Water Research 93 (2016) 121e132
Contents lists available at ScienceDirect
Water Research journal homepage: www.elsevier.com/locate/watres
Natural organic matter fouling behaviors on superwetting nanofiltration membranes Linglong Shan a, Hongwei Fan a, Hongxia Guo b, **, Shulan Ji a, Guojun Zhang a, * a Beijing Key Laboratory for Green Catalysis and Separation, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing 100124, PR China b College of Material Science and Engineering, Beijing University of Technology, Beijing 100124, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 November 2015 Received in revised form 19 January 2016 Accepted 24 January 2016 Available online xxx
Nanofiltration has been widely recognized as a promising technology for the removal of micro-molecular organic components from natural water. Natural organic matter (NOM), a very important precursor of disinfection by-products, is currently considered as the major cause of membrane fouling. It is necessary to develop a membrane with both high NOM rejection and anti-NOM fouling properties. In this study, both superhydrophilic and superhydrophobic nanofiltration membranes for NOM removal have been fabricated. The fouling behavior of NOM on superwetting nanofiltration membranes has been extensively investigated by using humic acid (HA) as the model foulant. The extended DerjaguinLandauVerweyOverbeek approach and nanoindentor scratch tests suggested that the superhydrophilic membrane had the strongest repulsion force to HA due to the highest positive total interaction energy (DGTOT) value and the lowest critical load. Excitation emission matrix analyses of natural water also indicated that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal thereof. Conversely, the superhydrophobic membrane showed resistance to fouling by hydrophilic substances and therefore high removal capacity. Long-term operation suggested that the superhydrophilic membrane had high stability due to its anti-NOM fouling capacity. Based on the different anti-fouling properties of the studied superwetting membranes, a combination of superhydrophilic and superhydrophobic membranes was examined to further improve the removal of both hydrophobic and hydrophilic pollutants. With a combination of superhydrophilic and superhydrophobic membranes, the NOM rejection (RUV254) and DOC removal rates (RDOC) could be increased to 83.6% and 73.3%, respectively. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Superhydrophilic Superhydrophobic Nanofiltration NOM Fouling
1. Introduction In natural water, natural organic matter (NOM) can affect the color, taste and odor of raw water, moreover, it has been identified as precursors of disinfection by-products. NOM can be divided into three categories: humic acid (HA), fulvic acid (FA) and humin, with molecular weights ranging from 500 kDa (Jaouadi et al., 2012; Lee et al., 2004; Ng et al., 2013). Nanofiltration has been widely recognized as a promising technology to remove NOM from natural water due to its high efficiency for removal of pollutants, low energy consumption, and easy accessibility. With the
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (H. Guo), [email protected] (G. Zhang). http://dx.doi.org/10.1016/j.watres.2016.01.054 0043-1354/© 2016 Elsevier Ltd. All rights reserved.
improvement of drinking water standards and the industrialized application of membrane technology, it is critical to develop novel nanofiltration membranes with high NOM removal and stability. Concurrently, the biggest challenge is the inevitable occurrence of membrane fouling, which is caused by the NOM accumulation on the membrane surface during the NOM removal process in drinking water treatment and production (Al-Amoudi, 2010; Cai et al., 2008; Cai and Benjamin, 2011; Howe and Clark, 2002; Huang et al., 2012; Jermann et al., 2007, 2008; Lahoussine-Turcaud et al., 1990; Saravia et al., 2006; Van der Bruggen et al., 2003; Vrouwenvelder et al., 1998; Yang et al., 2010; Zhang et al., 2003b; Zhou et al., 2009). In particular, HA has been considered as the most important foulant for the membrane flux decline. Membrane fouling is an issue of interface action between the foulant, membrane, and solution. NOM fouling is mainly determined by the interface relationship between NOM and the
122
L. Shan et al. / Water Research 93 (2016) 121e132
membrane surface in aqueous solution. Many studies have been carried out to investigate this point; however, the effect of hydrophilicity/-phobicity of NOM and the membrane surface on fouling behavior is still a controversial issue. Some researchers have studied the effect of NOM hydrophilicity/-phobicity on membrane fouling, since NOM can be fractionated into relatively hydrophilic and hydrophobic components (Hao et al., 2011; Kilduff et al., 2004; €ntta €ri et al., 2000; Thorsen, 2004). Amy and Nilson found that Ma the hydrophobic fraction was the major factor for flux decline, while the hydrophilic fraction was much less responsible for fouling of the membrane (Amy and Cho, 1999; Nilson and DiGiano, 1996). In contrast, Cho et al. (2000), Fan et al. (2001) and Yamamura et al. (2014) obtained completely opposite results, and they found that the hydrophilic fractions induced more severe loss of membrane permeability than the hydrophobic fractions. Although some previous studies indicated that hydrophilicity/-phobicity of the membrane surface was a predominant influence on the extent of fouling in aqueous systems (Maximous et al., 2009; Xiao et al., 2011), contradictory results were also often be obtained. Zhang et al. (2003a) reported that with the increase of surface hydrophobicity, a smaller amount of HA was deposited onto the membrane surface and the permeate flux reduction decreased. Conversely, many reports have indicated that a hydrophilic surface favors less fouling, due to a strong hydration layer that repels the €ntt€ adsorption of foulants on the membrane surface (Ma ari et al., 2000; Tiraferri et al., 2012). Thus, determining the effect of hydrophilicity/-phobicity of NOM and membrane surface wettability on its fouling behavior is still a challenge. To date, most studies in this field have concerned in the wettability of membranes with water contact angles in the range 30120 , covering the range from hydrophilic to hydrophobic surfaces. It is notable that some extraordinary membrane performances have been achieved when the wettability was extended to superhydrophilicity or superhydrophobicity. It is generally accepted that the solid surfaces with contact angles 150 are defined as superhydrophobic. Tiraferri et al. (2012) and Liang et al. (2014) investigated the fouling behavior of superhydrophilic membranes with model wastewater and found that these membranes showed lower overall flux decline and anti-fouling properties. Our previous study indicated that the pervaporation flux of a superhydrophilic membrane was twice than that of a hydrophilic membrane (Gong et al., 2014). We also demonstrated that the superhydrophobic pervaporation membrane showed both higher selectivity and higher permeability than a hydrophobic membrane (Li et al., 2014). However, little work has dealt with superwetting nanofiltration membranes for NOM removal from natural water. Additionally, there has been no systematic work on NOM fouling behavior on superwetting nanofiltration membranes. The aim of this study is to develop the superwetting nanofiltration membranes with both high NOM rejection and good antiNOM fouling properties. We extensively studied the NOM fouling behavior on superwetting nanofiltration membranes. The interaction forces and free energy of HA adhesion on superwetting membrane surfaces were elucidated by the extended DerjaguinLandauVerweyOverbeek (XDLVO) thermodynamic analyses. Then we carried out the nanoindentor scratch tests to quantify the adhesion between the HA fouling layer and membrane surface. In addition, the NOM fouling of hydrophilic and hydrophobic components on superhydrophilic and superhydrophobic nanofiltration membranes were analyzed by excitation emission matrix (EEM). The natural water treatment performances of the superwetting membranes and their long-term stability were investigated using real reservoir water.
2. Theoretical The XDLVO model has widely been used to explain how aqueous foulants interact with surfaces of polymeric membranes (Brant and Childress, 2002; Kang et al., 2004; Kim et al., 2006). According to XDLVO theory, in an aquatic environmental system, the interfacial energy between a membrane and foulants is the sum of the Lifshitzevan der Waals, Lewis acidbase, and electrostatic doublelayer interactions, which is given by (Brant and Childress, 2002): LW AB EL GTOT mwf ¼ Gmwf þ Gmwf þ Gmwf
(1)
where GTOT mwf is the total interaction energy between the membrane AB EL and foulants and GLW mwf , Gmwf and Gmwf represent the Lifshitzevan der Waals, Lewis acidbase, and electrostatic double-layer free energies, respectively. The subscripts m, w, and f correspond to the membrane, water, and the foulant, respectively. 2.1. Surface tension components The surface tension components of the membrane and foulants are determined from the extended Young equation (Bouchard et al., 1997; Gourley et al., 1994; Van Oss and Good, 1988), which gives their relationship with the contact angle of a liquid on a solid surface. The surface tension parameters of both the solid and liquid can be written as follows:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi LW þ gþ ð1 þ cosqÞgTOT ¼ 2 gLW g gþ s gl þ s gl l s
(2)
pffiffiffiffiffiffipffiffiffiffiffiffi gþ g
(3)
gTOT ¼ gLW þ gAB
(4)
gAB ¼ 2
where q is the contact angle, gTOT is the total surface tension, gLW is the Lifshitzevan der Waals component, and gþ and g are the electron-acceptor and electron-donor components, respectively. The subscripts s and l correspond to the solid surface and liquid, respectively. The surface tension parameters of foulants (gþ , g , f f þ , g , gLW ) can be determined from gLW ) and membranes (g m m m f equation (2) after measuring the contact angle data for three probe liquids with known surface tension parameters (gþ , g , gLW ). l l l 2.2. Adhesion free energy The free energy of adhesion per unit area was calculated as follows:
△GLW mwf ¼ 2
qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiqffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi gLW gLW gLW gLW w m w f
(5)
qffiffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi þ g Þ þ 2 g gþ gþ g ð g gþ w mþ m w w f f qffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ g gþ gþ w 2 m gf þ m gf
△GAB mwf ¼ 2
(6) △GEL mwf
# " ε0 εr k 2 2xm xf 2 xm þ xf 1 cothðkho Þ þ 2 ¼ cschðkho Þ 2 xm þ x2f (7)
AB EL 2 where △GLW mwf , △Gmwf and △Gmwf , expressed in mJ$m , are the
L. Shan et al. / Water Research 93 (2016) 121e132
Lifshitzevan der Waals, Lewis acidbase, and electrostatic doublelayer free energies at a separation distance of h0 (nm), respectively; h0 is the minimum equilibrium cut-off distance, usually assigned a value of 0.158 nm (±0.009 nm) (Brant and Childress, 2002; Chen et al., 2012; Kang et al., 2004); ε0εr is the dielectric permittivity of the suspending fluid, usually assigned as 6.95 1010 C2 J1 m1 (Brant and Childress, 2002; Kim et al., 2006); k is the inverse Debye screening length, assigned as 0.104 nm1 (Brant and Childress, 2002; Kim et al., 2006); and xm and xf, expressed in mV, are the surface potentials of the membrane and foulant, respectively. AB △GLW mwf and △Gmwf could be determined from the surface tension LW parameters of foulants (gþ , g , gLW ), membranes (gþ m , gm , gm ) f f f , gLW ) obtained in 2.1. and water (gþ , g w w w
2.3. Interaction energy between a spherical foulant and an infinite planar surface The interaction between a membrane and a foulant in an aqueous environment can be expressed as follows:
ULW mwf
¼
2p△GLW mwf
ah20 h
! (8)
h0 h AB ¼ 2pal△G exp UAB mwf mwf l " UEL mwf ¼ pε0 εr a 2xf xm In
1 þ ekh 1 ekh
!
(9) þ x2f þ x2m In 1 e2kh
#
(10) AB EL where ULW mwf , Umwf and Umwf are the Lifshitzevan der Waals, Lewis acid-base, and electrostatic double-layer interaction energies. a is the radius of the foulant (nm); h is the separation distance between the membrane and foulant (nm); and l is the characteristic decay length of the AB interaction in water, commonly measured as 0.6 nm for aqueous systems (Brant and Childress, 2002; Kim et al., 2006).
