Preparation of Crosslinked Highdensity Polyethylene Foam Using Supercritical CO 2

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 Preparation of Crosslinked Highdensity Polyethylene Foam Using Supercritical CO 2 Hongfu Zhou*, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen and Shanglin Jin School of Materials and Mechanical Engineering, Beijing Technology and Business University, Beijing 100048, People s Republic of China Received: 20 April 2016, Accepted: 21 September 2016 SUMMARY Different content of dicumyl peroxide (DCP) acting as a crosslinking agent was mixed with high-density polyethylene (HDPE) in a Haake internal mixer to improve the viscoelasticity and foamability of HDPE. The crosslinked HDPE samples were foamed in a high pressure stainless steel autoclave using CO 2 as the physical blowing agent. The molecular weight, crystallization behavior and rheological properties of various HDPE samples were examined by gel permeation chromatography, differential scanning calorimetry, rotational rheometer, and torque rheometer, respectively. The foaming properties of various samples were characterized by scanning electron microscope and densimeter. It was found that with the increasing content of DCP, the molecular weight, crystallization temperature, complex viscosity, and storage modulus of HDPE increased and the crystallization degree of HDPE decreased. When 0.2 phr of DCP was introduced into HDPE, the expansion volume ratio of HDPE showed the highest value, which could be more than 7 times. Keywords: Crosslinking; HDPE; Foams; CO 2 INTRODUCTION Polyethylene (PE) foam, which has been applied in many fields, such as packaging materials, chemical industry, construction, and so on [1, 2], possesses the properties of low density, low thermal conductivity, good Corresponding author: zhouhongfu1982@sina.com (Hongfu Zhou), Tel: +86 10 6898 3954 Smithers Information Ltd. 2017 Cellular Polymers, Vol. 36, No. 4, 2017 167

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin electrical insulation performance, and excellent resistance to chemical corrosion. Due to the low molecular weight, narrow molecular weight distribution, and easy crystallization nature of PE, its melt strength is low and the processing window is narrow, which are adverse to its applications in foaming. Therefore, in order to improve the above drawbacks of PE, numerous modifications have been employed, such as blending [3-5], filling [6, 7, 9] and crosslinking [10-13]. A blend of syndiotactic polypropylene (PP) and low-density polyethylene (LDPE) was extruded by Park [3] using isobutane as the blowing agent. It was reported that the foam product was flexible and dimensionally stable, which could withstand a higher temperature than a LDPE foam product. LDPE/EVA/ polyethylene-octene elastomer (POE) foams with different composition ratios were produced by Wang et al. [4] The results suggested that POE was more effective than EVA at improving the melt viscoelasticity of LDPE, which was useful to form closed cells in a continuous foaming process. Fukasawa et al. [7] reported that the shrinkage of LDPE foam using isobutene as a blowing agent was suppressed and the dimensional stability was improved by blending with stearyl stearamide as a permeation modifier. LDPE/silica nanocomposites were foamed by Saiz-Arroyo et al. [8] through two different processes, using a physical blowing agent and a chemical blowing agent, respectively. The results suggested that silica as a nucleating agent could be effective in reducing cell size, increasing cell density and achieving more homogeneous cellular structures. Xing et al. [10] investigated the foaming behaviors of γ-irradiated crosslinked LDPE sheets in an autoclave with CO 2 as the blowing agent. The results showed that γ-irradiation led to a wider foaming temperature range of LDPE. Danaei et al. [12] reported the effect of different additives on increasing the crosslinking efficiency of LDPE. The results confirmed that the addition of dicumyl peroxide (DCP) and trimethylolpropane trimethacrylate (TMPTA) were effective, especially at low doses, to improve the melt strength of LDPE and the foaming behavior of LDPE. Compared with LDPE, high-density polyethylene (HDPE) possesses the lower degree of branching and the stronger crystallization ability, which lead to narrower processing window and lower foamability of HDPE, thus it is meaningful to study the foaming behavior of HDPE [14-18]. The foams of pure HDPE and HDPE/clay nanocomposites were prepared by Khorasani et al. [14] The introduction of clay promoted the formation of more nucleation sites and expanded microcellular foams, and a better dispersed nanocomposite resulted in a more uniform nucleated system in the microcellular foaming process. The foam of HDPE/PP blends and their composites with wood fiber were fabricated by Rachtanapun et al. [16] The study suggested that the blending decreased the crystallinity of HDPE and PP, in addition, facilitated 168 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 the production of microcellular foam. The addition of wood fiber inhibited microcellular foaming. Lee and Park [18] studied the foaming behavior of HDPE using N 2 as a blowing agent and talc as a nucleating agent. The results revealed that the temperature of the die did not affect the cell density significantly and talc was very effective as a nucleating agent in increasing cell density, even for a small amount of N 2. Through the above literatures, it could be found that most studies on the HDPE foam were focused on the blending/filling foam. There are few studies focused on the foaming behavior of crosslinked HDPE using CO 2 as blowing agent in a high pressure stainless steel autoclave. Usually, PE could be crosslinked by high energy radiation, ultraviolet light, peroxide, silane and so on. Peroxide crosslinking (also called chemical crosslinking) is a series of free radical reactions induced by high temperature decomposition of peroxide. Active free radical could be generated when peroxide (ROOR) is heated, and then free radical captures hydrogen of PE molecular chain, generating PE molecular chain with free radical. Because PE molecular chain with free radical possesses high reaction activity, crosslinking reaction takes place and inter-molecular chemical bonds generate, when the molecular chain free radicals meet each other. The mechanism of crosslinking reaction is as follows [19]: where R is C(CH 3 ) 2 C 6 H 5. For the three modifications mentioned above, they all had their own deficiencies. Filling could not change the structure of molecule, and the improvement of melt strength is limited. For blending, the compatibility of blending resins is a problem needed to be solved. While excessive crosslinking would affect the second use of the material. In this paper, we will propose a new HDPE foaming methodology. Different contents of DCP were mixed into HDPE, in order to make HDPE have a microcrosslinking structure, which would be helpful to enhance viscoelasticity and foaming property of HDPE. The effects of DCP content on the molecular Cellular Polymers, Vol. 36, No. 4, 2017 169

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin structure, crystallization behavior, rheological properties, and foaming performance of crosslinked HDPE were also investigated. EXPERIMENTAL Materials HDPE (5000S) used as the matrix polymer was provided by PetroChina Co. Ltd. Dicumyl Peroxide (DCP) used as the crosslinking agent was purchased from Sinopharm Chemical Reagent Co. Ltd. Xylene was supplied by Tianjin Yongda Chemical Reagent Co. Ltd. Fabrication of Various HDPE Samples The HDPE and different content of DCP were mixed in a Haake internal mixer at 190 C, with a mixing time of 15 min and mixing speed of 60 rpm. All components used the unit of parts per hundred resin (phr) for their formulation. The ratios of DCP were 0 phr, 0.05 phr, 0.10 phr, 0.15 phr, 0.20 phr, and 0.25 phr, respectively. After HDPE was fully reacted with DCP, the resultant samples were then transferred to a mold and preheated at 160 for 5 min, then pressed at 10 MPa, and cooled to room temperature to obtain the composite sheets for further measurements and foaming experiment. The samples were denoted as 1#-6#, respectively. Foaming Processing of Various HDPE Samples In order to investigate the differences of foaming behaviors of various HDPE foams, all the samples were prepared and foamed under the same conditions. First, the HDPE sheet samples were put into a high pressure stainless steel autoclave at a temperature of 200 C and a high pressure of 20 MPa with supercritical CO 2 as a blowing agent. The solubilization time was maintained for 2 h to ensure the equilibrium absorption of CO 2 by the samples. Then the samples were cooled to the foaming temperature of 130 C. Finally, the pressure was quenched to atmosphere pressure within 3 seconds. The foam structure could be allowed to grow completely during the rapid depressurization. The foaming samples were prepared for further characterizations. 