Chitosan−Silica Complex Membranes from Sulfonic Acid

Biomacromolecules 2005, 6, 368-373

368

Chitosan-Silica Complex Membranes from Sulfonic Acid Functionalized Silica Nanoparticles for Pervaporation Dehydration of Ethanol-Water Solutions Ying-Ling Liu,* Chih-Yuan Hsu, Yu-Huei Su, and Juin-Yih Lai Department of Chemical Engineering and R&D Center for Membrane Technology, Chung Yuan Christian University, Chungli, Taoyuan 320, Taiwan Received August 11, 2004; Revised Manuscript Received October 11, 2004

Nanosized silica particles with sulfonic acid groups (ST-GPE-S) were utilized as a cross-linker for chitosan to form a chitosan-silica complex membranes, which were applied to pervaporation dehydration of ethanolwater solutions. ST-GPE-S was obtained from reacting nanoscale silica particles with glycidyl phenyl ether, and subsequent sulfonation onto the attached phenyl groups. The chemical structure of the functionlized silica was characterized with FTIR, 1H NMR, and energy-dispersive X-ray. Homogeneous dispersion of the silica particles in chitosan was observed with electronic microscopies, and the membranes obtained were considered as nanocomposites. The silica nanoparticles in the membranes served as spacers for polymer chains to provide extra space for water permeation, so as to bring high permeation rates to the complex membranes. With addition of 5 parts per hundred of functionlized silica into chitosan, the resulting membrane exhibited a separation factor of 919 and permeation flux of 410 g/(m2 h) in pervaporation dehydration of 90 wt % ethanol aqueous solution at 70 °C. Introduction Recently, chitosan has been widely studied for use in clinics,1 drug delivery systems,2 solid polyelectrolytes,3 surfactants,4 and membranes on ultrafiltration, reverse osmosis, and pervaporation.5 Considerable efforts are especially directed at modifying chitosan to improve its solubility in water6 and other physicochemical properties.7 One of the convenient and effective approaches to improve the physical and mechanical properties of chitosan is to cross-link the polymer.8 The hydroxyl and amino groups on glucosamine units of chitosan provide the reactive sites for cross-linking reactions. Various reagents, including gluteraldehyde,8a sulfuric acid,8b epoxy compound,8c dialdehyde starch,8d and nontoxic nature agents,8e-g etc., have been used as crosslinkers for chitosan. More recently, irradiation is also used to cross-link chitosan.8h Another approach to improve the physico/chemical properties of chitosan and other biomaterials is through the formation of organic-inorganic hybrid materials.2a,9 Pervaporation shows its efficient performance on separating an azeotropic mixture or a liquid mixture with close boiling points. Separation of such a mixture is difficult to achieve by conventional distillation process. Chitosan is widely used in pervaporation membranes due to its high hydrophilicity, good film-forming characteristic, and excellent chemical-resistant properties.10 However, the swelling of the chitosan membrane in an aqueous solution results in both an increase in solubility and diffusivity of alcohol, and consequently lowers the water permselectivity. Great efforts * To whom correspondence should be addressed. Telephone: +886-32654130. Fax: +886-3-2654199. E-mail: [email protected].

are reported to control the membrane swelling, including introducing cross-linking structure to the membrane, blending chitosan with other polymers, and developing organicinorganic hybrid membrane.11 However, these modifications on the membrane preparation and membrane structures often significantly reduce the hydrophilicity and permeation fluxes of the membranes.11,12 Therefore, it is highly critical that one should be able to enhance the water permeation rate in a membrane while maintaining the water permselectivity. In this study, silica nanoparticles possessing sulfonic acid groups (ST-GPE-S, Figure 1) on the particle surface were first prepared. Homogeneously dispersing the silica particles within chitosan resulted in a chitosan-silica nanocomposite. The sulfonic acid groups on the particle surface served as reactive sites to cross-link chitosan. Therefore, two distinct features of cross-linking and organic-inorganic nanocomposite in a membrane were achieved via a one-pot process. The physical properties and swollen stability of the chitosan membrane were highly improved by the cross-linking and organic-inorganic structures. As a result, high water permselectivity could be achieved from the resulting membrane. Moreover, the introduction of the silica nanoparticles into the chitosan membrane would bring about extra free volumes in the polymer domains to increase the permeation rate of the membrane in pervaporation dehydration. Experimental Section Materials. Nanoscale silica particles were purchased from Nissan Chemical Co., Japan. The commercial product of MIBK-ST, in which 30-31 wt % of silica (particle size: 10-20 nm) was dispersed in methyl isobutyl ketone (MIBK),

