Hydrogen bond interaction in poly(acrylonitrile-co-methylacrylate)/attapulgite nanocomposites.INTRODUCTION
Recently, compared with well developed inorganic filler of two dimensional layered silicates (such as montmo-rillonite), rod-like clay has attracted more attention in polymer/clay nanocornposite research. Apart from its natural and unique fibrillar structure, it is believed that these rod-like nanoparticles can be more efficiently dispersed in polymer matrix than traditional layered silicates due to its weak interactions between individual particles and low exchangeable sodium percentage (1, 2). Furthermore, it has been reported that a significant number of reactive silanol groups (Si--OH) are situated at the edges of its tunnel structure, which may provide sites to form H-bond with some functional groups on polymer chains (3), (4).
Previous researches on polymer/layered silicate nanocomposites has suggested that the dispersion state of nanofiller in the matrix (5-9) and interfacial adhesion interactions (10-12) are two key factors in high performance nanocomposites. For example, Wagner (13) suggested that the strong interactions at the interface will facilitate stress transfer from one to the other and lead to the increased strength and stiffness of composite. Balazs and coworkers. (14) numerically studied the conditions for driving the polymer to penetrate the gap between the clay surfaces and showed the effect of employing end-functionalized chains to promote the clay dispersion in the matrix. To improve dispersion and compatibility with polymer matrix, various kinds of organic modification have been done on clay surface, especially for some non-polar polymers, such as PP (15). Therefore, in these kinds of nanocomposites, we may say the 'interaction' between organic modified clay and polymer matrix actually means the contact between two organic matters and this kind of interaction have been recently evaluated by Katti et al. (16) using both photoacoustic Fourier Transform Infrared (FTIR) and computational techniques. Nevertheless, for the polymer containing functional groups (such as -OH in PVA, C=O in PA-6), it is possible for these electron withdrawing groups to interact the hydrophilic surface of clay via H-bonding. Unfortunately, to the best of our knowledge, there is no detail analysis on this subject.
In this work, we choose the typical rod-like silicate, attapulgite (AT), to prepare polymer/AT nanocomposites. Former studies of polymer/AT nanocomposites were mainly on in situ polymerization, mechanical, and crystalline properties (17-22). Poly(acrylonitrilc-co-methylacry-late) (P(AN-MA)), as a commercially important copolymer, has drawn attention for many years. It contains two different side functional groups in the polymer chains (i.e., nitrile group and carboxyl group). However, there was no detail report on the interfacial interaction mechanism of P(AN-MA)/AT nanocornposite. The main object of this work is to investigate H-bond formed on the surface of unmodified AT and the corresponding dynamic mechanical and crystalline properties of the nanocornposite.
Our previous work has investigated the rheological properties of PAN/AT solution, showing that there may be some interactions between PAN and AT to form local network structure (23). In this article, P(AN-MA) copolymer solutions with various AT content were prepared in dimethylsulfoxide (DMSO) and then casted into films. The H-bonding behavior between AT and the copolymer chains was investigated by FTIR spectra. IR spectroscopy has been proved especially helpful in determining the stability and type of bridges between the organic and inorganic phases (24). The H-bond effect on the dynamic mechanical properties and crystallization behavior of the nanocomposites were subsequently studied by dynamic mechanical analysis (DMA) and wide-angle X-ray diffraction (XRD), respectively.
P(AN-MA) copolymer, used in this study was purchased from Sinopec Shanghai (Jin Shan) Petro-Chemical Co. The copolymer (Acrylonitrile: Methacrylate (MA) = 96:4) were synthesized by a continuous free radical polymerization technique with weight-average molecular weight ([M.sub.w]) in the range of 70,000-80,000. The pristine AT powder (commercial name "Atta-Gel200") was supplied by Jiangsu Junda Attapulgite Material Co., (purity >90%). A maximum coarseness of AT powder was about 200 mesh. DMSO (analytically pure) was purchased from Shanghai Wulian Chemical Industry Co., and deionized water was used in the study.