3. Experimental 3.1. Chemicals and materials A polyacrylonitrile (PAN) ultrafiltration membrane with a molecular weight cut-off of 100 kDa (PAN-50) and a polysulfone (PS) ultrafiltration membrane with a cut-off of 20,000 were purchased from Sepro Membranes (Oceanside, CA, USA). Unless otherwise specified, all reagents and chemicals were of analytical grade. HA, poly(ethyleneimine) (PEI) (Mw ¼ 750,000), and poly(sodium 4styrenesulfonate) (PSS) (Mw ¼ 700,000) were obtained from Aldrich (St. Louis, MO, USA). Hydrochloric acid (HCl), glutaraldehyde, potassium chloride (KCl), sodium hydroxide (NaOH), tetraethyl orthosilicate, n-heptane, dibutyltin dilaurate, ethanol, and glycerol were provided by Beijing Chemical Factory (Beijing, China). Dihydroxypolydimethylsiloxane (PDMS) with viscosity of 2500 Pa S was purchased from China Bulestar Chengrand Chemical Co., Ltd. (Chengdu, China). Silica nanoparticles (7e40 nm particle size), titanium oxide nanoparticles (5e10 nm particle size), and diiodomethane were purchased from Shanghai Crystal Pure Biological Technology Co., Ltd. (Shanghai, China). Natural water was collected from Miyun reservoir in Beijing on March 13th, 2015.
123
3.2. Preparation of the membranes Superhydrophilic and superhydrophobic membranes were prepared, along with hydrophilic and hydrophobic membranes for comparison. The superhydrophilic membrane was prepared as illustrated in Scheme 1a, a PAN substrate was hydrolyzed with 2.0 mol/L NaOH, rinsed with deionized water, then immersed in 1.0 mg/mL PEI polycation solution for 20 min. It was then rinsed three times with deionized water, and the pH of the PEI solution was adjusted to 1. The substrate was then immersed in PSS-TiO2 mixed polyanion solution for 20 min and rinsed once more with deionized water; the pH of the PSS-TiO2 solution was 7. The concentration of TiO2 in the 1.0 mg/mL PSS polyanion solution was 0.25 mg/mL. This procedure was repeated six times and the membrane was then crosslinked by immersing in 0.02 mg/mL aqueous glutaraldehyde solution at 30 C for 48 h. Finally, the membrane was irradiated for 40 min using a UV lamp to form the superhydrophilic surface. The superhydrophobic membrane was prepared as illustrated in Scheme 1b, a PS substrate was immersed in a 30 wt% ethanol solution for 2 h at room temperature. The residue on the surface of the membrane was then rinsed off using deionized water. A vacuum pump was used to create a negative pressure to extract air from the membrane pores over a period of 4 h. In addition, 1.0 wt% of PDMS was dissolved in n-heptane and the solution was stirred for 1 h. Specified amounts of silica nanoparticles were then dispersed in n-heptane by stirring for 1 h (PDMS/silica mass ratio 1:0.5). Finally, 0.1 wt% tetraethyl orthosilicate and 0.05 wt% dibutyltin dilaurate were also dissolved in n-heptane and the resulting mixture was continuously stirred for 1 h. To better disperse the inorganic particles in the hybrid membrane, the surface of the PS substrate was coated with the suspension by sonication during the assemble process. For this, the PS substrate was dip-coated in the suspension for 30 s. Subsequently, the membrane was fixed onto a substrate and irradiated with a lamp to accelerate the removal of solvent from the film surfaces. The above steps were repeated twice more to obtain a multilayer membrane. The samples were then kept in a convection oven set at 80 C for 8 h to complete crosslinking of the PDMS. The hydrophilic membrane was prepared in the same way with superhydrophilic membrane, except for the addition of TiO2 nanoparticles and UV lamp irradiation. For hydrophilic membrane, the procedure was repeated four times. The hydrophobic membrane was prepared in the same way with superhydrophobic membrane, except for the addition of SiO2 nanoparticles and sonication assistance. 3.3. Fouling experiments Prior to the fouling experiments with real natural water, HA was selected as a model foulant to study the interaction energy between the membranes and NOM. To obtain the surface tensions of the membranes and HA, their contact angles (CAs) were measured with three different probe liquids of known surface tension values. The surface tension properties of each probe liquid are listed in Table S1. The membrane samples were typically dried for 48 h in an oven at 30 C and then fixed onto glass slides prior to the CA measurements. The instantaneous CA was obtained within 3 s, and the volume of the liquid drop was 5 mL. The CA of each sample was measured at least ten times. To measure HA surface tension, HA solution was deposited onto a glass slide and dried for 24 h in an oven at 30 C to form a flat solid film. CA measurements were then performed as described above. The HA-fouled membranes used for XDLVO theory evolution and scratch tests were obtained by filtering the model HA solution for 10 h to ensure the formation of a
124
L. Shan et al. / Water Research 93 (2016) 121e132
Scheme 1. (a) Schematic illustration of the preparation of the superhydrophilic membrane and its surface wettability; (b) schematic illustration of the preparation of the superhydrophobic membrane and its surface wettability.
cake layer on the membrane surface. Natural water collected from Miyun reservoir was filtered using a 0.45 mm porosity membrane and stored in dark at 4 C. Nanofiltration performance was studied with a home-made cross-flow nanofiltration system, which has been described in our previous research (Shan et al., 2015). The pressure of nanofiltration was maintained at 0.6 MPa. The membranes were cut into disks with an effective nanofiltration area of 22.9 cm2. The rejection ratio, R, was calculated using Eq. (11), where Cf and Cp are the concentrations of the solute in the feed and permeate, respectively. The permeance, J, was calculated by Eq. (12), where V is the volume of the permeate liquid passed across the membrane of area A (m2) in the time period T (h) at operative pressure P (MPa). After a stable permeance was achieved by eluting with deionized water, the membrane fouling potential was determined by calculating the relative permeance (J/J0) in terms of the ratio of permeance (J) to the initial membrane permeance (J0).
R¼
Cf Cp 100% Cf
VðLÞ J ¼ 2 A m TðhÞ PðMPaÞ
(11)
(12)
3.4. Analytical methods The zeta potential and the mean hydrodynamic diameter of HA were determined by a laser scattering size analyzer (NICOMPTM 380ZLS, USA). The zeta potentials of the membrane surfaces were
determined using a SurPASS electrokinetic analyzer (Anton Paar GmbH, Graz, Austria). During the process of measuring the zeta potential, the KCl solution concentration was maintained at 0.83 mmol/L, while the operation pressure was 0.03 MPa. The pH values of the solutions were measured with a pH meter (Leici PHS3C, Shanghai, China). The membrane surface morphologies were observed by scanning electron microscopy (SEM) observation, all membrane samples were dried under vacuum and gold-coated before observation. The images were obtained using a SU-8020 instrument (Hitachi, Japan) at an acceleration voltage of 20 keV. Wide-angle X-ray diffraction (XRD) experiments were conducted on a D8 ADVANCE X-ray diffractometer (Bruker/AXS, Germany). Water contact angle measurements were performed using a contact angle analyzer (DSA100, Germany). The interfacial adhesion between the foulant and the membrane was measured by a scratch test with a nanoindentor (G200). Scratch tests were performed using a Nano Indentor XP system with options for lateral-force measurements. The maximum scratch load of 0.45 mN was applied, with profiling load and profiling velocity of 20 mN and 30 mm/s, respectively. The scratch length and scratch velocity are 500 mm and 30 mm/s, respectively. UV254 concentration was measured using a UV/Vis spectrophotometer (T6, Shanghai, China). The concentrations of dissolved organic carbon (DOC) were determined with a total organic carbon analyzer (TOC-VCPN, Shimadzu, Japan). The relative precision of the DOC analyses was hydrophilic (112 L/ m2 h MPa) > hydrophobic (75 L/m2 h MPa) > superhydrophobic (53 L/m2 h MPa). The corresponding NOM rejections of the membranes were 79.1%, 64.2%, 61.2%, and 59.7%, respectively. The DOC removal rates of the membranes were 37.3%, 36.8%, 16.5%, and 13.3%, respectively. The superhydrophilic membrane exhibited both high initial permeance and high rejection. Permeance decline was monitored by the ratio of the permeance to the initial permeance (J/J0). The curve for the superhydrophilic membrane remained unchanged, whereas that for the hydrophilic membrane showed a continuous decline. The curves for the hydrophobic and superhydrophobic membranes exhibited an initial slow decrease, then appeared to be maintained, and finally showed a sharp decrease. Many researchers have found that rapid membrane fouling occurs at high permeation rates (Childress and Elimelech, 1996; Hong and Elimelech, 1997), whereas in our study the superhydrophilic membrane showed both high permeance and exceptionally low fouling. The morphologies, CAs, and XRD patterns of the membrane surfaces after nanofiltration were investigated to understand the changes after fouling experiments. It can be seen in Fig. S2 that the rod-like foulant was deposited on the superhydrophobic membrane surface, which might be the reason for its serious permeance decrease. After the fouling experiments, TiO2 and SiO2 nanoparticles could still be detected on the superhydrophilic and superhydrophobic membrane surfaces, respectively (Fig. S1). The water CAs of both membranes remained at initial level (Figs. 1 and S2). The results suggested that the superwetting membrane structures were stable in the nanofiltration tests, because both membranes were chemically crosslinked during their formation processes (Scheme 1). After the fouling experiments, the permeate was collected for EEM analysis, which was used to probe the chemical structure of NOM because of its ability to distinguish between different classes of organic matter (Coble et al., 1990; Henderson et al., 2009; Peiris et al., 2010). EEM fluorescence spectra were measured and
L. Shan et al. / Water Research 93 (2016) 121e132
127
Fig. 3. Variation of interaction energy profiles between the (a) clean membranes and HA; (b) fouled membranes and HA.
Fig. 4. Scratch test of the fouled membrane.
Fig. 5. (a) Variation of natural water permeance and RUV254; (b) long-term nanofiltration behavior of superhydrophilic membrane with natural water.
Table 1 The water quality indexes of the natural water and the permeate.
Natural water Superhydrophilic Hydrophilic Hydrophobic Superhydrophobic Double membranes
Turbidity (NTU)
Alkalinity (mg/L)
pH
UV254
DOC (mg/L)
RUV254 (%)
RDOC (%)
1.09 0 0 0 0 0
147.6 70.1 115.1 135.1 149.3 45.6
8.07 8.05 8.03 8.05 8.07 8.04
0.067 0.014 0.024 0.026 0.027 0.011
12.15 7.616 7.673 10.14 10.54 3.19
e 79.1 64.2 61.2 59.7 83.6
e 37.3 36.8 16.5 13.3 73.7
128
L. Shan et al. / Water Research 93 (2016) 121e132
representative examples were shown in Fig. 6. As reported in the
literature (Chen et al., 2003), horizontal and vertical lines were
Fig. 6. EEMs for (a) natural water; (b) natural water nanofiltrated by superhydrophilic membrane; (c) natural water nanofiltrated by hydrophilic membrane; (d) natural water nanofiltrated by hydrophobic membrane; (e) natural water nanofiltrated by superhydrophobic membrane; (f) natural water nanofiltrated by superhydrophilic membranesuperhydrophobic membrane.
