Effect of solution chemistry on the surface property of reverse osmosis membranes under seawater conditions

Transcription

1 Desalination 247 (2009) Effect of solution chemistry on the surface property of reverse osmosis membranes under seawater conditions Juhee Yang, Sangyoup Lee, Eunsu Lee, Joohee Lee, Seungkwan Hong* Department of Civil, Environmental and Architectural Engineering, Korea University, 1, 5-ka, Anam-Dong, Sungbuk-Gu, Seoul , Korea Tel ; Fax: ; Received 13 November 2008; revised 17 December 2008; accepted 24 December 2008 Abstract Recent studies have shown that the rougher, less negatively charged, and more hydrophobic membranes are prone to organic fouling. These surface characteristics of RO membranes, however, have been usually determined at very low TDS conditions, while seawater contains ten to thousand times more TDS than surface and even brackish waters. In this study, three aromatic polyamide thin-film composite (TFC) RO membranes were characterized for chemical and physical properties. Membrane characterization was performed under both the low (i.e., 10 mg/l) and high (i.e., 35,000 mg/l) TDS conditions to see how these surface characteristics are affected by seawater-level TDS. Results showed that both the chemical and physical surface properties were altered significantly under the high TDS condition with becoming more favorable to membrane fouling, namely, less negatively charged, more hydrophobic, and rougher. Mechanisms responsible for these changes such as charge screening and osmotic swelling are delineated. The way of changing in surface characteristics under the high TDS condition was substantially different with respect to the surface characteristics determined at the low TDS condition. It was confirmed that the chemical and physical properties were interrelated and, thus, variations in chemical properties with respect to the solution chemistry applied led to changes in physical properties and vice versa. Keywords: Membrane surface characterization; Seawater TDS; RO membranes; Surface charge; Hydrophobicity; Surface roughness *Corresponding author. Presented at the 2nd joint workshop between the Center for Seawater Desalination Plant and the European Desalination Society, Gwangju Institute of Science and Technology, Korea, October 8 9, /09/$ See front matter 2009 Published by Elsevier B.V. doi: /j.desal

2 J. Yang et al. / Desalination 247 (2009) Introduction Desalination based on reverse osmosis (RO) is exponentially being in the spotlight of treatment alternatives to overcome worldwide water scarcity and the lack of safe water. Although RO desalination is currently the most promising technology, there are still several challenges to be solved for efficient application. Among the major challenges of RO desalination is membrane fouling. There are several factors affecting membrane fouling such as foulant properties, membrane characteristics, and feed water chemistries [1 3]. Among these factors, membrane characteristics are first concerned practically since the others are rather natural factors with bearing difficulties in handling. Systematic and rigorous characterization of RO membranes could give useful insight into the better understanding of fouling phenomena, leading to an efficient control of membrane fouling. Surface characteristics are of paramount importance in RO membranes since polymeric TFC RO membranes are considered to be non-porous and, thus, the adhesion of foulants on the non-porous membrane surface is a key fouling mechanism [4,5]. Important membrane surface characteristics affecting the foulant adhesion on the membrane surface are surface roughness, charge, and hydrophobicity. Several researches have been performed to investigate RO membrane surface characteristics as well as find out the relationship between the membrane surface characteristics and the rate and extent of membrane fouling [6,7]. It has been shown that RO membrane surface roughness plays an important role in colloidal fouling [8,9]. The RO membranes with rougher surface were prone to colloidal fouling as the valleys created by the rough surface produced wells of low interaction energy in which colloidal particles preferentially deposited [10]. In case of surface charge, it has been known that RO membranes with negative surface charge exhibited low fouling tendency. Most foulants were negatively charged and, thus, the electrostatic repulsion between evenly charged foulants and membrane surface prevented foulant adhesion [11 13]. It has been also known that hydrophobic membranes more severely suffered from membrane fouling than hydrophilic membranes due to the strong hydrophobic interaction, which could allow multi-fouling layers on the membrane surface [14 16]. Therefore, the smoother, more negatively charged, and less hydrophobic membranes have been considered to be enviable for reducing organic fouling of RO membranes. The most eminent feature of RO desalination compared with other membrane-based water treatments such as drinking water treatment, wastewater reuse, and brackish water treatment is that seawater contains ten to thousand times higher TDS than surface, waste, and even brackish waters. Most previous researches dealing with surface characterization of RO membranes, however, have been carried out under very low TDS conditions significantly lower than seawater TDS. Therefore, there is a possibility of misunderstanding during the analyses of RO membrane surface characteristics since the aforementioned characteristics (i.e., roughness, charge, and hydrophobicity) may change upon high TDS environment. The altered surface characteristics due to high TDS conditions could affect foulant membrane interactions resulting in subsequent alteration in the rate and extent of membrane fouling. Therefore, it is of paramount importance to determine RO membrane surface characteristics at seawaterlevel TDS condition as well as investigate how these altered characteristics affect the rate and extent of membrane fouling. In this study, the widely used commercial RO membranes were characterized under two substantially different TDS levels of 10 and

