Pervaporation Of Toluenealcohol Mixtures Through A Coextruded

386

Ind. Eng. Chem. Res. 2003, 42, 386-391

Pervaporation of Toluene/Alcohol Mixtures through a Coextruded Linear Low-Density Polyethylene Membrane J. P. G. Villaluenga,* M. Khayet, P. Godino, B. Seoane, and J. I. Mengual Department of Applied Physics I, Faculty of Physics, Complutense University of Madrid, 28040 Madrid, Spain

The pervaporation characteristics of toluene/methanol, toluene/ethanol, and toluene/propanol mixtures through a linear low-density polyethylene membrane were investigated at different feed compositions. These characteristics were obtained from quantities such as the swelling ratio, permeation rate, and selectivity. In all cases toluene permeates preferentially through the membrane, with a separation factor up to 66 and fluxes varying between 0.1 and 1.4 kg/ m2‚h. Predictions based on the Flory-Huggins theory have shown that toluene is preferentially sorbed by the membrane. The pervaporation flux increases and the selectivity decreases with the toluene content in the feed. The experimental results show that pure-alcohol fluxes decrease with the molecular size. Introduction Pervaporation is a membrane technique used for the separation of liquid mixtures by means of partial vaporization across a permselective membrane. Because pervaporation makes use of dense membranes, the transmembrane mass transport can be described by the solution-diffusion model widely accepted in gas separation operations.1,2 The transport process can be divided into three steps: solution of the liquid mixture at the upstream membrane surface, diffusion of the permeants through the membrane, and desorption of the permeate, in vapor form, at the downstream membrane side. However, the study of the transport in pervaporation is more complicated than that in gas separation because the concentration in the membrane is generally higher, which causes the swelling and plasticization of the polymer matrix and strong coupling effects in the sorption and diffusion processes.3,4 Coupling phenomena are difficult to measure. It is also difficult to estimate beforehand the extent of the phenomena in relation to the separation properties, as several researchers have pointed out.5,6 Nevertheless, a lot of progress has been made in modeling the sorption step according to the Flory-Huggins theory7-9 and UNIQUAC model.10 The main approaches to describe mass transfer through the membrane employ the Fick equations,11,12 the thermodynamics of irreversible processes,1,4,13 or StefanMaxwell equations.14,15 The applications of pervaporation processes for dehydration from alcoholic solutions and removal of organics from aqueous solutions have been carried out commercially for several years.16 However, there has been no application of pervaporation for the separation of organic mixtures in the chemical industry. Among these separations, the removal of methanol from anhydrous organic liquid mixtures is required in many situations such as methanol/methyl tert-butyl ether, methanol/ethylene glycol, methanol/heptane, and methanol/propanol.2,16 The literature on the separation of toluene and alcohol by pervaporation is rather scarce. Extraction of the alcohol with water and subsequent * To whom correspondence should be addressed. Tel: (34) 91 394 4454. Fax: (34) 91 394 5191. E-mail: [email protected].

distillation of the alcohol/water mixture may be used to separate these mixtures. In comparison with distillation, pervaporation is usually a more energy-saving process because the selectivity is largely improved because of the permselectivity of the membrane. The performance of the membrane plays a key role in determining the feasibility in the process. Several efforts have been made to develop new materials in order to obtain a suitable membrane for this particular organic/ organic mixture. For example, Park et al.17,18 investigated the pervaporation of methanol/toluene and ethanol/ toluene mixtures through polymer blend membranes of poly(acrylic acid) (PAA) and poly(vinyl alcohol) (PVA). It was observed that alcohols permeated preferentially through all tested membranes. The flux decreased gradually as the PVA content in the blend increased, whereas the selectivity values increased. This offered a convenient way of optimizing the separation performance of the membranes by adjusting the blend composition of the polymer solution used for the preparation of the membranes. Separation of the toluene/ethanol mixtures by pervaporation, using adsorbent-filled membranes, was investigated by Duval et al.19 Various activated carbons and two different zeolites were used in this study. For all of the carbons tested, the total flux through the membrane decreased drastically with the addition of the active carbon particles; however, the porous structure of carbon seemed to have no effect on the selectivity. The methanol removal of methanol/toluene mixtures by pervaporation using polypyrrole (PPy) membranes was also investigated by Zhou et al.20 Results showed that the membranes displayed preferential permeation of methanol and exhibited a good selectivity and an acceptable permeation rate. Huang et al.21 prepared composite chitin membranes, supported by a porous poly(ether imide) substrate, for the pervaporation separation of ethanol/toluene and methanol/toluene mixtures. The chitin was obtained by modifying chitosan to its original form chitin by the N-acetylation reaction. It was reported that the total flux, for both types of mixtures, decreased with the incorporation of an acetyl functional group due to the larger inter- or intramolecular bonding among the acetyl groups. However, the N-acetylation made the chitosan

