- Email: [email protected]
High flux polyamide thin film composite forward osmosis membranes prepared from porous substrates made of polysulfone and polyethersulfone blends
Desalination 336 (2014) 72–79
Contents lists available at ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
High flux polyamide thin film composite forward osmosis membranes prepared from porous substrates made of polysulfone and polyethersulfone blends Yuanna Sun a, Lixin Xue a,b,⁎, Yaojian Zhang a, Xiulan Zhao a,b, Yan Huang a, Xudong Du a a b
Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 519 Zhuangshi Road, Ningbo, Zhejiang Province 315201, PR China Ningbo Lianhua Environmental Science and Technology Co., Ltd, Yuyao City East Road Kechuang Center, Ningbo, Zhejiang Province 315400, PR China
H I G H L I G H T S • High flux TFC FO membranes were made from polyamide on PSf/PES blend supports. • The ratio of PSf/PES affected the porous structure and top skin section thickness. • The initial roughness from the substrate was amplified in the interfacial copolymerization process.
a r t i c l e
i n f o
Article history: Received 17 October 2013 Received in revised form 13 December 2013 Accepted 30 December 2013 Available online 23 January 2014 Keywords: Forward osmotic High water flux Polymer blends TFC membranes
a b s t r a c t High water flux polyamide thin film composite (TFC) forward osmosis (FO) membranes with porous supporting substrate layers made from polysulfone (PSf) and polyethersulfone (PES) blends of varied PSf/PES ratio were prepared. The impact of PSf/PES ratio on the porous structure of the supporting substrates and resulting properties of TFC-FO membranes formed were investigated. There were three distinguished sections observed in the cross-sections of the substrates, a top dense skin section, a central porous section, and a bottom open section according to SEM images. With the joint favoring effects from maximum open bottom section percentage in the cross-section and maximum surface roughness, membrane samples made from substrates with a PSf/PES ratio of 2/3 exhibit a highest water flux of 27.6 L/m2 h for 2 M NaCl draw solution from a de-ionized water feed solution. And their water flux reached 12.7 L/m2 h under PRO mode and 10.6 L/m2 h under FO mode from an ocean seawater feed solution containing 3.5 wt.% of NaCl. This high flux is related to the existence of heavily populated cone shaped protrusions observed on the surfaces of formed TFC-FO membranes, resulting from amplification in initial surface roughness of the porous substrate during the active layer forming interfacial co-polymerization process. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Having adequate clean water resources are critical for the wellbeing of almost all aspects of modern society including economy, public health, national security, ecosystem and others. However, with an exponentially growing population, water scarcity is becoming a serious global challenge in our time. It is reported that more than 1.2 billion people in the world are lacking access to clean and safe drinking water [1,2]. Extra incentives have been put in to search for technological approaches to produce clean water under controllable cost, e.g. desalination of seawater or brackish water [3,4] and reuse of reclaimed water [5]. ⁎ Corresponding author at: Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 519 Zhuangshi Road, Ningbo, Zhejiang Province 315201, PR China. Tel.: +86 574 86685256; fax: +86 574 86685186. E-mail address: [email protected] (L. Xue). 0011-9164/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.desal.2013.12.036
Since more than 97% of the global water resource is presented as seawater and most of the water shortage in the earth happened in the coastal heavily populated regions, desalination of sea water is becoming more and more important as an approach in generating clean and safe water to ease the global water shortage problem. Reverse osmosis (RO) is a familiar process widely used in the seawater desalination, but the applications are greatly thwarted by its high cost resulting from high energy consumption, equipment requirement, and membrane fouling propensity. FO, as an emerging water treatment technology, has gained increasing interest in recent years. In the process, a draw solution containing specifically designed solute, e.g. ammonium bicarbonate and glucose, which have higher osmotic pressure to draw water from the feed solution such as sea water, and then clean water may be generated by separating water from the draw solution. Compared to RO, FO has many potential advantages: (1) FO process is a spontaneous low pressure process that needs lower external energy; (2) even it has
Y. Sun et al. / Desalination 336 (2014) 72–79
high rejection for a wide range of contaminants, there is much less membrane fouling [6–9]. In addition, this osmotic-driven low pressure membrane separation process possesses great potential to achieve energy-efficient separations in many areas, not only for seawater desalination [10,11], but also for waste water treatment [12–14], food processing [15–17], pharmaceutical industry [18,19], electric power production [20–22] and so on. Even though many researchers pay much attention to develop the new draw solution [23,24], one of the key issues to increase the efficiency of the process still lie in the development of high water flux FO membrane with optimized structures. M. Elimelech found that neither polyamide RO composite membrane, nor cellulose triacetate RO membrane could be used for FO processes due to low water flux of 1–2 L/ m2 h [11]. He also found that commercialized cellulose triacetate (CTA) FO membranes with optimized structure showed higher water flux of 6–8 L/m2 h. However, these water flux values are still far too low from what is expected for applications in practical FO systems. In a conventional RO system, the active layer of the membrane is the crux of the separation while the substrate of the membrane was mainly to provide support under higher pressure. However, in FO system, both the solute and water need to diffuse in the porous substrate before reaching the thin active layer, the structure and property of the substrate play greater role. The thin film composite membranes used for high flux FO process should have their own optimized substrate structure other than those for RO processes. Currently a lot of efforts have been put in to explore the high water flux in FO process. Recently, interesting works focusing on using novel materials from copolymerization, grafting, surface modification and polymer blend, have been reported in optimizing the porous structures of TFC membranes substrate for higher water flux FO processes [25–27]. Compared to the development of new membrane materials, modifying the existing materials is relatively simple and effective. Among them, blending a novel component into existing one during the pore formation process, adjusting the porous structure and properties of the supporting substrate simultaneously, has become one of the research focuses in the search for high flux FO membranes. Here, we reported our recent results in the efforts to prepare high water flux polyamide TFC-FO membranes by blending less polar polyethersulfone (PES) component into polysulfone (PSf) component in the porous supporting structures. Morphologies and physical characteristics of the resulting PSF/PES blending membranes were investigated. It is demonstrated that high water flux FO membrane with stable high salt rejection may be obtained with minimum twist from the current TFC membrane preparation process.
73
2. Experimental 2.1. Materials Polysulfone (PSf, P-3500 LCD MB7, Mn = 80,000) and Polyethersulfone (PES, veradel3000P, Mn = 62,000) were purchased from Trump Chemical Corp., China, and dried in a vacuum oven at 80 °C for 24 h before use. N-methyl-2-pyrrolidone (NMP, anhydrous, N99%), polyethylene glycol 600 (PEG600, anhydrous, N99%), 1,3-phenylenediamine (MPD, N99.5%), 1,3,5-benzenetricarbonyl trichloride (TMC, 98%), n-hexane (97%), NaCl (crystals) were all purchased from Aladdin Reagent Co., Ltd., China.
2.2. Membrane preparation and characterization 2.2.1. Fabrication of membrane substrates Polymer mixtures of PSf/PES with the weight ratios of 5/0, 4/1, 3/2, 2/3, 1/4 and 0/5, were prepared and named as M1s, M2s, M3s, M4s, M5s and M6s, respectively. To fabricate membrane substrates for interfacial polymerization, polymer casting solutions containing 18 wt.% of the above polymer mixtures and 10 wt.% of PEG were prepared using NMP as solvent. The casting solutions were de-aerated under reduced pressure for 1 h before casting onto a glass plate with a casting knife of 130 μm in thickness. The nascent membranes were then immersed into a water coagulation bath immediately at room temperature. After residual solvent was removed by washing in a DI water bath for 1 h, the samples were kept in another DI water bath before interfacial polymerization was carried out.
