Performance evaluation of novel nanostructured modified mesoporous silica/polyetherimide composite membranes for the treatment of oil/water emulsion

Separation and Purification Technology 205 (2018) 32–47 Contents lists available at ScienceDirect Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur Performance evaluation of novel nanostructured modified mesoporous silica/polyetherimide composite membranes for the treatment of oil/water emulsion T Noel Jacob Kaleekkala, , Ramakrishnan Radhakrishnanb, Vishnu Sunilc, Geethanzali Kamalanathanc, Arijit Senguptad, Ranil Wickramasinghed ⁎ a Department of Chemical Engineering, School of Chemical and Biotechnology, SASTRA Deemed University, Tanjore 613401, India Catalysis Laboratory, Department of Applied Science and Technology, ACT, Anna University, Chennai 600031, India Department of Chemical Engineering, ACT, Anna University, Chennai 600031, India d Ralph E. Martin Department of Chemical Engineering, University of Arkansas, Fayetteville, AR, USA b c A R T I C LE I N FO A B S T R A C T Keywords: Composite membranes Mesoporous silica Hydrophilicity Oil-water emulsion Bacterial anti-adhesion The discharges from industries particularly petroleum industry at any stage from exploration/drilling to transportation are of major environmental concerns in the present scenario. In this work, mesoporous silica (SBA-15) was modified using a bio-inspired coating and was used as filler in the fabrication of efficient, robust polyetherimide composite ultrafiltration membranes. The membranes prepared by this facile two-step approach were seen to be highly hydrophilic with improved porosity and pore-interconnectivity with a thinner skin layer. The rougher top surfaces of these membranes imparted an oleophobic character under water. The composite membranes exhibited improved emulsion/water flux while maintaining > 99.8% rejection of oil from a synthetic motor oil/water/surfactant emulsion. The composite membranes were investigated for its long-term efficiency in the removal of oil from produced water. The flux declination was only < 15% for up to 9 h with three intermittent backwashes and the oil/grease content of the permeate was lower than 10 ppm, well below the discharge limits. A significant improvement in antimicrobial characteristic was found to be achieved by the composite membranes against the gram-positive (Bacillus subtilis) and gram-negative (Pseudomonas aeruginosa) bacteria. The results from this study indicate that the polydopamine decorated SBA-15incorporatedpolyetherimide membranes hold a promising potential to be employed for oil-water emulsion separation. 1. Introduction The environmental destruction and threat to the ecological imbalance caused by the oil spills during the Gulf of Mexico Oil Spill (2010, USA) [1] or the recent one near the coast of Chennai (2017, India) [2] have highlighted the difficulty and the cost of separating the oil from water. Apart from these unexpected threats to the environment, point sources like deep drilling for fossil fuels - oil, gas, coal bed methane and more recently hydraulic fracking for extraction of these fuels also generate large amounts of wastewater as byproducts. This water, referred to produced water, process water or flow backwater [3] and generally contains high concentrations of light naphtha (liquefied petroleum gas, jet fuel, kerosene, and gasoline) or heavy naphtha (fuel oil, lube cut, grease, and diesel oil), apart from minor toxic substances and a great concentration of suspended or dissolved solids [4,5]. The ⁎ Corresponding author. E-mail address: noeljacob89@gmail.com (N.J. Kaleekkal). https://doi.org/10.1016/j.seppur.2018.05.007 Received 11 January 2018; Received in revised form 6 May 2018; Accepted 6 May 2018 Available online 07 May 2018 1383-5866/ © 2018 Elsevier B.V. All rights reserved. sequential stages of treatment of the produced water include (1) Removal of particulates including suspended sand, clay, etc. (2) Removal of organics, which include dispersed, dissolved and emulsified oil and grease and (3) Removal of dissolved inorganics referred to as total dissolved solid (TDS). Free or floating oils can be easily separated by conventional techniques like air-floatation or gravity settling followed by chemical treatment, skimming, adsorption onto activated carbon/ silicon sponges or centrifugation [6,7]. However, their use is curtailed by low separation efficiency, high energy consumption, the requirement of complex equipment, production of toxic substances or secondary by-products. Moreover, the existence of stable oil microemulsions, (droplet size < 10 µm dia) which cannot be separated by these techniques, makes it imperative to employ membrane-based technologies [8-10]. The United States and European regulations limit the oil content in Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. the synthesis. The optimized gel composition was 1:5.87:194:0.017 TEOS:HCl:H2O:P123 [23]. Briefly, 4.0 g of P123 was dissolved by a high shear stirrer into a solution of 20 ml of conc. hydrochloric acid (37%, Merck, Germany) and 130 g of water in a PP autoclavable bottle (NALGENE, Sigma, St. Louis, MO) maintained at 318 K for 2 h. 9.20 ml of TEOS was added dropwise to the transparent mixture and then the solution was vigorously stirred at 318 K for 24 h. The resulting mixture was then transferred into a Teflon lined autoclave bottle and aged at 373 K for 24 h under static condition. After being cooled to room temperature, the white precipitate was filtered and washed with deionized (DI) water (twice) to remove the unreacted acid residues. Finally, it was dried at 393 K for 12 h. The white material was ground and calcined in a muffle furnace at 823 K for 5 h to remove the triblock copolymer organic template and yield SBA-15. effluents (5–20 mg/L) prior to the discharge into the environment [11]. Polyetherimide (PEI) based tight ultrafiltration (UF) membranes can be used to address these issues due to their efficiency in the separation of emulsified oil, ease of scaling up, ability to be cascaded in series with reverse osmosis (RO)/nanofiltration (NF) membranes and minimum use of chemicals (only for membrane cleaning) [12,13]. The properties like good resistance to different chemical environments and UV irradiation, easy tunability of pores; lead PEI membranes to an advantageous position for their applications [14,15]. Application of ultrafiltration membranes is fast evolving in oil and gas industries to meet highquality water requirements and thus, a reasonably substantial number of publications have been devoted to studying the membrane technology in oily wastewater treatment. Fouling by the oil resulting deterioration of membrane performance is one of the most prominent challenges required to be addressed. To mitigate the fouling; superwetting membranes with ultra-low adhesion properties, which are superoleophobic/superoleophilic, underwater, are being designed to separate water/oil from the emulsion. Recent modifications of the polymeric membranes include hydrogel coatings[14], alkaline induced hierarchical membranes [14], zwitterionic polymeric component [10], introduction of modified graphene oxide [16], inducing formation of amidoxime group by altering the composition of coagulation bath [17], addition of amphiphilic block copolymers [18] incorporation of modified silica materials[9,19] etc. Biomimetic, mussel-inspired technologies are gaining momentum in imparting superhydrophilicity and underwater superoleophobicity to the polymeric membranes, which alleviates irreversible fouling by oils, enhances separation performance, improves water permeability, ensures longer membrane utilization with excellent recyclability [20]. The self-polymerization of dopamine and its subsequent anchoring onto a membrane surface have a few inherent limitations- the extent of polymerization cannot be easily controlled as the rate of polymerization of dopamine depends on the reaction temperature, concentration and time of reaction [21]. This modification could lead to a flux decline because of reduction in pore sizes and an increase in thickness of separation layer [20]. Moreover, in hollow fiber geometry, this modification could be difficult to implement. In view of these, novel polydopamine (pDA) decorated mesoporous silica SBA-15 incorporated PEI composite membrane has been synthesized for the treatment of produced water. Due to the ease of synthesis, high surface area, adjustable pore size, thick pore wall, remarkable hydrothermal stability; SBA-15, a mesoporous silica was selected as the inorganic filler in the present investigation [22]. A 2-step approach based tight ultrafiltration membranes - synthesis of SBA-15 followed by deposition of polydopamine and further incorporation of the modified fillers into the membrane matrix has been adopted to prepare hydrophilized polyetherimide UF membranes. The effect of incorporation of pDA-SBA-15 was investigated in terms of surface morphology, surface chemistry, surface wettability, water permeation rate and tensile strength. The composite membranes showed superior ability in the separation of oil from simulated oil/water emulsions. They were also found to exhibit extraordinary antifouling characteristics. The dopamine coating was found to be responsible for the inhibition of bacterial attachment onto the membrane surface. These composite membranes were investigated for the treatment of produced water from the Manali Refinery (Chennai Petroleum Corporation Limited, Chennai, India). 