10.1021/ie020603e CCC: $25.00 © 2003 American Chemical Society Published on Web 12/21/2002

Ind. Eng. Chem. Res., Vol. 42, No. 2, 2003 387 Table 1. Performance of Membranes in the Separation of Toluene/Alcohol Mixtures by Pervaporation polymer

alcohol

PAA-PVA MeOH EtOH PDMS EtOH EPDA EtOH PPy MeOH chitin EtOH MeOH CA MeOH PDMS MeOH CTA MeOH LLDPE MeOH EtOH

feed concn Ta Jb (wt %) (°C) (kg/m2‚h) 50 50 10 10 10 10 10 70 50 50 50 60

30 30 30 30 57.5 35 35 30 30 30 30 30

1.0 0.15 0.8 0.06 0.29 0.40 0.68 0.5 1.7 0.8 0.46 1.0

R 83c 82c 13 62 420c 126c 607c 5c 7 4c 22 27

ref 17 19 20 21 22 23 this work

a Feed temperature. b Overall flux. c Selectivity of alcohol over toluene.

more efficient for the separation of alcohol/toluene mixtures because the selectivity was increased significantly depending on the alcohol content in the feed. Bhat and Pangarkar22 have used polyimide and blends of cellulosic membranes for the pervaporative separation of methanol/toluene mixtures. All of the membranes studied were methanol selective with acceptable permeation rates. Recently, Mandal and Pangarkar23 carried out a systematic study on the pervaporation separation of methanol/toluene mixtures by using different hydrophilic and hydrophobic membranes. They showed that the hydrophilic membranes were methanol selective, whereas the hydrophobic membranes preferentially permeated toluene. Some pervaporation performance data for the separation of toluene/alcohol mixtures are given in Table 1. The aim of the present study is to investigate the pervaporation features of a coextruded linear lowdensity polyethylene (LLDPE) membrane for the separation of toluene/alcohol mixtures. The effect of the feed composition on the membrane performance was studied for toluene/methanol, toluene/ethanol, and toluene/ propanol mixtures over the entire range of mixture composition. In addition, pervaporation experiments with pure methanol, ethanol, propanol, butanol, and octanol were performed to study the influence of the nature of alcohol on the permeation through the LLDPE membrane. The swelling process of the membrane was investigated as a function of the concentration of toluene in the liquid mixture. Experimental Section Membrane. The membrane is made of 1-octene-coethylene copolymers with roughly 8% mole content of the first comonomer. The membrane, which was obtained by coextrusion, is composed of two layers of Dowlex 2247 (F ) 0.917 g/cm3), with thicknesses of 3.5 and 16.0 µm, respectively, and a third layer of Dowlex 2291 (F ) 0.912 g/cm3) of 3.5 µm thickness. Both materials are additive-free. The thicker Dowlex 2247 layer is sandwiched between the other two layers. The extrusion rate is 88 rpm for the thinner layers and 29 rpm for the thicker one. The distance between the die exit and the chill roll is 15 mm, with the temperature at these points being 543 and 293 K, respectively. Further details about the coextruded LLDPE membrane have been reported elsewhere.24 Sorption Experiments. The equilibrium swelling properties of the membrane in toluene/alcohol mixtures

