Lab on a Chip

PAPER

Motorized actuation system to perform droplet operations on

printed plastic sheets†
Cite this: Lab Chip, 2016, 16, 1861
Taejoon Kong,‡ Riley Brien,‡ Zach Njus, Upender Kalwa and Santosh Pandey*
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

We developed an open microfluidic system to dispense and manipulate discrete droplets on planar plastic sheets. Here, a
superhydrophobic material is spray-coated on commercially-available plastic sheets followed by the printing of
hydrophilic symbols using an inkjet printer. The patterned plastic sheets are taped to a two-axis tilting platform, powered by
stepper motors, that provides mechanical agitation for droplet transport. We demonstrate the following droplet operations:
transport of droplets of different sizes, parallel transport of multiple droplets, merging and mixing of multiple droplets, dispensing
of smaller drop- lets from a large droplet or a fluid reservoir, and one-directional transport of droplets. As a proof-of-
concept, a colorimetric assay is implemented to measure the glucose concentration in sheep serum. Com- pared to silicon-based
Received 5th February 2016, Accepted digital microfluidic devices, we believe that the presented system is appealing for various biological experiments because of
7th April 2016 the ease of altering design layouts of hydrophilic symbols, rela- tively faster turnaround time in printing plastic sheets, larger area
to accommodate more tests, and lower operational costs by using off-the-shelf products.
DOI: 10.1039/c6lc00176a

www.rsc.org/loc

Introduction operations are often conceptualized from test tube

experi- ments performed in a wet chemistry laboratory, and
Generally speaking, microfluidic platforms consist of the se- quence of operations can be easily altered
closed- channel networks where liquid flow is controlled by depending on the actual experiment being performed.
mechani- cal, pneumatic or electrokinetic means. Today, with The general strategy of producing and actuating discrete
emphasis on higher experimental throughput, microfluidic droplets on open surfaces relies on methods to modulate
platforms incorporate several on-chip components (e.g. the surface tension between the liquid droplet and the
microvalves, micropumps, and microelectrodes) that solid sur- face it rests on. The current literature on this
increase the complex- ity in fabricating the different layers, topic can be grouped into two categories – methods that
integrating the micro- and macroscale components, and employ electrical fields to modulate the wettability of
controlling the individual sensing or actuation parts.1,2 In droplets3–6 and non- electrical methods that employ
contrast to closed-channel microfluidics, open microfluidic mechanical, magnetic, acous- tic or gravitational forces to
platforms obviate the use of polymeric channels and generate directional movement of droplets.7–15
continuous liquid flow; thereby relaxing the fabrication The electrical or ‘electrowetting-on-dielectric’ method
process, easing the system integration to fewer components, of droplet actuation has gained popularity in the last
and promising a cheaper alternative to robotic micro-handling decade primarily because of the ease of programmability
systems.3,4 In open microfluidics, liq- uid is dispensed from and porta- bility.16,17 Here, the conductive liquid droplet
a reservoir as discretized droplets and transported to sits on pat- terned electrodes coated with a hydrophobic
desired locations for further manipulation. Typical dielectric layer. An electric field applied to the target
operations to be performed with discrete droplets may electrode increases the contact angle of the droplet placed
include transport of a single or multiple droplets, merg- over it, and thus alters the wettability of the liquid surface
ing and mixing of two droplets, incubation and affinity to the solid surface. This electrowetting phenomenon can
bind- ing within droplets, extraction of solid particles from be scaled up to move and control multiple droplets over an
the liq- uid phase, and removal of waste droplets.3,5 These array of electrodes, thereby performing any desired
droplet sequence of operations including transport, merging,
mixing, splitting, and dispensing. Analo- gous to digital
microelectronics where pockets of electrons are
Department of Electrical and Computer Engineering, Iowa State University, 1050
Coover Hall, Ames, IA 50011, USA. E-mail: pandey@iastate.edu transferred between devices (e.g. in charged coupled de-
† Electronic supplementary information (ESI) available: Supplementary figures vices), several groups have realized electrowetting-based ‘digi-
and videos of droplet manipulation included. See DOI: 10.1039/c6lc00176a tal microfluidic platforms’ having electrodes of precisely-
‡ Joint first authors.

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1
 View Article

Pap Lab on a

controlled geometry, on-chip control electronics to PDMS membrane has been used to activate and move
energize individual electrodes, and software programs to droplets in two dimensions on a superhydrophobic
automate the droplet operations.3,18,19
Even though the electrowetting method is widely accepted
as the gold standard for droplet handling systems, it is
re- strained by the need for high electrical voltages (in the
range of 100 volts to 400 volts) that have unknown
effects on the biomolecules or cells within droplets.18–20
For instance, the electric actuation force can interfere
with the adsorption of biomolecules on a surface.21
Furthermore, droplet actuation is dependent on the
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

conductivity of the droplet and the di- electric properties

of the insulating layers (e.g. Teflon and Parylene) that are
expensive for large-scale deposition. Be- cause each
electrode is electrically addressed, there are only a finite
number of electrodes that can be addressed on a dig- ital
microfluidics platform.22 To get around this last issue, it
has been shown that the electrodes can be optically
stimu- lated (and thereby producing on-demand optical
intercon- nects) by incorporating photoconductive and high
dielectric constant layers underneath the Teflon coating. 8,23
Active ma- trix arrays of thin film transistor (TFTs) have
also been dem- onstrated as an alternate digital
microfluidic testbed where many thousand individually
addressable electrodes could sense, monitor, and
manipulate droplets.22 Similarly, electrodes can be
selectively energized to reposition water volumes in an
otherwise liquid paraffin medium to create
reconfigurable, continuous-flow microfluidic channels. 24 As
these innovations in digital microfluidics technology extend
the functionalities to newer arenas of portable
diagnostics, much of the fabrication protocol still
requires access of industrial-grade microelectronics foundry
and is thus limited to select users.
To eliminate some of the limitations of electrowetting
mentioned above, non-electrical methods of droplet actuation
have been pursued.9,11–15 In the ‘textured ratchet’
method, movement of liquid droplets is achieved on
textured micro- structures (i.e. ratchets) fabricated in
silicon or elastomeric substrates.15 The textured ratchets are
placed on a level stage that is vertically vibrated using a
linear motor. At the reso- nant frequency of vertical
oscillations, the liquid droplet is able to advance or recede
on the textured ratchets. The move- ment of different
droplets can be individually controlled, both in linear and
closed tracks, by manipulating the volume and viscosity of
droplets. In the ‘superhydrophobic tracks’ method,
shallow grooves are cut in zinc plates or silicon sub- strates. 14
This is followed by a superhydrophobic coating step by
depositing silver and fluorinated thiol surfactant on metal
plates or a fluoropolymer on silicon substrates. The produced
superhydrophobic tracks are able to confine liquid
droplets and guide their movement in trajectories
defined by the tracks. In the ‘surface acoustic waves
(SAW)’ method, a high frequency source connected to
interdigitated gold electrodes generates acoustic waves that
is able to transport fluid drop- lets on a piezoelectric
substrate.25 Recently, pneumatic suc- tion through a

