7JE, UK, UNIVERSITY COLLEGE LONDON TEAM, MEMPHIS COLLABORATION — Two-colour micro-Particle Image Velocimetry (micro-PIV) is a relatively new technique that provides velocity fields simultaneously in both phases of a two-phase flow system. In this work, a laser emitting at two different wavelengths was used to excite two different types of particles, each added in one of the liquid phases of a two-phase, oil-water, system. The two types of particles emitted signals at separate wavelengths that were captured simultaneously by two different cameras. Instantaneous velocity fields could thus be obtained in both phases at the same time. This technique was used to study liquid-liquid plug flows in microchannels. Both plug propagation in the main channel and plug formation in the T-shaped inlet junction have been investigated. During plug propagation analysis of the velocity fields reveals recirculation patterns inside the dispersed plug and the continuous slug. These will be related to dimensionless numbers. The results on plug formation will be discussed against current models on plug size. 1 This work is undertaken as part of the UK Engineering and Physical Sciences Research Council Programme Grant MEMPHIS. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M21 Instability: Interfacial and Thin Films II 8:00AM M21.00001 ABSTRACT WITHDRAWN — 2010 - Sandra Troian, California Institute of Technology — 8:13AM M21.00002 Instabilities of evaporating non-isothermal ultra-thin film with insoluble surfactant , ALEXANDER MIKISHEV, Sam Houston State Univ, ALEXANDER NEPOMNYASHCHY, Technion-IIT — The stability of an evaporating ultra-thin liquid layer with insoluble surfactant spreading over a free deformable interface is investigated within lubrication theory. The evaporation process is described by 2D one-sided model based on the assumptions of density, viscosity and thermal conductivity of the gaseous phase being small compared to the same properties of the liquid phase. It is assumed that the thermal resistance to the evaporation at the interface is an increasing linear function of surfactant concentration. The evaporation mass flux depends on the interface temperature and vapor pressure. Using the long-wave approach and assumption of slow time evolution, a system of nonlinear equations governing the nonequilibrium evaporation is obtained. The system retains main physical effects which take place in the system. A linear stability analysis is also carried out. Both monotonic instability mode and oscillatory one are found and analyzed. The analysis does not include the Born repulsion force in intermolecular interactions. 8:26AM M21.00003 Three-dimensional modelling of film flows over spinning disks1 , KUN ZHAO, ALEX WRAY, JUNFENG YANG, OMAR MATAR, Imperial College London — Film flows over spinning disks are of central importance to a wide array of industrial processes, such as the augmentation of heat and mass transfer in chemical reactors, or power production in metallurgy. As a result they have been extensively investigated experimentally. Theoretically they constitute an interesting problem due to the interplay of inertial, capillary, centrifugal and Coriolis forces. However, modelling efforts have typically been restricted to the consideration of the one-dimensional axisymmetric situation. We extend the existing models to incorporate azimuthal variations. The resultant system is solved via the use of an operator-splitting method. In addition, we have performed Direct Numerical Simulations of the system. We compare the low order model, the direct simulations and the results of experiments, to reveal a wide variety of different flow regimes in accordance with existing literature. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1 8:39AM M21.00004 Buckling instabilities in photopolymerised gels1 , MATTHEW HENNESSY, JOAO CABRAL, OMAR MATAR, Imperial College London — Frontal photopolymerisation (FPP) is a process whereby solid polymer networks are created by illuminating a monomer-rich bath with collimated light. In practice, FPP can be used to rapidly fabricate intricate small-scale structures. Due to the attenuation of light as it propagates through the bath, polymerisation occurs in a wave-like fashion from the illuminated surface into the bulk. At low temperatures, the polymerisation front remains planar; however, at higher temperatures, it can undergo large deformations. We believe this is due the buckling of a thin gel layer that forms between the polymer-rich and solvent-rich phases. The gel is thought to buckle due to high compressive stresses that are generated as it absorbs solvent and swells. In this talk, we will present a mathematical model for gel formation which captures the phenomenon of buckling due to swelling. The gel is treated as a deformable porous medium and the solvent is assumed to flow according to Darcy’s law. We will also examine the buckling patterns that emerge from the model and compare them with experiments. 1 EPSRC Grant number EP/L022176/1 8:52AM M21.00005 Elastic membranes in confinement , JOSHUA BOSTWICK, MICHAEL MIKSIS, STEPHEN DAVIS, Northwestern University — An elastic membrane stretched between two walls takes a shape defined by its length and the volume of fluid it encloses. Many biological structures, such as cells, mitochondria and DNA, have finer internal structure in which a membrane (or elastic member) is geometrically “confined” by another object. We study the shape stability of elastic membranes in a “confining” box and introduce repulsive van der Waals forces to prevent the membrane from intersecting the wall. We aim to define the parameter space associated with mitochondria-like deformations. We compare the confined to ‘unconfined’ solutions and show how the structure and stability of the membrane shapes changes with the system parameters. 9:05AM M21.00006 Free surface dynamics of nematic liquid crystal1 , LINDA CUMMINGS, LOU KONDIC, MICHAEL LAM, New Jersey Institute of Technology, TE-SHENG LIN, National Chiao Tung University, Taiwan — Spreading thin films of nematic liquid crystal (NLC) are known to behave very differently to those of isotropic fluids. The polar interactions of the rod-like molecules with each other, and the interactions with the underlying substrate, can lead to intricate patterns and instabilities that are not yet fully understood. The physics of a system even as simple as a film of NLC spreading slowly over a surface (inclined or horizontal) are remarkably complex: the outcome depends strongly on the details of the NLC’s behavior at both the substrate and the free surface (so-called “anchoring” effects). We will present a dynamic flow model that takes careful account of such nematic-substrate and nematic-free surface interactions. We will present model simulations for several different flow scenarios that indicate the variety of behavior that can emerge. Spreading over a horizontal substrate may exhibit a range of unstable behavior. Flow down an incline also exhibits intriguing instabilities: in addition to the usual transverse fingering, instabilities can be manifested behind the flowing front in a manner reminiscent of Newtonian flow down an inverted substrate. 1 NSF DMS-1211713 9:18AM M21.00007 Electrified film flows at moderate Reynolds number1 , RICHARD CRASTER, ALEX WRAY, DEMETRIOS PAPAGEORGIOU, OMAR MATAR, Imperial College London — We examine the flow of a thin, inclined film sandwiched between two parallel electrodes. We follow the Weighted Residual Integral Boundary Layer method, which has been shown via comparison with both direct numerical simulations and experiments to give good results in both the drag-gravity and drag-inertia regimes. We extend existing models to give an accurate model of electrostatic effects via a similar separation of variables approach. A disparity in material properties between the liquid and gas regions induces a Maxwell stress at the interface, which affords a significant degree of control over the behaviour of the film. In one dimension, linear stability comparisons are made with a full Orr-Sommerfeld calculation, and nonlinear comparisons are made with direct numerical simulations, both showing excellent agreement in large parts of parameter space. The model is also extended to fully two-dimensional simulations. 1 EPSRC Programme Grant, MEMPHIS, EP/K0039761/1, EPSRC DTG Studentship (AWW) 9:31AM M21.00008 Phase diagram for the onset of rolling waves and flow reversal in inclined falling films , WILKO ROHLFS1 , RWTH - Aachen, BENOIT SCHEID2 , University Libre de Bruxelles, REINHOLD KNEER3 , RWTH - Aachen — The onset of rolling waves and the onset of flow reversal in inclined falling films is investigated in dependence of the Reynolds and the inclination number. For this, the weighted integral boundary layer model (WIBL) and direct numerical simulations (DNS) are used. Analytical criteria for the onset of rolling waves and flow reversal based on the wave celerity, the average film thickness and the maxi-mum/minimum film thickness have been approximated using self-similar parabolic velocity profiles. This approximation has been validated by second-order WIBL and DNS simulations. It is shown that the various transitions in the phase diagram for homoclinic solutions (waves of infinite wave length) are strongly dependent on the inclination, but independent on the streamwise viscous dissipation effect. Compared to the onset of flow reversal, the onset of rolling waves occurs for higher Reynolds numbers, resulting in a regime in which flow reversal and non-rolling waves coexist. Furthermore, simulation results for limit cycles (finite wave length) reveal a strong increase of the critical Reynolds number with the excitation frequency. 1 Institute of Heat and Mass Transfer, Augustinerbach 6, 52056 Aachen, Germany 165/67, Avenue F.D. Roosevelt 50, 1050 Bruxelles, Belgium 3 Institute of Heat and Mass Transfer, Augustinerbach 6, 52056 Aachen, Germany 2 C.P. 9:44AM M21.00009 Improved Measurements of the Dominant Mode Wavelength in Viscous Nanofilms Undergoing 3D Pillar Growth Via Bénard Type Instability1 , SANDRA TROIAN, KEVIN FIEDLER, California Institute of Technology, MC 128-95, Pasadena, CA 91125 — Free surface viscous nanofilms exposed to an initial uniform and very large transverse thermal gradient are prone to spontaneous formation and growth of nanopillars typically separated by tens of microns or less. Linear stability analyses of various interface equations in the long wavelength limit suggest these formations can result either from fluctuations in electrostatic forces between the fluid interface and induced image charge distribution,2 radiation pressure induced by acoustic phonon reflections,3 or thermocapillary stresses leading to Bénard-like deformations.4 Here we review improvements over previous comparison to theoretical predictions5 which suggest even closer agreement with the thermocapillary model; however, systematic discrepancies persist. We have therefore redesigned our experimental system for more accurate thermal flux control and estimation and will discuss our newest results. 1 Financial support from a 2013 NASA Space Technology Research Fellowship is gratefully acknowledged. Y. Chou and L. Zhuang, J. Vac. Sci. Technol. B 17, 3197 (1999) 3 E. Schäffer et al., Macromolecules 36, 1645 (2003) 4 M. Dietzel and S. M. Troian, Phys. Rev. Lett. 103 (7), 074501 (2009); M. Dietzel and S. M. Troian, J. Appl. Phys. 108, 074308 (2010) 5 E. McLeod, Y. Liu and S. M. Troian, Phys. Rev. Lett. 106, 175501 (2011) 2 S. 9:57AM M21.00010 Lyapunov Analysis of the Stability of Nanodroplet Arrays Arising From Steady State Bénard Flow in the Long Wavelength Limit , ZACHARY NICOLAOU, SANDRA TROIAN, California Institute of Technology, MC 128-95, Pasadena, CA 91125 — Previous work in our group has focused on a novel Bénard like instability leading to nanopillar arrays in ultrathin viscous films subject to a transverse thermal gradient.1,2,3 The shape and size of these formations is influenced by the relative strength of the thermocapillary to capillary stresses. In turn, this ratio is dependent on the system geometry, fluid material properties, overall magnitude of the applied thermal gradient, and whether volume is conserved. Here we examine the parameter regime corresponding to steady state solutions resembling either isolated or extended sinusoidal-like states. The linear stability of rectilinear and axisymmetric formations is investigated by a combination of Lyapunov analysis, asymptotic methods, and numerical simulations. Our findings indicate that radially symmetric arrays with small peak heights are linearly stable. The existence of stable axisymmetric states for parameter values accessible to experiment offers an intriguing route for non-contact fabrication of microlens arrays. 1 M. Dietzel and S. M. Troian, Phys. Rev. Lett. 103 (7), 074501 (2009) Dietzel and S. M. Troian, J. Appl. Phys. 108, 074308 (2010) 3 E. McLeod, Y. Liu and S. M. Troian, Phys. Rev. Lett. 106, 175501 (2011) 2 M. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M22 Instability: Multiphase Flow — 2012 - Snezhana Abarzhi, Carnegie Mellon University 8:00AM M22.00001 Stability of a hydrodynamic discontinuity1 , SNEZHANA I. ABARZHI, Carnegie Mellon University, YASUHIDE FUKUMOTO, Kyushu University, Japan, LEO P. KADANOFF, The University of Chicago, USA — While looking from a far field at a discontinuous front separating two incompressible ideal fluids of different densities, we identify two qualitatively different behaviors of the front (unstable and stable) depending upon whether the energy flux produced by the perturbed front is large or small compared to the flux of kinetic energy across the planar front. Landau’s solution for the Landau-Darriues instability is consistent with one of these cases, whether the gravity is present or not. 1 The work is supported by the US National Science Foundation. 8:13AM M22.00002 Phase Change Effects on Immiscible Flow Displacements in Radial Injection , MAJID AHMADLOUYDARAB, JALEL AZAIEZ, ZHANGXIN CHEN, university of Calgary — We report a systematic simulation of immiscible fluid-fluid displacements in radial injection in the presence of phase change. Due to the presence of two fluid-fluid interfaces in the system, a special treatment has been adopted. To track the leading interface position, two highly accurate methods including Level Set and Immersed Interface Method were used, while for locating the trailing interface an energy equation was adopted assuming the existence of a constant thin condensate layer. Dimensional analysis led to three important dimensionless groups including capillary number (Ca), Jacob number (Ja) and viscosity ratios (M) of the three fluids. Simulation results indicate significant influences of these parameters on the development of the instability and the interfacial morphology of fingers. Increasing Ca or M tends to amplify the interfacial instability, fingertip splitting, and results in longer fingers. In contrast, increasing Ja has stabilizing effects due to an increase of the thickness of the condensate layer. On the other hand at lower viscosity ratios as well as lower Ca, because of compensation effects of the phase change, both leading and trailing interfaces are found to be less unstable. Moreover accumulated condensate and oil saturation depletion curves show increasing and decreasing trends, respectively, when the Ca increases. Although viscosity ratio and Ja have similar effects on the accumulated condensate, they do not show any effect on the oil depletion saturation. 8:26AM M22.00003 Instability of a particle-laden jet in a confined environment , FLORENT P.M. SHARPIN1 , Ecole Polytechnique, JULIEN R. LANDEL, DAMTP, University of Cambridge, C.P. CAULFIELD, BPI & DAMTP, University of Cambridge — The dynamics of particle-laden jets is relevant to many geophysical events and industrial applications from volcanic eruptions to chemical reactors and oil refinement. We consider experimentally the dynamically rich behavior of a vertical momentum jet, constrained in a narrow gap whose length is two orders of magnitude smaller than the length-scales of the other two dimensions, and constrained to flow through, from below, a bed of small heavy particles. In the regime where the jet has eroded a large triangular region of the particle bed, a dense particle-laden jet develops, as the initially pure jet continually entrains, and carries to some height above the bed, a certain concentration of particles. This coupled particle-laden jet is unstable and oscillates from side to side in the confined environment. A large vortical structure forms as the particle-laden jet tilts sideways, at a well-defined frequency. Using an analogy with turbulent, single-phase fountains, we model the maximum height of rise of the particle-laden jet using a ratio between the single-phase jet source volume flux, and its coupled, particle-laden negative source buoyancy flux, which we determine using a novel non-intrusive technique. We also model the frequency of the particle-laden jet instability using the characteristic travel time of a particle in the jet, which also depends on the reduced gravity of the particle-laden jet. 1 This research was undertaken while at the DAMTP, University of Cambridge (UK) during an internship. 8:39AM M22.00004 Spatio-temporal evolution of interfacial instabilities in vertical gas-liquid flows , PATRICK SCHMIDT, PRASHANT VALLURI, University of Edinburgh, LENNON Ó NÁRAIGH, University College Dublin, MATHIEU LUCQUIAUD, University of Edinburgh — Vertical gas-liquid flows are characteristic for process engineering and widely employed in various technical applications. However, the dynamic behaviour of the liquid interface in such flows is still not fully understood. We focus in our work on characterising the interfacial instability as well as associated interfacial waves in vertical laminar-laminar gas-liquid flows over a wide range of parameters covering different flow regimes, i.e. counter-current, zero-interface velocity (loading) and partial-to-full liquid flow reversal (flooding). High-resolution direct numerical simulations using the TPLS flow solver (http://sourceforge.net/projects/tpls/) reveal the existence of weakly nonlinear interfacial waves, which are in good agreement with Stuart-Landau theory. These waves travel down- or upstream, depending on the flow regime. Furthermore, spatio-temporal linear stability analysis indicates the occurrence of absolute instability within the investigated parameter range. DNS is used to analyse this feature in more detail whereby agreement with linear theory has been established. 8:52AM M22.00005 Control strategy for a double-diffusive two-fluid channel flow: A stability analysis , SUKHENDU GHOSH, R. USHA, KIRTI SAHU, Indian Institute of Technology Madras — The effect of velocity slip (symmetric and asymmetric) at the walls on the linear stability characteristics of miscible two-fluid channel flow is considered in the presence of double diffusive (DD) phenomenon. The channel walls are made of same material or different materials; this suggests symmetric or asymmetric slip condition at the walls. The fluids are miscible, and consist of two solute species having different rate of diffusion. Both the fluids are assumed to be of the same density, but varying viscosity, which depends on the concentration of the solute species. This flow system is more unstable than the corresponding single component (SC) flow as well as unstratified flow, due to the presence of double-diffusive (DD) effect. When the mixed region of the fluids moves towards the channel walls a new unstable mode (namely the DD mode) arises at low Reynolds numbers. The slip parameter has nonmonotonic effect on the stability characteristics in this system. The trend of slip effect is influenced by other flow parameters. The effects of wall slip on the flow stability is weak or strong depending on the slip condition, only for the upper wall or only for the lower layer or for both the walls. Increasing the value of the slip parameter delays the first occurrence of the DD-mode. 9:05AM M22.00006 The Effect of Varying Stokes Number on the Growth Rate of Instabilities in Particle-Laden Shear Layers , SEAN DAVIS, GIACOMO SENATORE, GUSTAAF JACOBS, San Diego State University — The growth rates of instabilities in the shear layer of a stratified particle-laden flow are analyzed using both a Linear Stability Analysis (LSA) of a stochastic Eulerian-Eulerian (EE) model and a high-order Eulerian-Lagrangian (EL) computation. In the LSA, a modified Rayleigh equation is derived, which governs the linear growth rate of a spatially periodic disturbance while the EL method solves the inviscid Euler equations using a particle source in cell method. A study of the particle response time shows that small-inertia particles (St < 0.2) may destabilize the inviscid mixing layer development as compared to a pure-gas flow. Energy is transferred globally from the particle phase to the fluid phase triggering this destabilization. The maximum stabilizing effect occurs at intermediate St (1 < St < 10), while multiple unstable modes coexist at larger St. These small, medium and large St effects are validated with numerical experiments using the EL code, showing very good agreement with the growth rates computed using the LSA. 9:18AM M22.00007 Turbulent transition modification in dispersed two-phase pipe flow1 , KYLE WINTERS, ELLEN LONGMIRE, University of Minnesota — In a pipe flow, transition to turbulence occurs at some critical Reynolds number, Rec , and transition is associated with intermittent swirling structures extending over the pipe cross section. Depending on the magnitude of Rec , these structures are known either as puffs or slugs. When a dispersed second liquid phase is added to a liquid pipe flow, Rec can be modified. To explore the mechanism for this modification, an experiment was designed to track and measure these transitional structures. The facility is a pump-driven circuit with a 9m development and test section of diameter 44mm. Static mixers are placed upstream to generate an even dispersion of silicone oil in a water-glycerine flow. Pressure signals were used to identify transitional structures and trigger a high repetition rate stereo-PIV system downstream. Stereo-PIV measurements were obtained in planes normal to the flow, and Taylor’s Hypothesis was employed to infer details of the volumetric flow structure. The presentation will describe the sensing and imaging methods along with preliminary results for the single and two-phase flows. 1 Supported by Nanodispersions Technology. 9:31AM M22.00008 Studies of Interfacial Perturbations in Two Phase Oil-Water Pipe Flows Induced by a Transverse Cylinder1 , MAXIME CHINAUD, KYEONG PARK, Department of Chemical Engineering, University College London, JAMES PERCIVAL, Department of Earth Science and Engineering, Imperial College London, OMAR MATAR, Department of Chemical Engineering, University College London, CHRISTOPHER PAIN, Department of Earth Science and Engineering, Imperial College London, PANAGIOTA ANGELI, Department of Chemical Engineering, University College London — Droplet detachment from interfacial waves has been the subject of many studies. To observe this phenomenon experimentally it is necessary to spatially localize the drop formation and enable quantitative measurements. In this study, a novel approach is followed where a transverse cylinder is introduced close to the mixing point of the two phases in oil-water flows which induces waves. The introduction of the cylinder induces interfacial waves that lead to drop detachment. High speed visualization has been used to generate flow pattern maps with this new system. The dispersed patterns induced by the cylinder will be linked to pressure drop measurements. The interface downstream the cylinder is affected by three different contributions: the vortices shed by the cylinder, the wall effects due to the pipe itself and the interface fluctuations due to the mixing of the two phases. These contributions will be quantified through a numerical study. A mesh adaptive multiphase finite element Navier Stokes solver, Fluidity, will be used to obtain flow pattern maps for 2D channel flow. The numerical findings will be compared against the experimental results. 1 This work is undertaken as part of the UK Engineering and Physical Sciences Research Council Programme Grant MEMPHIS. 9:44AM M22.00009 Numerical study of interface stability in presence of surfactant in two phase couette flow using a multiphase lattice boltzmann method , V.P.T.N.C.SRIKANTH BOJJA, Norwegian Univ Tech (NTNU) — The multiphase lattice Boltzmann approach is used to study the dynamics of interface between two immiscible fluids of different densities and viscosities in presence of a insoluble surfactant in Coutte flow. The simulations are performed on muti-cpu cluster using MPI. The effects of inertia (Reynolds number) and surfactant (Marangoni number) on the stability of the interface at arbitary wave numbers are investigated. Neutral-stability and growth-rate curves are plotted at different wave-numbers, Reynolds numbers and Marangoni numbers. Interesting phenomenon of surfactant accumulation on the crest of interfacial waves is observed and subsequent breakdown of the interfacial wave, droplet formation,entrainment are also observed in 3D simulations. 9:57AM M22.00010 Thermal dispersion effects on the two-phase zone with evaporation in a porous medium , MANUEL PERALTA GUTIÉRREZ, Univ Nacl Autonoma de Mexico, OSCAR BAUTISTA, Instituto Politécnico Nacional — The one-dimensional steady-state heat transfer in a two-phase zone of a water-saturated porous medium is studied numerically by including thermal dispersion effects. The physical system consists of a porous medium-liquid-vapor mixture that is heated from above and maintaining a fixed temperature on the bottom surface. Under certain conditions, a two-phase zone of both vapor and liquid exists in the middle of the region of the porous medium. A mathematical model for the temperature and the liquid saturation profiles within this two-phase zone is formulated by allowing for explicit temperature dependence for the saturation vapor pressure together with explicit saturation dependence for the capillary pressure. The set of resultant equations is numerically integrated by using a conventional fourth order Runge-Kutta scheme. The results evidence the strong influence of the thermal dispersion, porosity and pore diameter on the two-phase zone. R1 R3 Tuesday, November 25, 2014 8:00AM - 10:10AM — Session M23 Geophysical Fluid Dynamics: Atmospheric and Oceanic Applications 2001 - Alberto Scotti, University of North Carolina at Chapel Hill 8:00AM M23.00001 Tornado-like flows driven by magnetic body forces1 , GUNTER GERBETH, ILMARS GRANTS, TOBIAS VOGT, SVEN ECKERT, Helmholtz-Zentrum Dresden - Rossendorf, Institute of Fluid Dynamics — Alternating magnetic fields produce well-defined flow-independent body forces in electrically conducting media. This property is used to construct a laboratory analogue of the Fiedler chamber with a room-temperature liquid metal as working fluid. A continuously applied rotating magnetic field (RMF) provides the source of the angular momentum. A pulse of a much stronger travelling magnetic field drives a converging flow at the metal surface, which focuses this angular momentum towards the axis of the container. The resulting vortex is studied experimentally and numerically. In a certain range of the ratio of both driving actions the axial velocity changes its direction in the vortex core, resembling the subsidence in an eye of a tropical cyclone or a large tornado. During the initial deterministic spin-up stage (T. Vogt et al., JFM 736, 2013, pp. 641) the vortex is well described by axisymmetric direct numerical simulation. Being strong enough the flow develops a funnel-shaped surface depression that enables visual observation of the vortex structure. As the RMF strength is increased the eyewall diameter grows until it breaks down to multiple vortices. A number of further observed similarities to tornado-like vortices will be discussed. 1 The work is supported by the German Helmholtz Association in frame of the LIMTECH alliance. 8:13AM M23.00002 A Minimal Model for Precipitating Turbulent Convection1 , LESLIE SMITH, University of Wisconsin, Madison, GERARDO HERNANDEZ-DUENAS, National University of Mexico, SAM STECHMANN, University of Wisconsin, Madison — To construct a minimal model for precipitating turbulent convection, we consider simplified bulk cloud physics assuming infinitely fast condensation, evaporation and auto-conversion from cloud to rain water. The model sacrifices all microphysics but retains important conservations principles. It is demonstrated numerically that the model is able to capture convective organization, such as squall lines. Linear analysis of a saturated base state identifies the stable, unstable and conditionally stable regions of parameter space. The two delineating parameters are established numerically in the general case of finite rainfall speed. Each parameter is also derived in an appropriate limiting scenario: the condition sufficient for instability (stability) is analytically found for the limit of zero (infinite) rainfall speed. Energy considerations further support the numerical and limiting analytical calculations. 1 Supported by NSF CMG 1025188 8:26AM M23.00003 Turbulent growth of cloud droplets without collisions , ALBERTO DE LOZAR, LUKAS MUESSLE, JUAN PEDRO MELLADO, Max Planck Institute for Meteorology — It is believed that the increase of droplet collisions due to turbulence is key for the initiation of rain in warm clouds. In particular, the turbulence enhancement of collisions might explain how some lucky droplets grow from 20 to 50 micrometers, a regime in which neither condensation nor collisions due to settling are effective. Stratocumulus clouds, however, do not fit in this picture because typical turbulence dissipation rates are too low to enhance collisions appreciably, but at the same time these clouds produce significant drizzle. We explore the possibility that long-wave radiation causes a significant part of the droplet growth in stratocumulus. In our simulations the bulk properties of the cloud are calculated in the Eulerian field, while at the same time some droplets are tracked using a Lagrangian scheme. The advantage of our formulation is that condensation-evaporation processes are assumed to be infinitely fast, and do not need to be resolved explicitly. This allows us to investigate domains hundreds of meters wide for several minutes, thus resolving the relevant scales for radiative cooling. In this talk we will show results of how the droplet size distribution evolves due to radiation and turbulence, using one billion Lagrangian droplets. 8:39AM M23.00004 The Walker circulation, diabatic heating, and outgoing longwave radiation1 , REED OGROSKY, SAMUEL STECHMANN, University of Wisconsin-Madison — The Matsuno-Gill model, derived from the forced shallow- water equations in the tropics, has been widely used to describe the large-scale overturning circulation in the tropical atmosphere. This model contains damping terms in the form of Rayleigh friction and Newtonian cooling. Here, using new data analysis techniques, evidence suggests that damping is actually negligible. Specifically, near the equator, the east–west overturning circulation is in agreement with the undamped wave response to atmospheric heating. To estimate the heating, satellite observations of outgoing longwave radiation (OLR) are used. Frequently OLR is used as a heuristic indicator of cloudiness. Here, the results further suggest that OLR variations are actually proportional to total diabatic heating variations, with a proportionality constant of 18 W m−2 (K day−1 )−1 . While the agreement holds best over long time averages of years or decades, it also holds over shorter periods of one season or one month. Consequently, it is suggested that the strength of the Walker circulation—and its evolution in time—could be estimated using satellite data. 1 We gratefully acknowledge support from ONR N00014-12-1-0744 and ONR N00014-12-1-0912. 8:52AM M23.00005 Coordinated in-situ observation of developing hurricanes using atmospheric balloons – a Model Predictive Control approach , GIANLUCA MENEGHELLO, THOMAS BEWLEY, Flow Control Lab, UC San Diego — Current operational methods used to monitor the development of hurricanes and typhoons include radar and satellite imagery as well as dropsondes parachuted from repeated aircraft flights above the hurricane itself. The accurate in-situ measurements provided by dropsondes are especially valuable for generating an accurate forecast of a hurricane’s evolution and landfall. Unfortunately, the data from dropsondes is expensive to obtain (requiring many hazardous high-altitude flights) and limited both spatially (to the vertical profile of its path) and temporally (to the ten or twenty minutes it takes to fall). We show in the present work how receding-horizon MPC can be used to coordinate a formation of sensor-laden atmospheric balloons, distributing them quasi-uniformly across a realistic developing hurricane flowfield for days at a time. Several atmospheric balloons can be released from a high-altitude aircraft, or launched from a ship at sea level, and distributed over the hurricane thereafter. Certain target orbits of interest in the hurricane can be continuously sampled by some balloons, while other balloons make continuous sweeps between the eye and the spiral rain bands. Various solution methods for the optimal control problem arising within the MPC framework are considered. 9:05AM M23.00006 Enabling high-resolution simulations of atmospheric flow over complex terrain in the WRF model , KATHERINE LUNDQUIST, JEFF MIROCHA, Lawrence Livermore National Laboratory, DAVID WIERSEMA, JINGYI BAO, University of California, Berkeley, MEGAN DANIELS, Lawrence Livermore National Laboratory, FOTINI CHOW, University of California, Berkeley — As model grid resolution increases, atmospheric models are able to represent fine scale terrain, which can result in steep terrain slopes. The standard terrainfollowing coordinates used by models such as WRF (Weather and Research Forecasting) are unable to handle very steep terrain because of the grid distortion and related numerical errors. This has prompted the development of an alternative gridding technique in the WRF model, known as the immersed boundary method (IBM), which eliminates terrain-following grids and the associated errors (Lundquist et al. 2010,2012). This implementation, WRF-IBM, has been validated for idealized cases and real urban cases with excellent results; however, to date WRF-IBM has been applied with idealized lateral boundary conditions, and uses a no-slip boundary condition. In this work, we detail a multi-year effort to develop WRF-IBM for real, multi-scale simulations, including full atmospheric physics. Results from three aspects of this project are presented: initializing IBM domains using real meteorological and surface data, developing a nest interface between domains using terrain-following and IBM coordinates, and modifying the IBM boundary condition to include a wall model. 9:18AM M23.00007 Urban Flow and Pollutant Dispersion Simulation with Multi-scale coupling of Meteorological Model with Computational Fluid Dynamic Analysis , YUSHI LIU, HEE JOO POH, Institute of High Performance Computing — The Computational Fluid Dynamics analysis has become increasingly important in modern urban planning in order to create highly livable city. This paper presents a multi-scale modeling methodology which couples Weather Research and Forecasting (WRF) Model with open source CFD simulation tool, OpenFOAM. This coupling enables the simulation of the wind flow and pollutant dispersion in urban built-up area with high resolution mesh. In this methodology meso-scale model WRF provides the boundary condition for the micro-scale CFD model OpenFOAM. The advantage is that the realistic weather condition is taken into account in the CFD simulation and complexity of building layout can be handled with ease by meshing utility of OpenFOAM. The result is validated against the Joint Urban 2003 Tracer Field Tests in Oklahoma City and there is reasonably good agreement between the CFD simulation and field observation. The coupling of WRF- OpenFOAM provide urban planners with reliable environmental modeling tool in actual urban built-up area; and it can be further extended with consideration of future weather conditions for the scenario studies on climate change impact. 9:31AM M23.00008 An experimental study of the flow pattern and heat transport behavior in horizontal convection with large Rayleigh number and small aspect ratio1 , KE-QING XIA, SHI-DI HUANG, Department of Physics, The Chinese University of Hong Kong, Shatin, Hong Kong, China — Horizontal convection is a simple conceptual model to understand the role of buoyancy in the Meridional Overturning Circulation (MOC). Here we report an experimental study of the flow pattern and heat transport behavior in horizontal convection with Rayleigh number Ra up to 2 × 1012 and aspect ratio of 0.1 using a long apparatus. Flow visualization studies reveal that it is not necessary for the returning flow to penetrate the strong stratification in the thermal BLs, suggesting that much less energy may be required to maintain a global circulation than is generally believed. Moreover, both the heat transport efficiency and thermal BL thicknesses are found to follow a 0.3 power law, which indicates a stronger heat transport in horizontal convection with large Ra number than is suggested in the literature. These findings on horizontal convection may be relevant to the driving mechanism of the MOC. 1 This work is supported by the Hong Kong Research Grants Council under Grant No. CUHK403811. 9:44AM M23.00009 DNS of Horizontal Convection1 , BRIAN WHITE, ALBERTO SCOTTI, Dept. of Marine Sciences, UNC Chapel Hill — We perform three-dimensional DNS of Horizontal Convection in a rectangular tank with idealized boundary conditions. The flow is driven by imposing the profile for the buoyancy b at the surface, where it ranges from b0 to b0 + ∆b and the transition region is confined to a very small area. The Rayleigh based on the domain depth ranges from 105 to 1012 . The scaling observed for the Nusselt number and the strength of the circulation is consistent with Rossby’s scaling across the range of Rayleigh numbers considered, indicating that the dynamics in the boundary layer under the “warming” side throttles the flow. Energetically, we find that Available Potential Energy (APE) is generated along the surface, and converted to Kinetic Energy (KE). Along the descending plume energy goes from APE to KE up to Ra ∼ 1011 . For higher Rayleigh numbers the plume becomes a net sink of APE. When the switch occurs, a stagnant layer develops near the bottom, and the overall circulation becomes characterized by a narrow plume which retroflects rapidly towards the surface, with a shallow recirculation to close the flow. This may indicate the beginning of a Sandström regime characterized by a stagnant abyssal region and a shallow circulation. 1 Work supported by the National Science Foundation 9:57AM M23.00010 An experimental study of bottom heating effects in horizontal convection1 , FEI WANG, SHI-DI HUANG, Department of Physics, The Chinese University of Hong Kong, Shatin, Hong Kong, China, SHENG-QI ZHOU, State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou 510301, China, KE-QING XIA, Department of Physics, The Chinese University of Hong Kong, Shatin, Hong Kong, China — We report an experimental study of bottom heating effects in horizontal convection. The horizontal convection is driven by a surface heat flux Qs and a small amount of heat flux Qb is applied at the bottom boundary as a perturbation. It is found that while the bottom heating has negligible effect on the thermal properties at the top surface, its influences on the interior temperature and the strength of the downwelling flow are remarkable and such influences are more significant for stronger bottom heating and larger Rayleigh number Ra. Most importantly, direct velocity measurements at Ra around 5 ×109 reveal that the overturning rate, characterized by the maximum stream function, is increased by up to 111% and 256% for η = Qb / Qs = 2% and 6.8% cases, respectively, which is consistent with previous numerical studies. These results might be helpful to understand how geothermal heating will affect the oceanic circulation. 1 This work is supported by the Hong Kong Research Grants Council under Grant No. CUHK403811. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M24 Granular Flows: General — 2003 - Christine M. Hrenya, University of Colorado Boulder 8:00AM M24.00001 Imaging Forces in a Three-Dimensional Granular Material , JONATHAN BARES, Duke University, JOSHUA DIJKSMAN, Wageningen University, NICOLAS BRODU, INRIA Bordeau, ROBERT BEHRINGER, Duke University — We experimentally study the quasi-static deformation of a three-dimensional sphere packings subjected to macroscopic deformation. We perform these experiments on slightly polydisperse and nearly frictionless soft hydrogel spheres in a modified tri-axial shear apparatus. We resolve the microscopic force and displacement network in a this three dimensional packing through imaging the entire packing at different loading steps. By resolving particle deformations via custom written image analysis software, we extract all particle contacts and contact forces with a very good accuracy. In addition, we measure boundary stresses during compression and shear. We address the non-linear force response of a disordered packing under compression and shear, force network dynamics and explore the plastic rearrangements inside cyclically sheared and compressed packings. 8:13AM M24.00002 Three-dimensional particle tracking velocimetry applied to granular flows down rotating chutes , HERMAN CLERCX, SUSHIL SHIRSATH, JOHAN PADDING, HANS KUIPERS, Eindhoven University of Technology — We report on the cross-validation of 3D particle tracking velocimetry (3D-PTV) with other measurement techniques, such as particle image velocimetry (PIV), electronic ultrasonic sensor measurements for bed height and the discrete element model (DEM), for gaining more insight into the behavior of granular flows down inclined rotating chutes. In particular we aim at gaining access to Lagrangian displacement data of surface particles in granular flows and to obtain independent measurements of both the surface velocity and the bed height in the chute. The 3D-PTV method is based on imaging and tracking colored tracer particles that are introduced in the granular material, which are viewed from three directions. The three cameras collect consecutive frames a known ∆t apart and the PTV algorithm for locating and tracking particles is used to determine particle trajectories and velocities. The PTV and PIV results are in good mutual agreement with regard to the streamwise and spanwise surface velocity. The particle bed height obtained from 3D-PTV was compared with data from an ultrasonic bed-height sensor and it is found to be in good mutual agreement, as was the case for the comparison between the experimental findings from 3D-PTV and simulations by DEM. 8:26AM M24.00003 Electrical charging in shaken granular media , FREJA NORDSIEK, DANIEL LATHROP, University of Maryland at College Park — Results are presented on the electrical charging of granular media shaken between two conducting plates. Voltage measurements were taken between the plates for both monodisperse and bidisperse sets of particles of different materials with diameters in the 100 micron to 1 mm range. Particle charging was observed through capacitive coupling with the plates and electrical discharges. The following results were observed: 1) a monotonic increase in charging with shaking strength, 2) a threshold in the number of particles of filling the cell with about one layer of particles to see charging, 3) material and diameter differences causing an order of magnitude spread in measured signal, and 4) long time scale transients. The influence of collective effects and the potential relevance to natural charging phenomena seen in sand storms, volcanic ash clouds, thunderstorms, and thundersnow are discussed. 8:39AM M24.00004 Statistics from granular stick-slip experiments1 , AGHIL ABED ZADEH, JONATHAN BARES, ROBERT P. BEHRINGER, Duke University — We carry out experiments to characterize stick-slip for granular materials. In our experiment, a constant speed stage pulls a slider which rests on a vertical bed of circular photoelastic particles in a 2D system. The stage is connected to the slider by a spring. We measure the force on the spring as well as the slider’s acceleration by a force sensor attached to the spring and accelerometers on the slider. The distributions of energy release and time duration of avalanches during slip obey power laws. We apply a novel event recognition approach using wavelets to extract the avalanche properties. We compare statistics from the wavelet approach with those obtained by typical methods, to show how noise can change the distribution of events. We analyze the power spectrum of various quantities to understand the effect of the loading speed and of the spring stiffness on the statistical behavior of the system. Finally, from a more local point of view and by using a high speed camera and the photoelastic properties of our particles, we characterize the internal granular structure during avalanches. 1 This work supported by NSF Grant DMR1206351 and NASA grant NNX10AU01G 8:52AM M24.00005 Preventing shear thickening in granular suspensions by enhancing hydrodynamic interactions , QIN XU, James Franck Institue and Physics Department, University of Chicago, SAYANTAN MAJUMDAR, James Franck Institue, University of Chicago, HEINRICH JAEGER, James Franck Institue and Physics Department, University of Chicago — As a critical volume fraction is approached, granular suspensions can increase their viscosity dramatically under rapid shear; i.e., they exhibit Discontinuous Shear Thickening (DST). Previous works show that this phenomenon is related to frictional particle interactions and the formation of force chains that span the system, similar to dry granular materials. However, frictional contacts can be possibly reduced by lubrication. We experimentally studied the flow dynamics of dense granular suspensions in highly viscous liquids. By combining rheological measurements and fast imaging techniques, we characterized the flow curves for different liquid viscosities η0 . We found that shear thickening becomes weaker with η0 and eventually disappears for highly viscous solvent. In this regime, the suspensions show a Newtonian-like behavior with constant viscosity under shear. The crossover from granular to Newtonian regimes reflects the competition between friction and hydrodynamics. 9:05AM M24.00006 Non-Newtonian stress tensor and thermal conductivity tensor in granular plane shear flow , MEHEBOOB ALAM, SAIKAT SAHA, Jawaharlal Nehru Centre for Advanced Scientific Research — The non-Newtonian stress tensor and the heat flux in the plane shear flow of smooth inelastic disks are analysed from the Grad-level moment equations using the anisotropic Gaussian as a reference. Closed-form expressions for shear viscosity, pressure, first normal stress difference (N1 ) and the dissipation rate are given as functions of (i) the density or the area fraction (ν), (ii) the restitution coefficient (e), (iii) the dimensionless shear rate (R), (iv) the temperature anisotropy [η, the difference between the principal eigenvalues of the second moment tensor] and (v) the angle (φ) between the principal directions of the shear tensor and the second moment tensor. Particle simulation data for a sheared hard-disk system is compared with theoretical results, with good agreement for p, µ and N1 over a large range of density. In contrast, the predictions from a Navier-Stokes order constitutive model are found to deviate significantly from both the simulation and the moment theory even at moderate values of e. We show that the gradient of the deviatoric part of the kinetic stress drives a heat current and the thermal conductivity is characterized by an anisotropic 2nd rank tensor for which explicit expressions are derived. 9:18AM M24.00007 ABSTRACT WITHDRAWN — 9:31AM M24.00008 Predictive simulation of granular flows applied to compressible multiphase flow modeling , RYAN J. GOETSCH, JONATHAN D. REGELE, Iowa State University — Multiphase flows have been an active area of research for decades due to their complex nature and occurrence in many engineering applications. However, little information exists about the dense compressible flow regime. Recent experimental work [Wagner et al., Exp. Fluids 52, 1507 (2012)] using a multiphase shock tube has studied gas-solid flows with high solid volume fractions (α = 0.2) by measuring shock wave-particle cloud interactions. It is still unclear what occurs at the particle scale inside and behind the particle cloud during this interaction. The objective of this work is to perform direct numerical simulations to understand this phenomena. With this goal in mind, a discrete element method (DEM) solver was developed to predict the properties of a particle cloud formed by gravity driven granular flow through a slit opening. For validation purposes, the results are compared with experimental channel flow data. It is found that the mean velocity profile and mass flow rates correlate well with the experiment, however the fluctuation velocities are significantly under-predicted for both smooth and rough wall cases. 9:44AM M24.00009 Anomalous effects in granular Poiseuille flow: temperature bimodality and Knudsen minima , DEEPTHI SHIVANNA, MEHEBOOB ALAM, Jawaharlal Nehru Centre for Advanced Scientific Research — Two well- known rarefaction effects, the Knudsen minima and the bimodality of the temperature profile, are investigated in the granular analog of the Poiseuille flow via event-driven simulations of smooth inelastic hard-disks under gravity. The appearance of the bimodal-shape of the granular temperature is found to depend crucially on wall conditions: the bimodality is most prominent for the intermediate case between the specular and the bounce-back wall-particle collisions. The dependences of the height of the temperature maxima and its location (from the center of the channel) on the restitution coefficient are in variance with the related kinetic theory predictions (Tij & Santos, J. Stat. Phys. 2004). For the Poiseuille flow of a rarefied gas, it is known that the mass flow rate decreases with increasing Knudsen number (Kn), reaches a minimum at Kn ∼ O(1) and increases again with further increase in Kn– this is dubbed Knusden minima. In a granular Poiseuille flow we show that this Knudsen minima is absent. The origin of these anomalous behaviour is shown to be tied to dissipation-induced particle-clustering in the channel. 