10.1021/bm049531w CCC: $30.25 © 2005 American Chemical Society Published on Web 11/19/2004

Chitosan-Silica Complex Membranes

Biomacromolecules, Vol. 6, No. 1, 2005 369

Figure 1. Preparation of surface-functionalized silica nanoparticles.

was used. Chitosan polymer (number-averaged molecular weight: 300 000) with a degree of deacetylation of 85% was received from Sigma Chemical Co. Measurements. (1) Instrumental. FTIR spectra were measured with a Perkin-Elmer Spectrum One FTIR. 1H NMR spectra were recorded with a Bru¨ker MSL 300 NMR (300 MHz). Samples were prepared from dissolving the surfacefunctionalized silica particles in CDCl3 or DMSO-d6 (DMSO ) dimethyl sulfoxide) to result in a homogeneous solution. Elemental analysis and element mapping on sulfur were conducted with a Hitachi S-3000N scanning electronic microscope equipped with an energy dispersive X-ray spectroscopy (EDX) of Horiba EDX-250. The contact angles of water on the membranes were measured at room temperature using the sessile drop method by means of an angle meter of Automatic Contact Angle Meter, Model CA-VP (Kyowa Interface Science Co., Japan). Distilled water was dropped onto at least 10 different sites on each sample. An average value was obtained for the measured contact angle. Transmission electronic micrographs were observed with an instrument of JEOL JEM-200 FX II. Elemental analysis on carbon was conducted with a Heraeous CHN-O rapid elementary analyzer with benzoic acid as a standard. Differential scanning calorimetry (DSC) thermograms were recorded with a Thermal Analysis DSC-Q10 at a heating rate of 10 °C/min under nitrogen atmosphere. Thermogravimetric analysis (TGA) was performed by a Thermal Analysis TGA-2050 at a heating rate of 10 °C/min under nitrogen or air atmosphere. The gas flow rate was 100 mL/min. (2) Degree of Swelling. The membrane was dried under vacuum at 40 °C for 24 h. The dried membrane was weighed (Wo) and then immersed in a test liquid (water and various alcohol solutions) at 30 °C until an equilibrium was reached. The membrane was removed from the liquid, wiped with filter paper, and then weighed (Ws). The degree of swelling (DS) was determined from DS (%) ) (Ws -Wo)/Wo × 100. (3) Pervaporation Measurements. Pervaporation was conducted with a conventional process.11 The feed is 90 wt % of ethanol aqueous solution at 25 °C. The effective membrane area is 7.0 cm2. The operation temperature was at 70 °C, and the downstream pressure was kept at 5-8 Torr. The compositions of the permeation were measured by gas

chromatography (China Chromatography GC-8700T). The separation factor (water/ethanol) was calculated from the concentration ratio of the permeate solution over that of the feed solution. Surface-Functionalization of Silica Nanoparticles. The reaction of silica particles with glycidyl phenyl ether (GPE) was carried out as follows.13 The silica particle solution and GPE were mixed together. After adding 1000 ppm of SnCl2 as a catalyst, the mixture was stirred at 140 °C for 3 h. The solvent was removed out with a rotary evaporator, and a condensed product (ST-GPE) was obtained with centrifugation. The sulfonation on ST-GPE was carried out with treating ST-GPE with fuming sulfuric acid. The reaction was performed at room temperature for 18 h. The reaction mixture was poured into plenty of ice-water mixture and the precipitate was collected with a centrifuge. The collected product (ST-GPE-S) was then dispersed in an acetic acid aqueous solution to result in a homogeneous solution for further application. Preparation of Chitosan-Silica Complex Membranes. Chitosan was dissolved in a 2 wt % acetic acid aqueous solution to form a chitosan solution with a concentration of 1.5 wt %. The solution was stirred at room temperature for 24 h, filtered, and then degassed. After another 24 h, a certain amount of ST-GPE-S was added into the solution under stirring. The mixture was stirred for 1 h to resulted in a homogeneous solution and then cast on a glass plate with a casting knife. The thickness set for the casting knife was 1200 µm. The resulting membrane was obtained from immersing the glass plate into a water bath. The free-standing membranes were dried under vacuum for 24 h to result in the product named CSCM-X, in which X indicates the content of ST-GPE-S in the membranes (parts per hundred, phr). The membranes obtained were not treated with NaOH(aq) to increase the membranes’ affinity for water.12d Results and Discussion Preparation of Sulfonic Acid Functionalized Silica Nanoparticle. The surfaces of nanoscale silica particles were functionalized with sulfonic acid group via a two-step reaction, as shown in Figure 1. First, the nanoscale colloidal

