- Email: [email protected]
Interaction energy evaluation of the role of solution chemistry and organic foulant composition on polysaccharide fouling of microfiltration membrane bioreactors
Chemical Engineering Science 104 (2013) 1028–1035
Contents lists available at ScienceDirect
Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces
Interaction energy evaluation of the role of solution chemistry and organic foulant composition on polysaccharide fouling of microfiltration membrane bioreactors Yi Ding b, Yu Tian a,b,n, Zhipeng Li b, Haoyu Wang b, Lin Chen c a
State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (SKLUWRE, HIT), Harbin 150090, China School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China c School of Civil and Environmental Engineering and Singapore Membrane Technology Centre, Nanyang Technological University, Singapore 639798, Singapore b
H I G H L I G H T S
Effect of feed solution on polysaccharide fouling of MF membrane was studied. Polysaccharide fouling of MF membrane was evaluated by XDLVO theory. Measured interaction energies were related to fouling rate of MF membrane. Greater attraction and lower repulsion energies induced higher MF membrane fouling.
art ic l e i nf o
a b s t r a c t
Article history: Received 13 May 2013 Received in revised form 19 October 2013 Accepted 22 October 2013 Available online 28 October 2013
Microfiltration membrane bioreactors (MF MBRs) have been extensively applied in wastewater treatment. Polysaccharide has been known to contribute significantly to organic fouling of MF membranes. In this study, the influences of ionic strength, divalent cations (Ca2 þ and Mg2 þ ) and organic foulant composition on polysaccharide fouling of MF membranes were investigated. Interfacial interaction parameters analysis showed that the cohesion and adhesion energies of polysaccharide increased with higher ionic strength, presence of divalent cations, and higher mass ratio of polysaccharide to protein. Fouling experiments indicated that MF membrane fouling by polysaccharide was also enhanced with increasing ionic strength, divalent cations addition, and higher polysaccharide concentration. Measured interaction energies confirmed the trends of the fouling profiles. Since the MF membrane surfaces were not completely coated with polysaccharide and some of the clean membrane remained exposed after fouling, it was consistently shown that feed solutions that induced higher fouling rates were associated with greater attractive and lower repulsive interaction energies among the polysaccharide molecules and between polysaccharide molecules and clean MF membrane. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Microfiltration membrane Polysaccharide fouling Sodium alginate (SA) XDLVO theory Interaction energy
1. Introduction Microfiltration membrane bioreactors (MF MBRs) have been actively employed for municipal and industrial wastewater treatment applications (Arabi and Nakhla, 2010). The obstacle to more widespread use of membrane technology is membrane fouling, which decreased permeability and added operational cost (Bouhabila et al., 2001; Tian et al., 2011). Polysaccharide was assumed to be one of the major membrane foulants (Drews, 2010). Moreover, there n Corresponding author at: School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China. Tel.: þ 86 451 8628 3077, þ 86 13804589869; fax: þ 86 451 8628 3077. E-mail address: [email protected] (Y. Tian).
0009-2509/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ces.2013.10.036
was a linear relationship between fouling rate and polysaccharide concentration in the sludge supernatant (Rosenberger et al., 2006). Therefore, understanding polysaccharide fouling is critical for successful and widespread application of MF membrane in wastewater treatment. Polysaccharide fouling is a complicated process due to the complex interaction between polysaccharide and membrane and between polysaccharide molecules. It has been reported that for reverse osmosis (RO) membrane, since monolayer coverage on the RO membrane surface through foulant–membrane interactions was attained in a very short period of time, the rate and extent of organic fouling were determined by the foulant–foulant interactions (Lee and Elimelech, 2006). Thus the analysis of foulant– foulant cohesion could provide molecular-level understanding the
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
organic fouling mechanisms of RO membrane. However, it is worthwhile to note that the fouling behavior of MF membranes is likely to be very different from that of RO membranes. Pore blocking has been reported as an important fouling mechanism for porous MF membranes but is unlikely to be important for nonporous RO membranes (Ang and Elimelech, 2007; Wang and Tarabara, 2008). The fouling mechanisms of MF membranes have been attributed to membrane pore blocking and/or pore constriction during initial fouling, followed by cake formation on the membrane surface during long-term fouling (Güell and Davis, 1996). The formation of membrane fouling mainly consisted of two processes: first the foulants adsorbed to the clean membrane, and then the foulants adhered to the foulants on the membrane. Since a fraction of organic foulants could pass in MF membrane pores (Liang et al., 2007), the membrane surfaces were not completely coated with organic foulants and some of the clean membrane remained exposed after fouling in a period of time (Subramani et al., 2009). Hence, the polysaccharide fouling process of MF membrane including cake buildup and pore plugging may be controlled by the combined interaction energies among polysaccharide molecules and between polysaccharide molecules and MF membrane. During the MF operation, periodic membrane backwash and chemical cleaning were utilized to recover the membrane from “reversible” fouling (typically due to cake formation) and “irreversible” fouling (typically due to pore plugging), respectively. Furthermore, characterization of the physicochemical interactions between polysaccharide and MF membrane and between polysaccharide molecules would be helpful to increase the membrane cleaning efficiency. Thus, this study focused on interactions between polysaccharide and membrane surface and between polysaccharide molecules, which were important stages of the much more complex overall process of MF membrane fouling. These interactions could be interpreted through extended Derjaguin–Landau–Verwey–Overbeek (XDLVO) models (Kang et al., 2004; Subramani and Hoek, 2008; Wang et al., 2005). Sodium alginate (SA) has been used as a model foulant to study organic fouling processes (Lee and Elimelech, 2006). Moreover, the ionic strength, divalent cations, and organic composition played important roles in adhesion force between SA molecules in bulk solution and SA molecules on the RO membrane (Lee and Elimelech, 2006). Nevertheless, the influence of solution characteristics on MF membrane fouling and interaction energies among SA molecules and between the SA molecules and MF membrane had been scarcely assessed. In previous study (Childress and Elimelech, 1996; Shim et al., 2002), they observed that the membrane showed different characteristics with different ionic strengths and ionic species. Meanwhile, membrane foulants readily adsorbed to the membrane surface and markedly influenced the membrane surface charge. This meant that the membrane surface charge would be continuously changed with the foulants adsorbed to the membrane. Therefore, the interaction between membrane surface and foulants was hardly determined due to the continuously changed membrane surface charge. It is widely known that the formation of a fouling layer during initial stages of membrane filtration leads to subsequent fouling layer development on the membrane surface. In this study, the initial membrane characteristics were used to determine the membrane fouling tendency. The motivation of this research was, therefore, to contribute toward a better understanding the effect of feed solution characteristics on polysaccharide fouling of MF membrane based on XDLVO theory. Specifically, the changed fouling characteristics of polysaccharide with different ion strengths, divalent cations and feed foulant compositions were investigated by batch filtration experiments; the physicochemical interactions between polysaccharide and MF membrane and between polysaccharide molecules
1029
were characterized through XDLVO models. The results obtained from the bench-scale fouling experiments and the corresponding interaction energies were used to elucidate the fouling mechanisms of MF membranes.
2. Materials and methods 2.1. Organic foulants Sodium alginate (SA) and Bovine serum albumin (BSA) (SigmaAldrich, St. Louis, MO) were used as model organic foulants. Both BSA and SA have been used extensively to represent the more complex proteinaceous and heteropolysaccharide materials present in wastewater (de Kerchove and Elimelech, 2007; Kim and Hoek, 2007). Both organic foulants were received in powder form, and stock solutions (2 g/L SA and 10 g/L BSA) were prepared by dissolving the SA and BSA in deionized (DI) water. Mixing of the stock solutions was performed for over 24 h to ensure complete dissolution of foulants, followed by the filtration using a 0.45 μm cellulose acetate membrane. The filtered stock solutions were stored in sterilized glass bottles at 4 1C. Reagent grade chloride salts of sodium, calcium and magnesium were used to adjust ionic strength of feed water. Solution chemistries used for fouling experiments and interaction energy calculations included variations in electrolyte concentrations (NaCl) and divalent cations (Mg2 þ and Ca2 þ ). In all cases, identical solution chemistries were employed in both fouling experiments and interaction energy calculations. When investigating the effect of divalent cations (Mg2 þ or Ca2 þ ), the feed solution was amended to desired value by adding 0.1 M MgCl2 or CaCl2 stock solutions. Meanwhile, the total ionic strength was maintained constant by adjusting NaCl concentration. The organic foulant composition was varied as needed by adjusting the ratio of the SA to BSA concentration, while the total organic foulant concentration was kept constant. Measured physicochemical properties of organic foulants are listed in Table S1. 2.2. Microfiltration membranes Polyvinylidene fluoride (PVDF) is advantageous over other membrane materials due to its high mechanical strength and excellent chemical resistance (Liu et al., 2011). This relatively wellcharacterized membrane has been extensively applied in microfiltration for general separation purposes (Liu et al., 2011). Therefore, the commercial PVDF (millipore) flat sheet membrane with a nominal pore size of 0.22 μm was selected as a model membrane for the fouling experiments. The membrane characteristics are listed in Table S2. PVDF membranes were immersed in 75% (v/v) alcohol for ca. 2 h, ensuring the membranes were sufficiently wetted and degassed. Prior to use, all membranes were soaked in DI water for 24 h with several intermediate water changes to remove impurities or additives. 2.3. Filtration apparatus To investigate the influence of feed solution characteristics on fouling, the fouling propensities of SA with different ionic strengths, divalent cations and organic foulant compositions were evaluated using a stirred dead-end cell (MSC300, Mosu Corp., China) at room temperature (20 71 1C). The flat sheet PVDF membranes were employed for filtration with an effective membrane area of 19.62 cm2. Before each experiment, DI water was filtered through the membrane for 1 h to stabilize the filtration system. The filtration pressure was maintained constant at 10 kPa and the stirring speed in cell was set at 250 rpm throughout the
1030
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
experiments. Permeate flux data were continuously logged using a top-loading electronic balance (BL-1200S, Setra Systems, USA) connected to a personal computer.
