2018, Vol. 29 
In last decade, polymer nanocomposites containing graphene, its derivatives graphene oxide (GO) [1] or reduced GO (rGO) have attracted great research interest owing to their high aspect ratio, remarkable mechanical, electrical, and thermal properties. These unique properties make them suitable for a wide range of valuable applications, such as mechanical reinforced polymer nanocomposites [2], electronic devices [3], energy storage [4], and sensors [5]. However, graphene and its derivatives are not compatible with polymers and tend to agglomerate, compromising the properties of the composites. Therefore, it is necessary to functionalize these high-performance fillers using organic molecular or polymer to improve their compatibility with polymers.
Up to date, composites containing graphene or its derivatives with polar polymer matrix, such as poly(vinyl alcohol) (PVA) [6, 7], poly(methyl methacrylate) (PMMA) [8, 9], polyamide (PA) [10, 11], and polyurethane (PU) [12, 13] have been extensively studied. However, reports on adding graphene or its derivatives to non-polar polyolefins, for example polyethylene (PE) [14-18] and polypropylene (PP) [19-23], are relatively rare. This is due to the fact that the dispersion of graphene and its derivatives within nonpolar polymers by solution or melt blending is a significant challenge since these nanosheets are thermodynamically driven to aggregate and exhibit poor interfacial adhesion comparing to those in polar polymers. In order to get graphene and its derivatives dispersed well in polyolefins, the surface modification is highly desired. Usually, there are two compatibilization strategies employed to improve the interfacial interaction between the fillers and polyolefins. GO can be alkylated by alkylamine [14, 24] or silane [25] then blended with polyolefins directly. The other way is that the functionalization of polymer matrix is first carried out then blended with graphene or its derivatives with or without chemical functionalization [15, 16, 18, 20, 23]. Their good compatibility is realized by chemical reaction, π-π stacking interaction, and polar-polar interaction. Previously, functionalization of graphene or its derivatives using organic reactive molecules then grafted with functionalized polyolefins has been proposed [15, 20]. The grafted polyolefins could effectively enhance the interfacial interactions and then improve the compatibility.
In this study, we presented an effective protocol to covalently functionalize GO (fGO) with HDPE-g-(MAH-co-St). GO was first modified by macromolecules with reactive epoxide sites through dual monomers (GMA and St) grafting method, and then it was covalently linked with HDPE-g-(MAH-co-St) through the reaction between the epoxides and anhydrides. The fGO was then melt blended with HDPE. The dispersion of GO in the fGO/HDPE nanocomposites and the mechanical properties of the nanocomposites were discussed. The dual monomers St and MAH grafted HDPE modified GO (GO-g-HDPE-g-(MAH-co-St)) has St block and anhydride group in MAH, which has chemical affinity towards styrene based polymers and can react with many engineering plastics with terminal functional group such as PA6 with terminal amine group. Thus, the GO-g-HDPE-g-(MAH-co-St) has great potential to be used to compatibilize and reinforce HDPE/styrene-based polymers/engineering plastics ternary blends in which each component represents a typical plastic of polyolefin (PE, PP, etc.), engineering plastic (PA6, PC, PET, etc.), and styrene polymer (PS, SEBS, ABS, etc.), respectively. These three kinds of polymers constitute nearly 80% of the world's plastics and occupy the largest share of the scrap or waste plastics and are the most recycled polymeric materials.
Here, GO was prepared according to a modified Hummers' method [26]. HDPE-g-(MAH-co-St) was prepared according to the method we reported previously [27-29], which was synthesized by multi-monomer free radical melt grafting. The synthetic pathway was illustrated in Scheme 1. First GO was functionalized with m-isopropenyl-α, α'-dimethylbenzylisocyanate (TMI) through the reaction between hydroxyl and carboxyl groups of GO with −NCO group of TMI [9, 30], resulting in the formation of GO-TMI. As previously reported [31], GMA can be easily grafted onto basal GO nanosheets by free radical polymerization. Moreover, GMA possesses epoxide group that can react with anhydride group [32]. Through dual monomers GMA and St free radical grafting, GO-g-(GMA-co-St) was obtained in the presence of GMA, St and initiator AIBN with the mole ratio of GMA/St 1:1. The role of St in the polymerization reaction is to modulate the quantity of grafting copolymer poly(GMA-co-St) [33, 34]. Finally, GO-g-HDPE-g-(MAH-co-St) was prepared through the reaction between the GO-g-(GMA-co-St) and the HDPE-g-(MAH-co-St). The covalent functionalization of GO was characterized by atomic force microscope (AFM), transmission electron microscopy (TEM), Fourier transform infrared (FTIR), X-ray diffraction (XRD), differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). The details of the experiment are provided in the Supporting information.