370

Biomacromolecules, Vol. 6, No. 1, 2005

silica particles (MIBK-ST) were reacted with glycidylphenyl ether through the addition reaction between Si-OH and oxirane group to result in silica particles with phenyl pendent groups on their surface (ST-GPE).13 Subsequently, sulfonation on the pendent phenyl groups on the silica particle surface was performed by treating the obtained ST-GPE with fuming sulfuric acid through the electrophilic substitution reaction. The functionalization reaction and the chemical structure of the silica particles were characterized with FTIR and 1H NMR. Based on the FTIR spectrum of ST-GPE, the occurrence of the reaction between the silanol group and oxirane ring was observed with the disappearance of absorption peak at 915 cm-1 (oxirane ring). With the presence of absorption bands at 1600 and 1500 cm-1 (phenyl ring), 2840 cm-1 (C-H stretching in methylene group), and 3008 cm-1 (aromatic C-H stretching), it was confirmed that the glycidylphenyl ether groups were covalently incorporated onto the silica particles. Moreover, strong absorption at 1168 cm-1 (-SO3H) and 611 cm-1 (-C-S) were observed for STGPE-S. This indicates that the occurrence of sulfonation reaction and introduction of -SO3H groups onto the GPE moieties of ST-GPE were achieved. 1H NMR spectroscopy further corroborates the above-mentioned reaction and the chemical structures of products obtained. As the mixture of silica particles and GPE showed only absorption peaks associating to GPE in their 1H NMR spectra, the reaction between silica particles and GPE was confirmed with the disappearance of absorption peaks of oxirane ring and the appearance of absorption peaks at 1.21 (2H), 3.67 (1H), and 4.05 (2H) ppm observed with the 1H NMR spectrum of STGPE. The incorporated phenyl group was observed with two absorption peaks at 7.03 (3H) and 7.34 ppm (2H). For STGPE-S, absorption peaks at 6.91 (1H), 7.23 (1H), and 7.54 (1H) ppm indicate that three types of aromatic protons are present in the product. As the alkoxy substituent is an activating group and an ortho-/para-director for electrophilic aromatic substitution reaction, ST-GPE-S should possess two -SO3H groups at the ortho and para positions of the phenyl moieties. Because of this, the chemical structure of STGPE-S is shown like it is in Figure 1. Moreover, the 1H NMR spectrum of ST-GPE-S also indicates that there was no hydrolysis occurring on the Si-GPE linkages after sulfonation. The carbon contents of ST-GPE and ST-GPE-S were found to be 12.9 and 4.56 wt %, respectively. The value of the carbon content of ST-GPE-S was coincident to the calculated one. This further demonstrates that the Si-GPE linkages were sustainable under the conditions of sulfonation reaction. The successful incorporation of GPE onto the nanoscale silica particles and the sulfonation on the particles were also demonstrated with energy-dispersive X-ray spectroscopy (EDX). As shown in Figure 2, the incorporation of GPE moieties were directly observed from the presence of the carbon absorption peak. Moreover, the sulfur peak that appeared in the EDX spectrum of ST-GPE-S confirmed the occurrence of the sulfonation reaction on the aromatic groups. Figure 3 shows the EDX mapping spectra of ST-GPE-S. Homogeneous distributions of silicon and sulfur were observed with the ST-GPE-S obtained. From elemental

Liu et al.

Figure 2. EDX spectrum for elemental analysis on ST-GPE-S.