L. Shan et al. / Water Research 93 (2016) 121e132
drawn to divide the EEM into five regions according to the nature of the fluorescent organic matter. In general, regions I and II are related to aromatic proteins, and the peak at 335/220 nm (Em/Ex) is indicative of the presence of protein-like substances. Regions III and V are related to fulvic acid-like materials and HA-like organics, respectively, and show maxima at approximately 410/225 nm (Em/ Ex) and 415/310 nm (Em/Ex) respectively (Coble et al., 1990; Sierra et al., 2005). Region IV is related to soluble microbial by-productlike substances (or SMP-like substances), and shows a maximum at approximately at 310/265 nm (Em/Ex). The sums of the fluorescence intensities for each region are listed in Table S6. A general decrease in the fluorescence intensity of all components was observed after nanofiltration. The intensities of regions I, II, and III seem to show little relationship with membrane wettability. The intensities of region V increased in the order: superhydrophilic < hydrophilic < hydrophobic < superhydrophobic, whereas the opposite sequence was found for region IV. SMP-like substances contain more hydrophilic than hydrophobic groups (Yamamura et al., 2014), and HA-like substances have greater hydrophobicity. These results suggested that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. In the same way, the superhydrophilic membrane showed better DOC removal rate as many foulants in natural water are hydrophobic (Table 1). In previous studies, the effects of NOM hydrophilicity/-phobicity and the membrane surface on fouling behavior were a controversial issue. Some researchers compared the fouling behavior of hydrophilic/ hydrophobic NOM on one type of membrane, whereas others compared the fouling behavior between hydrophilic/hydrophobic membranes with one type of NOM. The controversy originates from the unilateral consideration of membrane wettability or foulant hydrophilicity/-phobicity. In fact, both of the factors should be considered. In light of the aforementioned analyses, the superhydrophilic membrane showed higher NOM removal and better antihydrophobic NOM fouling ability than the some other types of membranes. It is noteworthy that the main limitation to the use of
129
Scheme 2. Combination of superhydrophilic and superhydrophobic membranes for nature water treatment.
membranes in drinking water treatment is their stability during long-term operation. A nanofiltration test of 60 h was conducted to examine the long-term stability of the superhydrophilic membrane. As shown in Fig. 5b, the membrane permeance and RUV254 were maintained at 154 L/m2$h$MPa and 79.1%, respectively. The former result was gratifying and surprising, showing that the membrane had excellent stability. As shown in Fig. 7, both the filtration time and J/J0 of the superhydrophilic membrane in this study were much higher than those of membranes described in the
Fig. 7. Comparisons of membrane permeance decline with the reported data. M presents the feed solution is model solution, N presents the feed solution is natural water (see Table S7).
130
L. Shan et al. / Water Research 93 (2016) 121e132
literature (Ates et al., 2009; Cui and Choo, 2013; Leiknes et al., 2004; Listiarini et al., 2010; Lowe and Hossain, 2008; Peeva et al., 2011; Shao et al., 2011, 2013; Song et al., 2011; Xia and Ni, 2015; Zhao et al., 2013). This excellent stability is due to the anti-fouling membrane surface and the stable chemically crosslinked structure. The anti-fouling properties of the superhydrophilic membrane were originated from the barrier provided by the tightly €ntt€ bound hydration layer on its surface (Ma ari et al., 2000; Tiraferri et al., 2012), which repels the approach of hydrophobic substances. The membrane inner structure is also important, in the superhydrophilic membrane, the positively charged PEI and negatively charged PSS are electrostatically crosslinked, causing PEI migrate into the PSS layers. In addition, PEI is crosslinked with glutaraldehyde, generating a very stable interwoven structure (Shan et al., 2015). The EEM analysis suggested that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. Therefore, it was expected that the superhydrophilic and superhydrophobic membranes could be used to remove hydrophobic and hydrophilic foulants, respectively. In a subsequent experiment, we attempted to combine the superhydrophilic membrane and superhydrophobic membrane to further improve the NOM removal from natural water. As shown in Scheme 2, the feed water was first nanofiltered through a superhydrophilic membrane to remove hydrophobic substances and then nanofiltered through a superhydrophobic membrane to remove hydrophilic substances. As shown in Fig. 6f and Table 1, the permeate from this combination of superhydrophilic and superhydrophobic membranes showed low fluorescence intensities and higher RUV254 and RDOC values. The RUV254 values for a single superhydrophilic membrane and a single superhydrophobic membrane were 79.1% and 59.7%, respectively. The combination of superhydrophilic and superhydrophobic membranes increased the RUV254 to 83.6%. The RDOC values of a single superhydrophilic membrane and a single superhydrophobic membrane were 37.3% and 13.3%, respectively. Gratifyingly, the RDOC increased to 73.3% with the combination of superhydrophilic and superhydrophobic membranes, almost double that of the superhydrophilic membrane. This suggested that the combination of superhydrophilic and superhydrophobic membranes had a great effect on most of the impurities in natural water, not just the NOM. This is ascribed to the formation of a hydration layer on the superhydrophilic membrane that prevents the approach of hydrophobic substances to its surface, in addition to a gas layer on the superhydrophobic membrane surface that prevents the approach of hydrophilic substances (Ishida et al., 2000; Jung and Bhushan, 2007; Meyer et al., 2006; Nosonovsky and Bhushan, 2005, 2006; Tyrrell and Attard, 2001; Yoshimitsu et al., 2002). From these results, the combination of superhydrophilic and superhydrophobic membranes is evidently a simple way to improve the water quality of nanofiltration permeate. 5. Conclusions Both superhydrophilic and superhydrophobic membranes with stable structure have been successfully obtained by using two facile hybrid routes. The XDLVO analyses and nanoindentor scratch test results indicated that superhydrophilic membrane showed the strongest repulsion to HA, and weakest adhesion strength of the HA-fouling layer on this membrane surface. Nanofiltration results with natural water showed the superhydrophilic membrane had an NOM permeance of 157 L/m2$h$MPa and a NOM rejection of 79.1%. Long-term (60 h) experiments suggested that the superhydrophilic
membrane had excellent stability without a decline in permeance during a long-term run, which is crucial for practical application. EEM analyses indicated that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. The RUV254 and RDOC could be increased to 83.6% and 73.3%, respectively, by a combination of superhydrophilic and superhydrophobic membranes. In view of the hydrophilic and hydrophobic NOM fouling behaviors on superwetting membrane surfaces, the superwetting strategy can be regarded as a very important approach for obtaining a membrane with both high NOM rejection and good anti-NOM fouling properties. Acknowledgments This work was supported by the New Century Excellent Researcher Award Program from Ministry of Education of China (NCET-12-0604); Fok Ying Tung Education Foundation (No.131068); and the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20121103110010); the Natural Science Foundation of Beijing, China (8122010). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.watres.2016.01.054. References Al-Amoudi, A.S., 2010. Factors affecting natural organic matter (NOM) and scaling fouling in NF membranes: a review. Desalination 259, 1e10. Amy, G.L., Cho, J., 1999. Interactions between natural organic metter (NOM) and membranes: rejection and fouling. Water Sci. Technol. 40 (9), 131e139. Ates, N., Yilmaz, L., Kitis, M., Yetis, U., 2009. Removal of disinfection by-product precursors by UF and NF membranes in low-SUVA waters. J. Membr. Sci. 328, 104e112. Bouchard, C., Jolicoeur, J., Kouadio, P., Britten, M., 1997. Study of humic acid adsorption on nanofiltration membranes by contact angle measurements. Can. J. Chem. Eng. 75 (2), 339e345. Brant, J.A., Childress, A.E., 2002. Assessing short-range membrane-colloid interactions using surface energetics. J. Membr. Sci. 203, 257e273. Cai, Z., Benjamin, M.M., 2011. NOM fractionation and fouling of low-pressure membranes in microgranular adsorptive filtration. Environ. Sci. Technol. 45, 8935e8940. Cai, Z., Kim, J., Benjamin, M.M., 2008. NOM removal by adsorption and membrane filtration using heated aluminum oxide particles. Environ. Sci. Technol. 42, 619e623. Chen, L., Tian, Y., Cao, C., Zhang, J., Li, Z., 2012. Interaction energy evaluation of soluble microbial products (SMP) on different membrane surfaces: role of the reconstructed membrane topology. Water Res. 46, 2693e2704. Chen, W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence excitationemission matrix regional integration to quantify spectra for dissolved organic matter. Environ. Sci. Technol. 37, 5701e5710. Childress, A.E., Elimelech, M., 1996. Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. J. Membr. Sci. 119 (2), 253e268. Cho, J., Amy, G.L., Pellegrino, J., 2000. Membrane filtration of natural organic matter: factors and mechanisms affecting rejection and flux decline characteristics with charged ultrafiltration (UF) membrane. J. Membr. Sci. 164, 89e110. Coble, P.G., Green, S.A., Blough, N.V., Gagosian, R.B., 1990. Characterization of dissolved organic matter in the Black Sea by fluorescence spectroscopy. Nature 348, 432e435. Cui, X., Choo, K.-H., 2013. Granular iron oxide adsorbents to control natural organic matter and membrane fouling in ultrafiltration water treatment. Water Res. 47, 4227e4237. Fan, L., Harris, J.L., Roddick, F.A., Booker, N.A., 2001. Influence of the characteristics of natural organic matter on the fouling of microfiltration membranes. Water Res. 35 (18), 4455e4463. Garp, O., Huisman, C.L., Reller, A., 2004. Photoinduced reactivity of titanium dioxide. Prog. Solid State Chem. 32, 133e177. Gong, L., Zhang, L., Wang, N., Li, J., Ji, S., Guo, H., Zhang, G., Zhang, Z., 2014. In situ ultraviolet-light-induced TiO2 nanohybrid superhydrophilic membrane for pervaporation dehydration. Sep. Purif. Technol. 122, 32e40.