The crude AT powder was pretreated with hydrogen peroxide solution and stirred with deionized water for 12 hr. The resulting slurry was then diluted and gravitational sedimentated for 24 hr to get rid of impurities. The purified AT was subsequently activated in 1 mol*[L.sup.-1] hydrochloric for 5 hr and the suspension was separated through filtration and washed with deionized water until pH value retained around 7.
Before the nanocomposites preparation, HC1 activated AT powder was dispersed in DMSO and subjected to son-ication in a bath sonicator for 4 hr (KUDOS-SK250LH, 59 kHz) at room temperature. Optically homogeneous AT/DMSO dispersion was added to the PAN/DMSO solution. The nanocomposites were prepared by mechanical stirring at 60[degrees]C for 8 hr and the polymer mass fraction of the solution was controlled at 16 wt% (25).
The nanocomposite films with 0, 1, 2, 3, and 5 wt% AT, which were named as AT0 (Neat copolymer), ATI, AT2, AT3, and AT5, respectively, were obtained by casting the blend solution on glass substrates. The excess amount of solvent was evaporated in blast oven at 80[degrees]C for 2 hr and then peeled off from those glass substrates after cooled in the air. The films were further dried in high vacuum oven for 24 hr. To remove residual DMSO and other unknown impurities on the surface of the film, all samples were successively washed by acetone and then dried in 50[degrees]C vacuum oven for about 30 min.
FTIR spectra were recorded on a Nicolet 8700 FTIR spectrometer at a resolution of 4 [cm.sup.-1] by a deuterate triglycine sulfate detector. The spectra for all samples were characterized in the film transmission mode. Thirty two scans were conducted to achieve an adequate signal-to-noise ratio. 'Automatic Self-Deconvolution' in OMNIC 7.0 software has been used to preliminarily identify the number of peaks in one overlapped peak. Curve fitting for the spectra was performed by 'Peak Resolve' program using Gaussian-Lorentzian function in OMNIC 7.0 software and the iteration was repeated until a best fit was obtained.
Dynamic mechanical properties were performed with the film extension mode at frequency of 10Hz on TA Q800 instruments. The measurements were done from 30-160[degrees]C with a heating rate 2[degrees]C/min. The thickness of the specimen was in the range of 0.01-0.03 mm.
A Rigaku (Japan) DMAX-2550 X-ray diffractometer (XRD) was used to investigate the crystallization behavior of the film samples. Each film has been washed and aged in 50[degrees]C oven for 5 min. The X-ray beam was Cu-[K.sub.[alpha]] ([lambda] = 0.154 nm) radiation, operated at 40 kV and 300 mA. The scanning angle (2[theta]) ranged from 0 to 60[degrees]. The data were collected at 0.02[degrees] intervals with counting for 0.12 sec at each step.