2.2.2. Interfacial polymerization of TFC-FO membrane The PSf/PES substrates were first immersed in an aqueous solution of 2 wt.% MPD for 2 min. A filter paper was used to remove the excess MPD solution. Next, the membrane was immersed into a 0.1 wt.% TMC in n-hexane solution for 1 min. The composite membranes were then cured at 95 °C for 3 min, followed by cooling in the air for 1 min. Thereafter the fabricated TFC-FO membranes were rinsed thoroughly to remove the residual solvent and stored in DI water at room temperature. We named the composite membranes after polymerization as M1c, M2c, M3c, M4c, M5c and M6c corresponding to M1s to M6s respectively.
Fig. 1. Schematic diagram of FO system.
74
Y. Sun et al. / Desalination 336 (2014) 72–79
Table 1 Characteristic data of TFC-FO membranes and substrates.⁎ Sample
Ml
M2
M3
M4
MS
M6
FO membranes Reverse solute flux (g/m2 h) FO flux (L/m2 h) FO surface roughness (nm)
5.6 (±0.2) 5.0 (±0.1) 17.8 (±1.9)
26.1 (±0.3) 10.0 (±0.4) 22.5 (±3.2)
31.7 (±0.5) 20.0 (±0.8) 73.6 (±4.8)
37.5 (±0.6) 27.6 (±0.9) 130 (±5.2)
20.4 (±0.3) 9.9 (±0.8) 86.4 (±3.5)
5.8 (±0.1) 6.0 (±0.1) 85.4 (±3.2)
Substrates PSf/PES Substrate surface roughness (nm) Substrate surface pore size (nm) Dense top skin thickness (μm) Open bottom section (%)
5/0 12.5 (±0.6) 70.3 (±1.3) 2.10 25
4/1 17.0 (±1.3) 66.2 (±0.9) 1.06 35
3/2 18.1 (±1.0) 56.5 (±1.4) 0.58 50
2/3 21.2 (±1.4) 46.0 (±1.2) 0.54 65
1/4 18.7 (±1.5) 42.1 (±1.3) 0.36 55
0/5 18.8 (±0.8) 38.9 (±0.8) 0.12 45
⁎ The operating conditions are as followed: the volume of both feed and draw solutions was 2.0 L at the start of each experimental run; a 2.0 M NaCl solution was used as draw solution and the DI water was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop driven by gear pumps.
2.2.3. Characterization of membrane substrates and formed TFC membranes The morphology of the cross section and top surface of the membranes substrates was characterized using a Hitachi S#4800 SEM (Japan) system. After immersing in ethanol, the cross sectioned samples were analyzed by fracturing the membrane in liquid nitrogen. All samples were sputtered gold for 2 min before observation. AFM images of the top surfaces were taken on a Dimension 3100V SPM system from Veeco (US) operated in tapping mode. The samples
were operated at room temperature with the scanning size of 20 μm ∗ 20 μm. The through-pore sizes of the membranes were characterized by a Liquid–Liquid Porometer (LLP-1200A, Porous Materials Inc. US). At first, Silwick solution (Porous Materials Inc. US) with surface tension adjusted to 20.1 dyn/cm (20 °C) was used to wet all the samples. After the sample was placed in a sealed sample chamber, a certain amount of IPA was injected into the chamber, and pure N2 was then
Fig. 2. The cross-section of membrane substrates made from varied PSf/PES ratios (a) M1s:5/0; (b) M2s:4/1; (c) M3s:3/2; (d) M4s:2/3; (e) M5s: 1/4; (f) M6s:0/5.
Y. Sun et al. / Desalination 336 (2014) 72–79
allowed to flow into the chamber behind the sample. With the increase of the pressure, the Silwick was pushed out by IPA from the occupied through-pores. The flow rate was measured until all through-pores were empty and the sample was considered dry. The pore diameter is calculated according to the following equation:
d¼
Cτ p
ð1Þ
where d = pore diameter (μm), C = constant: 0.415 when p is in psi, τ = differential surface tension of liquids (dynes/cm), p = differential pressure (psi).
75
2.2.4. Testing of membrane performance in forward osmosis mode The performance of the membrane was measured by a cross flow filtration system [28,29], (effective membrane area = 64.26 cm2). The volume of both feed and draw solutions was 2.0 L at the start of each experimental run, a 2.0 M NaCl solution was used as draw solution and the DI water was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop driven by gear pumps (Fig. 1). A balance (SE6001F,USA) connected to the computer recorded down the mass of water permeating into draw solution while a conductivity meter was used to record down the concentration of NaCl in feed solution. The water permeation flux ( Jv, L/m 2 h, abbreviated as LMH) is calculated from the weight changes draw solution. JV ¼
ΔV : Sm Δt
ð2Þ
Fig. 3. (a) The surface AFM 3D images of different membrane substrates from M1s to M6s and their mean surface roughness. (b) The surface AFM 3D images of different FO membranes from M1c to M6c and their mean surface roughness. (c) The “lotus-effect” of formation of M4c FO membrane.
76
Y. Sun et al. / Desalination 336 (2014) 72–79
Fig. 3 (continued).
Where ΔV is the change in volume of permeation water collected over a predetermined time Δt in the FO process duration; Sm is the effective membrane surface area (m2). The salt concentration in the feed water was determined from the conductivity measurement using a calibration curve for the single salt solution. And the reverse solute flux was calculated by the NaCl concentration changes in the feed solution, Js is thereafter determined from the increase of the feed conductivity: Js ¼
ΔðC t V t Þ Sm Δt
ð3Þ
where Ct and Vt are the salt concentration and the volume of the feed at the end of FO tests, respectively.
Characteristic data of TFC-FO membrane and substrate samples from above measurement and experiments are summarized in Table 1 for comparison. At least five samples were tested and the average value was obtained for each membrane.