2.2. Synthesis of polydopamine decorated SBA-15 100 mg of the prepared SBA-15 was introduced into a 100 ml (12.5 mM, pH = 8.5) of tris buffer (HiMedia India) solution and was dispersed mildly using a sonication bath. 250 mg of dopamine hydrochloride (Sigma, Aldrich, St. Louis, MO) was added to it and the solution was stirred continuously at 35 °C for 8 h. The product was allowed to settle down and separated by centrifuging. Subsequently, the product washed with DI water to remove the unreacted dopamine. The product is further dried in a vacuum oven at 80 °C for 24 h and the final weight obtained was 165 mg (Scheme 1 and Fig S1). 2.3. Preparation of composite membranes Polydopamine decorated SBA-15 (pDA-SBA-15) incorporated polyetherimide (PEI; Sigma Aldrich, St. Louis, MO) was prepared by dry-wet phase inversion technique. The casting solution was prepared (Table S1) by addition of pDA-SBA-15 into the solvent, N-methyl-2-pyrrolidone (NMP; SRL Chemicals, India) followed by 1.5 wt.% of polyvinylpyrrolidone K-90 (PVP, Sigma Aldrich, St. Louis, MO) and 16 wt.% of PEI under continuous stirring conditions to obtain a homogenous solution. This was then degassed and cast onto a glass substrate using a semi-automatic casting unit maintaining a constant thickness of 200 ( ± 10) µm. The casted solution was held for 30 s in a chamber with a relative humidity of 20 ± 2% before being introduced into a coagulation bath (15 °C, 0.2 wt.% NMP). The membrane (Fig S1) was kept overnight in DI water and then stored for further testing and analysis. 2.4. Characterization 2.4.1. Characterization of the SBA-15 and pDA decorated SBA-15 The SBA-15 and the pDA decorated mesoporous silica was confirmed by Low angle X-ray diffraction (XRD) analysis (Bruker-AXS diffractometer, Germany), high resolution scanning electron microscopy (HR-SEM) (FEI Quanta FEG 200, USA), high resolution transmission electron microscopy (HR-TEM) (JEM-3010 TEM, Jeol Ltd., Tokyo, Japan), atomic force microscopy (AFM) (Park XE-100, South Korea) Optical imaging, Fourier transformed infrared (FT-IR) Spectroscopy (Nicolet Avatar 370, Thermo Electron Corp., Madison, WI, USA), BET Analysis (Micrometrics ASAP 2020, USA), Pore size and pore volume analysis and X-ray Photoelectron Spectroscopy (XPS) (PHI 5000C ESCA system, PHI CO., USA). The d-spacing is obtained by Bragg’s Law as follows 2. Experimental 2.1. Preparation of SBA-15 A typical synthesis of hexagonal, highly ordered, large pore, mesoporous silica filler was carried out using poly(ethylene glycol)-b-poly (propylene glycol)-b-poly(ethylene glycol) (P123, Aldrich St. Louis, MO) and tetraethyl orthosilicate (synthesis, TEOS, Merck Germany) as a structure directing agent and as the silica precursor, respectively, in highly acidic conditions. The hydrothermal method was employed for d100 = n·λ 2sinθ (1) where n = integer; λ = wavelength of the incident light (Cu Kα = 1.54 Å);dhkl = lattice Spacing; θ = angle of incidence. The unit cell parameter can be calculated using the formula 33 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Scheme 1. Preparation schematic of pDA decorated mesoporous SBA-15. composite membranes were evaluated by (2) a0 = 2d100/ 3 And the pore wall thickness is determined by Pw = a0−Dp ; where Dp is the average pore diameter determined by BJH method. Jw (l/m2 h) = (3) where Ww, Wd correspond to the wet and dry weights of the membrane samples; V, ρ correspond to the volume of the dry membrane and density of water respectively. The membrane-free volume corresponds to the porosity of the membrane, which could be determined by the gravimetric analysis by wetting the membrane with water. Al Kα (1486.6 eV) was used as a source for XPS analysis of the membranes in order to understand the inorganic filler composition on the membrane surface. The cross-sectional morphology of the membranes was imaged using an HR-SEM, while the top-surface morphology could be determined by AFM operated in the non-contact mode (n = 3). The wettability of the membrane surface by water (in air) and 1,2 dichloroethane (underwater) were carried out by the contact angle goniometer (OCA-15 EC, Dataphysics, Germany) and the measurements were carried out in at least 8 different spots to ensure reliability. The work of adhesion (water-air-membrane interface) was calculated using the following equation. where γlv is surface tension of the water in air (72 mN·m and θ is the contact angle. 2.4.4. Flux decline and anti-fouling studies Membranes were pre-compacted and soaked in DI water for 24 h prior to the experiment. The pure water flux (JW) was measured at 0.2 MPa at a temperature of 25 °C. The oil-in-water emulsion was used as a feed solution and was permeated through the membrane for 60 min, followed by back-flushing with DI water for 5 min. The pure water fluxes were determined again (after backwash) and denoted as Ji (i = 1, 2, 3, 4). The reduction in the permeate flux for the emulsion was also determined to understand the fouling propensities of the membranes. The flux recovery ratio (F.R.R) was determined by; (4) Wa = γlv (1 + Cosθ) −1 (5) where Qp is the permeate volume (l), A denotes the effective surface area of the membranes (m2) and t is the sampling time (h). The recycle experiments were carried out for a fixed volume of 100 ml, which was allowed to permeate. The membranes were then rinsed thoroughly, and the flux was measured again. This was repeated for 10 cycles, and the membranes were rinsed after each cycle for membrane S3. Highly stable synthetic oil in water emulsions were prepared by dispersing 10 ml of light motor oil (Castrol, UK) in 990 ml of DI water using 15 ml of sodium dodecyl sulfate (SDS; Merck, Kenilworth, NJ) by stirring at 3500 rpm for 5 h to obtain a milky white emulsion used as feed in rejection and anti-fouling experiments. Oil filtration was carried out using the same testing kit and the rejection studies were carried out at a pressure of 0.2 MPa and the permeate quality was analyzed by a total organic carbon (TOC) analyzer (Shimadzu, TOC-V CPH). 2.4.2. Membrane characterization The free volume of the membrane (∈ %) was evaluated as follows ∈ (%) = (Ww−Wd )/(V ·ρ) Qp A·t at 25 °C), 2.4.3. Membrane testing and performance using simulated oil-in-water emulsion Membrane samples were cut into discs of 30 cm2 to fit into the membrane cell holder of the cross-flow testing kit and the experiments were carried out at a constant pressure of 0.2 MPa, after compaction at 0.35 MPa. The pure water flux Jw of the pristine membrane and the F . R. R (%) = Ji × 100 JW (6) The total fouling (RT) can be constituted as the sum of irreversible and reversible fouling. The irreversible fouling (Rir) is the cumulative 34 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Fig. 1. High resolution SEM Images of pristine SBA-15 (a) and pda-SBA-15 (b). Hr-TEM Images of pda-SBA-15 at higher magnifications showing the lattice structure and polydopamine deposition (c and d). local regulation and hence treatment prior to disposal became essential [24]. The efficiency of the composite membranes in the removal of oil from the emulsion was evaluated using a similar procedure as described earlier. The flux recovery ratio was determined for up to 3 cycles. fouling, which cannot be removed by backwashing after 3 flux-cycles. The reversible fouling (Rr) is the cumulative fouling, that can be removed. Rir = JW −J4 × 100% JW RT = Jw−Jsol × 100% Jw Rr = (RT −Rir ) × 100% (7) 2.6. Bacterial adhesion test (8) Bacterial adhesion on to membrane surface is detrimental as it forms colonies and bio-films which affects the membrane performance. The isolates of as obtained B. subtilis and P. aeruginosa (National Center for Microbial Resource, Pune, India) were initially incubated individually in 100 ml of sterile nutrient broth at 37 °C at 200 rpm using a temperature controlled orbital shaker for 24 h. The bacterial suspension was washed with double distilled water during the log phase and resuspended in water without any nutrients to achieve a half-McFarland standard (almost 108 cfu/mL). Membrane discs were prewashed with 70% ethanol and then incubated with 1 ml of the bacterial solution/ water for 24 h in a micro-well titer plate. After incubation, the (9) 2.5. Oil removal from produced water Produced water was collected from Manali Refinery, Chennai Petroleum Corporation Limited, Chennai, India. The collected water was allowed to settle for 5 days in order to separate the sand and clay particles. The characterization of this wastewater is summarized in Table S2 and the discharge of this wastewater was unacceptable by the 35 