Figure 1. Schematic representation of the pervaporation unit: (a) constant-temperature bath, (b) pervaporation cell, (c) manometer, (d) feed circulating pump, (e) pressure transducer, (f) cold trap, (g) venting, (h) vacuum pump.

were measured by a gravimetric method. The dry LLDPE membrane was weighted and, then, immersed in a closed bottle containing either toluene, methanol, ethanol, propanol, butanol, octanol, or a mixture of toluene and alcohol. The bottle was placed in a thermostated bath at 30 °C. After 24 h of immersion, the swollen membrane was taken out of the liquid and reweighed after the superfluous liquid was wiped out with tissue paper. The increase in weight was equal to the weight of the liquid sorbed by the membrane. Pervaporation Experiments. The vacuum-pervaporation setup used is shown in Figure 1. The effective membrane surface area was 28 cm2. The feed was circulated over the membrane sample through meandertype channels. The feed pressure was kept at 1.0 bar. The permeate stream was evacuated by a vacuum pump, and the permeate was collected alternatively in two traps, cooled by liquid nitrogen. The downstream pressure was monitored by a pressure transducer and was maintained below 10-2 mmHg in all of the experiments. The fluxes were determined by weighing the sample collected in a determined time interval. The feed and permeate compositions of all of the mixtures were determined by measuring their refractive indexes with a refractometer. The pervaporation selectivity of the membrane was studied in terms of the separation factor, R, which is defined as

R)

y1/y2 x1/x2

(1)

where x and y are the weight fractions of the relevant component in the feed and the permeate solutions, respectively. Indexes 1 and 2 refer to the more permeable component (toluene) and less permeable one (alcohol), respectively. In each run, it was observed that the concentration in the feed solution after the completion of the experiment was almost the same as that of the initial feed solution. So, it could be considered constant throughout the experiment. After the initial transient operation, at least three samples were collected at each run to determine the total flux, as well as the composition of the permeate. The alternative use of two cold traps allowed the permeate to be continuously sampled without interrupting the experiment.

388

Ind. Eng. Chem. Res., Vol. 42, No. 2, 2003

Table 2. Properties of Different Compounds compound

formula

boiling point (°C)

molar volume (cm3/mol)

toluene methanol ethanol propanol butanol octanol

C7H8 CH4O C2H6O C3H8O C4H10O C8H18O

111 64 78 82 117 194

106 40 58 77 92 158

Table 3. Pure-Component Sorption and Pervaporation Data compound

J (10-6 kg/m2‚s)

S (g/100 g)

toluene methanol ethanol propanol butanol octanol

3.89 ( 0.01 0.308 ( 0.012 0.273 ( 0.011 0.265 ( 0.013 0.251 ( 0.011 0.249 ( 0.012

6.5 1.4 2.2 0.7 <0.1 <0.1

χip 2.1 3.3 2.9 3.8

δ (MPa1/2)

∆δ (MPa1/2)