where A ¼ 12π h0 ΔGLW h0 at the right-hand side of Eq. (5) is the qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi LW is Hamaker constant; ΔGLW r LW r LW r LW h0 ¼ 2 2 1 Þð r 3 2
2.4. Surface thermodynamics analysis
the LW energy per unit area between the surfaces; h0 is the minimum cut-off distance due to Born repulsion; r LW is the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Lifshitz–van der Waals component; 1=k ¼ ε0 εr Rg T=ð2F 2 I S Þ is
Surface tension components were determined from the extended Young equation using a contact angle approach (Van Oss, 2006). qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi LW þ ð1 þ cos θÞγ TOT ¼2 γ LW γ sþ γ l þ γ s γ lþ ð1Þ s γl l pffiffiffiffiffiffiffiffiffiffiffiffiffi
γ AB ¼ 2 γ þ γ
ð2Þ
γ TOT ¼ γ LW þ γ AB
ð3Þ
is the total surface tension, γ LW where θ is the contact angle, γ is the Lifshitz–van der Waals component, and γ þ and γ are the electron-acceptor and electron-donor components, respectively. The subscripts s and l represent the solid surface and the liquid, respectively.
2
the Debye length; T is the absolute temperature in Kelvins; ε0 is the permittivity of free space; εr is the dielectric constant; Rg is the gas constant; F is Faraday's constant; I s is the ionic strength; ξ1 and ξ3 represent the surface potentials of membrane and foulant, respectively; λ is the decay length of AB interaction; qffiffiffiffiffiffiqffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi r 1þ þ r 3þ r 2þ Þ ΔGAB r 2þ ð r 1 þ r 3 r 2 þ 2 r 2 h0 ¼2 qffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi 2 r 1þ r 3 r 1 r 3þ is the acid–base free energy per unit area between the surfaces at contact.
TOT
2.5. Extended DLVO theory The surface tension components were converted into free energies following the model described by (Brant and Childress, 2002). The XDLVO theory (Eq. (4)) describes the total interaction energy per unit area (E) of particle-surface in terms of Lishitz–van der Waals force (LW), electrostatic force (EL) energy and acid–base (AB) interaction energy. O EL AB EXDLV ¼ ELW 123 123 þ E123 þ E123
ð4Þ
where the subscripts 1–3 represent the foulant, water and membrane, correspondingly. The interaction energy per unit area for LW, EL and AB was estimated from Eqs. (5–7) (Van Oss, 2006) as the function of separation distance (h) ELW 123 ðhÞ ¼
A 12π h
2
"
EEL 123 ðhÞ ¼
ξ2 þ ξ2 1 εr ε0 kξ1 ξ3 1 3 ð1 coth khÞ þ sinh kh 2ξ 1 ξ 3
AB EAB 123 ðhÞ ¼ ΔGh0 exp
3. Results and discussion 3.1. Interfacial interaction parameters for organics and MF membranes
ð6Þ
The free energy of cohesion (ΔGcoh) is the interaction free energy (per unit area) when two surfaces of the same material are immersed in a solvent (water) and brought into contact (Brant and Childress, 2002). The free energy of cohesion gives a quantitative representation of the hydrophobicity–hydrophilicity of organics
ð7Þ
λ
The zeta potentials (ζ) and size distribution of the test solutions were monitored using a zetasizer (Zetasizer 3000 HS type A, Malverin, England). The size of SA with the condition of ionic strength ¼10 mM NaCl, pH ¼6.8 and temperature¼20 1C was used to calculate the interaction energy to determine the fouling tendency. Surface tensions of organic foulants and MF membranes were determined from contact angles of DI water, formamide and diiodomethane by the sessile drop on filtered lawns of organic foulants or on a clean MF membrane. Contact angle measurements were performed using an NRL Contact Angle Goniometer (Rame Hart, Mountain Lakes, NJ), which is a standard goniometer with image analysis attachments (i.e. video camera, computer with a monitor, and image analysis software). At least 12 contact angles were determined for each test solutions with the highest and lowest values discarded, and the average value was taken as the contact angle.
ð5Þ #
h0 h
2.6. Analytical items
Table 1 Calculated surface energetic parameters for SA with different ion strengths, divalent cations and BSA proportions. SA
γLW
γþ
γ
ΔGLW sls
ΔGAB sls
ΔGEL sls
ΔGcoh
10 mM NaCl 30 mM NaCl 50 mM NaCl 100 mM NaCl 0.05 mM Ca2 þ 0.1 mM Ca2 þ 0.3 mM Ca2 þ 0.5 mM Ca2 þ 1.0 mM Ca2 þ 0.5 mM Mg2 þ SA:BSA ¼ 3:7a SA:BSA ¼ 7:3a
38.23 39.51 40.41 41.94 39.71 40.65 41.42 42.98 43.93 39.05 36.63 39.26
1.68 1.25 1.00 0.59 1.41 1.14 0.88 0.42 0.27 1.39 0.96 1.47
6.53 4.97 3.95 3.65 4.02 3.54 3.09 2.4 1.47 6.08 22.61 5.45
0.3122 0.4956 0.6474 0.9472 0.5277 0.6911 0.8396 1.1796 1.4114 0.4252 0.1398 0.4567
37.4515 44.3561 49.6061 53.7646 46.9371 50.4647 54.141 61.6333 69.5343 40.0081 4.7986 41.6768
0.1064 0.1175 0.1231 0.1181 0.0792 0.0653 0.0543 0.043 0.0316 0.046 0.0376 0.0403
37.6573 44.7342 50.1304 54.5937 47.3856 51.0905 54.9263 62.7699 70.914 40.3872 4.9007 42.0932
pH¼ 6.8, temperature¼ 20 1C. a
Ionic strength¼ 0.5 mM CaCl2 þ8.5 mM NaCl.
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
1031
Table 2 Calculated surface energetic parameters for SA with different ion strengths, divalent cations and BSA proportions. SA
ΔGLW slm
ΔGAB slm
ΔGEL slm
ΔGadh
10 mM NaCl 30 mM NaCl 50 mM NaCl 100 mM NaCl 0.05 mM Ca2 þ 0.1 mM Ca2 þ 0.3 mM Ca2 þ 0.5 mM Ca2 þ 1.0 mM Ca2 þ 0.5 mM Mg2 þ SA:BSA ¼ 3:7a SA:BSA ¼ 7:3a
0.7508 0.9458 1.0811 1.3076 0.976 1.1169 1.2311 1.4592 1.5962 0.8761 0.5023 0.908
50.1626 54.4964 57.6114 59.9847 56.0536 58.1266 60.2117 64.2372 68.3187 51.839 31.7217 52.8356
0.0763 0.1056 0.1228 0.143 0.0659 0.0598 0.0545 0.0486 0.0416 0.0502 0.0454 0.047
50.837 55.3367 58.5697 61.1493 56.9637 59.1837 61.3883 65.6479 69.8732 52.6649 32.1786 53.6966
pH¼ 6.8, temperature¼ 20 1C. a
Ionic strength¼ 0.5 mM CaCl2 þ8.5 mM NaCl.
(Subramani et al., 2009). Positive values for ΔGcoh represent the degree of hydrophilicity and negative values represent the degree of hydrophobicity (van Oss, 2007). The calculated surface tension parameters and cohesion free energies for SA with different ion strengths, divalent cations and organic foulant compositions are listed in Table 1. It could be seen that the cohesion energy of SA became larger with increasing ionic strength and divalent ion addition. Hence, the hydrophobicity of SA was enhanced with the increasing ionic strength. Compared with that in the absence of divalent cations (10 mM NaCl), the cohesion energies of SA in the presence of 0.5 mM Ca2 þ and 0.5 mM Mg2 þ were increased by 66.7% and 7.2%, respectively. Therefore, the hydrophobicity of SA in the presence of divalent cations was higher compared to that in the absence of divalent cations (10 mM NaCl). Moreover, the SA in the presence of 0.5 mM Ca2 þ showed stronger hydrophobicity than that in the presence of 0.5 mM Mg2 þ . Meanwhile, the cohesion energy of the SA became weaker with the decreasing mass ratio of SA to BSA, indicating that hydrophobicity of SA was reduced. The hydrophilic/hydrophobic interaction between the organics and membranes may be of much importance since hydrophobic substances demonstrated stronger interaction with hydrophobic membrane, which was supported by previous researches (Maximous et al., 2009; Van der Bruggen et al., 2004). The interfacial free energy (per unit area) between organics and clean MF membrane at contact is accurately described by the free energy of adhesion (ΔGadh), as presented in Table 2. The SA with different ion strengths and divalent cations expressed significantly different characters. The adhesion energy between SA and MF membrane became more substantial with increasing ionic strength. Compared with the absence of divalent cations (10 mM NaCl), the adhesion energies between SA and MF membrane were increased by 29.1% and 3.6% in the presence of 0.5 mM Ca2 þ and 0.5 mM Mg2 þ , respectively. With varying proportions of SA to BSA, the adhesion energy between SA and MF membrane was reduced with decreasing SA concentration. 3.2. Influence of ionic strength on SA fouling and interaction energy between foulant and MF membrane surfaces The influence of ionic strength on the flux decline behavior during SA fouling of MF membrane and the corresponding interaction energy are presented in Fig. 1. As shown in Fig. 1a, the membrane permeate flux of SA at 10 mM NaCl declined by approximately 72.7% when 400 ml of permeate was collected, while SA at 30 and 50 mM NaCl resulted in ca. 75.7% and 76.9% decrement, respectively. With respect to the SA at 100 mM NaCl,
Fig. 1. Effect of ionic strength on (a) flux decline of MF membrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L and pH¼ 6.8). The ionic strength of the test solutions was adjusted by varying indifferent salt (i.e., NaCl) concentration.
the flux decline was ca. 77.9% over a period of cumulative volume 400 ml. It was clearly shown that flux decline of MF membrane became more significant with increasing ionic strength.