|
Download:
|
| Scheme 1. Synthesis route for GO-TMI, GO-g-(GMA-co-St) and GO-g-HDPE-g-(MAH-co-St). | |
HDPE (45 g) and GO-g-HDPE-g-(MAH-co-St) masterbatch (5 g, GO contents: 0, 1, 2, 5 and 10 wt%) were melting mixed to obtain fGO/HDPE nanocomposites by using a batch mixer of Rheocord Haake type. The blending temperature, screw speed and blending time were set for 180 ℃, 60 rpm and 10 min, respectively. The GO contents of the resulting fGO/HDPE nanocomposites were 0, 0.1, 0.2, 0.5 and 1 wt%. As a contrast, HDPE/HDPE-g-(MAH-co-St) = 90/10 and 0.1 wt% pristine GO was melt blended to prepare GO/HDPE nanocomposite under the same condition. Scanning electron microscopy (SEM) images of cryogenically fractured surface of the nanocomposites were taken on a JEOL JSM-7401F. Tensile test was carried out on a GOTECH-2000 universal testing machine at a crosshead speed of 50 mm/min according to ASTM D638.
Fig. 1a shows the AFM image of GO nanosheets with relatively smooth surface and the thickness of GO ranges from 0.8–1.0 nm, corresponding to monolayer sheets. As poly(GMA-co-St) grafted on the two sides of GO nanosheets, the surfaces coated with the polymer become rough and the thickness of GO-g-(GMA-co-St) (Fig. 1b) increases to 5–6 nm, indicating a high grafting ratio of poly(GMA-co-St). After HDPE-g-(MAH-co-St) has been grafted onto GO-g-(GMA-co-St) due to the reaction between anhydride group of MAH and the epoxide group of GMA, many protuberances on the nanosheet backbones are observed and further the thickness of GO-g-HDPE-g-(MAH-co-St) (Fig. 1c) increases to 8–21 nm. The changes in the thickness and morphology of the functionalized GO nanosheets clearly demonstrates the success of the different polymer chains grafting onto the surface of GO. Furthermore, Figs. 1d–f shows the TEM images of the morphology of GO, GO-g-(GMA-co-St) and GO-g-HDPE-g-(MAH-co-St), respectively. Compared with exfoliated and transparent GO nanosheet as shown in Fig. 1d, the images of the GO grafted with polymer chains show a quite different morphology. With poly(GMA-co-St) grafted (Fig. 1e), the GO nanosheets are dark and flat. While, they become darker when the HDPE-g-(MAH-co-St) grafted to GO-g-(GMA-co-St) (Fig. 1f). It also implies the successful grafting.
|
Download:
|
| Fig. 1. AFM images (top) and TEM images (below) of GO (a) (d), GO-g-(GMA-co-St) (b) (e), GO-g-HDPE-g-(MAH-co-St) (c) (f). | |
The direct evidence for modifications of GO can be further demonstrated by the FTIR spectra of GO, GO-TMI, GO-g-(GMA-co-St), poly(GMA-co-St), GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St) in Fig. 2a. Compared with the FTIR spectrum of GO, there is obvious characteristic absorption bands at 2861 cm−1 and 2926, 2954 cm−1 respectively (—CH3 asymmetric stretching vibration and C—H stretching vibration) in the spectrum of the GO-TMI, implying the incorporation of TMI groups onto GO. Comparing the spectra of the GO-g-(GMA-co-St) and free Poly(GMA-co-St), the same characteristic absorption bands appear at 3059, 906, and 845 cm−1 (epoxide group from GMA) and 758 and 702 cm−1 (mono-substituted benzene from St), suggesting the success of the grafting. In the spectra of the GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St), almost the same characteristic absorption bands appear, but the characteristic absorption bands at 3059, 906, and 845 cm−1 of the epoxide group in the GO-g-HDPE-g-(MAH-co-St) decrease obviously, indicating the reaction between the epoxide group of GO-g-(GMA-co-St) and the anhydride groups of HDPE-g-(MAH-co-St).
