Bioresource Technology 100 (2009) 25692574

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Green composite lms prepared from cellulose, starch and lignin in room-temperature ionic liquid
Rong-Lan Wu a,b, Xiu-Li Wang a, Fang Li a, Hui-Zhang Li a, Yu-Zhong Wang a,*
a

Center for Degradable and Flame-Retardant Polymeric Materials, College of Chemistry, State Key Laboratory of Polymer Materials Engineering, Sichuan University, Chengdu 610064, China b Key Lab of Oil and Gas Fine Chemicals, Ministry of Education, Xinjiang University, Urumqi, Xinjiang, 830046, China

a r t i c l e

i n f o

a b s t r a c t
A series of novel biobased composite lms derived from cellulose, starch and lignin were prepared from an ionic liquid (IL), 1-allyl-3-methylimidazolium chloride (AmimCl) by coagulating in a nonsolvent condition. The ionic liquid can be recycled with a high yield and purity after the green lm was prepared. The uniform design method was applied to investigate mechanical properties of the biobased composite lms. The effect of each component and their associated interactive effects were investigated. The experimental results showed that contents of cellulose, lignin and starch had a signicant inuence on the mechanical properties of composite lms. The composite lms showed relatively excellent mechanical properties in dry and wet states owing to the mutual property supplement of different components. The composite lms were characterized via FT-IR, X-ray diffraction (XRD) and scanning electron microscope (SEM). Their thermal stability and gas permeability were also investigated, and the results showed that the composite lms had good thermal stability and high gas barrier capacity and give a CO2:O2 permeability ratio close to 1. 2008 Elsevier Ltd. All rights reserved.

Article history: Received 2 September 2008 Received in revised form 15 November 2008 Accepted 18 November 2008 Available online 24 January 2009 Keywords: Cellulose Starch Lignin Ionic liquid Homogenous blends

1. Introduction Synthetic polymers have provided undeniable benets for the development of the society. However, non-biodegradable petroleum-based plastics have become a threat to the environment and the diminishing petrochemical resources. Therefore, an increasing attention has been paid on the development of environmentally benign natural biobased materials, which is a solution not only to burgeoning environmental threats but also to the uncertainty of petroleum supply (Cakmakli et al., 2004). Cellulose, starch and lignin are three kinds of abundant biopolymers in the nature, The potential advantages of such biobased materials, apart from their environmental gains, are low cost, not depending on petroleum sources, available from renewable resources, and enabling to replace some synthetic polymers, etc. However, their inherent disadvantages such as the relatively low tensile property and high water absorbency of starch, the poor solubility and processability of cellulose, have limited their wider applications. Furthermore, all of them cannot melt, and cannot be processed like thermoplastic polymers. This makes the process much different from the synthetic polymer. Traditionally, the three
* Corresponding author. Address: Center for Degradable and Flame-Retardant Polymeric Materials, College of Chemistry, Sichuan University, Chengdu 610064, China. Fax: +86 28 85410259. E-mail address: yzwang@scu.edu.cn (Y.-Z. Wang). 0960-8524/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.11.044

biobased polymers have been incorporated into other polymeric materials as lower cost and biodegradable llers, respectively. Many systems based on these natural polymers have been investigated recently, such as PCL/starch (Ikeo et al., 2006), PHB/lignin (Kai et al., 2004), PCL/cellulose (Sanchez-Garcia et al., 2008), etc. However, these synthetic biodegradable polymers are expensive and/or rely on the petroleum resources. In addition, because of the poor compatibility between hydrophobic polymers and natural polymers, reactive compatibilizers or coupling agents are often used to improve interfacial interactions of these composites. Recently, many attentions had been paid on preparing the totally biodegradable blends with the natural polymers and their derivatives, such as cellulose diacetate/starch (Lee et al., 2006), starch acetate/native starch (Guan and Hanna, 2004), cationic starch/cellulose (Bratskaya et al., 2006) and chitosan/cellulose acetate (Liu and Bai, 2005), etc. These blends present good compatibility and retain biodegradability. However, compared with the preparation of environmental friendly materials based on pure natural polymers, using their derivatives will increase the cost and complicacy. To preserve biodegradability, low cost and to improve the mechanical resistance of nal products, associations between cellulose and plasticized starch have been tested. It is well known that cellulose bers can reinforce plasticized starch (Kumar and Singh, 2008; Averous et al., 2001) but only little attention has been paid