1032
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
SA fouling process might be controlled by the interactions between SA and clean MF membrane and among SA molecules. The XDLVO theory can provide a quantitative assessment of these interactions. The interaction energy profiles between SA molecules and clean MF membrane are shown in Fig. 1b. It could be observed that the interaction energy profile between SA molecules and clean MF membrane at 10 mM exhibited an energy barrier and a secondary energy minimum. The energy barrier means the foulants in suspension should have sufficient kinetic energy to overcome this barrier to approach the MF membrane (Redman et al., 2004). The secondary energy minimum represents the ability of foulants being sucked onto MF membrane surface (Hoek et al., 2003). The interaction energy profiles between SA molecules and clean MF membrane at 30, 50 and 100 mM showed entirely attractive interaction energy. Meanwhile, it can be seen that the attractive energy became much greater with increasing ionic strength. The interaction energies between SA molecules and SA molecules were also calculated. As seen in Fig. 1c, the interaction energy profile among SA molecules at 10 mM exhibited an energy barrier and a secondary energy minimum, and the interaction energies among SA molecules at 30, 50 and 100 mM were entirely attractive. The attractive energy between SA molecules and SA molecules became larger with increasing ionic strength. It could be inferred that with the increasing ionic strength, the compression of electric double layer due to charge screening reducing the electrostatic repulsion among SA molecules and between SA molecules and MF membrane surface (Fig. S1). This enabled van der Waals attraction to completely overwhelm electrostatic repulsion. Therefore, the ability of SA being sucked onto the MF membrane surface was enhanced with increasing ionic strength. The changes in interaction energies among SA molecules and between SA molecules and clean MF membrane could be related to the flux decline trend of fouling runs. It was possible that the MF membrane surfaces were not completely coated with SA and some of the clean MF membrane remained exposed after fouling. Thus, the SA fouling process was controlled by the combined interaction energies among SA molecules and between SA molecules and MF membrane. 3.3. Influence of divalent cations on SA fouling and interaction energy between foulant and MF membrane surfaces Recent studies have demonstrated that divalent cations have a dramatic effect on SA fouling (Braghetta et al., 1998; Hong and Elimelech, 1997; Yuan and Zydney, 2000). To investigate the influence of divalent ions on alginate fouling of MF membrane, fouling experiments were performed in the presence of CaCl2 and MgCl2, because they are the major divalent cations in wastewaters. In these experiments, the total ionic strength was kept constant at 10 mM by adjusting the background NaCl concentration. The permeate flux behavior during fouling of MF membrane and the corresponding interaction energy curves are presented in Fig. 2. As shown in Fig. 2a, the flux decline in the presence of Mg2 þ was much rapider compared to that in the absence of divalent cations. However, the fouling extent of SA with 0.5 mM Ca2 þ was more severe than the one observed with 0.5 mM Mg2 þ , even though both were divalent ions. The interaction energies between SA with different divalent cations and clean MF membrane surface were calculated (Fig. 2b). It was seen that all the interaction energy profiles of SA approaching the clean MF membrane exhibited an energy barrier and a secondary energy minimum. Compared to the absence of divalent cations (10 mM NaCl), the secondary energy minimum between SA and clean MF membrane surface was higher in the presence of Mg2 þ , but the energy barrier was lower. It could also been seen that the energy barrier of SA in the presence of Ca2 þ was much less than that determined in the presence of Mg2 þ , however, the secondary energy
Fig. 2. Effect of divalent cations on (a) flux decline of MF membrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L, total ionic strength¼ 10 mM, and pH¼6.8). The total ionic strength of the test solution is fixed at 10 mM by varying NaCl concentration.
minimum was stronger. The interaction energies among SA molecules with different divalent cations are plotted in Fig. 2c. It was noted that the cations had a similar effect on SA–MF membrane and SA–SA interactions, with the magnitude of the attractive energies
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
following the order of Ca2 þ 4Mg2 þ 4Na þ , and the repulsive energies following the order of Ca2 þ oMg2 þ oNa þ . This result was in agreement with the fouling rate of MF membrane observed in bench-scale fouling experiments, with the flux decline rate of MF membrane following the order of Ca2 þ 4Mg2 þ 4Na þ (Fig. 2a). According to previous study (Hong and Elimelech, 1997; Li and Elimelech, 2004), Na þ , Ca2 þ and Mg2 þ ions all played a role in charge screening (or double layer compression). Essentially, the effect of charge screening on reducing electrostatic repulsion among SA molecules is expected to be similar in all the three electrolyte solutions because the total ionic strength was kept constant. This difference in fouling profile of MF membrane in the presence and absence of divalent cations was probably attributed to other interaction apart from charge screening, which was most likely charge neutralization due to the divalent cations binding with SA molecules. Divalent cations of Ca2 þ and Mg2 þ were reported to form complex with carboxylic (–COO–) moieties of polysaccharide, which neutralized the negative charges carried by these macromolecules or bound them together via divalent ion bridging (Ang and Elimelech, 2007; Lee et al., 2005; Tang et al., 2009). Thus, the electrostatic repulsive energy in the presence of divalent cations was weakened, and the attractive energy was enhanced (Fig. S2a). Hence, the fouling rate of SA with divalent cations was higher compared to that without divalent cations. Calcium ions are known to be more favorable divalent cations to form complexes with SA molecules than magnesium ions, and thus, should be more effective in charge neutralization (Davis et al., 2003). Thus, the electrostatic repulsive energy in the presence of Ca2 þ was less than that in the presence of Mg2 þ (Fig. S2b). Meanwhile, ionic bridge between SA molecules in the presence of calcium resulted in the formation of a cross-linked organic gel layer on the membrane surface, which produced significant hydraulic resistance to permeate flow and, thus, a severe flux decline (Mo et al., 2008; Yang et al., 2010). As a result, the fouling layer on MF membrane formed in the presence of Ca2 þ was very compact and highly resistant to hydrodynamic forces. Based on above results that Ca2 þ had important impact on SA fouling of MF membrane, the influence of calcium ion concentration on SA fouling of MF membrane and the corresponding interaction energy were further investigated. Calcium concentration varied from 0.05 to 1.0 mM, keeping the total ionic strength constant at 10 mM. The results in Fig. 3a clearly show that the flux decline of MF membrane became greater with increasing Ca2 þ concentration. As seen in Fig. 3b, it was noted that the attractive interaction energy between SA and clean MF membrane increased significantly with increasing Ca2 þ concentration, but the repulsive interaction energy decreased correspondingly. The interaction energies between SA in bulk solution and SA on MF membrane surface (i.e., in the fouling layer) with different calcium ion concentrations were also calculated. As seen from Fig. 3c, the repulsive energy barrier between SA and SA molecules decreased significantly and the attractive energy increased remarkably with the increase of calcium ionic concentration. With increasing Ca2 þ concentration, the charge screening, charge neutralization and ionic bridge formed by Ca2 þ reduced electrostatic repulsion and increased attractive energies between SA molecules and MF membrane surface and among SA molecules (Fig. S3a and S3b). Thus, the stoichiometry between Ca2 þ and SA concentrations was also expected to play an important role in the interaction energies between SA molecules and MF membrane surface and among SA molecules. Obviously, the flux decline trends of MF membrane with different divalent cations and calcium ion concentrations can be related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane.
1033
Fig. 3. Effect of calcium concentration on (a) flux decline of MF mebrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L, total ionic strength¼ 10 mM, pH¼6.8).