RESULTS AND DISCUSSION
Characterization of H-bond in P(AN-MA)IAT Nanocomposites Through FTIR Analysis
The FTIR spectrum of HC1 activated AT nanoparticles is shown in Fig. 1. The absorption peaks at 3421-3614 [cm.sup.-1] are hydroxyl groups (--OH) stretching bands of coordinated water in the octahedral structure tunnels of AT. The peak at 1642 [cm.sup.-1] corresponds to the typical bending vibration of zeolite water (26). The sharp absorptions at 1029 [cm.sup.-1] and 998 [cm.sup.-1] are due to bending and the stretching vibration of Si--O--Si bond (16). The spectra of P(AN-MA) copolymer and the nanocomposites are displayed in Fig. 2. The peak at 2243 [cm.sup.-1] is a typical -CN absorption (27) and the peak at 1727 [cm.sup.-1] is attributed to the carboxyl groups (C=0) stretching, which confirms the MA in the copolymer (28), (29), It is interesting to find that neither the peak position nor the peak widths showed any notable difference for -CN stretching vibration. From literatures on PAN/clay investigations (27),(30), some proposed that hydrogen bond interaction between -CN and -OH may exist, however, there was no direct spectrum evidence on this postulation. Shin and Kondo have characterized the H-bond in PAN/celluloses blend and found 'hydroxyl region-chemistry' effect', which actually indicated, it was possible to form H-bond between two groups but not always the case (31). In this article, the remarkable peak shift of the zeolite water vibrations at 1642 [cm.sup.-1] has also been observed, we deduce that this kind of peak shift could not be attributed to the H-bond formed between -CN and -OH groups on AT nanorod, in other words, -CN in the copolymer may not participate in H-bonding engagement. Instead, it can be found in Fig. 2 that the most obvious changes of the spectra locate in two regions: 3800-3050 [cm.sup.-1], 1800-1550 [cm.sup.-1] (labeled by frame) and we deduce these two regions may characterize the H-bonds in the nanocomposites and will be discussed in detail in the following paragraphs.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
It is well-known that carbonyl region in the FTIR spectra is most applicable to analysis of the H-bond in polymer matrix (32). Figure 3 shows the spectra of C=O of the copolymer and the nanocomposites that are extracted from Fig. 2. In the stack curves, we note the C=O peaks shift significantly towards lower wavenumber and the C=O absorbance profile become broad when compared with that of the neat copolymer. This is the direct evidence for the H-bond forming between C=O and the -OH in the matrix (in Fig. 4).
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
As stated above, H-bond will broaden the absorption peak profile and the overlapped peaks can be further identified by peak resolution. The extent of the C=O participating in H-bond can be expressed by H-bond index (HBI), which is defined as the ratio of two carbonyl absorbance peaks area.
HBI = [[[C.sub.bonded][[epsilon].sub.bonded]]/[[C.sub.free][[epsilon].sub.free]]] = [[A.sub.bonded]/[A.sub.free]] (1)
where C and [epsilon] represent concentration of carboxyl groups and extinction coefficient of hydrogen bonding effect, respectively. A is the absorbance peak area of characteristic band, which is calculated by the 'Peak Resolve' program. The subscript 'bonded' and 'free' assign to the H-bonded and non-H-bonded carbonyl, respectively. Generally speaking, the value of [[epsilon].sub.bonded]/[[epsilon].sub.free] is between 1 and 1.2, and is assumed to be one in this study. The deconvolution and Gaussian-Lorentzian functions were used in our work to separate overlapping bands. For simplicity in data manipulation, the straight baselines were adopted and half-widths and intensities were allowed to vary in the iteration process.
The results of peak resolution for all nanocomposites samples and the change of HBI value with increasing clay content are illustrated in Figs. 5 and 6, respectively. In Fig. 6, the degree of H-bond of carboxyl groups in nanocomposites monotonically decreases with increasing content of AT in the matrix. This phenomenon is interesting and contrary to the previous conclusion for the H-bond in nanocomposite system (33). We deduce that the aggregation effect of nanorods will greatly influence H-bond behavior in the nanocomposite. It has been understood that the hydroxyl groups of AT are mainly distributed at grooves and tunnels, and the nanometer scale silicate dispersed in the matrix result in some interfacial areas, which provide sites to form H-bond with carboxyl groups (1), (15). However, parallel aggregation of the AT nanoparticles will not only reduce the specific area, but also shield hydroxyl groups located at the grooves. Therefore, the number of available sites for H-bonding decrease and this may be the reason for the HBI values of AT2, AT3, and AT5 are smaller than that of ATI.