2570

R.-L. Wu et al. / Bioresource Technology 100 (2009) 25692574

to the plasticized starch toughening cellulose (Dufresne and Vignon, 1998). Furthermore, it has been reported that lignin can increase the plasticity of starch matrix due to its lower molecular weight fraction (Morais et al., 2005; Pouteau et al., 2004), and can form a cementing material for cellulose brils, which can improve the mechanical properties of the composite material of lignin/cellulose (Rohella et al., 1996). Therefore, lignin can be used as a plasticizer of starch, a reinforcing agent of cellulose and a compatibilizer of cellulose and starch, to prepare natural biopolymer composites composed of cellulose, starch and lignin. However, because of the poor solubility and processability of them, it is difcult to dissolve the three biopolymers in a common solvent or process them in melt state. Recently, room-temperature ionic liquids having the imidazolium structure have been found to be nonderivatizing solvents for cellulose and starch (Seoud et al., 2007) For example, it has been demonstrated that 1-allyl-3-methylimidazolium chloride (AmimCl) could dissolved cellulose in high concentrations (Zhang et al., 2005), another study reported the dissolution of starch in BmimCl (Stevenson et al., 2007). Fortunately, in this work, it has been found that lignin can be dissolved directly in AmimCl with heating. To obtain novel natural biopolymer composite materials, we attempted to prepare cellulose/starch/lignin composite lms (CSL lms) using the ionic liquid, AmimCl as solvent. The uniform design method was applied to investigate the effect of each component and their interactive effects on the mechanical properties of the lms. The relationship between the structure and properties of the composite lms were investigated using FT-IR, XRD, TG analyses and gas permeability. Furthermore, the recovery of AmimCl was also discussed.

obtained by this method. IL can be recovered by evaporating water from the coagulated liquid. 2.3. Characterization 2.3.1. FT-IR spectroscopy analysis The IR spectra were recorded with a Fourier transform IR (FT-IR) spectrometer (6700, NICOLET). The composite lms were ground into powder and dried under vacuum for 24 h before the infrared (IR) measurement. 2.3.2. X-ray diffraction studies The X-ray diffraction patterns with Cu Ka radiation at 40 kV and 30 mA were recorded in the range of 2h = 550 with an X-ray diffractometer (X0 pert, Philips). The composite lms were cut into powder for the measurement of X-ray diffraction. 2.3.3. Microscopic analysis The dry lms were frozen in liquid nitrogen, snapped immediately and vacuum-dried. The free surface (side in direct contact with the coagulant) and the fracture surface of the lms were sputtered with gold and observed using a scanning Electron Microscope Philips XL-3 (FEI Company, Oregon, USA). 2.3.4. Thermal analysis Thermogravimetric analysis was carried out using a SDT Q600 (TA Co., USA) under nitrogen atmosphere. Samples (about 10 mg) were heated from room-temperature to 500 C at a heating rate of 10 C/min. 2.3.5. Mechanical properties measurement The tensile strength (rb) and elongation at break (eb) of the lms in both dry and wet states were measured on a universal testing machine (CMT6503, Shenzhen SANS Test Machine Co., Ltd.) according to ISO6239, at a tensile rate of 10 mm min1. The wet lms were measured immediately after immersed in water for 30 min. The rb and eb values are the averages of ve measurements. 2.3.6. Gas permeability Oxygen permeability and carbon dioxide permeability were measured using differential-pressure method with a VAC-VBS gas permeability tester (Labthink Inc., Ltd.). This equipment was operated according to Standard Method D3985-81 (ASTM, 1989). Testing was performed at 25 C under dry environment (0% RH). Oxygen and carbon dioxide transmission rates were obtained in cm3/m2 day atm. Multiplying these values by lm thickness provided the diffusion coefcients of oxygen and carbon dioxide. Three samples were measured for each test. 2.4. Uniform design method Uniform design is an experimental design, which allocates experimental points uniformly scattered on the domain, proposed by Fang and Wang in 1980s, based on quasi-Monte Carlo method or numbertheoretic method (NTM) (Fang, 1994; Liang et al., 2001). It is one powerful tool in studying the interaction of mixtures has been widely used in chemistry, chemical engineering and biotechnology (Ban et al., 2006, 2007). The uniform design was performed to study the effects of cellulose, starch and lignin on the mechanical properties in this study. The experimental program used three independent variables, which were named, X1, cellulose (C, 3085 wt%); X2, starch (S, 1065 wt%); and X3, lignin (L, 0.510 wt%). The following functions were considered as dependent variables: Y1, Y2 (the tensile strength of dry and wet lms, respectively) and Y3, Y4 (the elonga-