|
Download:
|
| Fig. 2. (a) FTIR spectra of GO, GO-TMI, GO-g-(GMA-co-St), poly(GMA-co-St), GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St); (b) XRD patterns of graphite, GO, GO-TMI, GO-g-(GMA-co-St), GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St); (c) DSC curves of GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St); (d) TGA curves of GO, GO-TMI, GO-g-(GMA-co-St), poly (GMA-co-St), GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St). | |
As shown in Fig. 2b, XRD characterization was carried out to reveal the structure change during the chemical modification of GO. Graphite shows a sharp and intense peak at 2θ = 26.6° (002) with a corresponding spacing of 0.33 nm, while GO exhibits a lower angle 2θ = 12.4° (002) with a corresponding spacing of 0.71 nm due to the oxidation of graphite. After functionalized with TMI, the parent peak of GO-TMI shifts downward to a lower angle 2θ = 8.3° with the increasing spacing of 1.06 nm, indicating the intercalation and the covalent linkage of TMI onto GO. While, in the patterns of GO-g-(GMA-co-St), there are no obvious peaks to be observed, suggesting the structural disorder and successful grafting. For GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St), their XRD patterns have similar profiles. Two peaks at about 2θ = 21.6° and 2θ = 23.9° were observed for them, corresponding to the (110) and (200) Bragg reflections of HDPE, respectively. The result implies that HDPE-g-(MAH-co-St) has been grafted onto GO-g-(GMA-co-St) successfully.
Fig. 2c depicts DSC plots for GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St). Crystallization temperature Tc and melting temperature Tm were determined by cooling and second heating process respectively. Compared with Tc = 117.0 ℃ and Tm = 131.7 ℃ of HDPE-g-(MAH-co-St), GO-g-HDPE-g-(MAH-co-St) has depressed Tc = 100.0 ℃ and Tm = 118.3 ℃. According to the reports before, when HDPE-g-MAH grafted onto GO functionalized with small molecules gamma-aminopropyltriethoxysilane [15] and HDPE crystallized on reduced graphene oxide (RGO) [35], their crystallization temperature increases by about 10 and 5.3 ℃ respectively, which are attributed to the heterogeneous nucleation induced by the 2D GO nanosheets. However, in our case, HDPE-g-(MAH-co-St) is covalently bonded to the amorphous poly(GMA-co-St) chains that grafted onto the surface of GO nanosheets. The amorphous poly(GMA-co-St) as an interlayer not only shields the heterogeneous nucleation effect of GO nanosheets but also restrict the crystallization behavior of HDPE-g-(MAH-co-St) [36-39]. Therefore, the grafted HDPE-g-(MAH-co-St) crystallizes under the confinement due to the molecular ends bonded with poly(GMA-co-St) interlayer grafted onto GO. It also indicates the success of grafting HDPE-g-(MAH-co-St) onto GO-g-(GMA-co-St).
To quantitatively determine the grafting amount of the polymer grafted onto GO nanosheets, TGA measurements were carried out. TGA curves of GO, GO-TMI, GO-g-(MAH-co-St), poly(GMA-co-St), GO-g-HDPE-g-(MAH-co-St) and HDPE-g-(MAH-co-St) under N2 atmosphere are plotted in Fig. 2d. For GO, the obvious weight loss stage (~40 wt %) below 350 ℃ is attributed to the decomposition of labile oxygen-contained groups. Due to the introduction of less stable isocyanate, the thermal stability of GO-TMI is deteriorated and the residual weight of GO-TMI at 800 ℃ is less than that of GO. There are two major weight loss stages for GO-g-(GMA-co-St). The first appearing at about 210 ℃ should be ascribed to the loss of labile oxygen functional groups. And the second between 300 and 450 ℃ corresponds to the decomposition of the grafted copolymers, which is similar to the decomposition of the corresponding free copolymer poly(GMA-co-St). Comparing to the rapid weight loss of poly(GMA-co-St) at 390 ℃, GO-g-(GMA-co-St) shows a rapid weight loss at lower temperature of 380 ℃. The reason for this might be that the chain length of the grafted polymer is much shorter than that of free poly(GMA-co-St). For GO-g-HDPE-g-(MAH-co-St), two weight loss stages at about 320–440 ℃ and 440–510 ℃ are observed. The first stage is resulted from the decompositions of the grafted poly(GMA-co-St) with a rapid weight loss at higher temperature 390 ℃ due to the protection of the outside grafted HDPE-g-(MAH-co-St). Another stage belongs to the main decomposition of the grafted HDPE-g-(MAH-co-St), which is almost the same with HDPE-g-(MAH-co-St). On the basis of residual weight of GO, GO-g-(GMA-co-St) and GO-g-HDPE-g-(MAH-co-St), the weight percent of grafted poly(GMA-co-St) and grafted HDPE-g-(MAH-co-St) onto GO nanosheets are calculated to be about 50.4 wt% and 70.4 wt%, respectively.