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Following the discussion above, the peak profiles ranged from 3100 to 3800 [cm.sup.-1], which represents various kinds of hydroxyl groups in AT nanorod crystal, were further investigated. In Fig. 7, three typical peaks of hydroxyl groups are shown in the pure AT curve (labeled by dash line). The peak (I) at 3614 [cm.sup.-1] is due to the vibration of hydroxyl group attached to magnesium. The peak (II) at 3553 [cm.sup.-1] is assigned to the vibration of bound water, located at the edge of internal pore and surface grooves, and the peak (III) at 3422 [cm.sup.-1] is ascribed to hydroxyl groups of crystal water inside (34). We observed that the profiles of peak (I)-(III) become board in nanocomposite samples. This is because when H-bonds were formed between proton and oxygen atom of car-bonyl, the proton will be closer to the oxygen atom and then O--H bond length will increase with deceasing of bond energy. Although it is difficult to define peak shift amplitude due to the diffused profile, it still can be seen that the curves of ATI and AT2 have almost the same profiles, while the curve of AT5 clearly exhibits the three typical peaks that represent--OH groups located along the edges of the AT nanorod. This may be another evidence for the lowest HBI of the AT5 sample. It is known that nanoclay aggregation should be diminished both in length and diameter with long time stirring and sonication, however, the smaller absorbed impurities, which cannot be completely removed in preparing process, acted as a linkage and stimulated clay aggregation in the matrix. This phenomenon can be more obvious in the sample with high AT content. Another reason for the reaggregation of AT may be the preparation protocol. In this case, direct blending may lead to inefficient dispersion of large amount clay in the viscous matrix. Furthermore, we deduce that the aggregation will subsequently affect other properties of the nanocomposite.
[FIGURE 7 OMITTED]
H-bond Effect on Dynamic Mechanical Properties of the Nanocomposites
DMA was utilized to investigate the thermal and mechanical properties of P(AN-MA)/AT nanocomposite films. It has been understood that DMA can provide information on clay dispersion, filler-matrix and filler-filler interactions in polymer matrix (10). The storage modulus of neat copolymer and the nanocomposites were measured in heating process. The storage moduli vs. temperature are shown in Fig. 8A. It can be seen that the storage moduli for all nanocomposite samples follow a clear trend, i.e., storage modulus monotonically increase with increasing AT content. Compared with the neat P(AN-MA) copolymer sample, about 40% increment of storage modulus was observed and it reached to 3.93GPa at 5 wt% AT. In addition, we also found that the onset values of plateau modulus for P(AN-MA)/AT nanocomposites are larger than that of the control sample and increased with increasing AT content. It has been known that the storage modulus characterizes the elastic property of the nanocomposites. In this work, the elastic property of the soft polymer matrix will be enhanced when cooperated with rigid AT nanoparticles. The result is consistent with previous reports on the dynamic mechanical properties of other polymer/clay nanocomposites (35) and we think the dispersion state of AT in PAN matrix may not greatly influence the storage modulus.
[FIGURE 8 OMITTED]
Heat distortion temperature (HDT) plays an important role in the characterization of engineering thermoplastics and it is used to describe heat resistance of a polymeric material under applied constant stress. In the standard testing method, the HDT is defined by ASTM standards (ASTM D648-672) as the temperature, at which the deflection of a standard specimen reaches 0.25 mm under an applied maximum stress of either 0.46 or 1.82 MPa. Apart from this method, DMA data of storage modulus vs. temperature can also provide the same information, which was developed by Scobbo (36). According to the reference, the HDT is the temperature, at which the storage modulus reaches 1 GPa or 250 MPa. In our work, 1 GPa was chose to determine the HDT value of neat copolymer and nanocomposite samples with different AT contents. As seen in Fig. 9, the experimental value of HDT gradually increases with increasing AT content, and in case of AT5, the value increase upto 100.8[degrees]C. The enhancing of HDT is due to the heat resistance effect brought by AT.