2. Methods 2.1. Materials The IL, 1-allyl-3-methylimidazolium chloride (AmimCl) was synthesized according to the literature (Zhang et al., 2005). Natural cellulose (bamboo pulp) with a degree of polymerization of 650, obtained from Leshan Beiya Yasong Paper Company (Chengdu, China). Corn starch was obtained from Langfang Starch Company (technical grade, Langfang, Hebei, China). According to the supplier, amylose and amylopectin contents are 25% and 75%, respectively. Lignin was purchased from Wuhan East China Chemical Co., Ltd. (Wuhan, China). 1-Methylimidazole was purchased from Kaile Chemical Reagent Co. (Linhai, China). Allyl chloride (analytical grade P 99.5%) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), and used in the experiment without further purication. 2.2. Preparation of composite lms All the materials were dried at 80 C under vacuum for 24 h before used. The weight ratios of cellulose, starch and lignin are given in Tables 1 and 2, respectively. The whole procedure was described as follows: all materials were dispersed into AmimCl, heated to 80 C with continuous stirring; they became transparent solution in which biopolymer concentration is about 6 wt%. The complete dissolution procedure of the materials was conrmed by an optical microscopy. The mixed solution was cast onto a glass plate with thickness of about 0.5 mm, after which air bubble was taken off in a vacuum oven, and the cellulose solution was coagulated in the distilled water to obtain a regenerated biopolymer gel. Subsequently, the obtained transparent membranes were washed with running water and dried in the air. A series of cellulose/starch/lignin blend lms (CSL lms) with different weight ratios were

R.-L. Wu et al. / Bioresource Technology 100 (2009) 25692574 Table 1 Uniform design data for the composite lms. No. of expt. Content of cellulose (wt%) 1 2 3 4 5 6 7 8 85.0 78.6 69.3 65.6 59.3 51.2 44.3 34.4 Content of starch (wt%) 10.0 19.0 23.4 33.2 34.1 44.1 55.0 61.5 Content of lignin (wt%) 5.0 2.4 7.3 1.2 6.6 4.7 0.7 4.1 Tensile strength (rb) of the composite lms (MPa) Dry 35.2 2.1 25.3 2.0 28.3 1.3 14.5 1.0 24.1 1.1 16.9 2.0 15.1 1.8 15.8 1.3 Wet 16.2 1.9 8.9 0.9 9.8 0.8 3.5 0.6 7.8 0.5 5.9 0.8 3.0 0.4 4.2 0.8