2.2.5. Texting on osmotic seawater desalination The membranes were tested under two different modes: (1) pressure retarded osmosis (PRO mode) where the draw solution faces against the dense selective layer and (2) forward osmosis (FO) mode where the feed solution side faces against the dense selective layer. The volume of both feed and draw solutions was 2.0 L at the start of each experimental run, a 2.0 M NaCl solution was used as draw solution and the seawater (3.5 wt.%) was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop
Y. Sun et al. / Desalination 336 (2014) 72–79
77
Fig. 3 (continued).
driven by gear pumps. At least five samples were tested and the average value was obtained for each membrane. 3. Results and discussion 3.1. Characteristics of membrane substrates and TFC membranes As shown in Fig. 2, there are three distinguished sections in the typical cross-sections of the substrates of the TFC-FO membranes, a porous central section, an open bottom section, and a dense top skin section from the phase inversion process. The porous central section containing finger-like pores was formed by delayed phase separation caused by mutual diffusion of water and organic solvents in the phase inversion process. The pores were formed by the polymer poor phase while the walls were formed by the polymer rich phase. The open bottom section was formed mainly from polymer precipitation in the organic solution due to extraction of solvent by water, whose thickness was controlled by the speed of solvent extraction and the solubility of polymer in the solvent. A thicker open bottom section will be formed from faster solvent diffusion and/or less polymer solubility. Adding a less polar component such as PES into more polar PSf will decrease not only the solubility of polymer in polar organic solvent, but also the speed of solvent extraction in the phase inversion process. With these two counteracting effects, as shown in Table 1, the percentage of the open bottom section of the membranes substrates increased from 25% to 65% from M1s to M4s due to reduction in solubility of the
polymer blends in polar solvents, and then decreased from 65% to 45% from M4s to M6s due to reduction in the speed of solvent extraction. Therefore, M4s possess the most open structure with a maximum open bottom section percentage, whose PSF/PES ratio was 2/3. The dense top sponge skin sections were formed by immediate extraction of solvent from solution and precipitation of polymer skins by water. Obviously, faster diffusion tends to form thinner top skin sections with smaller surface pores which means the faster the diffusion is, the thinner the top skin is. We can speculate that with the increase of the ratio of PES to PSf, i.e. from M1s to M6s, the membrane formation speeded up, then the top skin became thinner and the pore size became smaller. According to Jiang [30], the thickness of the top skin section may also reduced by speed-up solvent extraction processes caused by higher affinity between solvent and water, which is consistent with our results. From M1s to M6s, as shown in Fig. 2 and Table 1, when PSf/PES ratio was changed from 5/0 to 0/5, the thickness of the top dense skin section decreased from 2.10 μm to 0.12 μm. At the same time, the top skin sections became thinner with smaller pores from quicker polymer precipitation. As shown in Table 1, the pore size on the top skin section of the substrates decreased gradually from 70.3 nm to 38.9 nm from M1s to M6s. The variation of polymer mass fraction in solution impact not only micro-structures, but also the surface morphology of the formed membranes. As summarized in Table 1, with the increase of PES content, the roughness of the TFC-FO membranes increased from M1 to M4 and then decreased from M4 to M6. Obviously, there is interaction between PSf
78
Y. Sun et al. / Desalination 336 (2014) 72–79
and PES, the variation of PES mass fraction in solution impact the diffusion, so impact the surface morphology of the formed membranes. The interaction between them could be changed by the mass proportion. However, the detail mechanism to explain why the support membranes became rough first and then smooth after more PES addition in the casting solutions is interesting and deserve further investigation. There is obvious amplification on membrane surface roughness from the interfacial copolymerization process. As shown in the 3D AFM images of Fig. 3, the top surface of form FO membranes M4c had a distinguishably higher surface roughness (Ra = 130 ± 5.2 nm) than other membrane samples, relating to the existence of heavily populated cone shaped protrusions pre-formed on the surfaces of M4s. The roughness of M4c (PSf/PES 2/3) was 16.7 times that of M1c (PSf/PES 5/0), but the roughness of corresponding initial support M4s was only twice of M1s. The amplification may be explained by certain levels of “lotus-effect” [31] in the two-step interfacial copolymerization process. As shown in Fig. 3(c), for the rough M4s surface, after removal of excess, the first loaded aqueous solution of MPD tend to form uneven hydrophilic surfaces on tops of the heavily populated cone shaped protrusions. During the copolymerization step, due to the repelling nature of the two solutions, only the taller points of M4s hydrophilic surfaces could have more contact with the hydrophobic TMC hexane solution and have more copolymer formed, resulting in bigger protrusion and more surface roughness in M4. Study on mechanism of this surface roughness amplifying effect during interfacial surface copolymerization is interesting and deserves additional investigation in the future. 3.2. Impact of PSf/PES ratio in substrate on FO membrane performance Fig. 4(a) shows the flux of TFC-FO membranes on substrate M1c to M6c using 2 M NaCl as draw solutions. M4c has the higher water flux of 27.6 L/m2 h in FO mode. These results agreed well with the structure characteristics of the membranes and their substrates discussed in last sections. First, from SEM, M4c substrate has the largest open bottom section percentage, therefore, least concentration polarization in the FO process. Second, from the AFM images, M4c has the highest surface roughness, therefore, largest contact area with water. With the combined favoring effects from the highest surface roughness, and maximum open bottom section percentage of the substrates, M4c membranes were created to have exceptionally high water flux values. Fig. 4(b) shows the reverse solute flux among the membranes. All membranes showed low reverse solute flux, indicating the active top layers formed in all membranes were dense and homogeneous without leakage. Compared to TFC-FO membranes with other substrate systems [32–34], the TFC-FO membranes made from substrates made of PSf and PES blends in the ratio of about 2/3 shows both higher water flux and lower salt permeation, leading to potential applications in lower cost FO water treatment systems. In FO applications for desalination and water treatment, the active layer of the membrane faces the feed solution and the porous support layer faces the draw solution. As water permeates the active layer, the draw solution within the porous substructure becomes diluted. This is referred to as dilutive internal concentration polarization (ICP). According to Tang et al. [35], FO water flux is highly non-linear at high draw solute concentration where solute reverse diffusion may cause enhanced ICP of the back-diffused draw solutes. No obvious ICP was observed for M4c at 2 M NaCl as draw solution (See Fig. 5.), which may relate to the open structure caused by the large open bottom section percentage as observed SEM of M4c in Fig. 2.
Fig. 4. (a) The water flux of different TFC-FO membranes with varied PSf/PES ratios in the membrane substrate, (b) the reverse salt flux of different TFC-FO membranes with varied PSf/PES ratios in the membrane substrate, using a 2 M NaCl solution as draw solution, DI water as feed solution.
M4c in PRO and in FO modes with varied NaCl concentration in draw solutions. Using 2 M NaCl as a draw solution, the water flux reached 12.7 L/m2 h under PRO mode and 10.6 L/m2 h under FO mode, respectively. When tested the draw solution from 2 M to 5 M, the membrane showed certain level of ICP, because the line was not straight. As shown in Fig. 6, even with simulated seawater as feed, the ICP caused flux deviation under higher drawing solute concentrations is
3.3. Osmotic seawater desalination In order to test the feasibility of using these high performance FO membranes in sea water desalination processes, a simulated seawater solution was used as feed solution. Fig. 6 showed the water fluxes of
Fig. 5. The water flux and reverse solute flux of M4c using a 2 M NaCl solution as draw olution, DI water as feed solution.