18.2 29.7 26.5 24.4 23.1 21.2

2.2 13.7 10.5 8.4 7.1 5.2

The pervaporation experiments were conducted at a feed temperature of 30 °C using, as the feed, either toluene, methanol, ethanol, propanol, butanol, octanol, or binary mixtures of toluene/methanol, toluene/ethanol, and toluene/propanol. Toluene forms an azeotrope with methanol with a 0.883 mole fraction of methanol at 63.5 °C. Azeotrope also exists in toluene/ethanol and toluene/ propanol with a 0.81 mole fraction of ethanol at 76.7 °C and with a 0.773 mole fraction of propanol at 80.6 °C, respectively. Physicochemical properties of the compounds used in this work are given in Table 2. Results and Discussion The amounts of pure liquids sorbed in the coextruded LLDPE membrane are given in Table 3. The degree of swelling of the membrane is small, suggesting that plasticizing effects of liquid molecules in the LLDPE membrane may be considered as negligible. From Table 3, the amount of toluene sorbed was 6.5 g/100 g of dry membrane. The degree of swelling of the membrane was considerably reduced when the feed was methanol, ethanol, or propanol, ranging from 0.7 to 2.2%. When the feed consisted of pure butanol or octanol, it was not possible to get precise experimental data of the sorption of butanol and octanol in the membrane because they were very low (less than 0.1/100 g of membrane). A common approach, for describing polymer/liquid interactions, is the solubility parameter model developed by Hildebrand and Scott,4 which can be used to obtain a rough estimation of the affinity between the LLDPE polymer and the liquids used in this study. In the case of liquids, the solubility parameter, δ, can be easily calculated if thermodynamic data or characteristic physicochemical properties are available.4 The method of Hoftyzer-Van Krevelen25 can be applied to polymers and allows the estimation in a very simple way. The difference between the solubility parameters of a polymer and a liquid, ∆δ, is a measurement of the affinity between them. When ∆δ decreases, the affinity between the polymer and liquid increases. The solubility parameter estimated for the LLDPE polymer is δ ) 16.0 MPa1/2, whereas the parameters calculated for alcohols are given in Table 3. The solubility parameter difference between the membrane and toluene is smaller than that of membrane and alcohols. Thus, it can be hypothesized that the compatibility between the LLDPE membrane and toluene is better than that of LLDPE and alcohol.

Figure 2. Overall solubility of toluene/methanol, toluene/ethanol, and toluene/propanol mixtures in a LLDPE membrane, as a function of the liquid mixture composition.

The solubility parameter approach also predicts that the affinity between large alcohols, such as butanol and octanol, and the membrane is higher than that of small alcohols, such as methanol or ethanol. The fact that the membrane absorbs very little butanol and octanol, as compared to methanol and ethanol, suggests that the alcohol solubility in the membrane depends not only on differences in the affinity liquid/membrane but also on differences in the molar volume of the alcohols. Other researchers23 have pointed out that the LLDPE polymer is a nonpolar material due to the presence of only C-C and C-H bonds in the backbone polymer chains, whereas toluene is a nonpolar compound, and alcohols can be considered as polar compounds to a different extent. Thus, swelling results reveal that the solubility rule of “like dissolves like” seems to be obeyed because the coextruded LLDPE membrane sorbs much more toluene than alcohol. The extent of solubility or miscibility of a liquid with the polymer membrane can also be explained by the Flory-Rehner theory.26 Based on this approach, the affinity between the polymer and a solvent can be expressed in terms of one interaction parameter, χip, which, in the case of equilibrium sorption of a pure liquid in a polymer, can be calculated from the following equation:

∆µi ) ln(1 - φp) + φp + χipφp2 ) 0 RT

(2)

where φp is the volume fraction of the polymer in the polymer-liquid solution. As interaction between the polymer and the solvent increases, the amount of liquid inside the polymer membrane increases and χip decreases. Table 3 shows the interaction parameter values obtained from the sorption data of pure liquids in the LLDPE membrane. Affinity of toluene toward LLDPE is higher than that of alcohols. Hence, sorption of toluene in LLDPE is higher, as has been shown previously. Liquid sorption experiments with mixtures of toluene and methanol, ethanol, and propanol, respectively, were also performed in order to study the preferential sorption of the toluene/alcohol system in the coextruded LLDPE membrane at 30 °C. Figure 2 shows the overall solubility for mixtures of toluene/methanol, toluene/ ethanol, and toluene/propanol as a function of the composition of the external solution. A quasi-linear

Ind. Eng. Chem. Res., Vol. 42, No. 2, 2003 389

increase in the total amount of liquid sorbed in the membrane can be observed as the toluene concentration increases, which indicates that toluene has a higher solubility than do any of the alcohols. The composition of the liquid sorbed in the membrane can be estimated by a model derived from the FloryHuggins theory. At a given volume fraction of liquid in the feed, vi, the volume fraction in the membrane, φi, can be obtained by the following expressions:4

ln

() ()

()

φ1 v1 φ2 - ln ) (l - 1) ln - χ12[(v1 - v2) + φ2 v2 v2 (φ2 - φ1)] - φ3(χ13 - lχ23) (3) φ1 + φ2 + φ3 ) 1