2571

Elongation (eb) of the composite lms (%) Dry 92.0 2.5 94.1 3.1 72.9 2.3 81.5 2.6 83.9 3.9 43.4 2.0 27.1 1.9 34.2 2.1 Wet 91.7 4.1 41.8 2.6 84.6 2.9 16.9 1.2 62.9 2.0 60.0 2.6 15.9 1.0 40.1 1.1

tion at break of dry and wet lms, respectively) were also sued. The design of the experiments and corresponding results obtained are listed in Table 1. 3. Results and discussion 3.1. Dissolution of biopolymer into ionic liquid The dissolving process of cellulose, starch and lignin in AmimCl at 80 C was real time monitored using a hot stage optical microscopy. It was very interesting to observe that the mixed biopolymers without being pretreated or activated can be dissolved in AmimCl within 30 min. The solubilization efciency of them in IL was found in an order of starch > lignin P cellulose. It is probably due to the high crystallinity of cellulose and the aromatic character of lignin (Kilpelainen et al., 2007). It was also observed that the viscosity of the solution mainly depends on the content of cellulose, and the higher the content of cellulose, the higher the viscosity of the solution. 3.2. Effect of component on the mechanical properties Tensile strength (rb) and elongation at break (eb) are important mechanical properties of the CSL lms, which mainly depend on components in CSL lms. Table 2 shows the rb and eb values of binary and ternary composite lms consisted of cellulose, starch and lignin. The values of rb and eb of the CSL lms are better than those of cellulose/starch and cellulose/lignin blended lms. This result indicates that the synergistic interactions among cellulose, starch and lignin reinforce the composites. Therefore, the CSL lms with different components were prepared and the uniform design method was used to conduct the statistical analyses for mechanical properties of the composite lms. The experimental data were analysed using the Data Processing Software of Uniform Design 3.0. The generalized regression models of the wet and dry composite lms were obtained. Table 3 gives the summary of stepwise regression equations which described the contributions of each component and their inuence on mechanical properties of the lms. Thus, the corresponding surface diagrams can be derived from

regression model by holding one variable xed (shown in Figs. 1 and 2). The results indicate that the synergistic interactions among cellulose, starch and lignin show a signicant effect on rb and eb of CSL lms in wet and dry states. However, the components of cellulose, starch and lignin show different dependence on rb and eb of CSL lms in different states. Fig. 1 shows the effects of cellulose, starch and lignin contents on the strength of dry and wet lms, respectively. It can be seen that with the increase of cellulose concentration, rb values of wet and dry CSL lms increase. However, when starch content goes up to a high level (>35 wt%), the strength of dry lms will not increase with the increase of cellulose content, indicating that starch has a negative effect on strength at high cellulose content. Similar to the addition of cellulose, lignin also plays an important role in reinforcing dry and wet composite lms. Furthermore, the inuence of lignin on strength is signicant at a high cellulose content compared to low cellulose content. It is found that in dry state rb values of CSL lms decrease monotonically with the increase of the starch content, and in wet state rb values at rst increase slightly then decrease. These results suggest that a proper proportion among cellulose, starch and lignin levels can enhance the strength of lms. More specically, interactions among cellulose, starch and lignin demonstrate that optimal strength can be achieved by higher cellulose and lignin contents with lower level of starch content. Fig. 2 shows the effects of cellulose, starch and lignin contents on the elongation at break of dry and wet composite lms. It can be observed that cellulose can greatly enhance the exibility of dry and wet lms especially at high starch contents. Regarding eb values of the CSL lms in dry and wet states, there is an almost similar dependence on the starch and lignin. As the starch and lignin contents increase, eb values of lms increase at rst, then decrease. These results suggest that the optimal exibility can be achieved at higher cellulose content with middle level of lignin and starch contents. In summary, each biopolymer not only contributes to the properties of composite lms alone but also engages in biopolymer biopolymer interactions which affect the mechanical properties of overall systems. In this multiple biopolymer system, cellulose, starch and lignin have good afnity to each other and obtain

Table 2 Mechanical properties of binary and ternary composite lms. Cellulose/starch/lignin composite lms Tensile strength (rb) of the composite lms (Mpa) Dry 85/10/5 85/15/0 85/0/15 95/1/4 95/5/0 95/0/5 35.2 2.1 28.9 1.8 29.7 1.5 40.1 2.2 31.3 1.9 34.5 1.2 Wet 16.2 1.9 6.98 1.2 13.61 1.5 17.5 1.2 8.6 1.0 15.2 1.1 Elongation (eb) of the composite lms (%) Dry 92.0 2.5 65.6 2.8 64.7 2.9 72.8 2.3 60.5 3.3 62.3 2.0 Wet 91.7 4.1 69.0 3.6 75.2 3.5 94.0 3.9 62.9 3.2 91.2 2.7