Y. Sun et al. / Desalination 336 (2014) 72–79
Fig. 6. The water flux of PRO and FO system of M4c tests with varied draw solution concentration using seawater (3.5 wt.%) as feed solution.
not severe for M4c, so we conclude that no severe ICP happens for M4c at a high draw solution concentration level from 2 M to 5 M. 4. Conclusions Three distinguished sections were observed in the cross-sections of porous substrates prepared with PSF/PES blend using phase inversion processes, a top dense skin section, a central porous section, and a bottom open section. The structures and properties of formed polyamide TFC-FO membranes could be greatly impacted by the PES content in the substrates. More PES content led to thinner top skin sections with smaller surface pores. There was an obvious amplification on initial substrate membrane surface roughness in the active layer forming two-step interfacial co-polymerization process and the existence of heavily populated cone shaped protrusions were observed. The bottom open section percentage and the roughness of the formed FO membranes both reached maximum at a PSf/PES ratio of 2/3, leading to TFC-FO membranes with exceptionally high water flux with excellent salt rejection. For these special membranes, no severe ICP was observed in FO operation under higher salt concentration because of the existence of open bottom section with extended large pores. Acknowledgments The authors are grateful for financial support of Natural Science Foundation of Ningbo City, China (Nos. 2012A610083 and 2013C910012), National Natural Science Foundation of China (Grant No. 51273211) and the Grants from Ministry of Science and Technology of China (2012DFR50470 and SQ2013BAJY4228). References [1] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Marinas, A.M. Mayes, Science and technology for water purification in the coming decades, Nature 452 (2008) 301–310. [2] M. Elimelech, The global challenge for adequate and safe water, J. Water Supply Res. Technol. AQUA 55 (2006) 3–10. [3] R. Bond, S. Veerapaneni, Zeroing in on ZLD technologies for inland desalination, J. Am. Water Works Assoc. 100 (2008) 76–89. [4] A.D. Khawaji, I.K. Kutubkhanah, J.M. Wie, Advances in seawater desalination technologies, Desalination 221 (2008) 47–69. [5] T. Asano, F. Burton, H. Leverenz, R. Tsuchihashi, G. Tchobanoglous, Water Reuse: Issues, Technologies and Applications, McGraw-Hill, New York, 2007.
79
[6] W.C.L. Lay, T.H. Chong, C.Y. Tang, A.G. Fane, J. Zhang, Y. Liu, Fouling propensity of forward osmosis: investigation of the slower flux decline phenomenon, Water Sci. Technol. 61 (2010) 927–936. [7] B. Mi, M. Elimelech, Chemical and physical aspects of organic fouling of forward osmosis membranes, J. Membr. Sci. 320 (2008) 292–302. [8] M. Baoxia, M. Elimelech, Organic fouling of forward osmosis membranes: fouling reversibility and cleaning without chemical reagents, J. Membr. Sci. 348 (2010) 337–345. [9] E.R. Cornelissen, D. Harmsen, K.F. de Korte, C.J. Ruiken, J.-J. Qin, H. Oo, L.P. Wessels, Membrane fouling and process performance of forward osmosis membranes on activated sludge, J. Membr. Sci. 319 (2008) 158–168. [10] J.O. Kessler, C.D. Moody, Drinking-water form sea-water by forward osmosis, Desalination 18 (1976) 297–306. [11] M. Elimelech, J.R. McCutcheon, R.L. McGinnis, A novel ammonia-carbon dioxide forward (direct) osmosis desalination process, Desalination 174 (2005) 1–11. [12] F. Votta, S.M. Barnett, D.K. Anderson, Concentration of industrial waste by direct osmosis, Completion Report, 1974, University of Rhode Island. [13] D.K. Anderson, Concentration of Dilute Industrial Wastes by Direct Osmosis, University of Rhode Island, 1977. [14] R.W. Holloway, A.E. Childress, K.E. Dennett, T.Y. Cath, Forward osmosis for concentration of anaerobic digester centrate, Proceedings of the AWWA Membrane Technology Conference and Exposition, Phoenix, AZ, 2005. [15] E.G. Beaudry, K.A. Lampi, Membrane technology for direct-osmosis concentration of fruit juices, Food Technol. 44 (1990) 121–126. [16] M.I. Dova, K.B. Petrotos, H.N. Lazarides, On the direct osmotic concentration of liquid foods. Part I: impact of process parameters on process performance, J. Food Eng. 78 (2007) 422–430. [17] M.I. Dova, K.B. Petrotos, H.N. Lazarides, On the direct osmotic concentration of liquid foods: part II. Development of a generalized model, J. Food Eng. 78 (2007) 431–437. [18] F. Theeuwes, S.I. Yum, Principles of the design and operation of generic osmotic pumps for the delivery of semisolid or liquid drug formulations, Ann. Biomed. Eng. 4 (1976) 343–353. [19] J. Wright, S.I. Yum, R.M. Johnson, DUROS® osmotic pharmaceutical systems for parenteral & site-directed therapy, Drug Deliv. Technol. 3 (2003) 64–73. [20] H.H.G. Jellinek, H. Masuda, Osmo-power — theory and performance of an osmo-power pilot-plant, Ocean Eng. 8 (1981) 103–128. [21] K.L. Lee, R.W. Baker, H.K. Lonsdale, Membranes for power-generation by pressure-retarded osmosis, J. Membr. Sci. 8 (1981) 141–171. [22] S. Loeb, Large-scale power production by pressure-retarded osmosis, using river water and sea water passing through spiral modules, Desalination 143 (2002) 115–122. [23] S.K. Yen, F.M. Haja, N.M. Su, K.Y. Wang, T.S. Chung, Study of draw solutes using 2-methylimidazole-based compounds in forward osmosis, J. Membr. Sci. 364 (2010) 242–252. [24] A. Razmjou, G.P. Simon, H. Wang, Effect of particle size on the performance of forward osmosis desalination by stimuli-responsive polymer hydrogels as a draw agent, Chem. Eng. J. 215 (2013) 913–920. [25] J.T. Arena, B. McCloskey, B.D. Freeman, J.R. McCutcheon, Surface modification of thin film composite membrane support layers with polydopamine: enabling use of reverse osmosis membranes in pressure retarded osmosis, J. Membr. Sci. 375 (2011) 55–62. [26] C. Qiu, S. Qi, C.Y. Tang, Synthesis of high flux forward osmosis membranes by chemically crosslinked layer-by-layer polyelectrolytes, J. Membr. Sci. 381 (2011) 74–80. [27] A. Tiraferri, Y. Kang, E.P. Giannelis, M. Elimelech, Highly hydrophilic thin-film composite forward osmosis membranes functionalized with surface-tailored nanoparticles, ACS Appl. Mater. Interfaces 4 (2012) 5044–5053. [28] J.R. McCutcheon, M. Elimelech, Influence of membrane support layer hydrophobicity on water flux in osmotically driven membrane processes, J. Membr. Sci. 318 (2008) 458–466. [29] J.R. McCutcheon, R.L. McGinnis, M. Elimelech, A novel ammonia-carbon dioxide forward (direct) osmosis desalination process, Desalination 174 (2005) 1–11. [30] J. Peng, Y. Su, W. Chen, Q. Shi, Z. Jiang, Effects of coagulation bath temperature on the separation performance and antifouling property of poly(ether sulfone) ultrafiltration membranes, Ind. Eng. Chem. Res. 49 (2010) 4858–4864. [31] W. Barthlott, C. Neinhuis, Purity of the sacred lotus, or escape from contamination in biological surfaces, Planta 202 (1997) 1–8. [32] J. Wei, C. Qiu, C.Y. Tang, R. Wang, A.G. Fane, Synthesis and characterization of flat-sheet thin film composite forward osmosis membranes, J. Membr. Sci. 372 (2011) 292–302. [33] N.Y. Yip, A. Tiraferri, W.A. Phillip, J.D. Schiffman, M. Elimelech, High performance thin-film composite forward osmosis membrane, Environ. Sci. Technol. 44 (2010) 3812–3818. [34] 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. [35] 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.