3 150 J. Yang et al. / Desalination 247 (2009) ,000 mg/l. The latter TDS level is equivalent to common seawater ionic strength of 0.6 M. The surface characteristics investigated were divided by chemical and physical properties. The former included surface charge and hydrophobicity determined using zeta potential and contact angle measurements, respectively. The latter included surface roughness and roughness heterogeneity analyzed by atomic force microscopy (AFM) in conjunction with an image analysis system. Zeta potential measurements were performed only at 10 mg/l TDS condition with ph variation since no more than 10 mm of background electrolyte concentration is allowable within the current state of instrumental technique. It may be noteworthy that the method for determining membrane charge characteristics even at seawater-level TDS condition is being developed in our on-going study. The effect of solution ph on the membrane surface characteristics was investigated where both 10 and 35,000 mg/l TDS conditions were employed. The main idea of this study is to see if there are influential changes in RO membrane surface characteristics under seawater-level TDS condition as well as how these changes affect the foulant membrane interactions. Since there has been no study on analyzing RO membrane surface characteristics under seawaterlevel TDS condition, the results obtained in this study are expected to be useful for further studies of investigating the influence of RO membrane surface characteristics on the rate and extent of organic fouling during seawater desalination. 2. Materials and methods 2.1. RO membranes Three commercial RO membranes were used in this study. The RO membranes were Toray TM-820 (Chiba, Japan), Hydranautics SWC-5 (Oceanside, CA), and Dow-Filmtec SW-30HR (Minneapolis, MA). All membranes were polyamide TFC RO membranes with an average salt rejection over 99.5%. All membranes were stored in deionized (DI) water at 4 C with water replaced regularly prior to each measurement. The membranes were first characterized in a conventional way (i.e., low TDS condition) for chemical and physical properties such as surface roughness, zeta potential, and contact angle. The results are summarized in Table 1. The influence of seawater-level TDS as well as ph variation on these surface characteristics is discussed later in this study. As listed in Table 1, the surface characteristics differed with each membrane and the most substantial difference was observed in the membrane hydrophobicity (i.e., TM-820 and SWC-5 are relatively hydrophobic, while SW-30HR is noticeably hydrophilic). Table 1 Surface characteristics of the membranes determined under ambient conditions a Membrane RMS roughness b (nm) Surface charge c (mv) Contact angle ( ) TM SWC SW-30HR a Test solution conditions: ph = 5.5, Temperature = 25 C, and TDS = 10 mg/l NaCl. b See the description in Section 2.3. c Zeta potential was determined at a background electrolyte concentration of 10 mm KCl.