9:57AM M24.00010 The Effect of Particle Size on the Erosion of Lunar Regolith from a Spacecraft Landing , KYLE BERGER, BRENDAN BROWN, Univ of Colorado - Boulder, PHILIP METZGER, NASA Kennedy Space Center, CHRISTINE HRENYA, Univ of Colorado - Boulder — The ejection of regolith from a spacecraft landing on an extraterrestrial body (Moon, Mars, etc.) can be extremely hazardous to anything near or possibly even far from the landing point. Models currently being used to describe this phenomenon use single particle trajectories and thus ignore the effects of inter-particle collisions. We seek to improve those models by incorporating the effects of collisions. We model the system using the discrete element method (DEM), which models the particles individually using Newton’s laws and thus explicitly includes inter-particle collisions. The current study focuses on the effect of particle size, both in monodisperse systems, as well as polydisperse systems using binary and continuous particle size distributions (PSDs). While collisions above the surface are rare in the monodisperse case (about 0.0001% of eroded particles), they are relatively frequent in the binary case, particularly between unlike particle species (about 1-5% of eroded large particles). It is expected that as the size disparity becomes larger, which is the case for lunar regolith as it spans at least three orders of magnitude in size, this effect becomes enhanced. Differences in particle size can result in differences in velocity, leading to interesting phenomena. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M25 Turbulence: Curvature and Flow Instability — 2005 - Leonardo Chamorro, University of Illinois at Urbana-Champaign 8:00AM M25.00001 Parametric study of the perturbation dynamics in cross-flow boundary layer , FRANCESCA DE SANTI, STEFANIA SCARSOGLIO, Department of Mechanical and Aerospace Engineering, Politecnico di Torino, Italy, WILLIAM O. CRIMINALE, Department of Applied Mathematics, University of Washington, Seattle, WA, DANIELA TORDELLA, Department of Mechanical and Aerospace Engineering, Politecnico di Torino, Italy — A parametric study on the life of small perturbations acting on a three-dimensional boundary layer in cross flow is discussed. Five parameters are considered: - Reynolds number, - angle of cross flow, - external pressure gradient, - wavelength and - waveangle. We use almost optimal initial conditions and solve an initial value problem to obtain information on the initial transient and temporal long-term behavior. A good agreement with other studies is observed (Breuer & Kuraishi 1994, Corbett & Bottaro 2001). Results concern the substantial role that asymptotically stable perturbations with a transient growth could have in triggering non-linear processes that lead to transition to turbulence. In fact for positive external pressure gradients a subset of waves those have an intense transient growth show phase speeds directed upstream. It is then possible that wave packets made with such waves propagate upstream and contaminate the line of attack of the flow on the wing. We also computed the perturbed pressure field near the wall. In general, when a transient growth occurs, the perturbed pressure reaches its maximum value in significant advance in respect to the perturbed kinetic energy. But, for accelerated external flow and long wave perturbations, the opposite occurs. 8:13AM M25.00002 Development of second mode instability in a Mach 6 flat-plate boundarylayer with two-dimensional roughness , QING TANG, CHUANHONG ZHANG, CUNBIAO LEE, None — The PCB pressure sensors and particle image velocimetry (PIV) are used to study the development of the second mode instability in a Mach 6 flow over a flat plate with two-dimensional roughness. A two-dimensional transverse wall blowing is used to enhance the second mode instability in the boundary layer and seeding tracer particles for PIV measurement. Three roughness elements with different heights are mounted at 125mm downstream the leading edge of the flat plate. It is proved that two-dimensional roughness could enhance the second mode fluctuation upstream the roughness. The second mode instability waves in flat-plate boundary layer are clearly shown by PIV and the boundary layer separation zone upstream the roughness is carefully measured. The boundary layer then reattaches the wall and the second mode instability waves are found damping downstream the roughness. It is also proved that the amplification and damping effect of the second mode instability waves depend on the height of the roughness. 8:26AM M25.00003 Tollmien-Schlichting Wave Cancellation by Feedback Control1 , SH SANKARASARMA VEMURI, JONATHAN MORRISON, ERIC KERRIGAN, Imperial College London — Tollmien-Schlichting (TS) waves are primary instabilities in the boundary layer and by actively interfering with these naturally occurring waves, the transition could be delayed. The present active cancellation scheme involves a feedback control loop between an array of sensors and actuators to generate the desired actuation to attenuate the growth of these spatially evolving unstable waves. Experimental results of growing TS-waves from a point source on a flat-plate model will be presented. Numerical calculations based on linear stability theory have been carried out to predict the evolution of TS waves downstream to model the control system. We will present the spatial transfer functions between the excitation source and sensors and those between the actuators and sensors for various sensor-actuator configurations. A H∞ optimal controller is designed for each of these configurations to obtain an optimal sensor-actuator configuration and the controller will be implemented to attempt real-time cancellation of TS waves on a flat-plate model using these optimal configurations. 1 This work is supported by Airbus 8:39AM M25.00004 Edge states and the spatio-temporal transition dynamics in a boundary layer driven by free stream turbulence , BRUNO ECKHARDT, TOBIAS KREILOS, Philipps-Universitat Marburg, TARAS KHAPKO, PHILIPP SCHLATTER, Linne Flow Center, KTH Mechanics, Royal Institute of Technology, Stockholm, YOHANN DUGUET, LIMSI-CNRS, Universite Paris-Sud, Orsay, DAN S. HENNINGSON, Linne Flow Center, KTH Mechanics, Royal Institute of Technology, Stockholm — We present a cellular automaton model for the transition to turbulence in a boundary layer exposed to free stream turbulence. The model is based on the presence of an invariant flow structure (aka “edge state”) intermediate between laminar and turbulent (Phys. Rev. Lett. 108, 044501 (2012)) and replaces the complex initiation of turbulence by a stochastic process with parameters that are related to the properties of the free stream turbulence and the edge state. The model uses a discretization of space and time and includes spanwise and streamwise spreading of a turbulent nucleus. It reproduces the downstream variation of the nucleation rate, the intermittency factor, and the number and widths of turbulent spots, including their variation with the free stream turbulence intensity. The model thus connects the observed characteristics of boundary layer transition with the transition scenario that has been developed for parallel shear flows, such as pipe flow or plane Couette flow. 8:52AM M25.00005 Stability analysis of the experimental and the simulated flow past miniature vortex generators in a Blasius boundary layer1 , SIMONE CAMARRI, LORENZO SICONOLFI, Università di Pisa, JENS H.M. FRANSSON, Linne Flow Centre, KTH Mechanics — It is shown in the literature that Tollmien-Schlichting (TS) waves can be damped and transition delayed if properly shaped modulations of the streamwise velocity (streaks) are generated inside a Blasius boundary layer. In [1] velocity streaks are generated experimentally by installing miniature vortex generators (MVGs) on the plate wall so as to obtain a significant streak amplitude where the uncontrolled flow would be convectively unstable. When TS waves are excited upstream with respect to the MVGs, they undergo an amplification in the near wake past the MVGs and, if the streaks amplitude is sufficiently high, they decay further downstream, delaying transition. In order to investigate this behavior, representative experimental cases among those documented in [1] are selected and simulated by DNS, and local bi-global stability analysis is applied both to the experimental and to the DNS flow fields. As a result, stability curves for the BL with MVGs are computed and compared to that of an uncontrolled Blasius BL. It is shown that available experimental results agree with the computed stability curves and results from the stability analysis are used to investigate the involved stabilization mechanisms. [1] Shahinfar et al., Phys. Rev. Lett. 109, 074501, (2012) 1 PRACE is acknowledged for awarding us access to resource FERMI based in Italy at CINECA. 9:05AM M25.00006 Accelerated Transonic Flow past a curvature discontinuity , THOMAS DE COINTET, ANATOLY RUBAN, Imperial College London — The aim of this talk is to investigate High Reynolds number Transonic flow past a discontinuity in body curvature. Starting with the inviscid flow outside the boundary layer, our analysis will focus on the flow in a vicinity of the point of discontinuity, where a solution of the Euler equations will be sought in self-similar form. This reduces the Euler equations to an ordinary differential equation. The analysis of this equation shows that the pressure gradient on the airfoil surface develops a strong singularity, which is proportional to (x0 − x)−1/3 as the discontinuity point x0 is approached. We then study the response of the boundary layer to this extremely favourable pressure gradient. We show that the boundary layer splits into two parts, the main body of the boundary layer that becomes inviscid on approach to the singularity, and a thin viscous sublayer situated near the wall. The analysis of the behaviour of the solution in the viscous sublayer shows that Prandtl’s hierarchical concept breaks down in a small region surrounding the singular point, where the viscous-inviscid interaction model should be used. In the final part of this talk we present a full formulation of the viscous-inviscid interaction problem and discuss numerical results. 9:18AM M25.00007 Effect of curvature modulation on Gortler vortices in boundary layers1 , HUI XU, Department of Aeronautics, Imperial College London, PHILIP HALL, Department of Mathematics, Imperial College London, SPENCER SHERWIN, Department of Aeronautics, Imperial College London — The stability of a high-Reynolds-number flow over a curved surface with varying curvature is studied. The investigation is concentrated on spanwise-periodic vortices of wavelength comparable with the boundary layer thickness. Motivated by the amendment of Rayleigh’s criterion (Hall,2013), the effect of wavy-wall modulation on Gortler vortices is addressed. Both linear and nonlinear investigations are performed to understand the destabilization and stabilization mechanisms of the vortices. Furthermore, due to the wavy-wall curvature modulation, the growth or decay rate of the vortices is discussed. Finally, a control strategy of the vortices is proposed based on distributing the curvature. 1 This research was performed in the Laminar Flow Control Centre (LFC-UK) at Imperial College London. The Centre is supported by EPSRC, Airbus UK and EADS Innovation Works. 9:31AM M25.00008 Flow past a hump in subsonic and transonic regimes: Comparisons between triple deck theory and DNS1 , GIANMARCO MENGALDO, MARINA KRAVTSOVA, ANATOLY RUBAN, SPENCER SHERWIN, Imperial College London — The prediction of laminar-turbulent transition is a key factor for reducing the drag and for improving the aerodynamic performance of an aircraft. In the past few years several studies, theoretical, numerical and experimental, have been conducted on roughness elements and isolated humps in order to investigate their role in the transition process. Many comparisons already exist between numerical and experimental data while little work has been carried out in comparing theoretical and numerical results. In this work we present a comparative study between triple deck theory and DNS. Specifically we consider a flat plate with a hump of various heights in a compressible regime at a relatively high Reynolds number. Different Mach numbers are taken into account, ranging from subsonic to transonic regimes and various temperatures are applied to the wall for each Mach number considered. The main questions we aim to answer in this work are the following: • Are triple deck and DNS data comparable and how can we perform this comparison? • Which are the limits and the advantages of the first (triple deck) and the second (DNS) approach for the simple test case under investigation? 1 Work supported by LFC-UK Centre under grant EP/I037946 9:44AM M25.00009 Flow-field characterization over 1D and 2D periodic wavy walls using PIV in a refractive-index-matched channel , LEONARDO P. CHAMORRO, A.M. HAMED, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Chamapign, CARLO ZUNIGA ZAMALLOA, University of Illinois at Urbana-Chamapign — Flow over two wavy walls was experimentally investigated using high- and low-frame-rate particle image velocimetry (PIV). The first wall has 1D streamwise waves with amplitude-towavelength ratio a/λ = 0.05. The second wall has streamwise and spanwise waves with a/λ = 0.05 and 0.025, respectively. A refractive-index matching approach was used to minimize image distortion and reflections. It grants unobstructed optical access and allows for very near-wall velocity measurements. Flow-field measurements were acquired at multiple streamwise-wall-normal and wall-parallel planes. The low-frame-rate measurements were used to obtain high-resolution ensemble-averaged flow fields and turbulence statistics, while the high-frame-rate measurements were used to map the structure of the turbulence at various wall-normal locations and to determine scale-dependent correlations across the topological features of the walls. The results were studied to understand the link between the turbulence structure and wall undulations. Linking the turbulence to wall topology has many implications in environmental flows, sediment transport and advection diffusion of scalars. 9:57AM M25.00010 Turbulence structure over 1D and 2D periodic wavy walls: Coupling between coherent motions and large-scale undulations , NICOLAS TOBIN, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, PRANAV SURESH, Department of Ocean Engineering, IIT Madras, India, PRATAP VANKA, LEONARDO P. CHAMORRO, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — Understanding the turbulence dynamics over topographies with mild perturbations is of great relevance at geophysical scale and on a number of other wall-bounded flow phenomena. Turbulence statistics and its spectral distribution in the boundary layer are heavily modulated by the topological features of walls in a very complex fashion. In this study we aim to understand some of the basic processes modulating the interaction of near-wall turbulence and large-scale mild perturbations superimposed to smooth walls. Large Eddy Simulations (LES) of channel flows with 1D and 2D periodic wavy surfaces are performed at a Reynolds number of 104 based on the channel depth. The computational domain spans 10λ (where λ is the wavelength) in the streamwise direction and 4λ and 1λ in the spanwise and vertical directions respectively. The sinusoidal waves have amplitude of 0.1λ. Turbulence structures and high-order statistics as well as features of energetic coherent motions are discussed in terms of the wall topology. The effect of the wall shape on the flow is examined through the pre-multiplied spectral features of the turbulence at key locations. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M26 Turbulent Boundary Layers V — 2007 - Tamer A. Zaki, Johns Hopkins University 8:00AM M26.00001 Reduced-order model for near-wall dynamics with implications to wallmodels , PETER SCHMID, TARANEH SAYADI, Department of Mathematics, Imperial College London — The near-wall resolution requirements of wall-resolved large eddy simulations (LES) are almost as high as those of direct numerical simulations (DNS). This restriction severely limits the applicability of LES in high-Reynolds-number flows and complex geometries that are typical of engineering configurations. An alternative to the wall-resolved LES is the wall-modeled simulation, where the resolution requirement is relaxed by prescribing wall-stresses in the vicinity of walls. One such way of providing accurate values of wall-stresses is based on optimal flow-control techniques. In this study we propose models to extend the terminology of predictive control-based wallmodels to complex geometries, by defining transfer functions relating the mean velocity to the second moments at an optimal planar location. As a result the added calculation in the near-wall region (for example RANS) will be omitted and replaced by boundary conditions described by pre-existing transfer functions. The relevant transfer functions are extracted using a data-driven as well as model-based approach. The predicted transfer functions are then compared to their system-identified equivalent for verification. 8:13AM M26.00002 A nested-LES wall-modeling approach for computation of high Reynolds number equilibrium and non-equilibrium wall-bounded turbulent flows , YIFENG TANG, RAYHANEH AKHAVAN, The University of Michigan, Ann Arbor — A nested-LES wall-modeling approach for high Reynolds number, wall-bounded turbulence is presented. In this approach, a coarse-grained LES is performed in the full-domain, along with a nested, fine-resolution LES in a minimal flow unit. The coupling between the two domains is achieved by renormalizing the instantaneous LES velocity fields to match the profiles of kinetic energies of components of the mean velocity and velocity fluctuations in both domains to those of the minimal flow unit in the near-wall region, and to those of the full-domain in the outer region. The method is of fixed computational cost, independent of Reτ , in homogenous flows, and is O(Reτ ) in strongly non-homogenous flows. The method has been applied to equilibrium turbulent channel flows at 1000 ≤ Reτ ≤ 10000 and to non-equilibrium, shear-driven, 3D turbulent channel flow at Reτ ≈ 2000. In equilibrium channel flow, the friction coefficient and the one-point turbulence statistics are predicted in agreement with Dean’s correlation and available DNS and experimental data. In shear-driven, 3D channel flow, the evolution of turbulence statistics is predicted in agreement with experimental data of Driver & Hebbar (1991) in shear-driven, 3D boundary layer flow. 8:26AM M26.00003 Investigation of a Wall Shear-Stress Inner-Outer Interaction Model for Large-Eddy Simulations , WILLIAM SIDEBOTTOM, OLIVIER CABRIT, IVAN MARUSIC, The University of Melbourne, CHARLES MENE- VEAU, Johns Hopkins University, ANDREW OOI, The University of Melbourne, DAVID JONES, Defence Science and Technology Organisation — The very small turbulent motions in the thin layer of fluid immediately adjacent to a solid surface in a turbulent boundary layer make it difficult to effectively scrutinise the near-wall dynamics with physical and numerical experiments. These near-wall turbulent motions, and the no-slip condition, directly affect the tangential stress at the surface–the wall shear-stress. This study investigates a new wall-model for large-eddy simulations capable of predicting the fluctuating wall shear-stress from a large-scale velocity input, without the need to fully resolve the smallest structures in the flow. The model is based on the spectral structure of the turbulent boundary layer and the interaction between large-scale events in the logarithmic layer and small-scale events near the wall. Various methods have previously been used to predict the mean wall shear-stress with sufficient accuracy. There are, however, very few models available to predict the fluctuating component. Results from the new wall-model show that it has only a small effect on mean quantities, such as the skin-friction coefficient, but is able to resolve more of the wall shear-stress variance than a “standard” wall-model. 8:39AM M26.00004 Inference of the turbulent dissipation rates in wall-bounded turbulent flows1 , SUBHAS VENAYAGAMOORTHY, FARID KARIMPOUR, Colorado State University — Accurate prediction of the dissipation rate of the turbulent kinetic energy (ǫ) in turbulent flows is fundamental for modeling of such flows. However, measuring the dissipation rate of the turbulent kinetic energy has always been a challenge in laboratory experiments, especially near the wall. The focus of this study is to investigate and predict the dissipation rate of the turbulent kinetic energy (ǫ) in fully developed wall-bounded turbulent flows. To this end, new parameterizations for the mixing length (Lmix = (−u′ v ′ )1/2 /S) in fully developed wall-bounded turbulent flows are proposed and their relationship with the dissipation rate of the turbulent kinetic energy is investigated. Comparisons with different datasets of direct numerical simulation of canonical wall-bounded turbulent flows show remarkable agreement. These findings could be useful for the prediction of ǫ in wall-bounded turbulent flows, especially in the highly anisotropic near-wall region. 1 Funded by the National Science Foundation 8:52AM M26.00005 Validation of Reynolds Stress Transport Models with Velocity/PressureGradient Models in Wall- Bounded Flows1 , JUAN D.C. FERNANDEZ, SVETLANA POROSEVA, University of New Mexico, SCOTT MURMAN, NASA Ames Research Center — In the traditional formulation of Reynolds Stress Transport (RST) turbulence models, velocity/pressuregradient correlations are decomposed into pressure-strain correlations and pressure diffusion terms that are modeled separately. In our study, a potential of a different modeling approach for improving simulation results in the near-wall area is investigated. No decomposition of velocity/pressure-gradient correlations is attempted. New linear models for such correlations have been recently developed and successfully validated against DNS data in two-dimensional incompressible turbulent flows such as a zero-pressure gradient boundary layer over a flat plate and a fully-developed channel flow. The models correctly reproduce DNS profiles of velocity/pressure-gradient correlations up to the wall with the same model coefficients in different geometries and at different Reynolds numbers. These models are currently implemented in transport equations for Reynolds stresses. The compatibility of models for such correlations with existing models for the dissipation tensor and turbulent diffusion is investigated. Simulations are conducted with open-source software OpenFOAM and in-house code in two-dimensional wall-bounded flows. 1A part of the material is based upon work supported by NASA under award NNX12AJ61A. 9:05AM M26.00006 Space-time characteristics of wall-pressure fluctuation in wall-modeled large-eddy simulation , GEORGE ILHWAN PARK, PARVIZ MOIN, Stanford University, Center for Turbulence Research — Assessment of wallmodeled large-eddy simulation (WMLES) has always been based on the prediction quality of the mean velocity and Reynolds stresses. Secondary quantities from WMLES, such as wall pressure/stress fluctuations and their spectra received little attention, and they are usually not reported. Since they are directly related to the structural vibration and noise generation from the immersed bodies, identifying to what extent the near-wall pressure/stress field from WMLES can be utilized for the modeling purpose is of great importance. Here the r.m.s. and space-time characteristics of wall pressure/stress fluctuation obtained from WMLES are reported and analyzed for the first time. WMLES of a high Reynolds number turbulent channel flow at Reτ = 2000 by Park and Moin [Phys. Fluids 26, 015108, (2014)] is considered for this purpose. The r.m.s wall-pressure fluctuation in WMLES is generally under-predicted owing to the very coarse near-wall resolution, but improves with the mesh refinement. The convection velocity and wavenumber/frequency spectra of wall-pressure fluctuation show qualitative agreement with low Reynolds number data in the literature. Quantitative comparison to the Reτ = 2000 DNS data will hopefully be presented in the meeting. 9:18AM M26.00007 Liquid Jet Impingement Thermal Transport on a Superhydrophobic Surface1 , MATTHEW SEARLE, DANIEL MAYNES, JULIE CROCKETT, Brigham Young Univ - Provo — Thermal transport for an axisymmetric liquid jet impinging on a horizontal constant temperature superhydrophobic surface with an imposed isotropic hydrodynamic slip length and temperature jump length has been explored analytically. The flow is partitioned into three regions: 1) a region where the hydrodynamic and thermal boundary layers are developing, 2) a region where the hydrodynamic boundary layer is developed and the thermal boundary layer is still developing, and 3) a region where both boundary layers are developed throughout the thin film. An integral analysis has been performed, where third-order velocity and temperature profiles have been assumed. A system of differential equations are solved numerically to obtain boundary layer thicknesses, local shear stress and heat flux, thin film height, and free surface temperature as functions of radial position. The solution for the no-slip scenario shows excellent agreement with previous differential analysis of the same problem. The influence of the magnitude of the slip length and temperature jump length on the thermal transport is presented for a realizable range of slip lengths and typical jet Reynolds numbers. 1 NSF CBET-1235881 9:31AM M26.00008 Drag Reduction with Super-Hydrophobic Surfaces in Turbulent Channel Flow , AMIRREZA RASTEGARI, RAYHANEH AKHAVAN, University of Michigan, Ann Arbor — Drag reduction (DR) with super-hydrophobic (SH) surfaces is investigated using DNS in turbulent channel flow with SH walls. Both channel walls were covered with longitudinal arrays of SH micro-grooves of width g separated by distances of w. The liquid/gas interfaces on these walls were modeled as idealized, flat, shear-free boundaries. DRs of 5 − 83% were obtained with 4 ≤ g +0 ≤ 128 and g/w = 1, 7 & 15 at Rebulk = 3600. By analysis of the Navier-Stokes equations, it is shown that the magnitude of DR is given by DR = Uslip /Ubulk + O(ε), where Uslip /Ubulk represents the contribution of surface slip to DR, and the O(ε) term represents DR arising from other sources, such as modifications to turbulence dynamics. Comparison with DNS results shows surface slip to be the ‘dominant’ mechanism of DR even in turbulent flows, and responsible for over 80% of the DR in both the high and low DR regimes. The effect of the SH surface on the dynamics of turbulence is found to be small and confined to additional production of turbulence kinetic energy within a thin surface layer of size on the order of the width of the surface micro-grooves. Beyond this effect, the normalized dynamics of turbulence proceeds as with no-slip walls. 9:44AM M26.00009 Bio-inspired Gecko Micro-surface for Drag Reduction in Turbulent Flows1 , ISNARDO ARENAS, KENNETH CARRASQUILLO, The University of Texas at Dallas, GUILLERMO ARAYA, LUCIANO CASTILLO, Texas Tech University, STEFANO LEONARDI, The University of Texas at Dallas — Direct Numerical Simulations of a turbulent channel flow with a porous wall inspired from the Gecko lizard were performed at Reynolds number of Reτ = 450. Two superposed fluids were considered. As initial condition, one fluid fills the microfibrillar surface, the interface with the overlying fluid being flat and corresponding to the crests plane. The code is based on a finite difference scheme with a Runge Kutta and fractional step. The porous wall is modeled with the immersed boundary method, while the dynamic of the interface between the two fluids is solved with a level set method. A parametric study has been performed varying the viscosity ratio between the two fluids. Two cases have been considered, with and without surface tension. Without surface tension the microfibrillar wall acts as a rough wall increasing the drag. However, when the surface tension is large enough to maintain the interface stable, the external fluid cannot enter into the porous wall and an effective slip is produced. When the fluid in the porous wall has a viscosity 100 times smaller than that of the overlying fluid, a drag reduction of about 60% can be observed. In this case, the near wall coherent structures become significantly weaker. 1 The numerical simulations were performed on XSEDE TACC under Grant No. CTS070066. SL, IA and KC were supported by ONR MURI grant. 9:57AM M26.00010 Turbulent Taylor-Couette flow over liquid infused surfaces1 , BRIAN ROSENBERG, ALEXANDER SMITS, Princeton University — We experimentally study the flow of turbulent water over a textured surface that is impregnated with a second immiscible liquid. Two configurations are studied: (i) a configuration in which the impregnating fluid is contained within the texture, so that the turbulent flow sees a composite liquid/solid surface and (ii) a configuration in which the impregnating fluid overlies the texture. Experiments are performed in turbulent Taylor-Couette flow at a friction Reynolds number around 150. We characterize the impact the liquid infused surfaces have on the skin friction as well as the critical Reynolds number for transition to turbulence. Particular attention is focused on the influence of the texture geometry and length scale as well as the impregnating fluid properties. 1 This work was supported by the Office of Naval Research under MURI grant number N00014-12-1-0875 (Program Director Dr. Ki-Han Kim). Tuesday, November 25, 2014 8:00AM - 10:10AM Session M27 Turbulence Theory: Measurements — 2009 - Lester Su, Stanford University 8:00AM M27.00001 Full 3D derivative moments from an L-shaped SPIV experiment in the near wall region , JEAN-MARC FOUCAUT, Ecole Centrale de Lille, CHRISTOPHE CUVIER, University of Lille, MICHEL STANISLAS, Ecole Centrale de Lille, LABORATOIRE DE MÉCANIQUE DE LILLE TEAM — Stereoscopic PIV is now a relevant method to measure turbulent flow. This method allows the measurement of the three components of the velocity in a plane with an accuracy of about 1-2%. For turbulent flows usually only the large scale motions are investigated due to the limited spatial resolution of the PIV. The main difficulty comes when the derivative has to be computed due to the noise (Foucaut, 2002). The present communication proposes a method to determine the derivative moments which combines both the derivative and the statistic computation from a specific SPIV experiment. Balint et al (1991) published experimental results of derivative moments obtained by HWA in a turbulent boundary layer. A special multiwire probe was designed for this measurement. These results are globally of the same order as the DNS (Antonia et al. 1991), except for the derivative of the streamwise component for which Taylor’s hypothesis was used. This was not necessary with stereoscopic PIV. The experiment was performed in the LML 20 m boundary layer facility to determine all the derivative moments needed to determine the dissipation. The Reynolds number was Reθ = 7500. Measurements were taken in two normal planes in order to compute all the derivatives of the three components. For the PIV there is a trade-off between field of view and the interrogation window sizes, so the derivative filter choice and the measurement noise management are particularly discussed. 8:13AM M27.00002 Results of experimental investigation of the dissipation rate in the near wall region , WILLIAM GEORGE, Princeton University, JEAN-MARC FOUCAUT, Ecole Centrale de Lille, CHRISTOPHE CUVIER, University of Lille, MICHEL STANISLAS, Ecole Centrale de Lille, LABORATOIRE DE MÉCANIQUE DE LILLE TEAM, PRINCETON — The E D COLLABORATION Ei hD UNIVERSITY present idea is to propose a method to determine the dissipation rate from a specific SPIV experiment: ε = ν ∂ui ∂ui ∂xj ∂xj + ∂ui ∂uj ∂xj ∂xi is strongly linked to the small scales of a turbulent flow. It is indispensable for turbulence modelling. Yet it has seldom been measured. To obtain the full dissipation it is necessary to measure the full instantaneous gradient tensor and to compute the 15 moments. But, all are difficult to measure accurately. George and Hussain (1991) proposed to simplify this computation by using different hypotheses in order to reduce the number of terms. Their hypotheses were local homogeneity and local axisymmetry, both of which are more general than the usual assumptions of local isotropy. An experiment was performed in the LML boundary layer facility to determine all the necessary derivative moments. A detailed analysis of the errors in derivative measurements was carried out (see Foucaut et al. 2014 APS), as well as applying and using consistency checks derived from the continuity equation and local homogeneity. Local homogeneity estimates of the dissipation are accurate everywhere to within a few percent. Both local axisymmetry and local isotropy work almost as well outside of y+ = 100, but only local axisymmetry provides a reasonable estimate closer to the wall. The results are remarkably similar to those of Antonia et al (1991) from DNS results of a channel flow at low Reynolds number. 8:26AM M27.00003 Direct observation of energy cascade in three-dimensional turbulence , HAITAO XU, FABIO DI LORENZO, Max Planck Institute for Dynamics and Organization (MPIDS), Goettingen, Germany, ALAIN PUMIR, Max Planck Institute for Dynamics and Organization (MPIDS), Goettingen, Germany, and ENS-Lyon, Lyon, France, EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Organization (MPIDS), Goettingen, Germany — In three-dimensional turbulence, energy is supplied at large scales and cascaded down to smaller and smaller scales. The energy flux can be measured by, e.g., the velocity structure functions. On the other hand, the temporal process of the energy cascade, such as how fast it takes for energy at the forcing scales to transfer down to the dissipative scales, has received relatively little attention. Using novel laboratory turbulent flows and measurement techniques, we experimentally studied the response of turbulence in the inertial and dissipative scales to a sudden excess of energy in the forcing scales. Our measurements give us direct access to the energy cascade process. We also compare our observation with results from direct numerical simulations. 8:39AM M27.00004 Evolution of turbulent kinetic energy in the presence of a uniform kinetic energy gradient without mean shear1 , ADRIEN THORMANN, CHARLES MENEVEAU, Johns Hopkins University — In this work we study grid turbulence with a initial uniform spatial gradient of kinetic energy of the form k ∼ β(y − y0 ), where y is the spanwise position, while having no mean-velocity shear. Therefore, there is no production but only dissipation and spatial transverse diffusion of turbulent kinetic energy. The experiment is performed with the use of an active grid and screens mounted upstream of the wind-tunnel’s test section, iteratively designed to produce a uniform gradient of turbulent kinetic energy without mean velocity shear. Data are acquired using X-wire thermal anemometry at different spanwise and downstream locations. Profile measurements are used to quantify the constancy of the mean velocity and the linearity of the initial profile of kinetic energy. Measurements show that at all spanwise locations the decay in the streamwise direction follows a power-law but with exponents n(y) that depend upon the spanwise location. The results are consistent with a parameterization of decay of the form k/hui2 = β(x/xref )−n(y) (y − y0 )/M . Results for the development of the integral length scale, and for velocity skewness and flatness factors, which show significant deviations from Gaussianity, are also presented. 1 Research supported by NSF (CBET and CMMI), and Sardella chair at JHU. 8:52AM M27.00005 Comparison of fractal and classical grids with the same blockage , R. JASON HEARST, PHILIPPE LAVOIE, Univ of Toronto — Recently, the field of canonical grid turbulence has been reenergized by measurements in the wake of fractal grids. Fractals have produced turbulence that decays more rapidly than traditional grid turbulence experiments. In addition, in the wake of fractals, the normalized dissipation scaling, Cǫ , appears to grow rapidly, in sharp contrast with traditional expectations that Cǫ ∼ constant. In the present study, we compare a square-fractal-element grid, composed of a 12 × 8 mesh of small square fractal elements, to two regular grids with the same blockage, and similar mesh lengths and thicknesses. The same grid Reynolds number is used so that the results in the wake of the grids are comparable. We also employ a secondary contraction to marginalize anisotropy as a contributing factor to differences in the decay. Ultimately, we demonstrate that in the far-field the turbulence produced by all three grids is similar. However, the development region in the wake of the fractal is extended relative to the classical grids. One of the major conclusions of the present study is that certain classical grid configurations are able to produce higher turbulence intensities and Reynolds number than a fractal for the same blockage. 9:05AM M27.00006 Experimental Study of Homogeneous Isotropic Slowly-Decaying Turbulence in Giant Grid-Wind Tunnel Set Up , ALBERTO ALISEDA, LEGI, MICKAEL BOURGOIN, LEGI-CNRS, ESWIRP COLLABORATION1 — We present preliminary results from a recent grid turbulence experiment conducted at the ONERA wind tunnel in Modane, France. The ESWIRP Collaboration was conceived to probe the smallest scales of a canonical turbulent flow with very high Reynolds numbers. To achieve this, the largest scales of the turbulence need to be extremely big so that, even with the large separation of scales, the smallest scales would be well above the spatial and temporal resolution of the instruments. The ONERA wind tunnel in Modane (8 m-diameter test section) was chosen as a limit of the biggest large scales achievable in a laboratory setting. A giant inflatable grid (M=0.8 m) was conceived to induce slowly-decaying homogeneous isotropic turbulence in a large region of the test section, with minimal structural risk. An international team or researchers collected hot wire anemometry, ultrasound anemometry, resonant cantilever anemometry, fast pitot tube anemometry, cold wire thermometry and high-speed particle tracking data of this canonical turbulent flow. While analysis of this large database, which will become publicly available over the next 2 years, has only started, the Taylor-scale Reynolds number is estimated to be between 400 and 800, with Kolmogorov scales as large as a few mm. 1 The ESWIRP Collaboration is formed by an international team of scientists to investigate experimentally the smallest scales of turbulence. It was funded by the European Union to take advantage of the largest wind tunnel in Europe for fundamental research. 9:18AM M27.00007 Development and characterization of a Nano-scale temperature probe (TNSTAP) for turbulent temperature measurement , GILAD ARWATZ, YUYANG FAN, CARLA BAHRI, Princeton University, ALEXANDER J. SMITS, Princeton University and Monash University, MARCUS HULTMARK, Princeton University — A new nano-scale temperature probe (T-NSTAP) is presented. The T-NSTAP consists of a miniature, free-standing, platinum wire suspended between silicon supports. The sensor is designed for temperature measurements at high frequencies, operated in constant current mode. The design is based on the cold-wire model proposed by Arwatz et al. (2013) and is shown to have a bandwidth far superior that of conventional cold-wires. This minimizes the effect of temporal filtering as well as spatial filtering on the data and allows for a unique investigation of the full scalar spectrum, including the dissipation range. Data is acquired in a heated grid turbulence setup with constant mean temperature gradient using both the T-NSTAP and a conventional cold wire. It is shown that the cold wire is significantly attenuated over the full range of frequencies including low frequencies with a direct effect on the temperature variance and the scalar rate of dissipation. The model of Arwatz et al (2013) is used to correct the cold wire data and it is shown that the correction works well over the entire spectrum. In addition, the corrected data agrees closely with the T-NSTAP measurements. 9:31AM M27.00008 Scaling of spectra in grid turbulence with a mean cross-stream temperature gradient , CARLA BAHRI, GILAD ARWATZ, MICHAEL E. MUELLER, Princeton University, WILLIAM K. GEORGE, Imperial College London, MARCUS HULTMARK, Princeton University — Scaling of grid turbulence with a constant mean cross-stream temperature gradient is investigated using a combination of theoretical predictions, DNS, and experimental data. Conditions for self-similarity of the governing equations and the scalar spectrum are investigated, which reveals necessary conditions for self-similarity to exist. These conditions provide a theoretical framework for scaling of the temperature spectrum as well as the temperature flux spectrum. One necessary condition, predicted by the theory, is that the characteristic length scale describing the scalar √ spectrum must vary as ∝ t for a self-similar solution to exist. In order to investigate this, T-NSTAP sensors, specially designed for temperature measurements at high frequencies, were deployed in a heated passive grid turbulence setup together with conventional cold-wires, and complementary DNS calculations were performed to complement and complete the experimental data. These data are used to compare the behavior of different length scales and validate the theoretical predictions. 9:44AM M27.00009 On the Degeneration of Turbulence at High Reynolds Numbers , GREGORY BEWLEY, MICHAEL SINHUBER, EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Self-Organization — Turbulent fluctuations in a fluid wind down at a certain rate once stirring has stopped. The role of the most basic parameter in fluid mechanics, the Reynolds number, in setting this decay rate is not generally known. This talk concerns the high-Reynolds-number limit of the process. In a wind-tunnel experiment that reached higher Reynolds numbers than ever before and covered more than two decades in the Reynolds number (104 < Re = U M/ν < 5 × 106 ), we measured the decay rate with the unprecedented precision of about 2%. Here U is the mean speed of the flow, M the forcing scale, and ν the kinematic viscosity of the fluid. We observed that the decay was Reynolds number independent, which contradicts some models and supports others. 9:57AM M27.00010 Time-reversal-symmetry breaking in turbulence1 , JENNIFER JUCHA, HAITAO XU, Max Planck Institute for Dynamics and Self-Organization (MPIDS), Goettingen, Germany, ALAIN PUMIR, Max Planck Institute for Dynamics and Self-Organization (MPIDS), Goettingen, Germany, and ENS-Lyon, Lyon, France, EBERHARD BODENSCHATZ, Max Planck Institute for Dynamics and Self-Organization (MPIDS), Goettingen, Germany — In three-dimensional turbulent flows, the flux of energy from large to small scales breaks time symmetry. We show here that this irreversibility can be quantified by following the relative motion of several Lagrangian tracers. We find by analytical calculation, numerical analysis and experimental observation that the existence of the energy flux implies that, at short times, two particles separate temporally slower forwards than backwards, and the difference between forward and backward dispersion grows as t3 . We also find the geometric deformation of material volumes, surrogated by four points spanning an initially regular tetrahedron, to show sensitivity to the time-reversal with an effect growing linearly in t. We associate this with the structure of the strain rate in the flow. 1 We thank the support from Max Planck Society, the Humboldt Foundation, ANR, and PSMN at ENS-Lyon. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M28 Turbulence: Theory III — 2011 - Rene Pecnik, Delft University of Technology 8:00AM M28.00001 Numerical experiments of variable property turbulent channel flow , ASHISH PATEL, JURRIAAN PEETERS, BENDIKS BOERSMA, RENE PECNIK, Process and Energy Department, Delft University of Technology — We perform numerical experiments of turbulent channel flows with varying density and viscosity to investigate the validity of semi-local scaling as proposed by Huang, Coleman and Bradshaw (1995, J. Fluid Mech). Direct numerical simulations of the low Mach number approximation of the Navier-Stokes equations are used, whereby the fluid is internally heated and the temperature at the walls is set to constant. A pseudo-spectral discretization in the periodic directions and a 6th order compact finite difference in wall normal direction is used. The friction Reynolds number based on half channel height and wall friction velocity is Reτ = 395. Different relations for density and viscosity as a function of temperature are studied. A variable property case has been identified with turbulent statistics that are quasi-similar to constant property turbulence. This case corresponds to the condition when the semi-local scaling is equal to the classical scaling. For cases wherein the semi-local scaling differs from classical scaling in the channel core, we show that the near-wall turbulence deviates towards a state of increased/decreased anisotropy as compared to constant property turbulence. The above results show not only the validity but also the usefulness of the semi-local scaling. 8:13AM M28.00002 Energy transfer and dissipation in forced isotropic turbulence1 , MORITZ LINKMANN, University of Edinburgh, W. DAVID MCCOMB, Retired, ARJUN BERERA, University of Edinburgh, SAMUEL YOFFE, University of Strathclyde — A model for the Reynolds number dependence of the dimensionless dissipation rate Cε is derived from the dimensionless Kármán-Howarth equation, resulting in Cε = Cε,∞ + C/RL , where RL is the integral scale Reynolds number. The coefficients C and Cε,∞ arise from asymptotic expansions of the dimensionless second- and third-order structure functions. The model equation is fitted to data from direct numerical simulations (DNS) of forced isotropic turbulence for integral scale Reynolds numbers up to RL = 5875 (Rλ = 435), which results in an asymptote for Cε in the infinite Reynolds number limit Cε,∞ = 0.47 ± 0.01. Since the coefficients in the model equation are scale-dependent while the dimensionless dissipation rate is not, we modelled the scale dependences of the coefficients by an ad hoc profile function such that they cancel out, leaving the model equation scale-independent, as it must be. The profile function was compared to DNS data to very good agreement, provided we restrict the comparison to scales small enough to be well resolved in our simulations. 1 This work has made use of the resources provided by the UK supercomputing service HECToR, made available through the Edinburgh Compute and Data Facility (ECDF). A. B. is supported by STFC, S. R. Y. and M. F. L. are funded by EPSRC. 8:26AM M28.00003 On the distribution of local dissipation scales in turbulent flows , IAN MAY, KHANDAKAR MORSHED, KARAN VENAYAGAMOORTHY, LAKSHMI DASI, Colorado State University — Universality of dissipation scales in turbulence relies on self-similar scaling and large scale independence. We show that the probability density function of dissipation scales, Q(η), is analytically defined by the two-point correlation function, and the Reynolds number (Re). We also present a new analytical form for the two-point correlation function for the dissipation scales through a generalized definition of a directional Taylor microscale. Comparison of Q(η) predicted within this framework and published DNS data shows excellent agreement. It is shown that for finite Re no single similarity law exists even for the case of homogeneous isotropic turbulence. Instead a family of scaling is presented, defined by Re and a dimensionless local inhomogeneity parameter based on the spatial gradient of the rms velocity. For moderate Re inhomogeneous flows, we note a strong directional dependence of Q(η) dictated by the principal Reynolds stresses. It is shown that the mode of the distribution Q(η) significantly shifts to sub-Kolmogorov scales along the inhomogeneous directions, as in wall bounded turbulence. This work extends the classical Kolmogorov’s theory to finite Re homogeneous isotropic turbulence as well as the case of inhomogeneous anisotropic turbulence. 8:39AM M28.00004 Symmetry-plane model of 3D Euler flows: Mapping to regular systems and numerical solutions of blowup1 , RACHEL M. MULUNGYE, DAN LUCAS, MIGUEL D. BUSTAMANTE, Complex and Adaptive Systems Laboratory, School of Mathematical Sciences, University College Dublin — We introduce a family of 2D models describing the dynamics on the so-called symmetry plane of the full 3D Euler fluid equations. These models depend on a free real parameter and can be solved analytically. For selected representative values of the free parameter, we apply the method introduced in [M.D. Bustamante, Physica D: Nonlinear Phenomena, 240:1092-1099 (2011)] to map the fluid equations bijectively to globally regular systems. By comparing the analytical solutions with the results of numerical simulations, we establish that the numerical simulations of the mapped regular systems are far more accurate than the numerical simulations of the original systems, at the same spatial resolution and CPU time. In particular, the numerical integrations of the mapped regular systems produce robust estimates for the growth exponent and singularity time of the main blowup quantity (vorticity stretching rate), converging well to the analytically-predicted values even beyond the time at which the flow becomes under-resolved (i.e. the reliability time). In contrast, direct numerical integrations of the original systems develop unstable oscillations near the reliability time. We discuss the reasons for this improvement in accuracy, and explain how to extend the analysis to the full 3D case. 1 Supported under the programme for Research in Third Level Institutions (PRTLI) Cycle 5 and co-funded by the European Regional Development Fund. 8:52AM M28.00005 Large-Re asymptotics of the stream-wise normal stress in the ZPG turbulent boundary layer , PETER A. MONKEWITZ, Swiss Federal Institute of Technology, Lausanne, Switzerland, HASSAN M. NAGIB, Illinois Institute of Technology, Chicago, USA — Models for the stream-wise normal stress huui+ in wall-bounded turbulent flows have been proposed that lead to a log-law in the classical overlap layer (and part of the outer layer). Matching to the wall-layer immediately leads to huui+ inner ∼ ln(Re), i.e. to a mixed scaling in the inner layer. While this appears compatible with the observed Re− dependence of the inner peak, it is shown, in the case of the ZPG TBL, to be incompatible + will be presented which are with DNS data and the Reynolds-averaged momentum equation. Matching inner and outer expansions of huui+ in terms of 1/U∞ consistent with experimental data and DNS, and allow extrapolation to infinite Reynolds number. 