Figure 3. EDX mapping results of ST-GPE-S: (A) Si-mapping; (B) S-mapping.

analysis, the sulfur content of ST-GPE-S was 1.82 wt %, which corresponded to an equivalent concentration, 568.7 mmol/kg of the -SO3H group to ST-GPE-S. Both ST-GPE and ST-GPE-S were soluble in most polar organic solvents such as acetone, methyl ethyl ketone, methyl isobutyl ketone, tetrahydrofuran, chloroform, ethanol, and aprotic polar solvents. Moreover, ST-GPE-S could be homogeneously dissolved in water and acetic acid aqueous solution.

Chitosan-Silica Complex Membranes

Biomacromolecules, Vol. 6, No. 1, 2005 371

Figure 4. Model for the formation of chitosan-silica nanocomposite membrane with a cross-linking feature.

Preparation and Characterization of Chitosan-Silica Complex Membranes. Given the fact that sulfuric acid was used as a cross-linker for chitosan in the literature,8b the addition of ST-GPE-S into chitosan could also cross-link chitosan by means of forming amine-sulfuric acid ion pairs. A model for cross-linking chitosan and the formation of chitosan-silica nanocomposite membrane is shown in Figure 4. The addition of ST-GPE-S also means introducing silica nanoparticles into chitosan matrixes to form silica-chitosan nanocomposites. As a result, an improvement on the physical and mechanical properties of chitosan was achieved. Meanwhile, the chemical bonding formed between chitosan and sulfonic acid groups would enhance the compatibility between inorganic silica particles and organic chitosan. Therefore, in situ cross-linking in chitosan matrixes and constructing chitosan-silica complex membranes (CSCM) were successfully achieved by means of blending ST-GPE-S with chitosan. As seen in Figure 5A, all of the membranes exhibited good transparency. This indicates that ST-GPE-S did not aggregate in the membranes, leading to the absence of phase separation. Good miscibility between ST-GPE-S and chitosan was also demonstrated using SEM micrograph (Figure 5B). From the micrograph, one could infer that a dense membrane without macrovoids and phase separation was obtained. The silica distribution in chitosan matrixes was directly observed with a TEM micrograph, as shown in Figure 5C. Silica particles approximately ranged from 30 to 60 nm were homogeneously dispersed in chitosan matrixes. Therefore, the chitosan-silica complex membranes could be considered as nanocomposites consisting of homogeneously dispersed, nanosized inorganic silica reinforcements. Figure 6 shows the DSC thermograms of pristine chitosan membrane and CSCM samples. A broad endothermic peak was observed for all testing samples. As the silica contents in CSCM samples increased, the peak shifted to lower temperature region and the endothermic enthalpy decreased. This endothermic peak might result from the removal of the adsorbed water in the chitosan membranes under heating. Results from thermogravimetric analysis on the samples supported the above-mentioned inference (Figure 7). A weight loss before 200 °C was observed for all samples. This weight loss was considered to be corresponding to the absorbed water in the membranes. The amounts of weight loss at this temperature range decreased with increasing silica contents of the CSCM samples. This implies that the formation of a chitosan-silica complex would decrease the water adsorbability of chitosan membranes, i.e., decrease the hydrophilicity of the membranes. The hydrophilicity of the membranes was then estimated by measuring their surface

Figure 5. Morphology studies on CSCM samples: (A) photographs of CSCM-10 (left) and CSCM-30 (right); (B) cross-section SEM microphotograph of CSCM-10; (C) TEM microphotograph of CSCM1.

Figure 6. DSC thermograms of pristine chitosan (CSCM-0) and chitosan-silica complex membranes.

contact angles with water. The addition of 5 phr ST-GPE-S into chitosan dramatically increased its contact angle from

372

Biomacromolecules, Vol. 6, No. 1, 2005

Liu et al.

Table 1. Pervaporation Dehydration Performance of the Chitosan-Silica Nanocomposite Membranes on 90 wt % Ethanol Aqueous Solution

membrane

membrane thickness (µm)

temp (°C)

water conc in permeate (wt %)

separation factor (RH2O/EtOH)

permeation rate (g/(m2 h)

PSI (kg/m2 h)

CSCM-0 CSCM-3 CSCM-5 CSCM-10 CSCM-20 CSCM-30

28 23 27 29 33 32

70 70 70 70 70 70

98.4 98.6 99.0 98.8 99.0 99.2

554 639 919 735 919 1102

420 450 410 450 320 410

232.7 287.5 376.8 330.7 294.1 451.8

Figure 7. TGA thermograms in air: pristine chitosan (CSCM-0) and chitosan-silica complex membranes.