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. hexagonal mesoporous structure with interconnected micropores and mesopores. The peaks are indexed in a hexagonal lattice with the crystallographic space group of p6mm and are in good agreement with the literature [29]. A high-intensity diffraction peak at 2θ ∼ 0.98° can be indexed as (1 0 0) plane; the two smaller diffraction peaks at 1.68° and 1.91° corresponds to (1 1 0) and (2 0 0) planes respectively (Fig. 2a) [30]. The d-spacing of the (1 0 0) reflection is calculated to be 9.01 nm, which accounts for a lattice constant of 10.40 nm. All the peaks were still well resolved after the pDA coating confirming that the high order framework of the mesoporous silica was still maintained. A slight shift to a higher angle was observed for the (1 0 0) plane, which could be due to the pore-filling effects by pDA deposition which could reduce the xray scattering effect dopamine coating present within the pores [31]. The pDA coated SBA-15 retained all characteristic peaks of the pristine mesoporous material as seen in the FT-IR spectra in Fig. 2b. The high-intensity bands at 1064 cm−1 and 801 cm−1 can be attributed to the asymmetric and symmetric stretching vibrations of the SieOeSi present in the siliceous backbone, respectively. The SieOH bending vibration band is evidenced at 966 cm−1. The peak at 1697 cm−1 is an indication of absorbed water molecules rendering the material hydrophilic. An additional peak at 1506 cm−1 could be ascribed to stretching of the benzene ring present in the pDA. The broad peak at ∼3370 cm−1 can be attributed to the stretching frequency of eOH and NH groups of the pDA decorated mesoporous silica [26,32]. The results indicate a successful deposition of pDA on the SBA-15 material. Fig. 2c shows the multi-point BET isotherms of the nitrogen adsorption and desorption at 77 K. The pristine SBA-15 and the polydopamine-coated material exhibited classical type-IV isotherm, typical of characteristic capillary condensation of mesoporous material. Moreover, the distinguishable H1-type hysteresis loop (0.6–0.8) was observed [33]. The specific surface area was evaluated from the Brunauer –Emmett – Teller (BET) equation from the adsorption data in the relative partial pressure from 0.05 to 0.3. The total pore volume was derived from the BET plot by calculating the amount of N2 at a relative partial pressure of 0.95. The pore size distribution (PSD) was calculated from the desorption branch of the isotherm using Barrett – Joyner – Halenda (BJH) model. The specific surface area and cumulative specific pore volume of the pristine SBA-15 were evaluated as 745 m2/g and 1.02 cm3/g, respectively; and the significant peak of the BJH pore size (inset Fig. 2c, inset) distribution was found to be centered at 7.87 nm. The pDA coated SBA15 showed an appreciable reduction in pore size, pore volume and specific surface area, which reveals that the successful coating of the polydopamine was achieved [34] (see Table 1). The XPS scan of the pristine SBA-15 exhibited peaks at binding energies of 533.15 eV, 155 eV and 103 eV corresponding to O 1s, Si 2s, and Si 2p, respectively (Fig. 2d). A very small intensity carbon peak is also seen at a binding energy of 284 eV, which was attributed to carbon contaminant of the instrument or residual precursors [35,36]. The enhancement of C 1s peak and an appearance of an additional peak at 399.1 eV (ascribed to N 1s) indicated the coating of pDA onto pristine SBA-15 [37]. The XPS and FTIR analyses were found to be in mutual agreement. All these results confirm the mesoporous SBA-15 was uniformly decorated with pDA and that the pDA coating was stable even after repeated washings. membranes were rinsed with DI water and transferred to a new plate. A 1 ml of crystal violet was added to each well and incubated for further 15 min following which the membranes were washed to remove excess stain. 2 ml of a 95% (v/v) ethanol solution was added (15 min) to release the crystal violet from the bacterial cell walls. The optical density of the released crystal violet in each well was measured at 540 nm using a UV–Visible spectrophotometry (Libra S50, Biochrom Ltd., Cambridge, England). Relative bacterial attachment = OD S0 − P . C . OD S0 − N . C ./ OD SP . C OD SN . C . (10) where ODS0-P.C. and ODS0-N.C. correspond to the optical densities of positive control (bacterial solution) and negative control (water), respectively, for the membrane S0. ODS-P.C.and ODS-N.C. correspond to the positive control and negative control of the composite membranes. 3. Results and discussion 3.1. Characterization of prepared mesoporous fillers The color of the mesoporous silica material, upon modification, changed from pure white to brown indicating the self-polymerization and coating of dopamine onto the mesoporous SBA-15 as also observed by other researchers [25]. The modification of the SBA-15 is a facile process without employment of much (or no) chemicals. This benign modification involves self-polymerization of dopamine and its successive anchoring onto the mesoporous silica. However, the exact mechanism of these steps is still elusive and continue to be of interest to researchers. The mechanism of the decoration of pDA could be explained by the intermolecular H-bonding between catechol moiety of the pDA and germinal and isolated hydroxyl groups of the SBA-15[26]. The HR-SEM was used to visualize the surface morphology of the prepared mesoporous material as seen in Fig. 1(a and b). The mesoporous SBA-15 showed well-defined hexagonal type vermicular structures with a size of ∼480 nm [27]. This elongated mesoporous material has channel-like morphology running parallel to the main axis within the hexagonal symmetry. These are present joint as long-fiber like macrostructures with lengths up to few micrometers as evidenced by other researchers as well [25]. A uniform deposition of pDA with no changes in the skeletal structure of the SBA-15 can also be observed for the modified material. The highly ordered mesochannels in a hexagonal arrangement of the SBA-15 are seen (Fig. 1c) from the HR-TEM images when the imaging electron beam was parallel to the longer axis, while unidirectional channels are observed when the beam was perpendicular to the channel axis. This confirms its 2D hexagonal pore structure. The d100 ≈ 9 nm obtained from HR-TEM is in good agreement with the dspacing value obtained from the X-ray diffraction results (see Table 1). The pDA coating rendered the surface slightly uneven (Fig. 1d) without compromising the structural regularity or integrity of the mesopore [28]. In addition to the pDA deposition on the surface of the SBA-15, a fine coating within the pores is also evident. This observation concurs with the slight decrease in pore volume from the BET analysis. The small angle X-ray diffraction of the synthesized SBA-15 shows all three characteristic peaks indicating the formation of the predicted 2-D Table 1 Characteristic properties of the prepared filler. 3.2. Characterization of the fabricated membranes Filler 2θ (°) d100 (nm) a0 (nm) Dp (nm) SBET (m2/g) Vp (cm3/g) SBA-15 pDA SBA-15 0.98 1.03 9.01 8.57 10.40 09.89 7.87 7.01 745 672 1.02 0.72 3.2.1. Physico-chemical characteristics of the composite membranes The incorporation of modified mesoporous silica into the PEI membrane matrix was found to modify the chemical composition of the membrane surface successfully. The wide scan of the pristine membrane displayed signals at 532.3 eV, 400.2 eV and 285 eV corresponding to O 1s, N 1s, and C 1s, respectively [38]. The composite membranes displayed additional two emission spectra corresponding to Si 2s and Si 2p at 155.6 eV and 102.7 eV, respectively (Fig. 3a). This confirms the The 2θ corresponds to the (100) reflection; d100 denotes the d-spacing calculated by Bragg’s Law; a0 indicates the lattice parameters by geometric calculations; Dp is the BJH adsorption mean pore diameter; SBET is the specific surface area from multi-point BET analysis and Vp is the cumulative adsorptive pore volume. 36 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Fig. 2. Characterization of pristine SBA-15 and pda-SBA-15: Small angle X-ray diffraction (a); Fourier transform infrared spectroscopy (b); BET-Isotherms of N2 adsorption and desorption (c) and pore-size distribution (c, inset); XPS spectra (d). Fig. 3. Wide survey XPS spectra of the composite membranes and the elemental composition (a); Dope solution viscosity and porosity of the composite membranes (b). 37 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. formation of micropores near the fillers. A wide variety of fillers like Arabic gum [39], Fe2O3 [40], titania [41], Fe-Ag/MWCNTs [42], modified GO [43], etc. have been employed to improve membrane porosity and hydrophilicity. An earlier study reported no improvement in membrane porosity on the incorporation of SBA-15 [44]. It can be a synergistic effect of formation of micropores between the polymer chains and the preferential migration of the hydrophilic filler for the non-solvent (water) during the de-mixing process resulting in the enhancement of membrane porosity. presence of the fillers on the membrane surface. An increase in the concentration of N 1s and O 1s can be attributed to the increase in pDASBA-15 concentration on the membrane surface. The viscosity of the casting solutions was determined at a constant shear rate of 0.1 s−1. The viscosity of the casting solution is one of the factors determining the morphology of the membranes. The viscosity of the casting solution is seen to nearly double (Fig. 3b) on the addition of 0.1 wt.