(4)

In these equations indexes 1 and 2 refer to the liquid components, l is the ratio of molar volumes of liquids 1 (toluene) and 2 (alcohol). The polymer volume fraction, φ3, can be obtained from the overall sorption measurements. The interaction parameters between the liquids and the polymer, χ13 and χ23, were assumed to be concentration-independent and were calculated from the single liquid sorption data. The interaction parameter between toluene and alcohol, χ12, was also assumed to be concentration-independent and was calculated from the Wilson equation using data taken from the literature.27 For practical reasons, the volume of liquid i, ui, in the liquid mixture sorbed in the LLDPE membrane is used instead of the volume fraction of component i, φi, in the liquid-polymer phase. Values of ui can be easily obtained as

ui ) φi/(φ1 + φ2)

(5)

Parts a-c of Figure 3 show the predicted composition of the sorbed liquid in the swollen LLDPE membrane, as a function of the composition of the external liquid mixtures studied. The results reveal that the toluene content in the sorbed liquid was much higher than that of any alcohol; on the other hand, the relative proportion of toluene and alcohol in the membrane is nearly the same irrespective of the type of alcohol. It is worth noticing that the membrane sorbs more alcohol when they are pure than when they are in mixtures with toluene. In fact, the fraction of alcohol sorbed by the membrane is significantly reduced even with a low toluene content in the external mixture. Therefore, it can be concluded that the alcohol penetration in the LLDPE membrane is largely suppressed by the sorption of toluene. Pure-component pervaporation fluxes for toluene and five different alcohols are given in Table 3. It can be seen that the permeation rate of toluene is much higher than that of alcohols. It is generally accepted that pervaporation can be described by the solution-diffusion mechanism that occurs in the following three steps: sorption, diffusion, and desorption. According to this model and because of the low downstream pressure, the permselective properties of pervaporation of coextruded LLDPE membrane are determined by the relative solubility and diffusivity of the permeating components in the membrane. After comparison of the molar volume of the liquids, it seems reasonable to think that the high permeation rate of toluene through the LLDPE membrane can be explained by its high solubility in the membrane rather than its diffusivity.

Figure 3. Predicted composition of the sorbed liquid of (a) toluene/ methanol, (b) toluene/ethanol, and (c) toluene/propanol mixtures in a coextruded LLDPE membrane, as a function of the liquid mixture composition.

In the case of pure-alcohol pervaporation, the flux decreases as the alcohol molecular size, i.e., molar volume, decreases. A small molecular size favors the diffusivity of the permeate in the membrane. By considering the molecular size of the alcohols as their kinetic diameters, results suggest that the diffusivity of alcohol in the membrane decreases in the following order: methanol > ethanol > propanol > butanol > octanol. Based on sorption data, the solubility increases in the following order: ethanol > methanol > propanol > butanol ≈ octanol. Consequently, it seems that diffusivity and solubility increase in the same direction with respect to the alcohol molecular size.

390

Ind. Eng. Chem. Res., Vol. 42, No. 2, 2003

Figure 4. Total fluxes of toluene/alcohol mixtures through a coextruded LLDPE membrane, as a function of the feed composition.

Figure 5. Toluene selectivity for methanol/toluene, ethanol/ toluene, and propanol/toluene, as a function of the feed composition.

To study the permeation performance of the membrane, the separation of alcohol/toluene mixtures through a LLDPE membrane was evaluated in the whole composition range. Figures 4 and 5 show the variation of the total permeation rate and separation factor as a function of the feed composition, respectively, for mixtures of methanol/toluene, ethanol/toluene, and propanol/toluene. As can be observed, the flux increases and the selectivity decreases with the toluene concentration in the feed. In addition, the trend of the dependence of flux and selectivity on the feed composition is almost similar for all of the mixtures studied. As an example, in the case of methanol/toluene mixtures, the flux increases from 3.31 × 10-5 to 3.02 × 10-4 kg/m2‚s, and the selectivity decreases from 66 to 7, as the toluene volume fraction in the feed increases from 0.03 to 0.72. The partial fluxes of alcohol and toluene were calculated by using the permeate composition data. The results are given in Figure 6a-c, as a function of the feed composition. As has been pointed out previously, preferential sorption of toluene in the LLDPE membrane seems to be responsible for the difference in permeation rates between the aromatics and alcohols. This behavior agrees with previous studies,23,28 in which it was shown that preferential sorption determines the overall selectivity to a large extent for rubbery membranes, such as coextruded LLDPE. Furthermore, the partial fluxes are drastically reduced in the presence of