79 to 94°. The contact angles of CSCM with water continuously increased with increasing ST-GPE-S contents, indicating that CSCMs were less hydrophilic than the pristine chitosan membrane. Thermal stability of CSCM samples could be referred from their TGA thermograms (Figure 7). A rapid weight loss around 260 °C resulted from the chitosan chain degradation, whereas the second rapid weight loss derived from the oxidative degradation of char formed from the chitosan chain degradation. After the addition of ST-GPE-S, an enhancement on thermal stability and a retardation on the oxidative degradation were observed for these CSCM samples. High ST-GPE-S contents of the CSCM membranes resulted in high char yields at 800 °C. It is noteworthy that the values of char yields were almost coincident with the amounts of the ST-GPE-S additives. Therefore, the increased char ratios mainly resulted from the nonvolatile silica, and the addition of ST-GPE-S did not enhance char formation from the organic part (chitosan) of the complex membranes. Similar results were also reported for other polymer-silica hybrid materials and nanocomposites.14 On the basis of the above, one could conclude that silica in chitosan might not alter the thermal degradation mechanism of chitosan.14 Figure 8 shows the effects of the added contents of STGPE-S on the degree of swelling for the CSCM samples in water and various alcohol solutions. In all of the swelling tests, the degree of swelling decreased with increasing amounts of ST-GPE-S. This result is quite reasonable since a higher content of ST-GPE-S in the membrane would increase the restriction on molecular motion of the chitosan chains due to the increasing reinforcing inorganic silica nanoparticles and cross-linking density. Moreover, a decrease in the degree of swelling of the membranes in alcohol

Figure 8. Plots of degree of swelling vs ST-GPE-S contents of complex membranes.

solutions would enhance the operation stability in pervaporation dehydration of the alcohol-water mixtures. Pervaporation Dehydration of Ethanol-Water Solutions. The performance of the complex membranes on pervaporation dehydration was performed on a 90% ethanol/ water mixture at 70 °C, and the experimental results were summarized in Table 1. It was noticeable that formation of complex membranes significantly increased the values of the separation factor, i.e., to enhance the selectivity of the membranes on ethanol-water separation. The addition of ST-GPE-S into chitosan decreased the surface hydrophilicity of the membranes, as mentioned in a previous section, and consequently might result in decreasing the water selectivity in pervaporation. Therefore, an improvement on the separation factor should result from a decrease in the degree of swelling for the membranes. On the other hand, a decrease in the permeation rates was widely observed for the crosslinked chitosan membranes, since cross-linking seriously restricted the molecular mobility.12 However, in this work, the cross-linked chitosan membranes exhibited a similar or better permeation rate as compared to the original chitosan membrane. This property is highly appreciated in membrane pervaporation dehydration, especially for the mixture with a low water content, in which small flux rates in permeation was a critical problem. The high permeation rates of the cross-linked complex membranes in this work should be attributed from the presence of the silica nanoparticles in the membranes. The silica particles might provide extra free volumes to the polymer chains, consequently to offer spaces for water molecules to permeate through the membranes. Both high separation factor and permeation rate are expected for a permeation dehydration operation, and in most cases a tradeoff exists between these two factors. To better evaluate the performance of pervaporation separation for a