3.4. Influence of organic foulant composition on SA fouling and interaction energy between foulant and MF membrane surfaces Fouling experiments were performed with different mass ratios of SA to BSA to investigate the influence of organic foulant
1034
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
(iii) 14 mg/L SA plus 6 mg/L BSA (SA to BSA mass ratio of 7:3), and (iv) 20 mg/L SA and no BSA. The permeate flux decline of MF membrane and the corresponding interaction energy for these foulant combinations are presented in Fig. 4. It was expected that the flux decline profiles of MF membrane would vary correspondingly to the proportions of SA to BSA. It can be seen from Fig. 4a that feed solution in the presence of 20 mg/L SA resulted in the fastest flux decline of MF membrane. The SA with a small amount of BSA (14 mg/L SA and 6 mg/L BSA) caused a slower flux decline of MF membrane compared with 20 mg/L SA. The flux decline of the feed solution containing more BSA (6 mg/L SA and 14 mg/L BSA) was much less than SA with a small amount of BSA (14 mg/L SA and 6 mg/L BSA). Accordingly, the permeate flux decline of MF membrane for SA fouling was more substantial compared to BSA fouling. When the test solutions contained both SA and BSA, the flux decline was more severe for higher SA concentration. The interaction energies between SA and clean MF membrane surface were calculated with different mass ratios of SA to BSA, which are presented in Fig. 4b. It was noted that the attractive interaction energy between SA and clean MF membrane decreased with decreasing SA concentration, but the repulsive interaction energy increased correspondingly. The interaction energies among SA molecules with varying proportions of SA to BSA are plotted in Fig. 4c. With the decrease of proportions of SA to BSA, the maximum attractive energy (in magnitude and range) was also weakened, and the maximum repulsive energy was significantly enhanced. This was because the BSA molecules interact with Ca2 þ , unlike SA molecules, were hard to form a gel-type fouling layer in the presence of Ca2 þ (Ang and Elimelech, 2007). We attributed the much greater attractive energy and lower repulsive energy with SA than BSA to the gel forming nature of SA, where intermolecular cohesion needed to be strong enough to sustain the structural integrity of the gel network. It could be confirmed from the decreased attractive AB energy and LW energy with the decreasing proportions of SA to BSA (Fig. S4). The results indicated that the flux decline trends of MF membrane were related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane with varying proportions of SA to BSA.
4. Conclusions
Fig. 4. Effect of organic foulant composition on (a) flux decline of MF mebrane during fouling, (b) interaction energies profiles between foulant and the clean MF membrane, and (c) interaction energies profiles between foulant molecules (total ionic strength¼10 mM, pH¼6.8). The total organic foulant concentration of the test solution is fixed at 20 mg/L by varying the ratio of SA to BSA concentration.
compositions on MF membrane fouling. Four foulant compositions were examined with the total foulant concentration being maintained at 20 mg/L: (i) 20 mg/L BSA and no sodium alginate, (ii) 6 mg/L SA plus 14 mg/L BSA (SA to BSA mass ratio of 3:7),
The effects of ionic strengths, divalent cations, and organic foulant compositions on MF membrane fouling by sodium alginate (SA) were investigated. Measured physicochemical properties revealed that the hydrophobicity of SA was increased due to the increase of ionic strength, addition of divalent cations and the higher mass ratio of SA to BSA. Fouling experiments had shown that the MF membrane fouling by SA was higher when the ionic strength was increased or calcium ion was present in the feed solution. In the presence of magnesium, MF membrane fouling by SA was not as severe as the one observed with calcium ions. Since the MF membrane surfaces were not completely coated with SA and some of the clean MF membrane remained exposed after fouling, the flux decline trends of SA fouling with increasing ionic strength and different divalent cations addition could be related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane. When the test solutions contained both SA and BSA, the flux decline of MF membrane was more substantial for higher SA concentration. This could also be predicted by the greater attractive energy and lower repulsive energy among SA molecules and between SA molecules and MF membrane with higher SA concentration.
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
Acknowledgments This study was supported by the Major Science and Technology Program for Water Pollution Control and Management (No. 2013ZX07201007), The State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (No. 2011DX01) and The Specialized Research Fund for the Doctoral Program of Higher Education (No. 20112302110060). The authors also appreciate the Funds for Creative Research Groups of China (No. 51121062).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ces.2013.10.036. References Ang, W.S., Elimelech, M., 2007. Protein (BSA) fouling of reverse osmosis membranes: implications for wastewater reclamation. J. Memb. Sci. 296, 83–92. Arabi, S., Nakhla, G., 2010. Impact of molecular weight distribution of soluble microbial products on fouling in membrane bioreactors. Sep. Purif. Technol. 73, 391–396. Bouhabila, E.H., Ben Aïm, R., Buisson, H., 2001. Fouling characterisation in membrane bioreactors. Sep. Purif. Technol. 22, 123–132. Braghetta, A., DiGiano, F.A., Ball, W.P., 1998. NOM accumulation at NF membrane surface: impact of chemistry and shear. J. Environ. Eng. 124, 1087–1098. Brant, J.A., Childress, A.E., 2002. Assessing short-range membrane–colloid interactions using surface energetics. J. Memb. Sci. 203, 257–273. Childress, A.E., Elimelech, M., 1996. Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. J. Memb. Sci. 119, 253–268. Davis, T.A., Volesky, B., Mucci, A., 2003. A review of the biochemistry of heavy metal biosorption by brown algae. Water Res. 37, 4311–4330. de Kerchove, A.J., Elimelech, M., 2007. Impact of alginate conditioning film on deposition kinetics of motile and nonmotile Pseudomonas aeruginosa strains. Appl. Environ. Microbiol.73, 5227–5234. Drews, A., 2010. Membrane fouling in membrane bioreactors—characterisation, contradictions, cause and cures. J. Memb. Sci. 363, 1–28. Güell, C., Davis, R.H., 1996. Membrane fouling during microfiltration of protein mixtures. J. Memb. Sci. 119, 269–284. Hoek, E.M., Bhattacharjee, S., Elimelech, M., 2003. Effect of membrane surface roughness on colloid–membrane DLVO interactions. Langmuir 19, 4836–4847. Hong, S., Elimelech, M., 1997. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. J. Memb. Sci. 132, 159–181. Kang, S.T., Subramani, A., Hoek, E., Deshusses, M.A., Matsumoto, M.R., 2004. Direct observation of biofouling in cross-flow microfiltration: mechanisms of deposition and release. J. Memb. Sci. 244, 151–165. Kim, S., Hoek, E., 2007. Interactions controlling biopolymer fouling of reverse osmosis membranes. Desalination 202, 333–342.
1035
Lee, S., Cho, J., Elimelech, M., 2005. A novel method for investigating the influence of feed water recovery on colloidal and NOM fouling of RO and NF membranes. Environ. Eng. Sci. 22, 496–509. Lee, S., Elimelech, M., 2006. Relating organic fouling of reverse osmosis membranes to intermolecular adhesion forces. Environ. Sci. Technol. 40, 980–987. Li, Q., Elimelech, M., 2004. Organic fouling and chemical cleaning of nanofiltration membranes: measurements and mechanisms. Environ. Sci. Technol. 38, 4683–4693. Liang, S., Liu, C., Song, L., 2007. Soluble microbial products in membrane bioreactor operation: behaviors, characteristics, and fouling potential. Water Res. 41, 95–101. Liu, F., Hashim, N.A., Liu, Y., Abed, M.R.M., Li, K., 2011. Progress in the production and modification of PVDF membranes. J. Memb. Sci. 375, 1–27. Maximous, N., Nakhla, G., Wan, W., 2009. Comparative assessment of hydrophobic and hydrophilic membrane fouling in wastewater applications. J. Memb. Sci. 339, 93–99. Mo, H., Tay, K.G., Ng, H.Y., 2008. Fouling of reverse osmosis membrane by protein (BSA): effects of pH, calcium, magnesium, ionic strength and temperature. J. Memb. Sci. 315, 28–35. Redman, J.A., Walker, S.L., Elimelech, M., 2004. Bacterial adhesion and transport in porous media: role of the secondary energy minimum. Environ. Sci. Technol. 38, 1777–1785. Rosenberger, S., Laabs, C., Lesjean, B., Gnirss, R., Amy, G., Jekel, M., Schrotter, J.-C., 2006. Impact of colloidal and soluble organic material on membrane performance in membrane bioreactors for municipal wastewater treatment. Water Res. 40, 710–720. Shim, Y., Lee, H.-J., Lee, S., Moon, S.-H., Cho, J., 2002. Effects of natural organic matter and ionic species on membrane surface charge. Environ. Sci. Technol. 36, 3864–3871. Subramani, A., Hoek, E., 2008. Direct observation of initial microbial deposition onto reverse osmosis and nanofiltration membranes. J. Memb. Sci. 319, 111–125. Subramani, A., Huang, X., Hoek, E., 2009. Direct observation of bacterial deposition onto clean and organic-fouled polyamide membranes. J. Colloid Interface Sci. 336, 13–20. Tang, C.Y., Kwon, Y.-N., Leckie, J.O., 2009. 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. Memb. Sci. 326, 526–532. Tian, Y., Chen, L., Jiang, T., 2011. Characterization and modeling of the soluble microbial products in membrane bioreactor. Sep. Purif. Technol. 76, 316–324. Van der Bruggen, B., Kim, J.H., DiGiano, F.A., Geens, J., Vandecasteele, C., 2004. Influence of MF pretreatment on NF performance for aqueous solutions containing particles and an organic foulant. Sep. Purif. Technol. 36, 203–213. Van Oss, C.J., 2006. Interfacial Forces in Aqueous Media, 2nd ed. Taylor & Francis. Boca Raton. van Oss, C.J., 2007. Development and applications of the interfacial tension between water and organic or biological surfaces. Colloids Surf. B: Biointerfaces 54, 2–9. Wang, F., Tarabara, V.V., 2008. Pore blocking mechanisms during early stages of membrane fouling by colloids. J. Colloid Interface Sci. 328, 464–469. Wang, S., Guillen, G., Hoek, E.M.V., 2005. Direct observation of microbial adhesion to membranes. Environ. Sci. Technol. 39, 6461–6469. Yang, Q., Liu, Y., Li, Y., 2010. Control of protein (BSA) fouling in RO system by antiscalants. J. Memb. Sci. 364, 372–379. Yuan, W., Zydney, A.L., 2000. Humic acid fouling during ultrafiltration. Environ. Sci. Technol. 34, 5043–5050.