2572

R.-L. Wu et al. / Bioresource Technology 100 (2009) 25692574

Table 3 Relationship between lm mechanical property and content of biopolymers. Eq. no. 1 2 3 4 Multiple regression model Y1 = 31.7 + 0.000726X 2 0.00817X1X2 + 0.0269X1X3 0.0185X2X3 2 Y2 = 3.08 0.178X2 + 0.0526X1X3 + 0.00265X 2 + 0.313X 2 2 3 Y3 = 18.2 + 33.1X3 + 0.0466X1X2 0.0219X 2 0.197X2X3 3.35X 2 2 3 2 Y4 = 18.3 0.00828X 1 + 0.462X1X3 + 0.0357X2X3 2.56X 2 3 R2 0.976 0.992 0.997 0.995

mutual property supplement, which can achieve the excellent tensile strength without sacricing the extensibility of lms. 3.3. Structure and morphology of composite lms FT-IR measurements were carried out to identify the possible interactions among cellulose, starch and lignin in composite lms. Compared the spectrum of CSL lms to those of cellulose, starch and lignin, it can be found that the spectrum of CSL lms is similar to that of cellulose because the molecular structure of CSL lm is mainly based on cellulose. The absorption peaks at 3431, 3440 and 3427 cm1 attribute to the OH groups on cellulose, starch and lignin, respectively, which are broadened and shifted to around 3421 cm1 in composite lms. Furthermore, a corresponding effect is shown in the asymmetric COC stretching region (10201160 cm1) of the composite lms with different contents of cellulose, starch and lignin. The characteristic peaks of corn starch and the CO bond stretching of COH group appear at about 1156 and 1081 cm1, respectively; the CO bond stretching of COC group in the anhydroglucose ring appears at about 1020 cm1 (Wang et al., 2008), and these peaks shift to lower wave numbers (around 1150, 1070 and 1012 cm1, respectively) in composite lms and become broader. The peak at 1160 cm1 belongs to the CO stretching of cellulose (Kataoka and Kondo, 1998), which is also broadened and shifted to 1150 and 1151 cm1 in the composites lms. Peaks at 1120 and 1040 cm1 for lignin are

characteristic bands of the primary and secondary OH groups (Nada et al., 1998; Kadla and Kubo, 2003), which are moved to a lower wave numbers(around 1114 and 1012 cm1) in the composite lms. These results suggest that strong hydrogen bonds have been formed among the cellulose, starch and lignin. In this work, starch and lignin could be dissolved in AmimCl easily compared with cellulose, and it had been reported that cellulose could be dispersed molecularly in AmimCl (Wan et al., 2008). Therefore, we suggest that starch and lignin can also be dispersed molecularly in AmimCl. During the dissolution in AmimCl, the intra- and inter-molecular hydrogen bonds of each component were destroyed and the three components were dispersed uniformly in the IL. Furthermore, the rapid coagulation process was favorable to the formation of hydrogen bonds between the hydroxyl groups of three biopolymers, which combined the three components compactly. The original materials are bamboo pulp, corn starch and lignin, respectively. The regenerated samples were obtained through dissolving these original materials in AmimCl and then coagulated with distilled water, respectively. The XRD patterns of the original and regenerated materials reveal that the intensity of diffraction peaks of these regenerated materials is reduced signicantly. The bamboo pulp has the typical diffraction peaks of cellulose I at 2h = 14.8, 16.3 and 22.6 (Zhang et al., 2005; Reddy and Yang, 2008). However, after dissolved and regenerated, it exhibits the typical diffraction patterns of cellulose II at 2h = 20.1 and 21.2. This transformation from cellulose I to cellulose II occurred, which was also testied from the results of Zhang et al. (2005). The crystallinity of regenerated cellulose (RC) decreases greatly in comparison with the original cellulose. Corn starch is a semi-crystalline material, which shows ve diffraction peaks of A-type starch at 15.2, 17.2, 18.0, 19.9 and 23.1 (Liu et al., 2008). A broad peak appears in the pattern of regenerated starch (RS), indicating that regenerated starch has lost its crystallization. This phenomenon shows that, in the dissolution process, IL can break down the inter- and intra-molecular hydrogen bonds rapidly and destroy

