Ind. Eng. Chem. Res. 2002, 41, 531-537
531
MATERIALS AND INTERFACES Ethylene Propylene Diene Monomer (EPDM) Membranes for the Pervaporation Separation of Aroma Compound from Water R. Y. M. Huang,* G. Y. Moon, and R. Pal Department of Chemical Engineering, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
Ethylene propylene diene monomer (EPDM) rubber membranes were fabricated in thin-film composite form to separate an aroma compound, ethyl butyrate, from water and to study the effects of operating parameters on the pervaporation performance. A resistance-in-series model was applied to the pervaporation results to estimate the transport of the permeant. The partial flux of ethyl butyrate increased with the feed concentration. It was also found that the total flux decreased with increasing permeate pressure, whereas the organic flux increased. The effect of concentration polarization on the pervaporation performance was observed with changes in the feed flow rate. It was observed that the overall mass transfer coefficient was a function of the feed flow rate at fixed feed concentration. Introduction Among the major applications of pervaporation membrane processes, there has been a proliferation of papers dealing with dehydration. However, organic separation from organic/water mixtures as an application area is relatively less frequently addressed. Specifically, organic separation from water is important in the following two areas: alcohol or aroma compound separation and volatile organic compound (VOC) separation. The removal of VOCs from wastewater streams is becoming an important application of the pervaporation process. Traditional separation methods include packed-tower air stripping, adsorption, distillation, and oxidation, as well as the relatively new membrane distillation process.1 However, a major hurdle limits pervaporation commercialization, namely, the lack of proper membrane materials for this application. There has also been growing research interest in the application of pervaporation to biotechnology in areas such as aroma separation and fruit juice concentration.2-4 Generally, hydrophobic elastomers such as poly(dimethylsiloxane) (PDMS)5,6 and polyurethane,7,8 as well as block copolymers such as styrene-block-styrene (SBS)9 and acrylate,10 have been extensively investigated for applications in VOC removal. Because pervaporation separation occurs through sorption and diffusion processes, it was natural that researchers tried to increase the sorption selectivity of PDMS to organic compounds through the incorporation of hydrophobic fillers (such as zeolite, silicas, and carbon blacks)11,12 and the grafting of fluoroalkyl methacrylate onto PDMS.13,14 For alcohol or aroma compound separation from water, many hydrophobic materials used for VOC removal have been studied. Among them, polyether-block-amide (PEBA),2,3,15 polyoctylmethylsiloxane (POMS),4 and zeolite-filled PDMS16 were the materials that showed promise. However, in the case of alcohol separation, it * Corresponding author: R. Y. M. Huang. E-mail: ryhuang@ engmail.uwaterloo.ca.
Figure 1. Structure of EPDM.
was recognized that PDMS is not sufficiently permselective because alcohol compounds are generally not very hydrophobic compared to aroma compounds or VOCs. In a study of acetic acid/water separation, Lu et al.17 found that the advantage of silicalite addition into PDMS could be augmented by increasing the operating temperature because the enhanced thermal energy lessens the kinetic limitation on transport of silicalite in the membrane matrix. The effects of incorporated fillers on the sorption and diffusion of aroma compounds were studied by Vankelecom et al.12 Ethyl butyrate is a colorless compound giving rise to a fruit odor reminiscent of pineapple that is widely used in flavors of types such as pear, peach, apricot, and cherry for foods, beverages, wines, and tobacco. It is used as a model aroma compound because of its importance and moderate hydrophobicity. Ethylene propylene diene monomer (EPDM) rubber contains ethylidene norbornene (ENB) as a diene comonomer inserted in the chains. It is a terpolymer of ethylene and propylene having a saturated polymer backbone that gives it excellent resistance to ozone (Figure 1). It finds wide application in the roofing and roller industries. In the current study, we investigated EPDM rubber as a membrane material for the separation of the model aroma compound ethyl butyrate (EB) from water. In addition, the mass transport of the aroma compound through the EPDM membrane was investigated and correlated using a resistance-in-series model. In previous studies,18-20 EPDM was mainly used to separate volatile organic compounds from aqueous solution.