9:05AM M28.00006 Similarity of Turbulent Energy Scale Budget Equation of a Round Turbulent Jet1 , HAMED SADEGHI, Mr, PHILIPPE LAVOIE, ANDREW POLLARD, Prof — A novel extension to the similarity-based form of the transport equation for the second-order velocity structure function of h(δq)2 i along the jet centreline (see Danaila et al., 2004) has been obtained. This new self-similar equation has the desirable benefit of requiring less extensive measurements to calculate the inhomogeneous (decay and production) terms of the transport equation. According to this equation, the normalized third-order structure function can be uniquely determined when the normalized second-order structure function, the power-law exponent of hq 2 i and the decay rate constants of hu2 i and hv 2 i are available. In addition, on the basis of the current similarity analysis, the similarity assumptions in combination with power-law decay of mean velocity (U ∝ (x − x0 )−1 ) are strong enough to imply power-law decay of fluctuations (hq 2 i ∝ (x − x0 )m ). The similarity solutions are then tested against new experimental data, which were taken along the centreline of a round jet at ReD = 50, 000. For the present set of initial conditions, hq 2 i exhibits a power-law behaviour with m = −1.83. 1 This work was supported by grants from NSERC (Canada). 9:18AM M28.00007 High-order Boundary Behavior and the Incorporation of Spectral Hyperviscosity in Turbulence Models on General Bounded Regions in 3-D , JOEL AVRIN, Dept of Mathematics and Statistics, UNC Charlotte — In a bounded region in 3-D the velocity field u for the Navier-Stokes system satisfies in the no-slip case the familiar condition u = 0 on the boundary. We show further that if the boundary and the forcing data satisfy reasonably general smoothness assumptions then Au = 0 on the boundary as well where A is the Stokes operator (i.e. Au is the divergence-free part of −∇2 u). We apply this result to subgrid-scale modeling by noting that in a number of computational turbulence experiments hyperviscosity has been added to the NS system as an approximation to spectral eddy viscosity, but a rigorous definition of this technique and a qualitative theory for it has been restricted to the idealized case of box regions with periodic boundary conditions imposed on each face. But under the above smoothness assumptions the fact that Au = 0 on the boundary now allows us in the no-slip case to rigorously define adding hyperviscosity to the Navier-Stokes system on otherwise general bounded regions. We can also obtain a foundational qualitative theory for this system as well as for spectral hyperviscosity, which adds hyperviscosity only to the high frequencies past a cutoff wavenumber. 9:31AM M28.00008 Turbulent Flow past High Temperature Surfaces1 , IGBAL MEHMEDAGIC, U. S. Army ARDEC, Picatinny Arsenal, NJ, SIVA THANGAM, Stevens Institute of Technology, Hoboken, NJ, PASQUALE CARLUCCI, LIAM BUCKLEY, DONALD CARLUCCI, U. S. Army ARDEC, Picatinny Arsenal, NJ — Flow over high-temperature surfaces subject to wall heating is analyzed with applications to projectile design. In this study, computations are performed using an anisotropic Reynolds-stress model to study flow past surfaces that are subject to radiative flux. The model utilizes a phenomenological treatment of the energy spectrum and diffusivities of momentum and heat to include the effects of wall heat transfer and radiative exchange. The radiative transport is modeled using Eddington approximation including the weighted effect of nongrayness of the fluid. The time-averaged equations of motion and energy are solved using the modeled form of transport equations for the turbulence kinetic energy and the scalar form of turbulence dissipation with an efficient finite-volume algorithm. The model is applied for available test cases to validate its predictive capabilities for capturing the effects of wall heat transfer. Computational results are compared with experimental data available in the literature. Applications involving the design of projectiles are summarized. 1 Funded in part by U. S. Army, ARDEC. 9:44AM M28.00009 Long-range ordering of turbulent stresses in the 2D inverse energy cascade , YANG LIAO, NICHOLAS OUELLETTE, Yale University, OUELLETTE LAB TEAM — We report measurements of the spatial structure of the turbulent stress that couples motion on different length scales in a quasi-two-dimensional laboratory flow. We show that the range of scales over which we find net energy transfer to large scales—the inverse energy cascade—is associated with the appearance of long-range, system-spanning spatial order of the turbulent stress. Although the overall degree of order fluctuates in time, the form of the approach to ordering does not. Our results provide an unexpected example of turbulence-induced ordering, and suggest new pathways for modeling turbulence using geometric alignment. 9:57AM M28.00010 Filtering on the Sphere , HUSSEIN ALUIE, University of Rochester, MATTHEW HECHT, Los Alamos National Lab, GEOFFREY VALLIS, University of Exeter — The filtering approach has become an indispensable framework to analyzing and modeling turbulence, especially in the subject of Large-Eddy Simulation. However, applications have been mostly limited to flows in Euclidean spaces and generalizations to curvilinear domains suffer from several shortcomings, such as: dependence on the choice of coordinate system, commutation errors, or not preserving volume. Motivated by geophysical applications, we define a new generalized filtering operation for vector fields on the Sphere which is free from the aforementioned problems. We prove that our filter commutes with spatial derivatives, yielding simple and exact coarse-grained equations for flow on the Sphere. We demonstrate these tools with a-priori tests on flows from high-resolution Ocean simulations. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M29 Experimental Techniques: General — 2014 - Jun Chen, Purdue University 8:00AM M29.00001 Customized turbulent flow fields generated by means of an active grid , MICHAEL HOELLING, ForWind - Institute of Physics, University of Oldenburg, NICO REINKE, University of Oldenburg, ForWind, JOACHIM PEINKE, ForWind - Institute of Physics, University of Oldenburg — Wind tunnel experiments, which should clarify the interaction of wind energy converters and the ambient turbulent field, should be performed under realistic flow conditions. For the generation of realistic turbulent flow conditions we use an active grid. This grid allows for the generation of flows with high turbulence intensity and even to repeat those turbulent fields to a certain degree. Moreover, flow features are to a certain extent tuneable, e.g. velocity increments distributions or energy density spectrum, realized by individually controllable horizontal and vertical rotating axes, which are equipped with flaps. The rotation patterns of the axes over time are defined in an excitation protocol. The challenge is designing an excitation protocol, which generates a flow flied for a specific application. A general approach is still missing. Our approach allows estimating the flow features to given excitation protocols. The approach is based on the assumption that the flow field behind an active grid consists basically of different turbulent pulses, which belong to the excitation setting. Our approach gives a sequence of those pulses, which we call synthetic velocity time series, which is made on a computer. 8:13AM M29.00002 Measurements of the aerodynamic characteristics of the turbo-jav , KENTA YAMAMOTO, Kansai University Graduate School, TOMOYA NAKAJIMA, Osaka Prefecture University, TOMOAKI ITANO, MASAKO SUGIHARA-SEKI, Kansai University — The “turbo-jav” which is used for the javelic throw in the junior Olympic games has four tail fins. In order to investigate the aerodynamic characteristics of the turbo-jav with an emphasis on the effect of the fins, we performed wind tunnel tests, throwing experiments and numerical simulations of the flight for intact turbo-javs as well as turbo-javs with their fins cut. The wind tunnel tests showed that the drag and lift coefficients for the intact turbo-javs are larger than the corresponding values for the turbo-javs without fins. As the angle of attack increases from 0, the pitching moments for the intact turbo-javs decrease from 0, whereas the moments for the turbo-javs without fins increase. In accord with this property, the throwing experiments showed that intact turbo-javs fly stably with oscillating angle of attack around 0. The flight distance, the orbit and the variation of angle of attack for the intact turbo-javs launched by a launcher agree closely with the numerical simulation performed based on the wind tunnel tests. A comparison of throwing experiments by students and by the launcher suggested significant effects of the rolling motion of the turbo-jav on its flight characteristics. 8:26AM M29.00003 On the measurement of turbulence with unmanned aerial vehicles1 , BRANDON WITTE, MICHAEL THAMANN, SEAN BAILEY, University of Kentucky — We address the challenge of taking the novel approach of using highly instrumented and autonomous unmanned aerial vehicles (UAVs) to spatially interrogate the atmospheric boundary layer’s turbulent flow structure over a wide range of length scales. This approach will introduce new capabilities not available in contemporary micro-meteorological measurement techniques: the ability to spatially sample the flow field over a wide range of spatial scales; a reduced reliance on assumptions regarding the temporal evolution of the turbulence; the ability to measure in a wide range of boundary conditions and distance from the earth’s surface; the ability to gather many boundary layer thicknesses of data during brief periods of statistical quasi-stationarity; and the ability to acquire data where and when it is needed. We describe recent progress made in developing purpose-built airframes, integrating sensors into those airframes, and developing data analysis techniques to isolate the atmospheric turbulence from the measured velocity signal. 1 This research is supported by NASA Kentucky Award NNX10AL96H and NSF Award CBET-1351411 8:39AM M29.00004 Measurement of wall shear stress in a pulsatile pipe flow system using micro-pillar shear sensor (MPS3) , VRISHANK RAGHAV, CHRISTINE GARCIA, Georgia Institute of Technology, EBENEZER GNANAMANICKAM, Embry-Riddle Aeronautical University, AJIT YOGANATHAN, Georgia Institute of Technology, GT-EMBRY-RIDDLE COLLABORATION — The measurement of unsteady wall shear stress (WSS) in a pulsatile flow system is quite a challenge in experimental fluid mechanics. Recent developments in micro fabrication techniques have resulted in a novel measurement technique called the micro-pillar shear stress sensor (MPS3). It is a micro-pillar mounted on the surface of interest, which deflects an amount proportional to the shear stress it experiences. This technique has been widely used, validated and applied to measure turbulent WSS in several flow configurations. In this work, the MPS3 technique is used to measure WSS for a pulsatile fully developed pipe flow. The main objective here is to validate this technique for pulsatile pipe flow applications. For this purpose the WSS measurements obtained are compared with those obtained from analytical womersley solutions of the pulsatile flow system in the laminar flow regime. Statistical metrics will be used to better understand the measured WSS through the time period of the pulsatile flow. 8:52AM M29.00005 Further development of a wall-shear-stress sensor and validation in laminar and turbulent flows1 , LAURENT MYDLARSKI, PIERRE-ALAIN GUBIAN, JAMES MEDVESCEK, CRISTIAN TOMAZELA PRADO, B. RABI BALIGA, Department of Mechanical Engineering, McGill University — The present work involves the further development of a wall-shear-stress sensor, and its subsequent validation in both laminar and turbulent flows. Inspired by the works of Spazzini et al., Meas. Sci. Technol., 1999 and Sturzebecher et al., Exp. Fluids, 2001, the sensor consists of a tungsten hot-wire flush-mounted over a shallow rectangular slot, which serves to reduce heat loss to the substrate and therefore improve the frequency response of the sensor – a problem that frequently plagues hot-film wall-shear-stress sensors in many applications in air. Different aspects of the design, construction, operation and validation of the sensor will be presented. Particular attention will be paid to the performance of the sensor in fully developed turbulent channel flow, where measurements of statistical moments, probability density functions, and spectra of the wall-shear stress will be considered for turbulent Reynolds numbers (based on the friction velocity and half-height, Reτ ) in the range 200 ≤ Reτ ≤ 900. These measures will be compared with previous (experimental and numerical) work studying the wall-shear stress. The evolution of the statistics with Reτ will also be discussed. 1 Support for this work was provided by NSERC and Intel. 9:05AM M29.00006 Design of a High-Reynolds Number Recirculating Water Tunnel , LIBIN DANIEL, BRIAN ELBING, Oklahoma State University — An experimental fluid mechanics laboratory focused on turbulent boundary layers, drag reduction techniques, multiphase flows and fluid-structure interactions has recently been established at Oklahoma State University. This laboratory has three primary components; (1) a recirculating water tunnel, (2) a multiphase pipe flow loop, and (3) a multi-scale flow visualization system. The design of the water tunnel is the focus of this talk. The criteria used for the water tunnel design was that it had to produce a momentum-thickness based Reynolds number in excess of 104 , negligible flow acceleration due to boundary layer growth, maximize optical access for use of the flow visualization system, and minimize inlet flow non-uniformity. This Reynolds number was targeted to bridge the gap between typical university/commercial water tunnels (103 ) and the world’s largest water tunnel facilities (105 ). These objectives were achieved with a 152 mm (6-inch) square test section that is 1 m long and has a maximum flow speed of 10 m/s. The flow non-uniformity was mitigated with the use of a tandem honeycomb configuration, a settling chamber and an 8.5:1 contraction. The design process that produced this final design will be presented along with its current status. 9:18AM M29.00007 A new technique to linearly stratify a fluid , MICKAEL BOSCO, PATRICE MEUNIER, None, ROTATING AND GEOPHYSICAL FLOWS TEAM — Given that oceans and the atmosphere are stratified, most environmental flows like island and mountain range wakes are strongly influenced by the mean density gradient. Consequently, a great number of laboratory experiments have been run using stratified fluids to study geophysical flow. The double-bucket method is generally used to create a stable linearly stratified fluid. The water from the first bucket filled with salted water is slowly deposited at the surface of the tank with a floater and the density of the first bucket is gradually decreased by the addition of fresh water from a second bucket. Nevertheless, this method is not very convenient for large tank as the two buckets are very large and can easily be bulky. A simple method has been created which only needs two walls inside the tank. One plain barrier will ensure watertightness between the two sides of the tank and one holey barrier will allow density-driven exchanges at the origin of a stable linear stratification. One of the motivations was to analyze a stratified cylinder wake. The study has revealed four 3D unstable modes that appears behind the cylinder. 9:31AM M29.00008 Surrogate Immiscible Liquid Solution Pairs with Refractive Indexes Matchable Over a Wide Range of Density and Viscosity Ratios1 , RAJAT SAKSENA2 , KENNETH T. CHRISTENSEN3 , ARNE J. PEARLSTEIN, Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign — Use of laser diagnostics in liquidliquid flows is limited by refractive index mismatch. This can be avoided using a surrogate pair of immiscible index-matched liquids, with density and viscosity ratios matching those of the original liquid pair. We demonstrate that a wide range of density and viscosity ratios is accessible using aqueous solutions of 1,2-propanediol and CsBr (for which index, density, and viscosity are available), and solutions of light and heavy silicone oils and 1-bromooctane (for which we measured the same properties at 119 compositions). For each liquid phase, polynomials in the composition variables were fitted to index and density and to the logarithm of kinematic viscosity, and the fits were used to determine accessible density and viscosity ratios for each matchable index. Index-matched solution pairs can be prepared with density and viscosity ratios equal to those for water-liquid CO2 at 0o C over a range of pressure, and for water-crude oil and water-trichloroethylene, each over a range of temperature. For representative index-matched solutions, equilibration changes index, density, and viscosity only slightly, and chemical analysis show that no component of either solution has significant interphase solubility. 1 Partially supported by Intl. Inst. for Carbon-Neutral Energy Research at Dept. of Mechanical and Aerospace Eng., Ohio State Univ. 3 Now at Dept. of Aerospace and Mechanical Eng., Univ. of Notre Dame 2 Now 9:44AM M29.00009 Dynamic mode decomposition analysis of instability of the flow past rotating sphere , MACIEJ SKARYSZ, Warsaw University of Technology, Institute of Aeronautics and Applied Mechanics, Poland, SOPHIE GOUJON-DURAND, JOSE EDUARDO WESFREID, Ecole Superieure de Physique et de Chimie Industrielles de Ville de Paris, PMMH, UMR7636 CNRS ESPCI P6-P7, France — Dynamic mode decomposition (DMD) is an effective method to obtain the description of coherent features of fluid flow generated both by numerical simulations and experimental measurements. Extraction of dynamic modes connected with the frequency created by the method provide essential information and made our understanding of fluid-dynamical process more meaningful. The wake behind rotating sphere for low Reynolds number (lower than 400) was experimentally investigated. Different regimes depending both on rotating rate and Reynolds number appears and was characterized by different multiple frequencies observed in the wake. In this case the DMD analysis was very efficient to expose full spectra and new modes remaining undetectable for other methods. 9:57AM M29.00010 Real-time contaminant sensing and control in civil infrastructure systems1 , SARA RIMER, NIKOLAOS KATOPODES, Univ of Michigan - Ann Arbor — A laboratory-scale prototype has been designed and implemented to test the feasibility of real-time contaminant sensing and control in civil infrastructure systems. A blower wind tunnel is the basis of the prototype design, with propylene glycol smoke as the “contaminant.” A camera sensor and compressed-air vacuum nozzle system is set up at the test section portion of the prototype to visually sense and then control the contaminant; a real-time controller is programmed to read in data from the camera sensor and administer pressure to regulators controlling the compressed air operating the vacuum nozzles. A computational fluid dynamics model is being integrated in with this prototype to inform the correct pressure to supply to the regulators in order to optimally control the contaminant’s removal from the prototype. The performance of the prototype has been evaluated against the computational fluid dynamics model and is discussed in this presentation. Furthermore, the initial performance of the sensor-control system implemented in the test section of the prototype is discussed. 1 NSF-CMMI 0856438 Tuesday, November 25, 2014 8:00AM - 10:10AM Session M30 Aerodynamics: Unsteady Airfoils and Wings 2016 - — 8:00AM M30.00001 Buffeting of NACA 0012 airfoil at high angle of attack , TONG ZHOU1 , Beijing Institute of Technology, EARL DOWELL, Duke University — Buffeting is a fluid instability caused by flow separation or shock wave oscillations in the flow around a bluff body. Typically there is a dominant frequency of these flow oscillations called Strouhal or buffeting frequency. In prior work several researchers at Duke University have noted the analogy between the classic Von Karman Vortex Street behind a bluff body and the flow oscillations that occur for flow around a NACA 0012 airfoil at sufficiently large angle of attack. Lock-in is found for certain combinations of airfoil oscillation (pitching motion) frequencies and amplitudes when the frequency of the airfoil motion is sufficiently close to the buffeting frequency. The goal of this paper is to explore the flow around a static and an oscillating airfoil at high angle of attack by developing a method for computing buffet response. Simulation results are compared with experimental data. Conditions for the onset of buffeting and lock-in of a NACA 0012 airfoil at high angle of attack are determined. Effects of several parameters on lift coefficient and flow response frequency are studied including Reynolds number, angle of attack and blockage ratio of the airfoil size to the wind tunnel dimensions. Also more detailed flow field characteristics are determined. For a static airfoil, a universal Strouhal number scaling has been found for angles of attack from 30◦ to 90◦ , where the flow around airfoil is fully separated. For an oscillating airfoil, conditions for lock-in are discussed. Differences between the lock-in case and the unlocked case are also studied. 1 The second affiliation: Duke University 8:13AM M30.00002 Spanwise gradients in flow speed and leading edge vortex attachment on low Reynolds number wings , THIERRY JARDIN, ISAE, University of Toulouse, LAURENT DAVID, Institut Pprime, University of Poitiers — It is now accepted that the aerodynamic performance of low aspect ratio revolving wings, such as insect wings or maple seed membranes, largely relies on sustained leading edge vortex attachment. However, the mechanisms responsible for this sustained attachment are still poorly understood. Here, we compute the Navier-Stokes solution of the flow around a finite wing (i) subjected to a uniform oncoming flow, (ii) subjected to a spanwise varying oncoming flow and (iii) revolving about its root. Therefore, we are able to isolate the mechanisms associated with the spanwise gradient of the local wing speed from those associated with centrifugal and Coriolis effects. We show that over flapping amplitudes typical of insect flight the spanwise gradient of the local wing speed may suffice in maintaining leading edge vortex attachment. We correlate this result with the development of spanwise flow and we evaluate the sensitivity of such a mechanism to the Reynolds number. It is noted, however, that leading edge vortex attachment through the spanwise gradient of the local wing speed does not promote large lift, which ultimately arises from centrifugal and Coriolis effects. 8:26AM M30.00003 Exploring Unsteady Sail Propulsion in Olympic Class Sailboats , RILEY SCHUTT, C.H.K. WILLIAMSON, Cornell University — Unsteady sailing techniques, defined as “flicking,” “roll-tacking” and “roll-gybing” are used by athletes to propel their boats on an Olympic race course faster than using the wind alone. Body weight movements induce unsteady sail motion, increasing driving force and enhancing maneuvering performance. In this research, we explore the dynamics of an Olympic class Laser sailboat equipped with a GPS, IMU, wind sensor, and camera array. The velocity heading of a sailing boat is oriented at an apparent wind angle to the flow. In contrast to classic flapping propulsion, the heaving of the sail section (induced by the sailor’s body movement) is not perpendicular to the sail’s motion through the air. This leads to an “exotic heave,” with components parallel and perpendicular to the incident flow. The characteristic motion is recreated in a towing tank where the vortex structures generated by a representative 2-D sail section are observed, along with a measurement of thrust and lift forces. When combined with turning maneuvers, these heaving sail motions can lead to significant increases in velocity made good, a critical variable used when assessing racing performance. 8:39AM M30.00004 Three-dimensional flow fields and forces on revolving flat plates1 , MUSTAFA PERCIN, BAS W. VAN OUDHEUSDEN, Delft University of Technology — The evolution of three-dimensional flow structures of revolving low-aspect-ratio plates in the Reynolds number range of 10,000 to 20,000 was studied, combining Tomographic Particle Image Velocimetry with force measurements. Two motion kinematics were considered: (1) a revolving surge motion where the wing accelerates to a terminal velocity with a constant acceleration at a fixed angle of attack and then remains to revolve at a constant rate; (2) a revolving pitch motion which is initiated by a constant acceleration from rest to a terminal velocity at zero angle of attack, followed by a pitch-up motion at a constant pitch rate and revolution at a constant rate. In the experiments, the terminal velocity, acceleration, angle of attack and pitch rate were varied to study their effect on the resultant flow fields and forces. In general, a vortex system that consists of a leading edge vortex, a tip vortex and a trailing edge vortex is observed. The vortex system bursts into substructures as the motion progresses, which does not lead to a decrease in the forces. The evolution of spanwise flow and the effects of centrifugal acceleration and spanwise pressure gradient are discussed. 1 This research is supported by the Dutch Technology Foundation STW, project number 11023. 8:52AM M30.00005 Formation Flight: Modes of Interaction of a Streamwise Vortex with a Wing , CHRIS MCKENNA, MATTHEW BROSS, DONALD ROCKWELL, Lehigh University — Aircraft flying together in an echelon or V formation experience aerodynamic advantages. Impingement of the tip vortex of the leader (upstream) wing on the follower wing can yield an increase of lift to drag ratio. This enhancement is known to be sensitive to the location of vortex impingement on the follower wing. Particle image velocimetry is employed to determine patterns of velocity and vorticity in successive crossflow planes, which characterize the streamwise evolution of the vortex structure along the chord of the follower wing and into its wake. Different modes of vortex-follower wing interaction are created by varying the spanwise location of the leader wing. These modes are defined by differences in the development of, and interaction between, the incident tip vortex from the leader wing and the tip vortex along the follower wing. Modes of development/interaction of the tip vortices include bifurcation, attenuation, and mutual induction. The bifurcation and attenuation modes decrease the strength of the follower tip vortex. In contrast, the mutual induction mode increases the strength of the follower tip vortex. 9:05AM M30.00006 Vortex shedding and aerodynamic performance of an airfoil with multiscale trailing edge modifications , JOVAN NEDIC, J. CHRISTOS VASSILICOS, Imperial College London — An experimental investigation was conducted into the aerodynamic performance and nature of the vortex shedding generated by truncated and non-flat serrated trailing edges of a NACA 0012 wing section. The truncated trailing edge generates a significant amount of vortex shedding, whilst increasing both the maximum lift and drag coefficients, resulting in an overall reduction in the maximum lift-to-drag ratio (L/D) compared to a plain NACA0012 wing section. By decreasing the chevron angle (φ) of the non-flat trailing edge serrations (i.e. by making them sharper), the energy of the vortex shedding significantly decreases and L/D increase compared to a plain NACA0012 wing section. Fractal/multi-scale patterns were also investigated with a view to further improve performance. It was found that the energy of the vortex shedding increases with increasing fractal iteration if the chevron is broad (φ ≈ 65◦ ), but decreases for sharper chevrons (φ = 45◦ ). It is believed that if φ is too big, the multi-scale trailing edges are too far away from each other to interact and break down the vortex shedding mechanism. Fractal/multi-scale trailing edges are also able to improve aerodynamic performance compared to the NACA 0012 wing section. 9:18AM M30.00007 Identification of Scaling Parameters for Rotor-Induced Sediment Mobilization , GINO PERROTTA, Univ of Maryland-College Park — Flow imaging and particle imaging velocimetry experiments were conducted in a water tank to investigate the effects of rotor wake and sediment characteristics on rotor-induced sediment mobilization during hover in ground effect. The two-phase flow was separated into carrier phase and dispersed phase. The carrier phase was studied using PIV to acquire time-resolved planar velocity measurements for a field of view within the rotor wake. The rotor-induced flow was confirmed to be dominated by blade tip vortices and was thus characterized in terms of the vortex properties. Vortices were identified using a nonlocal function and were fit to the Lamb-Oseen vortex velocity profile to evaluate size and strength. The rotor-induced flow was also characterized in terms of wall jet velocity and turbulent kinetic energy. The dispersed phase was separated from the carrier phase using image filtering procedures and was quantified by identifying mobilized sediment particles visible in the field of view. Candidate scaling parameters were created by combining rotor-induced sediment mobilization system characteristics. These candidate parameters were inspected for correlation with sediment mobilization. Three new scaling parameters are proposed and evaluated. 9:31AM M30.00008 Use of passively actuated flaps for enhanced lift for pitching and heaving airfoils , FIRAS SIALA, CAMERON PLANCK, JAMES LIBURDY, Oregon State Univ — The enhanced lift and reduced drag obtained by applying passively actuated leading and trailing flaps to a low aspect ratio flat wing during heaving and pitching at moderate Reynolds numbers (104 ) is demonstrated. Direct force measurements are obtained during the cyclic motion and are synchronized with the tracking of the motion of the passive flaps. The flaps are controlled using torsion springs and their natural frequency is found to play a dominant role in determining the lift enhancement. Results are shown for a range of heaving and pitching conditions of amplitude and frequency, with the pitching phase offset ninety degrees from the heaving. Flow visualization is used to document the transient wake conditions. The lift and drag forces are shown to be enhanced near the peak effective angle of attack during the cycling motion resulting in a net mean lift increase. 9:44AM M30.00009 Force and vortical flow development on pitching wings at high rates , LUIS BERNAL, HUAI-TE YU, Univ of Michigan - Ann Arbor, MICHAEL OL, KENNETH GRANLUND, Air Force Research Laboratory — Recent experimental results of pitching flat plate wings are presented. High pitch-rate perching maneuvers are frequently used by birds for feeding and landing. Insects use very fast rotation rates at the end of each flapping stroke, which results in high thrust and precise flight. These wing motions are also of interest for engineered micro air vehicles to achieve semi-autonomous landing by unskilled operators. The wing motion considered is a constant rotation rate pitch motion from 0 to 45 degrees of an aspect-ratio-4 flat-plate wing. The goal is to gain a better understanding of force generation mechanisms and their relationship to two- and three-dimensional vortical flow structures. Leading edge, trailing edge, and tip vortices form with large separated flow regions over the wing, however comparison with linear potential flow theory gives good agreement. The evolution of the leading edge vortex is delayed for pivot axes locations downstream of the leading edge. Large forces at the end of the motion slowly return to the steady state value over more than 30 convective times. The flow in the near wake shows a brief period of vortex shedding and strong three dimensional effects. Two different three-dimensional flow features are observed: A rapid development of three-dimensionality in the core of the leading and trailing edge vortices and a swirl motion in the near wake. However the impact of these three-dimensional flow features on force development is small. 9:57AM M30.00010 Unsteady Airloads on Airfoils in Reverse Flow , ANDREW LIND, ANYA JONES, University of Maryland — This work gives insight into the influence of airfoil characteristics on unsteady airloads for rotor applications where local airfoil sections may operate at high and/or reverse flow angles of attack. Two-dimensional wind tunnel experiments have been performed on four airfoil sections to investigate the effects of thickness, camber, and trailing edge shape on unsteady airloads (lift, pressure drag, and pitching moment). These model rotor blades were tested through 360 deg of incidence for 104 ≤ Re ≤ 106 . Unsteady pressure transducers were mounted on the airfoil surface to measure the high frequency, dynamic pressure variations. The temporal evolution of chordwise pressure distributions and resulting airloads is quantified for each airfoil in each of the three unsteady wake regimes present in reverse flow. Specifically, the influence of the formation, growth, and shedding of vortices on the surface pressure distribution is quantified and compared between airfoils with a sharp geometric trailing edge and those with a blunt geometric trailing edge. These findings are integral to mitigation of rotor blade vibrations for applications where airfoil sections are subjected to reverse flow, such as high-speed helicopters and tidal turbines. Tuesday, November 25, 2014 8:00AM - 9:57AM Session M31 CFD: Uncertainty Quantification — 2018 - Onkar Sahni, Rensselaer Polytechnic Institute 8:00AM M31.00001 Statistical analysis and simulation of random shock waves in scalar conservation laws1 , DANIELE VENTURI, HEYRIM CHO, GEORGE KARNIADAKIS, Brown University — Hyperbolic conservation laws subject to additive random noise and random initial conditions can develop random shock waves at random space-time locations. The statistical analysis of such waves is quite complex, due to non-linearities, high-dimensionality and lack of regularity. By using the Mori-Zwanzig formulation of irreversible statistical mechanics, we derive formally exact reduced-order equations for the one- and two-point probability density function of the solution field. This allows us to perform numerical simulations and determine the statistical properties of the system. We consider the inviscid limit of the stochastic Burgers equation as a model problem and determine its solution in physical and probability spaces by using adaptive discontinuous Galerkin methods. In particular, we study stochastic flows generated by random initial states and random additive noise, yielding multiple interacting shock waves collapsing into clusters and settling down to a similarity state. We also address the question of how random shock waves in space and time manifest themselves in probability space. The mathematical framework is general and it can be applied to other systems, leading to new insights in high-dimensional stochastic dynamics and more efficient computational algorithms. 1 This work was supported by OSD-MURI grant FA9550-09-1-0613, DOE grant DE-SC0009247 and NSF/DMS-1216437 grant 8:13AM M31.00002 Quantification of the uncertainty of finite-time-average approximations of infinite-time-average statistics in turbulence simulations , POORIYA BEYHAGHI, THOMAS BEWLEY, University of California San Diego — Turbulent flows are often stationary and ergodic, which means that the time average of a quantity (TKE, total drag, etc) converges to a constant as the averaging interval is increased. This infinite-time-averaged statistic is of particular interest in many problems, such as aerodynamic shape optimization. Since taking an average over an infinite time horizon is not possible in simulation, some finite-time approximation of the infinite-time-average statistic of interest is generally used in practice. The error of this approximation decreases slowly, like the reciprocal of the square roots of the averaging time. In the present work, we develop a framework to quantify precisely the uncertainty of such a finite-time-average approximation of an infinite-time-average statistic of a stationary ergodic process. In the method used, different statistical models for stationary processes have been examined to model the statistical behavior of the time series derived from the turbulence simulation. It is observed that the statistical behavior of some of these models is sufficiently representative of that of the real time series that they provide an accurate estimate of the uncertainty associated with the finite time average approximation of the statistic of interest. 8:26AM M31.00003 Uncertainty Quantification of RANS dispersion modeling in Oklahoma City during the Joint Urban 2003 campaign , CLARA GARCIA-SANCHEZ, von Karman Institute for Fluid Dynamics, University of Antwerp, CATHERINE GORLE, Stanford University, University of Antwerp, JEROEN VAN BEECK, von Karman Institute for Fluid Dynamics, GIANLUCA IACCARINO, Stanford University — The high expansion rate of urban areas makes realistic predictions of dispersion within cities an important research topic. The transport of pollutants is influenced by wind flows that are affected by the large scale variability of the atmospheric boundary layer (ABL). In order to improve the predictive capabilities of Computational Fluid Dynamics simulations (CFD) of the ABL, this atmospheric variability should be included. This work focuses on representing this variability in the inflow boundary conditions using an uncertainty quantification framework for the Joint Urban 2003 experiment. The simulations focus on the Intensive Observation Period number 9, where a continuous release of SF6 took place in downtown Oklahoma. The RANS simulations with the k-epsilon turbulence model were performed with the code OpenFOAM, and an equation for passive scalar transport is solved, using a standard gradient diffusion model for the turbulent dispersion, to obtain the SF6 concentration. To define the inflow boundary conditions three uncertain parameters are used: wind speed, wind direction, and ABL roughness height. To propagate these uncertainties a tensor grid Clenshaw-Curtis Stochastic Collocation approach was used, and a polynomial chaos representation of the velocity and concentration at different field measurement locations was constructed to extract the mean and standard deviations. 8:39AM M31.00004 Statistically accurate low-order models for uncertainty quantification in turbulent dynamical systems , THEMISTOKLIS SAPSIS, MIT — A framework for low-order predictive statistical modeling and uncertainty quantification in turbulent dynamical systems will be presented. These reduced-order, modified quasilinear Gaussian (ROMQG) algorithms apply to turbulent dynamical systems in which there is significant linear instability or linear non-normal dynamics in the unperturbed system and energy-conserving non-linear interactions that transfer energy from the unstable modes to the stable modes where dissipation occurs, resulting in a statistical steady state; such turbulent dynamical systems are ubiquitous in geophysical and engineering turbulence. The ROMQG method involves constructing a low-order, nonlinear, dynamical system for the mean and covariance statistics in the reduced subspace that has the unperturbed statistics as a stable fixed point and optimally incorporates the indirect effect of non-Gaussian third-order statistics for the unperturbed system in a systematic calibration stage. This calibration procedure is achieved through information involving only the mean and covariance statistics for the unperturbed equilibrium. The performance of the ROMQG algorithm is assessed on two stringent test cases: the 40-mode Lorenz 96 model mimicking midlatitude atmospheric turbulence and two-layer baroclinic models for high-latitude ocean turbulence with over 125,000 degrees of freedom. 8:52AM M31.00005 Toward Uncertainty Quantification of Turbulence Closure Models , AASHWIN MISHRA, SHARATH GIRIMAJI, Texas A&M University — Predictive turbulence calculations require that the uncertainty in various constituent closures is quantified. We propose that uncertainty quantification must commence at the Reynolds stress closure level, specifically, with the pressure-strain correlation term. The Reynolds stress tensor provides an insufficient basis to describe the internal structure of a turbulent field, expressly its dimensionality. It is demonstrated that this leads to an inherent degree of uncertainty in classical models for turbulent flows. Using Interval Analysis, we quantify the propagation of this epistemic uncertainty for rapid pressure strain correlation models for different regimes of mean flow. It is exhibited that the magnitude of this uncertainty is dependent not just upon the dimensionality of the turbulent field, but to a greater degree upon the nature of the mean flow. In contrast to prior beliefs, we prove that such uncertainty is present (and even greater) in the absence of mean rotation. Finally, we analyze the qualitative and quantitative effects of the non-linear component of pressure on this systemic uncertainty. 9:05AM M31.00006 Stochastic modeling of jet in crossflow using dynamically orthogonal decomposition , HESSAM BABAEE, THEMISTOKLIS SAPSIS, Massachusetts Institute of Technology, GEORGE KARNIADAKIS, Brown University — In this numerical study the effect of stochastic perturbation on jet in crossflow is investigated. To efficiently quantify the evolution of stochasticity in such a system, the dynamically orthogonal method is used. In this methodology, the solution is approximated by a generalized Karhunen-Loeve (KL) expansion in PN the form of u(x, t; ω) = u(x, t) + y (t; ω)ui (x, t), in which u(x, t) is the stochastic mean, the set of ui (x, t)’s is a deterministic orthogonal basis and i=1 i yi (t; ω)’s are the stochastic coefficients. Explicit evolution equations for u, ui and yi are formulated. The elements of the basis ui (x, t)’s remain orthogonal for all times and they evolve according to the system dynamics to capture the energetically dominant stochastic subspace. In this study, the stochasticity is introduced at the crossflow boundary condition and, in particular, the effect of different time and length scales of the stochastic perturbation on the jet dynamics is investigated. The energy cascades and correlation between stochastic energy levels in the statistical sense are also analyzed. The relationship between the dynamic stochastic modes and the coherent structures present in jet in crossflow is discussed. 9:18AM M31.00007 Representing Model Inadequacy in Combustion Kinetics , REBECCA E. MORRISON, ROBERT D. MOSER, The University of Texas at Austin — An accurate description of the chemical processes involved in the oxidation of hydrocarbons may include hundreds of reactions and thirty or more chemical species. Kinetics models of these chemical mechanisms are often embedded in a fluid dynamics solver to represent combustion. Because the computational cost of such detailed mechanisms is so high, it is common practice to use drastically reduced mechanisms. But, this introduces modeling errors which may render the model inadequate. In this talk, we present a formulation of the model inadequacy in reduced models of hydrogen-methane combustion. Our goal is to account for the discrepancy between the high-fidelity model and its reduced version by incorporating an additive, linear, probabilistic inadequacy model. In effect, it is a random matrix, whose entries are characterized by probability distributions and which displays interesting properties due to conservation constraints. The distributions are calibrated via Bayesian inference using a hierarchical modeling scheme and high-dimensional MCMC. We apply this technique to a stand-alone reaction and also incorporate it within a one-dimensional laminar flame problem. 9:31AM M31.00008 Adaptive Discrete Equation Method for injection of stochastic cavitating flows , GIANLUCA GERACI, Stanford Univ and Inria Bordeaux, MARIA GIOVANNA RODIO, Inria Bordeaux, GIANLUCA IACCARINO, Stanford Univ, REMI ABGRALL, Universitat Zurich, PIETRO CONGEDO, Inria Bordeaux — This work aims at the improvement of the prediction and of the control of biofuel injection for combustion. In fact, common injector should be optimized according to the specific physical/chemical properties of biofuels. In order to attain this scope, an optimized model for reproducing the injection for several biofuel blends will be considered. The originality of this approach is twofold, i) the use of cavitating two-phase compressible models, known as Baer & Nunziato, in order to reproduce the injection, and ii) the design of a global scheme for directly taking into account experimental measurements uncertainties in the simulation. In particular, stochastic intrusive methods display a high efficiency when dealing with discontinuities in unsteady compressible flows. We have recently formulated a new scheme for simulating stochastic multiphase flows relying on the Discrete Equation Method (DEM) for describing multiphase effects. The set-up of the intrusive stochastic method for multiphase unsteady compressible flows in quasi 1D configuration will be presented. The target test-case is a multiphase unsteady nozzle for injection of biofuels, described by complex thermodynamics models, for which experimental data and associated uncertainties are available. 9:44AM M31.00009 Regression-based adaptive sparse polynomial dimensional decomposition for sensitivity analysis , KUNKUN TANG, PIETRO CONGEDO, Inria, REMI ABGRALL, Universität Zürich — Polynomial dimensional decomposition (PDD) is employed in this work for global sensitivity analysis and uncertainty quantification of stochastic systems subject to a large number of random input variables. Due to the intimate structure between PDD and Analysis-of-Variance, PDD is able to provide simpler and more direct evaluation of the Sobol’ sensitivity indices, when compared to polynomial chaos (PC). Unfortunately, the number of PDD terms grows exponentially with respect to the size of the input random vector, which makes the computational cost of the standard method unaffordable for real engineering applications. In order to address this problem of curse of dimensionality, this work proposes a variance-based adaptive strategy aiming to build a cheap meta-model by sparse-PDD with PDD coefficients computed by regression. During this adaptive procedure, the model representation by PDD only contains few terms, so that the cost to resolve repeatedly the linear system of the least-square regression problem is negligible. The size of the final sparse-PDD representation is much smaller than the full PDD, since only significant terms are eventually retained. Consequently, a much less number of calls to the deterministic model is required to compute the final PDD coefficients. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M32 Nonlinear Dynamics V — 2020 - Laurette S. Tuckerman, ESPCI ParisTech 8:00AM M32.00001 Burgers Turbulence on a Fractal Fourier set1 , MICHELE BUZZICOTTI, LUCA BIFERALE, Dept. Physics University of Rome “Tor Vergata”, URIEL FRISCH, CNRS, Observatory of Nice, SAMRIDDHI RAY, Tata Institute of Fundamental Research — We present a systematic investigation of the effects introduced by a fractal decimation in Fourier space on stochastically forced one-dimensional Burgers equations. The aim is to understand the statistical robustness of the shock singularity under different reductions of the number of the degrees of freedom. We perform a series of direct numerical simulations by using a pseudo-spectral code with resolution up to 16384 points and for various dimensions of the fractal set of Fourier modes DF <1. We present results concerning the scaling properties of statistical measures in real space and the probability distribution functions of local and non-local triads in Fourier space. 1 Partially supported by ERC Grant No 339032. 8:13AM M32.00002 Neurophysiology of pipe flow , DWIGHT BARKLEY, Univ of Warwick — This work explores the connection between the transition to turbulence in pipe flow and the dynamics of excitable media, as exemplified by nerve cells. The primary goal is to leverage years of extensive analysis of neural systems to understand the dynamics of transitional turbulence. To demonstrate the predictive nature of the approach, model simulations will be presented for puffs in pipe flow for cases not previously studied experimentally. 