Chinese Journal of Chemical Engineering, 19(5) 855862 (2011)

Integral PVA-PES Composite Membranes by Surface Segregation Method for Pervaporation Dehydration of Ethanol*
WU Hong ()1,2, LI Xianshi ()1, NIE Mingcheng ()1, LI Ben ()1 and JIANG Zhongyi ()1,**
1 2

Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China Tianjin Key Laboratory of Membrane Science and Desalination Technology, Tianjin University, Tianjin 300072, China

Abstract A facile surface segregation method was utilized to fabricate poly(vinyl alcohol)-polyethersulfone (PVA-PES) composite membranes. PVA and PES were first dissolved in dimethyl sulfoxide (DMSO), then casted on a glass plate and immersed in a coagulation bath. During the phase inversion process in coagulation bath, PVA spontaneously segregated to the polymer solution/coagulation bath interface. The enriched PVA on the surface was further crosslinked by glutaraldehyde. Scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and energy dispersive spectrometer (EDS) confirmed the integral and asymmetric membrane structure with a dense PVA-enriched surface and a porous PES-enriched support, as well as the surface enrichment of PVA. The coverage fraction of the membrane surface by PVA reached up to 86.8% when the PVA content in the membrane recipe was 16.7% (by mass). The water contact angle decreased with the increase of PVA content. The effect of coagulation bath type on membrane structure was analyzed. The membrane pervaporation performance was evaluated by varying the PVA content, the annealing temperature, feed concentration and operation temperature. The membrane exhibited a fairly good ethanol dehydration capacity and long-term operational stability. Keywords surface segregation, phase inversion, PVA-PES composite membrane, pervaporation, ethanol

INTRODUCTION

It is well known that bio-fuels including bio-ethanol have a number of environmental and economic benefits [1]. Among them, absolute ethanol containing 99.5% (by mass) or more ethanol is in the greatest demand [2]. Pervaporation has been demonstrated as an efficient membrane process for ethanol dehydration [3, 4] and as a promising alternative to conventional energy intensive technologies such as distillation [5, 6]. Composite membranes with a thin active layer supported on a porous substrate are often utilized for pervaporation dehydration due to their high permeation flux. However, if the hydrophilic active layer and hydroponic substrate do not swell in a coordinated manner, the active layer and support layer would peel off due to the interfacial stress between these two layers exceeding their maximum interfacial adhesion [7]. Therefore, considerable researches have focused on the structural stability of composite membranes [8, 9], which encompasses active layer cross-linking [10], multi-layer arrangement [11, 12], fusion [13] and interfacial polymerization [14, 15]. For instance, Shao et al. [13] fabricated SPEEK/PMMA-PVDF composite membranes with an integrated skin layer via a fusion technique. The SPEEK skin layer was prefabricated and then the substrate solution was applied to the skin layer to form a porous support layer by the wet phase inversion process. The SPEEK skin layer

was partially dissolved and integrated into the substrate layer. The long-term pervaporation operation showed that the structure stability of the resultant membranes was remarkably enhanced. The surface segregation is a common phenomenon of enrichment of one component at the blend surface [16] and has been widely used in fabricating porous membranes, such as ultrafiltration membrane [17-19]. The process can be described as that near the interface between the casting solution and the coagulation bath, hydrophilic component is transported more slowly into the casting solution than the hydrophobic component prior to precipitation, resulting in surface enrichment of the hydrophilic segments [20]. Therefore, polymer chains of the active layer and the support layer formed inter-penetrated network which enhance the structure stability. The flux and separation factor of asymmetric membrane used in pervaporation are about 0.127 kgm2h1 and 735 for 90% (by mass) ethanol solution [21]. PVA and PES are widely employed as membrane materials because of their superior comprehensive properties and membrane-forming property. In this study, PVA as a hydrophilic polymer and PES as a hydrophobic polymer are chosen to form an integrated composite membrane via surface segregation. The composition and structure of the resultant composite membrane were probed by several characterization techniques. The pervaporation dehydration of ethanol

Received 2011-06-10, accepted 2011-08-25. * Supported by the State Key Development Program for Basic Research of China (2009CB623404), Program for New Century Excellent Talents in University, the Programme of Introducing Talents of Discipline to Universities (B06006), State Key Laboratory for Modification of Chemical Fibers and Polymer Materials (Dong Hua University). ** To whom correspondence should be addressed. E-mail: [email protected]

