Desalination 424 (2017) 131–139
Contents lists available at ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
Modification of thin film composite hollow fiber membranes for osmotic energy generation with low organic fouling tendency
MARK
Ye Lia,b,1, Saren Qib,1, Yining Wangb, Laurentia Setiawanb, Rong Wanga,b,⁎ a b
School of Civil and Environmental Engineering, Nanyang Technological University, 639798, Singapore Singapore Membrane Technology Centre, Nanyang Environment and Water Research Institute, Nanyang Technological University, 637141, Singapore
A R T I C L E I N F O
A B S T R A C T
Keywords: Pressure retarded osmosis (PRO) Power density Thin film composite membrane Anti-fouling Polyelectrolytes
The emerging technology of pressure retarded osmosis (PRO) has been intensely studied over the past decades due to its potential in harvesting salinity gradient energy, but membrane fouling can deteriorate the PRO performance. To mitigate membrane fouling, the outer surface of the inner-selective thin-film composite (TFC) polyetherimide (PEI) hollow fiber membrane was modified by depositing poly-(allylamine hydrochloride) (PAH) and poly acrylic acid (PAA) polyelectrolytes. The results show that the outer surface of the modified membrane had less pores exposed and was more negatively charged, which was beneficial for preventing the entrance and adsorption of the negatively charged foulants into the membrane substrate. It was further proved that the modified membrane was effective in reducing PRO fouling by organic macromolecules such as alginate and bovine serum albumin, while maintaining the mechanical properties and intrinsic separation properties well. Interestingly, after 200 ppm alginate was introduced to the feed water, the power density of the modified membrane could be maintained constant at 16.2 W·m− 2 at 15 bar while that of the non-modified membrane reduced around 17% at the end of a 300-min testing period. The results suggest that the polyelectrolytes deposition is a feasible strategy for adding anti-fouling property to PRO membranes.
1. Introduction Renewable and clean sources of energy have drawn increasing attention over the past few decades due to the depletion of fossil fuel energy globally. Among various options, salinity gradient energy (SGE) is proposed to be one of the sources of prospective renewable energy that could be harnessed by the pressure retarded osmosis (PRO) technology upon the mixing of fresh water and salt water [1–5]. In the PRO process, water diffuses spontaneously through a semipermeable membrane, in a direction from a stream of low salinity (feed stream) to a pressurized stream of high salinity (draw stream). The pressurized draw stream with an increased amount of volume can drive a hydro-turbine for energy production [6]. Theoretically, the estimated global storage of osmotic energy can reach about 2 terawatts if major river water outfalls and sea water are mixed, of which about 980 gigawatts can be harnessed [4,7]. Membrane is one of the critical elements for the PRO process. There are two types of PRO membrane designs distinguished by their configurations, i.e., flat sheet and hollow fiber [8–15]. The current module design for flat sheet membranes faces challenges from spacers, as the
⁎
1
opening of the spacers shall be taken into consideration for both hydrodynamics and membrane deformation under high pressure [16–18]. As the hollow fiber membrane is self-supported and no spacers are present in the module, the deformation induced by feed spacers would not be a concern for hollow fibers. In addition, hollow fiber membranes can be made into a membrane module with higher packing density, thereby generating more energy based on the same sized membrane module than the flat sheet counterpart [1,19,20]. PRO membranes are usually run in the active layer facing draw solution (AL-DS) orientation. As compared with its counterpart, the active layer facing feed solution (AL-FS) orientation, the AL-DS orientation generally suffers from less internal concentration polarization (ICP) effect, resulting in a higher water flux and thus higher power density [21–23]. Nevertheless, membrane fouling is a major concern for the operation in the orientation of the AL-DS in real applications. When feed water contains organic matters, scaling precursors and/or other potential foulants, the membrane water flux may significantly decrease during the PRO operation, leading to the reduction in power density [24–26]. The foulants with relatively small size (compared to the pores on the membrane bottom surface) in the feed water could enter into the
Corresponding author at: School of Civil and Environmental Engineering, Nanyang Technological University, 639798, Singapore. E-mail address:
[email protected] (R. Wang). The authors contributed equally.
http://dx.doi.org/10.1016/j.desal.2017.10.005 Received 23 August 2017; Received in revised form 2 October 2017; Accepted 2 October 2017 0011-9164/ © 2017 Elsevier B.V. All rights reserved.
Desalination 424 (2017) 131–139
Y. Li et al.
hydrochloride) (PAH; Molecular weight (Mw) ~120,000, SigmaAldrich) and poly acrylic acid (PAA; Mw ~250,000, Sigma-Aldrich) were adopted as modification chemicals. Dextran ((C6H10O5)n, Mw 6000 to 500,000 Dalton (Da), Sigma-Aldrich) was used in measurement of molecular weight cut-off (MWCO) of the substrates. Sodium chloride (NaCl; Merck) was adopted in the draw stream preparations. Bovine serum albumin (BSA; Sigma-Aldrich, Mw ~66,000 Da) and sodium alginate (Sigma-Aldrich) were utilized as the model protein foulant and model polysaccharide foulant, respectively. A Milli-Q integral water purification system was utilized to supply the deionized (DI) water.