The pristine GO and the functionalized GO-g-HDPE-g-(MAH-co-St) masterbatch were melt blended with neat HDPE to prepare GO/HDPE nanocomposite (content of GO 0.1 wt%) and fGO/HDPE nanocomposite (contents of GO 0.1 wt%, 0.2 wt %, 0.5 wt% and 1 wt%) by using Haake mixer, respectively. In order to investigate the dispersion of the GO in the nanocomposites, the cryogenically fractured surfaces of the nanocomposites were observed by SEM. Fig. 3 shows the SEM images of the cryogenically fractured surfaces of the nanocomposites. As shown in Fig. 3a, for the GO/HDPE nanocomposite containing pristine GO 0.1 wt%, GO agglomerates are observed in the HDPE matrix (marked by the yellow circle). There is obvious interphase between GO and HDPE matrix, indicating the incompatibility and weak interfacial adhesion between GO and HDPE matrix. While, in the fGO/HDPE nanocomposite with the same content of GO 0.1%, GO is hardly to be found in Fig. 3b, implying the good compatibility between GO-g-HDPE-g-(MAH-co-St) and HDPE matrix. The SEM images of the fGO/HDPE nanocomposites containing GO 0.2 wt%, 0.5 wt% and 1 wt% are shown in Figs. 3c–e, respectively. The polymer grafted GO sheets are wrapped by the matrix and the cryogenically fractured surfaces are pretty rough. Thus, the agglomerates of GO are difficult to be found in the cryogenically fractured surfaces of the nanocomposites, suggesting the strong interfacial adhesion between the functionalized GO and the matrix.
|
Download:
|
| Fig. 3. SEM images of cryogenically fractured cross section of the nanocomposites: (a) GO/HDPE nanocomposite containing 0.1 wt% pristine GO; fGO/HDPE nanocomposite containing GO (b) 0.1 wt%, (c) 0.2 wt%, (d) 0.5 wt% and (e) 1 wt%. | |
From aforementioned morphology investigations to the nanocomposites, the good dispersion of GO nanosheets in HDPE matrix would result in potential reinforcement for mechanical properties. Fig. 4 shows the stress at break and strain at break of the GO/HDPE nanocomposite with 0.1 wt% pristine GO and fGO/HDPE nanocomposites with different GO contents including 0 wt%, 0.1 wt%, 0.2 wt%, 0.5 wt% and 1 wt%. As shown in Fig. 4a, compared to fGO/HDPE nanocomposite containing GO 0 wt%, the stress at break and strain at break of the GO/HDPE nanocomposite with 0.1 wt% pristine GO decrease from 18.8 MPa to 18.1 MPa and from 844.0% to 255.5%, respectively. The reason is that the incompatibility and weak interfacial adhesion between GO and HDPE matrix and the aggregation of GO deteriorating the energy dissipation capability, resulting in the strain at break decreasing sharply. It also can be seen from Fig. 3a that pristine GO disperses in HDPE matrix poorly. On the contrary, for the fGO/HDPE nanocomposites with the same GO content 0.1 wt%, both the stress at break and strain at break increase prominently to 23.6 MPa and 1402.6% respectively. This means that grafting HDPE-g-(MAH-co-St) onto GO can improve interfacial adhesion, which is an efficient way to enhance the interaction between the GO and HDPE matrix. Fig. 4b summarizes the stress at break and strain at break of the fGO/HDPE nanocomposites as a function of GO content. It can be found that both the stress at break and strain at break increase with increasing GO content firstly. With the content of GO up to 0.2%, the stress at break and strain at break reach to their maximum value 24.2 MPa and 1941.0%, respectively. Compared with fGO/HDPE nanocomposite containing GO 0 wt%, the stress at break is increased by 28.7% and the strain at break is increased significantly by 130%. The reason might be attributed to the strong interaction between fGO and the matrix, which enhances the stress and energy dissipation. On the other hand, as further increasing GO content from 0.5% to 1%, both the stress at break and strain at break decrease. Here, it is assumed that as further adding GO into a polymer matrix, the phenomenon of GO agglomerating occurs due to van der Waals force. Furthermore, in this work, we should pay attention to an interesting phenomenon that the improvement of strain at break is more pronounced than that of the stress at break. The crystallinity of the HDPE grafted onto GO is relatively less due to the confined crystallization indicated by DSC analysis in Fig. 2c. Thus, the matrix HDPE is easier to be stretched. This might be the reason for the strain at break improved more significant than the stress at break.