856

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

performance of PVA-PES composite membrane was investigated in detail. 2 2.1 EXPERIMENTAL Materials

2.3

Membrane characterization

Poly(vinyl alcohol) (PVA), with an average degree of polymerization of 175050, was purchased from Tianjin Kewei Chemical Co. (Tianjin, China). Polyethersulfone (PES) (catalogue no. 6020P, Mw = 29000) was purchased from BASF Co. (Germany) and dried at 110 C for 12 h before use. Glutaraldehyde [GA, 50% (by mass)], HCl and absolute ethanol were purchased from Tianjin Guangfu Fine Chemicals Co. (Tianjin, China). All the chemicals were of analytical grade and used without further purification. Doubledistilled water was used throughout the experiments. 2.2 Membrane preparation

2.3.1 Scanning electron microscope (SEM) The cross-section and surface morphology of as-prepared composite membranes and the element distribution of the membrane were observed by Philips XL-30M scanning electron microscope instrument operated at 10 kV after being freeze-fractured in liquid nitrogen and sputtered with gold. 2.3.2 Static contact angle measurement The static contact angles of water on the membranes were measured by the sessile drop method using a JC2000C Contact Angle Meter (Powereach Co., Shanghai, China) at room temperature. Water droplets (about 5 mm in diameter) were dropped carefully onto the sample surface. The average contact angle was obtained by measuring the same sample at 12 different sites, and then averaging. The error of measurement for each sample was around 5%. 2.3.3 X-ray photoelectron spectroscopy (XPS) analysis X-ray photoelectron spectroscopy patterns of the membranes were recorded using a PHI-1600 diffractometer (American PE Co.) with symmetric reflection geometry. Survey spectra were collected over a range of 0-1100 eV and high-resolution spectra of C 1s peak was also collected. The elemental content on the surface of the membrane was obtained from the XPS analysis. 2.3.4 Annealing treatment The as-prepared membranes were allowed to dry completely at room temperature. The membranes were then thermally annealed under specific temperature for 2 h in an oven prior to the pervaporation performance evaluation. 2.3.5 Pervaporation experiment The pervaporation experiments were carried out in the laboratory using pervaporation apparatus as previously reported and shown in Fig. 1 [23]. The feed solution [90% (by mass) ethanol solution] was supplied continuously across the surface of a membrane by pumping. The effective membrane area was about S = 25.6 cm2. The downstream pressure was maintained at 300 Pa by a vacuum pump. The operating temperature

PVA-PES composite membranes were prepared by phase inversion method [22]. PVA (A) solution [10% (by mass)] and PES (B) solution [25% (by mass)] were prepared by dissolving them separately in DMSO solvent at 80 C. Then, A, B solutions and dimethyl sulfoxide (DMSO) aqueous solution were mixed at different ratio and filtered to remove air bubble until uniformly mixed. Then, this casting solution was poured onto a horizontal glass plate, spread with a casting knife and immediately immersed into a coagulation bath for solidification. The surface crosslinking was obtained by immersing membranes into 2.5% (by mass, GA), 1 M HCl solution for 2 h. Then, the membrane was washed thoroughly with deionized water to remove the residues, and dried under set temperature 70, 90, 110, 130, 150 C. The thickness of prepared membranes was about 180 m and there morphology was shown in 3.1. The membranes were designated as PVA(X)-PES(Y) and the compositions were presented in Table 1, where X, Y was defined as mass fraction of PVA, PES in membrane casting solutions, respectively.

Table1

The composition of membrane casting solution for PVA-PES composite membranes

Composition/g A PES 0 5 7.5 10 15 B 36 34 33 32 30 DMSO Water 14 11 9.5 8 5 0 0 0 0 0 Mass fraction of PVA in dry membrane/% 0 5.56 8.33 11.11 16.67

Membrane

PVA(1)-PES(17) PVA(1.5)-PES(16.5) PVA(2)-PES(16) PVA(3)-PES(15)

Figure 1 tion cell

The schematic diagram of membrane permea-

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

857

was 80 C and the feed flow rate was 60 Lh1. The permeated vapor was collected in a cold trap with liquid nitrogen and was weighed after achieving ambient temperature to gain the permeation flux, J, as defined as follows: Q J= (1) ST where Q is the permeate collected in time interval T, and S is the effective membrane area. Then, the collected permeate was analyzed by HP4890 gas chromatography (GC) equipped with a thermal conductivity detector (TCD) and a column packed with GDX103 (Tianjin Chemical Reagent Co.,

