The effect of modified AlN on the thermal conductivity, mechanical and thermal properties of AlN/polystyrene composites

Abstract

Modified aluminum nitride particle/polystyrene (AlN/PS) composite was prepared by a powder processing technique. The thermal conductivities, dielectric and thermal properties of composites with different mass fraction of modified AlN were investigated. Compared to the pure PS, the thermally conductive properties of AlN/PS composites improved from 0.189 W (m⁻¹ K⁻¹) to 0.418 W (m⁻¹ K⁻¹) when the content of AlN was 25 wt%. The thermal stability of the AlN/PS composite was improved with the increasing addition of AlN. The dielectric constant and dielectric loss of the 25 wt% AlN/PS composite was 3.58 and 0.0036 at 10⁶ Hz, respectively. The dielectric constant of the AlN/PS composite showed a very small variation with the range of the frequency from 10² Hz to 10⁶ Hz. The tensile strengths of the AlN/PS composite increased with increasing filler content when it was no more than 5 wt%, and then decreased with the further increase of the filler content, whereas the elongations at break showed the similar trends. SEM analysis showed that AlN could pack tightly, and thermally conductive AlN–AlN channels could be generated with further addition of AlN. The thermal conductivity of the AlN/PS composite with the increase of AlN content tended to be higher at higher temperatures. The thermal conductivity of the AlN/PS composites at the same AlN content tended to be higher at higher temperatures. Several theoretical models were used in comparison with experimental data of the thermal conduction of the composites.

---

Introduction

Polystyrene (PS) has been widely applied in the fields of electronics industry, food packing, kitchen appliance, and so on, owing to its excellent dielectric properties, good dimensional stability, outstanding chemical resistance, easy fabrication and low cost.^1–3 Unfortunately, the intrinsically low thermally conductive properties of pure PS matrices (0.189 W (m⁻¹ K⁻¹)) have restricted their broader application in electronic packaging and encapsulation, satellite devices and in areas which require good heat dissipation, low thermal expansion and light weight.^4–7

Under this circumstance, the thermal conductivity property of pure PS matrix should be enhanced to transfer the heat to the outside of the PS matrix, which could effectively prolong their working life. Some relative reports showed that the addition of thermally conductive fillers (such as zinc oxide,^8,9 boron nitride,¹⁰ titanium,^11,12 silicon oxide,^13,14 and graphene^15,16) into PS matrix was emerging as the most economical and effective way to improve the thermal conductivity property of PS composite.^17–19 However, the thermal conductivity property of the PS composite which was modified by addition filler was still far below the expectation. The reason could be ascribed to the bad dispersion of thermal conductive filler and a strong interfacial thermal barrier of PS matrix/thermal conductive filler. Therefore, thermal conductive filler was treated by the surface functionalized to improve their interfacial compatibility to PS matrix, which could further enhance the good dispersion of thermal conductive filler and decrease the interfacial thermal barrier of PS matrix/thermal conductive filler.^20,21

Dielectric nanocomposites, also called nanodielectric, had attracted much attention as new materials because of their superior electrical, magnetic, and thermal properties in many fields.^22–25 For microelectronic packaging application, dielectric materials with a suitable dielectric constant, low dielectric loss, high thermal conductivity and good thermal stability are required. At present, such materials are mainly ceramics, which have drawbacks, such as high brittleness, low dielectric strength, high processing temperature and high density.^26–29 However, dielectric constant was improved, and at the same time, dielectric loss factor also often was increased because of more interface polarization between nanofillers and polymer. Therefore, a new type of nanofillers needs to be fabricated, which can improve dielectric constant and keep dielectric loss factor as low as possible.^30–33

Generally speaking, low value of dielectric constant and dissipation factor ensures high signal propagation speed and diminishes the effect of capacitive coupling. Therefore, a low dielectric constant and loss dissipation is required for polymer packaging materials since a high dielectric constant will cause a strong negative effect on the signal propagation by increasing the delay time. The literature concerning dielectric and thermal conductive polymer composites is particularly focused on the use of different kinds of fillers, such as aluminum nitride (AlN),^34,35 silicone nitride (Si₃N₄),^36,37 hexagonal boron nitride (hBN),^38,39 silicon carbide (SiC),⁴⁰ alumina (Al₂O₃),⁴¹ and SiO₂.^42–44 For example, Pan^34,35 et al. studied the structures and properties of AlN/PTFE composites. He³⁷ et al. investigated the preparation and properties of Si₃N₄/PS composites for electronic packaging. Kim^42–44 et al. studied the structures and properties of polyimide–silica composites. In order to develop suitable composite for packaging applications, the prime requirement was that the filler should have low dielectric relative permittivity, low dielectric loss, high thermal conductivity and good thermal stability.⁴⁵ Aluminum nitride (AlN) was a kind of material for dielectric application owing to high thermal conductive (319 W (m⁻¹ K⁻¹)), low thermal expansion coefficient, high electrical resistivity and high thermal resistance. Recently, many researchers^46,47 investigated the effect of AlN on the properties of various polymer composite. Wu⁴⁸ et al. reported that the introduction of silane grafted AlN effectively enhanced the thermal stability, thermal conductivity and mechanical properties of the polyetherimide composite. Zhou⁴⁹ et al. used AlN in PMMA matrix in order to prepare new composite with more excellent thermal conductivity and relative permittivity. The results showed that the thermal conductivity of PMMA composite was improved to 1.87 W (m⁻¹ K⁻¹).