4 J. Yang et al. / Desalination 247 (2009) Solution chemistry All the surface characterizations including AFM image analysis, zeta potential, and contact angle measurements were performed in liquid phase where the test solution TDS as well as ph were varied as needed. Two substantially different TDS values of 10 and 35,000 mg/l were employed where NaCl was used for TDS adjustment. The higher TDS of 35,000 mg/l is approximately equivalent to the seawater TDS. The test solution ph was adjusted to be 4.0, 7.0, and 10.0 using NaOH or HCl stock solutions. The test solution ph was fixed at 7.0 when investigating the effect of TDS on membrane surface characteristics. The effect of solution ph on membrane surface characteristics was investigated under both the low (i.e., 10 mg/l) and high (i.e., 35,000 mg/l) TDS conditions to see how TDS affects the ph impact on membrane surface characteristics Surface morphology Surface roughness Membrane surface roughness was determined by AFM imaging and analysis (PUCOStation AFM, Surface Imaging Systems, Herzogenrath, Germany). Liquid phase AFM imaging was performed in contact mode with silicon probes of which backside has a 30 nm thick aluminium reflex coating for better resolution and stability in liquid phase applications (APPNANO, Applied Nano Str uctures, Inc., Santa Clara, CA). The probe has a spring constant of 0.1 N/m (±0.08 N/m), resonance frequency of 28 khz (±10 μ m), tip radius of 5 6 nm, tip height of 14 μm (±2 μm), and cantilever length of 225 μm (±10 μm). The RO membranes were immersed in a liquid cell containing pre-adjusted test solution in terms of TDS and ph. All membranes were scanned three times with randomly selecting a scan position. Membranes surface roughness was quantified by root mean square (RMS) roughness, which is the RMS deviation of the peaks and valleys from the mean plane. Approaching force ranged from 4.0 to 6.0 N/m with a scan speed of 0.7 line/s and scan area of μm. Scanned images were analyzed using SPIP software (Surface Imaging Systems, Herzogenrath, Germany). Each image was flattened by a baseline prior to roughness analyses Surface heterogeneity In this study, surface heterogeneity was chosen as one of physical surface characteristics in addition to surface roughness. Scanned images were divided by nine sections with each section having the same surface area (see Fig. 1). The line correction followed by RMS calculation was performed with each section. This procedure was also performed three times by differing scan positions and, hence, totally 27 RMS roughness was obtained for each membrane. The standard deviation of 27 RMS roughnesses for each membrane was defined as the surface heterogeneity. The higher standard deviation can be obtained with more heterogeneous membrane surface Contact angle Contact angle measurements were performed with a goniometer (DM 500, Kyowa Interface Science, Japan). Equilibrium contact angle measurements as described by Marmur [17] were adopted. Equilibrium contact angle was the average of the left and right contact angles. Ten measurements for each membrane were carried out. The reported values are the average of 10 equilibrium contact angles.

5 152 J. Yang et al. / Desalination 247 (2009) μm 3.3 μm 10 μm 3.3 μm Fig. 1. Schematic illustration of the method for quantifying membrane surface heterogeneity Zeta potential Membrane zeta potential was determined by a streaming current electrokinetic analyzer (SurPass, Anton Paar GmbH, Graz, Austria) following the procedure described by Luxbacher [18]. Zeta potential value was calculated based on the Fairbrother and Mastin substitution [19]. For surface zeta potential analysis, 0.01 M KCl was used as a background electrolyte solution and solution ph was varied from 2 to 10. The operating pressure ranged from 0 to 500 milibar (mbar) and the temperature was about 25 C. 3. Results and discussion The influence of TDS on membrane surface characteristics is systematically investigated by dividing the surface characteristics into chemical and physical properties. The chemical properties include surface charge and hydrophobicity, while surface roughness and heterogeneity are classified as the physical property. In all cases except for zeta potential measurement, both the low (i.e., 10 mg/l) and high (i.e., 35,000 mg/l) TDS conditions were employed during each measurement, and the results were compared to see how the seawater-level TDS affects the membrane surface characteristics. When comparing the data obtained under the low and high TDS conditions, the solution ph was fixed at 7.0. Later, ph variation (i.e., 4.0, 7.0, and 10.0) was employed during each measurement where either the low or high TDS condition was employed. This allows determining how TDS affects the ph impact on membrane surface characteristics. In case of characterizing the membrane surface charge, measurements were performed by varying solution ph under the low TDS condition only due to the current instrumental limitation (i.e., no more than 10 mm ionic strength is applicable in most zeta potential analyzers). It is true that membrane charge character reduces with increasing the ionic strength of background electrolyte due to charge screening. It may be noted that our