Figure 6. Component fluxes of (a) methanol/toluene, (b) ethanol/ toluene, and (c) propanol/toluene mixtures through a coextruded LLDPE membrane, as a function of the feed composition.

toluene, even for a low toluene content in the feed. Similar results were found by Mandal and Pargarkar23 for the pervaporation separation of methanol/toluene mixtures. They reported that the sorbed toluene tends to create a hindrance to the transport of the alcohol, through the void space in the membrane, which was the reason for the lower extent of permeation of methanol. In the present study, the coextruded LLDPE membrane is preferably soluble to toluene rather to alcohols because of the nonpolarity of the polymer chains, leading to the toluene molecules to occupy the free volume available in the polymer. This fact favors the permeation of toluene and retards the permeation of alcohol.

Ind. Eng. Chem. Res., Vol. 42, No. 2, 2003 391

Conclusions A coextruded LLDPE membrane was evaluated for the separation of toluene/alcohol mixtures. The membrane was found to be toluene-selective but with a poor permeation rate. The flux of toluene/methanol, toluene/ethanol, and toluene/propanol mixtures through the LLDPE membrane decreases and the selectivity increases with the alcohol content in the feed solution. Furthermore, a small influence of the nature of alcohol on the membrane selectivity was observed. Results also indicated that the preferential sorption of the toluene in the coextruded LLDPE membrane contributes to a great extent to the membrane selectivity. Finally, it was observed that the flux of pure alcohols through the coextruded LLDPE membrane decreases as the molecular size of the alcohol increases. Literature Cited (1) Wijmans, J. G.; Baker, R. W. The solution-diffusion model: a review. J. Membr. Sci. 1995, 107, 1. (2) Feng, X.; Huang, R. Y. M. Liquid separation by membrane pervaporation: a review. Ind. Eng. Chem. Res. 1997, 36, 1048. (3) Huang, R. Y. M.; Rhim, J. W. Separation characteristics of pervaporation membrane separation processes. In Pervaporation Membrane Separation Process; Huang, R. Y. M., Ed.; Elsevier Publishers BV: Amsterdam, The Netherlands, 1991. (4) Mulder, M. H. V. Thermodynamics principles of pervaporation. In Pervaporation Membrane Separation Process; Huang, R. Y. M., Ed.; Elsevier Publishers BV: Amsterdam, The Netherlands, 1991. (5) Drioli, E.; Zhang, S.; Basile, A. On the coupling effect in pervaporation. J. Membr. Sci. 1993, 81, 43. (6) Kedem, O. The role of coupling in pervaporation. J. Membr. Sci. 1989, 47, 277. (7) Heintz, A.; Funke, H.; Lichtenthaler, R. N. Sorption and diffusion in pervaporation membranes. In Pervaporation Membrane Separation Process; Huang, R. Y. M., Ed.; Elsevier Publishers BV: Amsterdam, The Netherlands, 1991; Chapter 10. (8) Rhim, J.-W.; Huang, R. Y. M. Prediction of pervaporation separation characteristics for the ethanol-water-nylon 4 membrane system. J. Membr. Sci. 1992, 105, 105. (9) Favre, E.; Nguyen, Q. T.; Schaetzel, P.; Clement, R.; Neel, J. Sorption of organic solvents into dense silicone membraness validity and limitations of Flory-Huggins and related theories. J. Chem. Soc., Faraday Trans. 1993, 89, 4339. (10) Heintz, A.; Stephan, W. A generalized solution-diffusion model to the pervaporation process through composite membranes. Part I. Prediction of mixture solubilities in dense active layer using the UNIQUAC model. J. Membr. Sci. 1994, 89, 143. (11) Dutta, B. K.; Ji, W.; Sikdar, S. K. Pervaporation: principles and applications. Sep. Purif. Methods 1997, 25, 131. (12) Jonquieres, A.; Clement, A. R.; Roizard, D.; Lochon, P. Pervaporative transport modeling in a ternary system: ethyl tertiary butyl ether/ethanol/polyurethaneimide. J. Membr. Sci. 1996, 109, 65.