% of the filler, which could be explained by the inhibition of rotation or free movement of the polymeric chains due to the presence of the long mesoporous fillers. The increase in viscosity of the casting solution can lead to a delayed demixing during membrane formation. The addition of the fillers increases the overall free volume of the prepared membranes as seen in Fig. 3b. The free volume of the membranes was determined by the gravimetric method by water uptake measurements. The water uptake is not strictly ascertained by the free volume of the dried membrane as it depends also on the chemical nature of the membrane. The presence of these mesoporous fillers can explain both increases in membrane porosity as well as the hydrophilicity of the membranes. During vitrification of the dope solution, the relaxation of the interfacial stress of the system causes the 3.2.2. Morphology of the composite membranes The change in morphology of the membrane cross section due to the loading of the fillers could be observed using SEM imaging (Fig. 4). The composite membranes prepared by the dry-wet phase inversion showed a typical asymmetric structure with a top separating skin layer and vertical grown pores and macrovoids. The general morphology of the membranes remained almost invariable. However, the thickness of the skin layer was seen to decrease with increase in filler concentration up to membrane S3. A growth of the finger-like pores interconnected with the membrane bulk was observed, which is similar as observed by Fig. 4. Cross sectional morphology of the composite membranes using SEM imaging: S0 (a); S1 (b); S2 (c); S3 (d); S4 (e). 38 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Fig. 5. Top Surface morphology of S0 (a); S2 (b); S3 (c); S4 (d) composite membranes using Atomic force microscopy. dimensional image indicating the surface roughness for 5 × 5 µm area of the prepared composite membranes are displayed in Fig. 5.The surface roughness of composite membranes was evaluated in 5 distinct areas of the same membrane disc and the average value reported in Table2. A typical peak and valley structure was observed for all the composite membranes. The pristine membrane exhibited the lowest surface roughness of ∼32 nm and it progressively increased to ∼42.5 nm for membrane S3. This implies that the surface roughness of the composite membrane is influenced by the incorporation of fillers into the membrane matrix. There is a preferential migration of the hydrophilic fillers towards the membrane surface. As the concentration of the fillers increased to 0.1 wt.%, a much greater surface roughness was observed (S4); which could be elucidated by the presence of agglomerated dopamine decorated SBA-15 on the membrane top surface. It is widely reported that a lower surface roughness is associated with diminished fouling propensity when nano-materials are employed [47,48] or when leached with Candle soot [49]. It is widely believed that the valleys are the area where the foulants from the feed could settle and adhere irreversibly thus tuning the preparation to produce smoother membrane surfaces. However, in the present investigation, the increase in surface roughness is only by a few nanometers and in incorporation of ZrO2 into the PVDF matrix [45]. The membrane S4 exhibited no reduction in rejection efficiency indicating an excellent dispersion as well as compatibility of the pDA-SBA-15 into the membrane matrix even up to 0.1 wt.%. The decline in rejection would be more prominent if there is a great degree of incompatibility between the filler and the polymer matrix. This was attributed to the formation of a greater number of non-selective voids [46]. The inorganic fillers used are highly hydrophilic and is well dispersed into the solvent NMP, and which is responsible for the enhancement in the rate of de-mixing during membrane formation by the dry-wet phase inversion. These fillers also hinder the free movement of the polymeric chains due to an increase in solution viscosity, which can hinder the motion of the solvent. The maximum porosity is observed for membrane S2 as supported by SEM images. At lower filler loading ratio the rate of de-mixing is high and as the concentration of filler increases the dope solution viscosity plays a more prominent role. The membrane S4 displayed unexpected pore openings at the lower surface as well; indicating an excess concentration of the filler. However, there was no loss of membrane integrity during the tests. It is widely reported that the membrane surface roughness influences the membrane surface wettability by oil/water. The threeTable 2 Physico-chemical properties of the composite membranes. Membrane Average roughness Sa (nm) Water contact angle (°) Work of adhesion (mJ/m2) UWOCA (°) Pure water flux (LMH) @ 0.2 MPa S0 S1 S2 S3 S4 31.91 30.62 36.80 42.52 74.82 73.5 68.2 52.0 37.2 38.8 92.81 98.38 115.93 129.12 127.55 89.0( ± 2.5) 104.5 ( ± 2.0) 133.5 ( ± 3.5) 147.5 ( ± 5) 149.2 ( ± 5) 89.2 ( ± 4.5) 127.2 ( ± 3.5) 172.5 ( ± 2.0) 196.6 ( ± 4.2) 162.5 ( ± 10.5) ± ± ± ± ± 04.20 02.50 02.00 03.82 04.00 ( ± 1.5) ( ± 0.7) ( ± 0.5) ( ± 0.5) ( ± 1.0) 39 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Fig. 6. Dynamic water contact angles of the composite membrane and the representative droplet image. 15 (Table 2). The surface roughness also plays a key role in imparting superoleophobic character to the membrane surface and thereby deterring membrane fouling by oils. The fillers increase the surface roughness of the composite membranes causing water to be trapped in the micro/ nanostructures, whereby preventing wetting of the membrane surface with oil due to the Cassie-Baxter state existing at the membrane-wateroil interphase as evidenced by other researchers [52,53]. The intrinsic hydrophilicity of the membranes can be understood by water permeation through the membranes. The pure water fluxes (Table 2) were measured after pre-compacting the membranes at 0.35 MPa for 40 min until a steady state flux was observed. The pure membrane exhibited the lowest flux, which could be due to the fact that majority of the pores, as well as the surface, is hydrophobic. The membranes S1-S4 exhibit greater pure water permeability as the presence of the polydopamine decorated SBA-15 lines the surfaces as well as the pores of the membranes making it highly hydrophilic. Coating of membrane directly with polydopamine usually leads to a decline of water flux as the top surface pores often are blocked by the dopamine aggregates as evidenced in the literature [50,54]. The pDA decorated mesoporous silica fillers are well dispersed into the membrane matrix and no reduction in flux was observed due to aggregation of these fillers up to 0.075 wt.% loading. The migration of fillers to the membrane top surface enriches it with hydroxyl and amine groups and these fillers orient themselves to attract water molecules forming a hydration layer and thereby improving the water permeability. The rougher surface of the composite membranes further promotes trapping of water molecules on the membrane surface. Moreover, the fillers present in the dope solution is responsible for the formation of a more porous structure as evidenced by the porosity studies and SEM images. addition to membrane surface roughness, the surface free energy and chemical nature also play a prominent role in anti-fouling ability. Our results concur with some earlier works [8,50], which report an increase in underwater oleophobicity due to an increase in surface roughness. 