permeable membrane substrate, and their adsorption and deposition within the support layer cause a significant rise in the structural parameter (i.e., reduce porosity and increase tortuosity), and a reduction of water permeability [24–26]. Moreover, the membrane may suffer from structure changes under high hydraulic pressure (even though the changes are reversible for the mechanically stable membranes) [11,15,27], which could result in increased solute back diffusion from draw solution, thereby deteriorating the membrane further and reducing the power density. Different approaches have been proposed to control and alleviate the adverse effect of fouling in PRO process, such as pretreatment of feed water [28,29], chemical cleaning of fouled membrane [30], and membrane surface modification with added anti-fouling properties [31,32]. Among them, surface modification has the advantage of relatively low cost and easier scale-up [4]. Recently, grafting or coating with hydrophilic chemicals have been adopted to modify the PRO hollow fiber membrane for anti-fouling purposes, such as hyper-branched polyglycerol (HPG), zwitterionic copolymers, poly (vinyl alcohol) (PVA), 3-aminopropyltrimethoxysilane (APTMS) [31–37]. However, the grafting methods also faced challenges from the increased salt leakage [35], non-environmentally-friendly substances produced [37], and complicated chemical synthesis steps that may not be suitable for large-scale production [31,32,36]. Polyelectrolyte deposition has been adopted to develop a selective layer for nanofiltration and reverse osmosis membranes [38,39] and to coat membrane surface for controlling organic and bio-fouling [40,41]. In the polyelectrolyte deposition process, cationic and anionic polyelectrolytes are deposited on the membrane surface in an alternate sequence via electrostatic adsorption and form a nano-film on top of the substrate [42,43]. This method has the advantage of simple preparation procedures, environmentally friendly process and flexibility in changing surface properties such as wettability and zeta potential [44,45]. However, limited studies have been conducted to modify PRO membranes using the polyelectrolyte deposition method. Unlike pressuredriven membranes and forward osmosis membranes, where polyelectrolytes are usually deposited on the substrate to serve as the active skin layer itself or coating the active layer to control fouling, the polyelectrolytes coated on the bottom surface of a PRO membrane will play a different role. It is expected that this polyelectrolyte layer on the membrane bottom surface has potential to mitigate fouling during PRO operation through reduced pore blocking. In addition, well controlled thickness/density of polyelectrolyte layer on the bottom surface may induce minimal adverse effect on the intrinsic properties and mechanical strength, which is important for maintaining the high performance of the PRO membrane. In this study, the anti-fouling property of a PRO membrane was improved via polyelectrolytes deposition with two oppositely charged polyelectrolytes - poly-(allylamine hydrochloride) and poly acrylic acid. Specifically, the shell surface of a hollow fiber thin film composite (TFC) PRO membrane was post-treated using polyelectrolytes, which could potentially increase the charge density and hydrophilicity of the membrane outer surface for better retaining the organic foulants in feed water. The resultant membranes were characterized, and the antifouling behavior was evaluated with model foulants.
2.2. Fabrication of hollow fiber thin film composite membranes The PEI support layer was fabricated using dry-wet spinning method, and the detailed method can be referred to our previous work [15,46]. Based on the dimension measurement using microscope (Keyence VHX 500F Digital Microscope), the average wall thickness of the substrates was ~106 μm with inner and outer diameters of 398 μm and 611 μm, respectively. Prior to the formation of selective layer, fifteen fibers were placed in a module of 20 cm effective length and ½ inch diameter and were sealed with epoxy. A polyamide solute rejection skin was manufactured on the inner surface of the fiber by conducting interfacial polymerization with two monomers: MPD and TMC [46,47]. Briefly, 1.0 wt% MPD solution was in contact with the lumen side of the substrate for 30 min to thoroughly wet the surface. Subsequently, nhexane was used to rinse the lumen side to remove the excess MPD. After that, 0.1% (wt/v) TMC in n-hexane was adopted to react with the remaining MPD to synthesize an ultrathin skin. The unreacted chemical residue inside the membrane module was rinsed thoroughly with pure water. 2.3. Modification by polyelectrolytes deposition The shell surface modification was conducted by exposing the shell surface of the as-prepared TFC membrane or hollow fiber substrate to the chemicals. The membrane was firstly soaked in a 0.04 g·L− 1 NaOH solution for 60 min in order to introduce negative charged group on the membrane substrate [48]. Subsequently followed by soaking in pure water for 5 min to rinse the remaining alkaline solution off membrane surface. Then, the cationic PAH solution (1 g·L− 1) was adopted to soak the membrane for 60 min and subsequently DI water was to rinse off the remaining chemicals for 5 min. The same procedure was repeated for anionic PAA solution (1 g·L− 1). After the modification with the polyelectrolytes, the membrane was stored in a tank filled with pure water prior to use. The membrane substrates before and after polyelectrolytes deposition were named S-PEI and S-PEI-M, while the TFC membranes before and after the modification were named TFC-PEI and TFC-PEI-M, respectively. 2.4. Characterization of hollow fiber substrates The morphology of the substrates was characterized via a field emission scanning electron microscopy (FESEM, Joel JSM 7600F). The membrane cross section was prepared by fracturing after the sample was frozen in liquid nitrogen. The surface of membrane was coated with a thin layer of platinum (~ 5 nm) prior to FESEM imaging. The pure water permeability (PWP) and MWCO of the substrates were evaluated with a reverse osmosis (RO) cross-flow system at pressure of 1 bar. In the PWP tests, the permeate was collected at the shell/lumen side with pure water pressurized on the opposite surface. The PWP of the membrane substrate was calculated using Eq. (1):
2. Materials and methods 2.1. Chemicals and materials Commercially available polymer, polyetherimide (PEI, Ultem 1000, General Electric Plastics) was utilized as polymeric material and NMethyl-2-pyrrolidone (NMP; > 99.5% Merck) was adopted as the solvent in fabrication of the hollow fiber support layer. Trimesoyl chloride (TMC; > 99% Sigma-Aldrich) and m-Phenylenediamine (MPD; ≥ 99% Sigma-Aldrich) were employed for the formation of solute rejection layer. Sodium hydroxide (NaOH; Merck), poly-(allylamine
PWP =
ΔV Am ΔtΔP
(1)
where Am is the effective area of the lumen surface, ΔV is the volume of the permeate generated during the time period of Δt, and ΔP is the 132
Desalination 424 (2017) 131–139
Y. Li et al.
(a)
Fig. 1. FESEM images of membrane substrate S-PEI showing cross-section (a–b), lumen surface (c), outer surface (d), outer surface at higher magnification (e), and outer surface after polyelectrolytes modification (membrane substrate (S-PEI-M)) (f).