(13) Raghunath, B.; Hwang, S. T. General Treatment of liquidphase boundary layer resistance in the pervaporation of dilute aqueous organics through tubular membranes. J. Membr. Sci. 1992, 75, 29. (14) Heintz, A.; Stephan, W. A generalized solution-diffusion model to the pervaporation process through composite membranes. Part II. Concentration polarization, couple diffusion and the influence of the porous support layer. J. Membr. Sci. 1994, 89, 153. (15) Krishna, R.; Wesseling, J. A. The Maxwell-Stefan approach to mass transfer. Chem. Eng. Sci. 1997, 52, 861. (16) Fleming, H. L.; Slater, C. S. Applications and Economics. In Membrane Handbook; Ho, W. S. W., Winston, W. S., Eds.; Chapman and Hall: New York, 1992; Chapter 10. (17) Park, H. C.; Meertens, R. M.; Mulder, M. H. V.; Smolders, C. A. Pervaporation of alcohol-toluene mixtures through polymer blend membranes of poly(acrylic acid) and poly(vinyl alcohol). J. Membr. Sci. 1994, 90, 265. (18) Park, H. C.; Meertens, R. M.; Mulder, M. H. V. Sorption of alcohol-toluene mixtures in poly(acrylic acid) and poly(vinyl alcohol) blend membranes and its role on pervaporation. Ind. Eng. Chem. Res. 1998, 37, 4408. (19) Duval, J.-M.; Folkers, B.; Mulder, M. H. V.; Desgrandchamps, G.; Smolders, C. A. Separation of a toluene/ethanol mixture by pervaporation using active carbon-filled polymeric membranes. Sep. Sci. Technol. 1994, 29, 357. (20) Zhou, M.; Persin, M.; Sarrazin, J. Methanol removal from organics mixtures by pervaporation using polypyrrole membranes. J. Membr. Sci. 1996, 117, 303. (21) Huang, R. Y. M.; Moon, G. Y.; Pal, R. N-Acetylated chitosan membranes for the pervaporation separation of alcohol/toluene mixtures. J. Membr. Sci. 2000, 176, 223. (22) Bhat, A. A.; Pangarkar, V. G. Methanol-selective membranes for the pervaporation separation of methanol-toluene mixtures. J. Membr. Sci. 2000, 167, 187. (23) Mandal, S.; Pangarkar, V. G. Separation of methanolbenzene and methanol-toluene mixtures by pervaporation: effects of thermodynamics and structural phenomenon. J. Membr. Sci. 2002, 201, 175. (24) Villaluenga, J. P. G.; Seoane, B. Experimental estimation of gas-transport properties of linear low-density polyethylene membranes by an integral permeation method. J. Appl. Polym. Sci. 2001, 82, 3013. (25) Van Krevelen, D. W. Properties of Polymers; Elsevier: Amsterdam, The Netherlands, 1990; Chapter 7. (26) Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca, NY, 1953; Chapter 13. (27) Dewan, A. K.; Tao, L. C.; Weber, J. H. Correlation of Wilson parameters with number of carbon atoms for primary alcoholaromatic systems. Ind. Eng. Chem. Process Des. Dev. 1978, 17, 371. (28) Lee, M. Y.; Bougeois, D.; Belfort, G. Sorption, diffusion and pervaporation of organics in polymer membranes. J. Membr. Sci. 1989, 44, 161.

Received for review August 4, 2002 Accepted November 27, 2002 IE020603E