8:26AM M32.00003 The Self-Sustaining Process for Taylor-vortex flow , LAURETTE TUCKERMAN, PMMHCNRS-ESPCI, France, TOMMY DESSUP, MSC, Univ Paris 7, France, DWIGHT BARKLEY, University of Warwick, United Kingdom, JOSE EDUARDO WESFREID, PMMH-CNRS-ESPCI, France, ASHLEY WILLIS, University of Sheffield, United Kingdom — The Self-Sustaining Process (SSP) of Waleffe, like Hall’s Vortex-Wave Interaction theory, was proposed as the fundamental element of turbulence in low Reynolds number turbulence in wall-bounded shear flows and consists of three phases. (i) Streamwise vortices bend nd the streamwise velocity contours via advection. (ii) The undulating streamwise velocity leads to waviness in the vortices via Kelvin-Helmholtz instability. (iii) Nonlinear interaction of the wavy streamwise vortices promotes the streamwise vortices. We explore the SSP for Taylor-vortex flow, for which streamwise (azimuthal) and wavy vortices are genuine steady states resulting from linear instabilities with well-defined thresholds. In particular, we determine the circumstances under which wavy vortices reinforce Taylor vortices. 8:39AM M32.00004 The onset of turbulence in a square duct flow , GREGOIRE LEMOULT, BJORN HOF, IST Austria — Wall bounded shear flows experience a sudden transition from a laminar state to turbulence as Reynolds number, Re, increases. K. Avila et al. (Science 333, 2011) recently characterized the onset of turbulence in pipe flow. They measured the probability for a localized disturbance to decay or spread and defined the critical Reynolds number, Rec , where the characteristic time for both process is equal. Using the same methodology, we measure these probabilities, decay and splitting, as a function of Re in a 1200 D long square duct, where D is the width of the duct. We found the expected exponential probability distribution for both processes which underlines their memoryless character. From the characteristic time of these distributions, we estimate the point where turbulence first becomes sustained in a square duct flow. The main difference with pipe flow is that the characteristic time at Rec is shorter making it more suitable for measurements of critical exponents in the framework of phase transition. These results also emphasize the universal behavior of the transition to turbulence in wall bounded shear flows. 8:52AM M32.00005 Direct laminar-turbulent transition in Taylor-Couette flow: Experiments and simulations1 , CHRISTOPHER J. CROWLEY, MICHAEL KRYGIER, SAMUEL G. RABEN, DANIEL BORRERO, ROMAN O. GRIGORIEV, MICHAEL F. SCHATZ, Georgia Institute of Technology — The transition to turbulence in Taylor-Couette flow is frequently mediated by stable flow states (e.g. interpenetrating spirals). We describe a direct laminar-turbulent transition in a system with counterrotating cylinders and small aspect ratio of 5.26. In experiments probed using tomographic PIV and direct numerical simulations with realistic boundary conditions, we find the transition is hysteretic, yet highly reproducible with turbulence triggered by the growth of weak spiral flows. 1 This study was supported by NSF DMS-1125302 and NSF CMMI-1234436 9:05AM M32.00006 Fiber bundles and geometric phases of turbulent pipe flows , FRANCESCO FEDELE, Georgia Institute of Technology — In this talk, I will discuss the role of continuous translation symmetries in the dynamics of turbulent pipe flows. Drawing from differential geometry, the geometric structure of the N-dimensional state space V of the Navier Stokes pipe flow can be defined by means of a base manifold P of dimension N-1 (quotient space) and 1-D fibers attached to any point p of P (fiber bundle). In V, a trajectory can be observed in a special comoving frame, from which the motion is locally transversal to the fibers (horizontal transport). The proper shift along the fibers to bring the motion in the comoving frame is called dynamical phase. This is, for example, the translational shift induced by the constant speed of a traveling wave (TW), or relative fixed point. A TW in state space projects to a fixed point on the base manifold P, whereas a relative periodic orbit (RPO) reduces to a periodic orbit (PO). In this case, the shift along the fibers includes also a geometric phase, induced by curvature of the base manifold P. As an application, I will present results on symmetry reduction of experimental pipe flow data acquired by means of Laser Induced Fluorescence (LIF) techniques exploiting a generalization of Hopf fibrations and complex projective spaces. A chaotic Lorenz-type dynamics is unveiled in the desymmetrized state space. Moreover, the analysis reveals that the time-varying speed of a turbulent peak during bursts is related to the geometric phase associated with the motion in the fiber bundle. 9:18AM M32.00007 Streamwise-Localized Solutions with natural 1-fold symmetry , SEBASTIAN ALTMEYER, Institut of Science and Technology Austria, ASHLEY WILLIS, School of Mathematics and Statistics, University of Sheffield, Sheffield S3 7RH, UK, BJÖRN HOF, Institut of Science and Technology Austria, 3400 Klosterneuburg, Austria — It has been proposed in recent years that turbulence is organized around unstable invariant solutions, which provide the building blocks of the chaotic dynamics. In direct numerical simulations of pipe flow we show that when imposing a minimal symmetry constraint (reflection in an axial plane only) the formation of turbulence can indeed be explained by dynamical systems concepts. The hypersurface separating laminar from turbulent motion, the edge of turbulence, is spanned by the stable manifolds of an exact invariant solution, a periodic orbit of a spatially localized structure. The turbulent states themselves (turbulent puffs in this case) are shown to arise in a bifurcation sequence from a related localized solution (the upper branch orbit). The rather complex bifurcation sequence involves secondary Hopf bifurcations, frequency locking and a period doubling cascade until eventually turbulent puffs arise. In addition we report preliminary results of the transition sequence for pipe flow without symmetry constraints. 9:31AM M32.00008 Kolmogorov-like Flow: Effect of the Boundaries on Stability and Transition to Weak Turbulence1 , RAVI KUMAR PALLANTLA, BALACHANDRA SURI, JEFFREY TITHOF, SCHATZ MICHAEL, ROMAN GRIGORIEV, Center for Nonlinear Science and School of Physics, Georgia Institute of Technology — The dynamical description of turbulence in fluid flows using non-chaotic unstable solutions of the Navier-Stokes equation, called Exact Coherent Structures (ECS), is a promising approach to understand and control the turbulence. However, it has never been properly validated in experiment. This talk discusses a quasi-two-dimensional implementation of the Kolmogorov flow that enables validation of both dynamical and statistical aspects of the ECS-based description of weak turbulence. We use a numerical model of an experiment, which employs an electromagnetically-driven thin layer of electrolyte supported by a thin layer of a liquid dielectric, to describe the effects of the boundary conditions and the system size on the stability of the base flow as well as the properties of ECS which emerge in the turbulent regime. 1 This work is supported in part by the National Science Foundation under grants No. CBET-0853691, CBET-0900018, and CMMI-1234436. 9:44AM M32.00009 Search for Exact Coherent Structures in a Quasi-Two-Dimensional Kolmogorov-Like Flow1 , BALACHANDRA SURI, JEFFREY TITHOF, RAVI KUMAR PALLANTLA, ROMAN GRIGORIEV, SCHATZ MICHAEL, Center for Nonlinear Science and School of Physics, Georgia Institute of Technology — Recent theoretical advances suggest that turbulence can be characterized using unstable solutions of the Navier-Stokes equations having regular temporal behavior, called Exact Coherent Structures (ECS). Due to their experimental accessibility and theoretical tractability two-dimensional flows provide an ideal setting for the exploration of turbulence from a dynamical systems perspective. In our talk, we present a combined numerical and experimental study of electromagnetically driven flows in a shallow layer of electrolyte. On the numerical front we present our research concerning the search for ECS in a two-dimensional Kolmogorov-like flow. We discuss the change in the dynamics of the flow as the Reynolds number is varied. For a weakly turbulent flow, we show that the turbulent trajectory explores a region of state space which contains a number of ECS, including equilibria and periodic orbit solutions. We then discuss the occurrence of states similar to these numerically computed ECS in an experimental quasi-two-dimensional Kolmogorov-like flow. 1 This work is supported in part by the National Science Foundation under grants No. CBET-0853691, CBET-0900018, and CMMI-1234436. 9:57AM M32.00010 Topological selection mechanism for conservation laws in incompressible stratified Euler fluids , GIOVANNI ORTENZI, University of Bergamo, SHENGQIAN CHEN, University of Wisconsin at Madison, ROBERTO CAMASSA, University of North Carolina at Chapel Hill, GREGORIO FALQUI, University of Milano-Bicocca, MARCO PEDRONI, University of Bergamo — With his Kaffeeloeffel thought (“gedanken”) experiment, in 1910 Klein suggested that a topological change of an ideal fluid’s domain can provide a mechanism for breaking the conservation of circulation enforced by Kelvin’s Theorem. In our study, we extend this idea to more general conservation laws and explore the role of topological properties in the dynamics of an incompressible Euler fluid with stratification. In particular, we show that topologically non-trivial configurations of stratified fluid domains generate selection mechanisms for conserved quantities other than vorticity. In the talk we concentrate on the simple example of an air-water system in a channel, which encapsulates all the main points of these selection mechanisms. Among other examples, we show that the connection properties of the air domain affect total horizontal momentum conservation, despite the translational invariance of the system and its consequences by Noether’s Theorem. Tuesday, November 25, 2014 8:00AM - 9:57AM Session M33 Fluids Education II — 2022 - David Majerich, Georgia Institute of Technology 8:00AM M33.00001 Teaching Technical Competencies for Fluid Mechanics Research , RANDALL TAGG, University of Colorado Denver — We are developing an “on demand” framework for students to learn techniques used in fluid mechanics research. The site for this work is a university-grade laboratory situated next to Gateway High School in Aurora, Colorado. Undergraduate university students work with K-12 students on research and technical innovation projects. Both groups need customized training as their projects proceed. A modular approach allows particular competencies such as pump selection, construction of flow piping and channels, flow visualization, and specific flow measurement methods to be acquired through focused lessons. These lessons can be learned in either a stand-alone fashion or assembled into units for formal courses. A research example was a student project on diffusion of infectious material in micro-gravity in the event of an intestinal puncture wound. A curriculum example is a 9-week quarter of high-school instruction on instrumentation that uses small-scale water treatment systems as a case study. 8:13AM M33.00002 Applying the results of education research to help students learn more: peer instruction and clicker questions in upper-division courses , RACHEL E. PEPPER, University of Puget Sound, STEPHANIE V. CHASTEEN, STEVEN J. POLLOCK, KATHERINE K. PERKINS, University of Colorado Boulder — The physics faculty at the University of Colorado have transformed four upper-division courses: Classical Mechanics/Math Methods, Electricity and Magnetism (E&M) I and II, and Quantum Mechanics. We discuss these transformations as a model for other upper-division courses, such as fluid mechanics, focusing on one of the changes made in the transformation effort: the addition of peer instruction (“clicker questions”) to lecture. The goals of our course transformation were to improve student learning and to develop materials and approaches that other faculty could easily adopt or adapt. In this talk, we review the evidence for effectiveness of peer instruction, discuss our implementation, and present evidence of improved student learning in our transformed upper division courses. Tips for effective use of peer instruction and banks of clicker questions available for fluid mechanics will also be discussed. Our curriculum materials are free and available at http://per.colorado.edu/sei. 8:26AM M33.00003 Web-Based Problem-Solving Assignment and Grading System , GILES BRERETON, RONALD ROSENBERG, Michigan State University — In engineering courses with very specific learning objectives, such as fluid mechanics and thermodynamics, it is conventional to reinforce concepts and principles with problem-solving assignments and to measure success in problem solving as an indicator of student achievement. While the modern-day ease of copying and searching for online solutions can undermine the value of traditional assignments, web-based technologies also provide opportunities to generate individualized well-posed problems with an infinite number of different combinations of initial/final/boundary conditions, so that the probability of any two students being assigned identical problems in a course is vanishingly small. Such problems can be designed and programmed to be: single or multiple-step, self-grading, allow students single or multiple attempts; provide feedback when incorrect; selectable according to difficulty; incorporated within gaming packages; etc. In this talk, we discuss the use of a homework/exam generating program of this kind in a single-semester course, within a web-based client-server system that ensures secure operation. 8:39AM M33.00004 Inside Out: Active learning in fluid dynamics in and out of the classroom , NIGEL KAYE, LISA BENSON, BEN SILL, Clemson University — Active learning can be broadly defined as any activity that engages students beyond just listening. But is it worth the effort, when we can just lecture and tell students all they need to know? Learning theories posit that students remember far more of what they say and do than of what they hear and see. The benefits of active learning include increased attendance (because class is now something different and attending is more worthwhile) and deeper understanding of concepts (because students get to practice answering and generating questions). A recent meta-analysis of research on active learning has summarized evidence of real outcomes of active learning. Research is showing that students’ performance on exams are higher and that they fail at lower rates in classes that involve active learning compared to traditional lecturing. Other studies have shown evidence of improved performance in follow-on classes, showing that the improved learning lasts. There are some topics and concepts that are best taught (or at least introduced) through lecturing, but even lecturing can be broken up by short activities that engage students so they learn more effectively. In this presentation, we will review the findings of the meta study and provide examples of active learning both inside and outside the classroom that demonstrate simple ways of introducing this approach in fluid dynamics classes. 8:52AM M33.00005 Flippin’ Fluid Mechanics – Quasi-experimental Pre-test and Post-test Comparison Using Two Groups , D.R. WEBSTER, D.M. MAJERICH, J. LUO, Georgia Tech — A flipped classroom approach has been implemented in an undergraduate fluid mechanics course. Students watch short on-line videos before class, participate in active in-class problem solving (in dyads), and complete individualized on-line quizzes weekly. In-class activities are designed to achieve a trifecta of: 1. developing problem solving skills, 2. learning subject content, and 3. developing inquiry skills. The instructor and assistants provide critical “just-in-time tutoring” during the in-class problem solving sessions. Comparisons are made with a simultaneous section offered in a traditional mode by a different instructor. Regression analysis was used to control for differences among students and to quantify the effect of the flipped fluid mechanics course. The dependent variable was the students’ combined final examination and post-concept inventory scores and the independent variables were pre-concept inventory score, gender, major, course section, and (incoming) GPA. The R-square equaled 0.45 indicating that the included variables explain 45% of the variation in the dependent variable. The regression results indicated that if the student took the flipped fluid mechanics course, the dependent variable (i.e., combined final exam and post-concept inventory scores) was raised by 7.25 points. Interestingly, the comparison group reported significantly more often that their course emphasized memorization than did the flipped classroom group. 9:05AM M33.00006 Short storybooks to build conceptual understanding , EVAN VARIANO, University of California, Berkeley — To help students build intuitive or conceptual understanding of key fluids concepts, I present short stories written in the style of childrens’ books. The goal is to provide analogies with a strong visual component, in a format that allows students to return for a quick review. The content, philosophy, and initial student feedback will be discussed. 9:18AM M33.00007 A Comparison of the Development and Delivery of Two Short-Term Study-Abroad Thermal Sciences Courses , FRANK JACOBITZ, University of San Diego — Short-term study-abroad engineer- ing courses provide an opportunity to increase the international awareness and global competency of engineering students. Two different approaches have been taken in the past years in the development and delivery of two three-week long thermal sciences courses. A senior-level elective Topics in Fluid Mechanics course was taught twice in Marseille (France) in January 2010 and 2013. A sophomore-level Introduction to Thermal Sciences course was offered in London (United Kingdom) in July 2014. Both courses were developed due to a strong student desire for engineering study-abroad courses and an effort by the home institution to internationalize its curriculum. The common goals of the two courses are an effective teaching of their respective technical content combined with a meaningful international experience. The two courses differed in their respective settings: Topics in Fluid Mechanics was taught at Aix-Marseille University and included strong interactions with local faculty and students. Introduction to Thermal Sciences, however, was taught in a cluster of seven courses offered by the home institution in London. The courses were assessed using surveys, student reflection papers, course evaluations, and instructor observations. 9:31AM M33.00008 How to get students to love (or not hate) MATLAB and programming , SHANON RECKINGER, Fairfield University, SCOTT RECKINGER, Brown University — An effective programming course geared toward engineering students requires the utilization of modern teaching philosophies. A newly designed course that focuses on programming in MATLAB involves flipping the classroom and integrating various active teaching techniques. Vital aspects of the new course design include: lengthening in-class contact hours, Process-Oriented Guided Inquiry Learning (POGIL) method worksheets (self-guided instruction), student created video content posted on YouTube, clicker questions (used in class to practice reading and debugging code), programming exams that don’t require computers, integrating oral exams into the classroom, fostering an environment for formal and informal peer learning, and designing in a broader theme to tie together assignments. However, possibly the most important piece to this programming course puzzle: the instructor needs to be able to find programming mistakes very fast and then lead individuals and groups through the steps to find their mistakes themselves. The effectiveness of the new course design is demonstrated through pre- and post- concept exam results and student evaluation feedback. Students reported that the course was challenging and required a lot of effort, but left largely positive feedback. 9:44AM M33.00009 An integrated introduction to the mechanics of solids and fluids: Continuum mechanics as the first mechanics course , JENN STROUD ROSSMANN, Lafayette College, CLIVE DYM, LORI BASSMAN, Harvey Mudd College — We have developed an introduction to continuum mechanics for sophomore students without any prior knowledge of mechanics. The essence of continuum mechanics, the internal response of materials to external loading, is often obscured by the complex mathematics of its formulation. By building gradually from one- to two- and three-dimensional formulations, we are able to make the essence of the subject more accessible to undergraduates. From this gradual development of ideas, with many illustrative real-world case studies, students develop both physical intuition for how solids and fluids behave, and the mathematical techniques needed to begin to describe this behavior. At the same time they gain a unique appreciation for the connections between solid and fluid mechanics. It is particularly valuable for students interested in biological applications to appreciate the behavior of engineering materials as a spectrum with Hookean solids at one extreme, and Newtonian fluids at another, with many complex behaviors in between..This approach demonstrates the connections between solid and fluid mechanics, as well as the larger mathematical issues shared by both fields, to students who have not yet taken courses in fluid mechanics and/or strength of materials. The context and foundation provided by this educational strategy are available to students as they continue to study either solid or fluid mechanics, or specialize in the connections themselves by returning to a deeper study of the overarching field of continuum mechanics. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M34 Thermo-Acoustics and Flame Instabilities — 2024 - Tim Lieuwen, Georgia Institute of Technology 8:00AM M34.00001 Linear stability analysis of a premixed flame with lateral shear , CARLOS PANTANO, XIAOYI LU, University of Illinois at Urbana-Champaign — The hydrodynamic instability analysis of one-dimensional steady premixed planar flames, known as Darrieus-Landau, is extended to a lateral (side) uniform shear configuration. Here, in the steady planar flame, there is a transverse pressure gradient orthogonal to the density gradient and it is a situation of interest when a turbulent flame travels into a region of free-shear turbulence (such as a jet or shear layer). It is shown that the problem can be formulated analytically and a new dispersion relation can be determined. We were able to analytically solve Euler’s equation (with a constant shear parameter) and obtain the growth rate of flame front perturbation. The study of the dispersion relation shows that perturbations have two types of behavior as wavenumber increases. First, for negligible shear, we recover Darrieus-Landau result. Second, as the nondimensional shear parameter increases the flame becomes more unstable initially but eventually it completely stabilizes. There is a finite range of values of shear for which the flame remains stable. Finally, for sufficiently high shear, the flame becomes unstable again. Further details will be discussed at the talk. 8:13AM M34.00002 Hydrodynamic instabilities in swirl-stabilized combustion: experimental assessment and theoretical modelling1 , KILIAN OBERLEITHNER, Technical University Berlin, MICHAEL STÖHR, German Aerospace Center (DLR), Stuttgart, Germany, STEFFEN TERHAAR, OLIVER PASCHEREIT, Technical University Berlin — In gas turbine industry, it is common practice to implement swirling jets and associated vortex breakdown to stabilize the flame and to enhance turbulent mixing. The flow field of such swirl-stabilized combustors features a wide range of flow instabilities that promote the formation of large-scale flow structure. This talk presents recent experimental studies at the Technical University Berlin and the German Aerospace Center (DLR) targeting the impact of these instabilities on the combustion performance. Particular focus is placed on two types of instability: (i) a self-excited helical instability, typically known as the precessing vortex core, which crucially affects mixing and flame anchoring; (ii) the axisymmetric Kelvin-Helmholtz instability, which crucially affects the flame dynamics at thermo-acoustic oscillations. All experimental observations are correlated with analytic flow models utilizing linear hydrodynamic stability theory. This mathematical framework reveals the driving mechanisms that lead to the formation, saturation, and suppression of large-scale flow structures and how these mechanisms interact with the combustion process. 1 The authors kindly acknowledge the financial support of the German Research Foundation (DFG) and the Research Association for Combustion Engines (FVV). 8:26AM M34.00003 Global Stability Analysis of Jet Diffusion Flames , D. MORENO-BOZA, W. COENEN, A. SEVILLA, A.L. SÁNCHEZ, Dept. Ingenierı́a Térmica y de Fluidos, Universidad Carlos III de Madrid, Leganés 28911, Spain — This work investigates the global stability of axisymmetric laminar jet diffusion flames at moderately large Reynolds numbers, including effects of buoyancy, temperature increase due to chemical reaction and air coflow. The ultimate objective is to clarify the two different types of instabilities observed in experiments, as well as the connection of these instabilities with the phenomenon of diffusion-flame flickering. Quasi-isobaric conditions corresponding to low-Mach-number jets are considered and stability results regarding hot and light jets are also described. The limit of infinitely fast chemical reaction is used in the development, which assumes also a unity value of the fuel Lewis number, thereby enabling a simplified description of the temperature and composition fields in terms of a single mixture-fraction variable. A finite-element method is developed to integrate the steady equations of continuity, momentum and mixture fraction, which determine the basic steady flame structure as well as the associated perturbed equations that determine its 2D global stability. 8:39AM M34.00004 The dynamics of cellular two-dimensional flames , CHRISTOPHE ALMARCHA, JOEL QUINARD, BRUNO DENET, ELIAS AL-SARRAF, Aix Marseille Université, CNRS, Centrale Marseille, IRPHE UMR 7342, 13384, JEAN-MARIE LAUGIER, Aix Marseille Université, CNRS, PIIM UMR 7345, 13397, Marseille, France, EMMANUEL VILLERMAUX, Aix Marseille Université, CNRS, Centrale Marseille, IRPHE UMR 7342, 13384 — Premixed flames propagating in an initially quiescent medium undergo hydrodynamic instabilities that corrugate their shape, leading to non stationary cells. The shape of a flame is a critical issue as it rules its speed or the presence of incomplete reaction zones. We report here on experiments of premixed propane-air and methane-air flames freely propagating in a vertically oriented Hele-Shaw cell. In such configuration, the quasi two dimensional flames are easy to study by image analysis thanks to a high speed camera. The dynamics is favorably compared to numerical simulations of Michelson-Sivashinsky equation. The cell size distribution is analyzed and seems to be self similar whatever the gas mixture composition, provided that the dynamics is sufficiently rich, ie the flame is sufficiently unstable. We propose an explanation for this distribution. 8:52AM M34.00005 Parametrized mode decomposition for bifurcation analysis applied to a thermo-acoustically oscillating flame , TARANEH SAYADI, PETER SCHMID, Department of Mathematics, Imperial College London, FRANCK RICHECOEUR, DANIEL DUROX, EM2C Laboratory, Ecole Centrale Paris — Thermo-acoustic systems belong to a class of dynamical systems that are governed by multiple parameters. Changing these parameters alters the response of the dynamical system and causes it to bifurcate. Due to their many applications and potential impact on a variety of combustion systems, there is great interest in devising control strategies to weaken or suppress thermo-acoustic instabilities. However, the system dynamics have to be available in reduced-order form to allow the design of such controllers and their operation in real-time. As the dominant modes and their respective frequencies change with varying the system parameters, the dynamical system needs to be analyzed separately for a set of fixed parameter values, before the dynamics can be linked in parameter-space. This two-step process is not only cumbersome, but also ambiguous when applied to systems operating close to a bifurcation point. Here we propose a parametrized decomposition algorithm which is capable of analyzing dynamical systems as they go through a bifurcation, extracting the dominant modes of the pre- and post-bifurcation regime. The algorithm is applied to a thermo-acoustically oscillating flame and to pressure signals from experiments. A few selected mode are capable of reproducing the dynamics. 9:05AM M34.00006 Partial Extinction and the Rayleigh Index in Acoustically Driven Fuel Droplet Combustion1 , DARIO VALENTINI, University of Pisa, PHUOC HAI TRAN, BRETT LOPEZ, ARI EKMEKJI, OWEN SMITH, ANN KARAGOZIAN, UCLA — This experimental study examines burning liquid fuel droplets exposed to standing acoustic waves created within an atmospheric pressure waveguide. Building on prior studies which study relatively low-level excitation conditions in which the droplet is situated in the vicinity of a pressure node (PN), 2 the present experiments focus on higher amplitude excitation which can lead to periodic flame extinction. Phase-locked OH* chemiluminescence imaging reveals temporal oscillations in flame standoff distance from the droplet as well as chemiluminescent intensity in response to the applied acoustic perturbations. Temporal variation in the chemiluminescent intensity as well as pressure in the vicinity of the burning droplet enable quantification of combustionacoustic coupling via the Rayleigh index. While the sign of the Rayleigh index is consistent with oscillatory combustion during low-level acoustic excitation, when periodic partial extinction occurs at higher amplitude excitation, the Rayleigh index is insufficient to fully represent such coupling. Alternative metrics and methods are explored to enable a more robust study under such conditions. 1 Supported 2 Sevilla, by the University of Pisa and the UC CARE and MSD Scholars programs et al., Comb. Flame 161, pp. 1604-1619, 2014 9:18AM M34.00007 Finite amplitude wave interaction with premixed laminar flames , MOHAMAD ASLANI, JONATHAN D. REGELE, Iowa State University — The physics underlying combustion instability is an active area of research because of its detrimental impact in many combustion devices, such as turbines, jet engines, and liquid rocket engines. Pressure waves, ranging from acoustic waves to strong shocks, are potential sources of these disturbances. Literature on flame-disturbance interactions are primarily focused on either acoustics or strong shock wave interactions, with little information about the wide spectrum of behaviors that may exist between these two extremes. For example, the interaction between a flame and a finite amplitude compression wave is not well characterized. This phenomenon is difficult to study numerically due to the wide range of scales that need to be captured, requiring powerful and efficient numerical techniques. In this work, the interaction of a perturbed laminar premixed flame with a finite amplitude compression wave is investigated using the Parallel Adaptive Wavelet Collocation Method (PAWCM). This method optimally solves the fully compressible Navier-Stokes equations while capturing the essential scales. The results show that depending on the amplitude and duration of a finite amplitude disturbance, the interaction between these waves and premixed flames can produce a broad range of responses. 9:31AM M34.00008 Influence of flame-induced vorticity on acoustic wave generation , MATHIEU BLANCHARD, LadHyX - Ecole Polytechnique, PETER SCHMID, Department of Mathematics - Imperial College, DENIS SIPP, ONERA - Meudon, THIERRY SCHULLER, EM2C - Ecole Centrale Paris — An unsteady laminar premixed M-flame is examined using a linearized direct numerical simulation of a reactive compressible flow around a steady baseflow. Its response to a periodic acoustic forcing is considered. It is shown that the flame wrinkling process is associated with the generation and convection of vorticity waves. The impact of this vorticity on the upstream flow is examined. It is shown that vorticity waves have a strong impact on the flame tip dynamics. Results from optimal forcing computations are presented to illustrate this phenomenon. A simplified model equation, capturing the essential features of sound wave generation from vorticity, is then developed and analyzed. In particular, the impact of flame-induced unsteady vorticity on the generation of acoustic radiation at flame tip is emphasized. 9:44AM M34.00009 Blowoff characteristics of bluff-body stabilized syngas premixed flame in a meso-scale channel , BOK JIK LEE, HONG G. IM, King Abdullah University of Science and Technology, KAUST TEAM — Syngas has been actively studied recently for the application to Integrated Gasification Combined Cycle systems. It is also considered a candidate of fuels for combustion-based portable power-generating devices accompanied with a micro-reformer. In the present study, high-fidelity reacting flow simulations are conducted to investigate the instability near the blowoff limit of syngas premixed flames stabilized by a bluff-body in a meso-scale channel. Flames in a two-dimensional channel of 1 mm height and 10 mm length with a square box of 0.5 mm sides is considered. When the vortex shedding in a non-reacting flow at a mean inflow velocity remains symmetric as time passes, the flame at this inflow velocity tends to remain stable. By increasing the mean inflow velocity from the solution of this stable condition, the blowoff limit of a CO-to-H2 ratio is identified. At near-blowoff regime, the detail dynamics of flame instability and combustion characteristics associated to the instability are presented. The comparison with the simulations for lean hydrogen/air premixed flames is also discussed. 9:57AM M34.00010 Forced Response of Globally Unstable Reacting Wakes , BENJAMIN EMERSON, TIM LIEUWEN, Georgia Inst of Tech — In many practical combustors, a flame is stabilized in the confined wake of a bluff body. In such devices, the flame’s dynamics and its unsteady heat release are strongly governed by the fluid dynamics of the bluff body shear layers and wake. This unsteady heat release can couple with an acoustic mode of the combustor to cause a troublesome self-excited oscillation known as combustion instability. This coupling often occurs through the fluid dynamics, where the flame is dynamically wrinkled by acoustically excited vortical structures in the wake. This study experimentally investigates the acoustically excited hydrodynamic response of reacting bluff body wakes using time resolved PIV and chemiluminescence. The focus of the study is to understand how the flow responds to a varicose excitation on top of its globally unstable sinuous mode. In the experiment, the varicose mode is externally excited through harmonic, longitudinal acoustic forcing. The results show a varicose response. However, when forcing near the global mode frequency, the symmetrically arranged structures composing the varicose response quickly stagger to form a rapidly growing sinuous response. This resonant amplification of the sinuous mode is explained using linear spatial stability analysis and a bispectral analysis of the sinuous-varicose interaction. Tuesday, November 25, 2014 8:00AM - 10:10AM Session M35 Energy: Applications — 2001A - Mark Calaf, University of Utah 8:00AM M35.00001 A two-dimensional potential flow model of the interaction of a vortex ring passing over a flexible plate for energy harvesting applications , JIACHENG HU, SEAN PETERSON, Department of Mechanical and Mechatronics Engineering, University of Waterloo — Recent advancements in highly deformable smart materials have lead to increasing interest in small-scale energy harvesting research for powering low consumption electronic devices. One such recent experimental study by Goushcha et al. (APL, 2014) explored energy harvesting from a passing vortex ring by a cantilevered smart material plate oriented parallel to, and offset from, the path of the ring in an otherwise quiescent fluid. The present study focuses on modeling this experimental study using potential flow. The problem is modeled in two dimensions with the vortex ring represented as a pair of counter-rotating free vortices. Vortex parameters are determined to match convection speed of the ring and its pressure loading on the beam. The plateis approximated as a Kirchhoff-Love plate, and represented as a finite length vortex sheet in the fluid domain. The analytical model matches the experimentally measured strain at the clamped end of the beam well, suggesting that the model can be used as a tool to optimize this energy harvesting configuration. Results of a parametric study will be presented, as well as a discussion of the range of parameters for which the model is a good representation of the physical system. 8:13AM M35.00002 CFD Aided Design and Optimization of Francis Turbine Runners1 , FATMA AYANCIK, GIZEM DEMIREL, KUTAY CELEBIOGLU, ERDEM ACAR, SELIN ARADAG, TOBB University of Economics and Technology, ETU HYDRO RESEARCH CENTER TEAM — Francis turbines are commonly used for hydroelectric power plants with their wide range of flow rate and head values. They are composed of five main components and they generate energy with the help of the runner connected to the generator. Therefore, runner is the most important part of a Francis turbine. All components of turbines are linked and they are designed to maximize the turbine efficiency. The dimensions of the runner vary depending on the design discharge, head and the speed of the rotor of the generators. In this study, a design methodology is developed to design turbine runners with the help of computational fluid dynamics and is applied to the runner design of three different hydroelectric power plant turbines. Multi objective design optimization is also performed and the response surfaces are investigated to obtain maximum turbine efficiency and cavitation free design. 1 This study is financially supported by Turkish Ministry of Development. 8:26AM M35.00003 The effectiveness of a heated air curtain , DARIA FRANK, University of Cambridge — Air curtains are high-velocity plane turbulent jets which are installed in the doorway in order to reduce the heat and the mass exchange between two environments. The air curtain effectiveness E is defined as the fraction of the exchange flow prevented by the air curtain compared to the open-door situation. In the present study, we investigate the effects of an opposing buoyancy force on the air curtain effectiveness. Such an opposing buoyancy force arises for example if a downwards blowing air curtain is heated. We conducted small-scale experiments using water as the working fluid with density differences created by salt and sugar. The effectiveness of a downwards blowing air curtain was measured for situations in which the initial density of the air curtain was less than both the indoor and the outdoor fluid density, which corresponds to the case of a heated air curtain. We compare the effectiveness of the heated air curtain to the case of the neutrally buoyant air curtain. It is found that the effectiveness starts to decrease if the air curtain is heated beyond a critical temperature. Furthermore, we propose a theoretical model to describe the dynamics of the buoyant air curtain. Numerical results obtained from solving this model corroborate our experimental findings. 8:39AM M35.00004 Optimization of Transient Heat Exchanger Performance for Improved Energy Efficiency , GABRIELA BRAN ANLEU, PIROUZ KAVEHPOUR, ADRIENNE LAVINE, RICHARD WIRZ, University of California, Los Angeles — Heat exchangers are used in a multitude of applications within systems for energy generation, energy conversion, or energy storage. Many of these systems (e.g. solar power plants) function under transient conditions, but the design of the heat exchangers is typically optimized assuming steady state conditions. There is a potential for significant energy savings if the transient behavior of the heat exchanger is taken into account in designing the heat exchanger by optimizing its operating conditions in relation to the transient behavior of the overall system. The physics of the transient behavior of a heat exchanger needs to be understood to provide design parameters for transient heat exchangers to deliver energy savings. A numerical model was used to determine the optimized mass flow rates thermal properties for a thermal energy storage system. The transient behavior is strongly linked to the dimensionless parameters relating fluid properties, the mass flow rates, and the temperature of the fluids at the inlet of each stream. Smart metals, or advanced heat exchanger surface geometries and methods of construction will be used to meet the three goals mentioned before: 1) energy and cost reduction, 2) size reduction, and 3) optimal performance for all modes of operation. 8:52AM M35.00005 Numerical study of finned heat pipe-assisted latent heat thermal energy storage system , SAEED TIARI, SONGGANG QIU, MAHBOOBE MAHDAVI, Sustainable Energy and Energy Efficiency Laboratory, Department of Mechanical Engineering, Temple University, Philadelphia, PA — In the present study the thermal characteristics of a finned heat pipe-assisted latent heat thermal energy storage system are investigated numerically. A transient two dimensional finite volume based model employing enthalpy-porosity technique is implemented to analyze the performance of a thermal energy storage unit with square container and high melting temperature phase change material. The effects of heat pipe spacing, fin length and numbers as well as the influence of natural convection on the thermal response of the thermal energy storage unit have been studied. The obtained results reveal that the natural convection has considerable effect on the melting process of the phase change material. Increasing the number of heat pipes leads to the increase of melting rate and the decrease of base wall temperature. Also, the increase of fin length results in the decrease of temperature difference within the phase change material in the container, providing more uniform temperature distribution. Furthermore, it is showed that the number of fins does not affect the performance of the system considerably. 9:05AM M35.00006 Low-dimensional model of mixing in a liquid metal battery , DOUGLAS KELLEY, University of Rochester — Adding large-scale energy storage to Earth’s electrical grids would accommodate demand variations, enhance grid stability, and enable broad deployment of wind and solar generation. Liquid metal batteries are currently being commercialized as a promising and economically viable technology for grid-scale storage. Mass transport by mixing in their all-liquid electrodes affects battery performance, so predicting flow from known operating conditions (battery current and temperature) would allow for improved battery design. But accurate numerical simulation of these turbulent, three-dimensional, multi-phase flows, including electromagnetic forces and phase change, is challenging and computationally expensive. I will discuss a method for using experimental measurements to construct a simplified low-dimensional model with the potential to predict flow and battery performance. Initial results will also be presented. 9:18AM M35.00007 Hydrothermal Gasification for Waste to Energy , BRENDEN EPPS, MARK LASER, YEUNUN CHOO, Dartmouth College — Hydrothermal gasification is a promising technology for harvesting energy from waste streams. Applications range from straightforward waste-to-energy conversion (e.g. municipal waste processing, industrial waste processing), to water purification (e.g. oil spill cleanup, wastewater treatment), to biofuel energy systems (e.g. using algae as feedstock). Products of the gasification process are electricity, bottled syngas (H2 + CO), sequestered CO2, clean water, and inorganic solids; further chemical reactions can be used to create biofuels such as ethanol and biodiesel. We present a comparison of gasification system architectures, focusing on efficiency and economic performance metrics. Various system architectures are modeled computationally, using a model developed by the coauthors. The physical model tracks the mass of each chemical species, as well as energy conversions and transfers throughout the gasification process. The generic system model includes the feedstock, gasification reactor, heat recovery system, pressure reducing mechanical expanders, and electricity generation system. Sensitivity analysis of system performance to various process parameters is presented. A discussion of the key technological barriers and necessary innovations is also presented. 9:31AM M35.00008 Large eddy simulation of turbulent diffusion flame with hybrid fuel of CH4/H2 in various background conditions , SUNGMIN HONG, WOOK LEE, Dept. of Mechanical Engineering, Sogang University, Korea, HAN HO SONG, Dept. of Mechanical and Aerospace Engineering, Seoul National University, Korea, SEONGWON KANG, Dept. of Mechanical Engineering, Sogang University, Korea — A turbulent diffusion flame with hybrid fuel of methane and hydrogen is analyzed to investigate the effects of operating conditions on flame shape, rate of fuel consumption and pollutant formation. Various combinations of operating parameter, i.e. hydrogen concentration, background pressure and temperature, are examined in relatively high pressure and temperature conditions that can be found at the end of compression stroke in an internal combustion engine. A flamelet-progress variable approach (FPVA) and a dynamic subgrid scale (SGS) model are used for large eddy simulation (LES). A comparison with previous experiments and simulations in the standard condition shows a good agreement in the statistics of flow fields and chemical compositions, as well as in the resultant trends by similar parametric studies. As a result, the effects of added hydrogen are found to be consistent for most of the chemical species in the range of background pressure and temperature conditions. However, the flow fields of some species such as OH, NO, CO at a higher pressure and temperature state show a behavior different from the standard condition. Finally, hydrogen addition is shown to improve flame stability which is measured by the pressure fluctuations in all the tested conditions. 9:44AM M35.00009 A Non-linear Lifting Line Model for Design and Analysis of Trochoidal Propulsors , BERNARD ROESLER, BRENDEN EPPS, Dartmouth College — Flapping wing propulsors may increase the propulsive efficiency of large shipping vessels. A comparison of the design of a notional propulsor for a large shipping vessel with (a) a conventional ducted propeller versus (b) a flapping wing propulsor is presented. Calculations for flapping wing propulsors are performed using an open-source MATLAB software suite developed by the authors, CyROD, implementing an unsteady lifting-line model with free vortex wake roll-up to study the non-linear effects of foil-wake, and foil-foil interactions. Improvements to the traditional lifting line theory are made using further discretization of the wake vortex ring spacing near the trailing edge. Considerations of packaging options for a flapping wing propulsor on a large shipping vessel are presented, and compared with those for a conventional ducted propeller. 9:57AM M35.00010 An Experimental Investigation on the Wake Characteristics behind DualRotor Wind Turbines1 , ZHENYU WANG, AHMET OZBAY, WEI TIAN, ANUPAM SHARMA, HUI HU, Iowa State University, AEROSPACE ENGINEERING, IOWA STATE UNIVERSITY TEAM — We report an experimental study to investigate the aeromechanics and wake characteristics of dualrotor wind turbines (DRWTs) with co- and counter-rotating configurations, in comparison to those of a conventional single-rotor wind turbine (SRWT). The experiments were performed in a large-scale Aerodynamic/Atmospheric Boundary Layer (AABL) wind tunnel under neutral stability conditions. In addition to measuring the power output performance of DRWT and SRWT systems, static and dynamic wind loads acting on the SRWT and DRWT systems were also investigated. Furthermore, a high resolution PIV system was used for detailed wake flow field measurements (free-run and phase-locked) so as to quantify the characteristics of the turbulent turbine wake flow and to quantitatively visualize the transient behavior of the unsteady vortex structures in the wakes of DRWTs, in comparison with those behind a conventional SRWT systems. The detailed flow field measurements are correlated with the dynamic wind loads and power output measurements to elucidate underlying physics for higher total power yield and better durability of the wind turbines. 1 Funding support from the Iowa Energy Center with Grant No. 14-008-OG and National Science Foundation (NSF) with Grant No. CBET- 1438099 is gratefully acknowledged. Tuesday, November 25, 2014 8:00AM - 9:57AM Session M36 Nano Flows II — Alcove A - Sindy Tang, Stanford University 8:00AM M36.00001 Viscoelastic Flows in Simple Liquids Generated by Vibrating Nanostructures , JOHN SADER, The University of Melbourne, MATTHEW PELTON, University of Maryland, Baltimore County, DEBADI CHAKRABORTY, The University of Melbourne, EDWARD MALACHOSKY, PHILIPPE GUYOT-SIONNEST, University of Chicago — Newtonian fluid mechanics, in which the shear stress is proportional to the strain rate, is synonymous with the flow of simple liquids like water. We report the measurement and theoretical verification of non-Newtonian, viscoelastic flow phenomena produced by the high-frequency (>20 GHz) vibration of gold nanoparticles immersed in water-glycerol mixtures. The observed viscoelasticity is not due to molecular confinement, but is a bulk continuum effect arising from the short time scale of vibration. This represents the first direct mechanical measurement of the intrinsic viscoelastic properties of simple bulk liquids, and opens a new paradigm for understanding extremely high frequency fluid mechanics, nanoscale sensing technologies, and biophysical processes. 