China). The separation factor was calculated according to P /P = W E (2) FW / FE where FW and FE, PW and PE are the mass fractions of water and ethanol in the feed solution and the permeate, respectively. 3 RESULTS AND DISCUSSION

3.1 Membrane morphology of PVA-PES composite membrane

SEM images in Fig. 2 showed that PVA-PES

(a) Ethanol

(b) Isopropyl alcohol

(c) Butanol

(d) Isopropyl alcohol Figure 2 FESEM images of the cross-section and surface of PVA(3)-PES(15) membranes prepared in different coagulation bath

858

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

Figure 3

Schematic of the fabrication process and ideal structure of PVA-PES composite membrane

membranes exhibited the morphology of a typical asymmetric structure and their surface morphology changed greatly with varied coagulation bath. The thickness of compact layer for the composite membrane was about 1.2 m when isopropyl alcohol as coagulation bath. Table 2 showed the solubility parameters of DMSO, water, ethanol, isopropyl alcohol, butanol, PVA and PES which were calculated based on a simple mixing rule. The closer the solubility parameters of these two organic solvents were, the faster the solvent exchange between them. And the morphology of the composite membranes depended strongly on the solvent exchange rate when contacting the coagulation bath. As shown in Fig. 2, sponge-like porous top layers and finger-like porous bottom layer were formed when isopropyl alcohol and butanol were used as coagulation medium, while porous surface and fingerlike porous support layer were formed when ethanol was as coagulation medium. Similar phenomenon was also reported in the literature [24, 25].
Table 2 Solubility parameters of coagulation medium
Solubility parameter ()/MPa1/2 27.4 47.8 26.0 23.8 23.3 25.2 20.3 DMSO water ethanol isopropyl alcohol butanol PVA PES

support layer with a dense top layer to be used in subsequent experiments without special explanation. Surface segregation method was employed to fabricate the integral pervaporation membranes. To interpret the SEM photos, a schematic fabrication process and structure of PVA-PES composite membrane was shown in Fig. 3. PVA and PES were physically blended in the membrane casting solution. Upon immersion into the aqueous coagulant bath, PVA spontaneously segregated to all polymer-water interfaces of the forming membrane and self-organized, creating a hydrophilic brush-like layer. The integral structure could improve the interaction of polymer PVA and PES and then enhanced the membrane stability.
3.2 Water contact angle measurement

Coagulation medium

Water contact angles were introduced to evaluate the effect of the addition of PVA on the surface hydrophilicity, and the results were shown in Fig. 4. PES membrane had the highest contact angle, corresponding to the lowest hydrophilicity, and the water contact angle for PVA-PES membranes monotonously decreased with increasing in PVA content. Furthermore,

Since the sponge-like layer had the largest mass transfer resistance [26]. To obtain high permeation flux, membranes with finger-like porous were chosen as support layers for construction of composite membranes. In this study, Isopropyl alcohol was chosen as the coagulation medium to obtain a finger-like porous

Figure 4 PVA-PES composite membrane water contact angle data phase inversion; solvent evaporation

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

859

it could be seen that the water contact angles of PVA-PES membranes prepared through phase inversion were lower than the solvent evaporation ones, which indicated that the hydrophilic PVA chains were segregated to the membrane surface. Above all, it could be considered as supporting evidence of successful construction of a hydrophilic membrane surface by surface segregation of PVA. Increased the loading of PVA in the membrane accordingly increased the hydrophilicity of PVA-PES composite membrane.
3.3 XPS

XPS analysis was employed to quantitatively determine the chemical composition of the membrane surface. And the results were shown in Fig. 5. The S element concentration on membrane surface was 3.5%. The near-surface coverage of PVA () was calculated using the Equation (3) [20]: 1 N PES S 100% (3) = N PVA
where NPES is the total mole number of C, O and S in a PES unit, NPVA is the total mole numbers of C, O and S in a PVA unit, and aS is the mole fraction of element S on membrane surface.

Figure 6 The element analysis of PVA(3)-PES(15) composite membrane cross-section left: S; middle: C; right: O

reached a plateau under the depth of about 5.6 m.