Contents lists available at ScienceDirect
Desalination journal homepage: www.elsevier.com/locate/desal
High flux polyamide thin film composite forward osmosis membranes prepared from porous substrates made of polysulfone and polyethersulfone blends Yuanna Sun a, Lixin Xue a,b,⁎, Yaojian Zhang a, Xiulan Zhao a,b, Yan Huang a, Xudong Du a a b
Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 519 Zhuangshi Road, Ningbo, Zhejiang Province 315201, PR China Ningbo Lianhua Environmental Science and Technology Co., Ltd, Yuyao City East Road Kechuang Center, Ningbo, Zhejiang Province 315400, PR China
H I G H L I G H T S • High flux TFC FO membranes were made from polyamide on PSf/PES blend supports. • The ratio of PSf/PES affected the porous structure and top skin section thickness. • The initial roughness from the substrate was amplified in the interfacial copolymerization process.
a r t i c l e
i n f o
Article history: Received 17 October 2013 Received in revised form 13 December 2013 Accepted 30 December 2013 Available online 23 January 2014 Keywords: Forward osmotic High water flux Polymer blends TFC membranes
a b s t r a c t High water flux polyamide thin film composite (TFC) forward osmosis (FO) membranes with porous supporting substrate layers made from polysulfone (PSf) and polyethersulfone (PES) blends of varied PSf/PES ratio were prepared. The impact of PSf/PES ratio on the porous structure of the supporting substrates and resulting properties of TFC-FO membranes formed were investigated. There were three distinguished sections observed in the cross-sections of the substrates, a top dense skin section, a central porous section, and a bottom open section according to SEM images. With the joint favoring effects from maximum open bottom section percentage in the cross-section and maximum surface roughness, membrane samples made from substrates with a PSf/PES ratio of 2/3 exhibit a highest water flux of 27.6 L/m2 h for 2 M NaCl draw solution from a de-ionized water feed solution. And their water flux reached 12.7 L/m2 h under PRO mode and 10.6 L/m2 h under FO mode from an ocean seawater feed solution containing 3.5 wt.% of NaCl. This high flux is related to the existence of heavily populated cone shaped protrusions observed on the surfaces of formed TFC-FO membranes, resulting from amplification in initial surface roughness of the porous substrate during the active layer forming interfacial co-polymerization process. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Having adequate clean water resources are critical for the wellbeing of almost all aspects of modern society including economy, public health, national security, ecosystem and others. However, with an exponentially growing population, water scarcity is becoming a serious global challenge in our time. It is reported that more than 1.2 billion people in the world are lacking access to clean and safe drinking water [1,2]. Extra incentives have been put in to search for technological approaches to produce clean water under controllable cost, e.g. desalination of seawater or brackish water [3,4] and reuse of reclaimed water [5]. ⁎ Corresponding author at: Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 519 Zhuangshi Road, Ningbo, Zhejiang Province 315201, PR China. Tel.: +86 574 86685256; fax: +86 574 86685186. E-mail address: [email protected] (L. Xue). 0011-9164/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.desal.2013.12.036
Since more than 97% of the global water resource is presented as seawater and most of the water shortage in the earth happened in the coastal heavily populated regions, desalination of sea water is becoming more and more important as an approach in generating clean and safe water to ease the global water shortage problem. Reverse osmosis (RO) is a familiar process widely used in the seawater desalination, but the applications are greatly thwarted by its high cost resulting from high energy consumption, equipment requirement, and membrane fouling propensity. FO, as an emerging water treatment technology, has gained increasing interest in recent years. In the process, a draw solution containing specifically designed solute, e.g. ammonium bicarbonate and glucose, which have higher osmotic pressure to draw water from the feed solution such as sea water, and then clean water may be generated by separating water from the draw solution. Compared to RO, FO has many potential advantages: (1) FO process is a spontaneous low pressure process that needs lower external energy; (2) even it has
Y. Sun et al. / Desalination 336 (2014) 72–79
high rejection for a wide range of contaminants, there is much less membrane fouling [6–9]. In addition, this osmotic-driven low pressure membrane separation process possesses great potential to achieve energy-efficient separations in many areas, not only for seawater desalination [10,11], but also for waste water treatment [12–14], food processing [15–17], pharmaceutical industry [18,19], electric power production [20–22] and so on. Even though many researchers pay much attention to develop the new draw solution [23,24], one of the key issues to increase the efficiency of the process still lie in the development of high water flux FO membrane with optimized structures. M. Elimelech found that neither polyamide RO composite membrane, nor cellulose triacetate RO membrane could be used for FO processes due to low water flux of 1–2 L/ m2 h [11]. He also found that commercialized cellulose triacetate (CTA) FO membranes with optimized structure showed higher water flux of 6–8 L/m2 h. However, these water flux values are still far too low from what is expected for applications in practical FO systems. In a conventional RO system, the active layer of the membrane is the crux of the separation while the substrate of the membrane was mainly to provide support under higher pressure. However, in FO system, both the solute and water need to diffuse in the porous substrate before reaching the thin active layer, the structure and property of the substrate play greater role. The thin film composite membranes used for high flux FO process should have their own optimized substrate structure other than those for RO processes. Currently a lot of efforts have been put in to explore the high water flux in FO process. Recently, interesting works focusing on using novel materials from copolymerization, grafting, surface modification and polymer blend, have been reported in optimizing the porous structures of TFC membranes substrate for higher water flux FO processes [25–27]. Compared to the development of new membrane materials, modifying the existing materials is relatively simple and effective. Among them, blending a novel component into existing one during the pore formation process, adjusting the porous structure and properties of the supporting substrate simultaneously, has become one of the research focuses in the search for high flux FO membranes. Here, we reported our recent results in the efforts to prepare high water flux polyamide TFC-FO membranes by blending less polar polyethersulfone (PES) component into polysulfone (PSf) component in the porous supporting structures. Morphologies and physical characteristics of the resulting PSF/PES blending membranes were investigated. It is demonstrated that high water flux FO membrane with stable high salt rejection may be obtained with minimum twist from the current TFC membrane preparation process.
73
2. Experimental 2.1. Materials Polysulfone (PSf, P-3500 LCD MB7, Mn = 80,000) and Polyethersulfone (PES, veradel3000P, Mn = 62,000) were purchased from Trump Chemical Corp., China, and dried in a vacuum oven at 80 °C for 24 h before use. N-methyl-2-pyrrolidone (NMP, anhydrous, N99%), polyethylene glycol 600 (PEG600, anhydrous, N99%), 1,3-phenylenediamine (MPD, N99.5%), 1,3,5-benzenetricarbonyl trichloride (TMC, 98%), n-hexane (97%), NaCl (crystals) were all purchased from Aladdin Reagent Co., Ltd., China.
2.2. Membrane preparation and characterization 2.2.1. Fabrication of membrane substrates Polymer mixtures of PSf/PES with the weight ratios of 5/0, 4/1, 3/2, 2/3, 1/4 and 0/5, were prepared and named as M1s, M2s, M3s, M4s, M5s and M6s, respectively. To fabricate membrane substrates for interfacial polymerization, polymer casting solutions containing 18 wt.% of the above polymer mixtures and 10 wt.% of PEG were prepared using NMP as solvent. The casting solutions were de-aerated under reduced pressure for 1 h before casting onto a glass plate with a casting knife of 130 μm in thickness. The nascent membranes were then immersed into a water coagulation bath immediately at room temperature. After residual solvent was removed by washing in a DI water bath for 1 h, the samples were kept in another DI water bath before interfacial polymerization was carried out.