8:13AM M36.00002 Nanofluid Flow and Heat Transfer in Channel Entrance Region , JOSEPH T.C. LIU, School of Engineering, Brown University, GIANLUCA PULITI, Department of Aerospace and Mechanical Engineering, Notre Dame University — The present work uses the continuum description of nanofluid flow to study the flow, heat and mass transfer in the entrance and developing region of channels or tubes, where the viscous and heat conduction layers are thin and the heat transfer is much more intense than fully developed flow. Instead of supplementing the formulation with thermodynamic properties based on mixture calculations, use is made of recent molecular dynamical computations of such properties, specifically, the density and heat capacity of gold-water nanofluids. The more general formulation results, within the Rayleigh-Stokes (plug flow) approximation and perturbation for small volume fraction, show that the nanofluid density-heat capacity has an enormous effect in the inertia mechanism in causing the nanofluid temperature profile to steepen. The nanofluid thermal conductivity though has an explicit enhancement of the surface heat transfer rate has the almost hidden effect of stretching the nanofluid temperature profile thus giving the opposite effect of enhancement. Quantitative results for Gold-Water nanofluid is presented. 8:26AM M36.00003 Microscopic Description of Resonance in the Brownian Motion of Hydrophobic Nanoparticle in Harmonic Potential Trap1 , JAE HYUN PARK, Gyeongsang National University — Harmonic potential has been popular for the trapping of micro- and nanoparticles (e.g. optical tweezer). With the rapid development of harmonic potential trapping technology, its application is nowadays being extended to explore the fundamental nature in the random thermal fluctuation of particles in order to confirm the classical theory of Brownian motion. In this study, using extensive molecular dynamics simulations, we investigate the molecule-level features of dynamic response of hydrophobic C60 nanoparticle in harmonic potential trap with water medium. The time-averaged magnitudes of random fluctuation are measured for various trap stiffness and then the virtual mass, the amount of fluid moving together with particle, is extracted from curve fitting. The fluctuation is proportional to the inverse of trap stiffness. The virtual mass is mostly originated from the first hydration shell around the particle and it is not influenced by the stiffness. The resonance in frequency domain is observed as a result of coloured noise in the motion. The effect of stiffness on the resonance is weaker than that on the magnitude of fluctuation because the motion of particle is partially dissipated in the RDF valley between the first and the second hydration shell. 1 This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012R1A1A1042920). 8:39AM M36.00004 Translational and rotational diffusion of a single Janus nanoparticle in an explicit solvent1 , ALI KHARAZMI, Michigan State University, NIKOLAI PRIEZJEV, Wright State University — Molecular dynamics simulations are carried out to study the translational and rotational diffusion of a Janus particle in a Lennard-Jones fluid. We consider a spherical particle with two hemispheres of different wettability. The analysis of the particle dynamics is based on the time-dependent orientation tensor, particle displacement, as well as the translational and angular velocity autocorrelation functions. We show that both translational and rotational diffusion coefficients increase with decreasing surface energy at the nonwetting hemisphere. It was found that in contrast to uniform particles, the nonwetting hemisphere of the Janus particle rotates in the direction of the displacement vector during the rotational relaxation time. 1 NSF CBET-1033662 8:52AM M36.00005 Improved Proper Orthogonal Decomposition for Noise Reduction in Particle Flow Simulations1 , MALGORZATA ZIMON, University of Strathclyde, JASON REESE, University of Edinburgh, DAVID EMERSON, STFC Daresbury Laboratory — Proper orthogonal decomposition (POD), widely utilised for turbulent flows, has recently been explored for processing particle data. An extension of the method based on time-windows offers a useful approach for noise reduction in particle simulations. However, to successfully remove statistical noise from the system, large amounts of data need to be provided. Moreover, POD can fail to improve the quality of an ensemble mean (statistical average) when applied to steady-state simulations. In order to achieve a better efficiency of POD in processing non-stationary fields, we have combined the method with wavelet-based filtering. In this new procedure, the wavelet thresholding is performed within POD’s domain. In case of stationary problems, we will show how effectively POD can be applied to a matrix constructed from the mean, following the application of singular spectral analysis (SSA). The combination of POD and SSA is shown to successfully smooth time-dependent observables. Simulations were undertaken to illustrate the performance of the new tools applied to noisy velocity and density fields. Numerical examples include molecular dynamics, dissipative particle dynamics simulations of force-driven fluid flows and phase separation phenomena. 1 The research has received funding from the UK’s Engineering and Physical Sciences Research Council (EPSRC) under Grant EP/I011927/1. Results were obtained using the ARCHIE-WeSt High Performance Computer, under EPSRC grant EP/K000586/1. 9:05AM M36.00006 The filtration of colloidal gold nanopartilces in nanoporous media1 , FRANCISCUS DE JONG, MICHAEL SCHLUETER, Hamburg University of Technology - Institute for Multiphase Flows, INSTITUTE FOR MULTIPHASE FLOWS TEAM2 — Deep-bed filtration is connected to a wide variety of disciplines ranging from biology and medicine to engineering. A novel material with promising perspectives that can be used for deep-bed filtration are forests consisting of numerous multi-walled carbon nanotubes (MWCNTs). The filtration kinetics of particles within deep-beds is usually addressed using global investigations (i.e. the concentration of particles in the bulk solution). However, to optimize MWCNT forests for filtration purposes detailed information of the local filtration kinetics is indispensible. In the study presented here microbeam small-angle X-ray scattering (muSAXS) is used, for the first time, to study both, the spatial and the temporal local filtration kinetics of small-sized particles within MWCNT forests. The filtration is (globally) verified based on (I) scanning electron microscopy and (II) inductively coupled plasma atomic emission spectroscopy (ICPAES). Good agreement is observed between the local and the global measurements (i.e. a difference of 4.3%). The use of muSAXS to understand the local filtration kinetics of submicrometer particles opens up pathways to effectively optimize functionalized MWCNT forests and prepare them for specific filtering purposes. 1 We acknowledge the Max-Buchner Research Foundation for the financial support for project 2936 University of Technology 2 Hamburg 9:18AM M36.00007 Static and Alternating Field Magnetic Capture and Heating of Iron Oxide Nanoparticles in Simulated Blood Vessels1 , JOANNE HAEUN LEE, The City College of New York, CUNY, RHYTHM R. SHAH, CHRISTOPHER S. BRAZEL, The University of Alabama — Targeted drug delivery and localized hyperthermia are being studied as alternatives to conventional cancer treatments, which can affect the whole body and indiscriminately kill healthy cells. Magnetic nanoparticles (MNPs) have potential as drug carriers that can be captured and trigger hyperthermia at the site of the tumor by applying an external magnetic field. This study focuses on comparing the capture efficiency of the magnetic field applied by a static magnet to an alternating current coil. The effect of particle size, degree of dispersion, and the frequency of the AC field on capture and heating were studied using 3 different dispersions: 16 nm maghemite in water, 50 nm maghemite in dopamine, and 20-30 nm magnetite in dimercaptosuccinic acid. A 480G static field captured more MNPs than a similar 480G AC field at either 194 or 428 kHz; however, the AC field also allowed heating. The MNPs in water had a lower capture and heating efficiency than the larger, dopamine-coated MNPs. This finding was supported by dynamic light scattering data showing the particle size distribution and vibrating sample magnetometry data showing that the larger MNPs in the dopamine solution have a higher field of coercivity, exhibit ferrimagnetism and allow for better capture while smaller (16 nm) MNPs exhibit superparamagnetism. The dispersions that captured the best also heated the best. 1 NSF ECE Grant #1358991 supported the first author as an REU student. 9:31AM M36.00008 Pore dynamics in a liquid membrane1 , ALEXANDER NEPOMNYASHCHY, Department of Mathematics, Technion - Israel Institute of Technology, VLADIMIR VOLPERT, Department of Engineering Sciences and Applied Mathematics, Northwestern University — It is known that vesicles formed from lipid bilayer membranes are used for transportation of a toxic drug to a target, where the drug is released by pore creation. The pores in a membrane show a rather nontrivial dynamics, which thus far has been studied by means of simplified models. In the present talk, we describe the pore dynamics in a stretched membrane, which is considered as a two-dimensional viscous or viscoelastic liquid medium surrounded by a three-dimensional ambient viscous liquid. In the case of a viscoelastic membrane, a Lagrangian approach, which allows to account for large displacements, is applied. A closed equation for the pore radius is derived and investigated. 1 The work has been partially supported by the US-Israel Binational Science Foundation (grant No. 2008122). 9:44AM M36.00009 Effect of nanostructures and electrostatic interactions on disjoining pressure of ultra-thin liquid film , HAN HU, CHRISTOPHER WEINBERGER, YING SUN, Drexel Univ — Disjoining pressure, the excess pressure that stems from the long-range intermolecular interactions, plays a key role in the stability of thin films in applications such as lubrication, wetting, boiling, condensation and evaporation. In recent years, nanostructures have been introduced as a means to control the stability of thin films. However, the classic theory of disjoining pressure assumes atomically smooth surface and neglects the electrostatic interactions. In the present study, the effect of nanostructures and electrostatic interactions on disjoining pressure is examined with combined modeling and molecular dynamics simulations. A model of meniscus shape and disjoining pressure for a thin liquid film on a nanostructured surface is derived based on minimization of system free energy and Derjaguin approximation. The scaled healing length ξ/D (D the nanostructure depth) is used to characterize the competition between the liquid surface tension and solid-liquid intermolecular forces. The result shows disjoining pressure increases with D. The model prediction agrees well with molecular dynamics simulations for a water-gold system. The electrostatic interactions enhance the disjoining pressure effect but the strength of the electrostatic interactions becomes weaker as the aspect ratio of the nanostructures increases. Tuesday, November 25, 2014 10:30AM - 11:05AM — Session N8 Invited Session: Transition to Turbulence in a Soft-Walled Microchannel 3001/3003 - Peter Schmid, Imperial College, London 10:30AM N8.00001 Transition to turbulence in a soft-walled microchannel1 , VISWANATHAN KUMARAN, Indian Institute of Science — It is well known that the transition from a laminar to a turbulent flow in a rigid tube takes place at a Reynolds number of about 2100, and in a rigid channel at a Reynolds number of about 1200. Experimental results are presented to show that the transition Reynolds number could be as low as low as 200 in micro-channels of height 100 microns with a soft wall, provided the elasticity modulus of the wall is sufficiently low. At the point of transition, motion is observed in the walls of the channel, indicating that the instability is caused by a dynamical coupling between the fluid and the wall dynamics, which is qualitatively different from that in rigid tubes/channels. Theoretical calculations show that the transition Reynolds number depends on a dimensionless parameter Σ = (ρGR2 /µ2 ), where, ρ and µ are the fluid density and viscosity, G is the elastic modulus of the wall material, R is the cross-stream length scale and V is the maximum velocity. A low Reynolds number analysis indicates that there could be a transition even at zero Reynolds number when the dimensionless parameter (V µ/GR) exceeds a critical value. The mechanism of destabilisation is the transfer of energy from the mean flow to the fluctuations due to the shear work done at the fluid-solid interface. Two different types of instabilities are identified at high Reynolds number using asymptotic analysis, the inviscid mode instability for which the critical Reynolds number sales as Σ1/2 , and the wall mode instability for which the critical Reynolds number scales as Σ3/4 . Numerical continuation is used to extend the results to the intermediate Reynolds number regime. The low Reynolds number analysis is found to be in quantitative agreement with experiments. However, the high Reynolds number analysis is in agreement only if the wall deformation and consequent flow modification due to the applied pressure gradient is incorporated in the analysis. The flow after transition has transport coefficients up to five orders of magnitude higher than that in laminar flows, and so the transition could be used to significantly enhance transport rates in microfluidic and biomedical applications. 1 The author would like to thank the Department of Science and Technology, Government of India for financial support. Tuesday, November 25, 2014 10:30AM - 11:05AM — Session N14 Invited Session: Sand Waves in Environmental Flows - Insights Gained by LES 3009/3011 - Stephen Monismith, Stanford University 10:30AM N14.00001 Sand Waves in Environmental Flows: Insights gained by LES1 , FOTIS SOTIROPOULOS, University of Minnesota — In fluvial and coastal environments, sediment transport processes induced by near-bed coherent structures in the turbulent boundary layer developing over a mobile sediment bed result in the formation of dynamically rich sand waves, or bed forms, which grow and migrate continuously. Bed form migration alters streambed roughness and provides the primary mechanism for transporting large amounts of sediment through riverine systems impacting the morphology, streambank stability, and ecology of waterways. I will present recent computational advances, which have enabled coupled, hydro-morphodynamic large-eddy simulation (LES) of turbulent flow in mobile-bed open channels. Numerical simulations: 1) elucidate the role of near-bed sweeps in the turbulent boundary layer as the mechanism for initiating the instability of the initially flat sand bed; 2) show how near-bed processes give rise to aperiodic eruptions of suspended sediment at the free surface; and 3) clarify the mechanism via which sand waves migrate. Furthermore, in agreement with recent experimental observations, the computed spectra of the resolved velocity fluctuations above the bed exhibit a distinct spectral gap whose width increases with distance from the bed. The spectral gap delineates the spectrum of turbulence from that of slowly evolving coherent structures associated with sand wave migration. The talk will also present computational results demonstrating the feasibility of carrying out coupled, hydro-morphodynamic LES of large dunes migrating in meandering streams and rivers with embedded hydraulic structures and discuss future challenges and opportunities. 1 This work was supported by NSF Grants EAR-0120914 and EAR-0738726, and National Cooperative Highway Research Program Grant NCHRP-HR 24-33. Tuesday, November 25, 2014 11:10AM - 11:30AM Session P8 Invited Session: Francois Frenkiel Award — 3001/3003 - Sharath Girimaji, Texas A&M University 11:10AM P8.00001 Francois N. Frenkiel Award Talk: Relevance of the Thorpe length scale in stably stratified turbulence1 , BENJAMIN D. MATER, SIMON M. SCHAAD, S. KARAN VENAYAGAMOORTHY, Colorado State University — A relatively simple and objective measure of large-scale vertical overturns in turbulent oceanic flows is the Thorpe length scale, LT . Reliance on common scaling between the Ozmidov length scale LO (which is a measure of the size of largest eddy unaffected by buoyancy in stratified turbulence) and LT is commonplace in the field of oceanography to infer the dissipation rate of turbulent kinetic energy ε. In this study, we use direct numerical simulations (DNS) of stably stratified turbulence to compare the Thorpe overturn length scale, LT , with other length scales of the flow that can be constructed from large-scale quantities fundamental to shear-free, stratified turbulence. Quantities considered are the turbulent kinetic energy, k, its dissipation rate, ε, and the buoyancy frequency, N . Fundamental length scales are then the Ozmidov length scale, LO , the isotropic large scale, Lkε , and a kinetic energy length scale, LkN . Behavior of all three fundamental scales, relative to LT , is shown to be a function of the buoyancy strength parameter N TL , where TL = k/ε is the turbulence time scale. When buoyancy effects are dominant (i.e., for N TL > 1), LT is shown to be linearly correlated with LkN and not with LO as is commonly assumed for oceanic flows. Agreement between LO and LT is only observed when the buoyancy and turbulence time scales are approximately equal (i.e., for the critical case when N TL ≈ 1). The relative lack of agreement between LT and LO in strongly stratified flows is likely due to anisotropy at the outer scales of the flow where the energy transfer rate differs from ε. The key finding of this work is that observable overturns in strongly stratified flows are more reflective of k than ε. In the context of oceanic observations, this implies that inference of k, rather than ε, from measurements of LT is fundamentally correct when N TL ≈ 1 and most appropriate when N TL > 1. Furthermore, we show that for N TL < 1, LT is linearly correlated with Lkε , when mean shear is absent. 1 Support from ONR and NSF is gratefully acknowledged. Tuesday, November 25, 2014 11:10AM - 11:30AM Session P14 Invited Session: Andreas Acrivos Dissertation Award — 3009/3011 - Kausik Sarkar, George Washington University 11:10AM P14.00001 Andreas Acrivos Dissertation Award: Onset of Dynamic Wetting Failure - The Mechanics of High-Speed Fluid Displacement , ERIC VANDRE, University of Minnesota — Dynamic wetting is crucial to processes where a liquid displaces another fluid along a solid surface, such as the deposition of a coating liquid onto a moving substrate. Dynamic wetting fails when process speed exceeds some critical value, leading to incomplete fluid displacement and transient phenomena that impact a variety of applications, such as microfluidic devices, oil-recovery systems, and splashing droplets. Liquid coating processes are particularly sensitive to wetting failure, which can induce air entrainment and other catastrophic coating defects. Despite the industrial incentives for careful control of wetting behavior, the hydrodynamic factors that influence the transition to wetting failure remain poorly understood from empirical and theoretical perspectives. This work investigates the fundamentals of wetting failure in a variety of systems that are relevant to industrial coating flows. A hydrodynamic model is developed where an advancing fluid displaces a receding fluid along a smooth, moving substrate. Numerical solutions predict the onset of wetting failure at a critical substrate speed, which coincides with a turning point in the steady-state solution path for a given set of system parameters. Flow-field analysis reveals a physical mechanism where wetting failure results when capillary forces can no longer support the pressure gradients necessary to steadily displace the receding fluid. Novel experimental systems are used to measure the substrate speeds and meniscus shapes associated with the onset of air entrainment during wetting failure. Using high-speed visualization techniques, air entrainment is identified by the elongation of triangular air films with system-dependent size. Air films become unstable to thickness perturbations and ultimately rupture, leading to the entrainment of air bubbles. Meniscus confinement in a narrow gap between the substrate and a stationary plate is shown to delay air entrainment to higher speeds for a variety of water/glycerol solutions. In addition, liquid pressurization (relative to ambient air) further postpones air entrainment when the meniscus is located near a sharp corner along the plate. Recorded critical speeds compare well to predictions from the model, supporting the hydrodynamic mechanism for the onset of wetting failure. Lastly, the industrial practice of curtain coating is investigated using the hydrodynamic model. Due to the complexity of this system, a new computational approach is developed combining a finite element method and lubrication theory in order to improve the efficiency of the numerical analysis. Results show that the onset of wetting failure varies strongly with the operating conditions of this system. In addition, stresses from the air flow dramatically affect the steady wetting behavior of curtain coating. Ultimately, these findings emphasize the important role of two-fluid displacement mechanics in high-speed wetting systems. Tuesday, November 25, 2014 1:05PM - 2:49PM Session R1 General Fluid Dynamics IV — 3000 - Ralph Metcalfe, University of Houston 1:05PM R1.00001 Size-Dependent Couple Stress Fluid Mechanics: The Influence of Boundary Conditions , AREZOO HAJESFANDIARI, ALI HADJESFANDIARI, GARY DARGUSH, University at Buffalo, State University of New York — In size- dependent couple stress fluid mechanics, which involves a length parameter l, the corresponding modified Navier-Stokes equations are ρ Dv = −∇p + µ ∇2 v − Dt µl2 ∇2 ∇2 v. The term involving l is of fourth order, which then requires the prescription of additional boundary conditions compared to the classical case. Therefore, the boundary conditions in the size-dependent theory must include specification of either the tangential component of rotations ω on the boundary or the tangential moment-tractions m(n) . Here we concentrate on two-dimensional flows and explore the consequences of prescribing different boundary conditions in size-dependent couple-stresses fluid mechanics by using computational fluid dynamics. We investigate the characteristics of flow for the cavity problem based upon the equation above and the Boussinesq approximation for the Rayleigh-Benard problem. This provides us with interesting, unexpected results for various boundary conditions, when accounting for couple-stresses. These in turn might explain different mechanisms for energy dissipation, as well as for chaotic behaviors of fluid flow. 1:18PM R1.00002 Geometric and Dynamic Skewness in Passive Scalar Transport1 , RICHARD MCLAUGHLIN, MANUCHEHR AMINIAN, FRANCESCA BERNARDI, ROBERTO CAMASSA, UNC Chapel Hill, UNC JOINT FLUIDS GROUP TEAM — The evolution of a passive scalar in laminar shear flow is revisited in channel, pipe, and box geometries. Exact, explicit closed form, single sum formulae for the evolution of the skewness of a passive scalar along span-wise slices are derived and studied analytically in the case of channel flow. The largest skewness in time is interpreted using a geometric quantity. Surprisingly, the geometric quantity is seen to be absent in the smooth pipe geometries, but present in the box geometry, providing insight into the role of the wall mode versus center mode in assigning the sign to the instantaneous averaged skewness. 1 NSF and ONR 1:31PM R1.00003 Relativistic fluids: fundamentals and recent developments1 , A. SANDOVALVILLALBAZO, Department of Physics and Mathematics, Universidad Iberoamericana, A.L. GARCÍA-PERCIANTE, Department of Applied Mathematics and Systems, Universidad Autónoma Metropolitana-Cuajimalpa — Relativistic thermodynamics and kinetic theory have been subjects of intense research and debate recently. The topic has gained attention primarily due to its application in both astrophysical and experimental scenarios. In this talk I will review some of the challenges theorists have faced in search of a successful formalism capable of describing these systems and the alternatives proposed in order to avoid the well known instabilities and causality problems present in the first works on the subject published more than fifty years ago. Among these proposals I will focus on the first order in the gradients version of relativistic kinetic theory in order to describe special relativistic single component fluids in the presence of external forces. The main results obtained following this path will be shown including the relativistic expressions for dissipative fluxes and entropy production. Some consequences of relativistic modifications in the hydrodynamic equations will also be discussed. 1 This work is supported by CONACyT through grant CB2011/167563. 1:44PM R1.00004 Shape optimization for continuum with H1 gradient method , TAKASHI NAKAZAWA, Mathematical Institute, Graduate school of Science, Tohoku University, Japan — In this presentation, shape optimization of continuum is operated in the domain of interest which Navier-Stokes equation governs. And more, DNS is operated for the domain before and after shape optimization. By comparing the results of DNS in the before and after shape optimized domains, it is discussed that which domain is better. 1:57PM R1.00005 Numerical simulation of flow in a horizontal channel with multiple cross-flow inlets1 , PRANAB N. JHA, Dept. of Mechanical Engineering, University of Houston, CHUCK SMITH, Apache Corp., Houston, RALPH W. METCALFE, Dept. of Mechanical Engineering, University of Houston — Flow in a horizontal channel with multiple cross-flow inlets was studied numerically. Based on Reynolds and Mach number analysis of data obtained from a horizontal natural gas well having 31 completion stages, measured at two different times in the production cycle, it was determined that an incompressible flow model may be applied to study a large fraction of the wellbore. Using five cross-flow inlets, the existence of three basic flow regimes - trickle flow, partially blocked flow and fully blocked flow - were identified with respect to the blocking of upstream inlets by the downstream ones. The existence of these flow regimes is consistent with field data. A lumped-parameter model for pressure drop was used to simulate large axial distances between two inlets. A pressure boundary condition was employed at each inlet to simulate a linearly depleting reservoir. This was used to study the dynamic interaction between the inlets in the channel. The characteristic time scales related to the transient depletion were identified and analyzed. The transition of flow regimes is consistent with the trends observed from field data and gives an insight into the behavior of horizontal wells. 1 Supported in part by Apache Corp., Houston 2:10PM R1.00006 Vibration-Induced Rectified Motion of a Piston in a Liquid-Filled Cylinder with Bellows to Mimic Gas Regions , J.R. TORCZYNSKI, L.A. ROMERO, J.R. CLAUSEN, T.J. O’HERN, Sandia National Laboratories — The motion of a piston within a cylinder is investigated. A spring suspends the piston against gravity. The cylinder is filled with a viscous liquid and has compressible bellows at the top and bottom to mimic gas regions. A fixed post with protrusions extends into a hole through the piston with opposing protrusions. The length of the gap formed by the protrusions depends on the piston’s vertical position. The outer gap between the piston and the cylinder is extremely small. Hence, as the piston moves, the displaced liquid passes through the variable-length gap, and the liquid force on the piston depends on its position. When this system is subjected to vertical vibrations, this dependence can produce a nonzero net force. With bellows, this net force can become large enough for the piston to compress the spring. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. 2:23PM R1.00007 Simulating Rectified Motion of a Piston in a Housing Subjected to Vibrational Acceleration , JONTHAN CLAUSEN, JOHN TORCZYNSKI, LOUIS ROMERO, TIMOTHY O’HERN, Sandia National Laboratories — We employ ALE finite element simulations to investigate the behavior of a piston in a housing subjected to vertical vibrations. The housing is filled with a viscous liquid to damp the piston motion and has bellows at both ends to represent air bubbles present in real systems. The piston has a roughly cylindrical hole along its axis, and a post attached to the housing penetrates partway into this hole. Protrusions from the hole and the post form a gap with a length that varies as the piston moves and forces liquid through this gap. Under certain conditions, nonlinearities in the system can drive the piston to move downward and compress the spring that holds it up against gravity. This behavior is investigated using ALE finite element simulations, and these results are compared with theoretical predictions. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. 2:36PM R1.00008 Optimal Dynamics of Intermittent Water Supply , ANNA LIEB, JON WILKENING, University of California, Berkeley, CHRIS RYCROFT, Harvard University — In many urban areas of the developing world, piped water is supplied only intermittently, as valves direct water to different parts of the water distribution system at different times. The flow is transient, and may transition between free-surface and pressurized, resulting in complex dynamical features with important consequences for water suppliers and users. These consequences include degradation of distribution system components, compromised water quality, and inequitable water availability. The goal of this work is to model the important dynamics and identify operating conditions that mitigate certain negative effects of intermittent water supply. Specifically, we will look at valve parameters occurring as boundary conditions in a network model of transient, transition flow through closed pipes. Optimization will be used to find boundary values to minimize pressure gradients and ensure equitable water availability. Tuesday, November 25, 2014 1:05PM - 3:28PM Session R2 CFD: Turbulence Modeling II — 3002 - Katherine Lundquist, Lawrence Livermore National Laboratory 1:05PM R2.00001 Modelling the cut-off resolution parameter in the PANS method for turbulence simulation , BRANISLAV BASARA, AVL List GmbH, KEMAL HANJALIC, Delft University of Technology — The Partially-Averaged Navier-Stokes (PANS) approach, designed to resolve a part of the turbulence spectrum, adjusts seamlessly from the Reynolds-Averaged Navier-Stokes (RANS) to the Direct Numerical Solution (DNS) of the Navier-Stokes equations. This turbulence closure, derived from a RANS model, supports any filter width or scale resolution. We choose the PANS model as the basis for the present analysis of options for the model resolution parameter, but the conclusions derived are applicable to other partially resolved closure methods. Namely, in the conventional well-established PANS, the resolution parameter is obtained from the grid spacing and the integral turbulence length scale. The latter is obtained usually by summing up the resolved turbulence, while the unresolved motion is computed from the modelled equation. Several formulations have been shown to provide reliable and accurate results for many test flows. However, serious impediments have been noted in some applications such as moving domains and transient boundaries because too long calculations of the average velocity make this approach impractical. We analysed some recent alternative approaches which use the turbulent-to-mean-strain-rate time scale aimed at avoiding the on-line calculations of the resolved kinetic energy required for calculations of the input resolution parameter. Comparisons of several approaches will be shown in detail and conclusions drawn on the merits of each method. 1:18PM R2.00002 A novel VLES model accounting for near-wall turbulence: physical rationale and applications , SUAD JAKIRLIC, CHI-YAO CHANG, LUKAS KUTEJ, CAMERON TROPEA, Technische Universitaet Darmstadt — A novel VLES (Very Large Eddy Simulation) model whose non-resolved residual turbulence is modelled by using an advanced near-wall eddy-viscosity model accounting for the near-wall Reynolds stress anisotropy influence on the turbulence viscosity by modelling appropriately the velocity scale in the relevant formulation (Hanjalic et al., 2004) is proposed. It represents a variable resolution Hybrid LES/RANS (Reynolds-Averaged Navier–Stokes) computational scheme enabling a seamless transition from RANS to LES depending on the ratio of the turbulent viscosities associated with the unresolved scales corresponding to the LES cut-off and the ‘unsteady’ scales pertinent to the turbulent properties of the VLES residual motion, which varies within the flow domain. The VLES method is validated interactively in the process of the model derivation by computing fully-developed flow in a plane channel (important representative of wall-bounded flows, underlying the log-law for the velocity field, for studying near-wall Reynolds stress anisotropy) and a separating flow over a periodic arrangement of smoothlycontoured 2-D hills. The model performances are also assessed in capturing the natural decay of the homogeneous isotropic turbulence. The model is finally applied to swirling flow in a vortex tube, flow in an IC-engine configuration and flow past a realistic car model. 1:31PM R2.00003 A new algebraic wall model for LES based on the momentum integral approach: formulation and sample applications1 , CHARLES MENEVEAU, XIANG YANG, JASIM SADIQUE, RAJAT MITTAL, Johns Hopkins University — Inspired by the momentum integral boundary layer method of von Karman and Pohlhausen (VKP), we propose an integral Wall Model for LES (iWMLES). To capture near wall physics without assuming equilibrium conditions, a velocity profile with various parameters is proposed instead of numerical integration of the boundary layer equation in the near-wall zone. Since numerical solution of boundary layer equations on a refined mesh near the wall is not required, we preserve the essential simplicity of equilibrium-type wall models with a cost that is independent of Reynolds number. Two sets of test cases are presented here: (1) Fully developed half channel flows with a smooth wall at various Reynolds numbers, and with a rough wall in which roughness elements are numerically not resolved. The code we use for these cases is a pseudo-spectral code for fully developed channel flow with a Lagrangian dynamic subgrid model. (2) LES of flow over surface mounted cubes in a fully developed half channel for which detailed experimental data are available. A finite difference LES code with sharp immersed boundary method and dynamic Vreman eddy-viscosity model is used in this application. Results show that iWMLES provides a practical and accurate wall model for predicting the mean wall stress in LES. 1 This work is funded by ONR grant N00014-12-1-0582 (Dr. R. Joslin, program manager). 1:44PM R2.00004 Parallel Optimization with Large Eddy Simulations , CHAITANYA TALNIKAR, PATRICK BLONIGAN, Massachusetts Inst of Tech-MIT, JULIEN BODART, University of Toulouse, ISAE, QIQI WANG, Massachusetts Inst of Tech-MIT, ALEX GORODETSKY COLLABORATION, JASPER SNOEK COLLABORATION — For design optimization results to be useful, the model used must be trustworthy. For turbulent flows, Large Eddy Simulations (LES) can capture separation and other phenomena that traditional models such as RANS struggle with. However, optimization with LES can be challenging because of noisy objective function evaluations. This noise is a consequence of the sampling error of turbulent statistics, or long time averaged quantities of interest, such as the drag of an airfoil or heat transfer to a turbine blade. The sampling error causes the objective function to vary noisily with respect to design parameters for finite time simulations. Furthermore, the noise decays very slowly as computational time increases. Therefore, robustness with noisy objective functions is a crucial prerequisite to optimization candidates for LES. One way of dealing with noisy objective functions is to filter the noise using a surrogate model. Bayesian optimization, which uses Gaussian processes as surrogates, has shown promise in optimizing expensive objective functions. The following talk presents a new approach for optimization with LES incorporating these ideas. Applications to flow control of a turbulent channel and the design of a turbine blade trailing edge are also discussed. 1:57PM R2.00005 Optimization of a Turbine Blade Trailing Edge using Large Eddy Simulations , PATRICK BLONIGAN, CHAITANYA TALNIKAR, Massachusetts Inst of Tech-MIT, JULIEN BODART, University of Toulouse, ISAE, BRIAN PIERCE, Stanford University, SANJEEB BOSE, Cascade Technologies, QIQI WANG, Massachusetts Inst of Tech-MIT — As for many turbomachinery components, heat transfer and pressure loss are the key quantities influencing the design of turbine blades. To compute correct heat transfer and pressure loss data, flow features such as boundary layer transition and flow separation must be captured accurately. While traditional Computation Fluid Dynamics models such as Reynolds Averaged Navier-Stokes (RANS) struggle to capture these features accurately, Large Eddy Simulation (LES) is able to. This talk discusses an optimization study of a turbine blade trailing edge. The design of turbine blades involves two classical competing objectives: minimizing pressure loss and minimizing heat transfer to the blade. This trade-off is especially apparent for the design of the blade’s trailing edge. The study was conducted using a novel Bayesian optimization technique developed by the authors. The optimization algorithm is combined with a massively parallel LES solver and the results for a number of trailing edge designs including the optimal geometry will be presented and their implications for turbine blade design will be discussed. 2:10PM R2.00006 Modeling of near-surface generated turbulence in large-eddy simulation of microscale atmospheric flows , REY DELEON, University of Idaho, Boise State University, INANC SENOCAK, Boise State University — Large-eddy simulation (LES) is often used in microscale atmospheric boundary layer (ABL) flows. As a wall-resolved LES is not relevant in actual ABL flows due to surface roughness and very high Reynolds numbers, LES with wall-modeling has been widely adopted. But special attention must be given to the near-surface treatment in LES of ABL flows as several of the more commonly applied methods, e.g. hybrid RANS/LES or models, can provide unrealistic accelerations leading to the log-layer mismatch problem. Numerous studies have focused on the smooth-wall turbulent channel flow to address the log-law mismatch problem. However, several of these studies have yet to be extended to LES of ABL flows where terrain surface is aerodynamically rough and arbitrarily complex. We investigate different near-surface treatments to ABL flow in our GPU-accelerated LES framework with an immersed boundary method for complex terrain. We consider the constrained LES approach of Chen et al. (2012), and the mean wall shear stress boundary condition proposal of Lee et al. (2013) that have shown promising results. Additionally, we investigate a hybrid RANS/LES approach. Our goal is to identify a suitable near-surface treatment and extend it to LES of complex terrain winds using the immersed boundary method. 2:23PM R2.00007 A DG-FDF Large Eddy Simulator , SHERVIN SAMMAK, University of Pittsburgh, NASEEM ANSARI, University of Pittsburgh, ANSYS Inc, PEYMAN GIVI, University of Pittsburgh, MICHAEL J. BRAZELL, DIMITRI J. MAVRIPLIS, University of Wyoming — A new computational methodology is developed for large eddy simulation of turbulent flow in complex geometries. This is a hybrid methodology in which a discontinuous Galerkin (DG) base flow solver is combined with a Lagrangian Monte Carlo solver for the filtered density function (FDF). The advantage of the DG is that it provides high order accuracy with fewer degrees of freedom. It also provides flexibility of implementation on unstructured grids. The resulting DG-FDF solver is shown to be very useful for LES of turbulent flows. 2:36PM R2.00008 ABSTRACT WITHDRAWN — 2:49PM R2.00009 Implicit Large-Eddy Simulation of Transition and Turbulence Decay , FERNANDO GRINSTEIN, LANL — In ILES, energy-containing large scales are resolved, and physics capturing numerics are used to spatially filter-out unresolved scales and implicitly model subgrid scale effects. Analysis of transition and decay in the ILES context are the focus of the present work. Euler based ILES is based on using the LANL RAGE code [1] with triple-periodic boundary conditions on evenly spaced grids involving 64, 128, 256, and 512 cells in each direction; Navier-Stokes based isotropic turbulence data generated with the CFDNS code [2] provided initial conditions for ILES. Effects of grid resolution on the ILES unsteady turbulence measures are examined in detail. [1] Grinstein et al., PoF, 23, 034106, 2011. [2] Livescu et al., LANL LA-CC-09-100, 2009. 3:02PM R2.00010 Scale resolving computation of submerged wall jets on flat wall with different roughness heights1 , JOONGCHEOL PAIK, Gangneung-Wonju National University, FABIAN BOMBARDELLI, University of California, Davis — Scale-adaptive simulation is used to investigate the response of velocity and turbulence in submerged wall jets to abrupt changes from smooth to rough beds. The submerged wall jets were experimentally investigated by Dey and Sarkar [JFM, V. 556, p. 337, 2006] at the Reynolds number of 17500 the Froude number of 4.09 and the submergence ratio of 1.12 on different rough beds that were generated by uniform sediments of different median diameters The SAS is carried out by means of a second-order-accurate finite volume method in space and time and the effect of bottom roughness is treated by the approach of Cebeci (2004). The evolution of free surface is captured by employing the two-phase volume of fluid (VOF) technique. The numerical results obtained by the SAS approach, incorporated with the VOF and the rough wall treatment, are in good agreement with the experimental measurements. The computed turbulent boundary layer grows more quickly and the depression of the free surface is more increased on the rough wall than those on smooth wall. The size of the fully developed zone shrinks and the decay rate of maximum streamwise velocity and Reynolds stress components are faster with increase in the wall roughness. 1 Supported by NSF and NRF of Korea 3:15PM R2.00011 Drag and drop simulation: from pictures to full three-dimensional simulations , MICHEL BERGMANN, ANGELO IOLLO, INRIA Bordeaux Sud Ouest, France — We present a suite of methods to achieve “drag and drop” simulation, i.e., to fully automatize the process to perform thee-dimensional flow simulations around a bodies defined by actual images of moving objects. The overall approach requires a skeleton graph generation to get level set function from pictures, optimal transportation to get body velocity on the surface and then flow simulation thanks to a cartesian method based on penalization. We illustrate this paradigm simulating the swimming of a mackerel fish. Tuesday, November 25, 2014 1:05PM - 3:28PM Session R3 Electrokinetics: Surface and Particle Induced Flows — 3004 - Gilad Yossifon, Technion - Israel Institute of Technology 1:05PM R3.00001 Interplay of induced charge electroosmosis, electrothermal flow, and dielectrophoresis at insulating constrictions , NAGA NEEHAR DINGARI, QIANRU WANG, CULLEN BUIE, Massachusetts Inst of Tech-MIT — We present a theoretical and experimental study on the combined influence of induced charge electroosmotic flow (ICEO) and electrothermal flow on particle motion in an insulator based dielectrophoretic (iDEP) device. Strong electric fields used for particle trapping induce charges on the channel wall of low, but finite permittivity [1], and also induce strong temperature gradients [2] because of Joule heating. Consequently, the background fluid flow near the constriction is a superposition of these two effects. Our analysis presents a hitherto unexplored interplay between these two effects and how they influence particles which also experience dielectrophoresis. From our analysis, we find that for channels of low surface permittivity and conductivity, electrothermal effects are stronger near the constriction compared to ICEO effects, while the opposite is true when the surface permittivity or conductivity (or both) are comparable to that of bulk fluid. The analysis also includes the pH and electrolyte concentration dependent contributions of the dynamic Stern layer on ICEO flow. [1] Zhao, C.; Yang, C. AC Field Induced-Charge Electroosmosis over Leaky Dielectric Blocks Embedded in a Microchannel. Electrophoresis 2011, 32, 629–637. [2] Hawkins, B. G.; Kirby, B. J. Electrothermal Flow Effects in Insulating (electrodeless) Dielectrophoresis Systems. Electrophoresis 2010, 31, 3622–3633. 1:18PM R3.00002 3D experimental investigation of the interplay between dielectrophoresis and induced-charge electroosmosis , ALICIA BOYMELGREEN, MATAN ZEHAVI, GILAD YOSSIFON, TECHNION — It is well-known that the advent non-linear electrokinetic flows, such as induced-charge electroosmosis, are strongly dependent on the frequency of the applied field. However, to date, there exists no unifying theory which can exactly predict both the strength and frequency dispersion of such electrokinetic flows. Using microPIV and temperature sensitive dyes we demonstrate the presence of a number of competing non-linear effects including dielectrophoresis, electrothermal flow and wall effects which compete with induced-charge electrokinetic flow, potentially causing a distortion of both the strength and frequency dispersion predicted for pure induced-charge effects. In terms of the wall effects, we investigate the differences between channels in which the walls are conducting (the field is perpendicular to the wall) and insulating (the field is parallel to the wall). This work is of both fundamental and practical importance and may be used to further refine non-linear electrokinetic theory and optimize the flow parameters of electroosmotic pumps and the mobility of electrokinetically driven micromotors or carriers in lab-on-a-chip analysis systems. 1:31PM R3.00003 Induced- and alternating-current electro-osmotic control of the diffusion layer growth in a microchannel-membrane interface device , SINWOOK PARK, GILAD YOSSIFON, Technion - Israel Institute of Technology — The passage of an electric current through an ionic permselective medium under an applied electric field is characterized by the formation of ionic concentration gradients, which result in regions of depleted and enriched ionic concentration at opposite ends of the medium. Induced-current electro-osmosis (ICEO) and alternating-current-electro-osmosis (ACEO) are shown to control the growth of the diffusion layer (DL) which, in turn, controls the diffusion limited ion transport through the microchannel-membrane system. We fabricated and tested devices made of a Nafion membrane connecting two opposite PDMS microchannels. An interdigitated electrode array was embedded within the microchannel with various distances from the microchannel-membrane interface. The induced ICEO (floating electrodes) / ACEO (active electrodes) vortices formed at the electrode array stir the fluid and thereby suppress the growth of the DL. The intensity of the ACEO vortices is controlled by either varying the voltage amplitude or the frequency, each having its own unique effect. Enhancement of the limiting current by on-demand control of the diffusion length is of importance in on-chip electro-dialysis, desalination and preconcentration of analytes. 1:44PM R3.00004 Fluidic Dielectrophoresis of Aqueous Electrical Interfaces , ZACHARY GAGNON, Johns Hopkins University — To date, alternating current (AC) electric fields have been exploited to dielectrophoretically manipulate bubbles, liquid drops, particles, biomolecules and cells. Research and applications in this area, however, has been primarily limited to the interfaces formed between two immiscible metal-liquid, particle-liquid, or gas-liquid surfaces on particles. The influence of AC electric fields across aqueous liquid-liquid interfaces remains relatively unexplored. Fundamentally, many electrokinetic phenomena arise from discontinuities in ionic flux and charge accumulation at electrical interfaces, and here I explore the influence of AC electric fields on the electrical interface created between two aqueous liquids with disparaging electrical properties Using a microfluidic channel with embedded electrodes, two fluid streams - one with a greater electrical conductivity, the other a greater dielectric constant - were made to flow side-by-side. An AC electric field was applied across the flow channel and fluid was observed to displace across the phase interface. The displacement direction is AC frequency dependent, and is attributed to the Maxwell-Wagner interfacial polarization at the liquid-liquid electrical interface. At low AC frequency, below the interfacial charge relaxation time, the high conductive stream is observed to displace into the high dielectric stream. Above this frequency, the direction of liquid injection reverses, and the high dielectric stream injects into the high conductivity stream. An analytical model is presented for this liquid crossover frequency, and applied towards biosensing applications. 