(b)
10 µm
100 µm
(d)
(c)
1 µm
1 µm
(e)
(f)
100 nm
100 nm
trans-membrane pressure. For the measurement of MWCO, a 2 g·L− 1 dextran solution (Mw ranging from 6000 to 500,000 Da) was prepared as feed and circulated through the lumen surface. Both permeate and feed were collected and examined using a Laboratories-GPC 50 plus system [15,49]. The zeta potential (surface charge) of the substrate outer surface was determined using a SurPASS electrokinetic analyzer (Anton Paar GmbH, Austria) with a 0.01 M sodium chloride electrolyte solution. The chemistry of the shell surface of the substrate was analyzed using attenuated total reflection-Fourier transform infrared spectroscopy (ATRFTIR). ATR-FTIR spectra were obtained over a range of 650–4000 cm− 1 with the resolution of 4.0 cm− 1. The spectra analysis was conducted with FTIR software (IR Solution, Shimadzu). The dynamic contact angle was obtained based on the Wilhelmy method using a tensiometer (DCAT11 Dataphysics). The average value was obtained by repeating three runs with different fibers. The tensile modulus of the support was determined via a Zwick 0.5 kN universal testing machine. Samples were stretched by tensile force at a constant rate of 50 mm·min− 1.
A=
J ΔP
(2)
where J is the permeate flux and ΔP is the trans-membrane pressure. The NaCl rejection Rs was evaluated at 2 bar with a NaCl solution (10 mM) as the feed stream. The B value was determined from Eq. (3).
1 B=⎛ − 1⎞⋅J ⎝ Rs ⎠ ⎜
⎟
(3)
The structural parameter (S) was calculated with the aid of forward osmosis (FO) experiments using Eq. (4) [17,46].
( ) ( )⎞ ⎟ ( ) ⎟⎠ J
S=
⎛ πd − Aw + D ln ⎜ B Jw ⎜ πf + A ⎝
B A
(4)
where πd and πf are the osmotic pressures in draw and feed streams, correspondingly. D refers to the diffusivity of sodium chloride in water (1.61 × 10− 9 m2·s− 1 [17]). Jw is the water flux during FO tests. FO tests were performed using 1.0 M sodium chloride draw stream and pure water as the feed in the orientation of AL-DS. The crossflow rates in the feed and draw channels were 1 L·min− 1 and 0.15 L·min− 1, correspondingly.
2.5. Characterization of TFC membranes The membrane intrinsic properties in terms of the water permeability (A), salt rejection (Rs) and salt permeability coefficients (B) were determined using a reverse osmosis cross-flow system. Deionized water was pumped through the inner surface of the TFC membranes with applied pressure of 2 bar. The A value was calculated using Eq. (2).
2.6. PRO experiments The PRO testing system employed in this study could be referred to that reported in our previous research work [15,46]. In brief, the membrane was orientated with the selective skin facing draw stream for all the PRO tests. The effective area of a single membrane module was 133
Desalination 424 (2017) 131–139
Y. Li et al.
Js =
%Transmittance
47 cm2. The hydraulic pressure ranged from 2 to 20 bar was applied to the draw channel. The crossflow rates are the same as the FO tests (1 L·min− 1 in the shell side and 0.15 L·min− 1 in the lumen side). All the runs were conducted at 23 ± 1 °C. During the test, the feed water volume (10 L) was maintained by dosing DI water. A saturated stock solution was employed to dose the draw stream to keep the concentration constant. The water flux Jw was evaluated based on the change of the weight of the dosing water. The feed conductivity was recorded for the determination of solute flux Js using the following equation:
Δ (Cfe Vf ) Am Δt
(5)
S-PEI S-PEI-M
where Cfe is the salt concentration and Vf is the volume of the feed solution. The power density W can be determined by multiplying the trans-membrane pressure (ΔP) and the water flux (Jw) in Eq. (6):
W = ΔPJw
2800 3000 3200 3400 3600
1000
(6)
2000
3000
4000
5000
6000
7000
Wavelength (cm-1)
For measuring the power density of a fresh membrane, 1 M NaCl draw stream and DI feed water were applied. The flux calculation was usually based on ~ 30 min flux performance. PRO fouling experiments were conducted with foulants added to the feed water, i.e., 200 ppm sodium alginate or 200 ppm BSA, and each run last for ~300 min. The baseline test was also performed using pure water as feed. The draw was NaCl stream (1 M) for all the experiments.
Fig. 2. ATR-FTIR spectra for the S-PEI and the S-PEI-M. The analysis was conducted from the outer surface of the hollow fiber substrates.
40 S-PEI S-PEI-M
20 Zeta potential (mV)
3. Results and discussions 3.1. Characteristics of the hollow fiber substrates The morphology of the PEI supports is present in Fig. 1. The spongelike dense structure of the membrane cross-section can be observed in Fig. 1(a–b). The dense structure is believed to provide robust mechanical property and minimize the deformation under high pressure [3,50,51]. The pores on the inner surface could not be observed (Fig. 1(c)), likely due to the much denser structure close to/at the inner surface. In contrast, micropores are visible and evenly spread over the outer surface (Fig. 1(d)), with the surface pore size of about tens of nanometers as indicated in the magnified image (Fig. 1(e)). Interestingly, after polyelectrolyte modification, most of the pores seemed to be covered by a dense layer, although some pores are still exposed at a much smaller amount (Fig. 1(f)), indicating the successful deposition of the polyelectrolytes. The ATR-FTIR results for the pristine (S-PEI) and modified (S-PEIM) substrates are shown in Fig. 2. The typical imide bands are at 1778 cm− 1 and 1719 cm− 1, corresponding to the stretching of CeO of the pristine PEI substrates. The band at 1353 cm− 1 referred to the CeNeC stretching in phthalimide rings. After the modification, amine bands were detected at around 3380 cm− 1 and 2919 cm− 1, assigned to the eNH2 and eNH3+ groups, correspondingly. These amine bands confirmed the effective deposition of PAH on the PEI substrate. The zeta potential results of the two substrates (outer surface) at pH range of 3.5–9 are presented in Fig. 3. The zeta potential of the S-PEI-M surface was observed to be more negatively charged at pH > 4. The zeta potential at pH 6 was about − 30 mV and − 60 mV for the S-PEI and SPEI-M, respectively. The more negative charges on the S-PEI-M surface could be attributed to the negatively charged PAA deposited on the surface [52]. Again, the results reveal the successful modification of the membrane outer surface by the polyelectrolytes. The other properties of the substrates are summarized in Table 1. Tensile modulus can indicate the membrane pressure endurance in the PRO experiments. The tensile modulus of the substrate S-PEI was as high as 252 MPa, which is consistent with that of the PEI substrates reported previously [15,46]. The surface modification almost did not affect the mechanical property, as the substrate S-PEI-M had a tensile modulus of ~246 MPa. The surface contact angle after the modification
0 -20 -40 -60 -80
3
4
5
6
7
8
9
pH Fig. 3. Zeta potential of the outer surfaces of the S-PEI and S-PEI-M hollow fiber membrane substrates (0.01 M NaCl was employed as the electrolyte solution).