L. Shan et al. / Water Research 93 (2016) 121e132 Gourley, L., Britten, M., Gauthier, S., Pouliot, Y., 1994. Characterization of adsorptive fouling on ultrafiltration membranes by peptides mixtures using contact angle measurement. J. Membr. Sci. 97, 283e289. Guan, K., 2005. Relationship between photocatalytic activity, hydrophilicity and self-cleaning effect of TiO2/SiO2 films. Surf. Coatings Technol. 191, 155e160. Hao, Y., Moriya, A., Maruyama, T., Ohmukai, Y., Matsuyama, H., 2011. Effect of metal ions on humic acid fouling of hollow fiber ultrafiltration membrane. J. Membr. Sci. 376, 247e253. Henderson, R.K., Baker, A., Murphy, K.R., Hambly, A., Stuetz, R.M., Khan, S.J., 2009. Fluorescence as a potential monitoring tool for recycled water system: a review. Water Res. 43, 863e881. Her, N., Amy, G., Plottu-Pecheux, A., Yoon, Y., 2007. Identification of nanofiltration membrane foulants. Water Res. 41, 3936e3947. Hong, S., Elimelech, M., 1997. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. J. Membr. Sci. 132 (2), 159e181. Howe, K.J., Clark, M.M., 2002. Fouling of microfiltration and ultrafiltration membranes by natural waters. Environ. Sci. Technol. 36, 3571e3576. Huang, H., Cho, H.-H., Jacangelo, J.G., Schwab, K.J., 2012. Mechanisms of membrane fouling control by integrated magnetic ion exchange and coagulation. Environ. Sci. Technol. 46, 10711e10717. Ishida, N., Inoue, T., Miyahara, M., Higashitani, K., 2000. Nano bubbles on a hydrophobic surface in water observed by tapping-mode atomic force microscopy. Langmuir 16, 6377e6380. Jaouadi, M., Amdouni, N., Duclaux, L., 2012. Characteristics of natural organic matter extracted from the waters of Medjerda dam (Tunisia). Desalination 305, 64e71. Jermann, D., Pronk, W., Boller, M., 2008. Mutual influences between natural organic matter and inorganic particles and their combined effect on ultrafiltration membrane fouling. Environ. Sci. Technol. 42, 9129e9136. Jermann, D., Pronk, W., Meylan, S., Boller, M., 2007. Interplay of different NOM fouling mechanisms during ultrafiltration for drinking water production. Water Res. 41, 1713e1722. Jung, Y.C., Bhushan, B., 2007. Wetting transition of water droplets on superhydrophobic patterned surfaces. Scr. Mater. 57, 1057e1060. Kaiya, Y., Itoh, Y., Takizawa, S., Fujita, K., Tagawa, T., 2000. Analysis of organic matter causing membrane fouling in drinking water treatment. Water Sci. Technol. 41, 59e67. Kang, S.-T., Subramani, A., Hoek, E.M.V., Deshusses, M.A., Matsumoto, M.R., 2004. Direct observation of biofouling in cross-flow microfiltration: mechanisms of deposition and release. J. Membr. Sci. 244, 151e165. Kilduff, J.E., Mattaraj, S., Belfort, G., 2004. Flux decline during nanofiltration of naturally-occurring dissolved organic matter: effects of osmotic pressure, membrane permeability, and cake formation. J. Membr. Sci. 239, 39e53. Kim, B.R., Ko, M.J., 2009. The assessment of the fracture behavior in spin-on organosilicates by nanoindentation and nanoscratch tests. Thin Solid Films 517, 3216e3221. Kim, S., Marion, M., Jeong, B.-H., Hoek, E.M.V., 2006. Crossflow membrane filtration of interacting nanoparticle suspensions. J. Membr. Sci. 284, 361e372. Lahoussine-Turcaud, V., Wiesner, M.R., Bottero, J.-Y., 1990. Fouling in tangentialflow ultrafiltration: the effect of colloid size and coagulation pretreatment. J. Membr. Sci. 52, 173e190. , G., Pernot, E., Joud, J.C., 2006. Langlet, M., Permpoon, S., Riassetto, D., Berthome Photocatalytic activity and photo-induced superhydrophilicity of sol-gel derived TiO2 films. J. Photochem. Photobiol. A-Chem. 181, 203e214. , J.-P., Buisson, H., 2004. Identification and understanding of Lee, N.H., Amy, G., Croue fouling in low-pressure membrane (MF/UF) filtration by natural organic matter (NOM). Water Res. 38, 4511e4523. Leiknes, T., Ødegaard, H., Myklebust, H., 2004. Removal of natural organic matter (NOM) in drinking water treatment by coagulationemicrofiltration using metal membranes. J. Membr. Sci. 242, 47e55. Li, J., Wang, N., Yan, H., Ji, S., Zhang, G., 2014. Designing superhydrophobic surfaces with SAM modification on hierarchical ZIF-8/polymer hybrid membranes for efficient bioalcohol pervaporation. RSC Adv. 4, 59750e59753. Li, K., Xiong, J., Chen, T., Yan, L., Dai, Y., Song, D., Lv, Y., Zeng, Z., 2013. Preparation of graphene/TiO2 composites by nonionic surfactant strategy and their simulated sunlight and visible light photocatalytic activity towards representative aqueous POPs degradation. J. Hazard. Mater. 250e251, 19e28. Liang, S., Qi, G., Xiao, K., Sun, J., Giannelis, E.P., Huang, X., Elimelech, M., 2014. Organic fouling behavior of superhydrophilic polyvinylidene fluoride (PVDF) ultrafiltration membranes functionalized with surface-tailored nanoparticles: Implications for organic fouling in membrane bioreactors. J. Membr. Sci. 463, 94e101. Listiarini, K., Tor, J.T., Sun, D.D., Leckie, J.O., 2010. Hybrid coagulationenanofiltration membrane for removal of bromate and humic acid in water. J. Membr. Sci. 365, 154e159. Lowe, J., Hossain, MdM., 2008. Application of ultrafiltration membranes for removal of humic acid from drinking water. Desalination 218, 343e354. Luo, Y., Zhang, J., Sun, A., Chu, C., Zhou, S., Guo, J., Chen, T., Xu, G., 2014. Electric field induced structural color changes of SiO2 @ TiO2 core-shell colloidal suspensions. J. Mater. Chem. C 2, 1990e1994. Madaenia, S.S., Ghaemi, N., Alizadeh, A., Joshaghani, M., 2011. Influence of photoinduced superhydrophilicity of titanium dioxide nanoparticles on the antifouling performance of ultrafiltration membranes. Appl. Surf. Sci. 257, 6175e6180. €ntta €ri, M., Puro, L., Nuortila-Jokinen, J., Nystro €m, M., 2000. Fouling effects of Ma
131
polysaccharides and humic acid in nanofiltration. J. Membr. Sci. 165, 1e17. Maximous, N., Nakhla, G., Wan, W., 2009. Comparative assessment of hydrophobic and hydrophilic membrane fouling in wastewater applications. J. Membr. Sci. 339, 93e99. Meyer, E.E., Rosenberg, K.J., Isreaelachvili, J., 2006. Recent progress in understanding hydrophobic interactions. Proc. Natl. Acad. Sci. U. S. A. 103, 15739e15746. Mills, A., Elliott, N., Parkin, I.P., O'Neill, S.A., Clark, R.J., 2002. Novel TiO2 CVD films for semiconductor photocatalysis. J. Photochem. Photobiol. A-Chem. 151, 171e179. Ng, M., Liu, S., Chow, C.W.K., Drikas, M., Amal, R., Lim, M., 2013. Understanding effects of water characteristics on natural organic matter treatability by PACI and a novel PACI-chitosan soagulants. J. Hazard. Mater. 263, 718e725. Nilson, J.A., DiGiano, F.A., 1996. Influence of NOM composition on nanofiltration. J. Am. Water Works Assoc. 88 (5), 53e66. Nosonovsky, M., Bhushan, B., 2005. Roughness optimization for biomimetic superhydrophobic surfaces. Microsyst. Technol. 11, 535e549. Nosonovsky, M., Bhushan, B., 2006. Wetting of rough three-dimensional superhydrophobic surfaces. Microsyst. Technol. 12, 273e281. Peeva, P.D., Palupi, A.E., Ulbricht, M., 2011. Ultrafiltration of humic acid solutions through unmodified and surface functionalized low-fouling polyethersulfone membranes e effects of feed properties, molecular weight cut-off and membrane chemistry on fouling behavior and cleanability. Sep. Purif. Technol. 81, 124e133. Peiris, R.H., Budman, H., Moresoli, C., Legge, R.L., 2010. Understanding fouling behavior of ultrafiltration membrane processes and natural water using principal component analysis of fluorescence excitation-emission matrices. J. Membr. Sci. 357, 62e72. Saravia, F., Zwiener, C., Frimmel, F.H., 2006. Interactions between membrane surface, dissolved organic substances and ions in submerged membrane filtration. Desalination 192, 280e287. Shan, L., Guo, H., Qin, Z., Wang, N., Ji, S., Zhang, G., Zhang, Z., 2015. Covalent crosslinked polyelectrolyte complex membrane with high negative charges towards anti-natural organic matter fouling nanofiltration. RSC Adv. 5, 11515e11523. Shao, J., Hou, J., Song, H., 2011. Comparison of huimic acid rejection and flux decline during filtration with negatively charged and uncharged ultrafiltration membranes. Water Res. 45, 473e482. Shao, J., Zhao, L., Chen, X., He, Y., 2013. Humic acid rejection and flux decline with negative charged membranes of different spacer arm lengths and charge groups. J. Membr. Sci. 435, 38e45. Sierra, M.M.D., Giovanela, M., Parlanti, E., Soriano-Sierra, E.J., 2005. . Fluorescence fingerprint of fulvic and humic acids from varied origins as viewed by singlescan and excitation/emission matrix techniques. Chemosphere 58, 715e733. Song, H., Shao, J., He, Y., Hou, J., Chao, W., 2011. Natural organic matter removal and flux decline with charged ultrafiltration and nanofiltration membranes. J. Membr. Sci. 376, 179e187. Thorsen, T., 2004. Concentration polarization by natural organic matter (NOM) in NF and UF. J. Membr. Sci. 233, 79e91. Tiraferri, A., Kang, Y., Giannelis, E.P., Elimelech, M., 2012. Superhydrophilic thin-film composite forward osmosis membranes for organic fouling control: fouling behavior and antifouling mechanisms. Environ. Sci. Technol. 46, 11135e11144. Tyrrell, J.W.G., Attard, P., 2001. Images of nanobubbles on hydrophobic surfaces and their interactions. Phys. Rev. Lett. 87, 176104. Van der Bruggen, B., Vandecasteele, C., Gestel, T.V., Doyen, W., Leysen, R., 2003. A review of pressure-driven membrane processes in wastewater treatment and drinking water production. Environ. Prog. 22, 46e56. Van Oss, C., Good, R., 1988. Orientation of the water molecules of hydration of human serum albumin. J. Protein Chem. 7 (2), 179e183. Vrouwenvelder, H.S., van Paassen, J.A.M., Folmer, H.C., Hofman, J.A.M.H., Nederlof, M.M., van der Kooij, D., 1998. Biofouling of membranes for drinking water production. Desalination 118, 157e166. Wang, K., Zhang, J., Lou, L., Yang, S., Chen, Y., 2004. UV or visible light induced photodegradation of AO7 on TiO2 particles: the influence of inorganic anions. J. Photochem. Photobiol. A-Chem. 165, 201e207. Wang, Q., Wang, Z., Zhu, C., Mei, X., Wu, Z., 2013. Assessment of SMP fouling by foulant-membrane interaction energy analysis. J. Membr. Sci. 446, 154e163. Wei, W., Xia, S., Liu, G., Gu, X., Jin, W., Xu, N., 2010. Interfacial adhesion between polymer separation layer and ceramic support for composite membrane. AIChE J. 56, 1584e1592. Xia, S., Ni, M., 2015. Preparation of poly(vinylidene fluoride) membranes with graphene oxide addition for natural organic matter removal. J. Membr. Sci. 473, 54e62. Xiao, K., Wang, X., Huang, X., Waite, T.D., Wen, X., 2011. Combined effect of membrane and foulant hydrophobicity and surface charge on adsorptive fouling during microfiltration. J. Membr. Sci. 373, 140e151. Yamamura, H., Okimoto, K., Kinura, K., Yoshimasa, W., 2014. Hydrophilic fraction of natural organic matter causing irreversible fouling of microfiltration and ultrafiltration membranes. Water Res. 54, 123e136. Yang, Q., Liu, Y., Li, Y., 2010. Humic acid fouling mitigation by antiscalant in reverse osmosis system. Environ. Sci. Technol. 44, 5153e5158. Yoshimitsu, Z., Nakajima, A., Watanabe, T., Hashimoto, K., 2002. Effect of surface structure on the hydrophobicity and sliding behavior of water droplets. Langmuir 18, 5818e5822. Zhang, L., Chowdhury, G., Feng, C., Matsuura, T., Narbaitz, R., 2003a. Effect of surface-modifying macromolecules and membrane morphology on fouling of polyethersulfone ultrafiltration membranes. J. Appl. Polym. Sci. 88, 3132e3138.
132
L. Shan et al. / Water Research 93 (2016) 121e132
Zhang, M., Li, C., Benjamin, M.M., Chang, Y., 2003b. Fouling and natural organic matter removal in adsorbent/membrane systems for drinking water treatment. Environ. Sci. Technol. 37, 1663e1669. Zhang, X., Hu, L., Sun, D., 2006. Nanoindentation and nanoscratch profiles of hybrid films based on (g-methacrylpropyl) trimethoxysilane and tetraethoxysilane. Acta Mater. 54, 5469e5475. Zhao, Y., Zhou, S., Li, M., Xue, A., Zhang, Y., Wang, J., Xing, W., 2013. Humic acid
removal and easy-cleanability using temperature responsive ZrO2 tubular membranes grafted with poly(N-isopropylacrylamide) brush chains. Water Res. 47, 2375e2386. Zhou, M., Liu, H., Kilduff, J.E., Langer, R., Anderson, D.G., 2009. Belfort G. Highthoughput membrane surface modification to control NOM fouling. Environ. Sci. Technol. 43, 3865e3871.