1:57PM R3.00005 Bifurcation in the equilibrium height of colloidal particles near an electrode in oscillatory electric fields , TAYLOR WOEHL, BING-JIE CHEN, KELLEY HEATLEY, NICHOLAS TALKEN, CARI DUTCHER, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — Application of an oscillatory electric field is known to alter the equilibrium separation distance between micron-scale colloidal particles and an adjacent electrode. This behavior is believed to be partially due to a lift force caused by electrohydrodynamic (EHD) flow generated around each particle, with previous work focused on identifying a single equilibrium height of the individual particles over the electrode. Here we report the existence of a pronounced bifurcation in the equilibrium particle height in response to low frequency electric fields. Optical and confocal microscopy observations reveal that application of a ∼100 Hz field induces some of the particles to rapidly move several particle diameters up from the electrode, while the others move closer to the electrode. The fraction of particles that exhibit this “extreme levitation” increases with increased applied potential and decreased frequency, in a fashion qualitatively consistent with an energy landscape predicated on competition between EHD flow, colloidal interactions, and gravity. Taken together, the results provide evidence for the existence of a deep tertiary minimum in the electrode-particle interaction potential at a surprisingly large distance from the electrode. 2:10PM R3.00006 The influence of ionic strength on electrohydrodynamic aggregation of colloidal particles , SUKHLEEN SAINI, WILLIAM RISTENPART, Dept. Chemical Engineering and Materials Science, University of California Davis — Colloidal particles suspended in various electrolytes have been widely observed to aggregate near electrodes in response to oscillatory electric fields, a phenomenon believed to result from electrohydrodynamic flows induced around the particles. Most work has focused on elucidating the effects of the applied field strength, frequency, and electrolyte type on the aggregation rate, with less attention paid to the ionic strength. Here we demonstrate that the ionic strength of the electrolyte strongly affects both the aggregate morphology and aggregation dynamics. Optical microscopy observations reveal that an applied field causes micron-scale colloids in aqueous NaCl to rapidly aggregate over a wide range of ionic strengths, but with significant differences in aggregate morphology: at higher ionic strengths (∼1 mM), particles arrange as hexagonal close packed (HCP) crystals, but at lower ionic strengths(∼0.2 mM), the particle aggregates are randomly closed packed (RCP). We interpret these results in terms of the effect of the ionic strength on the height of the particles over the electrode and their corresponding diffusivity, and we discuss preliminary modeling efforts of the effect of ionic strength on the electrohydrodynamic driving force for aggregation. 2:23PM R3.00007 Wall-Induced Non-inertial Lift in Electrophoresis for Continuous Particle Separation , XIANGCHUN XUAN, XINYU LU, Clemson University — We present in this talk a novel continuous-flow electrokinetic method for particle separation based on intrinsic properties which may include size, surface charge, shape and potentially deformability. This method utilizes the wall-induced non-inertial lift force to deflect a sheath flow-focused particle mixture to property-dependent positions in a laminar flow through a straight microchannel. It is demonstrated through both a binary and a ternary separation of polymer particles by size. A numerical model is also developed to understand this separation and to study the parametric effects on it. The numerical predictions are found to agree reasonably with the experimental observations. 2:36PM R3.00008 Effect of the electric field ratio on electroosmotic flow patterns in crossshaped microchannels by the lattice-Boltzmann Method , ALVARO SOCIAS, DIEGO OYARZUN, Universidad de Santiago de Chile, AMADOR GUZMAN, Pontificia Universidad Catolica de Chile — The electroosmotic flow (EOF) pattern characteristics in cross-shaped microchannels flow are important features when either suppressing or enhancing flow features for injection and separation or mixing of multiple species are the wanted objectives. There are situations in EOF in cross-shaped microchannels where the fluid flows toward unexpected and unwanted directions under a given external electric field that depends of both the applied electric field and lengths of the different channels. This article describes the effect of the electric field ratio, defined as the ratio between longitudinal nominal electric field ELong =(VE -VW )/(LW +LE ) and the nominal electric field Ea =(VS -VE )/(VS +VE ), where E, S and W define the east, south and west directions of the cross-shaped microchannel; V is the externally applied voltage and L is the length, on the EOF characteristics in a cross-shaped microchannel. We use the lattice-Boltzmann method (LBM) for solving the discretized Boltzmann Transport Equation (BTE) describing the coupled processes of hydrodynamics and electrodynamic. Our numerical simulations allow us to determine the EOF pattern for a wide range of the electric field ratio and Ea such that inverted flow features are captured and described, which are very important to determine for flow separation or mixing. 2:49PM R3.00009 Electrical Power Generation by Mechanically Modulating Electrical Double Layers1 , HYUK KYU PAK, JONG KYUN MOON, Ulsan National Institute of Science and Technology, Korea — Since Michael Faraday and Joseph Henry made their great discovery of electromagnetic induction, there have been continuous developments in electrical power generation. Most people today get electricity from thermal, hydroelectric, or nuclear power generation systems, which use this electromagnetic induction phenomenon. Here we propose a new method for electrical power generation, without using electromagnetic induction, by mechanically modulating the electrical double layers at the interfacial areas of a water bridge between two conducting plates. We find that when the height of the water bridge is mechanically modulated, the electrical double layer capacitors formed on the two interfacial areas are continuously charged and discharged at different phases from each other, thus generating an AC electric current across the plates. We use a resistor-capacitor circuit model to explain the results of this experiment[1]. This observation could be useful for constructing a micro-fluidic power generation system and for understanding the interfacial charge distribution in solid-liquid interfaces in the near future. [1] J. K. Moon, J. Jeong, D. Lee, and H. K. Pak, Nature Communications, 4,1487 doi: 10.1038/ncomms2485 1 This work was supported by Center for Soft and Living Matter through IBS prgram in Korea. 3:02PM R3.00010 Magnetically Tuned Porous Electrode Formation in Electrochemical Flow Capacitor , HOWARD HU, EDWARD REILLY, University of Pennsylvania — In electrochemical flow capacitors (EFCs), high surface area, conducting, porous particles suspended in an electrolyte solution flow from one storage tank to another through a charging/discharging device between collecting electrodes (collector). In the collector, the particles quickly aggregate to form percolated, electrically conducting networks that facilitate electron flow. To achieve a highly conductive and rapidly assembling network, a high concentration suspension is needed. To facilitate easy pumping, a low concentration suspension is desired. To speed up the network formation process and overcome these conflicting requirements, it is possible to use magnetizable colloids. The particles will acquire a magnetic moment in the presence of an external magnetic field. The magnetic moment will reversibly disappear as soon as the magnetic field is removed. The magnetic field will be applied during the charge and discharge phases to accelerate the formation of electrically connected networks when desired and will be removed when it is time to flow the slurry and refresh the contents in the collector. In this study, we explore the network assembly process, and estimate the network connectivity and electric properties. 3:15PM R3.00011 Shape Oscillation of a Sessile Drop Under the Effect of AmplitudeModulated High Frequency Magnetic Field1 , ZUO-SHENG LEI, Shanghai Key Laboratory of Modern Metallurgy & Material Processing, Shanghai University, Shanghai, China, JIA-HONG GUO, Shanghai Institute of Applied Mathematics and Mechanics, Shanghai University, Shanghai, China, LI-JIE ZHANG, ZHONG-MING REN, Shanghai Key Laboratory of Modern Metallurgy & Material Processing, Shanghai University, Shanghai, China, YVES FAUTRELLE, JECQUELINE ETAY, SIMAP-EPM-Madylam/CNRS, ENSHMG, BP 95, 38402 St. Martin d’Heres Cedex, France — The shape oscillation of a sessile mercury drop under the effect of high frequency amplitude-modulated magnetic field (AMMF) is investigated experimentally. It is a new method to excite shape oscillation of a liquid metal sessile drop, which is different from the case in the presence of a low-frequency magnetic field. The high frequency AMMF is generated by a solenoid inductor fed by a specially designed alternating electric current. The surface contour of the sessile drop is observed by a digital camera. At a given frequency and magnetic flux density of the high frequency AMMF, the edge deformations of the drop with an azimuthal wave numbers (modes n =2, 3, 4, 5, 6) were excited. A stability diagram of the shape oscillation of the drop is obtained by analysis of the experimental data. It is found that the same oscillation mode is excited in different frequency range, and the corresponding frequencies have a ratio of 2. This is a typical character of Mathieu-type parametric instability of a liquid drop. 1 Project supported by the National Natural Science Foundation of China (Grant No. 11372174, 51274137) Tuesday, November 25, 2014 1:05PM - 3:41PM Session R4 Bubbles: Flow Dynamics and Clusters — 3006 - Carlos Martinez-Bazan, University of Jaen 1:05PM R4.00001 Numerical simulation of rising bubble with chemical reaction , KIRTI SAHU, MANOJ TRIPATHI, Indian Institute of Technology Hyderabad, OMAR MATAR, Imperial College London, GEORGE KARAPETSAS, University of Thessaly, Volos 38334, Greece — The dynamics of a rising bubble under the action of gravity and in the presence of an exothermic chemical reaction at the interface is investigated via direct numerical simulation using Volume-of-Fluid (VOF) method. The product of the chemical reaction, and temperature rise due to the exothermic chemical reaction influence the local viscosity and surface tension near the interfacial region, which in turn give rise to many interesting dynamics. The flow is governed by continuity, Navier-Stokes equations along with the convection equation of the volume fraction of the outer fluid and the energy equation. The effects of the Bond, Damkohler, and Reynolds numbers, and of the dimensionless heat of reaction are investigated. The results of this parametric study will be presented at the meeting. 1:18PM R4.00002 The dynamics of rising bubble inside a viscoplastic material , MANOJ TRIPATHI, KIRTI SAHU, Indian Institute of Technology Hyderabad, India, GEORGE KARAPETSAS, University of Thessaly, Volos 38334, Greece, OMAR MATAR, Imperial College London — The axisymmetric dynamics of a bubble rising in a Bingham fluid under the action of buoyancy is investigated. The equations of mass and momentum conservation, coupled to an equation for the volume fraction of the Bingham fluid, are solved used a Volume-of-Fluid (VOF) approach. A regularised constitutive model is used for the description of the viscoplastic behaviour of the material. We found that for large yield stresses, and for weak surface tension the bubble is highly deformable, and the rise is unsteady and is punctuated by periods of rapid acceleration, which separate stages of quasi-steady motion. During the acceleration periods, the bubble aspect ratio exhibits oscillations about unity, whose amplitude and wavelength increase with increasing yield stress and decreasing surface tension. These oscillations are accompanied by the alternating formation and destruction of unyielded zones at the bubble equator, as the bubble appears to “swim” upwards. 1:31PM R4.00003 Linear stability of the wake and path of a rising bubble with a realistic shape1 , JOSÉ CARLOS CANO-LOZANO, Universidad de Jaén, JOEL TCHOUFAG, Universite de Toulouse-IMFT, JACQUES MAGNAUDET, CNRSIMFT, DAVID FABRE, Universite de Toulouse-IMFT, CARLOS MARTÍNEZ-BAZÁN, Universidad de Jaén — A global linear stability analysis of the flow past a bubble rising in still liquid is carried out using the real bubble shape and the terminal velocity obtained for various sets of Galileo (Ga) and Bond (Bo) numbers in axisymmetric simulations performed with the multiphase software Gerris Flow Solver. Once the bubble shape is known, the axisymmetric, steady base flow is computed by means of an iterative Newton method with the finite element software FreeFem++, and the eigenvalue problem is solved with the shift-invert Arnoldi technique implemented in the SLEPc library. The critical curve separating stable and unstable regimes is obtained in the (Ga, Bo) and (Reynolds number, aspect ratio) spaces. This allows us to discuss the effect of the bubble shape and aspect ratio on the wake and path instabilities. We observe that the fore-and-aft asymmetry of the bubble has some influence on the stability since, for a given aspect ratio, bubbles with a realistic shape (i.e. a flatter front and a more rounded rear) are more stable that perfectly spheroidal bubbles. 1 Supported by the Spanish MINECO, Junta de Andalucı́a and EU Funds under projects DPI2011-28356-C03-03 and P11-TEP7495. 1:44PM R4.00004 Bubble formation dynamics in a planar co-flow configuration: Influence of geometric and operating characteristics1 , CÁNDIDO GUTIÉRREZ-MONTES, ROCÍO BOLAÑOS-JIMÉNEZ, CARLOS MARTÍNEZBAZÁN, University of Jaén, ALEJANDRO SEVILLA, University Carlos III of Madrid — An experimental and numerical study has been performed to explore the influence of different geometric features and operating conditions on the dynamics of a water-air-water planar co-flow. Specifically, regarding the nozzle used, the inner-to-outer thickness ratio of the air injector, β =Hi /Ho , the water-to-air thickness ratio, h=Hw /Ho , and the shape of the injector tip, have been described. As for the operating conditions, the water exit velocity profile under constant flow rate and constant air feeding pressure has been assessed. The results show that the jetting-bubbling transition is promoted for increasing values of β, decreasing values of h, rounded injector tip, and for uniform water exit velocity profiles. As for the bubble formation frequency, it increases with increasing values of β, decreasing values of h, rounded injector and parabolic-shaped water exit profiles. Furthermore, the bubble formation frequency has been shown to be lower under constant air feeding pressure conditions than at constant gas flow rate conditions. Finally, the effectiveness of a time-variable air feeding stream has been numerically studied, determining the forcing receptivity space in the amplitude-frequency plane. Experimental results corroborate the effectiveness of this control technique. 1 Work supported by Spanish MINECO, Junta de Andalucı́a, European Funds and UJA under projects DPI2011-28356-C03-02, DPI2011-28356-C03-03, P11-TEP7495 and UJA2013/08/05. 1:57PM R4.00005 Numerical and experimental analyses of the translation of bubbles due to non-spherical interface deformations , ELENA IGUALADA-VILLODRE, Universidad Carlos III de Madrid, DANIEL FUSTER, Institut Jean Le Rond D’Alembert, Universite Pierre et Marie Curie, JAVIER RODRÍGUEZ-RODRÍGUEZ, Universidad Carlos III de Madrid, HUGO DUTILLEUL, Institut Jean Le Rond D’Alembert, Universite Pierre et Marie Curie — Bubbles developing strong interface deformations (e.g. jetting) experience a strong net force that influences significantly their translational motion. In this work, the translation of bubbles as a result of non-spherical interface deformations is studied both numerically and experimentally. The Gerris flow solver is used to solve for a simplified model of the oscillation of a gas bubble in an incompressible liquid. In particular, we solve for the 3D conservation equations in both phases in a system where the total volume changes in the gas are imposed. Assuming a uniform pressure within the bubble, the conservation equations inside the bubble can be rewritten as a function of the temporal evolution of the bubble’s volume. Thus, using volume change rates experimentally measured, we identify different regimes in which the bubble deformation induces a net translation velocity significantly larger than the one obtained with models assuming spherical symmetry. We explore the effect of three parameters: Weber number, dimensionless intensity of the pressure wave and relative distance of the source of the non-spherical perturbation. We support the conclusions extracted from the numerical analyses with experimental measurements of the bubble translational velocity exposed to shock waves. 2:10PM R4.00006 Dynamics of a bubble bouncing at a compound interface , JIE FENG, Princeton University, METIN MURADOGLU, Koc University, HOWARD A. STONE, Princeton University — Bubbly flow is extensively involved in a wide range of technological applications, which generate a great demand for understanding of bubble physics. The collision, bouncing and coalescence of moving bubbles with liquid/gas and liquid/solid interfaces, as the first stage for the formation of foams and flotation aggregates, have been the subject of many studies, but there are still unanswered questions related to how the properties of the interface influence the dynamics. For example, Zawala et al. 2013 have tried to investigate how the kinetic energy of the bubble affects the liquid film drainage during the collision with an air-water interface. Inspired by Feng et al. 2014, we study the dynamics of an air bubble bouncing at a liquid/liquid/gas interface, in which a thin layer of oil is put on top of the water. The presence of the oil layer changes the interfacial properties and thus the entire process. Combined with direct numerical simulations, extensive experiments were carried out to investigate the effects of the oil layer thickness, oil viscosity, bubble size and initial impact velocity on the bouncing of the bubble at the compound interface. In addition, a mass-spring model is proposed to describe the bubble dynamics and interactions with the oil layer. 2:23PM R4.00007 Comparison of Detailed Bubble-Cluster Simulations with Reduced Models , ARPIT TIWARI1 , CARLOS PANTANO, JONATHAN B. FREUND, Univ of Illinois - Urbana — Reduced-physics models of bubble ensembles depend on lengthscale separation between the characteristic size of the cluster and the comprising bubbles. They have been remarkably successful in reproducing qualitatively the gross-scale development of the clusters. Studies based on such models, consistent with the experimental findings, suggest that the cluster collapse propagates inward, with pressure focusing toward the geometrical center (with particularly violent collapse of bubbles at its core). The bubble-scale dynamics near the focus are therefore anticipated to be particularly important in the damage of adjacent surfaces. Quantifying these dynamics is the goal of our three-dimensional simulations, which explicitly represent the non-spherical dynamics of each bubble within the cluster. We simulate collapse of a hemispherical cluster of 50 bubbles adjacent to a plane rigid wall for different initial configurations. Results show that the qualitative behavior matches predictions from the homogenized and particle-based reduced models. However, the peak pressures show strong dependence on bubble-scale dynamics. In the detailed simulations, they are typically only a small fraction of those predicted by the reduced models. A systematic comparison with these models will be presented. 1 Currently working at Gamma Technologies Inc. 2:36PM R4.00008 Bubble size measurements in a bubbly wake1 , ASHISH KARN, JIARONG HONG, CHRISTOPHER ELLIS, ROGER ARNDT, University of Minnesota — Measurements of bubble size distribution are ubiquitous in many industrial applications. Conventional methods using image analysis to measure bubble size are limited in their robustness and applicability in highly turbulent bubbly flows. These flows usually impose significant challenges for image processing such as a wide range of bubble size distribution, spatial and temporal inhomogeneity of image background including in-focus and out-of-focus bubbles, as well as the excessive presence of bubble clusters. This talk introduces a multi-level image analysis approach to detect a wide size range of bubbles and resolve bubble clusters from images obtained in a turbulent bubbly wake of a ventilated hydrofoil. The proposed approach was implemented to derive bubble size and air ventilation rate from the synthetic images and the experiments, respectively. The results show a great promise in its applicability for online monitoring of bubbly flows in a number of industrial applications. 1 Sponsored by Office of Naval Research and the Department of Energy 2:49PM R4.00009 Experimental study of interactions between bubbles and bubble wakes, via PIV/LIF , DAISUKE SHINOHARA, Graduate School of Engineering, Shizuoka University, TAKAYUKI SAITO, Research Institute of Green Science and Technology, Shizuoka University — A study of the interactions between bubbles and bubble wakes is essential to improve the efficiency of an industrial application. The knowledge of the bubbles and bubble wakes interactions in a bubble swarm, however, is still few. The purpose of this study is to experimentally investigate these interactions in a bubble column. For this specific purpose, a bubble-swarm generator that controls the formation and launch of the bubbles precisely was employed. Equivalent diameters of the bubbles was about 2.6 mm. The two bubbles were launched side by side and the bubble-bubble distance was 7 mm. The center bubble was launched between two leading bubbles 9.75 ms behind the leading bubbles. Using a high speed video camera and PIV, we visualized motion of the bubbles and their surrounding liquid motion. Those bubbles linearly ascended during 0.07 sec after launched. An aspect ratio of the center bubble in the time span from 0.03 to 0.07 sec after launched was smaller than those of the leading bubbles. The wakes of the leading bubbles are considered to enhance dynamic pressure acting on the center bubble. Hence, the bubble wake contribution is important to understand a bubbly flow. 3:02PM R4.00010 Multiple steady bubbles in a Hele-Shaw channel , CHRISTOPHER GREEN, University of California San Diego, GIOVANI VASCONCELOS, Federal University of Pernambuco — We construct analytical solutions, in the form of conformal mappings, solving the free boundary problem for the shapes of any finite number of steadily translating bubbles in a Hele-Shaw channel. These solutions can be decomposed into the sum of two analytic functions - corresponding to the complex potentials in the laboratory and co-travelling frames - which conformally map a bounded multiply connected circular domain onto respective degenerate polygonal domains (infinite strips with interior slits of finite-length which are either horizontally or vertically aligned). These functions are obtained using the generalised Schwarz-Christoffel formula for multiply connected domains in terms of the Schottky-Klein prime function. The solutions we have found are very general and make no assumptions on the geometrical arrangement of the bubbles. 3:15PM R4.00011 A Dynamic Testbed for Supercavitating Vehicles1 , DAVID SANABRIA, GARY BALAS, ROGER ARNDT, University of Minnesota — Underwater vehicles that travel inside a gas cavity offer possibilities for high-speed transportation as a result of reduced contact area with the fluid and drag reduction. Validation and testing of mathematical models and control systems for these vehicles is a challenge due to the cost and complexity of experimental facilities and procedures. In particular, planing forces generated when the vehicle back end pierces the supercavity, can lead to instability and are challenging for validation and testing. A cost efficient approach to the experimental validation of control systems for a supercavitating vehicle is presented in this talk. The test method uses a small scale supercavitating vehicle, free to rotate in a high-speed water tunnel, to evaluate control systems designed for stabilization and tracking of attack angle commands. The vehicle is equipped with a disk cavitator and two lateral fins used for control. The key feature of the validation approach is that planing forces and their effects are captured in the high-speed water tunnel. The proposed validation method is uniquely suitable to validate the robustness of control strategies in the presence of realistic flow conditions and planing. 1 Sponsored by Office of Naval Research 3:28PM R4.00012 The significance of electrically induced shear stress in drainage of thin aqueous films , VLADIMIR AJAEV, Southern Methodist University, CHRISTIAAN KETELAAR, University of Delaware — We develop a model of drainage of a microscale thin aqueous film separating a gas bubble and a solid wall. In contrast to previous studies, the electrostatic effects are accounted for not only in the normal but also in the shear stress balance at the liquid-gas interface. We show that the action of the tangential component of the electric field leads to potentially strong spatially variable shear stress at the deforming charged interface. This previously overlooked effect turns out to be essential for correctly estimating the long-time drainage rates. Time-dependent fluid interface shapes predicted by our model are in very good agreement with the experimental data. Tuesday, November 25, 2014 1:05PM - 3:41PM Session R5 Bubbles: General — 3008 - Sigurdur Thoroddsen, King Abdullah University of Science and Technology 1:05PM R5.00001 Particle induced air bubble break-up in a Hele-Shaw cell1 , PENG ZHANG, JOHN MINES, Department of Engineering Science and Mechanics, Virginia Tech, Blacksburg, VA 24061, SUNGYON LEE, Department of Mechanical Engineering, Texas A&M University, College Station, Texas 77843, SUNGHWAN JUNG, Department of Engineering Science and Mechanics, Virginia Tech, Blacksburg, VA 24061 — Hydrodynamic interactions of drops and bubbles with particles in viscous fluids are important in the multiphase separation and reaction processes. In the present work, we explore the fundamental mechanism of such complex processes by studying the collision of a single bubble with a fixed solid particle inside a Hele-Shaw cell. Physical experiments show that an air bubble either splits or slides around the particle depending on the initial transverse offset between the bubble and particle centroids. A bubble slides around the particle until the offset is decreased below a critical value, in which case the bubble splits into two daughter bubbles. We are able to predict this slide-split transition using a theoretical model that compares the relative change in surface energy, gravitational potential and viscous dissipation in the two regimes. 1 This research is partially supported by National Science Foundation (PHY-1205642, CBET- 1336038). 1:18PM R5.00002 Determination of the diffusion coefficient in a bubble suspension for large but finite Reynolds numbers , RODOLFO ALCALA, ALICIA AGUILAR, Universidad Michoacana de San Nicolas de Hidalgo, ROBERTO ZENIT, BERNARDO FIGUEROA, Universidad Nacional Autonoma de Mexico, JESUS CORREA, Universidad Michoacana de San Nicolas de Hidalgo — This paper presents experimental results concerning the study of a bubble suspension confined in a thin channel for different volume fractions of gas, in a regime characterized by Reynolds numbers of order 102 - 103 . The experimental set-up consists in a Hele-Shaw cell thin channel, a high speed camera, a diffused backlight system and a rotameter to control the flow of Nitrogen into the system. A trajectography technique is used to obtain the flow properties from series of digital images captured by the high speed camera. The trajectories and velocities of the bubbles were calculated from the instantaneous bubbles positions. With these measurements, the diffusion coefficients for different gas fractions were determined using the method of the autocorrelation function. 1:31PM R5.00003 Dissolution of spherical cap CO2 bubbles attached to flat surfaces in airsaturated water1 , PABLO PEÑAS, Universidad Carlos III de Madrid (UC3M), MIGUEL A. PARRALES, JAVIER RODRIGUEZ-RODRIGUEZ, UC3M — Bubbles attached to flat surfaces immersed in quiescent liquid environments often display a spherical cap (SC) shape. Their dissolution is a phenomenon commonly observed experimentally. Modelling these bubbles as fully spherical may lead to an inaccurate estimate of the bubble dissolution rate. We develop a theoretical model for the diffusion-driven dissolution or growth of such multi-component SC gas bubbles under constant pressure and temperature conditions. Provided the contact angle of the bubble with the surface is large, the concentration gradients in the liquid may be approximated as spherically symmetric. The area available for mass transfer depends on the instantaneous bubble contact angle, whose dynamics is computed from the adhesion hysteresis model [Hong et al., Langmuir, vol. 27, 6890-6896 (2011)]. Numerical simulations and experimental measurements on the dissolution of SC CO2 bubbles immersed in air-saturated water support the validity of our model. We verify that contact line pinning slows down the dissolution rate, and the fact that any bubble immersed in a saturated gas-liquid solution eventually attains a final equilibrium size. 1 Funded by the Spanish Ministry of Economy and Competitiveness through grant DPI2011-28356-C03-0. 1:44PM R5.00004 Gas dissolution in antibubble dynamics1 , BENOIT SCHEID, Université Libre de Bruxelles, JAN ZAWALA, Polish Academy of Sciences, STÉPHANE DORBOLO, Université de Liège — Antibubbles are ephemeral objects. Their lifetime is driven by the slow drainage of the air shell from the bottom to the top of the antibubble under the action of the hydrostatic pressure. We show in this work that this argument is only valid if the water used to make the surfactant mixture is saturated in air. Otherwise, two paths are used by the air that conducts to the thinning and the eventual collapse of the air shell: the drainage from the bottom to the top of the antibubble and the dissolution of the air by the liquid. Using degassed water dramatically shortens the lifetime of the antibubbles, as observed experimentally and rationalised by time-dependent simulations. Consequently, the antibubble lifetime is not only correlated to physical and chemical properties of the air-liquid interface but also to the gas content of the liquid. We also show that pure gas dissolution does not depend on the antibubble radius, a behaviour that allows to rationalise unexplained experimental data found in the literature. 1 We thank the F.R.S.-FNRS for financial support 1:57PM R5.00005 History effects on the diffusion-driven growth and dissolution of a gas bubble , JAVIER RODRIGUEZ-RODRIGUEZ, Carlos III Univ. Madrid (UC3M), MIGUEL A. PARRALES, PABLO PEÑAS, UC3M, OSCAR ENRIQUEZ, Univ. of Twente, ELENA IGUALADA-VILLODRE, UC3M, DEVARAJ VAN DER MEER, Univ. of Twente — A bubble of a gas that is soluble in the surrounding liquid may grow or dissolve depending on whether the saturation concentration at the bubble’s pressure is lower or higher than the gas concentration in the bulk liquid. In the limit of small Peclet, the (slow) diffusion-driven bubble growth or dissolution rates are commonly calculated using the Epstein-Plesset theory, despite the fact that it is only valid when the gas concentration field in the liquid is initially uniform. Here we show how to modify this theory to account for non-uniformities in the initial concentration field resulting from the past history of variations of the ambient pressure. In particular, we obtain a history term that closely resembles the Basset memory integral found in the unsteady translation of a sphere through a viscous fluid. The new formulation is applied to the particular example of a bubble, initially in diffusive equilibrium with the ambient, that is subjected to a depression and a later compression. The results are compared to numerical simulations as well as experiments. Funded by the Spanish Ministry of Economy and Competitiveness through grant DPI2011-28356-C03-02. 2:10PM R5.00006 Thermocapillary motion of bubble under the action of gravity in a selfrewetting fluid , OMAR MATAR, Imperial College London, MANOJ TRIPATHI, KIRTI SAHU, Indian Institute of Technology Hyderabad, India, GEORGE KARAPETSAS, University of Thessaly, Volos 38334, Greece, KHELLIL SEFIANE, University of Edinburgh, Edinburgh — The motion of a bubble driven under the action of buoyancy and thermocapillarity inside a tube with a non-uniformly-heated walls, containing a so-called “self-rewetting fluid” is investigated. The surface tension of the “self-rewetting fluid” exhibits a parabolic dependence on temperature with a well-defined minimum. We perform direct numerical simulation of axisymmetric bubble motion in a fluid whose temperature increases linearly with vertical distance from the bottom of the tube for a range of Bond and Gallileo numbers, and for various parameters that govern the functional dependence of surface tension on temperature. We demonstrate that bubble motion can be reversed and then arrested in self-rewetting fluids for sufficiently small Bond numbers; this in contrast with the linear fluid (surface tension linearly decreases with increasing temperature). We also demonstrate that considerable bubble elongation is possible under significant wall confinement, and for strongly self-rewetting fluids and large Bond numbers. In the Stokes flow limit, we derive the conditions under which a spherical bubble can come to rest in a self-rewetting fluid whose temperature varies linearly in the vertical direction, and demonstrate that this is possible for both positive and negative temperature. 2:23PM R5.00007 The effect of gravity-induced pressure gradient on bubble luminescence , OUTI SUPPONEN, Ecole Polytechnique Fédérale de Lausanne, DANAIL OBRESCHKOW, University of Western Australia, PHILIPPE KOBEL, NICOLAS DORSAZ, Ecole Polytechnique Fédérale de Lausanne, MARC TINGUELY, Imperial College London, MOHAMED FARHAT, Ecole Polytechnique Fédérale de Lausanne — The violent collapse of a bubble can heat up its gaseous contents to temperatures exceeding those on the sun’s surface, resulting in a short luminescence flash. Occurring at the very moment of the collapse, luminescence must be highly sensitive to the bubble geometry at the preceding final stage. This represents an important feature as any pressure anisotropy in the surrounding liquid will result in a deformation of an initially spherical bubble, inducing a micro-jet that pierces the bubble and makes it experience a toroidal collapse. We therefore present these as complementary phenomena by investigating the link between jets and luminescence of laser-generated single bubbles. Through ultra-high-speed imaging, the micro-jet formation and evolution of a single bubble are observed with unprecedented detail, whilst the bubble light emission is analyzed by means of a spectrometer. The bubble energy and the micro-jet size are controlled by adjusting the laser-pulse and by varying the gravity level aboard ESA parabolic flights, respectively. We here provide systematic evidence on how bubble-jets suppress luminescence in a considerable manner, even in normal gravity where the jet is barely observable. We conclude that gravity must be accounted for in accurate models of luminescence. 2:36PM R5.00008 Enhanced Condensation of Vapor Bubbles by Acoustic Actuation , THOMAS BOZIUK, MARC SMITH, ARI GLEZER, Georgia Institute of Technology — The effects of acoustic actuation on enhancement of the condensation rate of vapor bubbles in a liquid pool are investigated experimentally. Vapor bubbles are formed by direct injection into quiescent liquid in a sealed tank under controlled ambient pressure that varies from atmospheric to partial vacuum. The bubbles are injected vertically from a pressurized steam reservoir through nozzles of varying characteristic diameters, and the actuation is applied during different stages of the bubbles formation and advection. It is shown that kHz range acoustic actuation leads to excitation of high-amplitude surface capillary (Faraday) waves at the vapor-liquid interface that significantly increases the condensation rate. The concomitant controlled changes in bubble volume and in the structure of the vapor interface strongly affect bubble advection in the liquid pool. The increase in condensation rate is affected by the surface waves that increase the mixing in the thermal boundary layer surrounding the bubble, and on the advection of the bubble within the pool. High-speed image processing is used to quantitatively measure the scale of the capillary waves and their effect on vapor bubble dynamics at several ambient pressures that affect the global condensation rate. 2:49PM R5.00009 Investigation of a relationship between Spherical-shape particle flocculation and acoustic-cavitation-oriented bubbles (ACOBs) under kHz-band ultrasonic irradiation , SAYURI YANAI, Faculty of Engineering, Shizuoka University, YUKI MIZUSHIMA, Graduate school of Science and Technology, Shizuoka University, TAKAYUKI SAITO, Research Institute of Green Science and Technology, Shizuoka University — We investigated unprecedented spherical-shape particle flocculation with an effect of acoustic-cavitation-oriented bubbles (ACOBs) that were generated in water under kHz-band ultrasonic irradiation. In past studies, particle concentrations forming stripes under MHz-band ultrasonic irradiation have been investigated and reported by many previous researchers. However, the spherical-shape particle flocculation is very anomalous. Further, it is able to flocculate mm-order particles that are principally impossible to manipulate using of MHz-band ultrasonic. We focused on the mechanism of the spherical-shape particle flocculation under 20-kHz-ultrasonic irradiation in water. From our experimental results of spherical-shape particle flocculation, we found out that the flocculation factors were not only acoustic radiation force but also behavior of the ACOBs. The ACOBs adhering to the particle surface moved to a certain position with the particle, depending on acoustic pressure distribution in the water. In the present study, we report and discuss the results of the visualized relationship among the ACOB motion, the particle motion and acoustic pressure distribution in water. 3:02PM R5.00010 Theoretical and Experimental Investigation of Particle Trapping via Acoustic Bubbles , YUN CHEN, Texas A&M University, ZECONG FANG, BRETT MERRITT, DARIUS SAADAT-MOGHADDAM, Washington State University Vancouver, DILLON STRACK, Texas A&M University, JIE XU, University of Illinois at Chicago, SUNGYON LEE, Texas A&M University — One important application of lab-on-a-chip devices is the trapping and sorting of micro-objects, with acoustic bubbles emerging as an effective, non-contact method. Acoustically actuated bubbles are known to exert a secondary radiation force on micro-particles and trap them, when this radiation force exceeds the drag force that acts to keep the particles in motion. In this study, we theoretically evaluate the magnitudes of these two forces for varying actuation frequencies and voltages. In particular, the secondary radiation force is calculated directly from bubble oscillation shapes that have been experimentally measured for varying acoustic parameters. Finally, based on the force estimates, we predict the threshold voltage and frequency for trapping and compare them to the experimental results. 3:15PM R5.00011 Microscopic engine driven by laser-induced cavitation bubbles , PEDRO QUINTOSU, ICN-UNAM — In this work an analogue to a microscopic intermittent internal combustion engine is realized with a single microparticle periodically driven by cavitation bubbles at rates of up to 500 Hz. 3:28PM R5.00012 Satellite formation during bubble transition through an interface between immiscible liquids , ERQIANG LI, King Abdullah University of Science and Technology, SHABBAB AL-OTAIBI, Saudi Arabian Oil Company, IVAN VAKARELSKI, SIGURDUR THORODDSEN, King Abdullah University of Science and Technology — A bubble can pass through the interface between two immiscible liquids if it is energetically favourable. Once the intermediate film has drained sufficiently, the bubble makes contact with the interface, forming a triple-line and producing strong capillary waves which travel around the bubble and can pinch off a satellite on the opposite side, akin to the coalescence cascade dynamics. We identify the critical Ohnesorge number where such satellites are produced and characterize their sizes. The total transition time scales with the bubble size and differential surface tension, while the satellite pinch-off time scales with the capillary-inertial time of the pool liquid which originally surrounds the bubble. We also use high-speed video imaging to study the contact neck motion. For low viscosity it grows in time with a power-law exponent between 0.44 and 0.50, with a prefactor modified by the net sum of the three interfacial tensions. Increasing the receiving drop viscosity drastically slows down the triple-line motion, when the Ohnesorge number exceeds around 0.08. This differs qualitatively from the coalescence of two miscible drops of different viscosities, where the lower viscosity sets the coalescence speed. We thereby propose a strong resistance from the triple-line. Tuesday, November 25, 2014 1:05PM - 3:41PM Session R6 Biofluids: Predicting Effective Locomotion — 3010 - Geoffrey Spedding, University of Southern California 1:05PM R6.00001 Schooling of flapping wings: Experiments , LEIF RISTROPH, New York University, Courant Institute, ALEXANDER BECKER, New York University, Department of Mathematics, HASSAN MASOUD, New York University, Courant Institute, JOEL NEWBOLT, New York University, Department of Physics, MICHAEL SHELLEY, New York University, Courant Institute — The role of fluid dynamics in mediating schooling and flocking remains unclear because of the complex interactions between locomotors and their flow fields. We study such interactions for flapping wings that swim within the wakes of others in an array and discover “schooling modes” characterized by preferred spatial phase relationships. These modes are associated with surprising effects including the doubling of swimming speed for small changes in flapping kinematics and the coexistence of two possible speeds for identical kinematics. Flow visualization shows how these dynamics arise from repeated constructive or destructive interactions of a wing with the wave-like flow into which it swims. By establishing how coherent collective behavior emerges naturally for flapping locomotion, these results provide a physical basis to interpret the structure and dynamics of animal groups. 1:18PM R6.00002 Schooling of flapping wings: Simulations , HASSAN MASOUD, Courant Institute, NYU and Princeton University, ALEXANDER BECKER, LEIF RISTROPH, MICHAEL SHELLEY, Courant Institute, NYU — We examine the locomotion of an infinite array of wings that heave vertically with a prescribed sinusoidal motion and are free to translate in the horizontal direction. To do this, we simulate the motion of a freely translating flapping airfoil in a domain with periodic horizontal boundary conditions. These simulations indicate that the wings can “take advantage” of their collectively generated wake flows. In agreement with our experiments in a rotational geometry, we find ranges of flapping frequency over which there are multiple stable states of locomotion, with one of these swimming states having both higher speeds and efficiencies than an isolated flapping and locomoting wing. A simple mathematical model, which emphasizes the importance of history dependence in vortical flows, explains this multi-stability. These results may be important to understanding the role of hydrodynamic interactions in fish schooling and bird flocking. 1:31PM R6.00003 Performance characteristics of pitching flexible foil propulsors , CODY BROWNELL, BRENDAN EGAN, MARK MURRAY, United States Naval Academy — Performance characteristics of flexible foil propulsors are studied experimentally. The project investigates the dependence of thrust and efficiency on foil elasticity, Strouhal number, and flow velocity. The experiments took place in a large recirculating water channel, using full span flexible propulsor models to approximate a 2D geometry. The propulsor pitched about a fixed axis at its quarter chord, with a six-axis load cell measuring the forces and torques on the shaft. Propulsive efficiency is found to peak at an optimum Strouhal number for each foil tested. Varying elasticity did not produce a similar local maximum over the sampled parameter space. The ensemble data will facilitate the engineering of fish-like propulsion systems for future application of this technology. 1:44PM R6.00004 Survival of the fastest: Evolving wings for flapping flight , SOPHIE RAMANANARIVO, New York University, Courant Institute, THOMAS MITCHEL, New York University, Department of Mathematics, LEIF RISTROPH, New York University, Courant Institute — To optimize flapping flight with regard to wing shape, we use an evolutionary or genetic algorithm to improve the forward speed of 3d-printed wings or hydrofoils that heave up-and-down and self-propel within water. In this scheme, “genes” are mathematical parameters specifying wing shape, and “breeding” involves the merging and mutation of genes from two parent wings to form a child. A wing’s swimming speed is its “fitness”, which dictates the likelihood of breeding and thus passing on its genes to the next generation. We find that this iterative process leads to marked improvements in relatively few generations, and several distinct shape features are shared among the fastest wings. We also investigate the favorable flow structures produced by these elite swimmers and compare their shape and performance to biologically evolved wings, fins, tails, and flippers. 1:57PM R6.00005 PIV-based pressure, force, and torque measurements of a robotic model swimmer , JOHN DABIRI, California Institute of Technology, KELSEY LUCAS, PATRICK THORNYCROFT, GEORGE LAUDER, Harvard University — We apply a recently developed technique for non-invasive pressure measurement to study the dynamics of anguilliform swimming by a robotic flapping foil. The method is based on spatial integration of time-resolved particle image velocimetry measurements. The pressure gradient computed from the Navier-Stokes equations is integrated along multiple paths in the domain, and the local pressure is determined by the median value of the integration results. In addition, the pressure field is integrated on the surface of the foil to compute the instantaneous forces and torque exerted by the foil on the fluid. Direct force and torque measurements from a load cell are used to confirm the accuracy of the PIV-based measurements. Results for flapping foils of varying flexibility are compared to infer the role of the pressure field in the dynamics and energetic efficiency of locomotion. 2:10PM R6.00006 Underlying principle of efficient propulsion in flexible plunging foil1 , XING ZHANG, XIAOJUE ZHU2 , GUOWEI HE, Institute of Mechanics, Chinese Academy of Sciences — Recently, it has been reported that passive flexibility in flapping foils can result in the enhancement of propulsive performance. In this study, we investigate the relations among propulsive efficiency, structural resonance and hydrodynamic wake resonance in a flexible plunging foil. We conduct fluid-structure-interaction simulations on flows over flexible plunging foils by using the immersed boundary method. The wake resonant frequency is computed by performing a linear stability analysis on the averaged velocity profile. The results indicates that: (i) optimal efficiency is not necessarily achieved at the structural resonance point; and (ii) optimal efficiency always occurs when the driving frequency matches the wake resonant frequency. By dissecting the wake structures, we found that whether the optimal efficiency is achieved at the structural resonance point depends on the strength of the leading edge vortex (LEV) relative to that of the trailing edge vortex (TEV). In addition, the validity of the aforesaid principle under the condition of free-swimming (self-propulsion) is also discussed. 1 NSFC (11021262, 11023001, 11232011 and 11372331) address: Physics of Fluids Group, University of Twente, 7500 AE Enschede, The Netherlands 2 Present 2:23PM R6.00007 Swimming Efficiently: An Analytical Study of Optimal Swimming in Fish , A. JOSH WIENS, ANETTE HOSOI, Massachusetts Institute of Technology — The Strouhal Number (St), is widely considered to be the defining parameter for efficient undulatory swimming. Biological studies have shown that fish species across a broad range of shapes and sizes adhere to a narrow St range (0.2 < St < 0.4). Despite its significance, St alone provides an incomplete description of the kinematics and geometry of a swimming fish. The dimensionless speed and amplitude of the body wave, along with the size and shape of the body can also play a significant role in swimming performance. We apply Lighthill’s elongated body theory to construct a simple but powerful reduction of the steady-swimming problem. Through this reduction, the energetic efficiency of a swimming fish can be directly expressed as an analytical function of body geometry and kinematics. In this reduced form, the interplay between the parameters of the system, and their collective role in determining the performance of the swimmer can be readily observed and understood. In particular, the reduced model is applied to understand how wave amplitude, wave speed, and St must relate for optimal swimming efficiency. Following this, we then explore how these relationships are altered by geometric factors such as tail size and compliance. 