was reduced to ~58° from 75°, which can be explained by the hydrophilic property of the deposited polyelectrolytes. According to literature [24,53,54], the more hydrophilic surface is favorable for reducing membrane fouling. The pure water permeability (PWP) of S-PEI and SPEI-M was tested using a reverse osmosis (RO) cross-flow setup. The PWP-I was obtained with deionized water pressurized on the inner side at the hydraulic pressure of 1 bar and the permeate was obtained from the outer surface. The PWP-I of the S-PEI-M substrate was approximately 191 L·m− 2·h− 1·bar− 1, which was slightly lower than that of the pristine S-PEI substrate (201 L·m− 2·h− 1·bar− 1). This trend is consistent with the MWCO results. The S-PEI-M had smaller MWCO of 147 kDa, as compared to that of S-PEI substrate (168 kDa). The decrease in PWP and MWCO were also consistent with the FESEM images (Fig. 1(e–f)) where the narrowed pore size/decreased surface porosity was observed as a result of polyelectrolytes deposition. Another water permeability test was conducted with pressure applied to the outer surface of the membrane. The results show that the PWP-O of S-PEI-M was slightly higher than that of the S-PEI, when the permeate flowed from the shell to the lumen side of the membrane. This interesting result may suggest that the increased hydrophilicity could indeed enhance the water flux in spite of the narrowed pore size/decreased surface porosity, thereby
134
Desalination 424 (2017) 131–139
Y. Li et al.
Table 1 Properties of hollow fiber membrane substrates. Membrane
Tensile modulus (MPa)
Contact angle (°)a
Zeta potential @ pH 6a
PWP-Ib
PWP-Oc
MWCOb (kDa)
Structural parameter, S (μm)
S-PEI S-PEI-M
252 ± 17 246 ± 6
74.7 ± 4.3 58.2 ± 3.9
− 30 − 60
201 ± 16 191 ± 19
196 ± 16 202 ± 8
168 ± 48 147 ± 37
750 ± 16 620 ± 23
a b c
The membranes were measured at the outer surface. The membranes were pressurized from inner surface. The membranes were pressurized from outer surface.
decreasing the S value of the substrate, which was measured in the same flow orientation as the flow went from the shell to the lumen side. The structural parameter (S value) of the S-PEI-M substrate (~ 620 μm) was smaller than that of the pristine S-PEI substrate (~ 750 μm). In contrast, the conventional measurement with pressure applied at the lumen side could not effectively reflect the benefit of increased hydrophilicity at the outer membrane surface. Similar results have also been observed in prior studies [55,56] that the S value decreased despite the decreased PWP and MWCO of the substrates after the hydrophilic modification.
Table 2 Intrinsic separation properties of TFC membranes. Membrane
Water permeability, A (L·m− 2·h− 1·bar− 1)
Salt permeability, B (L·m− 2·h− 1)
Rejection (%)
TFC-PEI TFC-PEI-M
1.96 ± 0.16 2.00 ± 0.28
0.13 ± 0.02 0.10 ± 0.03
96.8 ± 0.2 97.5 ± 1.2
obvious at lower hydraulic pressure, as the corresponding higher water fluxes induced more significant internal concentration polarization effect, thereby the more important role of S value at higher flux. In the aspect of specific solute flux Js/Jw results (Fig. 5(b)), both membranes showed similar performance and trend. Only minimal increase in Js/Jw can be observed with rising pressure when the applied pressure was equal to or < 15 bar (i.e., 0.05 mol·L− 1 at 2 bar to 0.15 mol·L− 1 at 15 bar). This phenomenon has also been observed in other studies [11,17,46], which could be due to the combined effect of slightly increased Js as a result of reversible deformation and the decreased Jw at higher pressure [17,46,47]. The Js/Jw increased markedly to above 0.3 mol·L− 1 at pressure of 20 bar, which was consistent with the increased water flux at this pressure (Fig. 5(a)), indicating the irreversible deformation of the selective layer. Fig. 5(c) indicates that the power density increased with increasing pressure. The TFC-PEI-M membrane exhibited slightly higher power density than that of the control membrane as a result of the higher water flux. By considering the unwanted permanent deformation of membrane at 20 bar, the maximum power density attained by the TFC-PEI-M membrane was ~16.2 W·m− 2 at 15 bar.