2 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap
surface without any interference from an external energy cm × 0.5 cm); vertical column (1 cm ×
(e.g. heat, light, electricity).21
While the above non-electrical methods demonstrate
that mechanical machining the substrate can passively
move droplets, more results are needed to match the
level of drop- let handling operations achieved in digital
microfluidic plat- forms.3 To gauge the maturity of digital
microfluidics, an ex- citing example is a multi-
functional digital microfluidic cartridge by Advanced
Liquid Logic that can perform multiplexed real-time PCR,
immunoassays and sample prepa- ration.26 A group at
Sandia National Laboratories has devel- oped a digital
microfluidic distribution hub for next genera- tion
sequencing that is capable of executing sample
preparation protocols and quantitative capillary
electrophore- sis for size-based quality control of the DNA
library.27 With growing demand of lab on chip systems in
medicine, digital microfluidics has been used to extract
DNA from whole blood samples,28 quantify the levels of
steroid hormones from breast tissue homogenates, 29 and
screen for metabolic disor- ders and lysosomal storage
diseases from newborn dried blood spots.30–34 These
examples highlight the fact that digi- tal microfluidics is
revolutionizing the field of portable medi- cal diagnostics,
and any rival technology needs to achieve the basic
standards of droplet handling set by digital
microfluidics.
In an attempt to emulate the droplet operations
performed in digital microfluidics without the use of
high electrical voltages or micromachining steps, we present
a sys- tem where droplets are manipulated on a
superhydrophobic surface (created on plastic sheets) by
gravitational forces and mechanical agitation. The
superhydrophobic plastic sheets are further printed with
unique symbols using a hydrophilic ink. A
microcontroller controls the direction and timing of two
stepper motors which, in turn, provide mechanical agita-
tion for droplet transport. Droplets remain confined to
the hydrophilic symbols, and are able to ‘hop’ to
neighbouring symbols by gravity when the surface is
agitated and tilted to a certain degree. Using this basic
principle, we illustrate the following droplet operations:
transport of single and multiple droplets, transport of
larger-volume droplets, merging and mixing of multiple
droplets, dispensing of fixed-volume drop- lets from a large
droplet or liquid reservoir, and one- directional
movement of droplets. As a proof-of-concept, we show
the application of the system as a colorimetric assay to
detect the concentration of glucose in sheep serum.

Experimental
Design of the droplet actuation system
The motorized actuation system consists of a two-axis tilting
platform to manipulate movement of discrete liquid
droplets on hydrophilic symbols printed on a
superhydrophobic sur- face. Fig. 1a shows the system
configuration, including the three structural components:
base, vertical column, and up- per stage. The dimensions
of these components are as fol- lows: base (20 cm × 20
This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 3
 View Article

Lab on a Pap

Fig. 1 The droplet actuation system. (a) The system comprises three structural components: base, column, and stage with plastic sheet. The base is physically screwed to the
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

column. A universal joint connects the column to the stage. A microcontroller interfaces with two stepper motors (attached with individual timing belts) and controls the
mechanical tilting of the stage. The plastic sheet is taped on the top of the stage. Scale bar
= 2 cm. (b) A plastic sheet is spray-coated with a superhydrophobic chemical and printed with hydrophilic symbols using an inkjet printer. The im- age shows discrete
droplets, each coloured with food dyes for visual illustration, resting on the hydrophilic symbols. Scale bar = 2 mm.

1 cm × 10 cm); upper stage (9 cm × 9 cm × 1.3 cm). The Thereafter,

en- tire three-dimensional structure is designed in
AutoCAD (Autodesk™) and the separate components are
machined in acrylic glass (Plexiglas™). The stage is
connected to the col- umn by a universal joint that enables
two-axis rotation about a central pivot. Two stepper motors
(NEMA-17™, 200 steps per revolution, 12 volts, 350
milliamperes, bipolar mode) are connected with individual
timing belts to the stage and mounted to the base. Each
stepper motor controls one axis of rotation of the stage
through an Arduino microcontroller (Adafruit
Industries™). Single commands to tilt the stage up or
down, left or right, and any sequence of such commands
are programmed in a computer workstation and transmitted
through a universal serial bus (USB) connection to the
Arduino microcontroller. A graphical user interface (GUI)
is designed for remote access to the droplet actuation
system using a standard computer workstation (see ESI† Fig.
S1). For image recording and characterization of droplet
operations, a webcam (Logitech C920™) is positioned
above the stage to monitor and record the simultaneous
movement of multiple droplets.

Preparation of plastic sheets

After assembling the structural components of the droplet ac-
tuation system, we prepare the surface of plastic sheets
that will serve as an open microfluidic arena to hold
and move discrete droplets (Fig. 1b). Initially, letter-sized
transparency films (Staples Inc.™) are rinsed with
distilled water and spray-coated with a commercially
available superhydrophobic coating (Rust-Oleum NeverWet™).
The coating procedure is a two-step process that involves
depositing a base coat and a top coat provided by the
supplier. The base coat is applied by spraying on the surface
of the transparency film. Three appli- cations of the base
coat are performed with a wait time of two minutes
between successive applications. After drying for one hour,
four applications of the top coat are performed in a
similar fashion. The superhydrophobically-coated plastic
sheet is dried for 12 hours at room temperature.

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1863
 View Article

Pap Lab on a
hydrophilic symbols are printed on the plastic sheet by
ink- jet printing. For this step, the plastic sheet is loaded
into the document feeder of a commercial ink-jet printer
(Epson WF- 2540™). The layout of the desired symbols
are drawn in Adobe Illustrator, saved on the computer,
and printed using a black ink cartridge (Epson
T200120™). After printing, the plastic sheet is dried for
12 hours at room temperature. Using the above
procedure, a single letter-sized transparency film can
produce six printed templates (9 cm × 9 cm) in one run.