170 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 Characterizations Measurement of Gel Contents of Various HDPE Samples The gel contents of HDPE samples were determined by the Soxhlet extraction with xylene until the weight of insoluble polymers did not change. Then the insoluble parts were dried at 60 C for 6 h in oven. The test was carried out twice for every sample, and the average value was taken. The gel contents were calculated by the following equation: where W 0 and W g are the weights of the initial polymer and dried insoluble part of samples, respectively. (1) Gel Permeation Chromatography (GPC) The weight-average molecular weight (Mw) of various HDPE samples was measured by high temperature GPC (PL-220, Agilent, USA). Gel permeation chromatographic column adopted 3 series of PL gel Mixed-B (10 μm) columns (Polymer Laboratory Inc.). 10 mg of various HDPE samples was dissolved in 5 ml of 1, 2, 4-Trichlorobenzene at the test temperature of 150 C. The flow rate was 1.0 ml/min. Differential Scanning Calorimetry (DSC) The crystallization and melting behavior of HDPE samples was characterized by a DSC instrument (Q100, TA, USA) purged with nitrogen. The samples were tested as following: HDPE samples were equilibrated at 250 C for 3 min to eliminate the thermal and stress histories. After that, the samples were cooled down to 40 C at a rate of 10 C/min, and then heated to 250 C at the same rate to study the effect of DCP content on the HDPE samples. The crystallinity (χ) of various HDPE foam was computed by the following equation: (2) where DH c and DH 0 c are the melting enthalpy of various HDPE samples and HDPE that crystallized completely, respectively. DH 0 c is considered to be 290.0 J/g [20]. Cellular Polymers, Vol. 36, No. 4, 2017 171

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin Dynamic Rheometer Dynamic rheological behaviors of HDPE samples with different contents of DCP were tested using a rotational rheometer (ARES Rheometer, TA, USA) at 220 C with a parallel plates (20 mm in diameter with a gap of 1.0 mm). The frequency range was from 0.015 to 15.9 Hz, and the maximum strain was fixed at 0.5%, to confirm that these conditions were within the linear viscoelastic region under nitrogen. The complex viscosity (η*), storage modulus (G ) and loss factor (tanδ) were monitored at various frequencies. Scanning Electron Microscope (SEM) The micrographs of the resultant foams were observed using a SEM (FEI Quanta FEG) at an acceleration voltage of 5 kv. The SEM samples were prepared by cryogenically breaking the obtained foam samples and gold sputtering the fractured cross sections. Foaming Property The bulk densities of the prefoamed (r p ) and postfoamed (r f ) samples in g/ cm 3 were measured by a density balance (Sartorius, Goettingen, Germany). The Expansion volume ratio (Φ) was computed by the following equation: The cell densities, N 0, in cells/cm 3, are defined using Eq. 4 by assuming the cells are spherical: (3) where n is the number of cells in the SEM micrograph, M is the magnification factor, A is the area of the micrograph (in cm 2 ), and Φ is the volume expansion ratio of the polymer foam. (4) 172 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 RESULTS AND DISCUSSION Molecular Structure and Molecular Weight of Various HDPE Samples The torque curves, gel fraction, and molecular weight of various HDPE samples were measured to clarify the reaction mechanism as shown in Figures 1 and 2, as well as Table 1. Figure 1. Torque curves of various HDPE samples Figure 2. Gel contents of various HDPE samples Cellular Polymers, Vol. 36, No. 4, 2017 173

