University Malaysia Terengganu, Kuala Terengganu, Malaysia 2 Department of Chemical Science, Faculty of Science and Technology

Transcription

1 International Journal of Mechanical and Materials Engineering (IJMME), Vol.6 (2011), No.2, FOULING CONTROL STRATEGY OF UF-PES MEMBRANE USING THE SURFACE MODIFICATON TECHNIQUE A. Nora aini 1, M. T. S. Syaima 1 and M. Mazidah 2 1 Department of Engineering Science, Faculty of Science and Technology University Malaysia Terengganu, Kuala Terengganu, Malaysia 2 Department of Chemical Science, Faculty of Science and Technology University Malaysia Terengganu, Kuala Terengganu, Malaysia noraaini@umt.edu.my Received 14 December 2010, Accepted 17 July 2011 ABSTRACT In this study, the ultrafiltration polyethersulfone (UF-PES) membrane was modified by using a simple dip coating method, utilizing myoglobin as the coating reagent. This modification is done to render the membrane surface with hydrophilic properties, subsequently made the membrane more invulnerable to fouling. The ph of the coating solution was fixed to ph 7.0 and the concentration of myoglobin was varied to 30, 50 and 70 mg/l. Based on these, membranes were coded as MPH7-30, MPH7-50 and MPH7-70. These membranes were subjected to surface characterization (FTIR, contact angle and surface charge) and morphological analysis (SEM and MWCO). For all categorization, native membrane was used as control. The hydraulic stability of the membrane was estimated from the pure water flux assessment. Finally, all membranes were subjected to fouling analysis to further confirm its fouling durability and the effect of membrane as well as solute properties (feed concentration) on the fouling attribute were evaluated. Keywords: Ultrafiltration; Surface modification; Myoglobin; Hydrophilicity; Coating; Lysozyme. 1. INTRODUCTION In recent years, UF has become a tremendous tool for protein separation and fractionation, largely replacing the conventional methods (Susanto et al., 2009). However, despite the advantages, UF yet suffers from membrane fouling, thus restricting its practical applications is membrane fouling (Huisman et al., 2000). Fouling mainly caused the reversible adsorption of solutes on the membrane surface, consequently progressing to irreversible adsorption and cake build-up and leading to a flux decline over time (Ba et al., 2010). Since the key issues of fouling are the irreversible adsorption of foulants on the membrane surface, the properties of membrane surface must be altered to minimize the adsorptive interactions of these components with the membrane surface (Susanto and Ulbricht, 2009; Reddy et al., 2003; Shi et al., 2008). Among the approaches towards overcoming the fouling problem, membrane surface modification has shown its unique advantage in terms of understanding and affecting the fouling process at molecular 201 level. Surface modification is the most versatile way to change the surface properties of the membrane while preserving it macroporous structure. For years, many researchers had extensively devised various strategies to modify membrane surfaces, which can be generally divided into two groups which are: grafting and coating. Compared to grafting which consumes a large production cost, coating is more favourable, owing to its easy implementations since it can be performed in existing membrane installations, without requiring any additional processing (Ba et al., 2010). In this research, an investigation of surface modification of ultrafiltration polyethersulfone membrane (UF-PES) using myoglobin as the modification material was carried out. Coating solution concentration (i.e. myoglobin) was selected as the parameter of study and was varied in the pre-determined ranges. The influence of this surface modification on the membrane characteristics and fouling properties were later evaluated. 2. METHODOLOGY 2.1 Materials Polyethersulfone, PES (Manufacturer: Amoco, USA) was utilized as membrane-forming polymer and 1-Methyl-2- Pyrrolidone, NMP (Manufacturer: Fluka, Germany) was used as solvent. Myoglobin (from horse heart, MW 17kDa, pi = 7.0) was used as the coating material in surface modification process. For molecular weight cut-off (MWCO) estimation, lysozyme (14 kda), trypsin (23 kda), pepsin (35 kda), ovalbumin (45kDa) and bovine serum albumin (BSA) with MW of 67 kda were implied. All proteins were supplied by Sigma Aldrich and were well dispersed in the respective buffer prior to any application. Other organic solvents were reagent grade and used without further purification. 2.2 Apparatus and procedure Membrane fabrication Membranes were fabricated by the phase inversion techniques using the semi-automated electrically casting machine (Manufacturer: Atrish Techinic Services Sdn. Bhd). The prepared dope (consist of 15% PES and 85% NMP) was then poured onto a glass plate and spread with a