Contents lists available at ScienceDirect
Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces
Interaction energy evaluation of the role of solution chemistry and organic foulant composition on polysaccharide fouling of microfiltration membrane bioreactors Yi Ding b, Yu Tian a,b,n, Zhipeng Li b, Haoyu Wang b, Lin Chen c a
State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (SKLUWRE, HIT), Harbin 150090, China School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China c School of Civil and Environmental Engineering and Singapore Membrane Technology Centre, Nanyang Technological University, Singapore 639798, Singapore b
H I G H L I G H T S
Effect of feed solution on polysaccharide fouling of MF membrane was studied. Polysaccharide fouling of MF membrane was evaluated by XDLVO theory. Measured interaction energies were related to fouling rate of MF membrane. Greater attraction and lower repulsion energies induced higher MF membrane fouling.
art ic l e i nf o
a b s t r a c t
Article history: Received 13 May 2013 Received in revised form 19 October 2013 Accepted 22 October 2013 Available online 28 October 2013
Microfiltration membrane bioreactors (MF MBRs) have been extensively applied in wastewater treatment. Polysaccharide has been known to contribute significantly to organic fouling of MF membranes. In this study, the influences of ionic strength, divalent cations (Ca2 þ and Mg2 þ ) and organic foulant composition on polysaccharide fouling of MF membranes were investigated. Interfacial interaction parameters analysis showed that the cohesion and adhesion energies of polysaccharide increased with higher ionic strength, presence of divalent cations, and higher mass ratio of polysaccharide to protein. Fouling experiments indicated that MF membrane fouling by polysaccharide was also enhanced with increasing ionic strength, divalent cations addition, and higher polysaccharide concentration. Measured interaction energies confirmed the trends of the fouling profiles. Since the MF membrane surfaces were not completely coated with polysaccharide and some of the clean membrane remained exposed after fouling, it was consistently shown that feed solutions that induced higher fouling rates were associated with greater attractive and lower repulsive interaction energies among the polysaccharide molecules and between polysaccharide molecules and clean MF membrane. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Microfiltration membrane Polysaccharide fouling Sodium alginate (SA) XDLVO theory Interaction energy
1. Introduction Microfiltration membrane bioreactors (MF MBRs) have been actively employed for municipal and industrial wastewater treatment applications (Arabi and Nakhla, 2010). The obstacle to more widespread use of membrane technology is membrane fouling, which decreased permeability and added operational cost (Bouhabila et al., 2001; Tian et al., 2011). Polysaccharide was assumed to be one of the major membrane foulants (Drews, 2010). Moreover, there n Corresponding author at: School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China. Tel.: þ 86 451 8628 3077, þ 86 13804589869; fax: þ 86 451 8628 3077. E-mail address: [email protected] (Y. Tian).
0009-2509/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ces.2013.10.036
was a linear relationship between fouling rate and polysaccharide concentration in the sludge supernatant (Rosenberger et al., 2006). Therefore, understanding polysaccharide fouling is critical for successful and widespread application of MF membrane in wastewater treatment. Polysaccharide fouling is a complicated process due to the complex interaction between polysaccharide and membrane and between polysaccharide molecules. It has been reported that for reverse osmosis (RO) membrane, since monolayer coverage on the RO membrane surface through foulant–membrane interactions was attained in a very short period of time, the rate and extent of organic fouling were determined by the foulant–foulant interactions (Lee and Elimelech, 2006). Thus the analysis of foulant– foulant cohesion could provide molecular-level understanding the
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
organic fouling mechanisms of RO membrane. However, it is worthwhile to note that the fouling behavior of MF membranes is likely to be very different from that of RO membranes. Pore blocking has been reported as an important fouling mechanism for porous MF membranes but is unlikely to be important for nonporous RO membranes (Ang and Elimelech, 2007; Wang and Tarabara, 2008). The fouling mechanisms of MF membranes have been attributed to membrane pore blocking and/or pore constriction during initial fouling, followed by cake formation on the membrane surface during long-term fouling (Güell and Davis, 1996). The formation of membrane fouling mainly consisted of two processes: first the foulants adsorbed to the clean membrane, and then the foulants adhered to the foulants on the membrane. Since a fraction of organic foulants could pass in MF membrane pores (Liang et al., 2007), the membrane surfaces were not completely coated with organic foulants and some of the clean membrane remained exposed after fouling in a period of time (Subramani et al., 2009). Hence, the polysaccharide fouling process of MF membrane including cake buildup and pore plugging may be controlled by the combined interaction energies among polysaccharide molecules and between polysaccharide molecules and MF membrane. During the MF operation, periodic membrane backwash and chemical cleaning were utilized to recover the membrane from “reversible” fouling (typically due to cake formation) and “irreversible” fouling (typically due to pore plugging), respectively. Furthermore, characterization of the physicochemical interactions between polysaccharide and MF membrane and between polysaccharide molecules would be helpful to increase the membrane cleaning efficiency. Thus, this study focused on interactions between polysaccharide and membrane surface and between polysaccharide molecules, which were important stages of the much more complex overall process of MF membrane fouling. These interactions could be interpreted through extended Derjaguin–Landau–Verwey–Overbeek (XDLVO) models (Kang et al., 2004; Subramani and Hoek, 2008; Wang et al., 2005). Sodium alginate (SA) has been used as a model foulant to study organic fouling processes (Lee and Elimelech, 2006). Moreover, the ionic strength, divalent cations, and organic composition played important roles in adhesion force between SA molecules in bulk solution and SA molecules on the RO membrane (Lee and Elimelech, 2006). Nevertheless, the influence of solution characteristics on MF membrane fouling and interaction energies among SA molecules and between the SA molecules and MF membrane had been scarcely assessed. In previous study (Childress and Elimelech, 1996; Shim et al., 2002), they observed that the membrane showed different characteristics with different ionic strengths and ionic species. Meanwhile, membrane foulants readily adsorbed to the membrane surface and markedly influenced the membrane surface charge. This meant that the membrane surface charge would be continuously changed with the foulants adsorbed to the membrane. Therefore, the interaction between membrane surface and foulants was hardly determined due to the continuously changed membrane surface charge. It is widely known that the formation of a fouling layer during initial stages of membrane filtration leads to subsequent fouling layer development on the membrane surface. In this study, the initial membrane characteristics were used to determine the membrane fouling tendency. The motivation of this research was, therefore, to contribute toward a better understanding the effect of feed solution characteristics on polysaccharide fouling of MF membrane based on XDLVO theory. Specifically, the changed fouling characteristics of polysaccharide with different ion strengths, divalent cations and feed foulant compositions were investigated by batch filtration experiments; the physicochemical interactions between polysaccharide and MF membrane and between polysaccharide molecules
1029
were characterized through XDLVO models. The results obtained from the bench-scale fouling experiments and the corresponding interaction energies were used to elucidate the fouling mechanisms of MF membranes.
2. Materials and methods 2.1. Organic foulants Sodium alginate (SA) and Bovine serum albumin (BSA) (SigmaAldrich, St. Louis, MO) were used as model organic foulants. Both BSA and SA have been used extensively to represent the more complex proteinaceous and heteropolysaccharide materials present in wastewater (de Kerchove and Elimelech, 2007; Kim and Hoek, 2007). Both organic foulants were received in powder form, and stock solutions (2 g/L SA and 10 g/L BSA) were prepared by dissolving the SA and BSA in deionized (DI) water. Mixing of the stock solutions was performed for over 24 h to ensure complete dissolution of foulants, followed by the filtration using a 0.45 μm cellulose acetate membrane. The filtered stock solutions were stored in sterilized glass bottles at 4 1C. Reagent grade chloride salts of sodium, calcium and magnesium were used to adjust ionic strength of feed water. Solution chemistries used for fouling experiments and interaction energy calculations included variations in electrolyte concentrations (NaCl) and divalent cations (Mg2 þ and Ca2 þ ). In all cases, identical solution chemistries were employed in both fouling experiments and interaction energy calculations. When investigating the effect of divalent cations (Mg2 þ or Ca2 þ ), the feed solution was amended to desired value by adding 0.1 M MgCl2 or CaCl2 stock solutions. Meanwhile, the total ionic strength was maintained constant by adjusting NaCl concentration. The organic foulant composition was varied as needed by adjusting the ratio of the SA to BSA concentration, while the total organic foulant concentration was kept constant. Measured physicochemical properties of organic foulants are listed in Table S1. 2.2. Microfiltration membranes Polyvinylidene fluoride (PVDF) is advantageous over other membrane materials due to its high mechanical strength and excellent chemical resistance (Liu et al., 2011). This relatively wellcharacterized membrane has been extensively applied in microfiltration for general separation purposes (Liu et al., 2011). Therefore, the commercial PVDF (millipore) flat sheet membrane with a nominal pore size of 0.22 μm was selected as a model membrane for the fouling experiments. The membrane characteristics are listed in Table S2. PVDF membranes were immersed in 75% (v/v) alcohol for ca. 2 h, ensuring the membranes were sufficiently wetted and degassed. Prior to use, all membranes were soaked in DI water for 24 h with several intermediate water changes to remove impurities or additives. 2.3. Filtration apparatus To investigate the influence of feed solution characteristics on fouling, the fouling propensities of SA with different ionic strengths, divalent cations and organic foulant compositions were evaluated using a stirred dead-end cell (MSC300, Mosu Corp., China) at room temperature (20 71 1C). The flat sheet PVDF membranes were employed for filtration with an effective membrane area of 19.62 cm2. Before each experiment, DI water was filtered through the membrane for 1 h to stabilize the filtration system. The filtration pressure was maintained constant at 10 kPa and the stirring speed in cell was set at 250 rpm throughout the
1030
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
experiments. Permeate flux data were continuously logged using a top-loading electronic balance (BL-1200S, Setra Systems, USA) connected to a personal computer.