[FIGURE 9 OMITTED]
The ratio E"/E', which is called as loss factor (tan[delta]), was obtained from DMA and employed as a measurement of glass transition temperature ([T.sub.g]) for all of the samples in our work. Table 1 lists [T.sub.g] for all PAN/AT nanocomposite. It can be found the nanocomposite has the highest [T.sub.g] value when 1 wt% AT were incorporated. However, with further addition of AT, the [T.sub.g] values remain almost unaltered, indicating the repulsive interaction between two phases (37). Former reports have focused on two-dimensional (2-D) clay and the layer-constrained-polymer chains behavior (38), (39) and it has been reported that the [T.sub.g] value are largely dependent on the interfacial interactions and the matrix network density (19). In this case, high [T.sub.g] value of AT1 reflects the relatively strong interaction between the fillers and the matrix. The homogeneously dispersed AT nanoparticles may act as a barrier for chains motion in the matrix. What's more, it can be seen in Fig. 8B that the amplitudes of the tan[delta] peaks significantly suppressed and the peak width attained the maximum value when the clay content was 1 wt%. This is because the segments in the vicinity of nanorods will be confined by H-bond, thus their relaxation occurs at higher temperatures. In contrast, less confined segments that are distant from the nanorods will exhibit relaxation behavior similar to that of the bulk (19). Therefore the relaxation of segments in the system will span a wider temperature range. However, it should be pointed out that this kind of peak shift is not very significant, this may probably due to the few carboxyl groups in the copolymer chains. Combining with the conclusion of FTIR analysis, the DMA results also showed that H-bond in the nanocomposites may act as a junction and constrained motion of the chains. In other words, when the fraction of nanofillers is lower than a critical value for aggregation, the H-bond effect may become dominant in the confined motion of polymer chains.
TABLE 1. Glass transition temperatures of the copolymer and the nanocomposites with different AT contents. Samples AT0 AT1 AT2 AT3 AT5 [T.sub.g] ([degrees]C) 108.3 111.9 107.5 109.7 108.5
H-bond Effect on Crystallinity of the Nanocomposites
The exfoliation status of AT and crystallization in the series nanocomposite films were characterized by XRD. In Fig. 10, it can be observed that the location of the characteristic peak position of AT (110, 2[theta] = 8[degrees]) in AT1-AT5 samples remain unchanged. This is quite different from layered clay such as montmorillonite, whose diffraction peak shift significantly toward higher theta value when clay aggregated and this difference can be attributed to the different crystal geometry structure. For AT nanorods, (110) crystal plane is owing to intrinsic axis structure and aggregation, and exfoliation will not greatly affect the location of diffraction peak (110) in XRD pattern . This also indicates that XRD is not an efficient method to investigate exfoliation of rod-like silicates as that in layered silicates.
[FIGURE 10 OMITTED]
The crystallization behavior of P(AN-MA)/AT nanocomposites was also investigated by XRD. It is well known that P(AN-MA) chains display only two-dimensional order without periodicity along the chain axis, therefore, it may form paracrystalline morphology. In the spectrum, typical diffraction peak of P(AN-MA) located at about 20 = 17[degrees], which is indexed to (200) crystal plane (41). The diffused peaks at about 30[degrees] correspond to the distortion and irregularity of the crystal structure. Figure 11 shows the relationship between HBI and the crystallinity of the nanocomposites. The results differs from those, previously reported by Peng and chen, who believed that small amount of AT will act as a nucleation agent in the matrix and enhance the crystallinity of PVA matrix (17). In this case, we also recognize nucleation effect of AT in the polymer matrix, however, the relatively significant H-bond effect aforementioned should be taken into consideration and these two factors can be considered as a balance in the crystallization process of nanocomposite. From the previous literature about crystallinity of polymer blends, we have known that the formation of hydrogen bonds between the polymer chains usually induces compatibility as well as greatly suppress the crystallization of the component (42). Therefore, we speculate that H-bond interaction in AT1 may prevent the copolymer chains from moving into crystal lattices or the closely hexagonal packing process (43), thus decrease crystallinity of the matrix. However, as to the AT3 and AT5 samples, polymer chains confinement will be weakened as decreasing H-bond interaction.