2 casting knife. The glass plate was quenched immediately in a first coagulation bath to induce polymer precipitation. The formed membrane membranes were then post-treated in methanol as the second coagulation bath to ensure the excess solvent, water soluble polymer were totally removed and to strengthen the molecules structure built in the membrane Membrane surface modification Myoglobin solution was fixed to ph7.0 and was varied to three different concentrations of 30, 50 and 70 mg/l. The clean membrane was then dipped in the prepared myoglobin solution for 12 hours to ensure the formation of myoglobin film onto the outer layer of the membrane. After the immersion process, all coated membranes were air dried and compacted for until the steady state flux was reached. With regard to these, membranes were designated as MPH7-30, MPH7-50 and MPH7-70 with regard to the concentration of the coating solution. 2.3 Surface characterization Surface chemical composition evaluation Each individual coated membrane was subjected to Fourier Transform Infrared Spectroscopy (FTIR) analysis for the determination of surface chemical composition. FTIR is an important tool able to identify the types of chemicals from its chemical bonds (i.e. functional groups) (Ahmad et al., 2010). Thus, the successfulness of myoglobin coating onto the membrane surface could be detected. A total of 16 scans were recorded at 4.0cm -1 resolution Contact angle measurement The hydrophilicity of the membranes was measured by sessile drop method using face contact angle (Model: CA-A Face Contact Angle Meter, Manufacturer: Kyowa Kaimenkagaku Co. Ltd). Contact angle is the quantification of the degree of membrane hydrophilicity (Bolong et al., 2008). The lower contact angle value represent the higher hydrophilicity while the vice versa situation indicates the lower hydrophilicity (Arkhangelsky et al., 2007; Gon et al., 2001) 2.4 Morphological classification 2.4.1Molecular weight cut-off (MWCO) assessment The molecular weight cut-off (MWCO) of this pre-treated membrane was determined by measuring the separation values of various proteins with different molecular weight (MW). A logarithmic graph of the rejection of these protein series versus their MW was plotted and MWCO was calculated based on 90% of protein retention (Temmel et al., 2006) Scanning Electron Microscopy (SEM) The cross section and surface morphology of all membranes, including the modified and the native PES membrane were visualized using scanning electron microscopy (SEM) located in the Institute of Oceanography 202 (INOS), Universiti Malaysia Terengganu, Terengganu Darul Iman. All the flat sheet membrane were immersed in liquid nitrogen to initiate brittle fracture and splintered carefully so as to obtain a neat cross section. These samples were then attached to a carbon holder and sputtered with gold by an automatic auto-coater JFC1600 (INOS, UMT) to prevent any charging up of the surface by the electron beam (Wahit et al., 2010; Suherman et al., 20010; Kools, 1998). 2.5 Hydraulic and permeability determination Membrane permeability was determined by allowing the pure water to pass through the compacted membrane. The information of pure water permeability is important to explain the morphological attributes of the membrane. It is a fundamental quantity with respect to the membrane and could reflect the porosity of the membrane. The pure water permeability can be expressed as P m (L/m 2.h.bar), and can be obtained by plotting the graph of flux versus pressure (Hagen Poisseuille equation), J v = P m ΔP, where ΔP is the applied pressure and J v is the water flux. A constant value of the pure water permeability i.e. the linear dependence of the pure water flux, J v on pressure, points to the unchangeable membrane porosity. If the water flux dependence on pressure deviates from linearity, the pure water permeability of a membrane is not constant and it indicates the changes in the membrane s porous structure. Thus, by measuring a dependence of the membrane s pure water permeability on pressure, a state of the membrane s active layer porosity can be characterized (Kosutic et al., 2006). 2.6 Fouling analysis The fouling analysis step is important to evaluate the fouling tendency of the membranes. Permeate was collected for every five minutes. Then, volume permeated was measured and fluxes were calculated. The steps were repeated until consecutive flux was obtained and the graphs of flux versus time were then plotted. All fluxes were calculated using a simple mathematical equation (Table 1) developed by Capar et al., (2007). J cwi is the initial permeate flux, J ww is the steady state flux, J cwf is the flux of the pure water through the fouled membrane and J cwc is the water flux of the watercleaned membrane. Through the obtained results of flux declination trends, the degree of concentration polarization could be evaluated. Thus, the fouling level of the membrane could be estimated. After completing the above process, the flux recovery ration (FRR) was calculated using the following expression: FRR = (Jv 2 /Jv 1 ) x 100, where Jv 1 is the initial water flux and Jv 2 is the water flux after cleaned process. After the FRR calculation, the degree of fouling resistance of the membrane could be determined. The higher the FRR, the higher the fouling nature of the membrane will be. For the further characterization of the fouling attributes in terms of morphology, shape (aggregate/uniform), the protein fouledmembranes were visualized using SEM. The virgin membrane was used as control.