Remote control and GUI software

A graphical user interface (GUI) software is developed
in Matlab to remotely access and control the mechanical
move- ment of the droplet actuation system. The
Adafruit Motor Shield v1 communicates with the
Arduino microcontroller through the I2C (Inter IC)
protocol and controls each of the stepper motors. The
Arduino is further controlled from a computer
workstation using the Arduino Integrated Develop- ment
Environment™. The GUI enables commands to be eas- ily
sent to the Arduino microcontroller. The script accepts in-
puts to set the speed and number of steps taken by the
motors, which, in turn, controls the angular movement of the
stage about the central pivot. The GUI has options to
control motor parameters, such as the number of steps,
speed of ro- tation, and direction of rotation which
eventually control the angular movement of the stage about
the central pivot. In the default state, the position of the
stage is assumed horizontal and is calibrated using a
bubble level (Camco Manufacturing Inc.™). When the GUI
software is first run, the connection to the Arduino
microcontroller is established automatically by searching
active COM ports. Once the Arduino COM port is
confirmed to be connected, the user can enter the
sequence of mechanical operations to be performed. In
the GUI win- dow, pressing the double arrows increases
the stage's angle of rotation in the corresponding
direction (see ESI† Fig. S1). The single arrow button
rapidly tilts the stage to a specified angle, and then
returns it to the default horizontal position.

1864 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap

In addition, the GUI software communicates with a

slides down the superhydrophobic surface and rests on
webcam to display a live preview of the top surface and
the neighbouring hydrophilic symbol (right-side). In Fig.
record images or videos of droplet actuation.
2b, side-view images of a single droplet are shown as it
slides from the left symbol to the right one. The time for
Chemicals
trans- porting a single 10 μL droplet between two
Glucose assay kit (Sigma-Aldrich, GAGO20) is composed of consecutive sym- bols is approximately 100 milliseconds.
the following chemicals: glucose oxidase/peroxidase (Sigma- The stage is tilted at 100 revolutions per minute (r.p.m.)
Aldrich, G3660), and o-dianisidine reagent (Sigma- and the number of steps is 14.
Aldrich, D2679). Glucose standard (Sigma-Aldrich, G6918) The basic principle of droplet transport thus relies on
and sheep serum (Sigma-Aldrich, S3772) are also used. The po- sitioning a droplet on a hydrophilic symbol and
glucose oxi- dase/peroxidase reagent is dissolved in 39.2 providing a rapid tilting action (i.e. tilting the stage
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

ml of deionized water. Next, o-dianisidine reagent is added clockwise (or anti- clockwise) to a specific angle followed
in 1 mL of deion- ized water. The assay reagent is prepared by tilting the stage anti-clockwise (or clockwise) to the
by adding 0.8 mL of the o-dianisidine solution to the 39.2 mL horizontal position). The rapid tilting action allows us to use
of the glucose ox- idase/peroxidase solution and mixing small tilting angles (3–5°) with acceleration and deceleration
the solution thor- of a droplet. Alternatively, a single droplet can be transported
oughly. The glucose standard solution is diluted to create 0.7 by slowly tilting the stage in one direction which, however,
mg mL−1, 0.6 mg mL−1, 0.5 mg mL−1, 0.4 mg mL−1, 0.3 requires a larger tilting an- gle (9–20°) and provides no
mg mL−1, 0.2 mg mL−1, and 0.1 mg mL−1 standards in control on stopping the acceler- ated droplet.
deionized water. For control experiments, deionized We found that droplet transport can be controlled by a se-
water and black food dye (ACH Food Companies Inc.) are ries of hydrophilic symbols printed at regular intervals. Based
used. on initial tests, we chose to use ‘plus (+)’ symbols to
demon- strate single droplet transport. Other symmetric
Result and discussion symbols can also be used for this purpose. We printed plus
Transport of a single droplet symbols of dif- ferent line widths and inter-symbol
spacings (see ESI† Fig. S2a). The transport of single
Fig. 2a shows the side-view of a single droplet placed on a hy-
droplets on the different sym- bols is recorded, and an
drophilic symbol (left-side) printed on a superhydrophobic
average displacement error is mea- sured in each case.
layer. As the stage is tilted clockwise, the droplet remains
Negative displacement error occurs when a droplet fails to
on the hydrophilic symbol. But, as the stage is quickly
detach from the initial symbol. Conversely, positive
tilted anti-clockwise to the default horizontal position, the
displacement error occurs when the droplet travels
droplet
beyond the neighbouring symbol (see ESI† Fig. S2c). In
all cases, the droplet volume is 10 μL, tilting speed is 100
r.p.m., and number of steps is 14. The results indicate that
symbols with thicker line widths produce negative
displacement error as they have more surface area to hold the
droplet in its origi- nal position (see ESI† Fig. S2b). On the
other hand, symbols with thinner line widths produce
positive displacement error as they have insufficient
surface area to hold or capture a sliding droplet. The
optimal line width is 0.02 cm and the inter-symbol
spacing is 0.335 cm, which produces a negligi- ble
displacement error of 0.005 cm. We also found that,
using this optimal dimension of the plus symbol, we
can transport single droplets having a minimum and
maximum water volume of 8 μL and 38 μL, respectively.

Physical model for droplet detachment from a hydrophilic

symbol
Following the force balance analysis of Extrand and Gent, 35
Fig. 2 Transport mechanism of a single droplet. (a) In the cartoon, a droplet is
initially positioned on the left hydrophilic symbol printed on the superhydrophobic we assume the contact region of a liquid droplet on the
surface of a plastic sheet. The stage is tilted clockwise and then anti-clockwise to superhydrophobic surface is circular with a radius R.
return to its default horizontal position (depicted by red block arrows). This rapid The droplet is about to detach from the hydrophilic
tilting action en- ables the droplet to move to the right hydrophilic symbol. (b)
symbol and travel downwards as the stage is tilted from its
Time- lapsed images of an actual droplet show how the droplet is trans- ported
from the left symbol to the right symbol by the tilting action of the stage. The horizontal po- sition to a critical angle α (see ESI† Fig.
vertical dotted lines represents the starting position of the droplet. Scale bar = 1 S3a). If the angular speed of the stage is ω revolutions
mm. per minute (r.p.m.) and the time for rotation is Δt minutes,

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1865
 View Article

Pap Lab on a
then the critical angle α =

1866 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap

2π·ω·Δt radians. The parameterΔt can be further expressed as

force contributions of the hydrophilic ink and the
Δt = N·t1 minutes where N is the number of steps of the
super- hydrophobic surface.36 We denote the advancing
mo- tor and t1 is the time for one step rotation. The
and reced- ing contact angles on the hydrophilic ink as
‘advancing edge’ and ‘receding edge’ are labelled (see ESI†
cos θa,ink and cos θr,ink, respectively. Similarly, the
Fig. S3b). For the plus symbol, the hydrophilic line
advancing and receding contact angles on the
width is w and the length is 2 × R. The liquid droplet
superhydrophobic surface are denoted as cos θa,sub and cos
has a surface tension γ, contact angle θ, viscosity η,
θr,sub, respectively. The parameter ϕ1 indi- cates the azimuthal
density ρ, volume V, radius r (such that V = (4/3)·π·r3),
angle ϕ where the hydrophilic ink region changes to the
and linear velocity ν (such that ν = ω·ζ, where ζ = 3 cm is
superhydrophobic surface in the contact re- gion, and is
the distance from the pivot to the cen- ter of stage). The
given by ϕ1 = sin−1ijw/(2·R)].
azimuthal angle ϕ circumnavigates the pe- rimeter of the
Following from eqn (5), the force Fr acting on the rear
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

contact region between a value of ϕ = 0 at the rear end of

of the droplet can be written as a sum of three forces:
the droplet to a value to ϕ = π/2 at the advancing side of
the droplet.
There are three forces acting on the droplet as the stage is
tilted: surface tension FST, gravitation force FG, and (6)
viscous force FV. At the critical angle α of the stage, the
individual forces balance as:
where
FST + FV = FG (1)