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin Figure 1 was the torque curves of various HDPE samples. From Figure 1, it could be found that there was only a plastic peak in the torque curve of pure HDPE (1#). After DCP was introduced into HDPE, the second peak representing chemical reaction appeared in the torque curves of various HDPE samples (2#-6#). With the increase of DCP content, the torque curves of various HDPE samples (2#-6#) were enlarged gradually, which implied that branching and crosslinking reaction of HDPE occurred. As expected, with the DCP content increasing, the molecular weight and crosslinking degree of HDPE gradually increased, as shown in Figure 2 and Table 1. Figure 2 showed the gel contents of various HDPE samples. It could be seen that the gel contents of HDPE samples (1#-3#) were all at zero. From the HDPE sample (4#), the gel generated, and the gel content of HDPE increased with the content of DCP. The change of gel content of various HDPE samples implied that with the DCP contents increasing, the HDPE molecular chain structure was varied from branched, highly branched, to crosslinked. It also confirmed the inference discussed about Figure 1. Table 1. The molecular weight of various HDPE samples Samples 1# 2# 3# 4# 5# 6# Mw (10 5 g/mol) 1.3 1.6 1.7 - - - In order to further verify the conclusions above, GPC was carried out to measure the molecular weight of various HDPE samples, shown in Table 1. It can be found that with the increasing content of DCP, the molecular weight of HDPE (1#-3#) increased, which maybe due to the branching reaction of HDPE. Owning to the formation of crosslinked structure of HDPE samples (4#-6#), the GPC results of the crosslinked samples were not measured and the corresponding results were insignificance. Crystallization and Melting Behavior of Various HDPE Samples Solubilization temperature of foaming is associated with the melting temperature of polymer. The crystallization and melting properties of various HDPE samples were evaluated by DSC. The DSC curves including cooling and heating process are shown in Figures 3 and 4, and the corresponding results are summarized in Table 2, respectively. As shown in Figure 3, with the increasing content of DCP, the crystallization temperature of HDPE samples (1#-4#) increased slightly from 114.9 C to 117.5 C, due to the nucleation effect of branching point, and then the crystallization temperatures of HDPE samples (4#-6#) were unchanged, which may be due to the formation of 174 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 Table 2. Thermal properties of various HDPE samples Samples 1# 2# 3# 4# 5# 6# T c / C 114.9 115.8 116.7 117.5 117.4 117.7 H c /J g -1 199.7 195.4 184.7 184.0 183.4 183.9 T m / C 133.7 133.8 133.3 132.1 132.1 131.0 χ c /% 68.9 67.4 63.7 63.4 63.2 63.4 Figure 3. DSC curves of various HDPE samples at cooling rate of 10 C/min Figure 4. DSC curves of various HDPE samples at heating rate of 10 C/min Cellular Polymers, Vol. 36, No. 4, 2017 175

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin the crosslinking structure of HDPE. The branching point of HDPE acting as the heterogeneous nucleating sites was beneficial to the nucleation of the crystallization and the increase of the crystallization temperatures, but the crosslinking structure limited the motion of the HDPE molecular chain and the crystallization of HDPE [21]. The crystallinity of various HDPE samples decreased with the increasing content of DCP. The reason may be that the branching chains and the crosslinking structures of HDPE affected the structural regularity of HDPE chains. Rheological Property of Various HDPE Samples Sufficient melt strength is very important for the foaming process of polymers, which could be characterized indirectly by the viscosity, dynamic storage modulus, and loss angle of polymers. For a given polymer, high molecular weight and complex polymer architecture could lead to high melt strength. HDPE has a low melt strength resulted from its linear structure. In order to enhance the melt strength and improve the foaming behavior of HDPE, DCP was introduced into HDPE to fabricate the branching and/or crosslinking structure. Figure 5. Relationships between complex viscosity (η*) and frequency for various HDPE samples 176 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 Figure 6. Relationships between storage modulus (G ) and frequency for various HDPE samples Figure 7. Relationships between loss angle (tanδ) and frequency for various HDPE samples Cellular Polymers, Vol. 36, No. 4, 2017 177