3 Table 1 A simple mathematical equation that used to evaluate the flux decline behavior (Capar et al., 2007) Formula Type of flux decline (FD) amplified to more than 50 mg/l. The attained results had indirectly reveals that, 50 mg/l is the optimum myoglobin concentration for membrane surface modification. This will nevertheless proceed by the other analysis to further validate the conclusion. Table 2 Contact angles (J cwi -J ww )/J cwi Total flux decline Membranes ID Contact angle (J cwf -J ww )/J cwf Concentration polarization (J cwi -J cwf )/J cwi Fouling (irreversible+reversible) (J cwc -J cwf )/J cwc FD due to reversible fouling (J cwi -J cwc )/J cwi FD due to irreversible fouling MPH MPH MPH Native RESULTS AND DISSCUSSION 3.1 Surface characterization FTIR The influence of coating solution concentration on the extent of myoglobin adsorption on the membrane was evaluated by FTIR. From the plotted FTIR spectra (Figure 1), the typical infrared regions for PES as such aromatic bands at ~1578cm -1 and ~1485cm -1 from the benzene ring and 1240cm -1 and 1040cm -1 from ether groups of PES are clearly observed (Susanto and Ulbricht, 2009; Jung et al., 2004; Yu et al., 2009). Apparently, the bands in 3700cm -1 to 2500cm -1 (O-H and N-H), 2900cm -1 to 2800 cm -1 (H-bonds) and 1720cm -1 to 1730cm -1 (spectra for amino acid (AA) from carbonyl groups) all which denotes the peaks of (proteins) myoglobin, is highly intensified in MPH7-50 membrane. This might be the first indicator that signify that, the highest adsorption of myoglobin occur when MPH7 membranes were treated with coating solution having concentration of 50 mg/l. However, auxiliary analysis on hydrophilicity, surface charge, molecular weight cut-off (MWCO), pure water permeability (PWP) and protein filtration will be conducted to further verify this conclusion Hydrophilicity determination The contact angle measurement for MPH7-30, MPH7-50, and MPH7-70 membranes is depicted in Table 2. Although all the membranes were immersed in the myoglobin solution bears the same ph, there is a big difference in the contact angle when they were further modified by the myoglobin solution having various concentrations. This fairly suggesting that concentration of the coating material do plays a main role in membrane modification process. The smallest contact angle is attained for MPH7-50 membrane, which is However, when the concentration of myoglobin solution is increased to 70 mg/l, there is a dramatic increase in the contact angle value, implying that the membrane properties is gradually changes to the hydrophobic characteristic of the original un-modified PES membrane when concentration of the coating solution is Membrane surface charge The obtained zeta potential (Table 3) is lowest for MPH7-70 ( ), coherently with the discovery of the earlier scholars who declared that, increasing the coating solution concentration will lower the surface charge, as more myoglobin is adsorbed to cover the surface (Ghosh et al., 1998). However, the discrepancies in the surface charge for the three membranes are not significant and almost negligible. Table 3 Zeta potential Membranes ID Z p MPH MPH MPH Native Morphological characterization 3.2.1Molecular weight cut-off (MWCO) Table 4 represents the MWCO of MPH7-30, MPH7-50 and MPH7-70 membranes. MPH7-50 posses the lowest MWCO, which is 17kDa. This is likely due to the highest in the preceding FTIR result. A unique trend could be observed for MPH7-70, where its MWCO increase up to 36 kda, almost similar to the MWCO of the un-modified membrane (39kDa). From the findings, it is postulated that, the membrane surfaces were completely coated after treatment with 50mg/L myoglobin concentration and increasing the myoglobin concentration to adsorption of myoglobin for the MPH7-50, as illustrated in the preceding FTIR result.