where A ¼ 12π h0 ΔGLW h0 at the right-hand side of Eq. (5) is the qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi LW is Hamaker constant; ΔGLW r LW r LW r LW h0 ¼ 2 2 1 Þð r 3 2
2.4. Surface thermodynamics analysis
the LW energy per unit area between the surfaces; h0 is the minimum cut-off distance due to Born repulsion; r LW is the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Lifshitz–van der Waals component; 1=k ¼ ε0 εr Rg T=ð2F 2 I S Þ is
Surface tension components were determined from the extended Young equation using a contact angle approach (Van Oss, 2006). qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi LW þ ð1 þ cos θÞγ TOT ¼2 γ LW γ sþ γ l þ γ s γ lþ ð1Þ s γl l pffiffiffiffiffiffiffiffiffiffiffiffiffi
γ AB ¼ 2 γ þ γ
ð2Þ
γ TOT ¼ γ LW þ γ AB
ð3Þ
is the total surface tension, γ LW where θ is the contact angle, γ is the Lifshitz–van der Waals component, and γ þ and γ are the electron-acceptor and electron-donor components, respectively. The subscripts s and l represent the solid surface and the liquid, respectively.
2
the Debye length; T is the absolute temperature in Kelvins; ε0 is the permittivity of free space; εr is the dielectric constant; Rg is the gas constant; F is Faraday's constant; I s is the ionic strength; ξ1 and ξ3 represent the surface potentials of membrane and foulant, respectively; λ is the decay length of AB interaction; qffiffiffiffiffiffiqffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffi r 1þ þ r 3þ r 2þ Þ ΔGAB r 2þ ð r 1 þ r 3 r 2 þ 2 r 2 h0 ¼2 qffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi 2 r 1þ r 3 r 1 r 3þ is the acid–base free energy per unit area between the surfaces at contact.
TOT
2.5. Extended DLVO theory The surface tension components were converted into free energies following the model described by (Brant and Childress, 2002). The XDLVO theory (Eq. (4)) describes the total interaction energy per unit area (E) of particle-surface in terms of Lishitz–van der Waals force (LW), electrostatic force (EL) energy and acid–base (AB) interaction energy. O EL AB EXDLV ¼ ELW 123 123 þ E123 þ E123
ð4Þ
where the subscripts 1–3 represent the foulant, water and membrane, correspondingly. The interaction energy per unit area for LW, EL and AB was estimated from Eqs. (5–7) (Van Oss, 2006) as the function of separation distance (h) ELW 123 ðhÞ ¼
A 12π h
2
"
EEL 123 ðhÞ ¼
ξ2 þ ξ2 1 εr ε0 kξ1 ξ3 1 3 ð1 coth khÞ þ sinh kh 2ξ 1 ξ 3
AB EAB 123 ðhÞ ¼ ΔGh0 exp
3. Results and discussion 3.1. Interfacial interaction parameters for organics and MF membranes
ð6Þ
The free energy of cohesion (ΔGcoh) is the interaction free energy (per unit area) when two surfaces of the same material are immersed in a solvent (water) and brought into contact (Brant and Childress, 2002). The free energy of cohesion gives a quantitative representation of the hydrophobicity–hydrophilicity of organics
ð7Þ
λ
The zeta potentials (ζ) and size distribution of the test solutions were monitored using a zetasizer (Zetasizer 3000 HS type A, Malverin, England). The size of SA with the condition of ionic strength ¼10 mM NaCl, pH ¼6.8 and temperature¼20 1C was used to calculate the interaction energy to determine the fouling tendency. Surface tensions of organic foulants and MF membranes were determined from contact angles of DI water, formamide and diiodomethane by the sessile drop on filtered lawns of organic foulants or on a clean MF membrane. Contact angle measurements were performed using an NRL Contact Angle Goniometer (Rame Hart, Mountain Lakes, NJ), which is a standard goniometer with image analysis attachments (i.e. video camera, computer with a monitor, and image analysis software). At least 12 contact angles were determined for each test solutions with the highest and lowest values discarded, and the average value was taken as the contact angle.
ð5Þ #
h0 h
2.6. Analytical items
Table 1 Calculated surface energetic parameters for SA with different ion strengths, divalent cations and BSA proportions. SA
γLW
γþ
γ
ΔGLW sls
ΔGAB sls
ΔGEL sls
ΔGcoh
10 mM NaCl 30 mM NaCl 50 mM NaCl 100 mM NaCl 0.05 mM Ca2 þ 0.1 mM Ca2 þ 0.3 mM Ca2 þ 0.5 mM Ca2 þ 1.0 mM Ca2 þ 0.5 mM Mg2 þ SA:BSA ¼ 3:7a SA:BSA ¼ 7:3a
38.23 39.51 40.41 41.94 39.71 40.65 41.42 42.98 43.93 39.05 36.63 39.26
1.68 1.25 1.00 0.59 1.41 1.14 0.88 0.42 0.27 1.39 0.96 1.47
6.53 4.97 3.95 3.65 4.02 3.54 3.09 2.4 1.47 6.08 22.61 5.45
0.3122 0.4956 0.6474 0.9472 0.5277 0.6911 0.8396 1.1796 1.4114 0.4252 0.1398 0.4567
37.4515 44.3561 49.6061 53.7646 46.9371 50.4647 54.141 61.6333 69.5343 40.0081 4.7986 41.6768
0.1064 0.1175 0.1231 0.1181 0.0792 0.0653 0.0543 0.043 0.0316 0.046 0.0376 0.0403
37.6573 44.7342 50.1304 54.5937 47.3856 51.0905 54.9263 62.7699 70.914 40.3872 4.9007 42.0932
pH¼ 6.8, temperature¼ 20 1C. a
Ionic strength¼ 0.5 mM CaCl2 þ8.5 mM NaCl.
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
1031
Table 2 Calculated surface energetic parameters for SA with different ion strengths, divalent cations and BSA proportions. SA
ΔGLW slm
ΔGAB slm
ΔGEL slm
ΔGadh
10 mM NaCl 30 mM NaCl 50 mM NaCl 100 mM NaCl 0.05 mM Ca2 þ 0.1 mM Ca2 þ 0.3 mM Ca2 þ 0.5 mM Ca2 þ 1.0 mM Ca2 þ 0.5 mM Mg2 þ SA:BSA ¼ 3:7a SA:BSA ¼ 7:3a
0.7508 0.9458 1.0811 1.3076 0.976 1.1169 1.2311 1.4592 1.5962 0.8761 0.5023 0.908
50.1626 54.4964 57.6114 59.9847 56.0536 58.1266 60.2117 64.2372 68.3187 51.839 31.7217 52.8356
0.0763 0.1056 0.1228 0.143 0.0659 0.0598 0.0545 0.0486 0.0416 0.0502 0.0454 0.047
50.837 55.3367 58.5697 61.1493 56.9637 59.1837 61.3883 65.6479 69.8732 52.6649 32.1786 53.6966
pH¼ 6.8, temperature¼ 20 1C. a
Ionic strength¼ 0.5 mM CaCl2 þ8.5 mM NaCl.
(Subramani et al., 2009). Positive values for ΔGcoh represent the degree of hydrophilicity and negative values represent the degree of hydrophobicity (van Oss, 2007). The calculated surface tension parameters and cohesion free energies for SA with different ion strengths, divalent cations and organic foulant compositions are listed in Table 1. It could be seen that the cohesion energy of SA became larger with increasing ionic strength and divalent ion addition. Hence, the hydrophobicity of SA was enhanced with the increasing ionic strength. Compared with that in the absence of divalent cations (10 mM NaCl), the cohesion energies of SA in the presence of 0.5 mM Ca2 þ and 0.5 mM Mg2 þ were increased by 66.7% and 7.2%, respectively. Therefore, the hydrophobicity of SA in the presence of divalent cations was higher compared to that in the absence of divalent cations (10 mM NaCl). Moreover, the SA in the presence of 0.5 mM Ca2 þ showed stronger hydrophobicity than that in the presence of 0.5 mM Mg2 þ . Meanwhile, the cohesion energy of the SA became weaker with the decreasing mass ratio of SA to BSA, indicating that hydrophobicity of SA was reduced. The hydrophilic/hydrophobic interaction between the organics and membranes may be of much importance since hydrophobic substances demonstrated stronger interaction with hydrophobic membrane, which was supported by previous researches (Maximous et al., 2009; Van der Bruggen et al., 2004). The interfacial free energy (per unit area) between organics and clean MF membrane at contact is accurately described by the free energy of adhesion (ΔGadh), as presented in Table 2. The SA with different ion strengths and divalent cations expressed significantly different characters. The adhesion energy between SA and MF membrane became more substantial with increasing ionic strength. Compared with the absence of divalent cations (10 mM NaCl), the adhesion energies between SA and MF membrane were increased by 29.1% and 3.6% in the presence of 0.5 mM Ca2 þ and 0.5 mM Mg2 þ , respectively. With varying proportions of SA to BSA, the adhesion energy between SA and MF membrane was reduced with decreasing SA concentration. 3.2. Influence of ionic strength on SA fouling and interaction energy between foulant and MF membrane surfaces The influence of ionic strength on the flux decline behavior during SA fouling of MF membrane and the corresponding interaction energy are presented in Fig. 1. As shown in Fig. 1a, the membrane permeate flux of SA at 10 mM NaCl declined by approximately 72.7% when 400 ml of permeate was collected, while SA at 30 and 50 mM NaCl resulted in ca. 75.7% and 76.9% decrement, respectively. With respect to the SA at 100 mM NaCl,
Fig. 1. Effect of ionic strength on (a) flux decline of MF membrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L and pH¼ 6.8). The ionic strength of the test solutions was adjusted by varying indifferent salt (i.e., NaCl) concentration.
the flux decline was ca. 77.9% over a period of cumulative volume 400 ml. It was clearly shown that flux decline of MF membrane became more significant with increasing ionic strength.