3.2.3. Membrane wettability and water permeability of the composite membranes The wettability of the prepared membranes by water was evaluated by both dynamic water contact angle (WCA) as well as membrane water uptake (seen in earlier section). The low wettability of the membrane S0 is due to the inherent hydrophobicity of the polyetherimide polymer. As predicted in light of earlier research, improving the membrane hydrophilicity can impart exceptional oleophobicity under water. The static water contact angles of the prepared membranes are given in Table 2. Clearly, the WCA of the pDA-SBA-15 incorporated membranes is much lower than the pristine membrane. It indicates that these fillers provide a greater hydrophilicity to the membrane surface. The surface, as well as the pores, are lined with these fillers containing large amounts of hydroxyl and amine functional groups, which play an important role in forming a hydration layer with water via a hydrogen bonding. The work of adhesion increases as the surface wettability increases, which implies a greater amount of force is required to remove water from its surface. It indicates that the composite membranes possess a lower fouling tendency compared to pristine membrane due to the higher energy requirement in displacing water molecules. The dynamic WCAs recorded for 60 s (Fig. 6) further demonstrates the effect of incorporation of the polydopamine decorated mesoporous silica on membrane hydrophilicity. The pristine membrane showed only a marginal reduction in WCA even after 60 s compared to the composite membranes which displayed a near zero contact angle by 20 s (S3). The highly hydrophilic membrane S3 exhibited a water contact angle of 37° and according to Young’s equation, this membrane can be oleophilic (θ ≤ 90°) as the water surface tension is generally more than that of oil (in air). However, in the actual application as well as in the evaluation of the underwater oil contact angle (UWOCA), the membrane-air-liquid interface is replaced by the membrane-water-liquid interface. The hydrophilic membrane surface, having a greater liquid interfacial tension γlg(water) becomes oleophobic [51]. The UWOCAs of the composite membranes ranged from ∼89° to ∼150° indicating underwater superoleophobicity is imparted by the pDA-SBA- 3.2.4. Performance evaluation of the composite membranes using simulated feed Motor oil was used as the target oil for the preparation of synthetic oil-in-water emulsion as it is a significant waste product discarded in most automotive workshops. It can be understood that the permeate flux of the emulsion can increase with an increase in transmembrane pressure but drastically decreases with oil content. The feed solution used in this study contained 20 ppm of motor oil stabilized by SDS. The 40 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. droplets or surfactant on the membrane surface. The recyclability of membrane S3 was investigated by 10 filtration cycles of 100 ml feed followed by washing. The concentration of oil and the permeate flux was carried were evaluated using membrane S3 as shown in Fig. 8a. Membrane S3 indicated an excellent recyclability for the short-term test; as there was almost complete solution flux recovery for each run while maintaining an oil rejection efficiency of > 99.6%. The prepared composite membranes have great stability, excellent hydrophilicity, and superior underwater superoleophobicity. It can also be highly effective for long-term operations. The long-term fluxes of the prepared membranes were investigated with a thorough backwash every 4 h (Fig. 8b). Membranes S2 and S3 show remarkable pure water flux recovery ratio even after four cycles. These membranes also show a lower flux decline compared to the pristine membranes. The flux decline during filtration can arise due to concentration polarization at the membrane surface, cake layer formation, or the presence of a thin oil layer near the membrane top-surface or due to reversible/irreversible adsorption of the oil droplets/surfactant chains. The first two of these factors are dominated by hydrodynamics in the membrane module and can be tackled by optimizing operating conditions and careful module design. Flux recovery upon a water rinse, allows us to identify the contribution of adsorptive, irreversible membrane fouling, which is a mainly controlled by membrane surface chemistry. The excellent flux recovery, as well as the slight increase in oil rejection of the composite membranes, could also be explained by the electrostatic repulsion between the membranes and the emulsion droplets. The presence of pDA-SBA-15 on the membrane surface increases the negative surface charge on the membrane which can hinder attachment of the negatively charged anionic surfactant micelle. initial permeate fluxes increased from ∼ 2 LMH to ∼162 LMH on the incorporation of the pDA decorated mesoporous silica. The permeate flux is in congruence with the pure water flux results which attributed to the increase in membrane hydrophilicity as well as improvement of membrane porosity. The pDA coating on the SBA-15 is said to be stable in acidic, neutral and weakly alkaline solution [55]. This ensures that the membranes remain hydrophilic for most industrially relevant feeds. Transmembrane pressure was found to improve the permeate flux. However, this might lead to permeation of oil through the membrane. The operating pressure of 0.2 MPa is well below the breakthrough pressure. These composite membranes were having smaller pore openings in dozens of nanometers were still smaller than the emulsified oil droplet sizes (hundreds of nanometers). The pDA-SBA-15 incorporated composite membranes used in the cross-flow ultrafiltration system combine de-emulsification as well as separation into a single unit process. The feed solution of the oil-in-water emulsion is milky (Fig. 7a) attributed to the presence of stabilized micro-droplets. The sizes of the droplets were found to be in the range of 600–800 nm as observed by the dynamic-light scattering (Fig. 7b). The oil-emulsion selectivity of the prepared membranes was evaluated by the TOC analysis of the permeate. All membranes have a selectivity > 99.4% and membranes S2 and S3 exhibited the greatest selectivity against the permeation of oil while maintaining the highest permeate flux (Fig. 7c and d). This could be due to the larger number of ‘smaller-sized’ pore openings on the membrane surface created by the presence of the polymer poor phase (pDA-SBA-15) on the membrane surface. The increase in the surface roughness causes improvement in the rate of permeation of water for the composite membranes without allowing the oil molecules to come in contact with the membrane surface. A reduction in flux is generally observed due to the absorption of oil Fig. 7. Performance of the composite membranes using synthetic oil-in-water emulsion. Optical images of F-feed and P-permeate (a); size distribution of the emulsions using a particle size analyzer (DLS) (b); Solution permeate flux (c); oil selectivity (d). 41 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. Fig. 8. Recyclability of performance of S3 for fixed volume of feed (a); Long-term normalized solution flux recovery of the composite membranes (b) using synthetic oil-in-water emulsion. the suspended siliceous particles were easily separated, however, the dissolved ions could not be removed as these were ultrafiltration membranes. The oil and grease removal by the membranes were analyzed by the partition-gravimetric method as well as the TOC analysis. The organic carbon present in the permeate could be the smaller sized humic/fulvic acid moieties. The oil and grease content was reduced to below 10 ppm, which is below the industrial discharge limits. Leaching studies carried out for 30 days showed no loss of pDA-SBA-15 from the composite membranes. The above results indicate that the composite membranes provide considerable permeation rate with sufficient oil rejection efficiency as well as excellent anti-fouling ability. Table 3 summarizes a comparative study on the composite ultrafiltration and microfiltration membranes fabricated by electrospinning and phase inversion techniques. 3.3. Performance evaluation of the composite membranes using produced water The efficiency of the composite membranes for the removal of the oil from produced water was investigated at 0.2 MPa. Table S2 summarizes characterization of feed solution. The permeate fluxes of the membranes were slightly lower compared to that of simulated feed due to the presence of organic matter and other ions present in the produced water. The irreversible attenuation of the flux is could be due to the pore plugging effects of the various silica materials or oil present in the feed. However, these membranes showed only a small flux decline of ∼15% even after 9 h of operation where the volume of feed permeated is greater for the available membrane surface area (Fig. 9). Fux et al. have discussed three main characteristic stages of membrane fouling by oil: (1) droplet attachment and clustering, (2) axial droplet deformation, and (3) droplet coalescence [56]. It is observed that the composite membrane have greater flux and more stable performance during long-term ultrafiltration operation. This implies that the pDA-SBA-15 materials can repel adhesion of oil drops as well as humic foulants present in the feed. The higher flux can be attributed to the presence of greater number of pore openings, pore structure and the chemical nature (hydrophilicity) of the pores. We observed that most of 3.4. Bacterial anti-adhesion on the composite membranes Bio-fouling is another major challenge faced during ultrafiltration operations. A severe fouling of the membranes could occur due to the adhesion of the bacteria and subsequent biofilm formation. This results in (i) an additional mass transfer resistance due to the foulant layer, making it imperative to increase the operating pressures to achieve desired constant flux and the subsequent problems on higher pressure operations; (ii) formation of a cake layer which promotes concentration polarization leading to a significant decline in permeate flux [57]. The composite membranes were investigated for their efficiency to resist the attachment of a gram-negative bacterium Pseudomonas aeruginosa and gram-positive bacterium, Bacillus subtilis. In this work, the anti-adhesion efficiency of the membranes was carried out quantitatively in a dynamic contact condition. The DLVO (Derjaguin, Verwey, Landau and Overbeek) theory has been used to describe the net interaction between a bacterial cell and the membrane surface as a sum of the attractive van der Waals force and the repulsive force due to the presence of an electrical double layer between the cell and the membrane surface (both usually negatively charged). The interaction between these forces can account for Gibb’s free energy of adsorption [58]. It is also widely reported that improvement of hydrophilicity of membrane surfaces improves its potential to hinder the biofouling of the membranes. The reports of the anti-bacterial ability of polydopamine coated surfaces vary from moderate to excellent [59]. The dop-SBA15 imparts excellent membrane hydrophilicity as seen by the WCA and pure water flux studies. Fig. 9. Long term (9 h) permeation studies using produced water as feed and with 3 intermittent backwashes. 