4 Figure 1 FTIR of membrane (a) MPH7-50 (b) MPH7-30 (c) MPH7-70 (d) Native A unique trend could be observed for MPH7-70, where its MWCO increase up to 36 kda, almost similar to the MWCO of the un-modified membrane (39 kda). From the findings, it is postulated that, the membrane the membrane surfaces were completely coated after treatment with 50mg/L myoglobin concentration and increasing the myoglobin concentration to 70 mg/l will increase the MWCO up to the level of the MWCO of the native membrane, proposing that, the increment of myoglobin concentration more than 50 mg/l will only cause the myoglobin to be leached instead of accumulating on the membranes surface. Membranes ID Table 4 MWCO MPH MPH MPH Native 39 MWCO (kda) In a short explanation, myoglobin is no longer resides on the membranes surface, thus changing its character to the original un-modified membrane. These claims were strengthened by the earlier result from contact angle measurement, which shows that the contact angle of the membranes increase considerably to the contact angle of the virgin membrane after modified with 70mg/L coating solution Scanning electron microscopy (SEM) As expected, all the observed membranes have an asymmetric structure comprising of a dense skin layer, a porous intermediate layer and macrovoids at the bottom (Figure 2). The skin typically accounts for about 1-2 % of the entire thickness of the film and the pore size increase steadily from up to the bulk. It was anticipated since all membranes were produced using phase inversion technique and asymmetric configuration is a distinctive features for membranes prepared by phase inversion (Subramahyan, 2003). The most significant difference between the unmodified and modified membranes is the presence of a dense sponge like structure pitted between the two stratums of finger like and thick skin layer for the unmodified membrane. This spongy layer probably explained for the low water flux achieved for this membrane since it can creates resistance to water flow, thus resulting in rather low permeability rates (Idris et al., 2007; Boussu et al., 2006). Moreover, the native membrane has a rigid skin layer in contrast to the other membranes. Therefore, this is complementarily supporting its low pure water permeation level, since PWP is strongly dependant on the top and sublayer of the membranes. In comparison, all the myoglobin coated membranes (MPH7-30, MPH7-50 and MPH7-70) exhibit a less spongy constitution, with more porous finger and tear drop elements, consequently permitted a higher mobility to water, probably presenting the prime reason for their higher water permeability. Furthermore, the active layer of all the modified membranes is thinner, leading to higher fluxes (Jansen et al., 2006). 204

5 F lux, J v (L /m 2.h) MPH7-30 Native MPH7-50 MPH7-70 Figure 2 Cross sectional morphologies of membrane P res s ure (bar) J V = Δ P, R 2 = J V = Δ P, R 2 = J V = 8.30 Δ P, R 2 = ( J v = 9.68 Δ P, R 2 = ) Figure 3 Pure water flux and permeability of native and surface modified membranes 3.3 Pure water flux and permeability Permeability coefficient determination was implemented for MPH7-30, MPH7-50 and MPH7-70 membranes. All the plotted graphs show a linear profile, indicating that the pure water flux is directly proportional to the applied pressure. At applied pressure of 1 bar, all membranes gave the lowest flux and increased of flux occurred linearly as the pressure increased up to 5 bar. However, from the pure water flux results represented in Figure 3, it shows that, the P m of the MPH7-70 membrane does not lie in the range of permeability values for UF membrane specified for protein 205

6 Flux, Jv (L/m 2.h) Flux, Jv (L/m 2.h) Flux, Jv (L/m 2.h) Flux, Jv (L/m 2.h) separation. The highest P m is obtained for MPH7-50 membrane, which is 21.95L/m 2.h.bar. 3.4 Fouling analysis Flux behaviour analysis was performed for all membranes. In the first part, the investigation was observed for all membranes (Native, MPH7-50, MPH7-30 and MPH7-70) at three different concentrations of lysozyme: 300, 500 and 700 mg/l. Figure 4 and Table 5 clearly illustrate that, for all three concentrations, the decline in flux was lowest for MPH7-50 membrane. Although the pore size of MPH7-30 and MPH7-70 is larger and according to Yuan and Zydney (2000), the larger MWCO membrane will provide a less effect on flux decline, but the improvement of this membrane hydrophilicity after the surface modification may be the explanation for its fouling robustness, despite