1032
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
SA fouling process might be controlled by the interactions between SA and clean MF membrane and among SA molecules. The XDLVO theory can provide a quantitative assessment of these interactions. The interaction energy profiles between SA molecules and clean MF membrane are shown in Fig. 1b. It could be observed that the interaction energy profile between SA molecules and clean MF membrane at 10 mM exhibited an energy barrier and a secondary energy minimum. The energy barrier means the foulants in suspension should have sufficient kinetic energy to overcome this barrier to approach the MF membrane (Redman et al., 2004). The secondary energy minimum represents the ability of foulants being sucked onto MF membrane surface (Hoek et al., 2003). The interaction energy profiles between SA molecules and clean MF membrane at 30, 50 and 100 mM showed entirely attractive interaction energy. Meanwhile, it can be seen that the attractive energy became much greater with increasing ionic strength. The interaction energies between SA molecules and SA molecules were also calculated. As seen in Fig. 1c, the interaction energy profile among SA molecules at 10 mM exhibited an energy barrier and a secondary energy minimum, and the interaction energies among SA molecules at 30, 50 and 100 mM were entirely attractive. The attractive energy between SA molecules and SA molecules became larger with increasing ionic strength. It could be inferred that with the increasing ionic strength, the compression of electric double layer due to charge screening reducing the electrostatic repulsion among SA molecules and between SA molecules and MF membrane surface (Fig. S1). This enabled van der Waals attraction to completely overwhelm electrostatic repulsion. Therefore, the ability of SA being sucked onto the MF membrane surface was enhanced with increasing ionic strength. The changes in interaction energies among SA molecules and between SA molecules and clean MF membrane could be related to the flux decline trend of fouling runs. It was possible that the MF membrane surfaces were not completely coated with SA and some of the clean MF membrane remained exposed after fouling. Thus, the SA fouling process was controlled by the combined interaction energies among SA molecules and between SA molecules and MF membrane. 3.3. Influence of divalent cations on SA fouling and interaction energy between foulant and MF membrane surfaces Recent studies have demonstrated that divalent cations have a dramatic effect on SA fouling (Braghetta et al., 1998; Hong and Elimelech, 1997; Yuan and Zydney, 2000). To investigate the influence of divalent ions on alginate fouling of MF membrane, fouling experiments were performed in the presence of CaCl2 and MgCl2, because they are the major divalent cations in wastewaters. In these experiments, the total ionic strength was kept constant at 10 mM by adjusting the background NaCl concentration. The permeate flux behavior during fouling of MF membrane and the corresponding interaction energy curves are presented in Fig. 2. As shown in Fig. 2a, the flux decline in the presence of Mg2 þ was much rapider compared to that in the absence of divalent cations. However, the fouling extent of SA with 0.5 mM Ca2 þ was more severe than the one observed with 0.5 mM Mg2 þ , even though both were divalent ions. The interaction energies between SA with different divalent cations and clean MF membrane surface were calculated (Fig. 2b). It was seen that all the interaction energy profiles of SA approaching the clean MF membrane exhibited an energy barrier and a secondary energy minimum. Compared to the absence of divalent cations (10 mM NaCl), the secondary energy minimum between SA and clean MF membrane surface was higher in the presence of Mg2 þ , but the energy barrier was lower. It could also been seen that the energy barrier of SA in the presence of Ca2 þ was much less than that determined in the presence of Mg2 þ , however, the secondary energy
Fig. 2. Effect of divalent cations on (a) flux decline of MF membrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L, total ionic strength¼ 10 mM, and pH¼6.8). The total ionic strength of the test solution is fixed at 10 mM by varying NaCl concentration.
minimum was stronger. The interaction energies among SA molecules with different divalent cations are plotted in Fig. 2c. It was noted that the cations had a similar effect on SA–MF membrane and SA–SA interactions, with the magnitude of the attractive energies
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
following the order of Ca2 þ 4Mg2 þ 4Na þ , and the repulsive energies following the order of Ca2 þ oMg2 þ oNa þ . This result was in agreement with the fouling rate of MF membrane observed in bench-scale fouling experiments, with the flux decline rate of MF membrane following the order of Ca2 þ 4Mg2 þ 4Na þ (Fig. 2a). According to previous study (Hong and Elimelech, 1997; Li and Elimelech, 2004), Na þ , Ca2 þ and Mg2 þ ions all played a role in charge screening (or double layer compression). Essentially, the effect of charge screening on reducing electrostatic repulsion among SA molecules is expected to be similar in all the three electrolyte solutions because the total ionic strength was kept constant. This difference in fouling profile of MF membrane in the presence and absence of divalent cations was probably attributed to other interaction apart from charge screening, which was most likely charge neutralization due to the divalent cations binding with SA molecules. Divalent cations of Ca2 þ and Mg2 þ were reported to form complex with carboxylic (–COO–) moieties of polysaccharide, which neutralized the negative charges carried by these macromolecules or bound them together via divalent ion bridging (Ang and Elimelech, 2007; Lee et al., 2005; Tang et al., 2009). Thus, the electrostatic repulsive energy in the presence of divalent cations was weakened, and the attractive energy was enhanced (Fig. S2a). Hence, the fouling rate of SA with divalent cations was higher compared to that without divalent cations. Calcium ions are known to be more favorable divalent cations to form complexes with SA molecules than magnesium ions, and thus, should be more effective in charge neutralization (Davis et al., 2003). Thus, the electrostatic repulsive energy in the presence of Ca2 þ was less than that in the presence of Mg2 þ (Fig. S2b). Meanwhile, ionic bridge between SA molecules in the presence of calcium resulted in the formation of a cross-linked organic gel layer on the membrane surface, which produced significant hydraulic resistance to permeate flow and, thus, a severe flux decline (Mo et al., 2008; Yang et al., 2010). As a result, the fouling layer on MF membrane formed in the presence of Ca2 þ was very compact and highly resistant to hydrodynamic forces. Based on above results that Ca2 þ had important impact on SA fouling of MF membrane, the influence of calcium ion concentration on SA fouling of MF membrane and the corresponding interaction energy were further investigated. Calcium concentration varied from 0.05 to 1.0 mM, keeping the total ionic strength constant at 10 mM. The results in Fig. 3a clearly show that the flux decline of MF membrane became greater with increasing Ca2 þ concentration. As seen in Fig. 3b, it was noted that the attractive interaction energy between SA and clean MF membrane increased significantly with increasing Ca2 þ concentration, but the repulsive interaction energy decreased correspondingly. The interaction energies between SA in bulk solution and SA on MF membrane surface (i.e., in the fouling layer) with different calcium ion concentrations were also calculated. As seen from Fig. 3c, the repulsive energy barrier between SA and SA molecules decreased significantly and the attractive energy increased remarkably with the increase of calcium ionic concentration. With increasing Ca2 þ concentration, the charge screening, charge neutralization and ionic bridge formed by Ca2 þ reduced electrostatic repulsion and increased attractive energies between SA molecules and MF membrane surface and among SA molecules (Fig. S3a and S3b). Thus, the stoichiometry between Ca2 þ and SA concentrations was also expected to play an important role in the interaction energies between SA molecules and MF membrane surface and among SA molecules. Obviously, the flux decline trends of MF membrane with different divalent cations and calcium ion concentrations can be related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane.
1033
Fig. 3. Effect of calcium concentration on (a) flux decline of MF mebrane during fouling, (b) interaction energies profiles between SA and the clean MF membrane, and (c) interaction energies profiles between SA molecules (SA¼ 20 mg/L, total ionic strength¼ 10 mM, pH¼6.8).
3.4. Influence of organic foulant composition on SA fouling and interaction energy between foulant and MF membrane surfaces Fouling experiments were performed with different mass ratios of SA to BSA to investigate the influence of organic foulant
1034
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
(iii) 14 mg/L SA plus 6 mg/L BSA (SA to BSA mass ratio of 7:3), and (iv) 20 mg/L SA and no BSA. The permeate flux decline of MF membrane and the corresponding interaction energy for these foulant combinations are presented in Fig. 4. It was expected that the flux decline profiles of MF membrane would vary correspondingly to the proportions of SA to BSA. It can be seen from Fig. 4a that feed solution in the presence of 20 mg/L SA resulted in the fastest flux decline of MF membrane. The SA with a small amount of BSA (14 mg/L SA and 6 mg/L BSA) caused a slower flux decline of MF membrane compared with 20 mg/L SA. The flux decline of the feed solution containing more BSA (6 mg/L SA and 14 mg/L BSA) was much less than SA with a small amount of BSA (14 mg/L SA and 6 mg/L BSA). Accordingly, the permeate flux decline of MF membrane for SA fouling was more substantial compared to BSA fouling. When the test solutions contained both SA and BSA, the flux decline was more severe for higher SA concentration. The interaction energies between SA and clean MF membrane surface were calculated with different mass ratios of SA to BSA, which are presented in Fig. 4b. It was noted that the attractive interaction energy between SA and clean MF membrane decreased with decreasing SA concentration, but the repulsive interaction energy increased correspondingly. The interaction energies among SA molecules with varying proportions of SA to BSA are plotted in Fig. 4c. With the decrease of proportions of SA to BSA, the maximum attractive energy (in magnitude and range) was also weakened, and the maximum repulsive energy was significantly enhanced. This was because the BSA molecules interact with Ca2 þ , unlike SA molecules, were hard to form a gel-type fouling layer in the presence of Ca2 þ (Ang and Elimelech, 2007). We attributed the much greater attractive energy and lower repulsive energy with SA than BSA to the gel forming nature of SA, where intermolecular cohesion needed to be strong enough to sustain the structural integrity of the gel network. It could be confirmed from the decreased attractive AB energy and LW energy with the decreasing proportions of SA to BSA (Fig. S4). The results indicated that the flux decline trends of MF membrane were related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane with varying proportions of SA to BSA.