42 Modifier Incorporation Water Flux (LMH) Contact Angle Oil rejection (%) Emulsion Flux FRR PES Hydrous manganese oxide (HMO) and titanium dioxide (TiO2) Blending (NIPS) 28.48 LMH (1 bar) Water CA 65° to 15° in 22 s (dynamic contact angle 31.73% increase Nylon-6/SiO2 (MF) PVA 1348 LMH (0.28 bar) Thermoresponsive poly(Nisopropylacrylamide) (PNIPAM) hydrogel Water CA 21° (1 s to be 0°) UWOCA- 116° WCA 0° (room temperature) and 150.2° (45 °C) Water flux recovery rate (91.5%) 85% Thermoplastic polyurethane (MF) Electrospinning of the base polymer and then coating by NIPS Force spinning followed free-radical polymerization and subsequent coating of the hydrogen 98.57 (50 ppm solution) for 2000 ppm solution 97.5% > 98.80 for 250 mg/ l, 500 and 2000 (0.28 bar) Silicon oil-water > 99% at both temp WCA 0° WOCA- 155° Isooctane and hexadecane in water permeate < 10 ppm oil Surfactant free emulsion2000 LMH under 1 bar for surfactant stabilized – 372 LMH for Tween 80isooctane-in-H2O emulsion and 315 LMH for Tween 80hexadecane-in-H2O emulsion Permeation flux for SDS/ petroleum ether/H2O emulsion, SDS/nhexadecane/H2O emulsion, SDS/1,3,5trimethylbenzene/H2O emulsion, and SDS/diesel oil/H2O emulsion is 428 LMH, 605 LMH, 524 LMH, 382 LMH, respectively In FO mode water flux decreases from 16.5 LMH to 11.5 LMH (ultrafiltration) PAN 3646 LMH Alkaline-induced phase inversion 43 PVDf (commercial) 0.22 µm TiO2 nanomaterials Attachment via functional groups present by polymerization and cross linking of dopamine and KH550 785 LMH WCA (for 1:1 dop: silane) 26.9° to 0° in 1 s UWOCA 158.6° for dichloromethane > 99.10% for all oils PSf (TFC) Amine-terminated fully disulfonated poly(arylene ethers) (NH2− BPSH100) were synthesized via step growth polymerization PSfinsitu polymerization to yield an amide layer, followed by linking the NH2− BPSH100 via the amide bond 1.36 LMH/ bar WCA 10.2° and UWOCA of 155.4° Soybean oil/water emulsion > 99.9% PVDf (commercial) Chitosan-SiO2glutaraldehyde Affixing by dip coating WCA 0° and UWOCA 150° > 99% Oil flux at 45 °C is 503–2000 LMH under 1 bar Other Significant Properties Ref. [5] Stable in pH of 4, 7, 10 [62] The membrane is superhydrophilic/ underwater oleophobic at room temperature (25 °C) and becomes superhydrophobic at a temperature over the lower critical solution temperature of PNIPAM (LCST, 32 °C) -elongation of 288% and a toughness of 7536 kJ/ m3 [63] [8] 85% for 3 cycles Very high FRR Simple and eco-friendly fabrication [64] 99.9%, 98.1%, 95.9% of the initial water flux after 1st, 2nd, and 3rd cycle Used as FO membrane; water recovery as high as 80%, the 10 k-g-TFC membrane retains 69.8% of its initial water flux Oil adsorption on membrane 68.1 ng/cm2 [65] [19] (continued on next page) Separation and Purification Technology 205 (2018) 32–47 Base polymer N.J. Kaleekkal et al. Table 3 A comparative study on membranes employed for oil-water emulsion separation. 44 Base polymer Modifier Incorporation PAN GO PVDf copolymer poly (DMAEMA-r-HEMA) PVDf (microfiltration APTES functionalized MWCNTs PAN (MF) Nanocrystalline zeolite imidazole framework (ZIF8). Polydopamine decorated SBA-15 PEI (UF) Water Flux (LMH) Contact Angle Oil rejection (%) Emulsion Flux FRR Other Significant Properties Ref. Electrospinning followed by base assisted hydrolysis of the prepared membrane WCA 30° and UWOCA 120° 99% 3500 LMH in 0.1 bar 99% [16] Blending the copolymer followed by cross linking in the coagulation bath using glutaricdialdehyde followed by sulfonation of the prepared membrane using 1,3-propane sultone Adhesion of the fMWCNTs on the PVDf membrane using selfpolymerization of polydopamine WCA 61° goes to 0° in 10 s UWOCA 156° 2 ppm in filtrate For isooctane-in-water emulsion 6350 LMH/ bar 98% WCA 26.77° decreases to 0° in 1 s UWOCA 153.8° ± 2.0° > 99% for diesel in water emulsion 886 LMH at 0.09 MPa 90% upto 6 cycles Water and oil contact angles go to 0° in 0.1 s. > 99.5% for toluene in water > 900 LMH WCA 37.2° ( ± 0.5) UWOCA 147.5° ( ± 5) > 99.8% for motor oil/water/SDS ∼162.8 LMH GO-induced spindle-knot structure and hydrophilic chemical features could work together to enhance water flux The high recovery (∼91%) of flux for separating oil-in-water emulsion after long-term continuous filtration experiment For varying oils the rejection was constant. Under strong basic conditions the membranes were unstable. Increasing oil concentrations decreased the membrane performance Superoleophobic or superhydrophobic by prewetting with water/oil Underwater oleophobic, bacterial anti-adhesion property 1282.5 LMH at 0.09 MPa Blend followed by electrospinning Blend followed by phase inversion 196.6 LMH at 0.2 MPa ∼94% N.J. Kaleekkal et al. Table 3 (continued) [10] [66] [67] Current Work Separation and Purification Technology 205 (2018) 32–47 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. mesoporous silica is impressive filler which can be employed for the preparation of modified PEI membranes which can be commercially employed. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.seppur.2018.05.007. References [1] National Oceanic and Atmospheric Administration: Environmental Response Management Application (ERMA) | response.restoration.noaa.gov < https:// response.restoration.noaa.gov/maps-and-spatial-data/environmental-responsemanagement-application-erma > , n.d. (accessed December 6, 2017). [2] ESSO-INCOIS-Indian National Centre for Ocean Information Services. Bulletin No: 4; Trajectory Prediction on Ennore Oil Spill (28.01.2017) < http://www.incois. gov.in/WEBSITE_FILES/OSF_FILES/forecast/Bulletins/Bulletinonoil_spill_ trajectory.pdf > , n.d. (accessed December 6, 2017). [3] S. Munirasu, M.A. Haija, F. Banat, Use of membrane technology for oil field and refinery produced water treatment—a review, Process Saf. Environ. Prot. 100 (2016) 183–202, http://dx.doi.org/10.1016/j.psep.2016.01.010. [4] A. Salahi, T. Mohammadi, R.M. Behbahani, M. Hemati, PES and PES/PAN blend ultrafiltration hollow fiber membranes for oily wastewater treatment: preparation, experimental investigation, fouling, and modeling, 21494 (1–16), Adv. Polym. Technol. 34 (2015), http://dx.doi.org/10.1002/adv.21494. [5] G.S. Lai, M.H.M. Yusob, W.J. Lau, R.J. Gohari, D. Emadzadeh, A.F. Ismail, P.S. Goh, A.M. Isloor, M.R.-D. Arzhandi, Novel mixed matrix membranes incorporated with dual-nanofillers for enhanced oil-water separation, Sep. Purif. Technol. 178 (2017) 113–121, http://dx.doi.org/10.1016/j.seppur.2017.01.033. [6] J. Saadati, M. Pakizeh, Separation of oil/water emulsion using a new PSf/pebax/FMWCNT nanocomposite membrane, J. Taiwan Inst. Chem. Eng. 71 (2017) 265–276, http://dx.doi.org/10.1016/j.jtice.2016.12.024. [7] C.S. Ong, W.J. Lau, P.S. Goh, B.C. Ng, T. Matsuura, A.F. Ismail, Effect of PVP molecular weights on the properties of PVDF-TiO2 composite membrane for oily wastewater treatment process, Sep. Sci. Technol. 49 (2014) 2303–2314, http://dx. doi.org/10.1080/01496395.2014.928323. [8] F. Zhang, S. Gao, Y. Zhu, J. Jin, Alkaline-induced superhydrophilic/underwater superoleophobic polyacrylonitrile membranes with ultralow oil-adhesion for highefficient oil/water separation, J. Memb. Sci. 513 (2016) 67–73, http://dx.doi.org/ 10.1016/j.memsci.2016.04.020. [9] H.P. Ngang, A.L. Ahmad, S.C. Low, B.S. Ooi, Preparation of thermoresponsive PVDF/SiO2-PNIPAM mixed matrix membrane for saline oil emulsion separation and its cleaning efficiency, Desalination 408 (2017) 1–12, http://dx.doi.org/10.1016/j. desal.2017.01.005. [10] Y. Zhu, W. Xie, F. Zhang, T. Xing, J. Jin, Superhydrophilic in-situ-cross-linked zwitterionic polyelectrolyte/PVDF-blend membrane for highly efficient oil/water emulsion separation, ACS Appl. Mater. Interfaces 9 (2017) 9603–9613, http://dx. doi.org/10.1021/acsami.6b15682. [11] A. Salahi, T. Mohammadi, R.M. Behbahani, M. Hemmati, Experimental investigation and modeling of industrial oily wastewater treatment using modified polyethersulfone ultrafiltration hollow fiber membranes, Korean J. Chem. Eng. 32 (2015) 1101–1118, http://dx.doi.org/10.1007/s11814-014-0310-1. [12] R.S. Hebbar, A.M. Isloor, K. Ananda, A.F. Ismail, C.