Contents lists available at ScienceDirect
Water Research journal homepage: www.elsevier.com/locate/watres
Natural organic matter fouling behaviors on superwetting nanofiltration membranes Linglong Shan a, Hongwei Fan a, Hongxia Guo b, **, Shulan Ji a, Guojun Zhang a, * a Beijing Key Laboratory for Green Catalysis and Separation, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing 100124, PR China b College of Material Science and Engineering, Beijing University of Technology, Beijing 100124, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 November 2015 Received in revised form 19 January 2016 Accepted 24 January 2016 Available online xxx
Nanofiltration has been widely recognized as a promising technology for the removal of micro-molecular organic components from natural water. Natural organic matter (NOM), a very important precursor of disinfection by-products, is currently considered as the major cause of membrane fouling. It is necessary to develop a membrane with both high NOM rejection and anti-NOM fouling properties. In this study, both superhydrophilic and superhydrophobic nanofiltration membranes for NOM removal have been fabricated. The fouling behavior of NOM on superwetting nanofiltration membranes has been extensively investigated by using humic acid (HA) as the model foulant. The extended DerjaguinLandauVerweyOverbeek approach and nanoindentor scratch tests suggested that the superhydrophilic membrane had the strongest repulsion force to HA due to the highest positive total interaction energy (DGTOT) value and the lowest critical load. Excitation emission matrix analyses of natural water also indicated that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal thereof. Conversely, the superhydrophobic membrane showed resistance to fouling by hydrophilic substances and therefore high removal capacity. Long-term operation suggested that the superhydrophilic membrane had high stability due to its anti-NOM fouling capacity. Based on the different anti-fouling properties of the studied superwetting membranes, a combination of superhydrophilic and superhydrophobic membranes was examined to further improve the removal of both hydrophobic and hydrophilic pollutants. With a combination of superhydrophilic and superhydrophobic membranes, the NOM rejection (RUV254) and DOC removal rates (RDOC) could be increased to 83.6% and 73.3%, respectively. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Superhydrophilic Superhydrophobic Nanofiltration NOM Fouling
1. Introduction In natural water, natural organic matter (NOM) can affect the color, taste and odor of raw water, moreover, it has been identified as precursors of disinfection by-products. NOM can be divided into three categories: humic acid (HA), fulvic acid (FA) and humin, with molecular weights ranging from 500 kDa (Jaouadi et al., 2012; Lee et al., 2004; Ng et al., 2013). Nanofiltration has been widely recognized as a promising technology to remove NOM from natural water due to its high efficiency for removal of pollutants, low energy consumption, and easy accessibility. With the
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (H. Guo), [email protected] (G. Zhang). http://dx.doi.org/10.1016/j.watres.2016.01.054 0043-1354/© 2016 Elsevier Ltd. All rights reserved.
improvement of drinking water standards and the industrialized application of membrane technology, it is critical to develop novel nanofiltration membranes with high NOM removal and stability. Concurrently, the biggest challenge is the inevitable occurrence of membrane fouling, which is caused by the NOM accumulation on the membrane surface during the NOM removal process in drinking water treatment and production (Al-Amoudi, 2010; Cai et al., 2008; Cai and Benjamin, 2011; Howe and Clark, 2002; Huang et al., 2012; Jermann et al., 2007, 2008; Lahoussine-Turcaud et al., 1990; Saravia et al., 2006; Van der Bruggen et al., 2003; Vrouwenvelder et al., 1998; Yang et al., 2010; Zhang et al., 2003b; Zhou et al., 2009). In particular, HA has been considered as the most important foulant for the membrane flux decline. Membrane fouling is an issue of interface action between the foulant, membrane, and solution. NOM fouling is mainly determined by the interface relationship between NOM and the
122
L. Shan et al. / Water Research 93 (2016) 121e132
membrane surface in aqueous solution. Many studies have been carried out to investigate this point; however, the effect of hydrophilicity/-phobicity of NOM and the membrane surface on fouling behavior is still a controversial issue. Some researchers have studied the effect of NOM hydrophilicity/-phobicity on membrane fouling, since NOM can be fractionated into relatively hydrophilic and hydrophobic components (Hao et al., 2011; Kilduff et al., 2004; €ntta €ri et al., 2000; Thorsen, 2004). Amy and Nilson found that Ma the hydrophobic fraction was the major factor for flux decline, while the hydrophilic fraction was much less responsible for fouling of the membrane (Amy and Cho, 1999; Nilson and DiGiano, 1996). In contrast, Cho et al. (2000), Fan et al. (2001) and Yamamura et al. (2014) obtained completely opposite results, and they found that the hydrophilic fractions induced more severe loss of membrane permeability than the hydrophobic fractions. Although some previous studies indicated that hydrophilicity/-phobicity of the membrane surface was a predominant influence on the extent of fouling in aqueous systems (Maximous et al., 2009; Xiao et al., 2011), contradictory results were also often be obtained. Zhang et al. (2003a) reported that with the increase of surface hydrophobicity, a smaller amount of HA was deposited onto the membrane surface and the permeate flux reduction decreased. Conversely, many reports have indicated that a hydrophilic surface favors less fouling, due to a strong hydration layer that repels the €ntt€ adsorption of foulants on the membrane surface (Ma ari et al., 2000; Tiraferri et al., 2012). Thus, determining the effect of hydrophilicity/-phobicity of NOM and membrane surface wettability on its fouling behavior is still a challenge. To date, most studies in this field have concerned in the wettability of membranes with water contact angles in the range 30120 , covering the range from hydrophilic to hydrophobic surfaces. It is notable that some extraordinary membrane performances have been achieved when the wettability was extended to superhydrophilicity or superhydrophobicity. It is generally accepted that the solid surfaces with contact angles 150 are defined as superhydrophobic. Tiraferri et al. (2012) and Liang et al. (2014) investigated the fouling behavior of superhydrophilic membranes with model wastewater and found that these membranes showed lower overall flux decline and anti-fouling properties. Our previous study indicated that the pervaporation flux of a superhydrophilic membrane was twice than that of a hydrophilic membrane (Gong et al., 2014). We also demonstrated that the superhydrophobic pervaporation membrane showed both higher selectivity and higher permeability than a hydrophobic membrane (Li et al., 2014). However, little work has dealt with superwetting nanofiltration membranes for NOM removal from natural water. Additionally, there has been no systematic work on NOM fouling behavior on superwetting nanofiltration membranes. The aim of this study is to develop the superwetting nanofiltration membranes with both high NOM rejection and good antiNOM fouling properties. We extensively studied the NOM fouling behavior on superwetting nanofiltration membranes. The interaction forces and free energy of HA adhesion on superwetting membrane surfaces were elucidated by the extended DerjaguinLandauVerweyOverbeek (XDLVO) thermodynamic analyses. Then we carried out the nanoindentor scratch tests to quantify the adhesion between the HA fouling layer and membrane surface. In addition, the NOM fouling of hydrophilic and hydrophobic components on superhydrophilic and superhydrophobic nanofiltration membranes were analyzed by excitation emission matrix (EEM). The natural water treatment performances of the superwetting membranes and their long-term stability were investigated using real reservoir water.
2. Theoretical The XDLVO model has widely been used to explain how aqueous foulants interact with surfaces of polymeric membranes (Brant and Childress, 2002; Kang et al., 2004; Kim et al., 2006). According to XDLVO theory, in an aquatic environmental system, the interfacial energy between a membrane and foulants is the sum of the Lifshitzevan der Waals, Lewis acidbase, and electrostatic doublelayer interactions, which is given by (Brant and Childress, 2002): LW AB EL GTOT mwf ¼ Gmwf þ Gmwf þ Gmwf
(1)
where GTOT mwf is the total interaction energy between the membrane AB EL and foulants and GLW mwf , Gmwf and Gmwf represent the Lifshitzevan der Waals, Lewis acidbase, and electrostatic double-layer free energies, respectively. The subscripts m, w, and f correspond to the membrane, water, and the foulant, respectively. 2.1. Surface tension components The surface tension components of the membrane and foulants are determined from the extended Young equation (Bouchard et al., 1997; Gourley et al., 1994; Van Oss and Good, 1988), which gives their relationship with the contact angle of a liquid on a solid surface. The surface tension parameters of both the solid and liquid can be written as follows:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffi LW þ gþ ð1 þ cosqÞgTOT ¼ 2 gLW g gþ s gl þ s gl l s
(2)
pffiffiffiffiffiffipffiffiffiffiffiffi gþ g
(3)
gTOT ¼ gLW þ gAB
(4)
gAB ¼ 2
where q is the contact angle, gTOT is the total surface tension, gLW is the Lifshitzevan der Waals component, and gþ and g are the electron-acceptor and electron-donor components, respectively. The subscripts s and l correspond to the solid surface and liquid, respectively. The surface tension parameters of foulants (gþ , g , f f þ , g , gLW ) can be determined from gLW ) and membranes (g m m m f equation (2) after measuring the contact angle data for three probe liquids with known surface tension parameters (gþ , g , gLW ). l l l 2.2. Adhesion free energy The free energy of adhesion per unit area was calculated as follows:
△GLW mwf ¼ 2
qffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiqffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi gLW gLW gLW gLW w m w f
(5)
qffiffiffiffiffiffiffi qffiffiffiffiffiffi qffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi þ g Þ þ 2 g gþ gþ g ð g gþ w mþ m w w f f qffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffi þ g gþ gþ w 2 m gf þ m gf
△GAB mwf ¼ 2
(6) △GEL mwf
# " ε0 εr k 2 2xm xf 2 xm þ xf 1 cothðkho Þ þ 2 ¼ cschðkho Þ 2 xm þ x2f (7)
AB EL 2 where △GLW mwf , △Gmwf and △Gmwf , expressed in mJ$m , are the
L. Shan et al. / Water Research 93 (2016) 121e132
Lifshitzevan der Waals, Lewis acidbase, and electrostatic doublelayer free energies at a separation distance of h0 (nm), respectively; h0 is the minimum equilibrium cut-off distance, usually assigned a value of 0.158 nm (±0.009 nm) (Brant and Childress, 2002; Chen et al., 2012; Kang et al., 2004); ε0εr is the dielectric permittivity of the suspending fluid, usually assigned as 6.95 1010 C2 J1 m1 (Brant and Childress, 2002; Kim et al., 2006); k is the inverse Debye screening length, assigned as 0.104 nm1 (Brant and Childress, 2002; Kim et al., 2006); and xm and xf, expressed in mV, are the surface potentials of the membrane and foulant, respectively. AB △GLW mwf and △Gmwf could be determined from the surface tension LW parameters of foulants (gþ , g , gLW ), membranes (gþ m , gm , gm ) f f f , gLW ) obtained in 2.1. and water (gþ , g w w w
2.3. Interaction energy between a spherical foulant and an infinite planar surface The interaction between a membrane and a foulant in an aqueous environment can be expressed as follows:
ULW mwf
¼
2p△GLW mwf
ah20 h
! (8)
h0 h AB ¼ 2pal△G exp UAB mwf mwf l " UEL mwf ¼ pε0 εr a 2xf xm In
1 þ ekh 1 ekh
!
(9) þ x2f þ x2m In 1 e2kh
#
(10) AB EL where ULW mwf , Umwf and Umwf are the Lifshitzevan der Waals, Lewis acid-base, and electrostatic double-layer interaction energies. a is the radius of the foulant (nm); h is the separation distance between the membrane and foulant (nm); and l is the characteristic decay length of the AB interaction in water, commonly measured as 0.6 nm for aqueous systems (Brant and Childress, 2002; Kim et al., 2006).