10.1021/ie010246s CCC: $22.00 © 2002 American Chemical Society Published on Web 01/11/2002
532
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002
x ) 0, C ) Cm i x ) δ, C ) Cpi Ji ) Di
p (Cm i - Ci ) δ
(4)
where δ is the membrane thickness. The permeation flux in terms of the overall mass transfer coefficient can be expressed as
Ji ) Ki(Cbi - Cpi )
(5)
Equations 4 and 5 can be simplified into eqs 6 and 7, respectively, when Cpi ≈ 0 at low permeate pressure
Figure 2. Diagram of resistance-in-series model.
Resistance-in-Series Model for the Estimation of Transport Phenomenon A convenient model for describing the mass transport of a component is the resistance-in-series model (Figure 2), in which the overall resistance to mass transport (Ro, s/m) is written as the sum of the boundary-layer mass resistance (RLi), the membrane resistance (Rmi), and the permeate resistance (Rpi ≈ 0)
Ro ) RLi + Rmi )
1 1 1 ) + Ki kLi kmi
(1)
where Ki is the overall mass transfer coefficient (m/s), kLi is the boundary-layer mass transfer coefficient (m/s), and kmi is the membrane mass transfer coefficient. Previously, several research groups used the resistance-in-series model to interpret pervaporation data.9,18,19,21-25 The permeation flux of component i through the diffusion boundary layer on the feed stream can be expressed as
Ji ) kLi(Cbi - C/i )
(2)
where Ji is the permeation flux [kmol/(m2 s)] of component i and Cbi and C/i are the concentrations of component i in the bulk feed and at the interface with the membrane, respectively. Equation 2 can be rewritten in mole fraction, instead of concentration, form as
Ji ) kLiFL(xbi - x/i )
(3)
where x is the mole fraction and FL is the density of water, because the concentration of component i is assumed to be very low in this derivation. Note that Cbi is equal to the product of FL and xbi. The permeation flux of component i through the membrane can be expressed using Fick’s first law as
J ) -D with the boundary conditions
dC dx
DiCm i Ji ) δ
(6)
Ji ) KiCbi
(7)
where Ki is the overall mass transfer coefficient. From eqs 2 and 7, one can derive the equation
Ji ) Ki
(
Ji + C/i kLi
)
(8)
When the equilibrium sorption is assumed to be linear between the concentrations of component i in the interface and in the membrane, the partition coefficient, which is commonly obtained by the sorption experiments, can be defined as follows
H)
Cm i C/i
(9)
From eqs 6 and 9, one can write the intermediate equation
C/i )
J iδ HDi
(10)
Substituting eq 10 into eq 8 then gives
1 1 δ ) + Ki kLi DiH
(11)
or
1 1 δ ) + Ki kLi Pi where kLi is the liquid boundary-layer mass transfer coefficient () DLi/t) and Pi is the permeability of the permeant i () DiH). This is the overall mass transfer coefficient derived from the resistance-in-series model. Hence, the boundary-layer mass transfer resistance (the intercept of the plot) and the permeability of the permeant (the reciprocal slope) in addition to the membrane resistance can be obtained from a plot of the permeation flux data vs the membrane thickness. Sometimes, it is convenient to use the partial vapor pressure difference in expressing the permeation fluxes. The flux in the permeate stream can then be written
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002 533 Table 1. Analysis of EPDMs Used in This Study
Vistalon-8800 Vistalon-8609 Kelton-514
wt % of ethylene in ethylene/propylene
wt % of ENB (diene)
Mw (g/mol)
54 62 52
10 8 8
113 373 94 944 84 168
as
Ji ) kv(p/i - pi)
(12)
The resistance-in-series model derived in terms of the pressure difference can be found in the work of Ji et al.21 However, in this study, eq 11 was used to estimate the experimental results. Experimental Section Materials. The EPDMs used in this experiment are listed in Table 1. Polyacrylonitrile (MW 150 000) was obtained from Polysciences, Inc. (Warrington, PA). N,NDimethylformamide was purchased from Aldrich (St. Louis, MO). Ethyl butyrate (or ethyl butanoate) was a product of Aldrich. Ethylene glycol (monomethyl ether) from Sigma (St. Louis, MO) was used as received. A nonwoven polyester fabric donated by Veratec BBA Nonwovens (Ham Lake, MN) was used as the backing material for the composite membrane. PVDF ultrafiltration membranes from Millipore were utilized as porous substrates in most of the experiments. Water was deionized and distilled before use. Membrane Preparation. Porous polyacrylonitrile (PAN) membranes were prepared via the wet phaseinversion technique from casting solutions containing 12 wt % PAN, 83 wt % DMF, and 5% ethylene glycol as the porogen. The initial microporous PAN membranes showed a pure water permeation rate of 261 kg/(m2 h) at a transmembrane pressure of 689 kPa and an operating temperature of 22 °C. Most of the water flux tests were performed in replicate to achieve precision in the flux. For the preparation of composite membranes with different thicknesses, two EPDM solutions with concentrations of 9 and 11% in toluene were prepared. Each EPDM solution was cast onto a glass, and then the solution was allowed to evaporate for a certain time at room temperature. Before complete evaporation of the EPDM, a porous PVDF substrate was laid over the partially evaporated EPDM thin film (Figure 3), and then the composite membrane was kept in an oven for 12 h at 60 °C to remove the residual toluene. With this technique, mass transfer resistance due to the intrusion of the top layer solution into the porous substrate26 during fabrication of the composite membrane can be mitigated. Because the EPEM films were sticky after drying, it was hard to handle the thin dense membranes. The thicknesses of the membranes used were Mem 1, 32 µm; Mem 2, 76 µm; and Mem 3, 95 µm. Pervaporation. Pervaporation experiments were carried out using the apparatus and method described in a previous publication.27 The permeate was analyzed using a total organic carbon (TOC) analyzer (TOC-500) from Shimadzu, Kyoto, Japan. Scanning Electron Microscopy. Scanning electron microscopy was used to study the cross-sectional morphology of the various composite membranes and to measure the thicknesses of the membranes. Cryogenic fracturing of the membranes was done after the samples
Figure 3. Fabrication of EPDM thin-film composite membranes.
were frozen in liquid nitrogen. All specimens were coated with a conductive layer (400 Å) of sputtered gold. A Hitachi S-3000N scanning electron microscope was used to analyze the specimens at 20 kV. Results and Discussion The morphology of the composite membranes used in this study is presented in Figure 4. It is obvious from the pictures that the EPDM thin layer was properly cast on the top of the PAN and PVDF substrates. A closeup of the PAN-supported membrane (Figure 4b) reveals that no penetration of EPDM solution into the micropores of the substrates occurred as a result of the fabrication method used in this study. To determine the effect of the porous substrate on the pervaporation performance, composite membranes consisting of a top layer of about 76-µm in thickness and four different substrates were prepared and tested for aqueous aroma solutions with EB feed concentrations of 100-400 ppm. Porous PMMA and PVDF substrates were commercial products having 0.22- and 0.45-µm pore diameters, respectively. PAN and PSf substrates were prepared in our laboratory by the wet-phase inversion technique and had water fluxes of 261 and 154 kg/(m2 h), respectively, at a transmembrane pressure of 100 psi and an operating temperature 22 °C. The apparent differences in pore size and surface porosity were expected among the four different substrates. In Figure 5, the membrane supported by the PMMA substrate revealed the lowest EB flux, followed by the membrane supported by the PSf substrate. This result suggests that the EB flux is a function of the substrate porosity and pore diameter as the EB fluxes of the membranes supported by high-water-flux and largepore-diameter substrates are generally greater than those of the opposite. A general tradeoff between the EB flux and the separation factor is observed in Figures 5 and 6. On the basis of this experiment, it is believed that substrates affect the pervaporation performance of EPDM composite membranes in terms of the surface porosity rather than the type of material. Note that PMMA and PAN are relatively hydrophilic, whereas PVDF and PSf are relatively hydrophobic. It should be pointed out that, for the remainder of the experiments, the substrates were porous PVDF membranes, although the PAN substrates exhibited a comparable performance. From this experiment, it was also found that the EPDM rubber was reasonably selective for the aroma compound ethyl butyrate, although the organic flux of the EPDM membrane was slightly lower than those of PEBA and PDMS membranes investigated in our laboratory,3 as well as that of an EPDM membrane used
534
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002
Figure 5. Effect of substrate on the partial flux of ethyl butyrate (Mem 2 and Vistalon-8609).