that, the mean pore size has been decreased. On the basis of contact angle data, MPH7-50 has an exceptional hydrophilic property in contrast to the other two membranes; consequently it is expected to foul less (Bhatacharjee et al., 2006; Grandison et al., 2000). The results indicate that, the superior hydrophilicity of MPH7-50 membrane has suppressed the influence of MWCO on the fouling behaviour. This finding is further supported by the former result of membrane surface charge. MPH7-30 and MPH7-70 each have a negative charge, opposite to the model protein (lysozyme), which bears a positive charge. When a solute and membrane template exhibit a converse charge, a strong attraction will occur between them, leading to a rigid solute adsorption, consequently fouling will be more severe. In contrast, for MPH7-50, membrane is neutrally charged and it would ensure the avoidance of donnan effect between the positively charged molecules and negatively charged virgin membrane. Subsequently, the tendency of solute adsorption and cake formation on the membrane vicinity could be reduced and then fouling occurrence will be hindered (Bates, 2008). The lowest flux decline was achieved with the lowest-concentration of feed solution (300 mg/l), and the greatest decline in flux was observed for the feed concentration of 700 mg/l. This result is in accord with the fact that as feed concentration increases, the membrane surface concentration also increases, leading to an increase in the osmotic pressure close to the membrane surface (Kanani et al., 2007; Petrus et al., 2008; Sarkar et al., 2009). In other words, the concentration polarisation increases proportionally with the increment of feed concentration. The polarised layer acts as a secondary membrane that promotes a higher rejection of solutes, subsequently leading to the formation of a cake layer on the membrane surface. This is the phenomenon responsible for the reduction of solvent flux, which leads to a pronounced flux decline at higher initial feed concentrations (Bates, 2009). The above step was then continued to FRR analysis. To investigate the FRR, the fouled membranes were transferred to ultrafiltration cell. It was then exposed to distilled water for 30 minutes, and stirring was carried out with a magnetic stirrer. The process was conducted without any pressure Time (minutes) 300 mg/l 500 mg/l 700 mg/l Native MPH7-70 Time (minutes) 500 mg/l 700 mg/l 300 mg/l MPH Time (minutes) 500 mg/l 300 mg/l 700 mg/l MPH Time (minutes) 500 mg/l 300 mg/l 700 mg/l Figure 4 Flux decline at various feed concentration

7 Native: Control MPH7-30: Control Native: Used MPH7-30: Used MPH7-50: Control MPH7-70: Control MPH7-50: Used MPH7-70: Used Figure 5 SEM (500x) micrographs of membrane surface (Comparison between the fresh and the fouled membrane) 207

8 Table 5 Percentage of membrane flux decline Membrane ID Flux Decline Feed Concentration (mg/l) Total FD MPH7-50 FD due to CP Flux due to R Flux due to IR Total FD MPH7-30 FD due to CP Flux due to R Flux due to IR Total FD MPH7-70 FD due to CP Flux due to R Flux due to IR Native Total FD FD due to CP Flux due to R Flux due to IR * FD = Flux Decline CP = Concentration Polarization R = Reversible IR = Irreversible Thereafter, membranes were removed from the cell and rinse twice by immersing in distilled water. The pure water flux (J 2 ) was once again calculated using these washed membranes. FRR were calculated by denominating the initial pure water flux (before protein filtration), J 1 by J 2. The FRR for MPH7-50 is the most outstanding of all, with the ratio of J 2 /J 1 close to unity (Table 6). This signifies the ability of water to clean the membrane, thus pertaining that the hydrophilic structure of the membrane caused the weaker association of lysozyme residual with water (Rice et al., 2009). In whole, the FRR of all the surface modified membrane is consistently high as opposed to the unmodified one, with each having more than 0.85 permeability reinstatements. The excellent flux recovery property of the myoglobin coated membranes compared to the virgin one, prove that the surface 208 modification has significantly upgrade the fouling resistant of the membranes (Wang et al., 2006; Wang et al., 2006). Therefore, it will render the membrane to be run for a long time without significant decrease of separation performance. Such findings owe to the combination of the following possibilities; a) negligible irreversible fouling build-up due to increased surface hydrophilicity and b) the cleaning of the foulants (reversible) by water alone is effective. The exceptional FRR values ensured the elongated utilization time and operational reliability of the membranes. For SEM analysis, a piece of membrane that has been used for protein filtration was removed from the beneath of the stirring cell. The membrane was gold coated and visualized using a scanning electron microscope (SEM). This step is deemed necessary in order to establish a more detail characterization of the foulants