2.2.2. Interfacial polymerization of TFC-FO membrane The PSf/PES substrates were first immersed in an aqueous solution of 2 wt.% MPD for 2 min. A filter paper was used to remove the excess MPD solution. Next, the membrane was immersed into a 0.1 wt.% TMC in n-hexane solution for 1 min. The composite membranes were then cured at 95 °C for 3 min, followed by cooling in the air for 1 min. Thereafter the fabricated TFC-FO membranes were rinsed thoroughly to remove the residual solvent and stored in DI water at room temperature. We named the composite membranes after polymerization as M1c, M2c, M3c, M4c, M5c and M6c corresponding to M1s to M6s respectively.
Fig. 1. Schematic diagram of FO system.
74
Y. Sun et al. / Desalination 336 (2014) 72–79
Table 1 Characteristic data of TFC-FO membranes and substrates.⁎ Sample
Ml
M2
M3
M4
MS
M6
FO membranes Reverse solute flux (g/m2 h) FO flux (L/m2 h) FO surface roughness (nm)
5.6 (±0.2) 5.0 (±0.1) 17.8 (±1.9)
26.1 (±0.3) 10.0 (±0.4) 22.5 (±3.2)
31.7 (±0.5) 20.0 (±0.8) 73.6 (±4.8)
37.5 (±0.6) 27.6 (±0.9) 130 (±5.2)
20.4 (±0.3) 9.9 (±0.8) 86.4 (±3.5)
5.8 (±0.1) 6.0 (±0.1) 85.4 (±3.2)
Substrates PSf/PES Substrate surface roughness (nm) Substrate surface pore size (nm) Dense top skin thickness (μm) Open bottom section (%)
5/0 12.5 (±0.6) 70.3 (±1.3) 2.10 25
4/1 17.0 (±1.3) 66.2 (±0.9) 1.06 35
3/2 18.1 (±1.0) 56.5 (±1.4) 0.58 50
2/3 21.2 (±1.4) 46.0 (±1.2) 0.54 65
1/4 18.7 (±1.5) 42.1 (±1.3) 0.36 55
0/5 18.8 (±0.8) 38.9 (±0.8) 0.12 45
⁎ The operating conditions are as followed: the volume of both feed and draw solutions was 2.0 L at the start of each experimental run; a 2.0 M NaCl solution was used as draw solution and the DI water was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop driven by gear pumps.
2.2.3. Characterization of membrane substrates and formed TFC membranes The morphology of the cross section and top surface of the membranes substrates was characterized using a Hitachi S#4800 SEM (Japan) system. After immersing in ethanol, the cross sectioned samples were analyzed by fracturing the membrane in liquid nitrogen. All samples were sputtered gold for 2 min before observation. AFM images of the top surfaces were taken on a Dimension 3100V SPM system from Veeco (US) operated in tapping mode. The samples
were operated at room temperature with the scanning size of 20 μm ∗ 20 μm. The through-pore sizes of the membranes were characterized by a Liquid–Liquid Porometer (LLP-1200A, Porous Materials Inc. US). At first, Silwick solution (Porous Materials Inc. US) with surface tension adjusted to 20.1 dyn/cm (20 °C) was used to wet all the samples. After the sample was placed in a sealed sample chamber, a certain amount of IPA was injected into the chamber, and pure N2 was then
Fig. 2. The cross-section of membrane substrates made from varied PSf/PES ratios (a) M1s:5/0; (b) M2s:4/1; (c) M3s:3/2; (d) M4s:2/3; (e) M5s: 1/4; (f) M6s:0/5.
Y. Sun et al. / Desalination 336 (2014) 72–79
allowed to flow into the chamber behind the sample. With the increase of the pressure, the Silwick was pushed out by IPA from the occupied through-pores. The flow rate was measured until all through-pores were empty and the sample was considered dry. The pore diameter is calculated according to the following equation:
d¼
Cτ p
ð1Þ
where d = pore diameter (μm), C = constant: 0.415 when p is in psi, τ = differential surface tension of liquids (dynes/cm), p = differential pressure (psi).
75
2.2.4. Testing of membrane performance in forward osmosis mode The performance of the membrane was measured by a cross flow filtration system [28,29], (effective membrane area = 64.26 cm2). The volume of both feed and draw solutions was 2.0 L at the start of each experimental run, a 2.0 M NaCl solution was used as draw solution and the DI water was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop driven by gear pumps (Fig. 1). A balance (SE6001F,USA) connected to the computer recorded down the mass of water permeating into draw solution while a conductivity meter was used to record down the concentration of NaCl in feed solution. The water permeation flux ( Jv, L/m 2 h, abbreviated as LMH) is calculated from the weight changes draw solution. JV ¼
ΔV : Sm Δt
ð2Þ
Fig. 3. (a) The surface AFM 3D images of different membrane substrates from M1s to M6s and their mean surface roughness. (b) The surface AFM 3D images of different FO membranes from M1c to M6c and their mean surface roughness. (c) The “lotus-effect” of formation of M4c FO membrane.
76
Y. Sun et al. / Desalination 336 (2014) 72–79
Fig. 3 (continued).
Where ΔV is the change in volume of permeation water collected over a predetermined time Δt in the FO process duration; Sm is the effective membrane surface area (m2). The salt concentration in the feed water was determined from the conductivity measurement using a calibration curve for the single salt solution. And the reverse solute flux was calculated by the NaCl concentration changes in the feed solution, Js is thereafter determined from the increase of the feed conductivity: Js ¼
ΔðC t V t Þ Sm Δt
ð3Þ
where Ct and Vt are the salt concentration and the volume of the feed at the end of FO tests, respectively.
Characteristic data of TFC-FO membrane and substrate samples from above measurement and experiments are summarized in Table 1 for comparison. At least five samples were tested and the average value was obtained for each membrane.
2.2.5. Texting on osmotic seawater desalination The membranes were tested under two different modes: (1) pressure retarded osmosis (PRO mode) where the draw solution faces against the dense selective layer and (2) forward osmosis (FO) mode where the feed solution side faces against the dense selective layer. The volume of both feed and draw solutions was 2.0 L at the start of each experimental run, a 2.0 M NaCl solution was used as draw solution and the seawater (3.5 wt.%) was used as feed solution. The feed and draw solution was circulated at a rate of 0.5 L/min in a closed loop
Y. Sun et al. / Desalination 336 (2014) 72–79
77
Fig. 3 (continued).