2:36PM R6.00008 The hydrodynamic principle for the caudal fin shape of small aquatic animals , JEONGSU LEE, Department of Mechanical and Aerospace Engineering, Seoul Natl Univ, YONG-JAI PARK, Department of Mechanical Engineering, Sunmoon Univ, KYU-JIN CHO, HO-YOUNG KIM, Department of Mechanical and Aerospace Engineering, Seoul Natl Univ — The shape of caudal fins of small aquatic animals is completely different from that of large cruising animals like dolphin and tuna which have high aspect-ratio lunate tail. To unveil the physical principle behind natural selection of caudal fins of small aquatic animals, here we investigate the hydrodynamics of an angularly reciprocating plate as a model for the caudal fin oscillation. We find that the thrust production of a reciprocating plate at high Strouhal numbers is dominated by generation of two distinct vortical structures associated with the acceleration and deceleration of the plate regardless of their shape. Based on our observations, we construct a scaling law to predict the thrust of the flapping plate, which agrees well with the experimental data. We then seek the optimal aspect ratio to maximize thrust and efficiency of a flapping plate for fixed flapping frequency and amplitude. Thrust is maximized for the aspect ratio of approximately 0.7. We also theoretically explain the power law behaviors of the thrust and efficiency as a function of the aspect ratio. 2:49PM R6.00009 Maximizing the efficiency of a flexible propulsor using experimental optimization1 , DANIEL QUINN, Princeton University, GEORGE LAUDER, Harvard University, ALEXANDER SMITS, Princeton University, Monash University — Experimental gradient-based optimization is used to maximize the propulsive efficiency of a heaving and pitching flexible panel. Optimum and near-optimum conditions are studied via direct force measurements and Particle Image Velocimetry (PIV). The net thrust and power are found to scale predictably with the frequency and amplitude of the leading edge, but the efficiency shows a complex multimodal response. Optimum pitch and heave motions are found to produce nearly twice the efficiencies of optimum heave-only motions. Efficiency is globally optimized when (1) the Strouhal number is within an optimal range that varies weakly with amplitude and boundary conditions; (2) the panel is actuated at a resonant frequency of the fluid-propulsor system; (3) heave amplitude is tuned such that trailing edge amplitude is maximized while flow along the body remains attached; and (4) the maximum pitch angle and phase lag are chosen so that the effective angle of attack is minimized. 1 This work was supported by the Office of Naval Research under MURI grant number N00014-08-1-0642 (Program Director Dr. Bob Brizzolara), and the National Science Foundation under Grant DBI 1062052 (PI Lisa Fauci) and Grant EFRI-0938043 (PI George Lauder). 3:02PM R6.00010 New drag laws for flapping flight , NATALIE AGRE, New York University, Department of Mathematics and Department of Physics, JUN ZHANG, New York University, Courant Institute and Department of Physics, LEIF RISTROPH, New York University, Courant Institute — Classical aerodynamic theory predicts that a steadily-moving wing experiences fluid forces proportional to the square of its speed. For bird and insect flight, however, there is currently no model for how drag is affected by flapping motions of the wings. By considering simple wings driven to oscillate while progressing through the air, we discover that flapping significantly changes the magnitude of drag and fundamentally alters its scaling with speed. These measurements motivate a new aerodynamic force law that could help to understand the free-flight dynamics, control, and stability of insects and flapping-wing robots. 3:15PM R6.00011 Scaling macroscopic aquatic locomotion , MATTIA GAZZOLA, School of Engineering and Applied Sciences, Harvard University, USA, MEDERIC ARGENTINA, Universite Nice Sophia-Antipolis, Institut non lineaire de Nice, France, LAKSHMINARAYANAN MAHADEVAN, School of Engineering and Applied Sciences, Harvard University, USA — Inertial aquatic swimmers that use undulatory gaits range in length L from a few millimeters to 30 meters, across a wide array of biological taxa. Using elementary hydrodynamic arguments, we uncover a unifying mechanistic principle characterizing their locomotion by deriving a scaling relation that links swimming speed U to body kinematics (tail beat amplitude A and frequency ω) and fluid properties (kinematic viscosity ν). This principle can be simply couched as the power law Re ∼ Swα , where Re = U L/ν ≫ 1 and Sw = ωAL/ν, with α = 4/3 for laminar flows, and α = 1 for turbulent flows. Existing data from over 1000 measurements on fish, amphibians, larvae, reptiles, mammals and birds, as well as direct numerical simulations are consistent with our scaling. We interpret our results as the consequence of the convergence of aquatic gaits to the performance limits imposed by hydrodynamics. 3:28PM R6.00012 Modulation of a Flow Field by Dragonfly Nymph Valve Kinematics1 , CHRIS ROH, MORTEZA GHARIB, California Institute of Technology — Previously, we visualized a respiratory jet and a propulsive jet of a dragonfly nymph using laser induced fluorescence. A more quantitative measurement of the dragonfly nymph’s underwater breathing was investigated using digital particle image velocimetry. Simultaneously, dragonfly’s anal valve kinematics were recorded using high-speed videography. The result shows an active usage of the valve during exhalation and inhalation to modulate the flow field. Calculating a Lagrangian particle path by time integration of the velocity field showed that the exhaled fluid is not inhaled back. This result suggests that the anal valve modulation of the flow field prevents the rebreathing of the exhaled jet. 1 Supported by NSF-GRFP Tuesday, November 25, 2014 1:05PM - 3:28PM Session R7 Biofluids: Mechanics of Swallowing and Speech University — 3012 - Kartik Bulusu, The George Washington 1:05PM R7.00001 Mosquitoes drink with a burst in reserve: explaining pumping behavior with a fluid mechanics model1 , SOUVICK CHATTERJEE, JAKE SOCHA, MARK STREMLER, Virginia Tech — Mosquitoes drink using a pair of in-line pumps in the head that draw liquid food through the proboscis. Experimental observations with synchrotron x-ray imaging indicate two modes of drinking: a predominantly occurring continuous mode, in which the cibarial and pharyngeal pumps expand cyclically at a constant phase difference, and an occasional, isolated burst mode, in which the pharyngeal pump expansion is 10 to 30 times larger than in the continuous mode. We have used a reduced order model of the fluid mechanics to hypothesize an explanation of this variation in drinking behavior. Our model results show that the continuous mode is more energetically efficient, whereas the burst mode creates a large pressure drop across the proboscis, which could potentially be used to clear blockages. Comparisons with pump knock-out configurations demonstrate different functional roles of the pumps in mosquito feeding. 1 This material is based upon work supported by the NSF under Grant No. #0938047. 1:18PM R7.00002 How dogs drink water1 , SEAN GART, JAKE SOCHA, Virginia Tech, PAVLOS VLACHOS, Purdue, SUNGHWAN JUNG, Virginia Tech — Animals with incomplete cheeks (i.e. dogs and cats) need to move fluid against gravity into the body by means other than suction. They do this by lapping fluid with their tongue. When a dog drinks, it curls its tongue posteriorly while plunging it into the fluid and then quickly withdraws its tongue back into the mouth. During this fast retraction fluid sticks to the ventral part of the curled tongue and is drawn into the mouth due to inertia. We show several variations of this drinking behavior among many dog breeds, specifically, the relationship between tongue dynamics and geometry, lapping frequency, and dog weight. We also compare the results with the physical experiment of a rounded rod impact onto a fluid surface. 1 Supported by NSF PoLS #1205642. 1:31PM R7.00003 Influence of muscle activation and mucosal material property on esophageal transport: study based on a fully-resolved computational model1 , WENJUN KOU, Theoretical and Applied Mechanics Program, Northwestern University, JOHN PANDOLFINO, PETER KAHRILAS, Feinberg School of Medicine, Northwestern University, NEELESH PATANKAR, Department of Mechanical Engineering, Northwestern University — Esophageal transport involves interactions between food (bolus), the esophageal walls (composed of mucosal, circular muscle (CM) and longitudinal muscle (LM) layers), and neurally coordinated muscle activation including CM contraction and LM shortening. Due to the complexity of these interactions, few studies have been conducted on the mechanical role of the mucosal layer in esophageal transport. Also poorly understood are the collaborative roles of CM contraction and LM shortening and the influence of their synchronization. Here, based on a fully-resolved computational model that we developed, we investigated the individual roles of CM contraction and LM shortening, compared bolus transport with various levels of discoordination between CM and LM activation, and studied the role of the mucosa and how its stiffening influenced transport. These preliminary findings should help understand the synergy between LM, CM, and the mucosal layer in facilitating bolus transport, thereby providing insight into related physiology and pathophysiology. 1 The support of grant R01 DK56033 and R01 DK079902 from NIH is gratefully acknowledged 1:44PM R7.00004 3D separation over a wall-mounted hemisphere in steady and pulsatile flow1 , IAN A. CARR, MICHAEL W. PLESNIAK, George Washington University — Flow separation over a surface-mounted hemispheriod is prevalent in countless applications, both under steady (constant freestream velocity) and unsteady flow over the protuberance. Previous studies of 3D separation have been limited to steady inflow conditions. In biological and geophysical flows, pulsatile flow conditions are much more commonly observed, yet such conditions have not been well studied. Primarily motivated by previous studies of the flow observed in various human vocal fold pathologies, such as polyps, our research aims to fill the knowledge gap in unsteady 3D flow separation. This is achieved by characterizing surface pressure fields and velocity fields, focused primarily on the vortical flow structures and dynamics that occur around a hemispheroid protuberance under pulsatile flow conditions. Surface static pressure and two-dimensional, instantaneous and phase-averaged, particle image velocimetry data in steady and pulsatile flow are presented and compared. Coherent vortical flow structures have been identified using the λci vortex identification criterion. This analysis has revealed a novel set of flow structures dependent on the pulsatile flow forcing function. 1 This material is based in part upon work supported by the National Science Foundation under Grant Number CBET-1236351 1:57PM R7.00005 Wavelet analysis of hemispheroid flow separation toward understanding human vocal fold pathologies1 , DANIEL H. PLESNIAK, Haverford College, IAN A. CARR, KARTIK V. BULUSU, MICHAEL W. PLESNIAK, George Washington University — Physiological flows observed in human vocal fold pathologies, such as polyps and nodules, can be modeled by flow over a wallmounted protuberance. The experimental investigation of flow separation over a surface-mounted hemispheroid was performed using particle image velocimetry (PIV) and measurements of surface pressure in a low-speed wind tunnel. This study builds on the hypothesis that the signatures of vortical structures associated with flow separation are imprinted on the surface pressure distributions. Wavelet decomposition methods in one- and two-dimensions were utilized to elucidate the flow behavior. First, a complex Gaussian wavelet was used for the reconstruction of surface pressure time series from static pressure measurements acquired from ports upstream, downstream, and on the surface of the hemispheroid. This was followed by the application of a novel continuous wavelet transform algorithm (PIVlet 1.2) using a 2D-Ricker wavelet for coherent structure detection on instantaneous PIV-data. The goal of this study is to correlate phase shifts in surface pressure with Strouhal numbers associated with the vortex shedding. Ultimately, the wavelet-based analytical framework will be aimed at addressing pulsatile flows. 1 This material is based in part upon work supported by the National Science Foundation under Grant Number CBET-1236351, and GW Center for Biomimetics and Bioinspired Engineering (COBRE). 2:10PM R7.00006 Characterizing phonatory aeroacoustic sources using Lagrangian Coherent Structures , MICHAEL MCPHAIL, MICHAEL KRANE, Penn State University — Voice disorders that lead to changes in vocal fold geometry, or posturing, are known to substantially affect phonatory airflow topology. How these topology changes affect aeroacoustic sound sources is not well understood, however. This talk presents modelling aeroacoustic sources with Lagrangian Coherent Structures (LCS). Here we use the motion of dynamically distinct fluid regions, identified by the LCS, to predict sound. This approach provides a means to connect phonatory airflow topology changes to resulting changes in sound production. Simple validation cases of this approach will be shown. The application of LCS analysis to phonatory flows will be also presented. 2:23PM R7.00007 Self-oscillating Vocal Fold Model Mechanics: Healthy, Diseased, and Aging1 , ELIZABETH P. HIUBLER, LUCAS F. E. POLLOK, ADAM G. APOSTOLI, ADRIENNE B. HANCOCK, MICHAEL W. PLESNIAK, George Washington University — Voice disorders have been estimated to have a substantial economic impact of $2.5 billion annually. Approximately 30% of people will suffer from a voice disorder at some point in their lives. Life-sized, self-oscillating, synthetic vocal fold (VF) models are fabricated to exhibit material properties representative of human VFs. These models are created both with and without a polyp-like structure, a pathology that has been shown to produce rich viscous flow structures not normally observed for healthy VFs during normal phonation. Pressure measurements are acquired upstream of the VFs and high-speed images are captured at varying flow rates during VF oscillation to facilitate an understanding of the characteristics of healthy and diseased VFs. The images are analyzed using a videokymography line-scan technique. Clinically-relevant parameters calculated from the volume-velocity output of a circumferentially-vented mask (Rothenberg mask) are compared to human data collected from two groups of males aged 18-30 and 60-80. This study extends the use of synthetic VF models by assessing their ability to replicate behaviors observed in human subject data to advance a means of investigating changes associated with normal, pathological, and the aging voice. 1 Supported by the GWU Institute for Biomedical Engineering (GWIBE) and GWU Center for Biomimetics and Bioinspired Engineering (COBRE). 2:36PM R7.00008 Combining subject-specific and low-order modeling techniques to study fluid-structure interaction of rabbit phonation , SIYUAN CHANG, HAOXIANG LUO, CAROLYN NOVALESKI, BERNARD ROUSSEAU, Vanderbilt University — A subject-specific computational model has been developed to simulate flow-induced vocal fold vibration for evoked rabbit phonation. A freshly excised larynx was scanned using micro magnetic resonance imaging. Images were segmented to identify the vocal fold tissue and lumen surface. The 3D fluid-structure interaction (FSI) model was then constructed with experimentally measured flow parameters as input. The tissue deformation is assumed to be finite, and a previously developed FSI solver is used to simulate the coupled flow and nonlinear tissue mechanics. In addition, a one-dimensional flow model based on heuristic estimate of the flow separation point is used as an efficient tool to guide the full 3D simulation. This low-order model is motivated by presence of uncertainties in the tissue properties and boundary conditions, and it has proven to be very useful in our study. Similarities and differences in the vibration characteristics of the vocal fold predicted by these two models will be discussed. 2:49PM R7.00009 Study of non-linear deformation of vocal folds in simulations of human phonation1 , SHAKTI SAURABH, DANIEL BODONY, Univ of Illinois - Urbana — Direct numerical simulation is performed on a two-dimensional compressible, viscous fluid interacting with a non-linear, viscoelastic solid as a model for the generation of the human voice. The vocal fold (VF) tissues are modeled as multi-layered with varying stiffness in each layer and using a finite-strain Standard Linear Solid (SLS) constitutive model implemented in a quadratic finite element code and coupled to a high-order compressible Navier-Stokes solver through a boundary-fitted fluid-solid interface. The large non-linear mesh deformation is handled using an elliptic/poisson smoothening technique. Supra-glottal flow shows asymmetry in the flow, which in turn has a coupling effect on the motion of the VF. The fully compressible simulations gives direct insight into the sound produced as pressure distributions and the vocal fold deformation helps study the unsteady vortical flow resulting from the fluid-structure interaction along the full phonation cycle. 1 Supported by the National Science Foundation(CAREER Award Number 1150439) 3:02PM R7.00010 Simulations of acoustic waves in channels and phonation in glottal ducts , JUBIAO YANG, Rensselaer Polytechnic Institute, MICHAEL KRANE, Pennsylvania State University, LUCY ZHANG, Rensselaer Polytechnic Institute — Numerical simulations of acoustic wave propagation were performed by solving compressible Navier-Stokes equations using finite element method. To avoid numerical contamination of acoustic field induced by non-physical reflections at computational boundaries, a Perfectly Matched Layer (PML) scheme was implemented to attenuate the acoustic waves and their reflections near these boundaries. The acoustic simulation was further combined with the simulation of interaction of vocal fold vibration and glottal flow, using our fully-coupled Immersed Finite Element Method (IFEM) approach, to study phonation in the glottal channel. In order to decouple the aeroelastic and aeroacoustic aspects of phonation, the airway duct used has a uniform cross section with PML properly applied. The dynamics of phonation were then studied by computing the terms of the equations of motion for a control volume comprised of the fluid in the vicinity of the vocal folds. It is shown that the principal dynamics is comprised of the near cancellation of the pressure force driving the flow through the glottis, and the aerodynamic drag on the vocal folds. Aeroacoustic source strengths are also presented, estimated from integral quantities computed in the source region, as well as from the radiated acoustic field. 3:15PM R7.00011 Measurements of phonatory aeroacoustic source strengths in a physical model , MICHAEL KRANE, MICHAEL MCPHAIL, Penn State University — Aeroacoustic sources due to flow-induced vibration of a compliant constriction in a duct were characterized experimentally. The principal goal of this study is to estimate the character and level of the various sources of sound in human voice production. Measurements were performed in a model of the human airway, constructed to human dimensions, but with an idealized geometry. The airway duct models the passage from the trachea to the mouth, as a constant-area (7.64cm2 ) square cross-section, interrupted only by the model vocal folds. These were fabricated in two layers of soft silicone rubber. Time-resolved measurements included subglottal and supraglottal absolute pressure, sound pressure at the model vocal tract “mouth,” and high-speed video of the model vocal folds. These were sampled synchronously at 22 kHz. Steady-state measurements included subglottal pressure and volume flow rate. Measurements were conducted over a subglottal pressures range of 2.25-2.80 kPa. Source strengths were estimated by theoretical expressions, using the measured pressures and glottal area as inputs. Results show that the dipole source typically associated with vocal fold drag is the dominant source. Furthermore, for the vibration pattern observed in these experiments, glottal jet turbulence dominates the dipole source above approximately 1 kHz. Tuesday, November 25, 2014 1:05PM - 3:02PM Session R8 Biofluids: Microswimmers IV - Complex Fluids — 3001/3003 - Alexander Morozov, University of Edinburgh 1:05PM R8.00001 Swimming of bacteria in polymer solutions , ALEXANDER MOROZOV, VINCENT MARTINEZ, JANA SCHWARZ-LINEK, MATHIAS REUFER, School of Physics & Astronomy, University of Edinburgh, LAURENCE WILSON, Department of Physics, University of York, UK, WILSON POON, School of Physics & Astronomy, University of Edinburgh — The “standard model” of bacteria swimming in polymer solutions consists of experimental observations that the swimming speed first increases and then decreases as the function of the polymer concentration. This non-monotonic behaviour is usually explained by either swimming in pores in the polymer solutions or by its viscoelasticity. Using new, high-throughput methods for characterising motility, we have measured the swimming speed and the angular frequency of cell-body rotation of motile Escherichia coli as a function of polymer concentration in polyvinylpyrrolidone (PVP) and Ficoll solutions of different molecular weights. We find that non-monotonic speed-concentration curves are typically due to low-molecular weight impurities and, when cleaned, most molecular weight solutions exhibit Newtonian behaviour. For the highest molecular weight of PVP we observe non-newtonian effects. We present a simple theory that consists of the fast-rotating flagella “seeing” a lower viscosity than the cell body but otherwise Newtonian in nature. We show that our theory successfully describes the experimental observations and suggest that flagella can be seen as nano-rheometers for probing the non-newtonian behaviour of high polymer solutions on a molecular scale. 1:18PM R8.00002 Swimming of Taylor wavy sheets in viscoelastic fluids , ALEXANDER MOROZOV, School of Physics & Astronomy, University of Edinburgh — We consider a model swimmer, the Taylor wavy sheet, moving in a viscoelastic fluid. Based on the solution obtained by E. Lauga (Phys. Fluids ’97), we propose a mechanism for sheet’s propulsion in elastic fluids. We present a full numerical calculation of swimming at arbitrary amplitudes, and compute the most efficient and fastest waveforms for undulatory swimming in bulk and next to a boundary. 1:31PM R8.00003 Flexible polymers suppress wobbling and tumbling of E. coli cells , ALISON KOSER, PAULO ARRATIA, University of Pennsylvania — The run-and-tumble dynamics of swimming E. coli has been extensively studied. In this talk, we experimentally investigate the role of polymer concentration on the swimming dynamics of E. coli using tracking methods. We find that the addition of small amount of polymer to water drastically changes the run-and-tumble behavior of E. coli cells, enhancing translation while hindering rotational diffusion. Here, the cells are suspended in dilute solutions of carboxy-methyl cellulose (CMC) and imaged in a liquid film away from surfaces. The addition of polymer molecules to the fluid (water) leads to cell trajectories that are highly correlated in time; cells move in nearly straight lines and rotational diffusion is greatly reduced. By varying the polymer molecular weight, we show that trajectories are a result of two mechanisms: (1) suppression of cell wobbling due to elasticity and (2) enhancement of run times due to viscosity. Our experiments show that this combination of increased speed and suppressed reorientation dramatically changes overall cell dynamics in the presence of polymers. 1:44PM R8.00004 Enhanced helical swimming in Boger fluids , FRANCISCO GODINEZ, RODRIGO MENDEZROJANO, ROBERTO ZENIT, Universidad Nacional Autonoma de Mexico, ERIC LAUGA, University of Cambridge — We conduct experiments with force-free magnetically-driven helical swimmers in Newtonian and viscoelastic (Boger) fluids. In order assess the effect of viscoelasticity on the swimming performance, we conduct experiments for swimmers with different helical tail geometries. We use helices with the same wave length and total length but vary the angle of the helix. As previously reported by the computational study of Spagniole and collaborators, we found that the swimming performance can either increase, decrease or remain unchanged, depending on the geometry of the tail. With the right geometry, the enhancement can be up to a factor of two. 1:57PM R8.00005 Mechanisms of elastic enhancement and hindrance for finite length undulatory swimmers in viscoelastic fluids , BECCA THOMASES, ROBERT GUY, University of California, Davis — A computational model of finite-length undulatory swimmers is used to examine the physical origin of the effect of elasticity on swimming speed. We explore two distinct target swimming strokes, one derived from the motion of C. elegans, with large head undulations, and a contrasting stroke with large tail undulations. We show that both favorable stroke asymmetry and swimmer elasticity contribute to a speed-up, but a substantial boost results only when these two effects work together. We reproduce conflicting results from the literature simply by changing relevant physical parameters. 2:10PM R8.00006 Theory of locomotion in complex fluids , GWYNN ELFRING, University of British Columbia, ERIC LAUGA, University of Cambridge — Microorganisms often swim in environments that cannot be classified as Newtonian. Biological fluids can contain polymers or other heterogeneities which may yield complex rheology. For a given set of boundary conditions flows can be substantially different in complex fluids, while non-Newtonian stresses can alter the gait of the microorganisms themselves. Heterogeneities in the fluid may also occur on length scales on the order of the swimmer leading to additional complexity. In this talk we will discuss a theoretical description of the effects on locomotion of a non-Newtonian constitutive relation and discuss our current understanding of the interplay between swimming kinematics and the nonlinear response of the fluid. 2:23PM R8.00007 Ciliary kinematics of Chlamydomonas reinhardtii in Complex Fluids: Role of viscosity , ARVIND GOPINATH, Department of Physics and Astronomy, Haverford College, BOYANG QIN, PAULO ARRATIA, School of Engineering and Applied Sciences, University of Pennsylvania — The motility behavior of microorganisms can be significantly affected by the rheology of their fluidic environment. Guided by our experiments on the swimming gait of Chlamydomonas reinhardtii in viscoelastic fluids, we focus on ciliary waveforms in Newtonian fluids and systematically study the effect of increasing viscosity. We find that the beat frequency as well as the wave speed are both strongly influenced by fluid viscosity. Interestingly, ciliary waveforms at low viscosity show a larger influence of the cell body than waveforms at higher viscosity. We use slender body theory and principal component analysis to elucidate the role of fluid viscosity in regulating the kinematics of the swimming process. 2:36PM R8.00008 Swimming and pumping of helical structures in viscous fluids , LEI LI, SAVERIO SPAGNOLIE, UW-Madison — Many flagellated microorganisms including E. coli swim by rotating slender helical flagella, while ciliated organisms like Paramecia swim by passing helical waves along their surfaces. We will discuss a framework for studying such problems where the Stokes equations describing viscous flow are written in helical coordinates. Analytical predictions match well with full numerical simulations, and suggest optimal geometries. This work may also aid designs in microfluidic manipulation, microswimmer engineering, and the mixing of viscous fluids. 2:49PM R8.00009 Fluid transport by an unsteady microswimmer1 , PETER MUELLER, JEAN-LUC THIFFEAULT, University of Wisconsin - Madison — We study the drift caused by the microscopic algae Chlamydomonas reinhardtii. This microorganism swims by rapidly beating two frontal flagella. Previous studies of transport by microswimmers have neglected the ubiquitous time-dependence of their swimming. We model the organism by a time-dependent dumbbell consisting of two regularized Stokeslets. We study individual particle paths and their displacements in a region around the swimmer. Of particular interest are particle trajectories that remain trapped near the swimmer, forming the so-called “atmosphere” of the moving body. Atmospheres are common in the steady case, and they persist for unsteady motion though their size is reduced due to broken barriers. We vary the parameters in our model to determine their effect on the size and shape of the atmosphere. Finally we determine the importance of this atmosphere on overall fluid transport and mixing. 1 Supported by NSF grant DMS-1109315 Tuesday, November 25, 2014 1:05PM - 3:02PM Session R9 Biofluids: General — 3014/3016 - David Hu, Georgia Institute of Technology 1:05PM R9.00001 An experimental and theoretical approach to a simplified model of human birth1 , ALEXA BAUMER, ANDREA LEHN, George Washington University, JAMES GROTBERG, University of Michigan, MEGAN C. LEFTWICH, George Washington University — his study investigates the effects of amniotic fluid and vernix caseosa, as well as the uterine contraction wave dynamics, on the forces associated with human birth. An experimental model of the fetus passing through the birth canal is represented as concentric cylinders with a fluid filled gap. The rigid inner cylinder moves through the highly flexible outer cylinder while stabilized on a track. The inner cylinder is pulled through the system with constant velocity. As it moves, the rigid cylinder’s position is recorded with a high-speed camera and the force is simultaneously measured. A perturbation solution considers the upper boundary as the uterine wall with a peristaltic wave. The lower boundary is the fetus traveling at constant velocity. Assuming lubrication theory and a small Reynolds number, the Navier-Stokes Equations and conservation of mass are solved for an expression for shear stress at the wall. This solution, and the experimental results, are compared to the exact Couette flow solution for constant gap width. This model can be used as the foundation for predicting the force needed to deliver a fetus in the final stages of parturition. From the concentric cylinders representation of human delivery, more complex and geometrically accurate models can be generated. 1 NSF CBET-1437611 1:18PM R9.00002 Transport generated by mayfly nymphs to breathe , RODOLPHE CHABREYRIE, Department of Mechanical and Aerospace Engineering, The George Washington University, KHALED ABDELAZIZ, Department of Mechanical Engineering, University of Maryland, ELIAS BALARAS, Department of Mechanical and Aerospace Engineering, The George Washington University, KENNETH KIGER, Department of Mechanical Engineering, University of Maryland — In order to maintain their metabolism, many species of mayfly nymphs utilize an oscillating array of wingshaped gills to augment extraction of dissolved oxygen from the surrounding water. As a nymph develops, the kinematics of these gills have been observed to abruptly change from a rowing-like to a flapping-like motion. To better understand the role of this abrupt kinematic change, we study the transport of dissolved oxygen, viewed as a passive scalar surrounding the gills, for an in-silico mayfly nymph. In particular, through a Lagrangian and stochastic dynamical systems approach, we simulate the advection and diffusion of this passive scalar, and reveal the key structures of the transport generated by the gills for both flapping and rowing kinematics. In this talk, we show how the switch from rowing to flapping enables the generation of a better transport skeleton (i.e. breading of Lagrangian Coherent Structures) and how such a transport skeleton influences the oxygen uptake. 1:31PM R9.00003 Behavioral Response of Atlantic Mud Crab Megalopae to Coherent Shear Flows , D.R. WEBSTER, A.C. TRUE, M.J. WEISSBURG, J. YEN, Georgia Tech — Behavioral assays with megalopae of the Atlantic mud crab (Panopeus herbstii) were performed in a laboratory mimic of hydrodynamic structure associated with fronts and clines. A laminar, planar free jet was used to create finescale upwelling, downwelling, and horizontal shear flows. Analyses of digitized trajectories established orientation-specific behavioral shear strain rate thresholds in the range 0.04 – 0.1 s−1 . Changes in average kinematics revealed area-restricted searching in the vicinity of horizontal shear flows and excited area-restricted searching in the vicinity of both vertical shear flows, although not in the layers themselves. These responses could produce aggregations in the vicinity of coherent shear flows, although there is avoidance of vertical flow regions. Reduction in the net-to-gross displacement ratio (NGDR) and the vertical-net-to-gross displacement ratio (VNGDR) with respect to stagnant conditions indicate that trajectories become more sinuous and that larvae enhance depth-keeping in all shear flows. Collectively, this is consistent with foraging and sampling behaviors in which shear flow cues larvae to restrict search volume in hopes of exploiting some coincident cue or resource patch, typical in fronts and clines. Area-restricted searching along with depth-keeping seen here reveals that larvae may utilize the information contained in coherent shear flows to optimize needs operating on disparate space and time scales (e.g. foraging and site selection for settlement). 1:44PM R9.00004 Flow inside an eye under vitreous surgery1 , DAIKI KONO, SHUN SAKAMOTO, JUN SAKAKIBARA, Department of Mechanical Engineering, Meiji University — Vitreous is a clear gel filling the space between crystalline lens and retina in human eye. Under circumstances where the vitreous becomes opaque due to bleeding or other disease, ophthalmologist removes the vitreous from eye by cutting and sucking through a pipe named vitreous cutter, and meanwhile replaces fluid in the eye with a balanced salt solution by injecting it through the infusion port. Jet flow from the infusion port may cause intense flow. Consequently, this may generate a pressure and a shear stress on the retinal wall and possibly lead to the damage of retinal cell. In this study, we visualized the flow inside eye and estimated the shear stress on the retinal wall under the vitreous surgery. Instead of using human eye, we used a spherical shell model simulating human eyeball, and measured the two dimensional distribution of two-component velocity by PIV. Under the condition of Re=66 to 99, which meet in the actual operation, the maximum shear stress reaches 0.4 Pa. This value is insufficient to cause retinal detachment, while any physiological effect on the retinal endothelial cells is still unclear. Flow field under higher Re will be presented in the talk. 1 Supported by Grants-in-Aid for Scientific Research of Japan Society for the Promotion of Science under Grant No. 25289026. 1:57PM R9.00005 3D simulation of floral oil storage in the scopa of South American insects1 , ALEXANDER RUETTGERS, MICHAEL GRIEBEL, Institute for Numerical Simulation, University of Bonn, LARS PASTRIK, HEIKO SCHMIED, DIETER WITTMANN, Institute of Crop Science and Resource Conservation, University of Bonn, ANDREAS SCHERRIEBLE, ALBRECHT DINKELMANN, THOMAS STEGMAIER, Institute of Textile Technology and Process Engineering, Denkendorf, INSTITUTE FOR NUMERICAL SIMULATION TEAM, INSTITUTE OF CROP SCIENCE AND RESOURCE CONSERVATION TEAM, INSTITUTE OF TEXTILE TECHNOLOGY AND PROCESS ENGINEERING TEAM — Several species of bees in South America possess structures to store and transport floral oils. By using closely spaced hairs at their back legs, the so called scopa, these bees can absorb and release oil droplets without loss. The high efficiency of this process is a matter of ongoing research. Basing on recent x-ray microtomography scans from the scopa of these bees at the Institute of Textile Technology and Process Engineering Denkendorf, we build a three-dimensional computer model. Using NaSt3DGPF, a two-phase flow solver developed at the Institute for Numerical Simulation of the University of Bonn, we perform massively parallel flow simulations with the complex micro-CT data. In this talk, we discuss the results of our simulations and the transfer of the x-ray measurement into a computer model. 1 This research was funded under GR 1144 / 18-1 by the Deutsche Forschungsgemeinschaft (DFG) 2:10PM R9.00006 Physics of Weightlifting1 , CAROLINE COHEN, LadHyX, Ecole Polytechnique — In the footsteps of J.B. Keller who determined the optimal strategy to run a race [1], we investigate weightlifting records. We measure the dynamics of lifting barbells of different masses at Bench Press for different athletes. To understand the shape of experimental results, we need both a macroscopic mechanic model and microscopic description of muscle contraction. We dive into muscle in order to understand the relation between force generated by the muscle and its contraction velocity [2,3] and draw a capillary analogy of muscle contraction. Finally we use the Deshcherevskii kinetik model [4] and derive the dynamics of the barbell. From the fit between data and predictions, we extract microscopic characteristics of muscles. We consider to apply this protocole to diagnose muscle aging or dysfunctions. [1] Keller, J. B. (1973). iA theory of competitive running. Physics today, 43. [2] Hill, A. V. (1938). The heat of shortening and the dynamic constants of muscle. Proceedings of the Royal Society of London. Series B, Biological Sciences, 136-195. [3] Huxley, H. E. (1990). Sliding filaments and molecular motile systems. J. Biol. Chem, 265(15), 8347-8350. [4] Deshcherevskii, V. I. (1971). A kinetic theory of striated muscle contraction. Biorheology, 7(3), 147-170. 1 C. Cohen, B. Darbois Texier, D. Quere, G. Laffaye, C. Clanet 2:23PM R9.00007 Shape-assisted body reorientation enhances trafficability through cluttered terrain , CHEN LI, ANDREW PULLIN, DUNCAN HALDANE, RONALD FEARING, ROBERT FULL, University of California, Berkeley — Many birds and fishes have slender, streamlined bodies that reduce fluid dynamic drag and allow fast and efficient locomotion. Similarly, numerous terrestrial animals run through cluttered terrain where 3-D, multi-component obstacles like grass, bushes, trees, walls, doors, and pillars also resist motion, but it is unknown whether their body shape plays a major role. Here, we challenged discoid cockroaches that possess a rounded, thin, nearly ellipsoidal body to run through tall, narrowly spaced, grass-like beams. The animals primarily rolled their body to the side to maneuver through the obstacle gaps. Reduction of body roundness by artificial shells inhibited this side roll maneuver, resulting in a lower traversal probability and a longer traversal time (P <0.001, ANOVA). Inspired by this discovery, we added a cockroach-like, rounded exoskeleton shell to a legged robot of a nearly cuboidal body. The rounded shell enabled the robot to use passive side rolling to maneuver through beams. To explain the mechanism, we developed a simple physics model to construct an energy landscape of the body-terrain interaction, which allowed estimation of body forces and torques exerted by the beams. Our model revealed that, by passive interaction with the terrain, a rounded body (ellipsoid) rolled more easily than an angular body (cuboid) to access energy valleys between energy barriers caused by obstacles. Our study is the first to demonstrate that a terradynamically “streamlined” shape can reduce terrain resistance and enhance trafficability by assisting body reorientation. 2:36PM R9.00008 Mathematical Modeling of Tear Film Break up Modes and Fluorescent Intensity1 , JAVED SIDDIQUE, Penn State York, RICHARD BRAUN, University of Delaware, C.G. BEGLEY, School of Optometry, Indiana University, Bloomington, IN, P.E. KING-SMITH, College of Optometry The Ohio State University, Columbus, OH — We develop a mathematical model for variables of interest in tear film break up (TBU) to compare with experimental images of TBU to better predict local values of tear film (TF) osmolarity and fluorescence during and following the TBU. Models are developed for local changes in TF thickness, osmolarity and fluorescein concentration. Fluorescence concentration was converted to fluorescent intensity using the expression involving film thickness and the full range of fluorescence (Nichols et al (IOVS 2012). The fluorescent intensity response is a primary tool for visualizing the TF thickness, and it is qualitatively different in the dilute vs concentrated regimes. Computed results over a wide range of fluorescein concentrations show that evaporation rate led to thinner regions where TBU first occurs. The computed results will be closely compared with experimential fluorescence and other imaging techniques to help determine relevant parameters. The model predicts locally elevated concentration of osmolarity within areas of TBU and predicts osmolarity in these regions which can’t be measured experimental results to date. The osmolarity may increase from 50% to 1300% of the isosmolar value, depending sensitively on the corneal permeability and diffusivity of solutes in tear film. 1 J. I. Siddique would like to acknowledge the support from Simons Foundation. 2:49PM R9.00009 The hydrodynamics of defecation , PATRICIA YANG, DUC DAO, RICHARD LEHNER, MIKE TENNENBAUM, ALBERTO FERNANDEZ-NIEVES, DAVID HU, Georgia Institute of Technology — According to the U.S. Department of Health and Human Services, digestive disease affects 60 to 70 million people and costs over 140 billion annually. Despite the significance of the gastrointestinal tract to human health, the physics of both digestion and defecation remain poorly understood. In this combined experimental and theoretical study, we investigate the defecation of mammals, from mice to elephants. We film defecation events at Zoo Atlanta and apply plate-on-plate rheometry to measure the viscosity of mammalian feces. Among animals heavier than 3 kg, we find herbivores defecate for only 10 seconds (N = 13), while carnivores do so for 19 seconds (N = 8). We rationalize this surprising trend on the basis of the higher viscosity of carnivore feces. We compare defecation times to theoretical predictions based on a Poiseuille flow model of the rectum and parallel experiments with a synthetic defecator that extrudes pizza dough upon applied pressure. Our findings may help to diagnose digestive problems in animals. Tuesday, November 25, 2014 1:05PM - 3:28PM Session R10 Microscale Flows: Particle Sorting and Control — 3005 - Claire Hur, Harvard University 1:05PM R10.00001 Separation of polymers by length in rotational flow , FAIHAN ALFAHANI, School of Engineering, Computer Science, and Construction Managment, JENNIFER KREFT PEARCE, Department of Physics, Roger Williams University — We use a lattice-Boltzmann based Brownian dynamics simulation to determine if polymers of different lengths can be separated by a combination of a trapping force and fluid flow. We produce two counter-rotating vortices in the simulation, similar to the work of HIlgenfeldt, et al., that used rotational flow to separate colloids of different size. We can achieve separation of polymers that differ in length by as little as 30%. We expect that this technique could be used in a microfluidic device to analyze the size of long DNA fragments produced in common molecular biological tests. 1:18PM R10.00002 High throughput sorting of spherical particles in inertial microfluidics , PHANINDRA TALLAPRAGADA, SENBAGARAMAN SUDARSANAM, NILESH HASABNIS, Clemson University — The fundamental problem of sorting sphere like particles by size in flows at low Reynolds numbers in confined geometries is one which is frequently encountered in microfluidic engineering. The inertial sorting of particles in Dean flows, demonstrated in pioneering work by Papautsky and Bhagat et.al and DiCarlo et.al is specific to particular sizes of particles and it is not apparent how particles of different or larger sizes could be sorted. This is because the phenomena of particle focusing across a large parametric regime is poorly understood. Additionally the unexplored case where larger particles need to be sorted by size is especially important in applications involving large cells such as Islet cells whose diameter can vary from 50 µm to 200 µm. We characterize the transitions in particle focusing with changing channel Reynolds number, particle Reynolds number and the Dean number and exploit these transitions to sort particles by size. Based on such transitions, particles across size ranges of 3 µm to 100 µm in various 2-particle mixtures are sorted. We also find that this separation occurs in a narrow range of channel Reynolds number. We demonstrate our findings by sorting particles in different mixtures. 1:31PM R10.00003 Kinetics of colloidal gold nanoparticle chain assembly via in situ liquid cell electron microscopy observations1 , TAYLOR WOEHL, TANYA PROZOROV, Ames Laboratory, EMERGENT ATOMIC AND MAGNETIC STRUCTURES TEAM — Various types of colloidal nanoparticles are known to self-assemble into hierarchical mesostructures via anisotropic interparticle interactions. Previous modeling and experiments have suggested that dipolar interactions may be responsible for assembly of one dimensional nanoparticle chain structures; however, due to a lack of in situ observations little is known about the kinetics of the self-assembly. Here we use real-time nanoscale observations to measure the self-assembly kinetics of colloidal gold nanoparticles into one dimensional chains. Gold nanoparticles suspended in acetate buffer were observed via in situ liquid electron microscopy to self-assemble into chains of 5-10 nanoparticles over a time of minutes. Self-assembly is initiated upon irradiation of the nanoparticles with the imaging electron beam. Measurements of the self-assembly kinetics revealed that the chains formed via second order aggregation kinetics during the first tens of seconds. We investigate the effects of the electron beam current and ionic strength of the buffer solution on the effective aggregation rate and chain formation mechanism. Our observations suggest that the aggregation rate increases with the effective diffusivity of the nanoparticles. 1 T.P. acknowledges support from the Department of Energy Office of Science Early Career Research Award, Biomolecular Materials Program. This work was supported by the U.S. Department of Energy, Office of Basic Energy Science, Division of Materials Sciences 1:44PM R10.00004 Solution-based electric-field-assisted assembly of vertically aligned CNT membranes , RICHARD CASTELLANO, CEVAT AKIN, JERRY SHAN, Rutgers University — Carbon-nanotube (CNT) membranes are of interest due to experiments and simulations showing flow through nanotubes to be 3 to 5 orders of magnitude faster than predicted by viscous flow theory. Thus, membranes incorporating vertically aligned CNTs (VACNTs) as through-pores offer promise as highly efficient and permeable membranes for a variety of filter and separation processes. However, current membrane-fabrication techniques utilizing CVD-grown VACNT arrays are costly and difficult to scale up. We are developing a solution-based, electric-field-assisted approach as a cost-effective and scalable method to producing large-area VACNT membranes and composites. Post-growth nanotubes are first dispersed in a polymeric matrix and then aligned with an AC electric field. A DC component induces electrophoresis to the CNTs to significantly increase the VACNT number density. This composite field also introduces complex fluid motion caused by induced-charge electro-osmosis and the electrochemistry of the fluid/electrode interface. We experimentally probe all of these effects and consider factors affecting the number density and spatial uniformity of VACNT membranes. We also consider the basic electrokinetics of nanotube alignment under spatially uniform AC electric fields, making quantitative comparison with classical models of the dynamics of polarizable, 1D particles under the combined effects of electric fields, hydrodynamic drag, and Brownian motion. We conclude by discussing the implications of these fundamental electrohydrodynamic studies for producing large-area membranes containing aligned CNTs. 1:57PM R10.00005 Solution-Based Electro-Orientation Spectroscopy for the Automated, Quantitative Characterization and Sorting of 1D Nanomaterials , CEVAT AKIN, JERRY SHAN, JINGANG YI, LEONARD FELDMAN, Rutgers University, CORENTIN DURAND, AN-PING LI, Oak Ridge National Laboratory — The electrical-transport properties of 1D nanomaterials are often poorly known and vary with size and surface effects. Traditional quantitative characterization methods require specialized facilities and are usually slow, invasive and not suitable for the large number of measurements needed to statistically characterize samples with a heterogeneous distribution of properties. Here, we introduce a contactless, solution-based method to rapidly and quantitatively measure the electrical properties of 1D nanomaterials based on their transient alignment behavior in AC electric fields of different frequencies. The electro-orientation method can be automated and is compatible with further solution-based techniques for nanowire alignment and assembly, including electrophoresis, dielectrophoresis and flow control. We demonstrate the accuracy of the solution-based method using a variety of insulating, semiconducting and metallic nanowires, and show that electro-orientation spectroscopy can detect true nanoscale surface effects on the electrical conductivity of 1D nanomaterials. We further discuss our progress toward implementing the method in a microfluidic device capable of automated electrical characterization and sorting of nanowires and nanotubes. 2:10PM R10.00006 Rapid Electrokientic Patterning of Metal Nanoparticles and Nanowires , AVANISH MISHRA, Purdue Univ, STUART WILLIAMS, University of Louisville, STEVEN WERELEY, Purdue Univ — Rapid Electrokinetic Patterning (REP) combines electric field and laser induced heating for particle trapping on an electrode surface. This technique utilizes two planar transparent indium tin oxide (ITO) electrodes separated by a colloidal solution of suitable thickness. When an infrared (1064 nm) laser beam is projected on the electrode surface, due to interaction between AC electric field and laser induced heating, a toroidal electrothermal (ET) vortex is generated. It traps particles and brings them closer to the electrode surface where particles are captured by particle-electrode interactions. In this work, we demonstrate trapping of metal nanoparticles and discuss its application in Surface Enhanced Raman Scattering for trace analyte detection. 