Figure 6. Effect of substrate on separation factor (Mem 2 and Vistalon-8609).
Figure 4. SEM images of the composite membranes: (a) PANsupported membrane, (b) closeup of PAN-supported membrane, and (c) PVDF-supported membrane.
for VOC separation.21 Note that the membrane studied in this work is relatively less hydrophobic according to contact angle measurements than PEBA and PDMS. It is believed that EB is relatively more hydrophilic than VOCs (1 mL of EB is clearly soluble in 3 mL of 60% ethanol solution, unlike VOCs). Figure 7 shows the ethyl butyrate (EB) fluxes of three different composite membranes supported by porous PVDF substrates as a function of feed concentration. The EB fluxes increase linearly with increasing feed concentration. The overall mass transfer coefficients for
Figure 7. Effect of feed concentration on the aroma flux in EPDM membranes (Kelton-514) at 30 °C.
different membrane thicknesses, Ki, were obtained from the slopes of the flux vs concentration plots by means of eq 6. It was observed that the EB flux becomes larger with decreasing membrane thickness because the diffusion resistance of EB through the membrane is mitigated by the reduction of the membrane thickness. The water and organic fluxes as functions of the reciprocal membrane thickness are presented in Figure
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002 535
Figure 8. Water and organic fluxes for Vistalon-8609 EPDM membranes at a feed concentration of 600 ppm and a temperature of 30 °C.
Figure 9. Mass transfer resistances as a function of EPDM membrane thickness for Vistalon-8609 and Kelton-514.
8. The water flux, which is independent of the feed concentration, increases with decreasing membrane thickness. The fact that the water flux is independent of the feed concentration suggests that water permeation is not affected by concentration polarization in the case of the removal of small amounts of organic species from a water stream. The EB flux for EPDM membranes with thicknesses in the range of 32-96 µm is shown to increase almost linearly with reciprocal membrane thickness. The overall mass transfer resistance calculated from the plot of the organic flux vs the feed concentration is given in Figure 9 as a function of the membrane thickness. The permeabilities of the aroma compound obtained from the inverse slopes are slightly larger for the Kelton-514 EPDM membrane than for the Vistalon-
Figure 10. Effect of feed flow rate on EB flux (Vistalon-8609 EPDM membranes at 30 °C).