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin Rheological measurements were employed to evaluate the viscoelastic properties of various HDPE samples, as shown in Figures 5-7. Figure 5 shows the relationships between complex viscosity (η*) and frequency for various HDPE samples. With the increase of frequency, the complex viscosity of all samples decreased. There may be two reasons for this, one is the effect of shear-thinning, and the other is that higher frequency make the polymer molecular chain orientated. Moreover, the complex viscosity of HDPE increased with the contents of DCP. The enhancement of the complex viscosity may be attributed to the presence of the branching structure and/or the crosslinking structure, which made HDPE interchain slippage difficult. Similar phenomenon was observed in other study [22]. Figures 6 and 7 show the relationships between the storage modulus (G ) and frequency, as well as the relationships between the loss angle (tanδ) and frequency, respectively. G of various HDPE samples increased with the frequency and the DCP content, indicating the melt strength of HDPE was improved. In Figure 7, tanδ of HDPE samples decreased with the DCP content, however, the HDPE samples (4#-6#) were hardly changed with the frequency, indicating the formation of the gel structure [23]. With the increasing DCP content, G of various HDPE samples increased, while tanδ decreased, which implied that the melt elasticity of HDPE was enhanced [24]. This phenomenon suggested that HDPE with DCP had a longer relaxing process, which also testified that the branching structure improved melt elasticity of HDPE. Cellular Morphology of Various HDPE Foams Generally, the performance of polymer foams could be evaluated by many parameters, such as cell size, cell density, expansion volume ratio, and so on. Figure 8 shows the microstructure of various HDPE foam samples, and their detailed parameters are listed in Table 3. Because of its linear structure, the viscoelastic properties of HDPE melt is poor, which leaded the cell size of HDPE (1# in Figure 8) was only 56 μm, and the cells had some opencell structure. It suggested that the HDPE melt could not provide enough extensional viscosity for the expansion of cell walls, and the growing cells were not effectively stabilized. Therefore, the cell walls became thin, ruptured, and merged, leading to the formation of open cell. With the increase of DCP content, the cell size and expansion volume ratio of HDPE foam increased firstly, and then decreased. That may be because the branching structure could improve the melt strength of HDPE, which was beneficial to the growth of cell, but the crosslinking structure would restrict cell growth. From Figure 8, the cell shape (pyritohedron) of HDPE foam (2#-5#) was better than that (nearly oval) of HDPE foam samples (1#, 6#), which should be attributed to the effects of DCP. 178 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 Table 3. Cell morphology data of various HDPE samples Samples 1# 2# 3# 4# 5# 6# Cell size (μm) 56 64 70 110 74 57 Cell Density 3.47±0.24 2.67±0.25 3.23±0.19 1.70±0.24 2.88±0.18 5.32±0.32 (10 7 cell/cm 3 ) Density 0.30±0.01 0.26±0.02 0.19±0.01 0.15±0.01 0.12±0.01 0.14±0.01 (g/cm 3 ) Expansion volume ratio 3.1±0.1 3.5±0.2 4.9±0.4 6.0±0.6 7.7±0.7 6.5±0.5 Figure 8. SEM micrographs for foams of various HDPE samples (400 ) Cellular Polymers, Vol. 36, No. 4, 2017 179