6 J. Yang et al. / Desalination 247 (2009) on-going study focuses on developing a technique to determine charge characteristics even at the seawater-level TDS condition using a dynamic hysteresis method [20]. In the subsequent sections, results on the membrane chemical property followed by physical property are discussed, and each section includes, first, TDS effects and, then, ph effects Chemical property The results from zeta potential analysis for the three RO membranes are shown in Fig. 2. As shown in Fig. 2, TM-820 and SWC-5 membranes exhibited similar zeta potential over the ph range investigated (i.e., ph 2 10) with the almost identical isoelectric point (IEP) at ph 4.0. SW-30HR membrane was more negatively charged with no IEP point over the ph range investigated. It is interesting to note that TM-820 and SWC-5 membranes also showed similar hydrophobicity (see Table 1), while SW- 30HR membrane was much more hydrophilic compared to the other membranes. The possible explanations on the close connection between surface charge and hydrophobicity as well as the effect of TDS on this interrelationship are addressed later in this study Effects of solution TDS Contact angles of the RO membranes at different TDS conditions (i.e., 10 and 35,000 mg/l) were compared in Fig. 3. It is clearly TM-820 SWC-5 SW-30HR 0 Zeta potential (mv) Fig. 2. Membrane surface zeta potential plotted as a function of solution ph at a background electrolyte concentration of 10 mm KCl. Solution temperature was maintained at 25 C. ph

7 154 J. Yang et al. / Desalination 247 (2009) TDS 10 TDS 35, Contact angle ( ) TM-820 SWC-5 SW-30HR Fig. 3. Comparison of membrane contact angles determined at different TDS conditions. (ph = 7.0 and temperature = 20 C). shown that the contact angles at TDS of 35,000 mg/l are higher than those at 10 mg/l. The increase in hydrophobicity at the higher TDS conditions was further confirmed with dynamic contact angle measurements demonstrating that the advancing contact angles for all membranes increased with increasing the solution TDS. This implies that the RO membranes turn to be more hydrophobic under seawater condition. The increase in hydrophobicity at the higher TDS condition is attributed to electrostatic screening, which leads to the electrostatic double layer compression at the solid liquid interface. This results in the charge reduction at the polymer skin layer of the membranes causing the membranes to posses more non-polar character. Another possible reason for the increase in hydrophobicity at the higher TDS condition is associated to polymer wettability. Higher salinity could reduce the degree of polymer wettability [21], which is directly related to the increase in contact angle of the polyamide TFC membranes. The increase in hydrophobicity is more obvious for SW-30HR membrane compared to TM-820 and SWC-5 membranes. This observation is well accordance with the fact that the effects of electrostatic screening and wettability reduction due to the higher TDS condition tend to be more influential on the membranes with more charged and hydrophilic surfaces. As listed in Table 1, SW-30HR is more hydrophilic and negatively charged membrane than the others. Consequently, desalination

8 J. Yang et al. / Desalination 247 (2009) using RO membranes even with relatively hydrophilic surface could suffer from intense membrane fouling as the membrane hydrophobicity tends to increase under seawater condition, leading to an accelerated adhesion of foulants on the membrane surface due to the enhanced hydrophobic interaction Effects of solution ph Contact angles of the RO membranes at different ph conditions (i.e., ph 4.0, 7.0, and 10.0) were compared in Fig. 4. Note that ph variation was conducted under either the low (i.e., 10 mg/l) or high (i.e., 35,000 mg/l) TDS conditions. The result showed that the contact angles of all membranes generally decreased with increasing solution ph, implying the decrease in membrane hydrophobicity. This observation is opposite to the effect of TDS on membrane hydrophobicity, where the increase in TDS results in the increase in membrane hydrophobicity due to charge screening. Regarding to ph variation, however, the polymer skin layer of the membranes tends to posses more polar character with increasing ph as the deprotonation of functional groups (i.e., mostly carboxylic) in the polymer matrix was accelerated with increasing solution ph. Consequently, membrane TM-820 (low TDS) TM-820 (high TDS) SWC-5 (low TDS) SWC-5 (high TDS) SW-30HR (low TDS) SW-30HR (high TDS) 80 Contact angle ( ) Fig. 4. Membrane contact angles as a function of solution ph. TDS = 10 mg/l (closed symbol), TDS = 35,000 mg/l (open symbol), and temperature = 25 C. ph