9 in terms of their morphological (shape, size, etc) and their adsorption property (uniform/non- uniform) on the surface. The un-used disc for every type of membranes was used as a control. The lysozyme seems to be dispersed and they adsorb in a heterogeneous manner to the membrane surface (Figure 5). The surface of MPH7 membranes visibly shows the littlest amount of foulants, followed by MPH5, MPH9, MPH11 and the most rigorous fouling was discerned for native membrane, proving that the protein (lysozyme) adhesion onto the surface modified membrane has been considerably hindered due to the hydrophilicity enhancement. Table 6 Flux recovery ratio (FRR) Membrane ID Feed Concentration (mg/l) FRR (%) MPH MPH MPH Native CONCLUSION In summary, the hydrophilicity characteristics of PES-UF membrane have been enhanced render to the formation of myoglobin layer on the membrane surface. The development of myoglobin coating was readily obtained by 12 hours dipping of PES-UF membranes in the solution of myoglobin followed by a rinse step. The level of myoglobin adsorption and deposition on the top layer of membrane were controlled by adjusting parameters such as the concentration of the coating solution. This method of surface modification is relatively simple and economical since it does not require any pricey equipment, or elaborate material design. We really hope that this work can provide some facile approach toward the development of low-fouling, highly endurance membrane. ACKNOWLEDGEMENT The authors would like to thank all the individuals, involves directly or indirectly in this project, for their academics and technical supports. Utmost gratitude is for Ministry of Science, Technology and Innovation (MOSTI) Malaysia for the financial support. Appreciation also goes to Universiti Malaysia Terengganu, (UMT) Malaysia for the workplace and Malaysian Institute of Nuclear Technology (MINT) for their permission in using contact angle analyzer. 209

10 REFERENCES Ahmad, Z., Ansell, M.P., and Smedley, D Epoxy adhesives modified with nano and micro particles for in-situ timber bonding: effect of microstructure on bond integrity, International Journal of Mechanical and Materials Engineering 5(1): Arkhangelsky, E., Kuzmenko, D. and Gitis, V Impact of chemical cleaning on properties and functioning of polyethersulfone membranes, Journal of Membrane Science 305: Ba, C., Ladner, D.A., Economy, J Using polyelectrolyte coatings to improve fouling resistance of a positively charged nanofiltration membrane, Journal of Membrane Science 347: Bates Reducing the fouling rates of surface and wastewater RO system. IWC Retrieved from http//: Bhattacharjee, S., Ghosh, S., Datta, S. and Bhattacharjee, C Studies on ultrafiltration of casein whey using a rotating disk module: effects of ph and membrane disk rotation, Desalination 195: Bolong, N., Ismail, A.F., Salim, M.R. and Matsuura, T Miscibility characteristic of PES hollow fiber membranes with charged-surface modifying macromolecules (csmm) for water separation applications, Jurnal Teknologi, 49(F): Boussu, K., B. Van der Bruggen, Volodin, A., Haesendock, C.V., Delcour, J.A., P. Van der Meeren and Vandecasteele, C Characterization of nanofiltration membranes with self made polyethersulfone membrane, Desalination 191: Capar, G., Yetis, U. and Yilmaz, L The most effective pre treatment to nanofiltration for the recovery of print dyeing wastewater, Desalination 212: Ghosh, R. and Cui, Z.F Fractionation of BSA and lysozyme using ultrafiltration: effect of ph and membrane pre-treatment, Journal of Membrane Science 139: Gon, J., Belfer, S., Vwhenb, P. and Nystrom, M Effects of surface modification on antifouling and performance properties of reverse osmosis membranes, Desalination 140: Grandison, S. A, Youravong, W. and Lewis, M.J Hydrodynamic factors affecting flux and fouling during ultrafiltration of skimmed milk, Lait 80: Huisman I.H., Pradanos P. and Hernandez A The effect of protein protein and protein membrane interactions on membrane fouling in ultrafiltration, Journal