Chitosan-Silica Complex Membranes

membrane, the pervaporation separation index (PSI) obtained from multiplying separation factor with permeation rate was utilized. As seen in Table 1, all of the complex membranes exhibited higher PSI values than the pristine chitosan membrane did. Moreover, the performance of the CSCM membranes was also comparable to the commercially available alternative (Sulzer 2100 membrane based on crosslinked PVA, Germany), which showed a permeation flux of 450 g/(m2 h) and a separation factor of 171 in pervaporation dehydration of 90 wt % ethanol aqueous solution at 80 °C. While compared with other membranes reported in the literature,10,11,15 the CSCM membranes in this work showed relatively high permeation fluxes and comparable separation factors. It is important to note that CSCM complex membranes containing higher than 10 phr of ST-GPE-S would become brittle and could be easily cracked and broken during pervaporation operation. Therefore, it is concluded that CSCM-5 and CSCM-10 are more suitable in practical application. Conclusion Through a novel surface-functionalization approach, silica nanoparticles with sulfonic acid groups on their surfaces were obtained for cross-linking chitosan in situ and forming a series of chitosan-silica nanocomposite membranes. Crosslinking on chitosan reduces its degree of swelling in water and an alcohol/water mixture, so as to increase its permselectivity in pervaporation dehydration of an ethanol-water mixture. Moreover, the addition of silica nanoparticles to chitosan provided extra free volumes in polymer matrixes for water permeation, which would bring high permeation rates for the complex membranes. With the addition of 5 phr functionlized silica to chitosan, the resulting membrane exhibited a separation factor of 919 and permeation flux of 410 g/(m2 h) in pervaporation dehydration of 90 wt % ethanol aqueous solution at 70 °C. Acknowledgment. Financial support from the Ministry of Economic Affairs of Taiwan (Grant No. 92-EC-17-A-10S1-0004) is highly appreciated. References and Notes (1) (a) Thacharodi, D.; Panduanga, R. K. J. Chem. Technol. Biotechnol. 1993, 58, 177. (b) Badawy, M. E. I.; Rabea, E. I.; Rogge, T. M.; Stevens, C. V.; Smagghe, G.; Steurbaut, W.; Hofte, M. Biomacromolecules 2004, 5, 589. (c) Rabea, E. I.; Badawy, M. E. I.; Stevens, C. V.; Smagghe, G.; Steurbaut, W. Biomacromolecules 2003, 4, 1457. (2) (a) Park, S. B.; You, J. O.; Park, H. Y.; Hamm, S. J.; Kim, W. S. Biomaterials 2001, 22, 323. (b) Peniche, C.; Feranadez, M.; Gallardo, A.; Lopez-Barvo, A.; Roman, J. S. Macromol. Biosci. 2003, 3, 540. (c) Sashiwa, H.; Yajima, H.; Aiba, S. Biomacromolecules 2003, 4, 1244. (d) Hu, Q.; Li, B.; Wang, M.; Shen, J. Biomaterials 2004, 25, 779.