8609 EPDM membrane. This can be attributed to the higher propylene content in the Kelton-514 copolymer than in Vistalon-8609. Note that polypropylene is more hydrophobic than polyethylene. The mass transfer resistance values are listed in Table 2. From this table, it can be seen that the membrane resistance increases with increasing membrane thickness. The ratio of the membrane mass transfer coefficient to the liquid boundary-layer mass transfer coefficient, kmi/kLi, which represents the relative importance of hydrodynamics in the separation of the small amount of organic compound, increases with decreasing membrane thickness. From this result, it can be concluded that proper strategies to avoid any concentration polarization have to be taken into consideration to achieve the maximum performance with a given membrane material and membrane thickness for the separation of a small amount of an organic compound. The effect of feed flow rate on organic flux is shown in Figures 10 and 11. The organic flux increases with increasing feed flow rate. This suggests that concentration polarization plays an important role in the separation of small amounts of organic compounds using EPDM membranes. From Figure 10, it can be concluded that the concentration polarization depends on the feed concentration. When the feed concentration is 100 ppm, the change in the EB flux with the flow rate is not significant compared to the effect at other feed concentrations. However, the EB flux changes significantly with increasing feed concentration when the feed concentration is above 100 ppm. For feed concentrations ranging from 200 to 600 ppm, the EB flux is almost constant above a flow rate of 1.4 L/min. The separation factor increases with the feed flow rate, as shown in Figure 11. This is due to the reduced concentration polarization at higher flow rate. The overall mass transfer resistance was calculated from the experimental data and was plotted as a function of the feed flow rate in Figure 12. In this
Table 2. Overall Mass Transfer Resistance, Boundary-Layer Resistance, and Membrane Resistance for EPDM Membranes at 30 °C Vistalon-8609 Kelton-514
thickness (µm)
Ki (m/s)
1/Ki (s/m)
1/kLi (s/m)
1/km (s/m)
km/kLi
32 76 95 32 81 96
7.78 × 10-10 4.44 × 10-10 3.61 × 10-10 1.53 × 10-9 6.39 × 10-10 5.61 × 10-10
1.29 × 109 2.25 × 109 2.77 × 109 6.54 × 108 1.57 × 109 1.78 × 109
5.28 × 108 5.28 × 108 5.28 × 108 8.97 × 107 8.97 × 107 8.97 × 107
745 600 1 770 800 2 213 500 572 800 1 449 900 1 718 400
7.08 × 102 2.98 × 102 2.38 × 102 1.57 × 102 6.19 × 101 5.22 × 101
536
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002
Figure 11. Effect of feed flow rate on separation factor (Vistalon8609 EPDM membranes at 30 °C).
Figure 13. Effect of permeate pressure on the (a) total flux and (b) EB (ethyl butyrate) flux through EPDM membranes (Vistalon8800) at 30 °C.
Figure 12. Overall mass transfer resistance as a function of feed flow rate for Kelton-514 EPDM membranes at 30 °C.
experiment, the membrane resistance was assumed to be constant over the explored range of feed flow rates. From the linear decrease of the overall mass transfer resistance, it can be concluded that the liquid boundarylayer resistance can be significantly mitigated with an increase in the feed flow rate. Figure 13 shows the effects of the downstream pressure on the total flux and on the EB flux. The total permeation flux decreases with increasing downstream pressure as expected. Because of the reduced driving force, transport of the permeants was hindered. However, it is quite interesting that the EB flux in Figure 13b and the separation factor in Figure 14 increase with increasing permeate pressure. It is clear that only water transport was greatly reduced with decreasing driving force and that the transport of EB was not affected. When water molecules are excluded from the hydrophobic EPDM membrane, EB molecules with a higher affinity for EPDM replace the transport channel for water molecules and swell the membrane. The larger EB flux and the better separation factor can be rationalized according to this mechanism. In addition, the higher EB content in the feed solution makes the membranes more swollen and facilitates the transport of EB. However, the concurrent rapid transport of water
Figure 14. Effect of permeate pressure on separation factor through EPDM membranes (Vistalon-8800) at 30 °C.
molecules lowers the separation factor with increasing EB content in the feed solution. The effect of the feed temperature on the EB flux was investigated as shown in Figure 15. The temperature was increased from 15 to 35 °C. The reason that the feed temperature of 15 °C was explored is because the odor of the aroma compound can be preserved at lower operating temperature in the pervaporation separation of aroma. The EB flux increases with temperature because the fast thermal motion of the polymer chain results in more available free volume per unit time. The larger flux of the Vistalon-8609 membrane than of the Kelton-514 membrane can be attributed to the larger portion of ethylene, which makes the membrane less stiff.