(7)
In eqn (1), the surface tension force FST can be
divided into two components: force Fr acting on the rear of
the drop- let and force Fa acting on the advancing front of
the droplet. Plugging in the expressions for the gravitational
force FG act- ing parallel to the stage and the viscous force
FV, we get: (8)

(Fr − F a )+ 6·π·η·r·ν = ρ·V·g·sin α (2)

To compute the surface tension force, its component f per

unit length of the contact perimeter varies along the
perime- ter as:35 (9)
f = γ·cos θ·cos ϕ (3)

To simplify the calculation, we assume that cos θ inhomogeneity in the surface tension which is accounted
varies linearly around the perimeter of the contact region for by splitting the
between a receding value of cos θr at the rear end of the
droplet (where ϕ = 0) to an advancing value of cos θa at the
advancing side of the droplet (where ϕ = π/2). For the
case of a droplet on a homogeneous superhydrophobic
surface, the expression for the contact angle is given
by:35

(4)

Upon integration of eqn (3) and using eqn (4), the

force acting on the rear of the drop Fr can be evaluated
as:

(5)

In our design with plus symbols, we modify eqn (4) to

ac- commodate the role of hydrophilic symbol on the
surface tension acting on the droplet (see ESI† Fig.
S3b). In other words, the hydrophilic symbol produces an

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1867
 View Article

Pap Lab on a

Similarly, the force Fa acting on the advancing front of the

droplet can be written as a sum of three forces:36

(10)

Substituting eqn (6) and (10) into eqn (2), we can compute
the critical angle α of the inclined stage where the
gravita- tional force balances the surface tension and
the viscous forces; thereby allowing the droplet to detach
from the hydro- philic symbol and slide down the
superhydrophobic surface.
To validate the physical model, experiments are
conducted with water (density ρ = 1 g cm−3, viscosity η =
0.001 Pa s, sur- face tension γw = 72.8 mN m−1) and ethylene
glycol (density ρ
= 1.11 g cm−3, viscosity η = 0.0162 Pa s, surface tension γEG =
47.7 mN m−1) at temperature T = 20 °C. We measured the ad-
vancing and receding contact angles of the two liquids as: (a)
water: θa,ink = 147°, θr,ink = 81°, θa,sub = 157°, and θr,sub =
142°

1868 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap

and (b) ethylene glycol: θa,ink = 134°, θr,ink = 73°, θa,sub =

140°, and θr,sub = 126°. The radius of the contact region is
R = 0.12
mm. Table 1 shows the predicted and experimentally
mea- sured values of the critical angle α. The number of
experi- ments (n) for each combination of line width and
droplet vol- ume is 10. In all cases, the predicted values
lie within one standard deviation of the measured values.
It is worth noting that the viscosity of the liquid droplet
is dependent on the concentration of dissolved
electrolytes or sugars. The concentration-dependent
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

viscosity of various sugar solutions can be modelled as:37

η = η0·a·exp(E·X) (11)

where η0 is the viscosity of pure water (in centiPoise) and X is

the mole fraction in the solution. The parameters a and E are
numerically estimated from experiments. In the case of glu-
cose solutions, the values of the parameters are a = 0.954
Fig. 3 Transport of multiple droplets: a series of images are taken to illustrate the
and E = 27.93 for up to 60% maximum concentration at
movement of multiple droplets on an arrangement of plus symbols. The volume
tempera- ture T = 20 °C.37 of each droplet is 10 μL and they are uniquely coloured with a food dye for visual
illustration. The motor speed is 100 r.p.m. and the number of steps is 14. The
direction of tilting the stage at every step is denoted by a red arrow. The stage is
Transport of multiple droplets and large-volume droplets rapidly tilted twice in left direction (a, b) and three times in the downward
direction (c–e). The final positions of all droplets are shown in (f). This
Using the abovementioned principle, the droplet actuation demonstration shows that multiple droplets can be simultaneously moved in the
system can be used to transport multiple discrete same direction without any risk of cross-
droplets. As shown in Fig. 3, four droplets (each having 10 contamination. Scale bar = 0.5 cm.

μL volume and coloured with different food dyes for visual

illustration)
are initially placed on separate plus symbols. For each sym-
bol, the line width is 0.02 cm, line length is 0.24 cm,
and inter-symbol spacing is 0.335 cm. The motor speed is 100
r.p.m.
and the number of steps is 14. The red arrows in the figure 0.0178 cm, line length is 0.24 cm, and inter-array spacing is
in- dicate the direction of tilting the stage at each step. The 0.68 cm). Reducing the speed and increasing the number
stage is tilted to the right two times (Fig. 3a and b) and of steps of the motor (80 r.p.m., 20 steps) allows
then down- wards for three times (Fig. 3c–e). The final transport of the 80 μL droplet. Here, the stage is tilted
positions of the four droplets are shown in Fig. 3f. The once to the right (Fig. 4a-i and ii), once downwards (Fig.
images indicate that discrete droplets can be transported on 4a-iii), and once to the left as depicted by the red arrows.
a two-dimensional ar- rangement of plus symbols with The final position of the droplet is shown in Fig. 4a-iv.
virtually no risk of cross- contamination between Using a similar approach, Fig. 4b shows images of the
droplets. 300 μL droplet being moved using a 3 × 3 array of plus
To address the challenge of transporting droplets symbols (line width is 0.0178 cm, line length is 0.24 cm,
having volumes greater than 38 μL, we designed arrays of and inter-array spacing is 0.94 cm). The motor speed is
plus sym- bols. Fig. 4a shows images of the 80 μL droplet further reduced and the number of steps is increased to
being trans- ported using a 2 × 2 array of plus symbols move this large droplet (60 r.p.m., 25 steps).
(line width is

Table 1 Critical sliding angle α of a droplet (water and ethylene glycol) is predicted from the physical model and compared from experiments on the actuation system. Three
droplet volumes are tested (20 μL, 30 μL, and 40 μL); each droplet volume is tested on plus symbols having three different line widths (0.152 mm, 0.178 mm, and 0.203 mm). Every
combination of droplet volume and line width is tested 10 times