3. Experimental 3.1. Chemicals and materials A polyacrylonitrile (PAN) ultrafiltration membrane with a molecular weight cut-off of 100 kDa (PAN-50) and a polysulfone (PS) ultrafiltration membrane with a cut-off of 20,000 were purchased from Sepro Membranes (Oceanside, CA, USA). Unless otherwise specified, all reagents and chemicals were of analytical grade. HA, poly(ethyleneimine) (PEI) (Mw ¼ 750,000), and poly(sodium 4styrenesulfonate) (PSS) (Mw ¼ 700,000) were obtained from Aldrich (St. Louis, MO, USA). Hydrochloric acid (HCl), glutaraldehyde, potassium chloride (KCl), sodium hydroxide (NaOH), tetraethyl orthosilicate, n-heptane, dibutyltin dilaurate, ethanol, and glycerol were provided by Beijing Chemical Factory (Beijing, China). Dihydroxypolydimethylsiloxane (PDMS) with viscosity of 2500 Pa S was purchased from China Bulestar Chengrand Chemical Co., Ltd. (Chengdu, China). Silica nanoparticles (7e40 nm particle size), titanium oxide nanoparticles (5e10 nm particle size), and diiodomethane were purchased from Shanghai Crystal Pure Biological Technology Co., Ltd. (Shanghai, China). Natural water was collected from Miyun reservoir in Beijing on March 13th, 2015.
123
3.2. Preparation of the membranes Superhydrophilic and superhydrophobic membranes were prepared, along with hydrophilic and hydrophobic membranes for comparison. The superhydrophilic membrane was prepared as illustrated in Scheme 1a, a PAN substrate was hydrolyzed with 2.0 mol/L NaOH, rinsed with deionized water, then immersed in 1.0 mg/mL PEI polycation solution for 20 min. It was then rinsed three times with deionized water, and the pH of the PEI solution was adjusted to 1. The substrate was then immersed in PSS-TiO2 mixed polyanion solution for 20 min and rinsed once more with deionized water; the pH of the PSS-TiO2 solution was 7. The concentration of TiO2 in the 1.0 mg/mL PSS polyanion solution was 0.25 mg/mL. This procedure was repeated six times and the membrane was then crosslinked by immersing in 0.02 mg/mL aqueous glutaraldehyde solution at 30 C for 48 h. Finally, the membrane was irradiated for 40 min using a UV lamp to form the superhydrophilic surface. The superhydrophobic membrane was prepared as illustrated in Scheme 1b, a PS substrate was immersed in a 30 wt% ethanol solution for 2 h at room temperature. The residue on the surface of the membrane was then rinsed off using deionized water. A vacuum pump was used to create a negative pressure to extract air from the membrane pores over a period of 4 h. In addition, 1.0 wt% of PDMS was dissolved in n-heptane and the solution was stirred for 1 h. Specified amounts of silica nanoparticles were then dispersed in n-heptane by stirring for 1 h (PDMS/silica mass ratio 1:0.5). Finally, 0.1 wt% tetraethyl orthosilicate and 0.05 wt% dibutyltin dilaurate were also dissolved in n-heptane and the resulting mixture was continuously stirred for 1 h. To better disperse the inorganic particles in the hybrid membrane, the surface of the PS substrate was coated with the suspension by sonication during the assemble process. For this, the PS substrate was dip-coated in the suspension for 30 s. Subsequently, the membrane was fixed onto a substrate and irradiated with a lamp to accelerate the removal of solvent from the film surfaces. The above steps were repeated twice more to obtain a multilayer membrane. The samples were then kept in a convection oven set at 80 C for 8 h to complete crosslinking of the PDMS. The hydrophilic membrane was prepared in the same way with superhydrophilic membrane, except for the addition of TiO2 nanoparticles and UV lamp irradiation. For hydrophilic membrane, the procedure was repeated four times. The hydrophobic membrane was prepared in the same way with superhydrophobic membrane, except for the addition of SiO2 nanoparticles and sonication assistance. 3.3. Fouling experiments Prior to the fouling experiments with real natural water, HA was selected as a model foulant to study the interaction energy between the membranes and NOM. To obtain the surface tensions of the membranes and HA, their contact angles (CAs) were measured with three different probe liquids of known surface tension values. The surface tension properties of each probe liquid are listed in Table S1. The membrane samples were typically dried for 48 h in an oven at 30 C and then fixed onto glass slides prior to the CA measurements. The instantaneous CA was obtained within 3 s, and the volume of the liquid drop was 5 mL. The CA of each sample was measured at least ten times. To measure HA surface tension, HA solution was deposited onto a glass slide and dried for 24 h in an oven at 30 C to form a flat solid film. CA measurements were then performed as described above. The HA-fouled membranes used for XDLVO theory evolution and scratch tests were obtained by filtering the model HA solution for 10 h to ensure the formation of a
124
L. Shan et al. / Water Research 93 (2016) 121e132
Scheme 1. (a) Schematic illustration of the preparation of the superhydrophilic membrane and its surface wettability; (b) schematic illustration of the preparation of the superhydrophobic membrane and its surface wettability.
cake layer on the membrane surface. Natural water collected from Miyun reservoir was filtered using a 0.45 mm porosity membrane and stored in dark at 4 C. Nanofiltration performance was studied with a home-made cross-flow nanofiltration system, which has been described in our previous research (Shan et al., 2015). The pressure of nanofiltration was maintained at 0.6 MPa. The membranes were cut into disks with an effective nanofiltration area of 22.9 cm2. The rejection ratio, R, was calculated using Eq. (11), where Cf and Cp are the concentrations of the solute in the feed and permeate, respectively. The permeance, J, was calculated by Eq. (12), where V is the volume of the permeate liquid passed across the membrane of area A (m2) in the time period T (h) at operative pressure P (MPa). After a stable permeance was achieved by eluting with deionized water, the membrane fouling potential was determined by calculating the relative permeance (J/J0) in terms of the ratio of permeance (J) to the initial membrane permeance (J0).
R¼
Cf Cp 100% Cf
VðLÞ J ¼ 2 A m TðhÞ PðMPaÞ
(11)
(12)
3.4. Analytical methods The zeta potential and the mean hydrodynamic diameter of HA were determined by a laser scattering size analyzer (NICOMPTM 380ZLS, USA). The zeta potentials of the membrane surfaces were
determined using a SurPASS electrokinetic analyzer (Anton Paar GmbH, Graz, Austria). During the process of measuring the zeta potential, the KCl solution concentration was maintained at 0.83 mmol/L, while the operation pressure was 0.03 MPa. The pH values of the solutions were measured with a pH meter (Leici PHS3C, Shanghai, China). The membrane surface morphologies were observed by scanning electron microscopy (SEM) observation, all membrane samples were dried under vacuum and gold-coated before observation. The images were obtained using a SU-8020 instrument (Hitachi, Japan) at an acceleration voltage of 20 keV. Wide-angle X-ray diffraction (XRD) experiments were conducted on a D8 ADVANCE X-ray diffractometer (Bruker/AXS, Germany). Water contact angle measurements were performed using a contact angle analyzer (DSA100, Germany). The interfacial adhesion between the foulant and the membrane was measured by a scratch test with a nanoindentor (G200). Scratch tests were performed using a Nano Indentor XP system with options for lateral-force measurements. The maximum scratch load of 0.45 mN was applied, with profiling load and profiling velocity of 20 mN and 30 mm/s, respectively. The scratch length and scratch velocity are 500 mm and 30 mm/s, respectively. UV254 concentration was measured using a UV/Vis spectrophotometer (T6, Shanghai, China). The concentrations of dissolved organic carbon (DOC) were determined with a total organic carbon analyzer (TOC-VCPN, Shimadzu, Japan). The relative precision of the DOC analyses was hydrophilic (112 L/ m2 h MPa) > hydrophobic (75 L/m2 h MPa) > superhydrophobic (53 L/m2 h MPa). The corresponding NOM rejections of the membranes were 79.1%, 64.2%, 61.2%, and 59.7%, respectively. The DOC removal rates of the membranes were 37.3%, 36.8%, 16.5%, and 13.3%, respectively. The superhydrophilic membrane exhibited both high initial permeance and high rejection. Permeance decline was monitored by the ratio of the permeance to the initial permeance (J/J0). The curve for the superhydrophilic membrane remained unchanged, whereas that for the hydrophilic membrane showed a continuous decline. The curves for the hydrophobic and superhydrophobic membranes exhibited an initial slow decrease, then appeared to be maintained, and finally showed a sharp decrease. Many researchers have found that rapid membrane fouling occurs at high permeation rates (Childress and Elimelech, 1996; Hong and Elimelech, 1997), whereas in our study the superhydrophilic membrane showed both high permeance and exceptionally low fouling. The morphologies, CAs, and XRD patterns of the membrane surfaces after nanofiltration were investigated to understand the changes after fouling experiments. It can be seen in Fig. S2 that the rod-like foulant was deposited on the superhydrophobic membrane surface, which might be the reason for its serious permeance decrease. After the fouling experiments, TiO2 and SiO2 nanoparticles could still be detected on the superhydrophilic and superhydrophobic membrane surfaces, respectively (Fig. S1). The water CAs of both membranes remained at initial level (Figs. 1 and S2). The results suggested that the superwetting membrane structures were stable in the nanofiltration tests, because both membranes were chemically crosslinked during their formation processes (Scheme 1). After the fouling experiments, the permeate was collected for EEM analysis, which was used to probe the chemical structure of NOM because of its ability to distinguish between different classes of organic matter (Coble et al., 1990; Henderson et al., 2009; Peiris et al., 2010). EEM fluorescence spectra were measured and
L. Shan et al. / Water Research 93 (2016) 121e132
127
Fig. 3. Variation of interaction energy profiles between the (a) clean membranes and HA; (b) fouled membranes and HA.
Fig. 4. Scratch test of the fouled membrane.
Fig. 5. (a) Variation of natural water permeance and RUV254; (b) long-term nanofiltration behavior of superhydrophilic membrane with natural water.
Table 1 The water quality indexes of the natural water and the permeate.
Natural water Superhydrophilic Hydrophilic Hydrophobic Superhydrophobic Double membranes
Turbidity (NTU)
Alkalinity (mg/L)
pH
UV254
DOC (mg/L)
RUV254 (%)
RDOC (%)
1.09 0 0 0 0 0
147.6 70.1 115.1 135.1 149.3 45.6
8.07 8.05 8.03 8.05 8.07 8.04
0.067 0.014 0.024 0.026 0.027 0.011
12.15 7.616 7.673 10.14 10.54 3.19
e 79.1 64.2 61.2 59.7 83.6
e 37.3 36.8 16.5 13.3 73.7
128
L. Shan et al. / Water Research 93 (2016) 121e132
representative examples were shown in Fig. 6. As reported in the
literature (Chen et al., 2003), horizontal and vertical lines were
Fig. 6. EEMs for (a) natural water; (b) natural water nanofiltrated by superhydrophilic membrane; (c) natural water nanofiltrated by hydrophilic membrane; (d) natural water nanofiltrated by hydrophobic membrane; (e) natural water nanofiltrated by superhydrophobic membrane; (f) natural water nanofiltrated by superhydrophilic membranesuperhydrophobic membrane.