3.2. Characteristics of TFC membranes Fig. 4 presents the FESEM images of the solute rejection skin of the TFC-PEI PRO membrane. A typical ridge and valley pattern of a polyamide layer was presented, indicating the effective development of the solute rejection skin on the lumen surface of the PEI substrates [15,27,57]. The membrane intrinsic properties are listed in Table 2. The pristine membrane TFC-PEI and modified membrane TFC-PEI-M had similar water permeability (~ 2 L·m− 2·h− 1·bar− 1) and salt permeability (~ 0.1 L·m− 2·h− 1), suggesting that the outer surface modification with polyelectrolytes had minimal impact on the intrinsic properties of the selective layer, which is beneficial for the modified PRO membranes to achieve a high water flux. 3.3. PRO performance of the TFC membranes The performance of the TFC membranes TFC-PEI and TFC-PEI-M at different applied pressures (2–20 bar) are present in Fig. 5. Overall, the water flux decreased with the rise of hydraulic pressure (2–15) bar for both membranes (Fig. 5(a)), due to the increased retarded effect by hydraulic pressure. However, increased water fluxes were observed when the applied pressure was further raised from 15 to 20 bar for both membranes. It is generally expected that there is a critical pressure for a typical polymeric hollow fiber membrane, beyond which the membrane suffers from the irreversible deformation, accompanied by the increased water permeability and solute permeability [15,27]. In our case, the critical pressure for both membranes lied at around 15 bar. Compared with the control membrane TFC-PEI, the TFC-PEI-M exhibited slightly higher water flux, which can be elucidated by its smaller structural parameter (Table 1). This difference becomes more
3.4. PRO fouling test 3.4.1. Effect of foulant type on PRO performance The effect of organic foulants contained in the feed stream on the membrane performance is presented in Fig. 6. It is observed that the baseline fluxes (no foulant addition) for both membranes TFC-PEI and TFC-PEI-M are similar and stable during the 300-min test. When foulant alginate was added to the feed water (Fig. 6(a)), an obvious flux decline of ~17% occurred for TFC-PEI membrane, and the power density dropped to 13.5 W·m− 2 from 16.2 W·m− 2. In contrast, almost no flux drop occurred for TFC-PEI-M membrane, which proved the anti-fouling Fig. 4. FESEM images of TFC-PEI membrane showing the plan view of polyamide surface (a) and the cross-section view (b).
135
Desalination 424 (2017) 131–139
Y. Li et al.
Fig. 6. Effect of foulants on PRO membrane performance. 200 ppm alginate (a) or 200 ppm BSA (b) was present in the feed solution. Draw solution contained 1 M NaCl. The applied pressure was maintained at 15 bar.
effect of the modified membrane in reducing alginate fouling during PRO. Similarly, this effect was also demonstrated from the BSA fouling results (Fig. 6(b)). Despite flux decline for both membranes, the modified membrane TFC-PEI-M suffered less severe flux drop (16%) than TFC-PEI (41%). The corresponding power density at the end of 300-min test was 13.8 W·m− 2 and 9.7 W·m− 2 for membrane TFC-PEI-M and TFC-PEI, respectively. Therefore, the TFC-PEI-M with deposited polyelectrolytes showed better PRO performance for the feed water with the presence of alginate or BSA. The enhanced anti-fouling effect could be attributed to the more negatively charged and more hydrophilic modified back surface in addition to the narrowed pore size (as discussed in Section 3.1), leading to fewer foulants entering into/adsorbed to the support layer. It is worth pointing out that the decline of the water flux triggered by BSA was more severe than that of the alginate. One of the reasons could be due to the more negatively charged alginate at neutral pH range. According to prior studies [58,59], the zeta potential of alginate and BSA at pH 6 is −66 mV and −13 mV, respectively. Since the membrane outer surface was negatively charged (Fig. 3), the electrostatic repulsion between the outer surface of the membrane and the more negatively charged alginate was believed to be stronger. In addition, alginate is more hydrophilic [60] and has larger hydrodynamic diameter than BSA. The hydrodynamic diameter of alginate was 343.6 ± 33.3 nm and that of BSA was 10.0 ± 0.3 nm, measured by using Nano Zetasizer (NanoZS, Malvern, UK). The trend was consistent with previously reported study [61]. These factors may lead to the less
Fig. 5. PRO performance of the membrane TFC-PEI and TFC-PEI-M. Water flux (a), specific solute flux (b), and power density (c). Testing conditions: Pure water and NaCl (1 M) were utilized as feed and draw, correspondingly. The performance was obtained after the membranes were stabilized for 30 min at each pressure. Each error bar was calculated based on three replicates.
136
Desalination 424 (2017) 131–139
Y. Li et al.
Fig. 7. Water flux of TFC-PEI and TFC-PEI-M membranes under applied pressure of 2 bar (a), 10 bar (b) and 15 bar (c), and their power density at 0 and 300 min of alginate fouling (d). The PRO experiments were conducted using NaCl draw stream (1.0 M) and DI water as the feed for baseline test while 200 ppm alginate was present in the feed for fouling test.
Table 3 Comparison of the membranes in current work and other low-fouling PRO membranes in literature. Membrane
Foulants in FS
NaCl DS (M)
Initial W (W·m− 2)
Applied pressure (bar)
Decrease in W
Ref.
TFC-PEI-M
200 ppm alginate
1
16.2
15
Current work
TFC-PEI-M
200 ppm BSA
1
16.2
15
TFC-PEI-APTMS
1
7a
9
TFC-PES
3 mM CaCl2 200 ppm humic acid 200 ppm alginate
1
11.2
18
HPG-PES
200 ppm BSA
0.6
6.7
12.5
HTI-CTA
1 mM CaCl2 200 ppm alginate
1.2
6.2
15
~ 0% @ 300 min ~ 16% @ 300 min 16% @ 60 min 55.1% @ 3000 min 10% @ ~ 420 min 40% @ 300 min
a
Current work [35] [65] [31] [66]
Read from the figures.