3.5 The effect of PVA concentration on pervaporation performance

Figure 5 XPS survey scan of PVA(3)-PES(15) membrane surface C 1s 72.6%; O 1s 23.9%; S 2p 3.5%

The relationship between PVA concentration and pervaporation properties of PVA-PES composite membrane was investigated over the PVA concentration ranging from 0 to 17% (by mass). The permeation flux and separation factor as a function of PVA concentration were pictured in Fig. 7. It could be seen that the separation factor increased with the increasing of PVA concentration. This could be explained by increasing PVA content promoted the hydrophilicity of the membrane, and thus facilitated the sorption process on the upstream membrane surface and enhanced the diffusion of water molecules through the membrane. Furthermore, the compact PVA chains also increased the separation factor. However, the separation factor had no obvious change after PVA concentration reached 5.6% (by mass) because it was not dependent on the PVA concentration. Meanwhile, the permeation flux increased with the PVA concentration. This could be interpreted that the increased PVA concentration was favored to the membrane absorption. Therefore, water and ethanol were absorbed much faster.

It could be calculated that for PVA(3)-PES(15) membrane, the near-surface coverage of PVA was 79.7%, suggesting that the PVA chains which contain hydrophilic OH groups were segregated to membrane surface.
3.4 Energy dispersive spectrometer (EDS)

Figure 6 showed the element distribution of C, O, and S at the cross section of the membrane. The photograph indicated that the content of element S, which existed only in PES, increased obviously with the increasing depth under membrane surface. It could be inferred that the S-free PVA chains was enriched on the membrane surface. Furthermore, the S concentration was increased in an exponential manner and then

Figure 7 The effect of PVA concentration on the pervaporation performance of the composite membrane (dried at 90 C)

860

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

3.6 The effect of annealing temperature on pervaporation performance

The structure and the morphology of a membrane are crucial to its separation efficiency. Annealing temperature, as one important effective strategy, could control the membrane morphology and then changed the membrane performance especially for long-term operation performance. The effect of annealing temperature on PVA(3)-PES(15) composite membrane performance was shown in Fig. 8. It could be seen that there were significant changes of permeation flux and separation factor with the increase of annealing temperature. This could be explained that the increase of annealing temperature resulted in the compression of the polymer chains, which made the diffusion resistance for ethanol (kinetic radius of 0.26 nm) was much higher than that for water (kinetic radius 0.13 nm) [27]. Therefore, the separation factor increased and the permeation flux decreased.

(a) Original data

(b) Normalized data Figure 9 The effect of water concentration in feed on PVA(3)-PES(15) membrane performance

Figure 8 The effect of annealing temperature on PVA(3)PES(15) composite membrane performance

3.7 The effect of water concentration in feed on pervaporation

The permeance Qi and selectivity could be calculated as follows: Qi = Ji xi i pi0,feed yi ppermeate Qwater Qethanol (4) (5)

ized data]. It could be seen that the total permeation flux, ethanol flux and water flux were all increased with the water concentration, while the separation factor had a remarkable decrease due to the membrane swelling. The variation of water flux and total flux could be explained as follows: the water content in the membrane surface increased with the increasing of water concentration, which enhanced the diffusive flux of water through the membrane. Meanwhile, the membrane swelling facilitated the total flux.
3.8 The effect of operating temperature

where Ji is mass flux of component i (gm2h1), xi is the mole fraction of the component i in the feed liquid, i is the activity coefficient of component i in the feed liquid, pi0,feed is the pure component i feed vapor pressure under feed temperature (kPa) and yi is the mole fraction of the component i in the permeate, ppermeate is the pressure under the membrane (kPa). Figure 9 described the effect of water concentration (from 6% to 42%, with a flow rate of 60 Lh1 at 353 K) in feed on performance of the PVA(3)-PES(15) composite membrane [(a): original data, (b): normal-

Figure 10 depicted the effect of operating temperature (from 323 K to 353 K) on the permeation flux and separation factor of PVA(3)-PES(15) composite membrane. A flow rate of 60 Lh1 was kept for 90% (by mass) ethanol aqueous with a downstream pressure of 300 Pa. As shown in Fig. 9, both permeation flux and separation factor increased with operation temperature. With the temperature increased from 323 to 353 K, the permeation flux increased significantly from 600 to 1600 gm2h1, and separation factor considerably increased from 30 to 65, which could be attributed to the favorable effect of temperature on diffusion coefficient and saturation pressure. Diffusion increased with temperature due to higher mobility of water and ethanol molecules. Meanwhile, the saturation vapor pressure increased with temperature and the