Ind. Eng. Chem. Res., Vol. 41, No. 3, 2002 537
Figure 15. Effect of feed temperature on EB flux through two different EPDM membranes (membrane thickness of 76 µm).
Conclusions Various EPDM membranes having thin-film composite form were studied for the separation of ethyl butyrate, an aroma compound, from a water stream and were shown to be selective for the aroma compound. The EPDM membrane having the higher propylene content (Kelton-514) is more permselective to ethyl butyrate. However, because of the greater rigidity of propylene, the total flux of the Kelton-514 membrane was lower than that of the Vistalon-8609 EPDM membrane. A resistance-in-series model was used to calculate the mass transfer coefficient in the separation of an aroma compound from water. It was observed that the existence of liquid boundary-layer resistance in the separation of small amounts of ethyl butyrate was important, as was the membrane layer resistance. In addition, the effects of various operating parameters such as the feed flow rate, downstream pressure, and operating temperature were also investigated in terms of the organic flux and were shown to be significant in the separation performance of EPDM membranes. Literature Cited (1) Zander, A. K.; Semmens, M. J.; Narbaitz, R. M. Removing VOCs by membrane stripping. J. Am. Water Works Assoc. 1989, 81, 76. (2) Baudot, A.; Marin, M. Dairy aroma compounds recovery by pervaporation. J. Membr. Sci. 1996, 120, 207. (3) Sampranpiboon, P.; Jiraratananon, R.; Uttapap, D.; Feng, X.; Huang, R. Y. M. Pervaporation separation of ethyl butyrate and isopropanol with polyether block amide (PEBA) membranes. J. Membr. Sci. 2000, 173, 53. (4) Scha¨fer, T.; Bengtson, G.; Pingel, H.; Bo¨ddeker, K. W.; Crespo, J. P. S. G. Recovery of aroma compounds from a winemust fermentation by organophilic pervaporation. Biotechnol. Bioeng. 1999, 62, 412. (5) Almquist, C. B.; Hwang, S.-T. The permeation of organophosphorus compounds in silicone rubber membranes. J. Membr. Sci. 1999, 153, 57. (6) Yeom, C. K.; Kim, H. K.; Rhim, J. W. Removal of trace VOCs from water through PDMS membranes and analysis of their permeation behaviors. J. Appl. Polym. Sci. 1999, 73, 601. (7) Hoshi, M.; Ieshige, M.; Saitoh, T.; Nakagawa, T. Separation of aqueous phenol through polyurethane membranes by pervaporation. II. Influence of diisocyanate and diol compounds and crosslinker. J. Appl. Polym. Sci. 1999, 71, 439.
(8) Hoshi, M.; Ieshige, M.; Saitoh, T.; Nakagawa, T. Separation of aqueous phenol through polyurethane membranes by pervaporation. III. Effect of the methylene group length in poly(alkylene glycols). J. Appl. Polym. Sci. 2000, 76, 654. (9) Dutta, B. K.; Sikdar, S. K. Separation of volatile organic compounds from aqueous solutions by pervaporation using S-B-S block copolymer membranes. Environ. Sci. Technol. 1999, 33, 1709. (10) Hoshi, M.; Saitoh, T.; Toshioka, C.; Higuchi, A.; Nakagawa, T. Pervaporation separation of 1,1,2-trichloroethane-water mixture through cross-linked acrylate copolymer composite membranes. J. Appl. Polym. Sci. 1999, 74, 983. (11) Vankelecom, I. F. J.; Kinderen, J. D.; Dewitte, B. M.; Uytterhoeven, J. B. Incorporation of hydrophobic porous fillers in PDMS membranes for use in pervaporation. J. Phys. Chem. B 1997, 101, 5182. (12) Vankelecom, I. F. J.; De Beukelaer, S.; Uytterhoeven, J. B. Sorption and pervaporation of aroma compounds using zeolitefilled PDMS membranes. J. Phys. Chem. B 1997, 101, 5186. (13) Mishima, S.; Nakagawa, T. Plasma grafting of fluoroalkyl methacrylate onto PDMS membranes and their VOC separation properties for pervaporation. J. Appl. Polym. Sci. 1999, 73, 1835. (14) Mishima, S.; Nakagawa, T. Sorption and diffusion of volatile organic compounds in fluoroalkyl methacrylate-grafted PDMS membrane. J. Appl. Polym. Sci. 2000, 75, 773. (15) Djebbar, M. K.; Nguyen, Q. T.; Cle´ment, R.; Germain, Y. Pervaporation of aqueous ester solutions through hydrophobic poly(ether-block-amide) copolymer membranes. J. Membr. Sci. 1998, 146, 125. (16) Adnadjevlc, B.; Jovanovic, J.; Gajinov, S. Effect of different physicochemical properties of hydrophobic zeolites on the pervaporation properties of PDMS membranes. J. Membr. Sci. 1997, 136, 173. (17) Lu, S.-Y.; Chiu, C.-P.; Huang, H.-Y. Pervaporation of acetic acid/water mixtures through silicalite-filled poly(dimethylsiloxane) membranes. J. Membr. Sci. 2000, 176, 159. (18) Nijhuis, H. H.; Mulder, M. H. V.; Smolders, C. A. Removal of trace organics from aqueous solutions. Effect of membrane thickness. J. Membr. Sci. 1991, 61, 99. (19) Meuleman, E. E. B.; Bosch, B.; Mulder, M. H. V.; Strathmann, H. Modeling of liquid/liquid separation by pervaporation: Toluene from water. AIChE J. 1999, 45, 2153. (20) Pereira, C. C.; Habert, A. C.; Nobrega, R.; Borges, C. P. New insight in the removal of diluted volatile organic compounds from dilute aqueous solution by pervaporation process. J. Membr. Sci. 1998, 138, 227. (21) Ji, W.; Sirkar, S. K.; Hwang, S.-T. Modeling of multicomponent pervaporation for removal of volatile organic compounds from water. J. Membr. Sci. 1994, 93, 1. (22) Ji, W.; Sirkar, S. K.; Hwang, S.-T. Sorption, diffusion and permeation of 1,1,1,-trichloroethane through adsorbent-filled polymeric membranes. J. Membr. Sci. 1995, 103, 243. (23) Wijmans, J. G.; Athayde, A. L.; Daniels, R.; Ly, J. H.; Kamaruddin, H. D.; Pinnau, I. The role of boundary layers in the removal of volatile organic compounds from water by pervaporation. J. Membr. Sci. 1996, 109, 135. (24) Coˆte´, P.; Lipski, C. Mass Transfer Limitations in Pervaporation for Water and Wastewater Treatment. In Proceedings of the 3rd International Conference on Pervaporation Processes in the Chemical Industry; Nancy, France, Sep 1988; Bakish Materials Corp.: Englewood, NJ, p 19. (25) Raghunath, B.; Hwang, S.-T. Effect of boundary layer mass transfer resistance in the pervaporation of dilute organics. J. Membr. Sci. 1992, 65, 147. (26) Vankelecom, I. F. J.; Moermans, B.; Verschueren, G.; Jacobs, P. A. Intrusion of PDMS top layers in porous supports. J. Membr. Sci. 1999, 158, 289. (27) Huang, R. Y. M.; Pal, R.; Moon, G. Y. Cross-linked chitosan composite membrane for the pervaporation dehydration of alcohol mixtures and enhancement of structural stability of chitosan/ polysulfone composite membranes. J. Membr. Sci. 1999, 160, 17.
Received for review March 20, 2001 Revised manuscript received September 26, 2001 Accepted October 1, 2001 IE010246S