M membrane after 300 min fouling tests was about 10%, 7% and 0% at the pressure of 2, 10 and 15 bar. The results are consistent with prior studies [24,62–64] that lower initial flux induced less membrane fouling. The lower initial flux generally induces less hydrodynamic driving force on the foulant molecules and less concentration polarization of foulants on/within membrane support layer, leading to less foulant adsorption and more sustainable water flux. Similarly, the pristine TFC-PEI membrane also suffered slightly less flux drop with lower initial water flux operation. However, the overall flux drop was more significant compared with that of TFC-PEI-M. The corresponding power density results are presented in Fig. 7(d). Consistently, the modified membrane TFC-PEI-M showed higher power density than pristine membrane TFC-PEI after 300-min alginate fouling over the
adsorption of alginate on the membrane surface and/or in the porous substrate. 3.4.2. The effect of applied pressure on PRO performance The membrane performance under different applied hydraulic pressures (2, 10 and 15 bar) was also studied with feed water containing 200 ppm alginate. The results of water flux are shown in Fig. 7(a–c). As expected, increasing applied pressure led to the decline in water flux. The initial flux (at 0 min) for TFC-PEI-M membrane at 2 bar, 10 bar and 15 bar was 52, 44 and 38 L·m− 2·h− 1, respectively. A similar tendency was observed for TFC-PEI membrane as well. The presence of the alginate did not lead to flux drop at the beginning of the test, as compared with the baseline fluxes. The flux decline of TFC-PEI137
Desalination 424 (2017) 131–139
Y. Li et al.
[7] G.Z. Ramon, B.J. Feinberg, E.M.V. Hoek, Membrane-based production of salinitygradient power, Energy Environ. Sci. 4 (2011) 4423–4434. [8] X. Song, Z. Liu, D.D. Sun, Energy recovery from concentrated seawater brine by thin-film nanofiber composite pressure retarded osmosis membranes with high power density, Energy Environ. Sci. 6 (2013) 1199–1210. [9] G. Han, T.S. Chung, High performance thin-film composite pressure retarded osmosis (PRO) hollow fiber membranes for renewable osmotic energy generation, 24th Annual Meeting of the North American Membrane Society, NAMS 2014, 2014, pp. 148–149. [10] Y. Cui, X.Y. Liu, T.S. Chung, Enhanced osmotic energy generation from salinity gradients by modifying thin film composite membranes, Chem. Eng. J. 242 (2014) 195–203. [11] G. Han, T.S. Chung, Robust and high performance pressure retarded osmosis hollow fiber membranes for osmotic power generation, AICHE J. 60 (2014) 1107–1119. [12] R.A. Rica, R. Ziano, D. Salerno, F. Mantegazza, R. van Roij, D. Brogioli, Capacitive mixing for harvesting the free energy of solutions at different concentrations, Entropy 15 (2014) 1388–1407. [13] E. Curcio, G.D. Profio, E. Fontananova, E. Drioli, Membrane technologies for seawater desalination and brackish water treatment, Advances in Membrane Technologies for Water Treatment: Materials, Processes and Applications, 2015, pp. 411–441. [14] M. Tian, R. Wang, K. Goh, Y. Liao, A.G. Fane, Synthesis and characterization of high-performance novel thin film nanocomposite PRO membranes with tiered nanofiber support reinforced by functionalized carbon nanotubes, J. Membr. Sci. 486 (2015) 151–160. [15] Y. Chen, L. Setiawan, S. Chou, X. Hu, R. Wang, Identification of safe and stable operation conditions for pressure retarded osmosis with high performance hollow fiber membrane, J. Membr. Sci. 503 (2016) 90–100. [16] Y. Xu, X. Peng, C.Y. Tang, Q.S. Fu, S. Nie, Effect of draw solution concentration and operating conditions on forward osmosis and pressure retarded osmosis performance in a spiral wound module, J. Membr. Sci. 348 (2010) 298–309. [17] Q. She, X. Jin, C.Y. Tang, Osmotic power production from salinity gradient resource by pressure retarded osmosis: effects of operating conditions and reverse solute diffusion, J. Membr. Sci. 401–402 (2012) 262–273. [18] Y.C. Kim, M. Elimelech, Adverse impact of feed channel spacers on the performance of pressure retarded osmosis, Environ. Sci. Technol. 46 (2012) 4673–4681. [19] K. Nijmeijer, S. Metz, Chapter 5 salinity gradient energy, Sustainability Science and Engineering, 2010, pp. 95–139. [20] S.E. Skilhagen, Osmotic power - a new, renewable energy source, Desalin. Water Treat. 15 (2010) 271–278. [21] Y.N. Wang, F. Wicaksana, C.Y. Tang, A.G. Fane, Direct microscopic observation of forward osmosis membrane fouling, Environ. Sci. Technol. 44 (2010) 7102–7109. [22] C.Y. Tang, Q. She, W.C.L. Lay, R. Wang, A.G. Fane, Coupled effects of internal concentration polarization and fouling on flux behavior of forward osmosis membranes during humic acid filtration, J. Membr. Sci. 354 (2010) 123–133. [23] G.T. Gray, J.R. McCutcheon, M. Elimelech, Internal concentration polarization in forward osmosis: role of membrane orientation, Desalination 197 (2006) 1–8. [24] Q. She, Y.K.W. Wong, S. Zhao, C.Y. Tang, Organic fouling in pressure retarded osmosis: experiments, mechanisms and implications, J. Membr. Sci. 428 (2013) 181–189. [25] W.R. Thelin, E. Sivertsen, T. Holt, G. Brekke, Natural organic matter fouling in pressure retarded osmosis, J. Membr. Sci. 438 (2013) 46–56. [26] N.Y. Yip, M. Elimelech, Influence of natural organic matter fouling and osmotic backwash on pressure retarded osmosis energy production from natural salinity gradients, Environ. Sci. Technol. 47 (2013) 12607–12616. [27] G. Han, P. Wang, T.S. Chung, Highly robust thin-film composite pressure retarded osmosis (PRO) hollow fiber membranes with high power densities for renewable salinity-gradient energy generation, Environ. Sci. Technol. 