9 156 J. Yang et al. / Desalination 247 (2009) hydrophobicity decreased by the increase in membrane surface polarity. It is noticeable that changes in membrane hydrophobicity with ph variation was less obvious under the high TDS condition (i.e., 35,000 mg/l) for SW-30HR membrane of which surface charge and hydrophobicity was substantially different from those of TM-810 and SWC-5 membranes as shown in Table 1. This observation leads to the supposition that effect of solution ph on the chemical property of the membrane surface is less influential under seawater-level TDS condition, especially for the membranes with highly charged and hydrophilic surfaces. In case of TM-810 and SWC-5 membranes, which were less charged and more hydrophobic than SW-30HR membrane, the effect of solution ph on membrane hydrophobicity showed a similar trend regardless of the solution TDS. As discussed earlier, a possession of more polar character with increasing ph results in the increase in membrane wettability and, thus, the decrease in membrane contact angle. As a result, it is clear that there is an interplay between membrane surface charge and hydrophobicity, and both properties are simultaneously affected each other by the solution TDS and ph Physical property Effects of solution TDS The AFM images of the three RO membranes obtained under both the low (i.e., 10 mg/l) and high (i.e., 35,000 mg/l) TDS conditions are shown in Fig. 5. It is clearly visualized that TM-820 (i.e., Fig. 5(a) and 5(b)) and SWC-5 (Fig. 5(c) and 5(d)) membranes show rougher surfaces (i.e., increase in bright-spot area) at the higher TDS condition while slight decrease in surface roughness (i.e., increase in dark-spot area) is observed for SW-30HR membrane. The increase in surface roughness of TM-820 and SWC-5 membranes at the higher TDS condition is probably related to the osmotically driven polymer swelling. The swollen membrane surface, consequently, results in the increased peak and valley structure on the membrane surface. In addition to the increase in peak and valley structure, the structural deviation was also amplified, leading to an increase in surface heterogeneity (see Fig. 6). These observations are well accordance with other studies showing the changes in surface roughness of polymeric thin films by the osmotic swelling [22,23]. Interestingly, however, both the surface roughness and heterogeneity of SW-30HR membrane decreased at the higher TDS conditions (see Figs. 5(a), (b), and Fig. 6 for SW- 30HR membrane). These opposite changes in surface roughness and heterogeneity upon TDS increase for SW-30HR membrane are also due to the substantially different chemical surface properties (i.e., more negative and hydrophilic) of SW-30HR membranes compared to the other membranes (i.e., less negative and hydrophobic). There have been some studies showing the dependence of swelling behavior on the nature of polymers [24,25]. The different chemical properties of SW-30HR membrane compared to the other membranes are expected to affect the cross-linking density of polymer skin layer, which is an important factor for governing the degree of membrane swelling. The cross-linking density of polymer skin layer tends to increase at high TDS condition for the membrane with more negatively charged and hydrophilic due to the favorable wetting followed by charge screening. Charge screening following a complete wetting at the solid liquid interface reduces the electrostatic repulsive force between charged functional groups in polymer backbone and, thus, resulting in denser and more compact networking structure in the polymer skin layer. It should be addressed, however, that the present results are not further investigated with different

10 J. Yang et al. / Desalination 247 (2009) (a) (b) (c) (d) (e) (f) Fig. 5. AFM images of (a) TM-820 (TDS = 10 mg/l); (b) TM-820 (TDS = 35,000 mg/l); (c) SWC-5 (TDS = 10 mg/l); (d) SWC-5 (TDS = 35,000 mg/l); (e) SW-30HR (TDS = 10 mg/l); and (f) SW-30HR (TDS = 35,000 mg/l). Measurements were performed in liquid phase at ph 7.0 and the temperature was 25 C. membranes and, hence, more experimental work is required to confirm the current discussions Effects of solution ph The dependence of membrane surface roughness on the solution ph was investigated under the either low or high TDS conditions, and the results are shown in Fig. 7. It has been shown that the swelling of ionic polymer film was closely related to the solution ph, where swelling decreased with increasing ph until ph reaches to the IEP point of ionic polymer (i.e., neutralization occurs) and, then increased with further increase in ph [26]. The negative surface charge of polyamide TFC RO membranes is mostly attributed to the presence of carboxylic groups in the polymer backbone. The higher the ph above the IEP point in