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin The expansion volume ratio of HDPE samples (1#-5#) increased with the content of DCP, while the expansion volume ratio of HDPE samples (6#) decreased. That maybe because the introduction of DCP leads to the formation of branching and micro crosslinking structures, and improves the melt strength of HDPE samples (1#-5#), which was beneficial to the foaming behavior of HDPE. But HDPE sample (6#) had excessive crosslinked structure, which would restrict the growth of bubbles. The expansion volume ratio of HDPE sample (5#) showed the highest value of all the samples. From the SEM micrographs of the HDPE samples in Figure 8, the sample of 5# presented a better pyritohedron, which was the best sample of all. CONCLUSIONS It has been demonstrated that the introduction of DCP into HDPE could improve the viscoelasticity and foaming property of HDPE. With the content of DCP increasing, the molecular weight of HDPE increased from 1.3 10 5 to 1.7 10 5 g/mol, and the crystallization temperature of HDPE was increased from 114.9 C to 117.7 C. Due to the branching and crosslinking structure of HDPE, the complex viscosity and storage modulus of HDPE both increased rapidly at the low frequency. The SEM micrographs of the fracture surface of HDPE foam showed that a suitable amount of DCP was useful for the foaming behavior of HDPE (2#-5#), the cell structure turned oval to pyritohedron with the content of DCP, but excessive DCP would limit the foaming behavior of HDPE (6#). The highest expansion volume ratio of HDPE sample was more than 7 times. ACKNOWLEDGEMENT This work was supported by the Natural Science Foundation of Beijing (2164058 and 2162012), the National Key Research and Development Program of China (2016YFB0302203), Innovative research team of polymeric functional film of Beijing technology and business university (19008001071), and the Development Project of Beijing Municipal Education Commission (KM20131001002). REFERENCES 1. C. Xin, Y. He, Q. Li, Y. Huang, B. Yan, X. Wang. Journal of Applied Polymer Science, 119(3) (2011) 1275-1286. 2. P.S. Chum, K.W. Swogger. Progress in Polymer Science, 33(8) (2008) 797-819. 180 Cellular Polymers, Vol. 36, No. 4, 2017

Preparation of Crosslinked High-density Polyethylene Foam Using Supercritical CO 2 3. C.P. Park. Journal of Cellular Plastics, 38(2) (2002) 129-138. 4. W. Wang, W. Gong, B. Zheng. Journal of Vinyl and Additive Technology, 22(1) (2016) 61-71. 5. B.I. Chaudhary, R.P. Barry. Journal of Cellular Plastics, 35 (1999) 531-549. 6. M.A. Rodriguez-perez, J. Lobos, C.A. Perez-muǹoz, J.A.D. Saja. Journal of Cellular Plastics, 45(5) (2009) 389-403. 7. Y. Fukasawa, T. Okuda, T. Shimada, M. Hattori, H. Saito. Journal of Cellular Plastics, 44(2) (2008) 107-123. 8. C. Saiz-Arroyo, M. Rodríguez-Pérez, J.I. Velasco, J.A.D. Saja. Composites, 48 (2013) 40-50. 9. J.I. Velasco, M. Antunes, O. Ayyad, J.M. López-Cuesta, P. Gaudon, C. Saiz- Arroyo, M.A. Rodríguez-Pérez, J.A. De Saja. Polymer, 48(7) (2007) 2098-2108. 10. Z. Xing, G. Wu, S. Huang, S. Chen, H. Zeng. The Journal of Supercritical Fluids, 47(2) (2008) 281-289. 11. B. Wang, M. Wang, Z. Xing, H. Zeng, G. Wu. Journal of Applied Polymer Science, 127(2) (2013) 912-918. 12. M. Danaei, N. Sheikh, F.A. Taromi. Journal of Cellular Plastics, 41(6) (2005) 551-562. 13. J.C.A. Martínez-díez, M.A. Rodríguez-Pérez, J.A.D. Saja, L.O.A.Y. Rábago, O.A. Almanza. Journal of Cellular Plastics, 37(1) (2001) 21-42. 14. M.M. Khorasani, S.R. Ghaffarian, A. Babaie, N. Mohammadi. Journal of Cellular Plastics, 46(2) (2010) 173-190. 15. H. Zhang, G.M. Rizvi, C.B. Park. Advances in Polymer Technology, 23(4) (2004) 263-276. 16. P. Rachtanapun, S.E.M. Selke, L.M. Matuana. Journal of Applied Polymer Science, 88(12) (2003) 2842-2850. 17. Y. Zhang, D. Rodrigue, A. Aït-kadi. Journal of Cellular Plastics, 39(6) (2003) 451-474. 18. J.W.S. Lee, C.B. Park. Macromolecular Materials and Engineering, 291(10) (2006) 1233-1244. 19. R.P.D. Melo, V.D.O. Aguiar, M.D.F.V. Marques. Materials Research, 18(2) (2015) 313-319. 20. A. Choudhury. Materials Science and Engineering A, 491 (2008) 492-500. 21. H. Zhou, X. Wang, Z. Du, H. Li, K. Yu. Polymer Engineering and Science, 55(5) (2015) 988-994. Cellular Polymers, Vol. 36, No. 4, 2017 181

Hongfu Zhou, Zhanjia Wang, Guozhi Xu, Xiangdong Wang, Bianying Wen, and Shanglin Jin 22. H. Liu, X. Wang, W. Liu, B. Liu, H. Zhou, W. Wang. Cellular Polymers, 33(4) (2014) 189-212. 23. J. Xu, J. Zhang, Y. Liu, H. Sun, J. Wang. Carbohydrate Polymers, 139 (2016) 43-49. 24. H. Liu, X. Wang, H. Zhou, W. Liu, B. Liu. Cellular Polymers, 34(2) (2015) 63-94. 182 Cellular Polymers, Vol. 36, No. 4, 2017