L. Shan et al. / Water Research 93 (2016) 121e132
drawn to divide the EEM into five regions according to the nature of the fluorescent organic matter. In general, regions I and II are related to aromatic proteins, and the peak at 335/220 nm (Em/Ex) is indicative of the presence of protein-like substances. Regions III and V are related to fulvic acid-like materials and HA-like organics, respectively, and show maxima at approximately 410/225 nm (Em/ Ex) and 415/310 nm (Em/Ex) respectively (Coble et al., 1990; Sierra et al., 2005). Region IV is related to soluble microbial by-productlike substances (or SMP-like substances), and shows a maximum at approximately at 310/265 nm (Em/Ex). The sums of the fluorescence intensities for each region are listed in Table S6. A general decrease in the fluorescence intensity of all components was observed after nanofiltration. The intensities of regions I, II, and III seem to show little relationship with membrane wettability. The intensities of region V increased in the order: superhydrophilic < hydrophilic < hydrophobic < superhydrophobic, whereas the opposite sequence was found for region IV. SMP-like substances contain more hydrophilic than hydrophobic groups (Yamamura et al., 2014), and HA-like substances have greater hydrophobicity. These results suggested that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. In the same way, the superhydrophilic membrane showed better DOC removal rate as many foulants in natural water are hydrophobic (Table 1). In previous studies, the effects of NOM hydrophilicity/-phobicity and the membrane surface on fouling behavior were a controversial issue. Some researchers compared the fouling behavior of hydrophilic/ hydrophobic NOM on one type of membrane, whereas others compared the fouling behavior between hydrophilic/hydrophobic membranes with one type of NOM. The controversy originates from the unilateral consideration of membrane wettability or foulant hydrophilicity/-phobicity. In fact, both of the factors should be considered. In light of the aforementioned analyses, the superhydrophilic membrane showed higher NOM removal and better antihydrophobic NOM fouling ability than the some other types of membranes. It is noteworthy that the main limitation to the use of
129
Scheme 2. Combination of superhydrophilic and superhydrophobic membranes for nature water treatment.
membranes in drinking water treatment is their stability during long-term operation. A nanofiltration test of 60 h was conducted to examine the long-term stability of the superhydrophilic membrane. As shown in Fig. 5b, the membrane permeance and RUV254 were maintained at 154 L/m2$h$MPa and 79.1%, respectively. The former result was gratifying and surprising, showing that the membrane had excellent stability. As shown in Fig. 7, both the filtration time and J/J0 of the superhydrophilic membrane in this study were much higher than those of membranes described in the
Fig. 7. Comparisons of membrane permeance decline with the reported data. M presents the feed solution is model solution, N presents the feed solution is natural water (see Table S7).
130
L. Shan et al. / Water Research 93 (2016) 121e132
literature (Ates et al., 2009; Cui and Choo, 2013; Leiknes et al., 2004; Listiarini et al., 2010; Lowe and Hossain, 2008; Peeva et al., 2011; Shao et al., 2011, 2013; Song et al., 2011; Xia and Ni, 2015; Zhao et al., 2013). This excellent stability is due to the anti-fouling membrane surface and the stable chemically crosslinked structure. The anti-fouling properties of the superhydrophilic membrane were originated from the barrier provided by the tightly €ntt€ bound hydration layer on its surface (Ma ari et al., 2000; Tiraferri et al., 2012), which repels the approach of hydrophobic substances. The membrane inner structure is also important, in the superhydrophilic membrane, the positively charged PEI and negatively charged PSS are electrostatically crosslinked, causing PEI migrate into the PSS layers. In addition, PEI is crosslinked with glutaraldehyde, generating a very stable interwoven structure (Shan et al., 2015). The EEM analysis suggested that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. Therefore, it was expected that the superhydrophilic and superhydrophobic membranes could be used to remove hydrophobic and hydrophilic foulants, respectively. In a subsequent experiment, we attempted to combine the superhydrophilic membrane and superhydrophobic membrane to further improve the NOM removal from natural water. As shown in Scheme 2, the feed water was first nanofiltered through a superhydrophilic membrane to remove hydrophobic substances and then nanofiltered through a superhydrophobic membrane to remove hydrophilic substances. As shown in Fig. 6f and Table 1, the permeate from this combination of superhydrophilic and superhydrophobic membranes showed low fluorescence intensities and higher RUV254 and RDOC values. The RUV254 values for a single superhydrophilic membrane and a single superhydrophobic membrane were 79.1% and 59.7%, respectively. The combination of superhydrophilic and superhydrophobic membranes increased the RUV254 to 83.6%. The RDOC values of a single superhydrophilic membrane and a single superhydrophobic membrane were 37.3% and 13.3%, respectively. Gratifyingly, the RDOC increased to 73.3% with the combination of superhydrophilic and superhydrophobic membranes, almost double that of the superhydrophilic membrane. This suggested that the combination of superhydrophilic and superhydrophobic membranes had a great effect on most of the impurities in natural water, not just the NOM. This is ascribed to the formation of a hydration layer on the superhydrophilic membrane that prevents the approach of hydrophobic substances to its surface, in addition to a gas layer on the superhydrophobic membrane surface that prevents the approach of hydrophilic substances (Ishida et al., 2000; Jung and Bhushan, 2007; Meyer et al., 2006; Nosonovsky and Bhushan, 2005, 2006; Tyrrell and Attard, 2001; Yoshimitsu et al., 2002). From these results, the combination of superhydrophilic and superhydrophobic membranes is evidently a simple way to improve the water quality of nanofiltration permeate. 5. Conclusions Both superhydrophilic and superhydrophobic membranes with stable structure have been successfully obtained by using two facile hybrid routes. The XDLVO analyses and nanoindentor scratch test results indicated that superhydrophilic membrane showed the strongest repulsion to HA, and weakest adhesion strength of the HA-fouling layer on this membrane surface. Nanofiltration results with natural water showed the superhydrophilic membrane had an NOM permeance of 157 L/m2$h$MPa and a NOM rejection of 79.1%. Long-term (60 h) experiments suggested that the superhydrophilic
membrane had excellent stability without a decline in permeance during a long-term run, which is crucial for practical application. EEM analyses indicated that the superhydrophilic membrane showed resistance to fouling by hydrophobic substances and therefore high removal capacity, whereas the superhydrophobic membrane showed good resistance to fouling by hydrophilic substances and therefore high removal capacity. The RUV254 and RDOC could be increased to 83.6% and 73.3%, respectively, by a combination of superhydrophilic and superhydrophobic membranes. In view of the hydrophilic and hydrophobic NOM fouling behaviors on superwetting membrane surfaces, the superwetting strategy can be regarded as a very important approach for obtaining a membrane with both high NOM rejection and good anti-NOM fouling properties. Acknowledgments This work was supported by the New Century Excellent Researcher Award Program from Ministry of Education of China (NCET-12-0604); Fok Ying Tung Education Foundation (No.131068); and the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20121103110010); the Natural Science Foundation of Beijing, China (8122010). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.watres.2016.01.054. References Al-Amoudi, A.S., 2010. Factors affecting natural organic matter (NOM) and scaling fouling in NF membranes: a review. Desalination 259, 1e10. Amy, G.L., Cho, J., 1999. Interactions between natural organic metter (NOM) and membranes: rejection and fouling. Water Sci. Technol. 40 (9), 131e139. Ates, N., Yilmaz, L., Kitis, M., Yetis, U., 2009. Removal of disinfection by-product precursors by UF and NF membranes in low-SUVA waters. J. Membr. Sci. 328, 104e112. Bouchard, C., Jolicoeur, J., Kouadio, P., Britten, M., 1997. Study of humic acid adsorption on nanofiltration membranes by contact angle measurements. Can. J. Chem. Eng. 75 (2), 339e345. Brant, J.A., Childress, A.E., 2002. Assessing short-range membrane-colloid interactions using surface energetics. J. Membr. Sci. 203, 257e273. Cai, Z., Benjamin, M.M., 2011. NOM fractionation and fouling of low-pressure membranes in microgranular adsorptive filtration. Environ. Sci. Technol. 45, 8935e8940. Cai, Z., Kim, J., Benjamin, M.M., 2008. NOM removal by adsorption and membrane filtration using heated aluminum oxide particles. Environ. Sci. Technol. 42, 619e623. Chen, L., Tian, Y., Cao, C., Zhang, J., Li, Z., 2012. Interaction energy evaluation of soluble microbial products (SMP) on different membrane surfaces: role of the reconstructed membrane topology. Water Res. 46, 2693e2704. Chen, W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence excitationemission matrix regional integration to quantify spectra for dissolved organic matter. Environ. Sci. Technol. 37, 5701e5710. Childress, A.E., Elimelech, M., 1996. Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. J. Membr. Sci. 119 (2), 253e268. Cho, J., Amy, G.L., Pellegrino, J., 2000. Membrane filtration of natural organic matter: factors and mechanisms affecting rejection and flux decline characteristics with charged ultrafiltration (UF) membrane. J. Membr. Sci. 164, 89e110. Coble, P.G., Green, S.A., Blough, N.V., Gagosian, R.B., 1990. Characterization of dissolved organic matter in the Black Sea by fluorescence spectroscopy. Nature 348, 432e435. Cui, X., Choo, K.-H., 2013. Granular iron oxide adsorbents to control natural organic matter and membrane fouling in ultrafiltration water treatment. Water Res. 47, 4227e4237. Fan, L., Harris, J.L., Roddick, F.A., Booker, N.A., 2001. Influence of the characteristics of natural organic matter on the fouling of microfiltration membranes. Water Res. 35 (18), 4455e4463. Garp, O., Huisman, C.L., Reller, A., 2004. Photoinduced reactivity of titanium dioxide. Prog. Solid State Chem. 32, 133e177. Gong, L., Zhang, L., Wang, N., Li, J., Ji, S., Guo, H., Zhang, G., Zhang, Z., 2014. In situ ultraviolet-light-induced TiO2 nanohybrid superhydrophilic membrane for pervaporation dehydration. Sep. Purif. Technol. 122, 32e40.