|
Download:
|
| Fig. 4. (a) The stress-strain curves of GO/HDPE and fGO/HDPE nanocomposites; (b) The stress at break and strain at break of fGO/HDPE nanocomposites with increasing the GO contents. | |
In summary, we have developed an effective strategy to covalently graft HDPE-g-(MAH-co-St) onto GO. The GO surface modified with plenty of epoxide groups by dual monomers grafting method first, then HDPE-g-(MAH-co-St) was grafted onto it through the reaction between epoxide groups and anhydride groups. This modification of GO ensures the high grafting ratio of HDPE-g-(MAH-co-St) up to 70.4 wt%, which enhances the interfacial interaction between GO nanosheets and the matrix HDPE. With the addition of only 0.2 wt% functionalized GO, the resulting nanocomposite exhibits a prominent reinforcement effect. The stress at break and strain at break increased by 28.7% and by 130%, respectively. In addition, polyolefin grafted GO also has maleic anhydride grafted, so that it has the potential to be used as a compatibilizer for polyolefin/enngineering plastics blends. Therefore, this work provides a universal way for the modification of GO with function groups and high grafting efficiency, which can be used not only for the preparation of GO/non-polar polyolefin nanocomposites, but also for the compatibilization of GO/multi-component polymer system nanocomposites.
AcknowledgmentsWe sincerely thank the financial support from the National Natural Science Foundation of China (No. 51633003) and State Key Laboratory of Organic-Inorganic Composites, Beijing University of Chemical Technology (No. OIC-201601006).
Appendix A. Supplementary dataSupplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cclet.2017.06.001.
| [1] |
D.R. Dreyer, S. Park, C.W. Bielawski, et al., Chem. Soc. Rev. 39(2010) 228-240. DOI:10.1039/B917103G |
| [2] |
J.R. Potts, D.R. Dreyer, C.W. Bielawski, et al., Polymer 52(2011) 5-25. DOI:10.1016/j.polymer.2010.11.042 |
| [3] |
X.Y. Qi, C.L. Tan, J. Wei, et al., Nanoscale 5(2013) 1440-1451. DOI:10.1039/c2nr33145d |
| [4] |
Y.Q. Sun, G.Q. Shi, J. Polym. Sci. Part B:Polym. Phys. 51(2013) 231-253. DOI:10.1002/polb.23226 |
| [5] |
X. Huang, X. Qi, F. Boey, H. Zhang, Chem. Soc. Rev. 41(2012) 666-686. DOI:10.1039/C1CS15078B |
| [6] |
X. Zhao, Q. Zhang, D. Chen, et al., Macromolecules 43(2010) 2357-2363. DOI:10.1021/ma902862u |
| [7] |
C. Lin, Y.T. Liu, X.M. Xie, Aust. J. Chem. 67(2014) 121-126. DOI:10.1071/CH13339 |
| [8] |
J.R. Potts, S.H. Lee, T.M. Alam, et al., Carbon 49(2011) 2615-2623. DOI:10.1016/j.carbon.2011.02.023 |
| [9] |
Y. Yang, C.E. He, W. Tang, et al., J. Mater. Chem. A 2(2014) 20038-20047. DOI:10.1039/C4TA04543B |
| [10] |
R. Rafiq, D. Cai, J. Jin, et al., Carbon 48(2010) 4309-4314. DOI:10.1016/j.carbon.2010.07.043 |
| [11] |
Z. Xu, C. Gao, Macromolecules 43(2010) 6716-6723. DOI:10.1021/ma1009337 |