Water Ethylene glycol

Droplet volume (μL) Line width (mm) Predicted α Measured α Droplet volume (μL) Line width (mm) Predicted α Measured α
20 0.152 26.27° 24.1° ± 1.81° 20 0.152 20.99° 19.6° ± 1.36°
0.178 26.40° 26.2° ± 1.94° 0.178 21.06° 20.9° ± 1.70°
0.203 26.53° 28.5° ± 1.69° 0.203 21.13° 22.1° ± 1.42°
30 0.152 17.19° 15.7° ± 1.18° 30 0.152 14.32° 13.3° ± 0.93°
0.178 17.28° 17.3° ± 1.62° 0.178 14.37° 14.8° ± 0.79°
0.203 17.36° 18.2° ± 1.16° 0.203 14.41° 15.5° ± 0.81°

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1869
 View Article

Pap
40 0.152 12.83° 11.7° ± 1.04° 40 0.152 10.99° Lab on ±a 0.81°
10.5°
0.178 12.89° 12.7° ± 1.34° 0.178 11.02° 10.9° ± 0.81°
0.203 12.95° 13.4° ± 1.37° 0.203 11.05° 11.5° ± 0.72°

1870 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap

Fig. 5 Merging and mixing of multiple droplets. (a) A two-dimensional

arrangement of plus symbols is shown where the line width is thinnest in the left
Fig. 4 Transport of large droplets: (a) a large blue droplet (volume = 80 μL) is two columns, medium thickness in the middle two columns, and thickest in the
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

moved using a 2 × 2 array of plus symbols (line width is 0.0178 cm, line length is right two columns. Three droplets (yellow, red, and blue) are placed on the plus
0.24 cm, and inter-array spacing is 0.68 cm). Compared to Fig. 3 where 10 μL symbols. The red arrow indicates the di- rection of tilting the stage and the stage
droplets were moved, here the motor speed is decreased and the number of steps is is tilted in the following se- quence: downwards, right, right, downwards, and
increased (80 r.p.m., 20 steps) to move the 80 μL droplet. (b) A large green droplet right. The yellow droplet is moved to and merged with the red droplet (a–c). This
(volume = 300 μL) is being transported to the neighbouring pattern using a 3 × 3 merged droplet is now moved and merged with the blue droplet (d–e) and the
array of plus symbols. As in (a), the motor speed is decreased and the number of final product after merging all droplets is shown in (f). After the merging step, the
steps is increased (60 r.p.m., 25 steps) compared to those in Fig. 3. By using the stage can be agitated to mix the combined drop- lets. Scale bar = 0.5 cm.
same scheme and adjusting the parameters of
stepper motors, up to 1 mL droplets have been transported. Scale bar
= 0.5 cm.
droplet is immobile throughout all the tilting
operations. Thus, by adjusting the line widths of the
Here, the stage is tilted to the left and the droplet settles
plus symbols, we can selectively move one or more
on the neighbouring array of 3 × 3 symbols. Even though
droplets to accomplish se- quential merging operations. Post-
larger droplet volume can be transported by changing
merging, the mixing of two droplets is demonstrated in
the design layout, we feel that the droplet volume of 300 μL
Fig. 5c and f by letting the merged product stay put on
adequately represents the maximum threshold needed for
the symbol for some time (depending on the incubation
portable diag- nostic testbeds.29–33
time). This way of mixing by passive diffusion is
satisfactory in case of droplets having sol- uble compounds.
Merging and mixing of multiple droplets
For droplets having immiscible or water- insoluble
The ability to bring two droplets together, merge and compounds, one can mix the droplets by agitating the
mix them, and repeat these steps sequentially with a finite stage (i.e. rapidly tilting the stage in alternate right and
num- ber of discrete droplets is important for realizing left directions in small angles) or moving the droplet in a
on-chip chemical reactions. To achieve this ability, it is cir- cular pattern on neighbouring symbols.
required that some droplets remain stationary while other
droplets are be- ing transported, merged or mixed
together. This is accom- plished by using plus symbols of One-directional transport of droplets
different line widths, where symbols with thicker line While the plus symbols allow us to move droplets in two
widths have more holding force than symbols with di- mensions (i.e. left and right, upwards and
thinner line widths. Fig. 5 shows images of a two-step downwards) on the plastic sheet, there is also interest
merging and mixing performed on three drop- lets. The to control droplet transport in only one direction (i.e. left or
line widths of the plus symbols are thinnest in the left right only, upwards or downwards only). Previously, this
two columns (i.e. 0.015 cm holding the yellow droplet), transport mechanism was demonstrated on a texture
medium thickness in the middle two columns (i.e. 0.02 ratchet where vibrations at the resonance frequency
cm holding the red droplet), and thickest in the right produced directed motion of drop- lets.15 To accomplish
two col- umns (i.e. 0.025 cm holding the blue droplet). this task in our system, we used a
For all sym- bols, the line length is 0.24 cm and inter- ‘greater-than (>)’ symbol that allows us to move a droplet
symbol spacing is only to the right side (i.e. converging side of the symbol)
0.37 cm. The intent here is to merge the yellow droplet upon tilting the stage in that direction. For each symbol,
with the line width is 0.023 cm and the length of each line is 0.33
the red one, and subsequently merge their product with cm. The acute angle between the two lines of the
the blue droplet. The stage is tilted in the following greater-than symbol is 28°. Fig. 6a shows images of two
sequence: downwards, right, right, downwards, and droplets; one placed on a greater-than symbol and the other
right (Fig. 5a–e). The red arrows indicate the direction of on a plus sym- bol. The stage is tilted in the following
tilting the stage. The final product formed after merging sequence: right, left, right, and left. The droplet on the
all the three droplets is shown in Fig. 5f. It is interesting row of plus symbols fol- lows the direction of stage
to note that the red and blue droplets are stationary when tilting, and eventually returns to its original position. In
the yellow droplet is moved and merged with the red one comparison, the droplet on the
(Fig. 5a and b), and the blue
This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1871
 View Article