Fig. 1. Effects of content of cellulose (C), starch (S) and lignin (L) on tensile strength (rb) of the dry and wet lms: rbCL (a), rbCS (b), rbSL (c), rbCL (d), rbCS (e) and rbLS (f).

R.-L. Wu et al. / Bioresource Technology 100 (2009) 25692574

2573

Fig. 2. Effects of content of cellulose (C), starch (S) and lignin (L) on elongation at break (eb) of the dry and wet lms: ebCL (a), ebCS (b), ebLS (c), ebLC (d), ebCS (e) and ebLS (f).

the original crystalline forms of cellulose and starch. Furthermore, the coagulation process is so transitory that it is unfavorable to the crystallization of cellulose and starch. The X-ray diffraction pattern of the regenerated lignin (RL) is very similar with that of original lignin due to its amorphous and highly cross-linked morphology (Rohella et al., 1996). Different from the regenerated materials, all of the composite lms become quite amorphous after dissolved and blended in IL, probably due to the strong dissolving effect of AmimCl and the formation of inter-molecular hydrogen bond among cellulose, starch and lignin. The three biopolymers can be dispersed molecularly in AmimCl, and the molecular chains are aggregated in rapid coagulation process, which hinders the crystallization of cellulose and starch. By observing the surface morphologies of the regenerated cellulose and composite lms, it can be seen that the free and fracture surfaces of all lms display uniformity from the interior to the surface, that is, a dense and homogeneous texture. It is different from the structure of the cellulose regenerated from NaOH/H2O (Zhang et al., 2001, 2002), no porous structure can be observed. Moreover, with the addition of starch and lignin to cellulose, no obvious phase separation can be seen and there is not any phase interfaces among these three kinds of biopolymers. These results further

prove the excellent compatibility and a higher homogeneity in the ternary blend system. 3.4. Thermal properties of composite lms Samples for thermogravimetric analysis were the same as those of FT-IR spectroscopy tests. A small weight loss of about 25% below 100 C is assigned to the release of moisture from the samples. All of the samples show one-stage degradation within the range of 240400 C except lignin. The lignin sample exhibits a multi-step degradation process and the major weight loss started at 241 C, which is probably due to its complex structure. The residual weights of the composite lms beyond 500 C increase slightly with increasing cellulose and lignin content in the composite lms. The maximum decomposition temperature of the RC (Tmax = 326 C) is lower than that of pure cellulose (Tmax = 345 C). This result suggested that the strong intra- and inter-molecular hydrogen bonds of cellulose were partially broken by IL during the solution process, which is coincident with the result of the X-ray diffraction. The maximum decomposition temperatures of the composite lms (Tmax = 331335 C) are slightly lower than that of pure cellulose, but higher than that of starch (Tmax = 309 C), lignin and RC, showing relatively high thermal stability because of the enhancement of inter-molecular hydrogen bonds among biopolymers in the blend lms. 3.5. Gas permeability of composite lms

Table 4 Gas permeability of RC and composite lms.a Cellulose/starch/ lignin composite lms 50/40/10 73/25/2 80/15/5 90/1/9 RC
a

Oxygen permeability (103cm3/ m day atm) 12.49 0.65 12.95 1.01 13.16 0.98 13.08 0.53 11.08 0.88

Carbon dioxide permeability (103cm3/ m day atm) 13.20 0.95 21.37 1.13 15.64 0.87 21.84 0.77 13.30 0.79

Permeability ratio (PCO2:PO2) 1.05 1.65 1.19 1.67 1.2

Permeability evaluated at 25 C and 0% relative humidity.