Biomacromolecules, Vol. 6, No. 1, 2005 373 (3) (a) Wan, Y.; Creber, K. A. M.; Peppley, B.; Bui, C. T. Polymer 2003, 44, 1057. (b) Gandini, A.; Hariri, S.; Nest, J. F. L. Polymer 2003, 44, 7565. (4) Ngimhuang, J.; Furukawa, J.; Satoh, T.; Furuike, T.; Sakairi, N. Polymer 2004, 45, 837. (5) (a) Miya, M.; Iwamoto, R.; Yoshikawa, S.; Mima, S. Kobunshi Ronbunshu 1982, 39, 649. (b) Yang, A. T.; Zall, R. R. J. Food Sci. 1984, 49, 91. (c) Chanachai, A.; Jiraratananon, R.; Uttapap, D.; Moon, G. Y.; Anderson, W. A.; Huang, R. Y. M. J. Membr. Sci. 2000, 166, 271. (d) Ge, J.; Cui, Y.; Yan, Y.; Jiang, W. J. Membr. Sci. 2000, 165, 75. (6) (a) Sashiwa, H.; Yamamori, N.; Ichinose, Y.; Sunamoto, J.; Aiba, S. Macromol. Biosci. 2003, 3, 231. (b) Hirano, S.; Yamaguchi, Y.; Kamiya, M. Macromol. Biosci. 2003, 3, 629. (c) Zhou, Y.; Yang, Y.; Wang, D.; Liu, X. Chem. Lett. 2003, 32, 682. (d) Lu, S.; Song, X.; Cao, D.; Chen, Y.; Yao, K. J. Appl. Polym. Sci. 2004, 91, 3497. (7) (a) Tolaimate, A.; Desbrires, J.; Rhazi, M.; Alagui, A. Polymer 2003, 44, 7939. (b) Francis, S. J. K.; Matthew, H. W. T. Biomaterials 2000, 21, 2589. (c) Zeng, X.; Ruckenstein, E. J. Membr. Sci. 1998, 148, 195. (8) (a) Goissis, G.; Junior, E. M.; Marcantonio, R. A. C.; Lai, R. C. C.; Cancian, D. C. J.; De Caevallho, W. M. Biomaterials 1999, 20, 27. (b) Huang, R. Y. M.; Pal, R.; Moon, G. Y. J. Membr. Sci. 1999, 160, 17. (c) Wei, Y. C.; Hudson, S. M.; Mayer, J. M.; Kaplan, D. L. J. Polym. Sci., Part A: Polym. Chem. 1992, 30, 2187. (d) Schmidt, C. E.; Baier, J. M. Biomaterials 2000, 21, 2215. (e) Mi, F. L.; Sung, H. W.; Shyu, S. S.; Su, C. C.; Peng, C. K. Polymer 2003, 44, 6521. (f) Jin, J.; Song, M.; Hourston, D. J. Biomacromolecules 2004, 5, 162. (g) Kuboe, Y.; Tonegawa, H.; Ohkawa, K.; Yamamoto, H. Biomacromolecules 2004, 5, 348. (h) Ramnani, S. P.; Chaudhari, C. V.; Patil, N. D.; Sabharwal, S. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 3897. (9) (a) Mizushima, Y. J. Non-Cryst. Solids 1992, 144, 305. (b) Suzuki, T.; Mizushima, Y. J. Ferment. Bioeng. 1997, 84, 128. (c) Tian, D.; Dubois, P. H.; Jerome, R. J. Polym. Sci., Part A: Polym. Chem. 1997, 35, 2309. (d) Tian, D.; Dubois, P. H.; Grandfile, C. H.; Jermome, P.; Viville, P.; Lazzaroni, R.; Bredas, J. L.; Leprince, P. Chem. Mater. 1997, 9, 871. (e) Liu, Y. L.; Su, Y. H.; Lai, J. Y. Polymer 2004, 45, 6831. (10) (a) Wang, X. P.; Shen, Z. Q.; Zhang, F. Y.; Zhang, Y. F. J. Membr. Sci. 1996, 119, 203. (b) Uragami, T.; Kato, S.; Miyata, T. J. Membr. Sci. 1997, 124, 203. (c) Chanachai, A.; Jiraratananon, R.; Uttapap, D.; Moon, G. Y.; Anderson, W. A.; Huang, R. Y. M. J. Membr. Sci. 2000, 166, 271. (d) Jiraratananon, R.; Chanachai, A.; Huang, R. Y. M. J. Membr. Sci. 2002, 199, 211. (11) (a) Hung, M. Y.; Chen, S. H.; Liou, R. M.; Hsu, C. S.; Lai, J. Y. J. Appl. Polym. Sci. 2003, 90, 3374. (b) Hung, M. Y.; Chen. S. H.; Liou, R. M.; Hsu, C. S.; Tsia, H. A.; Lai, J. Y. Eur. Polym. J. 2003, 39, 2367. (c) Wang, Y. C.; Fan, S. C.; Lee, K. R.; Li, C. L.; Huang, S. H.; Tsai, H. A.; Lai, J. Y. J. Membr. Sci. 2004, 239, 219. (12) (a) Ghazali, M.; Nawawi, M.; Huang, R. Y. M. J. Membr. Sci. 1997, 124, 53. (b) Huang, R. Y. M.; Pal, R.; Moon, G. Y. J. Membr. Sci. 1999, 160, 17. (c) Huang, R. Y. M.; Pal, R.; Moon, G. Y. J. Membr. Sci. 2000, 167, 275. (d) Uragami, T.; Takuno, M.; Miyata, T. Macromol. Chem. Phys. 2002, 203, 1162. (e) Uragami, T.; Yamamoto, S.; Miyata, T. Biomacromolecules 2003, 4, 137. (13) Liu, Y. L.; Hsu, C. Y.; Wang, M. L.; Chen, H. S. Nanotechnology 2003, 14, 813. (14) Kashiwagi, T.; Morgan, A. B.; Antonucci, J. M.; VanLandingham, M. R.; Harris, R. H.; Awad, W. H. Shields, J. R. J. Appl. Polym. Sci. 2003, 89, 2072. (15) (a) Uragami, T.; Okazaki, H.; Matsugi, H.; Miyata, T. Macromolecules 2002, 35, 9156, and the references therein. (b) Chang, C. L.; Chang, M. S. J. Membr. Sci. 2004, 238, 117.

BM049531W