Pap Lab on a

held in its original position due to the asymmetry of the

greater-than symbol (on its left side compared to its
right side). During the right tilting, the droplet volume
concen- trates to the narrow point of the symbol (on its
right side) and is able to slide to the neighbouring
symbol. Fig. 6c shows images of a droplet placed at the
center of three con- verging greater-than symbols. Here
the line width is 0.023 cm, length of each line is 0.33
cm, and the acute angle of each greater-symbol is 28°.
Similar to Fig. 6a and b, this sym- bol also allows
movement of droplets only to the right side but is able to
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

hold the droplet on its central position even when the

stage is tilted left, up or down. Thus one greater- than
symbol prevents droplet movement in the left direction
while the three converging greater-than symbols
Fig. 6 (a) Each droplet is placed on two different symbols (plus and greater-than prevent droplet movement in the left, up, and down
sign). The droplet on the plus symbol moves to the left when rapidly tilted to the directions.
left, but the droplet on the greater-than symbol does not move. When the substrate
rapidly tilts to the right, both droplets on the plus symbol and the greater-than
symbol move to the right. The acute point of the greater-than symbol has less hy-
drophilic area to attract the droplet. (b) Slow motion images showing different
Dispensing smaller droplets from a large droplet
configurations of the droplet when the stage rapidly tilts to the left and right In wet chemistry experiments, it is often desired to
directions. When the stage tilts left, the two diagonal lines attached to the large
pipette small volumes of reagents or samples repeatedly for
area of the droplet prevent it from moving to next symbol. When the stage tilts
right, a sharp point (where two di- agonal lines meet) attaches to a small area of multiple tests. As such, there is a need to generate equal
the droplet and the droplet is released to the next symbol. (c) One directional volumes of smaller droplets from a large droplet (which
movement: the droplet only moves to the right due to the pattern of three con- may be a re- agent or test sample). Typically, this is
verging greater-than symbols pointing to the center. Scale bar = 0.5 cm. achieved in devices based on electrowetting16–21 or by
using a superhydrophobic blade to split a large droplet.14
We accomplish this task by moving the large droplet over a
series of circular dot symbols. Fig. 7a shows the side-view of a
large red droplet moving over four dot symbols, and leaving
behind a small droplet over
each traversed symbol. Besides circular dot symbols, we can
use rectangular or diamond-shaped symbols for dispensing
small droplets, as shown in Fig. 7b and c, respectively (in all
greater-than symbol is held at its original position when cases, the symbol area is 0.0097 cm2). In Fig. 7d, we
the stage is tilted to the left but moves to the right when show how dispensing and mixing are performed sequentially.
the stage is tilted to the right. Fig. 6b shows the dynamics Here, a large red droplet moves over a row of dot symbols,
of the droplet on the greater-than symbol during the left leaving behind small droplets over each symbol (Fig. 7d-i–
or right tilting of the stage. During the left tilting, the iii). After- wards, a water droplet is moved over the same
droplet is still set of dot

Fig. 7 Droplet dispensing from a large droplet (volume = 10 μL): (a) A small red droplet is dispensed on each circular dot hydrophilic symbol. While a large droplet
moves over the hydrophilic dots, each symbol attracts the droplet and a small volume is left on each symbol. (b) Small red droplets are dispensed on rectangular-shaped
hydrophilic symbols. (c) Small red droplets are dispensed on diamond-shaped hydrophilic symbol.

1872 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap
(d) After the dispensing operation on circular dot symbols, a clear water droplet is transported across the dot symbols, causing the red colour intensity to increase in the
clear droplet. Scale bar = 0.5 cm.

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1873
 View Article

Pap Lab on a
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

Fig. 8 Dispensing droplets from an external reservoir: (a) the reservoir is placed along the edge of the stage. (b) A dispenser tip is pressed by tilting the stage. (c) While
the tip is pushed up, liquid flows out through the opened entrance of the reservoir. (d) A dispensed droplet is transported to another symbol and the next droplet is dispensed.
(e, f) By tilting the stage twice, a larger droplet is dispensed in the same location. Scale bar = 1 cm.

symbols, thereby mixing the previously-left behind red drop-

lets with water (Fig. 7d-iv–vi). We conducted experiments to (12)
measure the actual volume of small droplets left behind as
a 10 μL water droplet travels over plus symbols and
different- sized dot symbols (see ESI† Fig. S4, Tables S1 and
S2). In ad- dition to the fluid properties, the volume of
(13)
droplets dis- pensed on the dot symbol is determined by
the surface area of the symbol or surface defect,38,39 which
can be increased or decreased depending on the desired
In the presence of glucose oxidase, D-glucose is
volume of dispensed droplets.
oxidized to D-gluconic acid and hydrogen peroxide. The
colorless o-dianisidine reacts with hydrogen peroxide, in
Dispensing droplets from an external reservoir
the presence of peroxidase, to form a brown-coloured
Besides dispensing smaller droplets from a large droplet, it is oxidized o-dianisidine.
beneficial to develop a mechanism to dispense finite droplets Initially, experiments are conducted in 24-well plates to
from an external liquid reservoir that may contain a characterize the colorimetric glucose assay. A standard glu-
much larger liquid volume (e.g. cartridges, tubes, and cose assay kit is used to prepare glucose solutions of different
syringes).7 To achieve this method of dispensing, a syringe- dilution factors. Around 250 μL of each solution is
based dispenser is realized. Here, the tip of a 20 mL syringe is loaded into separate well plates, followed by 500 μL of assay
cut, plugged by a 200 μL pipette tip, and then attached to reagent in each well. A webcam is used to record the colour of
a 1 mL syringe. The pipette tip is sealed with a all well solutions for 30 minutes (frame rate: 29 frames per
cyanoacrylate adhesive along with a steel wire to extend second). A Matlab script is written to extract the colour
the tip. This syringe-based dis- penser is positioned intensity of each well solution as a function of time.
above the plastic sheet on the stage (Fig. 8a). As the Specifically, the user selects different cropped areas in the
syringe tip faces downwards, gravitational force prevents first image. Then the script identifies the selected areas of
liquid from back-flowing through the 20 mL syringe. When all subsequent images in a video (see ESI† Fig. S8a). The
the stage is rapidly tilted, the steel wire is mo- mentarily 3-channel (RGB) images are converted into 1-channel
pushed up (Fig. 8b) to dispense a small droplet on the (i.e. grayscale) images using ITU-R Recommendation
hydrophilic symbol underneath (Fig. 8c). This step can be BT.601, and the average colour intensity values are
repeated several times to dispense a series of discrete drop- estimated as a function of time (see ESI† Fig. S8b). The
lets from the reservoir (Fig. 8d–f). colour intensity data are exported to a Micro- soft Excel
spreadsheet. The maximum slope for each solution (i.e.
Glucose detection maximum change in colour intensity per second) is de-
termined that correlates to the initial concentrations of
As a proof-of-concept, the droplet actuation system is
glu- cose.34 For each run with glucose samples, two
employed to determine the glucose concentration in
control sam- ples are used: deionized water with reagent
sheep serum using a colorimetric enzymatic test. The
and black food
following re- action details the chemical reactions involved
dye with reagent. The sheep serum is tested in a similar man-
in the colori- metric test for glucose.34
ner to give its glucose concentration (i.e. 0.59 mg mL−1,
see ESI† Fig. S8c), which is close to the value obtained
from a microplate reader (i.e. 0.63 mg mL−1).
1874 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

Fig. 9 Glucose detection on the droplet actuation system. (a) Glucose standards of different concentrations are placed on the middle column of plus symbols and the
glucose reagents are placed on the leftmost column. (b) The stage is tilted to the right and the two columns of droplets merge on the middle column. (c) The merged
droplets settle on the third column. (d–g) The stage is agitated in multiple directions to mix the combined droplets. (h–j) The merged droplets are incubated for the
chemical reaction and the colour change is visible after around 10 seconds of incubation. The colour intensity is darker for droplets having higher glucose concentrations.
Scale bar = 0.5 cm.