11 158 J. Yang et al. / Desalination 247 (2009) TDS 10 TDS 35, Heterogeneity (nm) TM-820 SWC-5 SW-30HR Fig. 6. Membrane surface heterogeneity at different TDS conditions (i.e., 10 and 35,000 mg/l). Other experimental conditions were the same as those in Fig. 5. Note that the method for quantifying the relative surface heterogeneity is described in Section terms of surface zeta potential (see Fig. 2), the more the carboxylic groups are ionized. This leads to a favorable association with water molecules in the cross-linked polymer network and, consequently, an enhance swelling. Similarly, TM-820 membrane (see Fig. 7(a)) shows the increase in surface roughness due to enhanced swelling with increasing ph. In addition, the influence of ph on the changes in surface roughness was comparable for both the low and high TDS conditions. This is attributed to the interplay between ph and TDS effects. At the low TDS condition, the osmotic swelling is unfavorable, thus the ionization of functional groups through ph change plays a key role in membrane swelling. It is interesting to note that, at the high TDS condition, not only the osmotic swelling but also charge screening is favorable and, thus, these two factors on swelling could be compensated as these factors affect the membrane swelling oppositely. The reason for the negligible dependence of membrane surface roughness on ph variation for SWC-5 (see Fig. 7(b)) and SW-30HR (see Fig. 7(c)) membranes is somewhat ambiguous. As discussed earlier, this is also probably due to the differences in chemical and physical properties of each membranes as well as the interplay between both properties. Up to this point, it is concluded that the physical property such as surface roughness and heterogeneity is affected by the alteration in chemical property of the membrane induced by the variation in solution chemistry.

12 J. Yang et al. / Desalination 247 (2009) (a) 160 Roughness (nm) (b) 160 Roughness (nm) (c) 160 Roughness (nm) TDS 10 TDS 35,000 TDS 10 TDS 35,000 TDS 10 TDS 35,000 ph 4 ph 7 ph 10 ph 4 ph 7 ph 10 ph 4 ph 7 ph 10 TM-820 SWC-5 SW-30HR Fig. 7. Membrane surface roughness as a function of ph at different TDS conditions (i.e., 10 and 35,000 mg/l): (a) TM-820, (b) SWC-5, and (c) SW-30HR. Other experimental conditions were the same as those in Fig Conclusions Systematic characterizations of desalination RO membrane surface were conducted to investigate the effect of solution chemistry on the chemical and physical properties of membrane surface. Emphasis was placed on the role of seawater-level TDS in the alteration of membrane surface characteristics since the most previous characterization data were obtained under very low TDS conditions. All membranes investigated in this study exhibited different chemical and physical surface properties compared with those obtained at low TDS condition. Among the chemical (i.e., surface charge and hydrophobicity) and physical (i.e., surface roughness and heterogeneity) surface properties investigated, the chemical property including surface charge and hydrophobicity became more favorable state to membrane fouling, namely, less negatively charged and more hydrophobic under the seawater-level TDS condition. Both surface roughness and heterogeneity increased at the high TDS conditions for the membranes with less charged and more hydrophobic surface, while opposite trend for the membrane with more charged and relatively hydrophilic surface. TDS as well as ph dependent charge screening and osmotic swelling were mainly responsible for these diverse changes in membrane surface characteristics. It was further confirmed that the interplay between chemical and physical surface properties played an important role in characteristic changes in membrane surface at the seawater-level TDS condition. It is suggested that the seawater-level TDS should be carefully considered when investigating the surface characteristics of RO membranes for better understanding of membrane fouling during desalination since a certain change in surface characteristics could affect the foulant membrane interactions substantially.