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

861

Figure 11 presented the Arrhenius plots of lnJp versus 1000/T for water and ethanol. A good linearity was observed in the given temperature range for water and ethanol permeation flux. The activation energy for water and ethanol in the temperature range of 323-353 K was 18.82 kJmol1 and 4.35 kJmol1, respectively. The activation energy of permeation for water was much higher than that of ethanol, indicating that the permeation rate of water increased faster than ethanol, so did the separation factor [28, 29].
3.9
(a) Original data

Operation stability test

water; ethanol; selectivity

(b) Normalized data

The as-prepared composite membrane was immersed into 90% (by mass) ethanol aqueous solution for the specific time and was then repeatedly evaluated by pervaporation process to test the membrane operation stability. The effect of immersion time on the pervaporation performance of PVA(3)-PES(15) composite membrane was shown in Fig. 12. It could be seen that the permeation flux was only slightly increased and the separation factor was only slightly decreased with the increase of immersion time. No peeling off and structural defect were found during long-term pervaporation, implying excellent operation stability for the prepared membranes.

Figure 10 Effect of operating temperature on the pervaporation performance of PVA(3)-PES(15) composite membrane

mass transport driving force for water and ethanol was enhanced accordingly. The dependence of permeation flux on temperature was fitted according to the Arrhenius equation: Ep J p = Ap exp (6) RT where Ap, Jp, Ep, R and T are the pre-exponential factor, permeation flux, apparent activation energy, gas constant and feed temperature, respectively.

Figure 12 Effect of immersion time on the pervaporation performance of PVA(3)-PES(15) composite membrane J;

CONCLUSIONS

Figure 11 Arrhenius plots of permeation flux for separation of ethanol/water mixture by the PVA(3)-PES(15) composite membrane water; ethanol

The PVA-PES composite membranes were prepared by blending polyvinyl alcohol (PVA) with polyethersulfone (PES) in the casting solution. A PES enriched porous support layer and a PVA enriched dense separation layer were formed during phase inversion process for membrane preparation. X-ray photoelectron spectroscopy (XPS) and Energy dispersive spectrometer (EDS) results confirmed the spontaneous surface segregation of hydrophilic PVA. The PVA(3)/ PES(15) membrane, which had a PVA content of 16.67% (by mass) and the surface aera coverage of 86.79% by PVA on the composite membrane surface, exhibited a flux of 297.6 gm2h1 and a separation factor of 86.1 for 90% (by mass) ethanol solution with

862

Chin. J. Chem. Eng., Vol. 19, No. 5, October 2011

15 Liu, Y., Yu, C., Lai, J., Poly(tetrafluoroethylene)/polyamide thin-film composite membranes via interfacial polymerization for pervaporation dehydration on an isopropanol aqueous solution, J. Membr. Sci., 315, 106-115 (2008). Lipatov, Y.S., Polymer blends and interpenetrating polymer networks at the interface with solids, Prog. Polym. Sci., 27, 1721-1801 (2002). Su, Y.L., Cheng, W., Li, C., Jiang, Z., Preparation of antifouling ultrafiltration membranes with poly(ethylene glycol)-graftpolyacrylonitrile copolymers, J. Membr. Sci., 329, 246-252 (2009). Shi, Q., Su, Y., Zhu, S., Li, C., Zhao, Y., Jiang, Z., A facile method for synthesis of pegylated polyethersulfone and its application in fabrication of antifouling ultrafiltration membrane, J. Membr. Sci., 303, 204-212 (2007). Fan, Z., Wang, Z., Sun, N., Wang, J., Wang, S., Performance improvement of polysulfone ultrafiltration membrane by blending with polyaniline nanofibers, J. Membr. Sci., 320, 363-371 (2008). Li, B., Zhao, W., Su, Y.L., Enhanced desulfurization performance and swelling resistance of asymmetric hydrophilic pervaporation membrane prepared through surface segregation technique, J. Membr. Sci., 326, 556-563 (2009). Wang, D., Lin, F., Wu, T.T., Lai, J.Y., Pervaporation of water-ethanol mixtures through symmetric and asymmetric TPX membranes, J. Membr. Sci. 123, 35-46 (1997). Nermen Maximous, Nakhla, G., Wan, W., Wong, K., Preparation, characterization and performance of Al2O3/PES membrane for wastewater filtration, J. Membr. Sci., 341, 67-75 (2009). Chen, J., Chen, X., Yin, X., Bioinspired fabrication of composite pervaporation membranes with high permeation flux and structural stability, J. Membr. Sci., 344, 136-143 (2009). Choi, S.H., Tasselli, F., Jansen, J.C., Effect of the preparation conditions on the formation of asymmetric poly(vinylidene fluoride) hollow fibre membranes with a dense skin, Eur. Polym. J., 46, 1713-1725 (2010). Xua, A., Yang, A.H., Young, S., Effect of internal coagulant on effectiveness of polyvinylidene fluoride membrane for carbon dioxide separation and absorption, J. Membr. Sci., 311, 153-158 (2008). Zhang, L.Z., Coupled heat and mass transfer through asymmetric porous membranes with finger-like macrovoids structure, Int. J. Heat Mass Tran., 52, 751-759 (2009). Tamai, Y., Tanaka, H., Nakanishi, K., Molecular simulation of permeation of small penetrants through membranes. 1. Diffusion coefficients, Macromolecules, 27, 4498-4508 (1994). Raisi, A., Aroujalian, A., Kaghazchi, T., A predictive mass transfer model for aroma compounds recovery by pervaporation, J. Food Eng., 95, 305-312 (2009). Wu, P., Field, R.W., England, R., Brisdon, B.J., A fundamental study of organo functionalised PDMS membranes for the pervaporative recovery of phenolic compounds from aqueous streams, J. Membr. Sci., 190, 147-157 (2001).