-M. Kim, I.S. Kim, B.D. Freeman, D.R. Paul, M. Whiteley, J.S. Vrouwenvelder, Fabrication of polydopamine functionalized halloysite nanotube/polyetherimide membranes for heavy metal removal, J. Mater. Chem. A 4 (2016) 764–774, http://dx.doi.org/10.1039/ C5TA09281G. [13] B. Govardhan, S.S. Chandrasekhar, S. Sridhar, Purification of surface water using novel hollow fiber membranes prepared from polyetherimide/polyethersulfone blends, J. Environ. Chem. Eng. 5 (2017) 1068–1078, http://dx.doi.org/10.1016/j. jece.2017.01.027. [14] E. Tuncer, Electrical properties of polyetherimide thin films: Nonparametric dielectric reponse analysis with distribution of relaxation times. < http://arxiv.org/ abs/1304.5536 > , 2013 (accessed May 1, 2017). [15] P. Kanagaraj, A. Nagendran, D. Rana, T. Matsuura, S. Neelakandan, K. Malarvizhi, Effects of polyvinylpyrrolidone on the permeation and fouling-resistance properties of polyetherimide ultrafiltration membranes, Ind. Eng. Chem. Res. 54 (2015) 4832–4838, http://dx.doi.org/10.1021/acs.iecr.5b00432. [16] J. Zhang, X. Pan, Q. Xue, D. He, L. Zhu, Q. Guo, Antifouling hydrolyzed polyacrylonitrile/graphene oxide membrane with spindle-knotted structure for highly effective separation of oil-water emulsion, J. Memb. Sci. 532 (2017) 38–46, http:// dx.doi.org/10.1016/j.memsci.2017.03.004. [17] Y. Peng, F. Guo, Q. Wen, F. Yang, Z. Guo, A novel polyacrylonitrile membrane with a high flux for emulsified oil/water separation, Sep. Purif. Technol. 184 (2017) 72–78, http://dx.doi.org/10.1016/j.seppur.2017.04.036. [18] Y. Liu, Y. Su, Y. Li, X. Zhao, Z. Jiang, Z. Jiang, Improved antifouling property of PVDF membranes by incorporating an amphiphilic block-like copolymer for oil/ water emulsion separation, RSC Adv. 5 (2015) 21349–21359, http://dx.doi.org/10. 1039/C4RA16290K. [19] J. Liu, P. Li, L. Chen, Y. Feng, W. He, X. Lv, Modified superhydrophilic and underwater superoleophobic PVDF membrane with ultralow oil-adhesion for highly Fig. 10. Bacterial adhesion challenge onto the composite membranes. During the membrane preparation, surface tension is lowered by the preferential migration of the hydrophilic fillers towards the membrane surfaces thus hindering bacterial adhesion. The lower adhesion seen for the composite membranes can be explained by the membrane hydrophilicity as well as the negative surface charge. Our bacterial anti-adhesion results (Fig. 10) are in concurrence with earlier studies employing a direct polydopamine coating on the membrane surface [60] or by incorporation of dopamine coated halloysite tubes [61]. The polydopamine is also reported to be anti-bacterial in nature and is widely understood that the protonated amine groups present is responsible for cleavage of the bacterial wall during operation as it forms a contact-active antibacterial surface over a wide range of pH [60]. 4. Conclusions The present work deals with the synthesis of novel composite PEI ultrafiltration membranes by phase inversion technique using polydopamine decorated mesoporous SBA-15 as the hydrophilic filler. The prepared SBA-15 had a surface area of 745 m2/g which decreased to 672 m2/g indicating the successful deposition of pDAon the pristine mesoporous material and it was further confirmed by SEM imaging and FTIR and XPS spectroscopy. HR-TEM micrograph revealed the structural integrity was intact even after the pDA coating with a fine coating present even within the inner pores. The fillers were uniformly dispersed in the membrane matrix and it was observed that during membrane formation the fillers migrated towards the membrane surface to reduce the surface tension. The membrane S3 had a flux 196 LMH at 0.2 MPa compared to the pristine membrane which had a flux of 89 LMH at the same operating conditions. This could be explained by the thinner skin layer with the formation of fully developed pores as evidenced by SEM imaging and the increase in hydrophilicity observed by lower water contact angles. The surface chemistry and the small increase in surface roughness imparted underwater oleophobicity to the prepared membranes. The composite membranes showed a > 99% removal of oil from a synthetic emulsion. Moreover, the membranes showed no leaching of the filler even up to 30 days and were stable in the operating pH. These composite membranes also demonstrated an excellent recyclability and long-term anti-fouling ability when challenged with produced water. The pDA-SBA-15 also reduced the bacterial adhesion to > 90% for membranes S2 and S3 owing to the hydrophilicity, surface charge, and surface composition. Moreover, dopamine can act as a bio-inspired adhesive and can be used to adhere metal nanoparticles or organic moieties onto the mesoporous silica to impart specific functional properties for suitable applications- bactericidal property, photocatalytic property etc. Overall the pDA decorated 45 Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] efficient oil/water emulsion separation, Mater. Lett. 185 (2016) 169–172, http:// dx.doi.org/10.1016/j.matlet.2016.08.124. B. Wang, W. Liang, Z. Guo, W. Liu, S.A. Mazzella, G.C. Rutledge, G.H. McKinley, R.E. Cohen, L. Jiang, Y.Q. Wen, X.J. Zhang, S.T. Wang, H. Zhang, Biomimetic superlyophobic and super-lyophilic materials applied for oil/water separation: a new strategy beyond nature, Chem. Soc. Rev. 44 (2015) 336–361, http://dx.doi.org/10. 1039/C4CS00220B. J. Jiang, L. Zhu, L. Zhu, B. Zhu, Y. Xu, Surface characteristics of a self-polymerized dopamine coating deposited on hydrophobic polymer films, Langmuir 27 (2011) 14180–14187, http://dx.doi.org/10.1021/la202877k. A. Martín, J.M. Arsuaga, N. Roldán, A. Martínez, A. Sotto, Effect of amine functionalization of SBA-15 used as filler on the morphology and permeation properties of polyethersulfone-doped ultrafiltration membranes, J. Memb. Sci. 520 (2016) 8–18, http://dx.doi.org/10.1016/j.memsci.2016.07.040. C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Ordered mesoporous molecular sieves synthesized by a liquid-crystal template mechanism, Nature 359 (1992) 710–712, http://dx.doi.org/10.1038/359710a0. G. of I. Ministry of Urban Development, CPHEEO, 2012. Manual on Sewerage and Sewage Treatment, Part A: Engineering Final Draft. Central Public Health and Environmental Engineering Organisation, Ministry of Urban Development, New Delhi - Google Search. < https://www.google.com.kw/search?q=CPHEEO %2C+2012.+Manual+on+Sewerage+and+Sewage+Treatment%2C+Part +A%3A+Engineering+Final+Draft.+Central+Public+Health+and +Environmental+Engineering+Organisation%2C+Ministry+of+Urban +Development%2C+New+Delhi&rlz=1C1RLNS_enI > , n.d. (accessed May 10, 2017). F. Rehman, P.L.O. Volpe, C. Airoldi, Free amino and imino-bridged centres attached to organic chains bonded to structurally ordered silica for dye removal from aqueous solution, J. Environ. Manage. 133 (2014) 135–143, http://dx.doi.org/10. 1016/j.jenvman.2013.11.042. J.-K. Gao, L.-A. Hou, G.-H. Zhang, P. Gu, Facile functionalized of SBA-15 via a biomimetic coating and its application in efficient removal of uranium ions from aqueous solution, J. Hazard. Mater. 286 (2015) 325–333, http://dx.doi.org/10. 1016/j.jhazmat.2014.12.061. R. Ghorbani-Vaghei, S. Hemmati, H. Veisi, Pd immobilized on amidoxime-functionalized Mesoporous SBA-15: a novel and highly active heterogeneous catalyst for Suzuki-Miyaura coupling reactions, J. Mol. Catal. A Chem. 393 (2014) 240–247, http://dx.doi.org/10.1016/j.molcata.2014.06.025. J. Tao, J. Xiong, C. Jiao, D. Zhang, H. Lin, Y. Chen, Hybrid mesoporous silica based on hyperbranch-substrate nanonetwork as highly efficient adsorbent for water treatment, ACS Sustain. Chem. Eng. 4 (2016) 60–68, http://dx.doi.org/10.1021/ acssuschemeng.5b00652. A. Vinu, P. Srinivasu, M. Takahashi, T. Mori, V.V. Balasubramanian, K. Ariga, Controlling the textural parameters of mesoporous carbon materials, Micropor. Mesopor. Mater. 100 (2007) 20–26, http://dx.doi.org/10.1016/J.MICROMESO. 2006.10.008. L.-L. Li, H. Sun, C.-J. Fang, J. Xu, J.-Y. Jin, C.-H. Yan, Optical sensors based on functionalized mesoporous silica SBA-15 for the detection of multianalytes (H+ and Cu2+) in water, J. Mater. Chem. 17 (2007) 4492, http://dx.doi.org/10.1039/ b708857d. Q. Yao, Z.-H. Lu, K. Yang, X. Chen, M. Zhu, Ruthenium nanoparticles confined in SBA-15 as highly efficient catalyst for hydrolytic dehydrogenation of ammonia borane and hydrazine borane, Sci. Rep. 5 (2015) 15186, http://dx.doi.org/10. 