2:23PM R10.00007 Viscous flow within an embedded serpentine channel as a mechanism to create time-dependent deformation patterns of elastic beams1 , YOAV MATIA, AMIR GAT, Technion - Israel Institute of Technology — We analyze the time dependent interaction between the flow-field and the elastic deformation-field of a viscous liquid within a long serpentine channel, embedded in an elastic beam. The channel is positioned asymmetrically with regard to the midplane of the elastic beam. We focus on creeping flows and small deformations of the elastic beam and obtain, in leading order, a diffusion equation governing the pressure-field within the serpentine channel. The deformation of the beam is then related to the propagation of pressure within the channel. We thus obtain a viscous-elastic equation governing the deformation of the beam due to the viscous flow within the serpentine channel. This equation enables to design complex time-dependent deformation patterns of beams with embedded channel networks, relevant to soft-robotic applications. Our theoretical results were illustrated and verified using numerical computations. 1 Israel Science Foundation 818/13 2:36PM R10.00008 Different modes of self-assembly for same-stream and cross-stream microparticles in inertial microflows , SOROUSH KAHKESHANI, Graduate student at UCLA, DINO DI CARLO, Associate professor of Bioengineering at UCLA — Understanding parameters affecting dynamic self-assembly of particles in microchannels can enable control of particle density for applications such as flow cytometry and tissue printing. Inertial lift forces and repulsive viscous interactions have been shown to have important effects on inter particle spacing and dynamic particle pairwise interactions. Based on the aspect ratio of the channel, particles inertially focus to two or four positions in finite Reynolds number flows. In this work, we show that in channels with aspect ratio of two or greater, where we have predominantly two focused streams of particles, there is a favored same-stream spacing as well as favored cross-stream spacing between particles. We studied how channel geometry, particle size, concentration of particles, orientation of the neighboring particles, and Reynolds number can affect both cross-stream and same-stream spacing. In addition, based on our simulations, for the first time we showed that particle size and position of the particle in the channel significantly affect the shape of reversing streamlines behind and in front of the particles, however Reynolds number does not have a controlling effect on shape of these reversing streamlines. These results show unique features of particle interactions. 2:49PM R10.00009 Rapid Electrokinetic Patterning for Vertical Stacking and Manipulation of Particles , KATHERINE CLAYTON, AVANISH MISHRA, STEVEN WERELEY, Purdue University — A variety of optical and optoelectrical-based microfluidics techniques have been used for the trapping and manipulation of particles in a colloidal solution. Rapid Electrokinetic Patterning (REP) is one such technique. It uses laser activated electrothermal flow to trap particles in a monolayer. Particles can be manipulated on the substrate by steering the laser. In this work we show that by a careful selection of parameters, particles can be rapidly trapped in a tower configuration instead of a monolayer. Moreover, this vertical tower can be manipulated and stationed at any desirable place on the chip. We intend to discuss underlying physical mechanism and potential applications in biology. 3:02PM R10.00010 Dielectric Decrement Effects on Nonlinear Electrophoresis of Ideally Polarizable Particles , JEFFREY L. MORAN, WAI HONG RONALD CHAN, CULLEN R. BUIE, Massachusetts Institute of Technology, BRUNO FIGLIUZZI, Ecole des Mines de Paris — We present numerical simulations of nonlinear electrophoresis of ideally polarizable particles that specifically include the effects of a spatially non-uniform dielectric permittivity near the particle surface. Models for this dielectric decrement phenomenon have been developed by several authors, including Ben-Yaakov et al. [J. Phys. Condens. Matter 2009] Hatlo et al. [EPL 2012], and Zhao & Zhai [JFM 2013]. We extend this work to ideally polarizable particles and include the effects of surface conduction and advective transport in the electric double layer. By numerically solving for the coupled velocity field, electric potential, and ionic concentration distributions in the bulk solution surrounding the particle, we demonstrate that the dielectric decrement model predicts ionic saturation around the particle and thus physical implications that resemble those resulting from the steric model developed by Kilic et al. [PRE 2007], albeit with differences that reflect the nonlinearity of the modified Poisson-Boltzmann equation. In addition, we develop a generalized condensed layer model that approximates both the steric and dielectric decrement models in the limits of strong electric fields and negligible surface conduction to obtain more physical insights into these models. We demonstrate that the mobility in both models asymptotically scales as the square root of the electric field at high fields, recovering the result of Bazant et al. [Adv. Colloid Interface Sci 2009]. 3:15PM R10.00011 Continuous Nanoparticle Size Separation Using Microfluidic Technology , BUSHRA TASADDUQ, GONGHAO WANG, WENBIN MAO, WILBUR LAM, ALEXANDER ALEXEEV, TODD SULCHEK, Georgia Institute of Technology — High throughput size based separation of nanoparticles is important to better understand and improve diagnosis of diseases that involve nanoparticles. We propose a novel microfluidic device capable of continuous size-dependent separation of particles. The separation device consists of a microchannel with periodically arranged diagonal ridges. The key to the separation is that these diagonal ridges create helical flow fields. Simultaneously, inertial particle migration alters the particle height in a size-dependent manner, which then exposes the particle to different secondary flows. The height-dependent secondary flows then cause particles with different sizes to migrate transversely with unique trajectories. We have characterized the separation results utilizing forward and side scatter flow cytometric analysis. We are able to separate 4 micrometer particles from 7 micrometer; 0.5 micrometer from 5 micrometer; and platelets from RBCs and WBCs with a substantial enrichments of number densities of 29-fold, 227-fold, and 53-fold respectively .We demonstrate there exists a z-position dependent phenomena which affects particle trajectories and hypothesize that controlling the particle z-position, we can further improve the efficiency of size based sorting. Tuesday, November 25, 2014 1:05PM - 3:28PM Session R11 Waves III: Nonlinear Waves and Turbulence — 3007 - Fabrice Veron, University of Delaware 1:05PM R11.00001 Turbulent dynamics of breaking internal gravity waves on slopes , ROBERT ARTHUR, OLIVER FRINGER, Stanford University — The turbulent dynamics of breaking internal gravity waves on slopes are investigated using a high-resolution numerical model. A Navier-Stokes code is employed in an idealized, three-dimensional domain where an internal solitary wave of depression impinges upon a sloping bottom. A bottom-following curvilinear grid is used to capture the bathymetry accurately, and the vertical grid spacing ∆z+ =O(1) near the bottom in the breaking region to resolve the near-wall flow. In order to understand the transition to turbulence as a result of wave breaking, flow variability is analyzed in the cross-stream dimension. In particular, streamwise vorticity, or secondary streamwise rolls that lead to the turbulent breakdown of the wave, is found to develop in regions of unstable stratification. Dissipation and irreversible mixing of the density field are analyzed as a function of time, and related to breaking dynamics; irreversible mixing is quantified in terms of the change in background potential energy inside the domain. The mixing efficiency is also calculated for various wave and slope conditions. These results have application to the nearshore coastal ocean, where breaking internal waves affect the distributions of ecologically important scalars such as temperature, oxygen, and nutrients. 1:18PM R11.00002 Modeling the damping rates of surface water waves , GIRISH KUMAR RAJAN, DIANE HENDERSON, The Pennsylvania State University — In this work, we investigate linear damping rates of surface water waves in both laboratory and ocean settings. We formulate two models that generalize previous work to include the effects of air and two interfacial conditions. The first is a two-fluid model (representing air and water) with a monomolecular film at the interface. The second is a three-fluid model that has a Newtonian fluid with variable, but thin, thickness at the interface of the water and air. Limiting cases of these models reduce to expressions that agree with previously published results. We compare predictions of damping rates to both laboratory and oceanographic data. These models have been developed for a general fluid system and may thus be used for fluids other than air and water, with any general fluid-film at the interface, whose properties are known. 1:31PM R11.00003 Large Eddy Simulation of Disturbance Waves and Heat Transfer in Annular Flows1 , GEOFF HEWITT, JUNFENG YANG, OMAR MATAR, Imperial College London — A numerical method for forced convective boiling in an annulus needs to be developed in order to elucidate the reason for nucleation enhancement by disturbance waves. The benchmark test case is the experiment of Barbosa et al., in which nucleate boiling in a liquid film, droplet entrainment, disturbance waves of the liquid film, and their interaction were observed. We first develop a numerical strategy to model the development of disturbance waves in annular flows in which the highly turbulent gas core flow drives the laminar liquid flow upwards using advanced CFD tool TransAT. Then, the heat transfer process in the non-boiling annular flow was investigated to provide insight into the temperature gradient underneath the wave region. Agreement with experimental data for the temperature field could be improved by accounting for phase change in the models. However, the modelling results are still indicative and show that heat transfer is hindered in the wave region. The local overheat zones underneath the disturbance wave could play key roles activating the nucleation boiling sites. 1 NURESAFE project, and ASCOMP Co. (for use of TransAT code) 1:44PM R11.00004 The Turbulent Wake and Internal Wave Field Due to a Sphere Moving in an Ocean Thermocline , LAURA BRANDT, JAMES ROTTMAN, CECILY TAYLOR, Leidos, DAVE BROUTMAN, Computational Physics Inc — We use ray theory to gain insight into the generation and propagation of internal waves produced by a sphere towed horizontally at constant speed in an idealized ocean thermocline. In particular, we seek to test a previously proposed model of internal wave generation produced by the flow over the body and the disturbances produced by the turbulent wake within the framework of a ray model of the internal waves. This model approximates the stratified flow over the sphere as a combination of a distribution of sources representing the steady internal wave flow, relative to the sphere, and a vertically oscillating sphere, representing the waves generated by the turbulent eddies in the wake. The frequency of oscillation of the sphere is based on laboratory observations of the frequency of shedding of coherent structures in the wake. We formulate a ray model of this flow that incorporates the source distribution as the initial conditions of the internal wave rays produced by the sphere. The results of our simulations are compared with laboratory experiments. The steady wave flow is well represented by this model, in agreement with previous studies. The model of the generation of the unsteady waves, which until now have been untested, requires some additional tuning of the parameters. 1:57PM R11.00005 Thermo-electrohydrodynamic internal waves in annular geometry1 , HARUNORI YOSHIKAWA, Universite Nice Sophia Antipolis, ANTOINE MEYER, OLIVIER CRUMEYROLLE, INNOCENT MUTABAZI, Universite du Havre — An electric field applied to a dielectric fluid with a temperature gradient generates a body force on the fluid, which can be regarded as thermal buoyancy associated with an electric effective gravity. We consider the internal waves due to this thermoelectric force in annular geometry, where the force field is centro-symmetric. The Earth’s gravity is neglected. This configuration is of relevance to large-scale geophysical flows. The dispersion relation of the waves is determined by a spectral method, with or without taking into account the fluid viscosity. The effects of geometry curvature and of a thermoelectric feedback are discussed. The oscillatory instability of the circular Couette flow under the thermoelectric body force and its relation with the waves will also be discussed. 1 Authors acknowledge the financial support from the CNRS under the program PEPS-PTI OndInterGE. 2:10PM R11.00006 Experiments on the interaction between hydrodynamic turbulence and surface waves1 , TIMOTHEE JAMIN, MICHAEL BERHANU, ERIC FALCON, MSC, Universite Paris Diderot, CNRS, UMR 7057 Paris — Different regimes of interaction between hydrodynamic turbulence and a free surface are investigated in a meter scale basin. A homogeneous and isotropic turbulence is generated by an 8x8 array of jets pointing upwards at the bottom of the tank. The 64 jets are driven individually to reach a random spatiotemporal forcing pattern and produce an intense turbulence. Using fluid velocity measurements, we characterize the turbulence obtained with this setup, then we investigate free-surface deformations induced by hydrodynamic turbulence. In a second stage an electromechanical shaker will generate gravity-capillary waves at the free surface. We aim to study reduction or amplification of surface waves and then measure energy exchange between hydrodynamic turbulence and wave turbulence. 1 This work was supported by the DGA-CNRS Ph.D program and ANR Turbulon 12-BS04-0005. 2:23PM R11.00007 Robust energy transfer mechanism via precession resonance in nonlinear turbulent wave systems1 , DAN LUCAS, MIGUEL BUSTAMANTE, Univ Coll Dublin, BRENDA QUINN, Tel-Aviv University — The precise mechanisms by which energy is most efficiently transferred in a turbulent system remain an important open question for the fluid mechanics community. In this talk we present a newly discovered resonance which is found to drive transfers across the spectrum of Fourier modes in a nonlinear wave system. Quadratic nonlinearity results in modes interacting in triads and, by considering the “truly dynamical degrees of freedom” (amplitudes and triad phases) and the precessional frequencies of the triads, we show transfers are maximal when the precession resonates with the nonlinear temporal frequencies. This can lead to a collective state of synchronised triads with intense cascades at intermediate nonlinearity; we find greatest transfer between the traditional weak and strong turbulence regimes and discover that this new mechanism is dominant here. We present the effect in a hierarchy of models including a full DNS of the Charney-Hasegawa-Mima equation and confirm analytical predictions. 1 Supported by Science Foundation Ireland (SFI) under Grant Number 12/IP/1491. 2:36PM R11.00008 Airflow separation events above surface waves , FABRICE VERON, MARC BUCKLEY, Univ of Delaware — Airflow dynamics above waves strongly influence exchanges of heat, momentum and mass between the Ocean and the Atmosphere. We present experimental results on the details of the airflow above surface gravity waves for a several wind speeds, wave ages and slopes. The bulk of the results presented were obtained from a series of laboratory experiments that took place at the University of Delaware’s Air-sea interaction facility. Airflow properties within and above the viscous sublayer were obtained using PIV, and wave profiles and spectra were measured by laser-induced fluorescence. We observe direct evidence of intermittent separation of the viscous sublayer past the crest of the wind waves. Despite the intermittent aspect of this phenomenon, ensemble averages of the wave phase-locked velocity products suggests the airflow separation yield significant flux of vorticity away from the surface thereby generating intense mixing and momentum transport within the airflow. These events, in turn, may affect wave growth and the air-water momentum balance. Our results hold for wind speeds that would normally be considered low to moderate. Implications for models of air-sea momentum flux will be discussed. 2:49PM R11.00009 Transverse instability and viscous dissipation of forced 3-D gravitycapillary solitary waves on deep water1 , YEUNWOO CHO, Korea Advanced Institute of Science and Technology — The shedding phenomena of 3-D viscous gravity-capillary solitary waves generated by a moving air-forcing on the surface of deep water are investigated. Near the resonance where the forcing speed is close to 23 cm/s, two kinds of shedding modes are possible; Anti-symmetric and symmetric modes. A relevant theoretical model equation is numerically solved for the identification of shedding of solitary waves, and is analytically studied in terms of their linear stability to transverse perturbations. Furthermore, by tracing trajectories of shed solitary waves, the decay rate of a 3-D solitary wave due to viscous dissipation is estimated. 1 This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2014R1A1A1002441). 3:02PM R11.00010 Experimental investigation of three-wave interactions of capillary surfacewaves , MICHAEL BERHANU, MSC Universite Paris Diderot, CNRS, UMR 7057 Paris, ANNETTE CAZAUBIEL, Ecole Normale Supérieure, Paris, LUC DEIKE, Scripps Institution of Oceanography, University of California San Diego, TIMOTHEE JAMIN, ERIC FALCON, MSC Universite Paris Diderot, CNRS, UMR 7057 Paris — We report experiments studying the non-linear interaction between two crossing wave-trains of gravity-capillary surface waves generated in a closed laboratory tank. Using a capacitive wave gauge and Diffusive Light Photography method, we detect a third wave of smaller amplitude whose frequency and wavenumber are in agreement with the weakly non-linear triadic resonance interaction mechanism. By performing experiments in stationary and transient regimes and taking into account the viscous dissipation, we estimate directly the growth rate of the resonant mode in comparison with theory. These results confirm at least qualitatively and extend earlier experimental results obtained only for unidirectional wave train. Finally we discuss relevance of three-wave interaction mechanisms in recent experiment studying capillary wave turbulence. 3:15PM R11.00011 Rogue Wave Modes for the Long Wave-Short Wave Resonance and the Derivative Nonlinear Schrödinger Models1 , HIU NING CHAN, KWOK WING CHOW, The University of Hong Kong, DAVID JACOB KEDZIORA, Australian National University, ROGER HAMILTON JAMES GRIMSHAW, Loughborough University, EDWIN DING, Azusa Pacific University — Rogue waves are unexpectedly large displacements of the water surface and will obviously pose threat to maritime activities. Recently, the formation of rogue waves is correlated with the onset of modulation instabilities of plane waves of the system. The long wave-short wave resonance and the derivative nonlinear Schrödinger models are considered. They are relevant in a two-layer fluid and a fourth order perturbation expansion of free surface waves respectively. Analytical solutions of rogue wave modes for the two models are derived by the Hirota bilinear method. Properties and amplitudes of these rogue wave modes are investigated. Conditions for modulation instability of the plane waves are shown to be precisely the requirements for the occurrence of rogue waves. In contrast with the nonlinear Schrödinger equation, rogue wave modes for the derivative nonlinear Schrödinger model exist even if the dispersion and cubic nonlinearity are of the opposite signs, provided that a sufficiently strong self-steepening nonlinearity is present. Extensions to the coupled case (multiple waveguides) will be discussed. 1 This work is partially supported by the Research Grants Council General Research Fund contract HKU 711713E. Tuesday, November 25, 2014 1:05PM - 3:41PM Session R12 Flow Control: Vortices and Turbulence — 3018 - Mohamed Gad-el-Hak, Virginia Commonwealth University 1:05PM R12.00001 Scaling the Response of Separating Turbulent Boundary Layer to Pulsed Excitation , SEIFERT AVRAHAM1 , VITALI PALEI2 , School of Mech. Eng., Tel Aviv University — The talk will start by offering the presenters view of active flow control current status, main challenges and future directions. Then recent experimental results of turbulent separating boundary layer, subjected to pulsed excitation will be presented and discussed. A search for instability mechanism did not result in any disturbances that were amplified. Therefore, pulsed excitation that intermittently enhances the skin friction with optimal time lag was sought. Scaling the response of the excited flow leads to dimensionless optimal magnitude and repetition rates. 1 Prof. and Head, Meadow Aerolab 2 Researcher 1:18PM R12.00002 Ekman and Taylor Vortices’ Destruction and Mixing Enhancement in a Taylor–Couette System , H. OUALLI, M. MEKADEM, A. BENTSABET, M. ABADA, École Militaire Polytechnique, Algiers, Algeria, A. BOUABDALLAH, Université des Sciences et de la Technologie Houari Boumediene, Algiers, Algeria, M. GAD-EL-HAK, Virginia Commonwealth University, Richmond, Virginia, USA — Suppression of Ekman and Taylor vortices is sought in several industrial processes such as cylindrical crystal growth and osmotic/photonic water purification. Last meeting, we investigated experimentally and numerically an active flow control strategy to obliterate vortices in a Taylor–Couette flow. The control consists of effecting minute radial pulsatile motion of the rotating inner cylinder’s cross-section. The results showed that destruction of either type of vortices occurs at different pulsatile frequencies, requiring one order of magnitude higher frequency to obliterate the Ekman type. This problem is revisited with identical parameters and conditions for the controlling strategy but the Taylor–Couette system is now inclined relative to the horizontal direction in such a way that gravitational effects are no longer negligible. It is found that body forces contribute to the complete destruction of Taylor and Ekman vortices, reducing the optimum frequency by more than 50% for even a modest inclination angle of θ = 15◦ . Furthermore, the axial and azimuthal velocity fluctuations are increased by one order of magnitude, thus yielding substantial enhancement in flow mixing. 1:31PM R12.00003 Estimation of the global modes in the wake of a low-aspect-ratio pyramid1 , ZAHRA HOSSEINI, ROBERT J. MARTINUZZI, University of Calgary, Canada, BERND R. NOACK, PPRIME, France — A pressure sensor based estimation technique is proposed to extract the most energetic global modes in the turbulent wake of a wall-mounted square-based pyramid with apex angle of ζ = 60◦ immersed partially in a thin turbulent boundary layer. A modified Extended Proper Orthogonal Decomposition (EPOD) technique is presented which exploits extracting the maximum pressure-velocity correlations for the optimal velocity estimation. The method is assessed based on the planar stereoscopic Particle Image Velocimetry data taken simultaneously with fluctuating pressure at the pyramid surface and the wall. The proposed modifications enable to recover significant dynamics that are otherwise lost in the EPOD estimation and greatly reduce the residual of the estimated coherent kinetic energy. The method will be used to estimate the three-dimensional coherent structures reconstructed from the dominant modes: mainly the fundamental harmonics associated to the periodic shedding and a slow-drift mode capturing the base flow modulations. Such three-dimensional description of the coherent structures helps to understand complex couplings between slow drift and harmonic fluctuations in tapered body wake. 1 This work is supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). 1:44PM R12.00004 The effect of butterfly-scale inspired patterning on leading-edge vortex growth1 , JACOB WILROY, AMY LANG, University of Alabama, REDHA WAHIDI, None — Leading edge vortices (LEVs) are important for generating thrust and lift in flapping flight, and the surface patterning (scales) on butterfly wings is hypothesized to play a role in the vortex formation of the LEV. To simplify this complex flow problem, we designed an experiment to focus on the alteration of 2-D vortex development with a variation in surface patterning. Specifically we are interested in the secondary vorticity generated by the LEV interacting at the patterned surface and how this can affect the growth rate of the circulation in the LEV. For this experiment we used rapid-prototyped longitudinal and transverse square grooves attached to a flat plate and compared the vortex formation as the plate moved vertically. The plate is impulsively started in quiescent water and flow fields at Re = 1500, 3000, and 6000 are examined using Digital Particle Image Velocimetry (DPIV). The vortex formation time is 0.6 and is based on the flat plate travel length and chord length. 1 Support for this research came from NSF REU Grant 1358991 and CBET 1335848. 1:57PM R12.00005 Interaction of a vortex ring and a bubble , NARSING K. JHA, RAGHURAMAN N. GOVARDHAN, Indian Institute of Science — Micro-bubble injection in to boundary layers is one possible method for reducing frictional drag of ships. Although this has been studied for some time, the physical mechanisms responsible for drag reduction using microbubbles in turbulent boundary layers is not yet fully understood. Previous studies suggest that bubble-vortical structure interaction seems to be one of the important physical mechanisms for frictional drag reduction using microbubbles. In the present work, we study a simplification of this problem, namely, the interaction of a single vortical structure, in particular a vortex ring, with a single bubble for better understanding of the physics. The vortex ring is generated using a piston-cylinder arrangement and the bubble is generated by connecting a capillary to an air pump. The bubble dynamics is directly visualized using a high speed camera, while the vorticity modification is measured using time resolved PIV. The results show that significant deformations can occur of both the bubble and the vortex ring. Effect of different non-dimensional parameters on the interaction will be presented in the meeting. 2:10PM R12.00006 Machine learning control (MLC) — a novel method for optimal control of complex nonlinear systems1 , BERND R. NOACK, LAURENT CORDIER, VLADIMIR PAREZANOVIC, KAI VON KRBEK, PPRIME, Poitiers, France, MARC SEGOND, MARKUS W. ABEL, Ambrosys GmbH, Germany, STEVEN BRUNTON, University of Washington, USA, THOMAS DURIEZ, Universidad de Buenos Aires, Argentinia — We propose a model-free closed-loop control strategy for complex nonlinear systems with a finite number of sensors and actuators (MIMO). This strategy yields a feedback law which optimizes a cost functional with machine learning methods. Thus, no dynamical model of the plant is required in contrast to model-based approaches, In addition, no working open-loop control is necessary in contrast to adaptive approaches. The approach is illustrated for strongly nonlinear dynamical systems which are not accessible to linear control design. Control studies of several shear-turbulence experiments will be presented in the talks of T. Duriez and V. Parezanović. 1 Funding of the ANR Chair of Excellence TUCOROM, of the ANR grant SepaCoDe, of the EC’s Marie-Curie ITN program and of Ambrosys GmbH is acknowledged. 2:23PM R12.00007 Closed-loop control of experimental shear flows using MLC1 , THOMAS DURIEZ, Universidad de Buenos Aires, Argentinia, VLADIMIR PAREZANOVIĆ, KAI VON KRBEK, LAURENT CORDIER, BERND R. NOACK, JEAN-PAUL BONNET, PPRIME, Poitiers, France, MARC SEGOND, MARKUS W. ABEL, Ambrosys GmbH, Germany, NICOLAS GAUTIER, JEAN-LUC AIDER, PMMH, ESPCI, France, CÉDRIC RAIBAUDO, CHRISTOPHE CUVIER, MICHEL STANISLAS, LML, Lille, France, ANTOINE DEBIEN, NICOLAS MAZELLIER, AZEDDINE KOURTA, PRISME, Orléans, France, STEVEN BRUNTON, Universty of Washington, USA — We employ a novel closed-loop control strategy for turbulent flows using machine learning methods in a model-free manner (see MLC talk of B. R. Noack). MLC yields in-time control of experimental shear flows and has advantages over the state-of-the-art control. In this talk, MLC is applied to four different experimental closed-loop control setups: (1) the TUCOROM mixing layer tunnel (talk of V. Parezanović), (2) the Görtler PMMH water tunnel with a backward facing step, (3) the LML Boundary-Layer wind tunnel with a separating turbulent boundary layer, and (4) the Malavard wind tunnel with the SepaCoDe ramp. In all cases, MLC finds a control which yields a significantly better performance with respect to the given cost functional as compared to the best previously tested open-loop actuation. 1 Funding of the ANR Chair of Excellence TUCOROM, of the ANR grant SepaCoDe, of the EC’s Marie-Curie ITN program, and of Ambrosys GmbH is acknowledged. 2:36PM R12.00008 Closed-loop control of an experimental mixing layer using MLC1 , VLADIMIR PAREZANOVIĆ, LAURENT CORDIER, BERND R. NOACK, ANDREAS SPOHN, JEAN-PAUL BONNET, PPRIME, Poitiers, France, THOMAS DURIEZ, Universidad de Buenos Aires, Argentinia, MARC SEGOND, MARKUS W. ABEL, Ambrosys GmbH, Germany, STEVEN BRUNTON, University of Washington, USA — A novel framework for closed-loop control of turbulent flows is tested for an experimental mixing layer flow. This framework, called Machine Learning Control (MLC), provides a model-free method of searching for the best control law (see talk of B. R. Noack). Here, MLC is benchmarked against classical open-loop actuation of the mixing layer. Results show that this method is capable of producing sensor-based control laws which can rival or surpass the best open-loop forcing, and be robust to changing flow conditions. Additionally, MLC can detect non-linear mechanisms present in the controlled plant, and exploit them to find a better type of actuation than the best periodic forcing. Other experimental shear-flow control studies with MLC will be presented in a talk by T. Duriez. 1 Funding of the ANR Chair of Excellence TUCOROM, of the ANR grant SepaCoDe, of the EC’s Marie-Curie ITN program, and of Ambrosys GmbH is acknowledged. 2:49PM R12.00009 Symmetries, multistability and stochastic dynamics of turbulent wakes , GEORGIOS RIGAS, AIMEE MORGANS, JONATHAN MORRISON, Imperial College — The dynamics of a turbulent wake generated behind a bluff threedimensional axisymmetric body are investigated experimentally at a diameter based Reynolds number ∼ 2 × 105 . Proper orthogonal decomposition of base pressure measurements indicates that the most energetic coherent structures retain the structure of the symmetry-breaking laminar instabilities and manifest as unsteady vortex shedding with azimuthal wavenumber m = ±1. In a rotating reference frame, the wake preserves the reflectional symmetry, as observed in the laminar and transitional regimes. Due to a slow and random rotation of the symmetry plane around the axis of the body, the turbulent wake explores an infinite number of metastable states and statistical axisymmetry is recovered in the time average. A simple dynamical model, where the deterministic part describes the broken symmetries of the flow and the stochastic part accounts for the incoherent fluctuations, shows excellent agreement with the experimental results for the spatiotemporal evolution of the turbulent wake. Finally, we show how these models can be obtained directly from the governing Navier-Stokes equations. 3:02PM R12.00010 Subcritical Transition in Channel Flows , JOSEPH MAESTRI, PHILIP HALL, Imperial College London — Exact-coherent structures, or colloquially non-linear solutions to the Navier-Stokes equations, have been the subject of great interest over the past decade due to their relevance in understanding the process of transition to turbulence in shear flows. Over the past few years the relationship between high Reynolds number vortex-wave interaction theory and such states has been elucidated in a number of papers and has provided a solid asymptotic framework to understand the so-called self-sustaining process that maintains such structures. In this talk, we will discuss this relationship before talking about recent work on solving the vortex-wave interaction equations using numerical techniques in order to propose laminar-flow control techniques. 3:15PM R12.00011 Vortex-Wave interaction in plane Poiseuille flow , LIAM DEMPSEY, ANDY WALTON, PHILIP HALL, Imperial College London — Our main interest is in the process of transition to turbulence at high Reynolds numbers in the flow in a plane channel. We will consider the basic flow to be driven by a uniform streamwise pressure gradient (plane Poiseuille flow). We will formulate a high Reynolds number asymptotic structure in the form of a nonlinear vortex-Tollmien-Schlichting-wave interaction (VWI)- or so-called self sustaining process. We look for starting solutions for the VWI by performing a weakly nonlinear analysis close to the lower branch neutral point and seek equilibrium solutions of the resulting nonlinear amplitude equation. We provide numerical solutions of the interaction equations which are localised in the spanwise direction at large enough amplitudes. 3:28PM R12.00012 Linear Mechanisms and Pressure Fluctuations in Wall Turbulence , KAMTHON SEPTHAM, JONATHAN MORRISON, Department of Aeronautics, Imperial College, London, SW7 2AZ, UK — Full-domain, linear feedback control of turbulent channel flow at Reτ ≤ 400 via vU ′ at low wavenumbers is an effective method to attenuate turbulent channel flow such that it is relaminarised. The passivitybased control approach is adopted and explained by the conservative characteristics of the nonlinear terms contributing to the Reynolds-Orr equation (Sharma et al. P hys. F luids 2011). The linear forcing acts on the wall-normal velocity field and thus the pressure field via the linear (rapid) source term of the Poisson ∂v equation for pressure fluctuations, 2U ′ ∂x . The minimum required spanwise wavelength resolution without losing control is constant at λ+ z = 125, based on the wall friction velocity at t = 0. The result shows that the maximum forcing is located at y + ≈ 20, corresponding to the location of the maximum in the mean-square pressure gradient. The effectiveness of linear control is qualitatively explained by Landahl’s theory for timescales, in that the control proceeds via the shear interaction timescale which is much shorter than both the nonlinear and viscous timescales. The response of the rapid (linear) and slow (nonlinear) pressure fluctuations to the linear control is examined and discussed. Tuesday, November 25, 2014 1:05PM - 3:28PM Session R13 Biofluids: Membranes, Vesicles and Micelles — 3020 - Petia Vlahovska, Brown University 1:05PM R13.00001 Lipid Bilayer Vesicle Dynamics in AC Electric Fields , LANE MCCONNELL, University of New Mexico, PETIA VLAHOVSKA, Brown University, MICHAEL MIKSIS, Northwestern University — Vesicles are closed, fluid-filled lipid bilayers which are mechanically similar to biological cells and which undergo shape transitions in the presence of electric fields. Here we model the vesicle membrane as an infinitely thin, capacitive, area-incompressible interface with the surrounding fluids acting as charge-advecting leaky dielectrics. We then implement the boundary integral method to numerically investigate the dynamics of a vesicle in various AC electric field profiles. Our numerical results are then compared with recent small deformation theory and experimental data. We also note our observation of a new theoretical vesicle behavior that has yet to be observed experimentally. 1:18PM R13.00002 Ellipsoidal Relaxation of Electrodeformed Vesicles , MIAO YU, Rutgers, the State University of New Jersey, RAFAEL LIRA, Max Planck Institute of Colloids and Interfaces, KARIN RISKE, Univ Fed Sao Paulo, RUMIANA DIMOVA, Max Planck Institute of Colloids and Interfaces, HAO LIN, Rutgers, the State University of New Jersey — Theoretical analysis and experimental quantification on the ellipsoidal relaxation of electrodeformed vesicles are presented. A closed-form solution is derived which predicts the aspect ratio as a function of time. Analysis of the solution and experimental data reveals good agreement, and two distinguishable regimes are identified. The “entropic” regime is dictated by the Helfrich constitutive relation, and in the “constant tension” regime the aspect ratio exhibits an exponential decay. Both the bending rigidity and initial membrane tension are accurately extracted. The relaxation of electroporated vesicles is also briefly discussed. This analytical approach provides a simple and powerful tool to query the mechanics of lipid membranes and similar soft materials. 1:31PM R13.00003 Thermal undulations of biomimetic bilayer membranes in external fields1 , NICO FRICKE, PETIA VLAHOVSKA, Brown University — We study the influence of an applied electric field on the physical properties of fluid bilayer membranes. Global and regional analyses of the shape fluctuations of a giant quasi-spherical vesicle are used to determine membrane tension, bending rigidity, and shear viscosity from a time series of video- microscopy images. The parameters of the uniform electric field (frequency and amplitude) are chosen such that there is no global ellipsoidal vesicle deformation, and hence any renormalization of the tension and bending rigidity arise only from electric stress in the membrane. Using this approach we examine the effect of the electrotension on the main phase transition temperature of lipid membranes, where we observe that increasing field strength decreases, albeit slightly (about 0.1K), the melting temperature. 1 supported by NSF-CMMI- 1232477 1:44PM R13.00004 Experimental Methods to Observe Asymmetric Instability of IntermediateReduced-Volume Vesicles in Extensional Flow , JOANNA DAHL, University of California, Berkeley, VIVEK NARSIMHAN, Stanford University, BERNARDO GOUVEIA, SANJAY KUMAR, University of California, Berkeley, ERIC SHAQFEH, Stanford University, SUSAN MULLER, University of California, Berkeley — Vesicles provide an attractive model system to understand the deformation of living cells in response to mechanical forces. These enclosed lipid bilayer membranes are suitable for complementary theoretical and experimental analysis. A recent study (Narsimhan et al., J. Fluid Mech. 750: 144-190, 2014) predicted that intermediate-aspect-ratio vesicles break up asymmetrically in extensional flow. Upon infinitesimal perturbation to its shape, the vesicle stretches into an asymmetric dumbbell. In this work, we present preliminary results from cross-slot microfluidic experiments observing this instability. The onset of breakup depends on two non-dimensional parameters: reduced volume (vesicle asphericity) and capillary number (ratio of viscous to bending forces). We will present strategies for accurately measuring these quantities in order to plot a stability diagram. Specifically, we will describe our synthesis of floppy, intermediate-reduced-volume vesicles and our measurement of their bending moduli by analyzing membrane thermal fluctuations. We will discuss coupling particle-image velocimetry (PIV) with cross-slot trapping of vesicles to ensure that breakup occurs at the stagnation point. A preliminary phase diagram for asymmetric breakup will be reported. 1:57PM R13.00005 Tethering and pearling of vesicles in flow , MARC LEONETTI, GWENN BOEDEC, Univ AixMarseille, MARC JAEGER, Centrale Marseille, IRPHE, SOFT MATTER TEAM, M2P2, SOFT MATTER TEAM — Vesicles are studied by their own for their extensive use as biomedical vehicles for drug delivery but, they are also considered as red blood cell model. Indeed, vesicles are drops bounded by a phospholipidic membrane, an interface with peculiar mechanical properties. The membrane is a 2D incompressible fluid with a resistance to bending. The surface is preserved and there is no shear resistance as in capsules. The dynamics in flow depends mainly on their geometric ability to deform measured by the excess area or the reduced volume and on the hydrodynamical forcing such as settling and extensional flows for example characterized notably by the capillary number., we study the stability of thin phospholipidic tubes in flow. Notably, we show that pearls appear along stretched tethers, stretching which appears in vesicles during settling and in elongational flow for example. Numerical investigations of such vesicles under stress support the theoretical analysis. [1] G. Boedec, M. Jaeger and M. Leonetti, Sedimentation-induced tether on a settling vesicle, Phys. Rev. E 88 (2013) 0107022. [2] G. Boedec, M. Jaeger and M. Leonetti, Pearling instability of a cylindrical vesicle, J. Fluid Mech. 743 (2014) 262-279. 2:10PM R13.00006 Nonlinear deformations of microcapsules in elongation flow , JULIEN DESCHAMPS, CLÉMENT DE LOUBENS, GWENN BOEDEC, MARC GEORGELIN, MARC LEONETTI, Univ Aix-Marseille, SOFT MATTER AND BIOPHYSICS GROUP TEAM — Soft microcapsules are drops bounded by a thin elastic shell made of cross-linked proteins. They have numerous applications for drug delivery in bioengineering, pharmaceutics and medicine, where their mechanical stability and their dynamics under flow are crucial. They can also be used as red blood cells models. Here, we investigate the mechanical behaviour of microcapsules made of albumine in strong elongational flow, up to a stretching of 180% just before breaking. The set-up allows us to visualize the deformed shape in the two perpendicular main fields of view, to manage high capillary number and to manipulate soft microcapsules. The steady-state shape of a capsule in the planar elongational flow is non-axisymmetric. In each cross section, the shape is an ellipse but with different small axis which vary in opposite sense with the stretching. Whatever the degree of cross-linking and the size of the capsules, the deformations followed the same master-curve. Comparisons between numerical predictions and experimental results permit to conclude unambiguously that the more properly strain-energy model of membrane is the generalized Hooke model. 2:23PM R13.00007 Effect of bending on the dynamics and wrinkle formation for a capsule in shear flow , ANNE-VIRGINIE SALSAC, Univ Technol Compiegne, CLAIRE DUPONT, Univ Technol Compiegne and Ecole Polytechnique, DOMINIQUE BARTHES-BIESEL, Univ Technol Compiegne, MARINA VIDRASCU, UPMC, PATRICK LE TALLEC, Ecole Polytechnique — When microcapsules are subjected to an external flow, the droplets enclosed within a thin hyperelastic wall undergo large deformations, which often lead to buckling of the thin capsule wall. The objective is to study numerically an initially spherical capsule in shear flow and analyze the influence of the membrane bending rigidity on the capsule dynamics and wrinkle formation. The 3D fluid-structure interactions are modeled coupling a boundary integral method to solve for the internal and external Stokes flows with a thin shell finite element method to solve for the wall deformation. Hyperelastic constitutive laws are implemented to model the deformation of the capsule mid-surface and the generalized Hooke’s law for the bending effects. We show that the capsule global motion and deformation are mainly governed by in-plane membrane tensions and are marginally influenced by the bending stiffness Ks. The bending stiffness, however, plays a role locally in regions of compressive tensions. The wrinkle wavelength depends on Ks following a power law, which provides an experimental technique to determine the value of Ks through inverse analysis. 2:36PM R13.00008 Swinging of two-domains vesicles in shear flow , ANNIE VIALLAT, CNRS, SIMON TUSCH, KAMEL KHELLOUFI, Aix Marseille University, MARC LEONETTI, CNRS — Giant lipid vesicles and red blood cells in shear flow at low shear rates tank tread (TT) at small viscosity ratio between the inner particle volume and the external fluid, and flip or tumble (T) at large viscosity ratio. The phase diagram of motion of red blood cells is however much more complex. Swinging superimposes to TT, cells wobble and roll rather than tumble with increasing shear rate and present a shear-rate driven transition between TT to T. These features are attributed to the shear elasticity and the non spherical stress-free shape of the cell membrane, which stores shear elastic energy as a function of the relative position of its elements. We have created vesicles with a phase diagram of motion comparable to that of red blood cells by preparing membranes with two lipids and cholesterol. These membranes present two domains separated by a contact line. The line has a tension energy that depends on its relative position on the vesicle. Similarly to red blood cells, two-domains vesicles swing and wobble. An analytical model where line tension energy is added to the Keller and Skalak’s model fits our experimental data without any adjustable parameter. Our experiments and model shed light on the motion of deformable particles in shear flow. 2:49PM R13.00009 Three-dimensional numerical simulation of red blood cell motion in Poiseuille flows1 , LINGLING SHI, TSORNG-WHAY PAN, ROLAND GLOWINSKI, Dept. of Mathematics, University of Houston — An immersed boundary method based on a finite element method has been successfully combined with an elastic spring network model for simulating the dynamical behavior of a red blood cell (RBC) in Poiseuille flows. This elastic spring network preserves the biconcave shape of the RBC in the sense that after the removal of the body force for driving the Poiseuille flow, a RBC with its typical parachute shape in a tube does restore its biconcave resting shape. As a benchmark test, the relationship between the deformation index and the capillary number of the RBCs flowing through a narrow cylindrical tube has been validated. For the migration properties of a single cell in a slit Poiseuille flow, a slipper shape accompanied by a cell membrane tank-treading motion is obtained for Re ≥ 0.03 and the cell mass center is away from the center line of the channel due to its asymmetric slipper shape. For the lower Re ≤ 0.0137, a RBC with almost undeformed biconcave shape has a tumbling motion. A transition from tumbling to tank-treading happens at the Reynolds number between 0.0137 and 0.03. In slit Poiseuille flow, RBC can also exhibit a rolling motion like a wheel during the migration. The lower Reynolds number is, the longer the rolling motion lasts. 1 This work is supported by an NSF Grant No. DMS-0914788. 3:02PM R13.00010 Mesoscopic Modeling of Thrombus Formation and Growth: Platelet Deposition in Complex Geometries , ALIREZA YAZDANI, GEORGE KARNIADAKIS, Brown University — Haemodynamics and blood rheology are important contributing factors to thrombus formation at a vulnerable vessel wall, and adhesion of platelets to a vascular surface, particularly in regions of flow stagnation, recirculation and reattachment is significantly important in formation of thrombi. For example, haemodynamic micro-environment can have effects on thrombosis inside the atherosclerotic plaques and aneurysms. To study these effects, we have developed and validated a model for platelet aggregation in blood flow using Dissipative Particle Dynamics (DPD) method. In this model platelets are considered as single DPD particles interacting with each other via Morse potential once activated. We assign an activation delay time to each platelet such that they remain passive during that time. We investigate the effect of different geometries on platelet aggregation by considering arterial stenosis at different levels of occlusion, and aneurysms of different shapes and sizes. The results show a marked increase in platelet aggregation within the boundaries of deceleration zone by increasing the degree of stenosis. Further, we observe enhanced platelet margination and wall deposition in the presence of red blood cells. 3:15PM R13.00011 Topology, Energetics and Rheology of Surfactant Micelles1 , RADHAKRISHNA SURESHKUMAR, SUBAS DHAKAL, ABHINANDAN SAMBASIVAM, Syracuse University — A rich variety of self-assembled structures of amphiphilic molecules, ranging from spherical and cylindrical shapes to topologically complex networks consisting of branches and loops, is unraveled through large scale Molecular Dynamic simulations that account for explicit solvent, electrostatic and hydrodynamic interactions. The simulations employ a coarse grained force field, benchmarked against atomistic simulations (Sangwai and Sureshkumar, Langmuir, 27, 6628 (2011); 28, 1127 (2012)), to describe inter-molecular forces. Analysis of these structures allows for the first time to directly determine certain fundamental length scales, e.g. persistence and contour lengths, mesh size, as well as the end cap energy, which dictate the rheological properties and flow phenomena in micellar fluids. The much debated anomalous viscosity variations with respect to salt concentration can be understood based on the underlying morphological changes (http://arxiv.org/abs/1407.5086). This, and the effect of nanoparticle addition to the network structure and flow properties, will be discussed. 1 NSF Grants 1049454, 1049489; NSF-supported Extreme Science and Engineering Discovery Environment (XSEDE) for computational resources Tuesday, November 25, 2014 1:05PM - 3:41PM Session R14 Drops: Wetting and Spreading II — 3009/3011 - Omar K. Matar, Imperial College, London 1:05PM R14.00001 Laws of spreading: why Tanner, Hoffman, Voinov, Cox and de Gennes were wrong, generally speaking , PIROUZ KAVEHPOUR, ALIREZA MOHAMMADKARIM, University of California, Los Angeles — For nearly 50 years, most of the researchers in the area of wetting and spreading have used a relationship between the dynamics contact angle and velocity, θ 3 − θ03 ∼ U , where θ is dynamics contact angle, θ0 is the equilibrium