13 160 J. Yang et al. / Desalination 247 (2009) Acknowledgement The authors would like to thank the Ministry of Land, Transport and Maritime Affairs (MLTM) for supporting this study through Seawater Engineering & Architecture of High Efficiency Reverse Osmosis (seahero) program. References [1] M. Elimelech, X. Zhu, A.E. Childress and S. Hong, Rapid communication; Role of membrane surface morphology in colloidal fouling of cellulose acetate and composite aromatic polyamide reverse osmosis membranes, J. Membr. Sci., 127 (1997) [2] S. Hong and M. Elimelech, Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes, J. Membr. Sci., 132 (1997) [3] S. Lee and M. Elimelech, Relating organic fouling of reverse osmosis membranes to intermolecular adhesion forces, Environ. Sci. Technol., 40 (2006) [4] S. Lee, W.S. Ang and M. Elimelech, Fouling of reverse osmosis membranes by hydrophilic organic matter: implication for water reuse, Desalination, 187 (2006) [5] S. Lee and M. Elimelech, Salt cleaning of organic-fouled reverse osmosis membranes, Water Res., 41 (2007) [6] J.A. Brant, K.M. Johnson and A.E. Childress, Characterizing NF and RO membrane surface heterogeneity using chemical force microscopy, Colloids Surfaces A: Physicochem. Eng. Aspects, 280 (2006) [7] S. Al-Jeshi and A. Neville, An investigation into the relationship between flux and roughness on RO membranes using scanning probe microscopy, Desalination, 189 (2006) [8] M. Elimelech, X. Zhu, A. Childress and S. Hong, Role of membrane surface morphology in colloidal fouling of cellulose acetate and composite polyamide RO membranes, J. Membr. Sci., 127 (1997) [9] E.M. Vrijenhoek, S. Hong and M. Elimelech, Influence of membrane surface properties on initial rate of colloidal fouling of reverse osmosis and nanofiltration membranes, J. Membr. Sci., 188 (2001) [10] E.M.V. Hoek, S. Bhattacharjee and M. Elimelech, Effect of membrane surface roughness on colloid-membrane DLVO interactions, Langmuir, 19 (2003) [11] A.E. Childress and S.S. Deshmukh, Effect of humic substances and anionic surfactants on the surface charge and performance of reverse osmosis membranes, Desalination, 118 (1998) [12] Y. Shim, H.J. Lee, S. Lee, S.H. Moon and J. Cho, Effects of natural organic matter and ionic species on membrane surface charge, Environ. Sci. Technol., 36 (2002) [13] J.A. Brant, K.M. Johnson and A.E. Childress, Examining the electrochemical properties of a nanofiltration membrane with atomic force microscopy, J. Membr. Sci., 276 (2006) [14] C. Jucker and M.M. Clark, Adsorption of aquatic humic substances on hydrophobic ultrafiltration membranes, J. Membr. Sci., 97 (1994) [15] J.A. Brant and A.E. Childress, Colloidal adhesion to hydrophilic membrane surfaces, J. Membr. Sci., 241 (2004) [16] K. Boussu, A. Belpaire, A. Volodin, C. Van Haesendonck, P. Van der Meeren, C. Vandecasteele and B. Van der Bruggen, Influence of membrane and colloid characteristics on fouling of nanofiltration membranes, J. Membr. Sci., 280 (2007) [17] A. Marmur, Equilibrium contact angles: theory and measurement, Colloid Surf. A, 116 (1996) [18] T. Luxbacher, Electrokinetic characterization of flat sheet membranes by streaming current measurement, Desalination, 199 (2006) [19] F. Fairbrother and H. Mastin, Studies in electroendosmosis. Part 1, J. Chem. Soc. Trans., 125 (1924) [20] E. Lee, S. Lee, J. Yang and S. Hong, Membrane characterization by dynamic hysteresis, J. Membr. Sci., submitted for publication. [21] G.B. Kaggwa, S. Froebe, L. Huynh, J. Ralston and K. Bremmell, Morphology of adsorbed

14 J. Yang et al. / Desalination 247 (2009) poly mers and solid surface wettability, Langmuir, 21 (2005) [22] L. Bay, T. Jacobsen, S. Skaarup and K. West, Mechanism of actuation in conducting polymers: osmotic expansion, J. Phys. Chem., B, 105 (2001) [23] S.T. Dubas and J.B. Schlenoff, Swelling and smoothing of polyelectrolyte multilayers by salt, Langmuir, 17 (2001) [24] A. Singh and M. Mukherjee, Swelling dynamics of ultrathin polymer films, Macromolecules, 36 (2003) [25] R. Toomey, D. Freidank and J. Ruhe, Swelling behavior of thin, surface-attached polymer networks, Macromolecules, 37 (2004) [26] J. E. Elliott, M. Macdonald, J. Nie and C.N. Bowman, Structure and swelling of poly (acrylic acid) hydrogels: effect of ph, ionic strength, and dilution on the crosslinked polymer structure, Polymer, 45 (2004)