Oxygen and carbon dioxide permeability values of RC lm and CSL composite lms have been presented in Table 4. CSL lms have higher gas permeability values than RC lm and give a wide range of CO2:O2 permeation ratios in the range of 1.051.67. Thus, CSL lms can create and maintain a suitable atmosphere, such as reducing oxygen and elevating carbon dioxide concentrations, so as to reduce respiration and prevent changes (Yun et al., 2006), which provides one promising applications as fresh food packaging.

2574

R.-L. Wu et al. / Bioresource Technology 100 (2009) 25692574 Ikeo, Y., Aoki, K., Kishi, H., Matsuda, S., Murakami, A., 2006. Nano clay reinforced biodegradable plastics of PCL starch blends. Polym. Adv. Technol. 17, 940944. Kadla, J.F., Kubo, S., 2003. Miscibility and hydrogen bonding in blends of poly(ethylene oxide) and kraft lignin. Macromolecules 36, 78037811. Kai, W.H., He, Y., Asakawa, N., Inoue, Y., 2004. Effect of lignin particles as a nucleating agent on crystallization of poly(3-hydroxybutyrate). J. Appl. Polym. Sci. 94, 24662474. Kataoka, Y., Kondo, T., 1998. FT-IR microscopic analysis of changing cellulose crystalline structure during wood cell wall formation. Macromolecules 31, 760 764. Kilpelainen, I., Xie, H.B., King, A., Granstrom, M., Heikkinen, S., Argyropoulos, D.S., 2007. Dissolution of wood in ionic liquids. J. Agric. Food Chem. 55, 91429148. Kumar, A.P., Singh, R.P., 2008. Biocomposites of cellulose reinforced starch: improvement of properties by photo-induced crosslinking. Bioresour. Technol. 99, 88038809. Lee, S.Y., Cho, M.S., Nam, J.D., Lee, Y., 2006. Melting processing of biodegradable cellulose diacetate/starch composites. Macromol. Symp. 242, 126130. Liang, Y.Z., Fang, K.T., Xu, Q.S., 2001. Uniform design and its applications in chemistry and chemical engineering. Chemom. Intell. Lab. Syst. 58, 4347. Liu, C.X., Bai, R.B., 2005. Preparation of chitosan/cellulose acetate blend hollow bers for adsorptive performance. J. Membr. Sci. 267, 6877. Liu, C., Shao, Y., Jia, D.M., 2008. Chemically modied starch reinforced natural rubber composites. Polymer 49, 21762181. Morais, L.C., Curvelo, A.A.S., Zambon, M.D., 2005. Thermoplastic starch lignosulfonate blends. 1. Factorial planning as a tool for elucidating new data from high performance size-exclusion chromatography and mechanical tests. Carbohyd. Polym. 62, 104112. Nada, A.A.M.A., El-Sakhawy, M., Kamel, S.M., 1998. Infra-red spectroscopic study of lignins. Polym. Degrad. Stab. 60, 247251. Pouteau, C., Baumberger, S., Cathala, B., Dole, P., 2004. Ligninpolymer blends: evaluation of compatibility by image analysis. C.R. Biologies 327, 935943. Reddy, N., Yang, Y.Q., 2008. Characterizing natural cellulose bers from velvet leaf (Abutilon theophrasti) stems. Bioresour. Technol. 99, 24492454. Rohella, R.S., Sahoo, N., Paul, S.C., Choudhury, S., Chakravortty, V., 1996. Thermal studies on isolated and puried lignin. Thermochim. Acta 287, 131138. Sanchez-Garcia, M.D., Gimenez, E., Lagaron, J.M., 2008. Morphology and barrier properties of solvent cast composites of thermoplastic biopolymers and puried cellulose bers. Carbohyd. Polym. 71, 235244. Seoud, O.A.E., Koschella, A., Fidale, C., Dorn, S., Heinze, T., 2007. Applications of ionic liquids in carbohydrate chemistry: a window of opportunities. Biomacromolecules 8, 26292647. Stevenson, D.G., Biswas, A., Jane, J., Inglett, G.E., 2007. Changes in structure and properties of starch of four botanical sources dispersed in the ionic liquid, 1butyl-3-methylimidazolium chloride. Carbohyd. Polym. 67, 2131. Wan, Z.Z., Li, L., Cui, S.X., 2008. Portrait of isolated individual natural cellulose molecules. Biopolymers 89, 11701174. Wang, N., Yu, J.G., Ma, X.F., 2008. Preparation and characterization of compatible thermoplastic dry starch/poly(lactic acid). Polym. Compos. 29, 551559. Yun, J.H., An, D.S., Lee, K.E., Jun, B.S., Lee, D.S., 2006. Modied atmosphere packaging of fresh produce using microporous earthenware material. Packag. Technol. Sci. 19, 269278. Zhang, L.N., Ruan, D., Zhou, J.P., 2001. Structure and properties of regenerated cellulose lms prepared from cotton linters in NaOH/urea aqueous solution. Ind. Eng. Chem. Res. 40, 59235928. Zhang, L.N., Ruan, D., Gao, S.J., 2002. Dissolution and regeneration of cellulose in NaOH/thiourea aqueous solution. Polym. Sci. Part B: Polym. Phys. 40, 1521 1529. Zhang, H., Wu, J., Zhang, J., He, J.S., 2005. 1-Allyl-3-methylimidazolium chloride room temperature ionic liquid: a new and powerful nonderivatizing solvent for cellulose. Macromolecules 38, 82728277.