1038/srep15186. H. Luo, C. Gu, W. Zheng, F. Dai, X. Wang, Z. Zheng, Facile synthesis of novel sizecontrolled antibacterial hybrid spheres using silver nanoparticles loaded with polydopamine spheres, RSC Adv. 5 (2015) 13470–13477, http://dx.doi.org/10.1039/ C4RA16469E. K.S.W. Sing, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity (recommendations 1984), Pure Appl. Chem. 57 (1985) 603–619, http://dx.doi.org/10.1351/ pac198557040603. N. Gargiulo, A. Peluso, P. Aprea, F. Pepe, D. Caputo, CO 2 adsorption on polyethylenimine-functionalized SBA-15 mesoporous silica: isotherms and modeling, J. Chem. Eng. Data. 59 (2014) 896–902, http://dx.doi.org/10.1021/je401075p. L. Yang, B. Wang, S. Lai, C. Jiang, H. Zhong, Enhancing photocatalytic degradation of phenol through nitrogen- and nitrogen/fluorine-codoped Ti-SBA-15, RSC Adv. 5 (2015) 53299–53305, http://dx.doi.org/10.1039/C5RA09922F. A.C.C. Chang, S.S.C. Chuang, M. Gray, Y. Soong, In-situ infrared study of CO2 adsorption on SBA-15 grafted with γ-(aminopropyl)triethoxysilane, Energy Fuels 17 (2003) 468–473, http://dx.doi.org/10.1021/ef020176h. B.H. Kim, D.H. Lee, J.Y. Kim, D.O. Shin, H.Y. Jeong, S. Hong, J.M. Yun, C.M. Koo, H. Lee, S.O. Kim, Mussel-inspired block copolymer lithography for low surface energy materials of teflon, graphene, and gold, Adv. Mater. 23 (2011) 5618–5622, http://dx.doi.org/10.1002/adma.201103650. M.C. Burrell, J.J. Chera, Polyetherimide (Ultem®) Spin Cast Films by XPS, < http:// oasc12039.247realmedia.com/RealMedia/ads/click_lx.ads/test.int.aip.org/adtest/ L23/2102859201/x01/AIP/Hiden_SSSCovAd_1640x440Banner_01_2016thru12_ 2016/1640x440px-BANNER-AD-GENERAL-26582.jpg/ 33722b58693165736c486741447a3563?x > , 1999. http://doi.org/10.1116/1. 1247898. Y. Manawi, V. Kochkodan, A.W. Mohammad, M. Ali Atieh, Arabic gum as a novel [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] 46 pore-forming and hydrophilic agent in polysulfone membranes, J. Memb. Sci. 529 (2017) 95–104, http://dx.doi.org/10.1016/j.memsci.2017.02.002. R.S. Hebbar, A.M. Isloor, K. Ananda, M.S. Abdullah, A.F. Ismail, Y. Dong, Y. Zhang, Y. Liang, J. Liu, Fabrication of a novel hollow fiber membrane decorated with functionalized Fe2O3 nanoparticles: towards sustainable water treatment and biofouling control, New J. Chem. 41 (2017) 4197–4211, http://dx.doi.org/10.1039/ C7NJ00221A. N.O. Rahmati, M. Pourafshari Chenar, H. Azizi Namaghi, Removal of free active chlorine from synthetic wastewater by MEUF process using polyethersulfone/titania nanocomposite membrane, Sep. Purif. Technol. 181 (2017) 213–222, http:// dx.doi.org/10.1016/j.seppur.2017.03.030. M.L. Masheane, L.N. Nthunya, S.P. Malinga, E.N. Nxumalo, B.B. Mamba, S.D. Mhlanga, Synthesis of Fe-Ag/f-MWCNT/PES nanostructured-hybrid membranes for removal of Cr(VI) from water, Sep. Purif. Technol. 184 (2017) 79–87, http://dx.doi.org/10.1016/j.seppur.2017.04.018. N.J. Kaleekkal, A. Thanigaivelan, D. Rana, D. Mohan, Studies on carboxylated graphene oxide incorporated polyetherimide mixed matrix ultrafiltration membranes, Mater. Chem. Phys. 186 (2017) 146–158, http://dx.doi.org/10.1016/j. matchemphys.2016.10.040. C. Liao, J. Zhao, P. Yu, H. Tong, Y. Luo, Synthesis and characterization of SBA-15/ poly (vinylidene fluoride) (PVDF) hybrid membrane, Desalination 260 (2010) 147–152, http://dx.doi.org/10.1016/j.desal.2010.04.050. X. Shen, T. Xie, J. Wang, P. Liu, F. Wang, B. Chu, B.M. Zhou, X.M. Qian, Z. Wang, X.M. Qian, An anti-fouling poly(vinylidene fluoride) hybrid membrane blended with functionalized ZrO2 nanoparticles for efficient oil/water separation, RSC Adv. 7 (2017) 5262–5271, http://dx.doi.org/10.1039/C6RA26651G. H. Wu, B. Tang, P. Wu, Development of novel SiO2–GO nanohybrid/polysulfone membrane with enhanced performance, J. Memb. Sci. 451 (2014) 94–102, http:// dx.doi.org/10.1016/j.memsci.2013.09.018. G. Zeng, Y. He, Y. Zhan, L. Zhang, H. Shi, Z. Yu, Preparation of a novel poly(vinylidene fluoride) ultrafiltration membrane by incorporation of 3-aminopropyltriethoxysilane-grafted halloysite nanotubes for oil/water separation, Ind. Eng. Chem. Res. 55 (2016) 1760–1767, http://dx.doi.org/10.1021/acs.iecr.5b04797. F.S. Halek, S.K. Farahani, S.M. Hosseini, Fabrication of poly(ether sulfone) based mixed matrix membranes modified by TiO2 nanoparticles for purification of biodiesel produced from waste cooking oils, Korean J. Chem. Eng. 33 (2016) 629–637, http://dx.doi.org/10.1007/s11814-015-0161-4. J. Li, Z. Zhao, D. Li, H. Tian, F. Zha, H. Feng, L. Guo, Smart candle soot coated membranes for on-demand immiscible oil/water mixture and emulsion switchable separation, Nanoscale 9 (2017) 13610–13617, http://dx.doi.org/10.1039/ C7NR04448H. Z. Wang, X. Jiang, X. Cheng, C.H. Lau, L. Shao, Mussel-inspired hybrid coatings that transform membrane hydrophobicity into high hydrophilicity and underwater superoleophobicity for oil-in-water emulsion separation, ACS Appl. Mater. Interfaces 7 (2015) 9534–9545, http://dx.doi.org/10.1021/acsami.5b00894. L. Hu, S. Gao, Y. Zhu, F. Zhang, L. Jiang, J. Jin, An ultrathin bilayer membrane with asymmetric wettability for pressure responsive oil/water emulsion separation, J. Mater. Chem. A 3 (2015) 23477–23482, http://dx.doi.org/10.1039/C5TA03975D. H.-C. Yang, J.-K. Pi, K.-J. Liao, H. Huang, Q.-Y. Wu, X.-J. Huang, Z.-K. Xu, Silicadecorated polypropylene microfiltration membranes with a mussel-inspired intermediate layer for oil-in-water emulsion separation, ACS Appl. Mater. Interfaces 6 (2014) 12566–12572, http://dx.doi.org/10.1021/am502490j. P.-C. Chen, Z.-K. Xu, L.L. Zheng, L.J. Wang, Z.Y. Jiang, Mineral-coated polymer membranes with superhydrophilicity and underwater superoleophobicity for effective oil/water separation, Sci. Rep. 3 (2013) 2776, http://dx.doi.org/10.1038/ srep02776. Z.-X. Wang, C.-H. Lau, N.-Q. Zhang, Y.-P. Bai, L. Shao, M. Shan, Y. Li, C. Yang, Z. Wang, X. Qian, A.A. Golriz, R. Berger, M. Wagner, A. del Campo, Mussel-inspired tailoring of membrane wettability for harsh water treatment, J. Mater. Chem. A 3 (2015) 2650–2657, http://dx.doi.org/10.1039/C4TA05970K. H.-C. Yang, K.-J. Liao, H. Huang, Q.-Y. Wu, L.-S. Wan, Z.-K. Xu, Mussel-inspired modification of a polymer membrane for ultra-high water permeability and oil-inwater emulsion separation, J. Mater. Chem. A 2 (2014) 10225–10230, http://dx. doi.org/10.1039/C4TA00143E. G. Fux, G.Z. Ramon, Microscale dynamics of oil droplets at a membrane surface: deformation, reversibility, and implications for fouling, Environ. Sci. Technol. 51 (2017) 13842–13849, http://dx.doi.org/10.1021/acs.est.7b03391. V. Kochkodan, D.J. Johnson, N. Hilal, Polymeric membranes: surface modification for minimizing (bio)colloidal fouling, Adv. Colloid Interface Sci. 206 (2014) 116–140, http://dx.doi.org/10.1016/J.CIS.2013.05.005. M. Hermansson, The DLVO theory in microbial adhesion, Colloids Surf. B Biointer. 14 (1999) 105–119, http://dx.doi.org/10.1016/S0927-7765(99)00029-6. L. Su, Y. Yu, Y. Zhao, F. Liang, X. Zhang, Strong antibacterial polydopamine coatings prepared by a shaking-assisted method, Sci. Rep. 6 (2016) 24420, http://dx. doi.org/10.1038/srep24420. H. Karkhanechi, R. Takagi, H. Matsuyama, Biofouling resistance of reverse osmosis membrane modified with polydopamine, Desalination 336 (2014) 87–96, http:// dx.doi.org/10.1016/J.DESAL.2013.12.033. R.S. Hebbar, A.M. Isloor, K. Ananda, A.F. Ismail, Fabrication of polydopamine functionalized halloysite nanotube/polyetherimide membranes for heavy metal removal, J. Mater. Chem. A 4 (2016) 764–774, http://dx.doi.org/10.1039/ C5TA09281G. Separation and Purification Technology 205 (2018) 32–47 N.J. Kaleekkal et al. [65] X. Zhang, J. Tian, S. Gao, Z. Zhang, F. Cui, C.Y. Tang, In situ surface modification of thin film composite forward osmosis membranes with sulfonated poly(arylene ether sulfone) for anti-fouling in emulsified oil/water separation, J. Memb. Sci. 527 (2017) 26–34, http://dx.doi.org/10.1016/j.memsci.2017.01.002. [66] X. Yang, Y. He, G. Zeng, X. Chen, H. Shi, D. Qing, F. Li, Q. Chen, Bio-inspired method for preparation of multiwall carbon nanotubes decorated superhydrophilic poly(vinylidene fluoride) membrane for oil/water emulsion separation, Chem. Eng. J. 321 (2017) 245–256, http://dx.doi.org/10.1016/j.cej.2017.03.106. [67] Y. Cai, D. Chen, N. Li, Q. Xu, H. Li, J. He, J. Lu, Nanofibrous metal–organic framework composite membrane for selective efficient oil/water emulsion separation, J. Memb. Sci. 543 (2017) 10–17, http://dx.doi.org/10.1016/j.memsci.2017.08. 047. [62] M.S. Islam, J.R. McCutcheon, M.S. Rahaman, A high flux polyvinyl acetate-coated electrospun nylon 6/SiO2 composite microfiltration membrane for the separation of oil-in-water emulsion with improved antifouling performance, J. Memb. Sci. 537 (2017) 297–309, http://dx.doi.org/10.1016/j.memsci.2017.05.019. [63] R. Ou, J. Wei, L. Jiang, G.P. Simon, H. Wang, Robust thermoresponsive polymer composite membrane with switchable superhydrophilicity and superhydrophobicity for efficient oil-water separation, Environ. Sci. Technol. 50 (2016) 906–914, http://dx.doi.org/10.1021/acs.est.5b03418. [64] H. Shi, Y. He, Y. Pan, H. Di, G. Zeng, L. Zhang, C. Zhang, A modified mussel-inspired method to fabricate TiO2 decorated superhydrophilic PVDF membrane for oil/water separation, J. Memb. Sci. 506 (2016) 60–70, http://dx.doi.org/10.1016/j.memsci. 2016.01.053. 47