[FIGURE 11 OMITTED]
In this article, P(AN-MA)/AT nanocomposites films with different clay content were successfully prepared. FTIR spectra showed the H-bond interaction between AT and the copolymer. It was found that--CN in PAN did not form H-bond with hydroxyl groups of AT and the main H-bond position located at C=O of the second monomer MA. HBI results indicated that the large amount of AT nanorods in nanocomposites would diminish specific surface area, thus decrease compatibility between two components. Results of DMA demonstrated that the addition of AT enhanced storage modulus of the polymer matrix and HDT value, whereas [T.sub.g] value reached to a maximum when 1 wt% AT was added. These results are quite consistent with the analysis in FTIR, i.e., homogeneously dispersed AT (1 wt% in our case) nanorods may result in a relatively strong H-bond interaction between the nanorods and the matrix. Crystallization behavior of the nanocomposites indicated that crystallinity of the matrix was also related to the H-bond interaction between the organic and inorganic phases.
(1.) J. Ma, E. Bilotti, T. Peijs, and J.A. Darr, Eur. Polym. J., 43, 4931 (2007).
(2.) A. Neaman and A. Singer, Geoderma, 123, 297 (2004).
(3.) K. William, M. Goertzen, and R. Kessler, Compos: Part A Appl. Sci., 39, 761 (2008).
(4.) K. Liang, G. Li, H. Toghiani, J.H. Koo, and C.U. Pittman Jr., Chem. Mater., 18, 301 (2006).
(5.) K.M. Dean, M.D. Do, E, Petinakis, and L. Yu, Compos. Sci. Technol., 68, 1453 (2008).
(6.) R.K. Bharadwaj, Macromolecules, 34, 9189 (2006).
(7.) J.S. Shelleya, P.T. Matherb, and K.L. DeVriesc, Polymer, 42, 5849 (2001).
(8.) A. Dasari, Z.Z. Yu, Y.W. Mai, and J.K. Kim, Nanotehnology, 19, 055708/1 (2008).
(9.) Z. Peng and L.X. Kong, Polym. Degrad. Stab., 92, 1061 (2007).
(10.) Y.Q. Rao and J.M. Pochan, Macromolecules, 40, 290 (2007).
(11.) D.W. Schaefer and R.S. Justice, Macromolecules, 40, 8501 (2007).
(12.) H.J. Lee, H.W. Choi, K.J. Kim, and S.C. Lee, Chem. Mater., 18, 5111 (2006).
(13.) H.D. Wagner, Nat. Nanotechnol., 2, 742 (2007).
(14.) C. Anna, A.C. Balazs, C. Singh, and E. Zhulina, Macromolecules., 31, 8370 (1998).
(15.) E. Bilotti, H.R. Fischer, and T. Peijs, J. Appl. Polym. Sci., 107, 1116 (2008).
(16.) K.S. Katti, D. Sikdar, D.R. Katti, P. Ghosh, and D. Verma, Polymer, 47, 403 (2006).
(17.) Z.Q. Peng and D.J. Chen, .J. Polym. Sci. Part A: Polym. Phys., 44, 534 (2006).
(18.) Y.S. Liu and Z.X. Su, Polym. Int., 57, 306 (2008).
(19.) Y.Z. Pan, Y. Xu, L. An, H.B. Lu, Y.L. Yang, W. Chen, and S. Nutt, Macromolecules, 41, 9245 (2008).
(20.) L. Shen, Y.J. Lin, Q.G. Du, W. Zhong, and Y.L. Yang, Polymer, 46, 5675 (2005).
(21.) S.Q. Lai, T.S. Liu, X.J. Liu, and R.G. Lv, Macromol. Mater. Eng., 289, 916 (2004).