water at 80 C with a downstream pressure of 300 Pa. In addition, the composite membrane showed desirable long-term operation stability.
REFERENCES

16 17 1 2 3 4 5 6 7 8 9 Balat, M., Balat, H., Recent trends in global production and utilization of bio-ethanol fuel, Appl. Energy, 86, 2273-2282 (2009). Frolkova, A.K., Raeva, V.M., Bioethanol dehydration: State of the art, Theor. Found. Chem. Eng., 44, 545-556 (2010). Neel, J. Huang., R.Y.M., Ed., Introduction to Pervaporation, Pervaporation Membrane Separation Processes, Elsevier, Amsterdam (1991). Feng, X., Huang, R.Y.M., Liquid separation by membrane pervaporation: a review, Ind. Eng. Chem. Res., 36, 1048-1066 (1997). Mulder, M., Basic Principles of Membrane Technology, second edition, Kluwer Academic Publishers, Netherlands (1998). Chapman, P.D., Oliveira, T., Livingston, A.G., Membranes for the dehydration of solvents by pervaporation, J. Membr. Sci., 318, 5-37 (2008). Shao, P., Huang, R.Y.M., Polymeric membrane pervaporation, J. Membr. Sci., 287, 162-179 (2007). Ma, X.H., Xu, Z.L., Liu, Y., Sun, D., Preparation and characterization of PFSA-PVA-SiO2/PVA/PAN difunctional hollow fiber composite membranes, J. Membr. Sci., 360, 315-322 (2010). Peng, N., Chung, T.S., Chng, M.L., Aw, W.Q., Evolution of ultra-thin dense-selective layer from single-layer to dual-layer hollow fibers using novel Extem polyetherimide for gas separation, J. Membr. Sci. 360, 48-57 (2010). Zhang, L., Yu, P., Luo, Y.B., Dehydration of caprolactam-water mixtures through cross-linked PVA composite pervaporation membranes, J. Membr. Sci., 306, 93-102 (2007). Huang, R.Y.M., Pal, R., Moon, G.Y., Crosslinked chitosan composite membrane for the pervaporation dehydration of alcohol mixtures and enhancement of structural stability of chitosan/polysulfone composite membranes, J. Membr. Sci., 160, 17-30 (1999). Ma, J., Zhang, M.H., Wu, H., Mussel-inspired fabrication of structurally stable chitosan/polyacrylonitrile composite membrane for pervaporation dehydration, J. Membr. Sci., 348, 150-159 (2010). Shao, P., Huang, R., Feng, X., Anderson, W., Pal, R., Burns, C., Composite membranes with an integrated skin layer: preparation, structural characteristics and pervaporation performance, J. Membr. Sci., 254, 1-11 (2005). Yu, C.H., Kusumawardhana, I., Lai, J.Y., Liu, Y.L., PTFE/polyamide thin-film composite membranes using PTFE films modified with ethylene diamine polymer and interfacial polymerization: Preparation and pervaporation application, J. Colloid Interf. Sci., 336, 260-267 (2009). 18

19 20

21 22 23 24

10 11

25 26 27 28 29

12 13

14