After conducting the well plate experiments, we performed

darker is the colour of the incubated droplet. The
a similar set of experiments on the droplet actuation system.
Matlab script accurately determines the average colour
After preparing the same dilutions of glucose solution, 5
intensity of the droplets (Fig. 10a), which is later used to
μL droplets are placed on the middle column of plus
estimate the glucose concentrations in each droplet (Fig.
symbols (line width = 0.015 cm) as shown in Fig. 9a. Another
10b). The sheep serum is also tested in parallel with other
set of 10 μL glucose reagents are placed on the leftmost
glucose samples. Using the standard curve equation, the
column of plus symbols (line width = 0.02 cm). When the
unknown glucose concentration of sheep serum is
stage is tilted to the right, the two columns of droplets (i.e.
calculated as 0.62 mg mL−1, which is close to the readings
of glucose sam- ples and reagents) merge on the middle
from the microplate reader and well plate experiments.
column (Fig. 9b). Upon further tilting the stage to the
Table 2 summarizes the system parameters for the various
right, the merged drop- lets settle on the rightmost
droplet operations. Table 3 shows the flexibility of the system
column of X-shaped symbols (Fig. 9c) where they are
in transporting droplets having different fluid properties and
agitated to be mixed thoroughly (Fig. 9d–g) and incubated
different volumes. The three fluids tested are: water,
for the chemical reaction (Fig. 9h–j). We found that agitating
milk, and ethylene glycol. Keeping the operating
the stage reduces the mixing time of a merged droplet
conditions fixed (i.e. motor speed = 100 r.p.m., number
(using 5 μL red droplet and 20 μL yellow droplet) from
of steps = 14), we found that a wide range of droplet
550 seconds with passive diffusion to 60 sec- onds with
volumes (7 μL to 40 μL of water) can be transported on
stage agitation (i.e. approximately a nine-fold re- duction
plus symbols (line width = 0.152 mm). However, under
in mixing time) (see ESI† Fig. S5–S7). As shown in Fig.
the same operating conditions, the range of droplet volumes
9h, the colour change is visible after around 10 seconds of
transported on plus symbols de- creases for a viscous
incubation. The higher the glucose concentration, the
liquid (12 μL to 26 μL of ethylene

Fig. 10 Determination of glucose concentrations in sheep serum. (a) The colour intensities of incubated droplets at different time points are shown. Each glucose
concentration is tested three times (n = 3). (b) The maximum slope of each colour intensity graph at different glucose concentrations is plotted to obtain the standard curve
equation and to determine the glucose concentration in sheep serum.

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1875
 View Article

Pap Lab on a

Table 2 Values of the system parameters for the different droplet operations

Droplet operation Figure number Volume (μL) Speed (r.p.m) Steps N Line width (cm) Inter-symbol spacing (cm)
Single droplet transport 2 10 100 14 0.02 0.335
Multiple droplet transport 3 10 100 14 0.02 0.335
Large droplet transport 4(a) 80 80 20 0.0178 0.68
4(b) 300 60 25 0.0178 0.94
Merging and mixing 5(a and b): left 2 columns 10 80 14 0.015 0.37
5(c–e): middle 2 columns 20 90 14 0.02 0.37
5(f): right 2 columns 30 0 0 0.025 0.37
One-directional transport 6(a and b) 10 100 14 0.023 0.37 (+)
6(c) 20 100 14 0.023 0.74 (>)
0.74
Dispending droplets 7(a–d) 10 100 14 Area = 0.0097 cm2 0.37
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

Glucose detection 9(a): left column 10 100 14 0.015 0.45

9(a): middle column 5 100 14 0.02 0.45
9(c): rigth column 15 100 14 0.038 0.45
9(d–g): rigth column 15 40 25 0.038 0.45
9(i and j): rigth column 15 0 0 0.038 0.45

Table 3 The range of droplet volumes that can be transported on plus symbols is helps to automatically extract the parametric data, thereby
shown. Three different fluids are tested: water, milk, and ethyl- ene glycol. The operating minimizing human bias.
conditions of the motors is fixed (speed = 100 rpm, number of steps = 14). Each
experiment on the minimum and maxi- mum droplet volume is conducted 5–7 times

Fluid droplet Fluid properties Line width (mm) Volume (μL)

Acknowledgements
Water η = 0.001 Pa s 0.152 7–40 This work is partially supported by the U.S. National Science
ρ =1 g cm−3 0.203 8–38 Foundation (NSF CBET-1150867 to S. P. and NSF DGE1247194
γw = 72.8 mN m−1 0.254 10–36 to R. B.). In addition, T. K. is partially supported by a
Milk η = 0.003 Pa s 0.152 7.5–38
ρ = 1.032 g cm−1 0.203 9–35 grant from the Defence Threat Reduction Agency (HDTRA1-15-1-
γm = 52.4 mN m−1 0.254 11–33 0053).
Ethylene glycol η = 0.0162 Pa s 0.152 12–26
ρ = 1.11 g cm−3 0.203 17–24
γEG = 47.7 mN m−1 0.254 20–22 References
1 K. S. Elvira, X. C. Solvas, R. C. Wootton and A. J. deMello,
Nat. Chem., 2013, 5, 905–915.
glycol). Supplemental videos show the real-time droplet oper- 2 E. K. Sackmann, A. L. Fulton and D. J. Beebe, Nature,
ations performed on the droplet actuation system (see 2014, 507, 181–189.
ESI† Videos S1–S3). 3 M. J. Jebrail, M. S. Bartsch and K. D. Patel, Lab Chip,
2012, 12, 2452–2463.
4 M. G. Pollack, A. D. Shendorov and R. B. Fair, Lab Chip,
2002, 2, 96–101.
Conclusion 5 K. Choi, A. H. C. Ng, R. Fobel and A. R. Wheeler, Annu.
We demonstrated a droplet actuation system where Rev. Anal. Chem., 2012, 5, 413–440.
discrete droplets are manipulated on hydrophilic patterns 6 W. C. Nelson and C.-J. Kim, J. Adhes. Sci. Technol., 2012, 26,
printed on a superhydrophobic plastic surface. Gravitational 1747–1771.
forces and mechanical agitation of the stage enable the 7 A. Ghosh, R. Ganguly, T. M. Schutzius and C. M. Megaridis,
transport of droplets. The system is designed for low- Lab Chip, 2014, 14, 1538–1550.
cost, resource lim- ited settings where large area, 8 S. Y. Park, M. A. Teitell and E. P. Y. Chiou, Lab Chip,
disposable plastic sheets can be printed from standard 2010, 10, 1655–1661.
inkjet printers and portable 9 V batteries power the 9 Z. Long, A. M. Shetty, M. J. Solomon and R. G. Larson,
motorized stage. We showed the possibil- ity of transporting Lab Chip, 2009, 9, 1567–1575.
multiple droplets (volumes: 8 μL to 300 μL) in parallel 10 J. Seo, S. Lee, J. Lee and T. Lee, ACS Appl. Mater. Interfaces,
and performing sequential fluidic reactions that will be 2011, 3, 4722–4729.
beneficial to a variety of biological experiments. With the 11 V. Jokinen, L. Sainiemi and S. Franssila, Adv. Mater.,
presented method, the design and layout of the hy- 2008, 20, 3453–3456.
drophilic symbols can be easily altered to specific 12 Z. Wang and J. Zhe, Lab Chip, 2011, 11, 1280–1285.
functional requirements of an experiment. Lastly, the 13 D. Foresti, M. Nabavi, M. Klingauf, A. Ferrari and D.
integration of smart image analysis tools with the droplet Poulikakos, Proc. Natl. Acad. Sci. U. S. A., 2013, 110,
actuation system 12494–12554.