L. Shan et al. / Water Research 93 (2016) 121e132 Gourley, L., Britten, M., Gauthier, S., Pouliot, Y., 1994. Characterization of adsorptive fouling on ultrafiltration membranes by peptides mixtures using contact angle measurement. J. Membr. Sci. 97, 283e289. Guan, K., 2005. Relationship between photocatalytic activity, hydrophilicity and self-cleaning effect of TiO2/SiO2 films. Surf. Coatings Technol. 191, 155e160. Hao, Y., Moriya, A., Maruyama, T., Ohmukai, Y., Matsuyama, H., 2011. Effect of metal ions on humic acid fouling of hollow fiber ultrafiltration membrane. J. Membr. Sci. 376, 247e253. Henderson, R.K., Baker, A., Murphy, K.R., Hambly, A., Stuetz, R.M., Khan, S.J., 2009. Fluorescence as a potential monitoring tool for recycled water system: a review. Water Res. 43, 863e881. Her, N., Amy, G., Plottu-Pecheux, A., Yoon, Y., 2007. Identification of nanofiltration membrane foulants. Water Res. 41, 3936e3947. Hong, S., Elimelech, M., 1997. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. J. Membr. Sci. 132 (2), 159e181. Howe, K.J., Clark, M.M., 2002. Fouling of microfiltration and ultrafiltration membranes by natural waters. Environ. Sci. Technol. 36, 3571e3576. Huang, H., Cho, H.-H., Jacangelo, J.G., Schwab, K.J., 2012. Mechanisms of membrane fouling control by integrated magnetic ion exchange and coagulation. Environ. Sci. Technol. 46, 10711e10717. Ishida, N., Inoue, T., Miyahara, M., Higashitani, K., 2000. Nano bubbles on a hydrophobic surface in water observed by tapping-mode atomic force microscopy. Langmuir 16, 6377e6380. Jaouadi, M., Amdouni, N., Duclaux, L., 2012. Characteristics of natural organic matter extracted from the waters of Medjerda dam (Tunisia). Desalination 305, 64e71. Jermann, D., Pronk, W., Boller, M., 2008. Mutual influences between natural organic matter and inorganic particles and their combined effect on ultrafiltration membrane fouling. Environ. Sci. Technol. 42, 9129e9136. Jermann, D., Pronk, W., Meylan, S., Boller, M., 2007. Interplay of different NOM fouling mechanisms during ultrafiltration for drinking water production. Water Res. 41, 1713e1722. Jung, Y.C., Bhushan, B., 2007. Wetting transition of water droplets on superhydrophobic patterned surfaces. Scr. Mater. 57, 1057e1060. Kaiya, Y., Itoh, Y., Takizawa, S., Fujita, K., Tagawa, T., 2000. Analysis of organic matter causing membrane fouling in drinking water treatment. Water Sci. Technol. 41, 59e67. Kang, S.-T., Subramani, A., Hoek, E.M.V., Deshusses, M.A., Matsumoto, M.R., 2004. Direct observation of biofouling in cross-flow microfiltration: mechanisms of deposition and release. J. Membr. Sci. 244, 151e165. Kilduff, J.E., Mattaraj, S., Belfort, G., 2004. Flux decline during nanofiltration of naturally-occurring dissolved organic matter: effects of osmotic pressure, membrane permeability, and cake formation. J. Membr. Sci. 239, 39e53. Kim, B.R., Ko, M.J., 2009. The assessment of the fracture behavior in spin-on organosilicates by nanoindentation and nanoscratch tests. Thin Solid Films 517, 3216e3221. Kim, S., Marion, M., Jeong, B.-H., Hoek, E.M.V., 2006. Crossflow membrane filtration of interacting nanoparticle suspensions. J. Membr. Sci. 284, 361e372. Lahoussine-Turcaud, V., Wiesner, M.R., Bottero, J.-Y., 1990. Fouling in tangentialflow ultrafiltration: the effect of colloid size and coagulation pretreatment. J. Membr. Sci. 52, 173e190. , G., Pernot, E., Joud, J.C., 2006. Langlet, M., Permpoon, S., Riassetto, D., Berthome Photocatalytic activity and photo-induced superhydrophilicity of sol-gel derived TiO2 films. J. Photochem. Photobiol. A-Chem. 181, 203e214. , J.-P., Buisson, H., 2004. Identification and understanding of Lee, N.H., Amy, G., Croue fouling in low-pressure membrane (MF/UF) filtration by natural organic matter (NOM). Water Res. 38, 4511e4523. Leiknes, T., Ødegaard, H., Myklebust, H., 2004. Removal of natural organic matter (NOM) in drinking water treatment by coagulationemicrofiltration using metal membranes. J. Membr. Sci. 242, 47e55. Li, J., Wang, N., Yan, H., Ji, S., Zhang, G., 2014. Designing superhydrophobic surfaces with SAM modification on hierarchical ZIF-8/polymer hybrid membranes for efficient bioalcohol pervaporation. RSC Adv. 4, 59750e59753. Li, K., Xiong, J., Chen, T., Yan, L., Dai, Y., Song, D., Lv, Y., Zeng, Z., 2013. Preparation of graphene/TiO2 composites by nonionic surfactant strategy and their simulated sunlight and visible light photocatalytic activity towards representative aqueous POPs degradation. J. Hazard. Mater. 250e251, 19e28. Liang, S., Qi, G., Xiao, K., Sun, J., Giannelis, E.P., Huang, X., Elimelech, M., 2014. Organic fouling behavior of superhydrophilic polyvinylidene fluoride (PVDF) ultrafiltration membranes functionalized with surface-tailored nanoparticles: Implications for organic fouling in membrane bioreactors. J. Membr. Sci. 463, 94e101. Listiarini, K., Tor, J.T., Sun, D.D., Leckie, J.O., 2010. Hybrid coagulationenanofiltration membrane for removal of bromate and humic acid in water. J. Membr. Sci. 365, 154e159. Lowe, J., Hossain, MdM., 2008. Application of ultrafiltration membranes for removal of humic acid from drinking water. Desalination 218, 343e354. Luo, Y., Zhang, J., Sun, A., Chu, C., Zhou, S., Guo, J., Chen, T., Xu, G., 2014. Electric field induced structural color changes of SiO2 @ TiO2 core-shell colloidal suspensions. J. Mater. Chem. C 2, 1990e1994. Madaenia, S.S., Ghaemi, N., Alizadeh, A., Joshaghani, M., 2011. Influence of photoinduced superhydrophilicity of titanium dioxide nanoparticles on the antifouling performance of ultrafiltration membranes. Appl. Surf. Sci. 257, 6175e6180. €ntta €ri, M., Puro, L., Nuortila-Jokinen, J., Nystro €m, M., 2000. Fouling effects of Ma
131
polysaccharides and humic acid in nanofiltration. J. Membr. Sci. 165, 1e17. Maximous, N., Nakhla, G., Wan, W., 2009. Comparative assessment of hydrophobic and hydrophilic membrane fouling in wastewater applications. J. Membr. Sci. 339, 93e99. Meyer, E.E., Rosenberg, K.J., Isreaelachvili, J., 2006. Recent progress in understanding hydrophobic interactions. Proc. Natl. Acad. Sci. U. S. A. 103, 15739e15746. Mills, A., Elliott, N., Parkin, I.P., O'Neill, S.A., Clark, R.J., 2002. Novel TiO2 CVD films for semiconductor photocatalysis. J. Photochem. Photobiol. A-Chem. 151, 171e179. Ng, M., Liu, S., Chow, C.W.K., Drikas, M., Amal, R., Lim, M., 2013. Understanding effects of water characteristics on natural organic matter treatability by PACI and a novel PACI-chitosan soagulants. J. Hazard. Mater. 263, 718e725. Nilson, J.A., DiGiano, F.A., 1996. Influence of NOM composition on nanofiltration. J. Am. Water Works Assoc. 88 (5), 53e66. Nosonovsky, M., Bhushan, B., 2005. Roughness optimization for biomimetic superhydrophobic surfaces. Microsyst. Technol. 11, 535e549. Nosonovsky, M., Bhushan, B., 2006. Wetting of rough three-dimensional superhydrophobic surfaces. Microsyst. Technol. 12, 273e281. Peeva, P.D., Palupi, A.E., Ulbricht, M., 2011. Ultrafiltration of humic acid solutions through unmodified and surface functionalized low-fouling polyethersulfone membranes e effects of feed properties, molecular weight cut-off and membrane chemistry on fouling behavior and cleanability. Sep. Purif. Technol. 81, 124e133. Peiris, R.H., Budman, H., Moresoli, C., Legge, R.L., 2010. Understanding fouling behavior of ultrafiltration membrane processes and natural water using principal component analysis of fluorescence excitation-emission matrices. J. Membr. Sci. 357, 62e72. Saravia, F., Zwiener, C., Frimmel, F.H., 2006. Interactions between membrane surface, dissolved organic substances and ions in submerged membrane filtration. Desalination 192, 280e287. Shan, L., Guo, H., Qin, Z., Wang, N., Ji, S., Zhang, G., Zhang, Z., 2015. Covalent crosslinked polyelectrolyte complex membrane with high negative charges towards anti-natural organic matter fouling nanofiltration. RSC Adv. 5, 11515e11523. Shao, J., Hou, J., Song, H., 2011. Comparison of huimic acid rejection and flux decline during filtration with negatively charged and uncharged ultrafiltration membranes. Water Res. 45, 473e482. Shao, J., Zhao, L., Chen, X., He, Y., 2013. Humic acid rejection and flux decline with negative charged membranes of different spacer arm lengths and charge groups. J. Membr. Sci. 435, 38e45. Sierra, M.M.D., Giovanela, M., Parlanti, E., Soriano-Sierra, E.J., 2005. . Fluorescence fingerprint of fulvic and humic acids from varied origins as viewed by singlescan and excitation/emission matrix techniques. Chemosphere 58, 715e733. Song, H., Shao, J., He, Y., Hou, J., Chao, W., 2011. Natural organic matter removal and flux decline with charged ultrafiltration and nanofiltration membranes. J. Membr. Sci. 376, 179e187. Thorsen, T., 2004. Concentration polarization by natural organic matter (NOM) in NF and UF. J. Membr. Sci. 233, 79e91. Tiraferri, A., Kang, Y., Giannelis, E.P., Elimelech, M., 2012. Superhydrophilic thin-film composite forward osmosis membranes for organic fouling control: fouling behavior and antifouling mechanisms. Environ. Sci. Technol. 46, 11135e11144. Tyrrell, J.W.G., Attard, P., 2001. Images of nanobubbles on hydrophobic surfaces and their interactions. Phys. Rev. Lett. 87, 176104. Van der Bruggen, B., Vandecasteele, C., Gestel, T.V., Doyen, W., Leysen, R., 2003. A review of pressure-driven membrane processes in wastewater treatment and drinking water production. Environ. Prog. 22, 46e56. Van Oss, C., Good, R., 1988. Orientation of the water molecules of hydration of human serum albumin. J. Protein Chem. 7 (2), 179e183. Vrouwenvelder, H.S., van Paassen, J.A.M., Folmer, H.C., Hofman, J.A.M.H., Nederlof, M.M., van der Kooij, D., 1998. Biofouling of membranes for drinking water production. Desalination 118, 157e166. Wang, K., Zhang, J., Lou, L., Yang, S., Chen, Y., 2004. UV or visible light induced photodegradation of AO7 on TiO2 particles: the influence of inorganic anions. J. Photochem. Photobiol. A-Chem. 165, 201e207. Wang, Q., Wang, Z., Zhu, C., Mei, X., Wu, Z., 2013. Assessment of SMP fouling by foulant-membrane interaction energy analysis. J. Membr. Sci. 446, 154e163. Wei, W., Xia, S., Liu, G., Gu, X., Jin, W., Xu, N., 2010. Interfacial adhesion between polymer separation layer and ceramic support for composite membrane. AIChE J. 56, 1584e1592. Xia, S., Ni, M., 2015. Preparation of poly(vinylidene fluoride) membranes with graphene oxide addition for natural organic matter removal. J. Membr. Sci. 473, 54e62. Xiao, K., Wang, X., Huang, X., Waite, T.D., Wen, X., 2011. Combined effect of membrane and foulant hydrophobicity and surface charge on adsorptive fouling during microfiltration. J. Membr. Sci. 373, 140e151. Yamamura, H., Okimoto, K., Kinura, K., Yoshimasa, W., 2014. Hydrophilic fraction of natural organic matter causing irreversible fouling of microfiltration and ultrafiltration membranes. Water Res. 54, 123e136. Yang, Q., Liu, Y., Li, Y., 2010. Humic acid fouling mitigation by antiscalant in reverse osmosis system. Environ. Sci. Technol. 44, 5153e5158. Yoshimitsu, Z., Nakajima, A., Watanabe, T., Hashimoto, K., 2002. Effect of surface structure on the hydrophobicity and sliding behavior of water droplets. Langmuir 18, 5818e5822. Zhang, L., Chowdhury, G., Feng, C., Matsuura, T., Narbaitz, R., 2003a. Effect of surface-modifying macromolecules and membrane morphology on fouling of polyethersulfone ultrafiltration membranes. J. Appl. Polym. Sci. 88, 3132e3138.
132
L. Shan et al. / Water Research 93 (2016) 121e132
Zhang, M., Li, C., Benjamin, M.M., Chang, Y., 2003b. Fouling and natural organic matter removal in adsorbent/membrane systems for drinking water treatment. Environ. Sci. Technol. 37, 1663e1669. Zhang, X., Hu, L., Sun, D., 2006. Nanoindentation and nanoscratch profiles of hybrid films based on (g-methacrylpropyl) trimethoxysilane and tetraethoxysilane. Acta Mater. 54, 5469e5475. Zhao, Y., Zhou, S., Li, M., Xue, A., Zhang, Y., Wang, J., Xing, W., 2013. Humic acid
removal and easy-cleanability using temperature responsive ZrO2 tubular membranes grafted with poly(N-isopropylacrylamide) brush chains. Water Res. 47, 2375e2386. Zhou, M., Liu, H., Kilduff, J.E., Langer, R., Anderson, D.G., 2009. Belfort G. Highthoughput membrane surface modification to control NOM fouling. Environ. Sci. Technol. 43, 3865e3871.