| [12] |
H. Kim, Y. Miura, C.W. Macosko, Chem. Mater. 22(2010) 3441-3450. DOI:10.1021/cm100477v |
| [13] |
K. Nawaz, U. Khan, N. Ul-Haq, et al., Carbon 50(2012) 4489-4494. DOI:10.1016/j.carbon.2012.05.029 |
| [14] |
T. Kuila, S. Bose, C.E. Hong, et al., Carbon 49(2011) 1033-1037. DOI:10.1016/j.carbon.2010.10.031 |
| [15] |
Y. Lin, J. Jin, M. Song, J. Mater. Chem. 21(2011) 3455-3461. DOI:10.1039/C0JM01859G |
| [16] |
H. Kim, S. Kobayashi, M.A. AbdurRahim, et al., Polymer 52(2011) 1837-1846. DOI:10.1016/j.polymer.2011.02.017 |
| [17] |
Y. Li, Polymer 52(2011) 2310-2318. DOI:10.1016/j.polymer.2011.03.025 |
| [18] |
A.A. Vasileiou, M. Kontopoulou, A. Docoslis, ACS Appl. Mater. Interfaces 6(2014) 1916-1925. DOI:10.1021/am404979g |
| [19] |
Y. Huang, Y. Qin, Y. Zhou, et al., Chem. Mater. 22(2010) 4096-4102. DOI:10.1021/cm100998e |
| [20] |
M.C. Hsiao, S.H. Liao, Y.F. Lin, et al., Nanoscale 3(2011) 1516-1522. DOI:10.1039/c0nr00981d |
| [21] |
P. Song, Z. Cao, Y. Cai, et al., Polymer 52(2011) 4001-4010. DOI:10.1016/j.polymer.2011.06.045 |
| [22] |
Y. Cao, J. Feng, P. Wu, J. Mater. Chem. 22(2012) 14997-15005. DOI:10.1039/c2jm31477k |
| [23] |
B. Yuan, C. Bao, L. Song, et al., Chem. Eng. J. 237(2014) 411-420. DOI:10.1016/j.cej.2013.10.030 |
| [24] |
X. Yang, T. Mei, J. Yang, et al., Appl. Surf. Sci. 305(2014) 725-731. DOI:10.1016/j.apsusc.2014.03.184 |
| [25] |
X. Yang, X. Wang, J. Yang, et al., Chem. Phys. Lett. 570(2013) 125-131. DOI:10.1016/j.cplett.2013.03.069 |
| [26] |
W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80(1958) 1339. DOI:10.1021/ja01539a017 |
| [27] |
Y. Li, X.M. Xie, B.H. Guo, Polymer 42(2001) 3419-3425. DOI:10.1016/S0032-3861(00)00767-9 |
| [28] |
X.M. Xie, Y. Li, J. Zhang, et al., Acta Polym. Sin. 1(2002) 7-12. |
| [29] |
Y. Li, Y. Li, X.M. Xie, Acta Polym. Sin. 4(2011) 347-353. |
| [30] |
S. Stankovich, R.D. Piner, S.T. Nguyen, et al., Carbon 44(2006) 3342-3347. DOI:10.1016/j.carbon.2006.06.004 |
| [31] |
L. Kan, Z. Xu, C. Gao, Macromolecules 44(2010) 444-452. |
| [32] |
H. Lee, K. Neville, Handbook of Epoxy Resins, McGraw-Hill, New York, 1967.
|
| [33] |
X.M. Xie, N.H. Chen, B.H. Guo, et al., Polym. Int. 49(2000) 1677-1683. DOI:10.1002/(ISSN)1097-0126 |
| [34] |
J.L. Li, X.M. Xie, Polymer 53(2012) 2197-2204. DOI:10.1016/j.polymer.2012.03.035 |
| [35] |
S. Cheng, X. Chen, Y.G. Hsuan, et al., Macromolecules 45(2012) 993-1000. DOI:10.1021/ma2021453 |
| [36] |
P.Q. Yu, X.M. Xie, Z. Wang, et al., Polymer 47(2006) 1460-1464. DOI:10.1016/j.polymer.2005.12.059 |
| [37] |
F. Zhang, X.M. Xie, J. Yuan, et al., Acta Polym. Sin. 4(2011) 354-359. |
| [38] |
P.Q. Yu, L.T. Yan, N. Chen, et al., Polymer 53(2012) 4727-4736. DOI:10.1016/j.polymer.2012.08.038 |
| [39] |
N. Chen, L. T. Yan, X.M. Xie, Macromolecules 46(2013) 3544-3553. DOI:10.1021/ma4005692 |