1876 | Lab Chip, 2016, 16, 1861–1872 This journal is © The Royal Society of Chemistry 2016
 View Article

Lab on a Pap

14 H. Mertaniemi, V. Jokinen, L. Sainiemi, S. Franssila, A.

27 H. Kim, M. S. Bartsch, R. F. Renzi, J. He, J. L. Van de
Marmur, O. Ikkala and R. H. Ras, Adv. Mater., 2011, 23,
Vreugde, M. R. Claudnic and K. D. Patel, J. Lab. Autom.,
2911–2914.
2011, 16, 405–414.
15 T. A. Duncombe, E. Y. Erdem, A. Shastry, R. Baskaran and
28 R. Sista, Z. Hua, P. Thwar, A. Sudarsan, V. Srinivasan, A.
K. F. Bohringer, Adv. Mater., 2012, 24, 1545–1550.
Eckhardt, M. Pollack and V. Pamula, Lab Chip, 2008,
16 J. Gong and C. J. Kim, J. Microelectromech. Syst., 2008, 17, 8, 2091–2104.
257–264. 29 N. A. Mousa, M. J. Jebrail, H. Yang, M. Abdelgawad, P.
17 C. Peng, Z. Zhang, C. J. Kim and Y. S. Ju, Lab Chip, 2014, Metalnikov, J. Chen, A. R. Wheeler and R. F. Casper, Sci.
14, 1117–1122. Transl. Med., 2009, 1, 1ra2.
18 S. C. Shih, P. C. Gach, J. Sustarich, B. A. Simmons, P. D. 30 C. Arnaud, Chem. Eng. News, 2011, 89, 13–17.
Adams, S. Singh and A. K. Singh, Lab Chip, 2015, 15, 225–236. 31 R. S. Sista, A. E. Eckhardt, T. Wang, C. Graham, J. L. Rouse,
Published on 08 April 2016. Downloaded by Iowa State University on 4/1/2022 7:50:10

19 S. Srigunapalan, I. A. Eydelnant, C. A. Simmons and A. S. M. Norton, V. Srinivasan, M. G. Pollack, A. A. Tolun, D.

R. Wheeler, Lab Chip, 2012, 12, 369–375. Bali, D. S. Millington and V. K. Pamula, Clin. Chem.,
20 M. Abdelgawad, S. L. Freire, H. Yang and A. R. Wheeler, Lab 2011, 57, 1444–1451.
Chip, 2008, 8, 672–677. 32 M. J. Jebrail, H. Yang, J. M. Mudrik, N. M. Lafreniere, C.
21 C. J. Huang, W. F. Fang, M. S. Ke, H. Y. E. Chou and J. T. McRoberts, O. Y. Al-Dirbashi, L. Fisher, P. Chakraborty and
Yang, Lab Chip, 2014, 14, 2057–2062. A. R. Wheeler, Lab Chip, 2011, 11, 3218–3224.
22 B. Hadwen, G. R. Broder, D. Morganti, A. Jacobs, C. Brown,
33 S. C. C. Shih, H. Yang, M. J. Jebrail, R. Fobel, N. McIntosh,
J. R. Hector, Y. Kubota and H. Morgan, Lab Chip, 2012,
O. Y. Al-Dirbashi, P. Chakraborty and A. R. Wheeler, Anal.
12, 3305–3313.
Chem., 2012, 84, 3731–3738.
23 P. Y. Chiou, H. Moon, H. Toshiyoshi, C.-J. Kim and M. C.
34 V. Srinivasan, V. K. Pamula and R. B. Fair, Anal. Chim. Acta,
Wu, Sens. Actuators, A, 2003, 104, 222–228.
2004, 507, 145–150.
24 R. Renaudot, V. Agache, Y. Fouillet, G. Laffite, E. Bisceglia,
35 C. Extrand and A. Gent, J. Colloid Interface Sci., 1990, 138,
L. Jalabert, M. Kumemura, D. Collard and H. Fujita,
431–442.
Lab Chip, 2013, 13, 4517–4524.
36 M. Elsharkawy, T. M. Schutzius and C. M. Megaridis, Lab
25 Z. Guttenberg, H. Müller, H. Habermüller, A. Geisbauer,
Chip, 2014, 14, 1168–1175.
J. Pipper, J. Felbel, M. Kielpinski, J. Scriba and A. 37 J. Chirife and M. P. Buera, J. Food Eng., 1997, 33, 221–226.
Wixforth, Lab Chip, 2005, 5, 308–317. 38 J. Schneider, A. Egatz-Gomez, S. Melle, S. Lindsay, P.
26 Z. Hua, J. L. Rouse, A. E. Eckhardt, V. Srinivasan, V. Dominguez-Garcia, M. A. Rubio, M. Marquez and A. A.
K. Pamula, W. A. Schell, J. L. Benton, T. G. Mitchell and Garcia, Colloids Surf., A, 2008, 323, 19–27.
M. G. Pollack, Anal. Chem., 2010, 82, 2310–2316. 39 N. A. Patankar, Langmuir, 2003, 19, 1249–1253.

This journal is © The Royal Society of Chemistry 2016 Lab Chip, 2016, 16, 1861–1872 | 1877