3.6. Recycling of the ionic liquid After the regeneration of biopolymer blends, the residual IL in the coagulation bath can be recovered by a simple process of reducing the pressure and subsequently distilling to remove water. NMR spectra gave the evidence that no degradation was found and the purity of ionic liquid was high enough for repeating use. 4. Conclusion In this study, we have successfully prepared novel cellulose/ starch/lignin composite lms in AmimCl and coagulated with water. The contents of cellulose and lignin can affect the mechanical properties of lms signicantly. Starch can make good contribution to the exibility of lms. The composite lms are amorphous and have good transparence. Due to the strong intermolecular hydrogen bonds, the thermal stability of the composite lms is better than that of RC. CSL lms have higher gas permeability values than RC lm and have a wide range of CO2:O2 permeation ratios (1.051.67), so they have great potential use as fresh food packaging. Therefore, IL can be used as a green solvent for preparing composite lms based on natural biomacromolecules, and the process is very facile and environmentally friendly. Acknowledgement The work was nancially supported by the National Science Fund for Distinguished Young Scholars (50525309). References
Averous, L., Fringant, C., Moro, L., 2001. Plasticized starchcellulose interactions in polysaccharide composites. Polymer 42, 65656572. Ban, W.P., Song, J.G., Argyropoulos, D.S., Lucia, L.A., 2006. Inuence of natural biomaterials on the elastic properties of starch-derived lms: an optimization study. Ind. Eng. Chem. Res. 45, 627633. Ban, W.P., Song, J.G., Lucian, L.A., 2007. Inuence of natural biomaterials on the absorbency and transparency of starch-derived lms: an optimization study. Ind. Eng. Chem. Res. 46, 64806485. Bratskaya, S., Schwarz, S., Petzold, G., Liebert, T., Heinze, T., 2006. Cationic starches of high degree of functionalization: 12. Modication of cellulose bers toward high ller technology in papermaking. Ind. Eng. Chem. Res. 45, 73747379. Cakmakli, B., Hazer, B., Tekin, I.O., Kizgut, S., Koksal, M., Menceloglu, Y., 2004. Synthesis and characterization of polymeric linseed oil grafted methyl methacrylate or styrene. Macromol. Biosci. 4, 649655. Dufresne, A., Vignon, M.R., 1998. Improvement of starch lm performances using cellulose microbrils. Macromolecules 31, 26932696. Fang, K.T., 1994. Uniform Design and Uniform Design Tables. Science Press, Beijing. Guan, J.J., Hanna, M.A., 2004. Extruding foams from corn starch acetate and native corn starch. Biomacromolecules 5, 23292339.