(22.) S.Q. Lai, T.S. Liu, X.J. Liu, and R.G. Lv, Macromol. Mater. Eng., 290, 195 (2005).
(23.) H. Yin, D. Mo, and D.J. Chen, J. Polym. Sci. Part A: Polym. Phys., 47, 945 (2009).
(24.) R.L. Frost and E. Mendelovici, J. Colloid. Interface Sci., 294, 47 (2006).
(25.) H.G. Chae, M.L. Minus, and S. Kumar, Polymer, 47, 3494 (2006).
(26.) C.H. Wang, M.L. Auad, N.E. Marcovich, and S. Nutt, J. Appl. Polym. Sci., 109, 2562 (2008).
(27.) R. Fernandez-Saavedra and P. Aranda, Adv. Func. Mater., 14, 77 (2004).
(28.) Y. Sugahara, S. Satokawa, K. Kuroda, and C. Kato, Clay Clay Miner., 36, 343 (1988).
(29.) J.C. Chen and I.R. Harrison, Carbon., 40, 25 (2002).
(30.) L.W. Ji, C. Saquing, S.A. Khan, and X.W. Zhang, Nanotechnology, 19, 085605 (2008).
(31.) J.H. Shin and T. Kondo, Polymer, 39, 6899 (1998).
(32.) X.H. Dai, J. Xu, X.L. Guo, Y.L. Lu, D.Y. Shen, N. Zhao, X.D. Luo, and X.L. Zhang, Macromolecules, 37, 5615 (2004).
(33.) H.C. Kuan, C.C.M. Ma, W.P. Chuang, and H.U. Su, J. Polym. Sci. Part A: Polym. Phys., 43, 1(2005).
(34.) J.H. Huang, Y.F. Liu, Q.Z. Jin, and X.G. Wang. Spectrosc. Spect. Anal., 27, 408 (2007).
(35.) B.L. Pan, Q.F. Yue, J.F. Ren, H.G. Wang, L.Q. Jian, J.Y. Zhang, and S.R. Yang, Polym. Test., 25, 384 (2006)
(36.) J.J. Scobbo Jr. and C.R. Hwang, Polym. Eng. Sci., 34, 1744 (1994).
(37.) L. An, Y.Z. Pan, X.W. Shen, H.B. Lu, and Y.L. Yang. J. Mater. Sci., 18, 4928 (2008).
(38.) A. Usuki, A. Koiwai, Y. Kojima, M. Kawasumi, A. Okada, T. Kurauchi, and O. Kamigaito, J. Appl. Polym. Sci., 55, 119 (1995).
(39.) S.D. Burnside and E.P. Giannelis, J. Polym. Sci. Part A: Polym. Phys., 38, 1595 (2000).
(40.) X.P. Yuan, C.C. Li, G.H. Guan, X.Q. Liu. Y.N. Xiao, and D. Zhang, J. Appl. Polym. Sci., 103, 1279 (2007).
(41.) H.Q. Hou, J.J. Ge, J. Zeng, Q. Li, D.H. Reneker, A. Greiner, and S.Z.D. Cheng, Chem. Mater., 17, 967 (2005).
(42.) Y. He, B. Zhu, and Y. Inoue, Prog. Polym. Sci., 29, 1021 (2004).
(43.) Z. Bashir, Polymer., 33, 4304 (1992).
Han Yin, Huifang Chen, Dajun Chen
State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, College of Materials Science and Engineering, Donghua University, Shanghai 200051, China
Correspondence to: Dajun Chen: e-mail: email@example.com
Contract grant sponsor: National Basic Research Program (973 Program); contract grant number: 2006CB606505; contract grant sponsor: National Natural Foundation of China; contract grant number: 50333050; contract grant sponsor: Shanghai Fundamental Theory Program; contract grant number: 07DJ14002; contract grant sponsor: Programme of Introducing Talents of Discipline to Universities: contract grant number: 111-2-04.