driven by gear pumps. At least five samples were tested and the average value was obtained for each membrane. 3. Results and discussion 3.1. Characteristics of membrane substrates and TFC membranes As shown in Fig. 2, there are three distinguished sections in the typical cross-sections of the substrates of the TFC-FO membranes, a porous central section, an open bottom section, and a dense top skin section from the phase inversion process. The porous central section containing finger-like pores was formed by delayed phase separation caused by mutual diffusion of water and organic solvents in the phase inversion process. The pores were formed by the polymer poor phase while the walls were formed by the polymer rich phase. The open bottom section was formed mainly from polymer precipitation in the organic solution due to extraction of solvent by water, whose thickness was controlled by the speed of solvent extraction and the solubility of polymer in the solvent. A thicker open bottom section will be formed from faster solvent diffusion and/or less polymer solubility. Adding a less polar component such as PES into more polar PSf will decrease not only the solubility of polymer in polar organic solvent, but also the speed of solvent extraction in the phase inversion process. With these two counteracting effects, as shown in Table 1, the percentage of the open bottom section of the membranes substrates increased from 25% to 65% from M1s to M4s due to reduction in solubility of the
polymer blends in polar solvents, and then decreased from 65% to 45% from M4s to M6s due to reduction in the speed of solvent extraction. Therefore, M4s possess the most open structure with a maximum open bottom section percentage, whose PSF/PES ratio was 2/3. The dense top sponge skin sections were formed by immediate extraction of solvent from solution and precipitation of polymer skins by water. Obviously, faster diffusion tends to form thinner top skin sections with smaller surface pores which means the faster the diffusion is, the thinner the top skin is. We can speculate that with the increase of the ratio of PES to PSf, i.e. from M1s to M6s, the membrane formation speeded up, then the top skin became thinner and the pore size became smaller. According to Jiang [30], the thickness of the top skin section may also reduced by speed-up solvent extraction processes caused by higher affinity between solvent and water, which is consistent with our results. From M1s to M6s, as shown in Fig. 2 and Table 1, when PSf/PES ratio was changed from 5/0 to 0/5, the thickness of the top dense skin section decreased from 2.10 μm to 0.12 μm. At the same time, the top skin sections became thinner with smaller pores from quicker polymer precipitation. As shown in Table 1, the pore size on the top skin section of the substrates decreased gradually from 70.3 nm to 38.9 nm from M1s to M6s. The variation of polymer mass fraction in solution impact not only micro-structures, but also the surface morphology of the formed membranes. As summarized in Table 1, with the increase of PES content, the roughness of the TFC-FO membranes increased from M1 to M4 and then decreased from M4 to M6. Obviously, there is interaction between PSf
78
Y. Sun et al. / Desalination 336 (2014) 72–79
and PES, the variation of PES mass fraction in solution impact the diffusion, so impact the surface morphology of the formed membranes. The interaction between them could be changed by the mass proportion. However, the detail mechanism to explain why the support membranes became rough first and then smooth after more PES addition in the casting solutions is interesting and deserve further investigation. There is obvious amplification on membrane surface roughness from the interfacial copolymerization process. As shown in the 3D AFM images of Fig. 3, the top surface of form FO membranes M4c had a distinguishably higher surface roughness (Ra = 130 ± 5.2 nm) than other membrane samples, relating to the existence of heavily populated cone shaped protrusions pre-formed on the surfaces of M4s. The roughness of M4c (PSf/PES 2/3) was 16.7 times that of M1c (PSf/PES 5/0), but the roughness of corresponding initial support M4s was only twice of M1s. The amplification may be explained by certain levels of “lotus-effect” [31] in the two-step interfacial copolymerization process. As shown in Fig. 3(c), for the rough M4s surface, after removal of excess, the first loaded aqueous solution of MPD tend to form uneven hydrophilic surfaces on tops of the heavily populated cone shaped protrusions. During the copolymerization step, due to the repelling nature of the two solutions, only the taller points of M4s hydrophilic surfaces could have more contact with the hydrophobic TMC hexane solution and have more copolymer formed, resulting in bigger protrusion and more surface roughness in M4. Study on mechanism of this surface roughness amplifying effect during interfacial surface copolymerization is interesting and deserves additional investigation in the future. 3.2. Impact of PSf/PES ratio in substrate on FO membrane performance Fig. 4(a) shows the flux of TFC-FO membranes on substrate M1c to M6c using 2 M NaCl as draw solutions. M4c has the higher water flux of 27.6 L/m2 h in FO mode. These results agreed well with the structure characteristics of the membranes and their substrates discussed in last sections. First, from SEM, M4c substrate has the largest open bottom section percentage, therefore, least concentration polarization in the FO process. Second, from the AFM images, M4c has the highest surface roughness, therefore, largest contact area with water. With the combined favoring effects from the highest surface roughness, and maximum open bottom section percentage of the substrates, M4c membranes were created to have exceptionally high water flux values. Fig. 4(b) shows the reverse solute flux among the membranes. All membranes showed low reverse solute flux, indicating the active top layers formed in all membranes were dense and homogeneous without leakage. Compared to TFC-FO membranes with other substrate systems [32–34], the TFC-FO membranes made from substrates made of PSf and PES blends in the ratio of about 2/3 shows both higher water flux and lower salt permeation, leading to potential applications in lower cost FO water treatment systems. In FO applications for desalination and water treatment, the active layer of the membrane faces the feed solution and the porous support layer faces the draw solution. As water permeates the active layer, the draw solution within the porous substructure becomes diluted. This is referred to as dilutive internal concentration polarization (ICP). According to Tang et al. [35], FO water flux is highly non-linear at high draw solute concentration where solute reverse diffusion may cause enhanced ICP of the back-diffused draw solutes. No obvious ICP was observed for M4c at 2 M NaCl as draw solution (See Fig. 5.), which may relate to the open structure caused by the large open bottom section percentage as observed SEM of M4c in Fig. 2.
Fig. 4. (a) The water flux of different TFC-FO membranes with varied PSf/PES ratios in the membrane substrate, (b) the reverse salt flux of different TFC-FO membranes with varied PSf/PES ratios in the membrane substrate, using a 2 M NaCl solution as draw solution, DI water as feed solution.
M4c in PRO and in FO modes with varied NaCl concentration in draw solutions. Using 2 M NaCl as a draw solution, the water flux reached 12.7 L/m2 h under PRO mode and 10.6 L/m2 h under FO mode, respectively. When tested the draw solution from 2 M to 5 M, the membrane showed certain level of ICP, because the line was not straight. As shown in Fig. 6, even with simulated seawater as feed, the ICP caused flux deviation under higher drawing solute concentrations is
3.3. Osmotic seawater desalination In order to test the feasibility of using these high performance FO membranes in sea water desalination processes, a simulated seawater solution was used as feed solution. Fig. 6 showed the water fluxes of
Fig. 5. The water flux and reverse solute flux of M4c using a 2 M NaCl solution as draw olution, DI water as feed solution.