of Membrane Science 179: Hurwitz, G., Gregory, R. Eric, G. and Hoek M.V Probing polyamide membrane surface charge, zeta potential, wettability, and hydrophilicity with contact angle measurements, Journal of Membrane Science 347: Idris, A., Nordin, M.Y., Ismail, A.F. and Shilton, S.J Study of shear rate influence on the performance of 210 cellulose acetate reverse osmosis hollow fiber membranes, Journal of Membrane Science 205: Idris, A., Zain, N.M. and Noordin, M.Y Synthesis, characterization and performance of asymmetric polyethersulfone (PES) ultrafiltration Membranes with polyethylene glycol of different molecular weights as additives, Desalination 207: Jansen, J.C., Buonomenna, M.G., Figoli, A. and Drioli, E Ultra thin asymmetric gas separation membrane of modified PEEK prepared by the dry wet phase inversion technique, Desalination 193: Jung B Preparation of hydrophilic polyacrylonitrile blends membranes for ultrafiltration, Journal of Membrane Science 229: Kanani, D.M. and Ghosh, R A constant flux based mathematical model for predicting permeate flux decline in constant pressure protein ultrafiltration, Journal of Membrane Science 290: Kools, W.F.C Membrane formation by phase inversion in multicomponent polymer system mechanisms and morphologies. Unpublished Doctoral Dissertation, University of Twente. Kosutic, K., Dolar, D. and Kunst, B On experimental parameters characterizing the reverse osmosis and nanofiltration membranes active layer, Journal of Membrane Science 282: Petrus, H.B., Li H., Chen V. and Norazman N Enzymatic cleaning of ultrafiltration membranes fouled by protein mixture solutions, Journal of Membrane Science 325: Reddy A.V.R., Mohan D.J., Bhattacharya A., Shah V.J. and Ghosh P.K Surface modification of ultrafiltration membranes by preadsoprtion of a negatively charged polymer. I. Permeation of water soluble polymers and inorganic salt solutions and fouling resistance properties, Journal of Membrane Science 214: Rice G., Barber A., O Connor A., Stevens G. and Kentish S Fouling of NF membranes by dairy ultrafiltration permeates, Journal of Membrane Science 330: Sarkar P., Ghosh S., Dutta S., Sen D. and Bhattacarjee C Effect of different operating parameters on the recovery of proteins from casein whey using a rotating disc membrane ultrafiltration cell, Desalination 249: Shi Q., Su Y., Zhao W., Li C., Hu Y., Jiang Z. and Zhu S Zwitterionic polyethersulfone ultrafiltration membrane with superior antifouling properties, Journal of Membrane Science 319: Subrahmanyan, S An Investigation of Pore Collapse in Asymmetric Polyethersulfone Membrane. Unpublished Doctoral Dissertation, Virginia Polytechnic Institute and State University. Suherman, H., Sulong, A.B. and Sahari, J Effect of filler loading concentration on the electrical

11 conductivity of CNTs/Graphite/Epoxy-nano composite at high loading of conductivity fillers, International Journal of Mechanical and Materials Engineering 5(1): Susanto H., Stahra, N., Ulbricht M High performance polyethersulfone microfiltration membrane having high flux and stable hydrophilic properties, Journal of Membrane Science 342: Susanto, H. and Ulbricht, M Characteristics, performance and stability of polyethersulfone ultrafiltration membrane prepared by phase separation method using different macromolecular additives, Journal of Membrane Science 327: Temmel, S., Kern, W. and Luxbacher, T Zeta potential of photochemically modified polymer surfaces, Progress in Colloid and Polymer Science 132: Wahit, M.U., Hassan, A., Koay, A.F., Othman, N. and Mokhtar, M Toughening of polyamide 6 nanocomposites: effect of organoclay and maleic anyhride grafted polyethylene octane loading on morphology and mechanical properties, International Journal of Mechanical and Materials Engineering 5(1): Wang, Y., Su, Y., Sun, Q., Ma, X., Ma, X. and Jiang, Z Improved permeation performance of pluronic F- 127 polyethersulfone blend ultrafiltration membranes, Journal of Membrane Science 282: Wang, Y.Q., Su, Y.L., Sun, Q., Ma, X.L. and Jiang, Z Generation of anti-biofouling ultrafiltration membrane surface by blending novel branched amphiphilic polymers with polyethersulfone, Journal of Membrane Science 286: Yu, H., Cao, Y., Kang, G., Liu, J., Li, M. and Yuan, Q Enhancing antifouling property of polysulfone ultrafiltration membrane by grafting zwitterionic copolymer via UV-initiated polymerization, Journal of Membrane Science 342: Yuan, W. and Zydney, A.L Humic acid fouling during ultrafiltration, Environmental Science Technology 34: