Fouling of Nanofiltration, Reverse Osmosis, and Ultrafiltration

ARTICLE pubs.acs.org/est

Fouling of Nanofiltration, Reverse Osmosis, and Ultrafiltration Membranes by Protein Mixtures: The Role of Inter-Foulant-Species Interaction Yi-Ning Wang†,‡ and Chuyang Y. Tang*,†,‡ †

School of Civil and Environmental Engineering and ‡Singapore Membrane Technology Centre, Nanyang Technological University, Singapore 639798

bS Supporting Information ABSTRACT: Protein fouling of nanofiltration (NF), reverse osmosis (RO), and ultrafiltration (UF) membranes by bovine serum albumin (BSA), lysozyme (LYS), and their mixture was investigated under crossflow conditions. The effect of solution chemistry, membrane properties, and permeate flux level was systematically studied. When the solution pH was within the isoelectric points (IEPs) of the two proteins (i.e., pH 4.7 10.4), the mixed protein system experienced more severe flux decline compared to the respective single protein systems, which may be attributed to the electrostatic attraction between the negatively charged BSA and positively charged LYS molecules. Unlike a typical single protein system, membrane fouling by BSA LYS mixture was only weakly dependent on solution pH within this pH range, and increased ionic strength was found to enhance the membrane flux as a result of the suppressed BSA LYS electrostatic attraction. Membrane fouling was likely controlled by foulant fouled-membrane interaction under severe fouling conditions (elevated flux level and unfavorable solution chemistry that promotes fouling), whereas it was likely dominated by foulant clean-membrane interaction under mild fouling conditions. Compared to nonporous NF and RO membranes, the porous UF membrane was more susceptible to dramatic flux decline due to the increased risk of membrane pore plugging. This study reveals that membrane fouling by mixed macromolecules may behave very differently from that by typical single foulant system, especially when the inter-foulant-species interaction dominates over the intra-species interaction in the mixed foulant system.

1. INTRODUCTION Reverse osmosis (RO) and nanofiltration (NF) membranes are increasingly used in wastewater reclamation and surface water treatment.1 The performance of these membranes can be severely affected as a result of membrane fouling by macromolecules such as proteins.2 5 Compared to the vast literature available on protein fouling of porous membranes (microfiltration (MF) and ultrafiltration (UF)), there have been only a handful number of studies on RO and NF fouling by proteins.6 10 In addition, many of these studies only involved a single type of protein in the feedwater (such as bovine serum albumin (BSA) due to its easy availability).5 Existing literature suggests that protein fouling is highly dependent on solution chemistry. For example, membrane flux is generally least stable when the solution pH is close to the isoelectric point (IEP) of a protein due to the lack of electrostatic repulsion between the molecules.6,9 12 Similarly, studies of RO and NF fouling with BSA revealed that fouling was more severe at higher ionic strength, which is attributed to the electrical double layer (EDL) compression effect and the associated suppression of intermolecular electrostatic repulsive force between BSA molecules.6,9,10 In addition to unfavorable solution conditions, r 2011 American Chemical Society

protein fouling can also be promoted by elevated membrane permeate flux.6,10,12,13 Similar effect of permeate flux has also been observed for other macromolecules.14 16 In practice, typical feedwater for RO and NF contains a mixture of foulants rather than a single well-defined foulant. Thus, investigations on fouling by mixed foulant species can be highly valuable. Compared to a single-foulant feed, a binaryfoulant system is much more complicated.5 Controlled laboratory studies on fouling by the mixture of BSA and alginate revealed accelerated membrane permeability loss compared to that induced by BSA or alginate alone.6 On the contrary, filtration of a feed consisting of humic acid and polysaccharides 17 showed that the foulant mixture may have less severe fouling compared to the individual foulant in some cases. Complicated fouling behavior was also reported for organic inorganic foulant mixtures.18,19 Presumably, the inter-foulant-species interaction may play a critical role in a mixed foulant system in addition to Received: April 18, 2011 Accepted: June 16, 2011 Revised: June 10, 2011 Published: June 16, 2011 6373

dx.doi.org/10.1021/es2013177 | Environ. Sci. Technol. 2011, 45, 6373–6379

Environmental Science & Technology

Figure 1. Zeta potential of BSA and LYS in 10 mM NaCl as a function of pH. At least 2 replicate measurements were performed for each condition.

the intra-foulant-species interaction and foulant membrane interaction.5 Up to date, systematic investigations on the role of foulant foulant interaction on membrane fouling by mixed foulants are still limited. In addition, the dependence of fouling on solution chemistry, membrane properties, and hydrodynamic conditions is poorly understood for such systems. The objective of the current study was to investigate the effect of solution chemistry, membrane properties, and hydrodynamic conditions on binary protein fouling of NF, RO, and UF membranes. BSA, lysozyme (LYS), and their binary mixture were used as the model foulants. The current study may provide important insights on the role of foulant foulant interactions and hydrodynamic forces in mixed foulant systems.

2. MATERIALS AND METHODS 2.1. Chemicals and Materials. Unless specified otherwise, all reagents and chemicals were of analytical grade with purity over 99%. Ultrapure water (resistivity of 18.2 Mohm.cm) was used to prepare working solutions. The pH and ionic compositions of the feed solution were adjusted by the addition of sodium chloride, hydrochloric acid, and sodium hydroxide. Bovine serum albumin (BSA, Sigma-Aldrich A7906) and lysozyme (LYS, Fluka 62971) were used as model protein foulants to represent two proteins of very different IEPs. On the basis of the zeta potential measurements performed in the current study (Malvern ZetaSizer Nano ZS, following the method reported in ref 10), the IEP for BSA is ∼pH 4.7 and that for LYS is ∼pH 10.4 (Figure 1). Both BSA and LYS were received in powder form with g98% purity and were stored at 4 °C in the dark. The molecular weights of BSA and LYS are 67 and 14.3 kDa, respectively.6,11 Protein working solutions were freshly prepared prior to each fouling experiment. Binary BSA LYS mixture was prepared by dissolving equal amount of BSA and LYS (i.e., BSA/LYS at 1:1 w/w ratio). The membranes investigated in the current study include three Dow FilmTec membranes (an RO membrane XLE and two NF membranes NF90 and NF270), and one GE Osmonics UF membrane (membrane GM). Their properties have been reported in our previous studies,10,20 22 and are summarized in Table 1 for the ease of reference. Briefly, the semi-aromatic piperazine-based

ARTICLE

membrane NF270 has a much smoother surface (root-meansquare roughness Rrms = ∼9 nm) than those of the fully aromatic polyamide membranes XLE and NF90 (Rrms on the order of 100 nm).20 23 In addition, NF270 is more hydrophilic and more negatively charged, and it has higher water permeability and lower salt rejection compared to XLE and NF90. The polyamidebased UF membrane GM has a molecular weight cutoff (MWCO) of ∼8 kDa, with a relatively smooth and hydrophilic membrane surface (contact angle ∼39.3°).24 2.2. Fouling Experiments. Details of the crossflow membrane test setup and fouling procedures can be found elsewhere.10 Crossflow test cells (CF042 Membrane Cell, Sterlitech, with an active membrane area of 42 cm2) were used to house flat sheet membrane coupons. Feed water was pumped using a variablespeed diaphragm pump (Model D-03, HydraCell, Minneapolis, MN). The pressure inside each test cell was adjusted using a back pressure regulator, and the crossflow velocity was adjusted with a needle valve. The pressure and permeate flux were measured using pressure transmitter (WIKA Transmitter, 0 40 bar, Germany) and digital flowmeter (Bronkhorst, Netherlands) respectively and were recorded with a computer data logging system. Manual gravimetric measurement of permeate flux was also performed at predetermined time intervals to ensure the flux measurement accuracy. For each experimental run, a new membrane coupon was soaked in ultrapure water overnight before being loaded into the test cell. The coupon was first compacted and equilibrated with the background electrolyte solution (with identical composition to that used for subsequent fouling test except that foulant was not added) for 2 days under pressure to ensure that any subsequent flux decline after the addition of foulant was not influenced by the mechanical compaction of the membrane.10 Desired amount of foulant was added to the feed solution at the end of the 2-day compaction stage to initiate membrane fouling, and the test was continued for another 4 days. During the entire 6-day test, the same applied pressure, crossflow velocity and background electrolyte solution chemistry (pH and ionic composition) were used. Unless specified otherwise, the following reference conditions were applied: • a total foulant concentration of 20 mg/L (10 mg/L BSA + 10 mg/L LYS for the BSA LYS mixture) • pH 7 and total ionic strength of 10 mM (adjusted by addition of NaCl) • initial permeate flux (the flux right before the addition of foulant to the feedwater) at either 75 or 120 L/m2 3 h • crossflow velocity at 9.5 cm/s with diamond patterned spacer used in the feedwater channel • feed tank temperature at 20 ( 1 °C

3. RESULTS AND DISCUSSION 3.1. Zeta Potential of BSA and LYS. The zeta potential values of BSA and LYS are presented in Figure 1. The IEPs of BSA and LYS were ∼pH 4.7 and pH 10.4, respectively, consistent with the values reported in the literature.11 For both proteins, the zeta potential increased at lower solution pHs. Around neutral pH, BSA was negatively charged, whereas LYS was positively charged. 3.2. Mixed Protein Fouling versus Single Protein Fouling. Figure 2 shows the flux behavior of an NF membrane (NF270) in the presence of a binary mixture of BSA and LYS (10 mg/L BSA and 10 mg/L LYS). For comparison purpose, the fouling behavior in the presence of a single protein (20 mg/L BSA or 6374

dx.doi.org/10.1021/es2013177 |Environ. Sci. Technol. 2011, 45, 6373–6379


ARTICLE

Table 1. Summary of Membrane Properties (Adapted from Ref 10) membrane XLE (RO)a a

NF90 (NF)

a

NF270 (NF) GM (UF)

a

water permeability

NaCl rejection

MWCO

contact angle

membrane zeta

root mean square

(L/m2 3 h 3 psi)b

(%)b

(Daltons)

(°)b

potential @ pH 7 (mV)

roughness (nm)

0.396

96.5

80% over 4 days). At pH 3.5 (pH below IEPBSA and IEPLYS), both BSA and LYS were positively charged (Figure 1). Thus, the more stable flux at this solution pH may be explained by the BSA LYS inter-species electrostatic repulsive force. In comparison, such BSA LYS electrostatic repulsion was not available at pH 4.7, 7.0, and 8.2 (IEPBSA e pH e IEPLYS). It is also interesting to observe that the final fluxes (J96 h) were nearly identical at pH 4.7, 7.0, and 8.2 (part a of Figure 4), which suggests that the BSA LYS mixture fouling had very weak pHdependence within this pH range. This is in direct contrast to the strong pH-dependence when there was either BSA or LYS alone in the feed solution (Figure S1 of the Supporting Information). Part b of Figure 4 plots J96 h as a function of pH. When there was only a single protein (either BSA or LYS) in the feedwater, improved J96 h was observed at pHs away from the protein’s IEP. For instance, the J96 h value increased from its minimum value ∼25 L/m2.h at pH 4.7 (i.e., IEP of BSA) to nearly 60 L/m2.h at pH 7.8, which may be attributed to the enhanced BSA BSA electrostatic repulsion.6,9,11 Similar flux enhancement was observed at a lower pH (J96 h = ∼85 L/m2.h at pH 3.8). The BSA LYS mixture had comparable J96 h values to those for BSA alone at pH around or below 4.7, where the BSA LYS electrostatic interaction was either repulsive (pH 3.5) or negligible (pH 4.7). This result may suggest that the BSA BSA interaction dominates over BSA LYS interaction when pH < IEPBSA. However, when the solution pH was within the IEPs of the two proteins, the pseudostable fluxes for the binary mixture were significantly lower compared to those for BSA alone. Within the two IEPs (pH 4.7 10.4), the final flux did not seem to benefit from the enhanced BSA BSA electrostatic repulsion, likely due to the presence of the BSA LYS attractive interaction. As discussed earlier (Figure 2 and related discussion on the effect of ionic strength), the flux behavior for the mixed proteins in this pH range was likely dominated by the BSA LYS interaction. Presumably, the mixed foulants may deposit in a fashion that the negatively charged BSA molecules may be surrounded by the oppositely charged LYS molecules (and vice versa) to achieve an 6376



Figure 5. Effect of membrane properties on flux performance during mixed protein fouling at (a) pH 7.0, (b) pH 4.7, and (c) pH 3.5. Other test conditions: total foulant concentration 20 mg/L (10 mg/L BSA + 10 mg/L LYS), ionic strength of 10 mM, crossflow velocity of 9.5 cm/s, and temperature at 20 ( 1 °C.

energetically more stable arrangement,26 which renders the BSA BSA (or LYS LYS) interaction less important. The above discussion may be captured by the simple conceptual model presented in part c of Figure 4. For a mixture containing proteins A and B (where IEPA < IEPB), the flux behavior may be dominated by the attractive A B interaction if the solution pH is within the two IEPs. In this region, a rise in ionic strength tends to increase the pseudostable flux as a result of the weakened A B electrostatic attraction. In contrast, at pH

ARTICLE

below IEPA or above IEPB, the fouling behavior is likely governed by the respective A A or B B interaction, such that the mixed protein fouling behaves in a similar manner as the corresponding single protein case. This conceptual model interprets the effect of solution chemistry via both inter-species and intra-species foulant foulant electrostatic interactions. Despite that other nonelectrostatic interactions may also play a role in a binary protein mixture, the conceptual model was able to provide an insightful and consistent explanation of the flux dependence on pH and ionic strength during mixed protein fouling. This is because that the electrostatic interactions for charged molecules and particles are much more strongly affected by solution pH and ionic strength compared to nonelectrostatic interactions.5 Future research is warranted to further address the role of both electrostatic and nonelectrostatic interactions in mixed foulant systems. 3.4. Effect of Membrane Properties on Mixed Protein Fouling. Figure 5 shows the flux behavior of the four membranes (NF270, XLE, NF90, and GM) fouled by the binary protein mixture at various solution pHs. To ensure that the experimental results are directly comparable, the membranes were subjected to the same initial permeate flux of 75 L/m2.h. Under all cases, the three nonporous membranes (NF270, XLE, and NF90) had higher pseudostable fluxes compared to the porous UF membrane GM. At pH 7.0 and 4.7 where severe membrane fouling had occurred (parts a and b of Figure 5), nearly identical fluxes were obtained for NF270, XLE, and NF90 at long filtration time, despite that the clean membranes had very different membrane properties (Table 1). This suggests that the long-term fouling tends to be controlled by the foulant fouled-membrane interaction, which in turn is a strong function of the solution chemistry. Under severe fouling conditions, the membrane surface can be completely masked by the foulant cake layer,23 which renders the pseudostable flux independent of the properties of the RO and NF membranes.10,15,27 In the current study, the contact angle results showed similar hydrophobicity of the fouled membranes, although they were very different for the clean membranes (Figure S2 of the Supporting Information). However, a strong membrane dependency was observed at pH 3.5 (part c of Figure 5), with J96 h increased in the following order: NF90 < XLE < NF270. Membrane fouling at pH 3.5 was relatively mild such that the foulant clean-membrane interaction may still play an important role. The current study revealed that the long-term flux behavior under mild fouling conditions was dominated by foulant clean-membrane interaction, whereas that under severe fouling conditions was controlled by foulant fouled-membrane interaction. The better antifouling performance of NF270 was likely attributed to its more hydrophilic and smoother membrane surface compared to XLE and NF90 (Table 1). As shown in Figure 5, the UF membrane GM suffered severe flux loss under a wide pH range (∼80% flux reduction within the first hour of fouling at pH 7.0 and 4.7, and ∼50% at pH 3.5). In the current study, the porous GM membrane has an MWCO of ∼8 kDa, which is comparable to the molecular weight of LYS (14.3 kDa) but significantly smaller than that of BSA (67 kDa). Thus, it might be hypothesized that the severe initial flux decline experienced by GM was due to membrane pore plugging by the small LYS molecules. In Figure S4 of the Supporting Information, the flux behavior of membrane GM due to BSA LYS mixture fouling was compared to the respective single protein cases. The fouling curve for LYS alone was nearly identical to that for the mixture. This observation is consistent with the 6377



Figure 6. Effect of initial flux on the flux performance during mixed protein fouling. (a) pH 7, ionic strength of 10 mM; (b) pH 7, ionic strength of 100 mM. Other test conditions: membrane NF270, total foulant concentration 20 mg/L (10 mg/L BSA + 10 mg/L LYS), crossflow velocity of 9.5 cm/s, and temperature at 20 ( 1 °C.

hypothesis that the dramatic flux decline of GM in the presence of the mixture was mainly caused by pore plugging by the small LYS molecules. When the feedwater contained only BSA, the membrane GM experienced less severe flux decline, and the final stable flux was almost identical to those of the nonporous membranes (NF270, XLE, and NF90, as seen in Figure S3 of the Supporting Information), suggesting that cake layer formation was likely playing a more significant role for fouling by the larger BSA molecules. 3.5. Effect of Initial Flux on Mixed Protein Fouling. The effect of initial permeate flux on the mixture fouling is presented in this section. Part a of Figure 6 shows the flux behavior of NF270 under two different initial flux values (120 and 75 L/m2.h), where the feed solution contained 10 mg/L BSA and 10 mg/L LYS in 10 mM NaCl at pH 7. Severe membrane fouling occurred in both cases, and the test with higher initial flux suffered greater flux decline. It is interesting to note the similar long-term flux behavior both reached an identical J96 h value of ∼22 L/m2.h. As also shown in part a of Figure 5, the same pseudostable flux value was attained for the other nonporous membranes XLE and NF90. Part b of Figure 6 shows the membrane flux behavior under initial flux of 15 120 L/m2.h at an ionic strength of 100 mM. As discussed in section 3.3, membrane fouling with oppositely charged proteins experienced a decrease in flux

ARTICLE

decline at the higher ionic strength due to the depression of BSA LYS electrostatic attraction. For the test with initial flux of 120 L/m2.h at 100 mM ionic strength, its J96 h was ∼44 L/m2.h. The same J96 h value was also attained when the initial flux was 50 L/m2.h. However, only mild flux decline was observed when the initial flux level was lower than or close to 44 L/m2.h. These results revealed that the permeate flux of a nonporous membrane was largely membrane-limited under low initial flux conditions (minimal membrane fouling and flux controlled by membrane resistance, ref 15) and foulant-limited under elevated initial flux conditions.15,27 In the latter case, severe flux reduction may occur, and there was a strong tendency to approach an identical limiting value (i.e., the limiting flux5,15,27) that was independent of membrane surface properties and initial flux conditions. 3.6. Implications. The current study has important implications for membrane fouling with protein mixtures (or mixture of other types of macromolecules), 1) For a binary system containing macromolecules A and B (where IEPA < IEPB), the inter-foulant-species interaction (A B interaction) may play a critical role in addition to the intra-foulant-species interactions (A A and B B interactions). When the pH is within the two IEPs of A and B, membrane permeate flux can be significantly destabilized by the A B electrostatic attraction. To achieve optimal flux stability (or enhanced cleaning efficiency), the solution pH needs to be either below IEPA or above IEPB. Although the current study focuses primarily on binary protein mixtures, results from the current study might also be applicable for membrane fouling by protein polysaccharide mixtures, where the protein polysaccharide interaction may dominated over the protein protein and polysaccharide polysaccharide interactions. Such systems may have important practical significance (e.g., fouling in membrane bioreactors involves both proteins and polysaccharides28), and further systematically studies are required. 2) The effect of solution chemistry in a mixed foulant system can be potentially different from (or even directly opposite to) a single foulant system. For example, whereras the majority of existing studies reported reduced flux stability at higher ionic strength in the presence of a single type of charged macromolecules, the current study clearly demonstrated the opposite trend when the inter-species electrostatic attraction dominates the system. Further investigations are necessary to achieve improved mechanistic understanding of the foulant behavior in such multifoulant systems. 3) The long-term flux behavior can be either membranecontrolled or foulant-controlled. In the foulant controlled region (e.g., this can be promoted by a higher initial flux or unfavorable solution conditions), the membrane surface properties had little influence of the long-term flux behavior, and this is usually accompanied with severe flux decline. From a practical point of view, this region needs to be avoided by either stabilizing the feed solution or reducing the permeate flux. 4) Compared to nonporous NF and RO membranes, porous membranes are more likely to experience dramatic flux decline due to the risk of pore plugging. One potential way to enhance the antifouling performance of a UF membrane is to control the surface pore size and pore size distribution via surface modification. 6378



’ ASSOCIATED CONTENT

bS

Supporting Information. S1. Effect of solution pH on flux performance during single protein fouling; S2. Contact angles of virgin and fouled membranes; S3. Effect of membrane properties on flux performance during single protein fouling; S4. Comparison of flux behavior of mixed protein fouling with single protein fouling for the UF membrane GM. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel: +65 6790 5267; Fax: +65 6791 0676; E-mail: cytang@ ntu.edu.sg.

’ ACKNOWLEDGMENT The authors thank the Environment and Water Industry Programme Office (EWI) under the National Research Foundation of Singapore (Project # MEWR C651/06/176) for funding support for the research work carried out in this manuscript. The Singapore Membrane Technology Centre is supported by both EWI and Nanyang Technological University. We also thank Dow FilmTec and GE Osmonics for providing free membrane samples. ’ REFERENCES (1) Fane, A. G.; Tang, C. Y.; Wang, R. Membrane technology for water: Microfiltration, ultrafiltration, nanofiltration, and reverse osmosis. In Treatise on Water Science; Wilderer, P., Ed.; Academic Press: Oxford, 2011; pp 301 335. (2) Barker, D. J.; Salvi, S. M. L.; Langenhoff, A. A. M.; Stuckey, D. C. Soluble microbial products in ABR treating low-strength wastewater. J. Environ. Eng. 2000, 126 (3), 239–249. (3) Jarusutthirak, C.; Amy, G. Role of soluble microbial products (SMP) in membrane fouling and flux decline. Environ. Sci. Technol. 2006, 40 (3), 969–974. (4) Her, N.; Amy, G.; Plottu-Pecheux, A.; Yoon, Y. Identification of nanofiltration membrane foulants. Water Res. 2007, 41 (17), 3936–3947. (5) Tang, C. Y.; Chong, T. H.; Fane, A. G. Colloidal interactions and fouling of NF and RO membranes: A review. Adv. Colloid Interface Sci. 2011, 164, 126–143. (6) Ang, W. S.; Elimelech, M. Protein (BSA) fouling of reverse osmosis membranes: Implications for wastewater reclamation. J. Membr. Sci. 2007, 296 (1 2), 83–92. (7) Kim, S.; Hoek, E. M. V. Interactions controlling biopolymer fouling of reverse osmosis membranes. Desalination 2007, 202 (1 3), 333–342. (8) Li, Q.; Xu, Z.; Pinnau, I. Fouling of reverse osmosis membranes by biopolymers in wastewater secondary effluent: Role of membrane surface properties and initial permeate flux. J. Membr. Sci. 2007, 290 (1 2), 173–181. (9) Mo, H.; Tay, K. G.; Ng, H. Y. Fouling of reverse osmosis membrane by protein (BSA): Effects of pH, calcium, magnesium, ionic strength and temperature. J. Membr. Sci. 2008, 315 (1 2), 28–35. (10) Wang, Y.; Tang, C. Y. Protein fouling of nanofiltration, reverse osmosis, and ultrafiltration membranes - the role of hydrodynamic conditions, solution chemistry, and membrane properties. J. Membr. Sci. 2011, 376 (1 2), 275–282. (11) Palecek, S. P.; Zydney, A. L. Intermolecular electrostatic interactions and their effect on flux and protein deposition during protein filtration. Biotechnol. Prog. 1994, 10 (2), 207. (12) She, Q.; Tang, C. Y.; Wang, Y. N.; Zhang, Z. The role of hydrodynamic conditions and solution chemistry on protein fouling during ultrafiltration. Desalination 2009, 249 (3), 1079–1087.

ARTICLE

(13) Wu, D.; Howell, J. A.; Field, R. W. Critical flux measurement for model colloids. J. Membr. Sci. 1999, 152 (1), 89. (14) Hong, S. K.; Elimelech, M. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. J. Membr. Sci. 1997, 132 (2), 159. (15) Tang, C. Y.; Leckie, J. O. Membrane independent limiting flux for RO and NF membranes fouled by humic acid. Environ. Sci. Technol. 2007, 41 (13), 4767–4773. (16) Lee, S.; Elimelech, M. Relating organic fouling of reverse osmosis membranes to intermolecular adhesion forces. Environ. Sci. Technol. 2006, 40 (3), 980. (17) M€antt€ari, M.; Puro, L.; Nuortila-Jokinen, J.; Nystr€om, M. Fouling effects of polysaccharides and humic acid in nanofiltration. J. Membr. Sci. 2000, 165 (1), 1–17. (18) Lee, S.; Cho, J.; Elimelech, M. Combined influence of natural organic matter (NOM) and colloidal particles on nanofiltration membrane fouling. J. Membr. Sci. 2005, 262 (1 2), 27–41. (19) Contreras, A. E.; Kim, A.; Li, Q. Combined fouling of nanofiltration membranes: Mechanisms and effect of organic matter. J. Membr. Sci. 2009, 327 (1 2), 87–95. (20) Tang, C. Y.; Kwon, Y.-N.; Leckie, J. O. Probing the nano- and micro-scales of reverse osmosis membranes—A comprehensive characterization of physiochemical properties of uncoated and coated membranes by XPS, TEM, ATR-FTIR, and streaming potential measurements. J. Membr. Sci. 2007, 287 (1), 146–156. (21) Tang, C. Y.; Kwon, Y.-N.; Leckie, J. O. Effect of membrane chemistry and coating layer on physiochemical properties of thin film composite polyamide RO and NF membranes. I. FTIR and XPS characterization of polyamide and coating layer chemistry. Desalination 2009, 242 (1 3), 149–167. (22) Tang, C. Y.; Kwon, Y.-N.; Leckie, J. O. Effect of membrane chemistry and coating layer on physiochemical properties of thin film composite polyamide RO and NF membranes II. Membrane physiochemical properties and their dependence on polyamide and coating layers. Desalination 2009, 242 (1 3), 168–182. (23) Tang, C. Y.; Kwon, Y.-N.; Leckie, J. O. Characterization of humic acid fouled reverse osmosis and nanofiltration membranes by transmission electron microscopy and streaming potential Measurements. Environ. Sci. Technol. 2007, 41 (3), 942–949. (24) Shim, Y.; Lee, H. J.; Lee, S.; Moon, S. H.; Cho, J. Effects of natural organic matter and ionic species on membrane surface charge. Environ. Sci. Technol. 2002, 36 (17), 3864–3871. (25) Kelly, S. T.; Zydney, A. L. Effects of intermolecular thioldisulfide interchange reactions on BSA fouling during microfiltration. Biotechnol. Bioeng. 1994, 44 (8), 972. (26) Hogg, R.; Healy, T. W.; Fuerstenau, D. W. Mutual coagulation of colloidal dispersions. Trans. Faraday Soc. 1966, 62, 1638–1651. (27) Tang, C. Y.; Kwon, Y. N.; Leckie, J. O. The role of foulant foulant electrostatic interaction on limiting flux for RO and NF membranes during humic acid fouling-theoretical basis, experimental evidence, and AFM interaction force measurement. J. Membr. Sci. 2009, 326 (2), 526–532. (28) Gao, D. W.; Zhang, T.; Tang, C. Y.; Wu, W. M.; Wong, C. Y.; Lee, Y. H.; Yeh, D. H.; Criddle, C. S. Membrane fouling in an anaerobic membrane bioreactor: Differences in relative abundance of bacterial species in the membrane foulant layer and in suspension. J. Membr. Sci. 2010, 364, 331–338.

6379


Fouling of Nanofiltration, Reverse Osmosis, and Ultrafiltration

Recommend Documents