4. Conclusions
Fig. 4. Effect of organic foulant composition on (a) flux decline of MF mebrane during fouling, (b) interaction energies profiles between foulant and the clean MF membrane, and (c) interaction energies profiles between foulant molecules (total ionic strength¼10 mM, pH¼6.8). The total organic foulant concentration of the test solution is fixed at 20 mg/L by varying the ratio of SA to BSA concentration.
compositions on MF membrane fouling. Four foulant compositions were examined with the total foulant concentration being maintained at 20 mg/L: (i) 20 mg/L BSA and no sodium alginate, (ii) 6 mg/L SA plus 14 mg/L BSA (SA to BSA mass ratio of 3:7),
The effects of ionic strengths, divalent cations, and organic foulant compositions on MF membrane fouling by sodium alginate (SA) were investigated. Measured physicochemical properties revealed that the hydrophobicity of SA was increased due to the increase of ionic strength, addition of divalent cations and the higher mass ratio of SA to BSA. Fouling experiments had shown that the MF membrane fouling by SA was higher when the ionic strength was increased or calcium ion was present in the feed solution. In the presence of magnesium, MF membrane fouling by SA was not as severe as the one observed with calcium ions. Since the MF membrane surfaces were not completely coated with SA and some of the clean MF membrane remained exposed after fouling, the flux decline trends of SA fouling with increasing ionic strength and different divalent cations addition could be related to the changes of interaction energies among SA molecules and between SA molecules and MF membrane. When the test solutions contained both SA and BSA, the flux decline of MF membrane was more substantial for higher SA concentration. This could also be predicted by the greater attractive energy and lower repulsive energy among SA molecules and between SA molecules and MF membrane with higher SA concentration.
Y. Ding et al. / Chemical Engineering Science 104 (2013) 1028–1035
Acknowledgments This study was supported by the Major Science and Technology Program for Water Pollution Control and Management (No. 2013ZX07201007), The State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (No. 2011DX01) and The Specialized Research Fund for the Doctoral Program of Higher Education (No. 20112302110060). The authors also appreciate the Funds for Creative Research Groups of China (No. 51121062).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ces.2013.10.036. References Ang, W.S., Elimelech, M., 2007. Protein (BSA) fouling of reverse osmosis membranes: implications for wastewater reclamation. J. Memb. Sci. 296, 83–92. Arabi, S., Nakhla, G., 2010. Impact of molecular weight distribution of soluble microbial products on fouling in membrane bioreactors. Sep. Purif. Technol. 73, 391–396. Bouhabila, E.H., Ben Aïm, R., Buisson, H., 2001. Fouling characterisation in membrane bioreactors. Sep. Purif. Technol. 22, 123–132. Braghetta, A., DiGiano, F.A., Ball, W.P., 1998. NOM accumulation at NF membrane surface: impact of chemistry and shear. J. Environ. Eng. 124, 1087–1098. Brant, J.A., Childress, A.E., 2002. Assessing short-range membrane–colloid interactions using surface energetics. J. Memb. Sci. 203, 257–273. Childress, A.E., Elimelech, M., 1996. Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. J. Memb. Sci. 119, 253–268. Davis, T.A., Volesky, B., Mucci, A., 2003. A review of the biochemistry of heavy metal biosorption by brown algae. Water Res. 37, 4311–4330. de Kerchove, A.J., Elimelech, M., 2007. Impact of alginate conditioning film on deposition kinetics of motile and nonmotile Pseudomonas aeruginosa strains. Appl. Environ. Microbiol.73, 5227–5234. Drews, A., 2010. Membrane fouling in membrane bioreactors—characterisation, contradictions, cause and cures. J. Memb. Sci. 363, 1–28. Güell, C., Davis, R.H., 1996. Membrane fouling during microfiltration of protein mixtures. J. Memb. Sci. 119, 269–284. Hoek, E.M., Bhattacharjee, S., Elimelech, M., 2003. Effect of membrane surface roughness on colloid–membrane DLVO interactions. Langmuir 19, 4836–4847. Hong, S., Elimelech, M., 1997. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. J. Memb. Sci. 132, 159–181. Kang, S.T., Subramani, A., Hoek, E., Deshusses, M.A., Matsumoto, M.R., 2004. Direct observation of biofouling in cross-flow microfiltration: mechanisms of deposition and release. J. Memb. Sci. 244, 151–165. Kim, S., Hoek, E., 2007. Interactions controlling biopolymer fouling of reverse osmosis membranes. Desalination 202, 333–342.
1035
Lee, S., Cho, J., Elimelech, M., 2005. A novel method for investigating the influence of feed water recovery on colloidal and NOM fouling of RO and NF membranes. Environ. Eng. Sci. 22, 496–509. Lee, S., Elimelech, M., 2006. Relating organic fouling of reverse osmosis membranes to intermolecular adhesion forces. Environ. Sci. Technol. 40, 980–987. Li, Q., Elimelech, M., 2004. Organic fouling and chemical cleaning of nanofiltration membranes: measurements and mechanisms. Environ. Sci. Technol. 38, 4683–4693. Liang, S., Liu, C., Song, L., 2007. Soluble microbial products in membrane bioreactor operation: behaviors, characteristics, and fouling potential. Water Res. 41, 95–101. Liu, F., Hashim, N.A., Liu, Y., Abed, M.R.M., Li, K., 2011. Progress in the production and modification of PVDF membranes. J. Memb. Sci. 375, 1–27. Maximous, N., Nakhla, G., Wan, W., 2009. Comparative assessment of hydrophobic and hydrophilic membrane fouling in wastewater applications. J. Memb. Sci. 339, 93–99. Mo, H., Tay, K.G., Ng, H.Y., 2008. Fouling of reverse osmosis membrane by protein (BSA): effects of pH, calcium, magnesium, ionic strength and temperature. J. Memb. Sci. 315, 28–35. Redman, J.A., Walker, S.L., Elimelech, M., 2004. Bacterial adhesion and transport in porous media: role of the secondary energy minimum. Environ. Sci. Technol. 38, 1777–1785. Rosenberger, S., Laabs, C., Lesjean, B., Gnirss, R., Amy, G., Jekel, M., Schrotter, J.-C., 2006. Impact of colloidal and soluble organic material on membrane performance in membrane bioreactors for municipal wastewater treatment. Water Res. 40, 710–720. Shim, Y., Lee, H.-J., Lee, S., Moon, S.-H., Cho, J., 2002. Effects of natural organic matter and ionic species on membrane surface charge. Environ. Sci. Technol. 36, 3864–3871. Subramani, A., Hoek, E., 2008. Direct observation of initial microbial deposition onto reverse osmosis and nanofiltration membranes. J. Memb. Sci. 319, 111–125. Subramani, A., Huang, X., Hoek, E., 2009. Direct observation of bacterial deposition onto clean and organic-fouled polyamide membranes. J. Colloid Interface Sci. 336, 13–20. Tang, C.Y., Kwon, Y.-N., Leckie, J.O., 2009. 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. Memb. Sci. 326, 526–532. Tian, Y., Chen, L., Jiang, T., 2011. Characterization and modeling of the soluble microbial products in membrane bioreactor. Sep. Purif. Technol. 76, 316–324. Van der Bruggen, B., Kim, J.H., DiGiano, F.A., Geens, J., Vandecasteele, C., 2004. Influence of MF pretreatment on NF performance for aqueous solutions containing particles and an organic foulant. Sep. Purif. Technol. 36, 203–213. Van Oss, C.J., 2006. Interfacial Forces in Aqueous Media, 2nd ed. Taylor & Francis. Boca Raton. van Oss, C.J., 2007. Development and applications of the interfacial tension between water and organic or biological surfaces. Colloids Surf. B: Biointerfaces 54, 2–9. Wang, F., Tarabara, V.V., 2008. Pore blocking mechanisms during early stages of membrane fouling by colloids. J. Colloid Interface Sci. 328, 464–469. Wang, S., Guillen, G., Hoek, E.M.V., 2005. Direct observation of microbial adhesion to membranes. Environ. Sci. Technol. 39, 6461–6469. Yang, Q., Liu, Y., Li, Y., 2010. Control of protein (BSA) fouling in RO system by antiscalants. J. Memb. Sci. 364, 372–379. Yuan, W., Zydney, A.L., 2000. Humic acid fouling during ultrafiltration. Environ. Sci. Technol. 34, 5043–5050.