47 (2013) 8070–8077. [28] S.C. Chen, G.L. Amy, T.S. Chung, Membrane fouling and anti-fouling strategies using RO retentate from a municipal water recycling plant as the feed for osmotic power generation, Water Res. 88 (2016) 144–155. [29] C.F. Wan, T.S. Chung, Osmotic power generation by pressure retarded osmosis using seawater brine as the draw solution and wastewater retentate as the feed, J. Membr. Sci. 479 (2015) 148–158. [30] X. Li, T. Cai, G.L. Amy, T.S. Chung, Cleaning strategies and membrane flux recovery on anti-fouling membranes for pressure retarded osmosis, J. Membr. Sci. 522 (2017) 116–123. [31] X. Li, T. Cai, T.S. Chung, Anti-fouling behavior of hyperbranched polyglycerolgrafted poly(ether sulfone) hollow fiber membranes for osmotic power generation, Environ. Sci. Technol. 48 (2014) 9898–9907. [32] T. Cai, X. Li, C. Wan, T.S. Chung, Zwitterionic polymers grafted poly(ether sulfone) hollow fiber membranes and their antifouling behaviors for osmotic power generation, J. Membr. Sci. 497 (2016) 142–152. [33] N. Lieu Le, M. Quilitzsch, H. Cheng, P.Y. Hong, M. Ulbricht, S.P. Nunes, T.S. Chung, Hollow fiber membrane lumen modified by polyzwitterionic grafting, J. Membr. Sci. 522 (2017) 1–11. [34] S. Zhang, Y. Zhang, T.S. Chung, Facile preparation of antifouling hollow fiber membranes for sustainable osmotic power generation, ACS Sustain. Chem. Eng. 4 (2016) 1154–1160. [35] L. Zhang, Q. She, R. Wang, S. Wongchitphimon, Y. Chen, A.G. Fane, Unique roles of aminosilane in developing anti-fouling thin film composite (TFC) membranes for pressure retarded osmosis (PRO), Desalination 389 (2016) 119–128. [36] X. Li, T. Cai, C. Chen, T.S. Chung, Negatively charged hyperbranched polyglycerol grafted membranes for osmotic power generation from municipal wastewater, Water Res. 89 (2016) 50–58. [37] D. Zhao, G. Qiu, X. Li, C. Wan, K. Lu, T.S. Chung, Zwitterions coated hollow fiber
tested pressures. The TFC-PEI-M still gained much higher power density at higher applied pressure despite the presence of foulants in the feed water (16.2 W·m− 2 at 15 bar vs. 11.6 W·m− 2 at 10 bar), owing to the sustainable water flux performance. 3.5. Comparison with low-fouling PRO membranes in literature In this study, the power density of the fabricated membranes is compared with the other documented PRO membranes as listed in Table 3. In general, the TFC-PEI-M membrane in current work exhibited relatively high and stable power density compared with the membranes reported in recent research work using the feed containing model organic foulants. The advantage of TFC-PEI-M is particularly obvious when the alginate was used as foulant, as almost no flux decline was observed and the power density could be maintained at 16.2 W·m− 2 during the 300-min run. In addition, the membrane modification method employed in this study had negligible effects on the intrinsic properties of the membranes (i.e., not sacrificing the water permeability), and is simple with high scalability compared with the grafting methods for making low-fouling membranes [31,32,36]. 4. Conclusions The membrane modification via polyelectrolytes deposition has been demonstrated as an effective method for imparting anti-fouling property to PRO membranes. This method has advantages of facile fabrication, environmental friendly, and ability to maintain membrane performance. The detailed conclusions are as follows: 1) The modified membrane after polyelectrolytes deposition showed even slightly better PRO performance than that of the control membrane, attributing to the well-maintained intrinsic properties, and the reduced structural parameter as a result of increased hydrophilicity. 2) The modified membrane could maintain its performance in the PRO process for the foulant containing feed streams, with better sustainable flux and higher power density achieved at increased pressure. 3) The in-house fabricated PRO membranes without and with surface modification had better fouling resistance toward alginate compared with BSA, presumably owing to the higher charge density, higher hydrophilicity and larger hydrodynamic diameter of the alginate. Acknowledgements This research work was supported by the Singapore National Research Foundation under its Environmental and Water Technologies Strategic Research Program (EWI) (Project number: 1102-IRIS-07-01). We are also grateful for the funding support from Singapore Economic Development Board to the Singapore Membrane Technology Centre. References [1] K. Gerstandt, K.V. Peinemann, S.E. Skilhagen, T. Thorsen, T. Holt, Membrane processes in energy supply for an osmotic power plant, Desalination 224 (2008) 64–70. [2] S.E. Skilhagen, J.E. Dugstad, R.J. Aaberg, Osmotic power - power production based on the osmotic pressure difference between waters with varying salt gradients, Desalination 220 (2008) 476–482. [3] T. Thorsen, T. Holt, The potential for power production from salinity gradients by pressure retarded osmosis, J. Membr. Sci. 335 (2009) 103–110. [4] B.E. Logan, M. Elimelech, Membrane-based processes for sustainable power generation using water, Nature 488 (2012) 313–319. [5] A. Achilli, A.E. Childress, Pressure retarded osmosis: from the vision of Sidney Loeb to the first prototype installation - review, Desalination 261 (2010) 205–211. [6] S. Loeb, F. Van Hessen, D. Shahaf, Production of energy from concentrated brines by pressure retarded osmosis. II. Experimental results and projected energy costs, J. Membr. Sci. 1 (1976) 249–269.
138
Desalination 424 (2017) 131–139
Y. Li et al.
[38]
[39]
[40]
[41]
[42] [43]
[44]
[45] [46]
[47]
[48] [49]
[50]
[51]
[52]
separator, RSC Adv. 4 (2014) 46940–46946. [53] G. Zuo, R. Wang, Novel membrane surface modification to enhance anti-oil fouling property for membrane distillation application, J. Membr. Sci. 447 (2013) 26–35. [54] A. Tiraferri, Y. Kang, E.P. Giannelis, M. Elimelech, Superhydrophilic thin-film composite forward osmosis membranes for organic fouling control: fouling behavior and antifouling mechanisms, Environ. Sci. Technol. 46 (2012) 11135–11144. [55] G. Han, S. Zhang, X. Li, N. Widjojo, T.S. Chung, Thin film composite forward osmosis membranes based on polydopamine modified polysulfone substrates with enhancements in both water flux and salt rejection, Chem. Eng. Sci. 80 (2012) 219–231. [56] X. Liu, S.L. Ong, H.Y. Ng, Fabrication of mesh-embedded double-skinned substrate membrane and enhancement of its surface hydrophilicity to improve anti-fouling performance of resultant thin-film composite forward osmosis membrane, J. Membr. Sci. 511 (2016) 40–53. [57] W. Mu, J. Liu, J. Wang, H. Mao, X. Wu, Z. Li, Y. Li, Bioadhesion-inspired fabrication of robust thin-film composite membranes with tunable solvent permeation properties, RSC Adv. 6 (2016) 103981–103992. [58] Y.N. Wang, C.Y. Tang, Fouling of nanofiltration, reverse osmosis and ultrafiltration membranes by protein mixtures: the role of inter-foulant-species interaction, Environ. Sci. Technol. 45 (2011) 6373–6379. [59] Y.N. Wang, C.Y. Tang, Nanofiltration membrane fouling by oppositely charged macromolecules: investigation on flux behavior, foulant mass deposition, and solute rejection, Environ. Sci. Technol. 45 (2011) 8941–8947. [60] N. Subhi, A.R.D. Verliefde, V. Chen, P. Le-Clech, Assessment of physicochemical interactions in hollow fibre ultrafiltration membrane by contact angle analysis, J. Membr. Sci. 403 (2012) 32–40. [61] S. Liang, G. Qi, K. Xiao, J. Sun, E.P. Giannelis, X. Huang, M. Elimelech, 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 (2014) 94–101. [62] Q. She, R. Wang, A.G. Fane, C.Y. Tang, Membrane fouling in osmotically driven membrane processes: a review, J. Membr. Sci. 499 (2016) 201–233. [63] Y.-N. Wang, X. Li, R. Wang, Silica scaling and scaling control in pressure retarded osmosis processes, J. Membr. Sci. 541 (2017) 73–84. [64] Y.N. Wang, C.Y. Tang, Protein fouling of nanofiltration, reverse osmosis, and ultrafiltration membranes-the role of hydrodynamic conditions, solution chemistry, and membrane properties, J. Membr. Sci. 376 (2011) 275–282. [65] S.C. Chen, C.F. Wan, T.S. Chung, Enhanced fouling by inorganic and organic foulants on pressure retarded osmosis (PRO) hollow fiber membranes under high pressures, J. Membr. Sci. 479 (2015) 190–203. [66] J. Kim, B. Kim, D. Inhyuk Kim, S. Hong, Evaluation of apparent membrane performance parameters in pressure retarded osmosis processes under varying draw pressures and with draw solutions containing organics, J. Membr. Sci. 493 (2015) 636–644.
membranes with enhanced antifouling properties for osmotic power generation from municipal wastewater, Water Res. 104 (2016) 389–396. S. Rajabzadeh, C. Liu, L. Shi, R. Wang, Preparation of low-pressure water softening hollow fiber membranes by polyelectrolyte deposition with two bilayers, Desalination 344 (2014) 64–70. L. Ouyang, R. Malaisamy, M.L. Bruening, Multilayer polyelectrolyte films as nanofiltration membranes for separating monovalent and divalent cations, J. Membr. Sci. 310 (2008) 76–84. O. Sanyal, Z. Liu, J. Yu, B.M. Meharg, J.S. Hong, W. Liao, I. Lee, Designing foulingresistant clay-embedded polyelectrolyte multilayer membranes for wastewater effluent treatment, J. Membr. Sci. 512 (2016) 21–28. W. Ma, M.S. Rahaman, H. Therien-Aubin, Controlling biofouling of reverse osmosis membranes through surface modification via grafting patterned polymer brushes, Proceedings, Annual Conference - Canadian Society for Civil Engineering, 2015, pp. 9–17. G. Decher, Fuzzy nanoassemblies: toward layered polymeric multicomposites, Science 277 (1997) 1232–1237. P. Bertrand, A. Jonas, A. Laschewsky, R. Legras, Ultrathin polymer coatings by complexation of polyelectrolytes at interfaces: suitable materials, structure and properties, Macromol. Rapid Commun. 21 (2000) 319–348. A.N.D. Egueh, B. Lakard, P. Fievet, S. Lakard, C. Buron, Charge properties of membranes modified by multilayer polyelectrolyte adsorption, J. Colloid Interface Sci. 344 (2010) 221–227. P.T. Hammond, Recent explorations in electrostatic multilayer thin film assembly, Curr. Opin. Colloid Interf. Sci. 4 (1999) 430–442. S. Chou, R. Wang, A.G. Fane, Robust and high performance hollow fiber membranes for energy harvesting from salinity gradients by pressure retarded osmosis, J. Membr. Sci. 448 (2013) 44–54. Y. Li, R. Wang, S. Qi, C. Tang, Structural stability and mass transfer properties of pressure retarded osmosis (PRO) membrane under high operating pressures, J. Membr. Sci. 488 (2015) 143–153. L.E. Stephans, A. Myles, R.R. Thomas, Kinetics of alkaline hydrolysis of a polyimide surface, Langmuir 16 (2000) 4706–4710. S. Chou, L. Shi, R. Wang, C.Y. Tang, C. Qiu, A.G. Fane, Characteristics and potential applications of a novel forward osmosis hollow fiber membrane, Desalination 261 (2010) 365–372. L. Huang, N.N. Bui, M.T. Meyering, T.J. Hamlin, J.R. McCutcheon, Novel hydrophilic nylon 6,6 microfiltration membrane supported thin film composite membranes for engineered osmosis, J. Membr. Sci. 437 (2013) 141–149. S. Zhang, T.S. Chung, Minimizing the instant and accumulative effects of salt permeability to sustain ultrahigh osmotic power density, Environ. Sci. Technol. 47 (2013) 10085–10092. M. Gu, J. Lee, Y. Kim, J.S. Kim, B.Y. Jang, K.T. Lee, B.S. Kim, Inhibiting the shuttle effect in lithium-sulfur batteries using a layer-by-layer assembled ion-permselective
139