Y. Sun et al. / Desalination 336 (2014) 72–79
Fig. 6. The water flux of PRO and FO system of M4c tests with varied draw solution concentration using seawater (3.5 wt.%) as feed solution.
not severe for M4c, so we conclude that no severe ICP happens for M4c at a high draw solution concentration level from 2 M to 5 M. 4. Conclusions Three distinguished sections were observed in the cross-sections of porous substrates prepared with PSF/PES blend using phase inversion processes, a top dense skin section, a central porous section, and a bottom open section. The structures and properties of formed polyamide TFC-FO membranes could be greatly impacted by the PES content in the substrates. More PES content led to thinner top skin sections with smaller surface pores. There was an obvious amplification on initial substrate membrane surface roughness in the active layer forming two-step interfacial co-polymerization process and the existence of heavily populated cone shaped protrusions were observed. The bottom open section percentage and the roughness of the formed FO membranes both reached maximum at a PSf/PES ratio of 2/3, leading to TFC-FO membranes with exceptionally high water flux with excellent salt rejection. For these special membranes, no severe ICP was observed in FO operation under higher salt concentration because of the existence of open bottom section with extended large pores. Acknowledgments The authors are grateful for financial support of Natural Science Foundation of Ningbo City, China (Nos. 2012A610083 and 2013C910012), National Natural Science Foundation of China (Grant No. 51273211) and the Grants from Ministry of Science and Technology of China (2012DFR50470 and SQ2013BAJY4228). References [1] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Marinas, A.M. Mayes, Science and technology for water purification in the coming decades, Nature 452 (2008) 301–310. [2] M. Elimelech, The global challenge for adequate and safe water, J. Water Supply Res. Technol. AQUA 55 (2006) 3–10. [3] R. Bond, S. Veerapaneni, Zeroing in on ZLD technologies for inland desalination, J. Am. Water Works Assoc. 100 (2008) 76–89. [4] A.D. Khawaji, I.K. Kutubkhanah, J.M. Wie, Advances in seawater desalination technologies, Desalination 221 (2008) 47–69. [5] T. Asano, F. Burton, H. Leverenz, R. Tsuchihashi, G. Tchobanoglous, Water Reuse: Issues, Technologies and Applications, McGraw-Hill, New York, 2007.
79
[6] W.C.L. Lay, T.H. Chong, C.Y. Tang, A.G. Fane, J. Zhang, Y. Liu, Fouling propensity of forward osmosis: investigation of the slower flux decline phenomenon, Water Sci. Technol. 61 (2010) 927–936. [7] B. Mi, M. Elimelech, Chemical and physical aspects of organic fouling of forward osmosis membranes, J. Membr. Sci. 320 (2008) 292–302. [8] M. Baoxia, M. Elimelech, Organic fouling of forward osmosis membranes: fouling reversibility and cleaning without chemical reagents, J. Membr. Sci. 348 (2010) 337–345. [9] E.R. Cornelissen, D. Harmsen, K.F. de Korte, C.J. Ruiken, J.-J. Qin, H. Oo, L.P. Wessels, Membrane fouling and process performance of forward osmosis membranes on activated sludge, J. Membr. Sci. 319 (2008) 158–168. [10] J.O. Kessler, C.D. Moody, Drinking-water form sea-water by forward osmosis, Desalination 18 (1976) 297–306. [11] M. Elimelech, J.R. McCutcheon, R.L. McGinnis, A novel ammonia-carbon dioxide forward (direct) osmosis desalination process, Desalination 174 (2005) 1–11. [12] F. Votta, S.M. Barnett, D.K. Anderson, Concentration of industrial waste by direct osmosis, Completion Report, 1974, University of Rhode Island. [13] D.K. Anderson, Concentration of Dilute Industrial Wastes by Direct Osmosis, University of Rhode Island, 1977. [14] R.W. Holloway, A.E. Childress, K.E. Dennett, T.Y. Cath, Forward osmosis for concentration of anaerobic digester centrate, Proceedings of the AWWA Membrane Technology Conference and Exposition, Phoenix, AZ, 2005. [15] E.G. Beaudry, K.A. Lampi, Membrane technology for direct-osmosis concentration of fruit juices, Food Technol. 44 (1990) 121–126. [16] M.I. Dova, K.B. Petrotos, H.N. Lazarides, On the direct osmotic concentration of liquid foods. Part I: impact of process parameters on process performance, J. Food Eng. 78 (2007) 422–430. [17] M.I. Dova, K.B. Petrotos, H.N. Lazarides, On the direct osmotic concentration of liquid foods: part II. Development of a generalized model, J. Food Eng. 78 (2007) 431–437. [18] F. Theeuwes, S.I. Yum, Principles of the design and operation of generic osmotic pumps for the delivery of semisolid or liquid drug formulations, Ann. Biomed. Eng. 4 (1976) 343–353. [19] J. Wright, S.I. Yum, R.M. Johnson, DUROS® osmotic pharmaceutical systems for parenteral & site-directed therapy, Drug Deliv. Technol. 3 (2003) 64–73. [20] H.H.G. Jellinek, H. Masuda, Osmo-power — theory and performance of an osmo-power pilot-plant, Ocean Eng. 8 (1981) 103–128. [21] K.L. Lee, R.W. Baker, H.K. Lonsdale, Membranes for power-generation by pressure-retarded osmosis, J. Membr. Sci. 8 (1981) 141–171. [22] S. Loeb, Large-scale power production by pressure-retarded osmosis, using river water and sea water passing through spiral modules, Desalination 143 (2002) 115–122. [23] S.K. Yen, F.M. Haja, N.M. Su, K.Y. Wang, T.S. Chung, Study of draw solutes using 2-methylimidazole-based compounds in forward osmosis, J. Membr. Sci. 364 (2010) 242–252. [24] A. Razmjou, G.P. Simon, H. Wang, Effect of particle size on the performance of forward osmosis desalination by stimuli-responsive polymer hydrogels as a draw agent, Chem. Eng. J. 215 (2013) 913–920. [25] J.T. Arena, B. McCloskey, B.D. Freeman, J.R. McCutcheon, Surface modification of thin film composite membrane support layers with polydopamine: enabling use of reverse osmosis membranes in pressure retarded osmosis, J. Membr. Sci. 375 (2011) 55–62. [26] C. Qiu, S. Qi, C.Y. Tang, Synthesis of high flux forward osmosis membranes by chemically crosslinked layer-by-layer polyelectrolytes, J. Membr. Sci. 381 (2011) 74–80. [27] A. Tiraferri, Y. Kang, E.P. Giannelis, M. Elimelech, Highly hydrophilic thin-film composite forward osmosis membranes functionalized with surface-tailored nanoparticles, ACS Appl. Mater. Interfaces 4 (2012) 5044–5053. [28] J.R. McCutcheon, M. Elimelech, Influence of membrane support layer hydrophobicity on water flux in osmotically driven membrane processes, J. Membr. Sci. 318 (2008) 458–466. [29] J.R. McCutcheon, R.L. McGinnis, M. Elimelech, A novel ammonia-carbon dioxide forward (direct) osmosis desalination process, Desalination 174 (2005) 1–11. [30] J. Peng, Y. Su, W. Chen, Q. Shi, Z. Jiang, Effects of coagulation bath temperature on the separation performance and antifouling property of poly(ether sulfone) ultrafiltration membranes, Ind. Eng. Chem. Res. 49 (2010) 4858–4864. [31] W. Barthlott, C. Neinhuis, Purity of the sacred lotus, or escape from contamination in biological surfaces, Planta 202 (1997) 1–8. [32] J. Wei, C. Qiu, C.Y. Tang, R. Wang, A.G. Fane, Synthesis and characterization of flat-sheet thin film composite forward osmosis membranes, J. Membr. Sci. 372 (2011) 292–302. [33] N.Y. Yip, A. Tiraferri, W.A. Phillip, J.D. Schiffman, M. Elimelech, High performance thin-film composite forward osmosis membrane, Environ. Sci. Technol. 44 (2010) 3812–3818. [34] 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. [35] 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.