The aim of the present study is to prepare a high thermal conductive polymer composite with low dielectric loss. Polystyrene was selected as the matrix because of its good thermal stability, excellent dielectric property, outstanding chemical resistance, easy fabrication. To enhance the filler–matrix interface bonding strength, a silane coupling agent, γ-glycidoxypropyl-trimethoxysilane, was employed to improve the interfacial adherence between the AlN and PS matrix. Modified treatment of aluminum nitride by coupling agent can greatly improve the interfacial properties between the polymer and filler, which is beneficial to improve the comprehensive properties of the composite materials. So, the modified AlN/PS composite with various filler content were prepared. The mechanical, thermal and dielectric properties of the modified AlN/PS composite were investigated. The degree of dispersion of the AlN was examined by observing the morphology of composite bulk samples. Compared to the conventional mixing and blending, a mixing method by the ball milling machine was adopted for preparing the composite in order to good dispersion and further improve the interfacial compatibility between AlN and PS matrix. In previous studies, the researchers mainly study the thermal conduction mechanism and model by the blending between polymer and nanofiller. However, in this work, the experimental methods are described comprehensively and several theoretical models were used to compare with experimental data of the thermally conductive of the composite. And the size of AlN is 2 μm compared to others size 10 μm, the size of AlN can an effect on the performance of polymer composite. Up until now, the investigations of the dielectric properties of the modified AlN/PS composite in a wide frequency range and the thermal conductivity properties at different temperature of the modified AlN/PS composite have seldom been reported until now. Especially, the thermal conductivity property of modified AlN/PS composite was proved by the Agari's semi-empirical model fitting. So, our purposes expect to give a deeper insight into the influence of the content, particle size on the thermal conductivity and dielectric properties of AlN/PS composite. And the preparation of modified AlN/PS composite can be applied in the reliablity of electric components, electronic packaging, encapsulations, satellite devices and in areas which required good heat dissipation, low thermal expansion and light weight.

Experimental

Materials

Polystyrene is received from Haimai Courier Chemical Technology Co. Ltd. (Shanghai, China); aluminum nitride particle (AlN) powder with average size in 2 μm was provided by Aladdin Co. Ltd. (Shanghai, China); (γ-aminopropyl-triethoxysilane) KH-550 is supplied by Jingzhou Jiangshan fine chemical Ltd (Jingzhou, China). Alcohol is supplied by Tianjin Fuyu Fine Chemical Co. Ltd. (Tianjin, China).

Surface functionalization of AlN

Surface functionalization of AlN could ensure good dispersion of AlN in PS matrix, which could improve the interfacial compatibility between AlN and PS matrix. AlN powder was dispersed in alcohol and stirred magnetically well. To the stirred suspension, calculated KH-550 and deionized water were added. The amount of KH-550 was 2 wt% of the weight of AlN powder and the amount of water was controlled exactly for the hydrolysis of KH-550. The stirring was continued for 6 h, then the slurry thus was vacuum dried at 80 °C and silane modified AlN powder was obtained.

Preparation of the AlN/PS composites

The AlN/PS composite were fabricated according to the following procedures: (i) mixing AlN and PS matrix using a ball milling machine for 24 h at room temperature, to embed the AlN at the interface of the PS matrix; (ii) hot-pressing (190 °C at 10 MPa) to fabricate the AlN/PS composites.

Characterization

The thermal conductivity property was measured on disk samples by using a LFA447 light flash system (NETZSCH, Selb, Germany) from 25 °C to 150 °C. The mechanical properties were investigated following ASTM D 1708-02a, and the test speed was 100 mm min⁻¹. The measurement results of tensile strengths and elongations at break were mean values for five separate experiments. The morphology of fractured cross sections of composite was examined by scanning electron microscopy (SEM, VEGA3 LMH). DSC was performed with a TA instrument Q1000 system at a heating rate of 10 °C min⁻¹ under a nitrogen atmosphere from room temperature to 380 °C. TGA analysis was performed with a TA instrument Q600ADT system at a heating rate of 10 °C min⁻¹ under a nitrogen atmosphere from room temperature to 700 °C. The dielectric properties of composite was tested by using a broadband dielectric spectrometer (CONCEPE 80, Novocontrol Technology Company, Germany) with an Alpha-A high-performance frequency analyzer from 10² Hz to 10⁶ Hz. To ensure good electrical contact, samples were evaporated with thin gold layers on both surfaces to serve as electrodes.

Results and discussions

Thermal conductivity property of AlN/PS composites

Fig. 1 shows the variation of thermal conductivity property of modified AlN/PS composite as a function of temperature. The effective thermal conductivity property for the composite of PS was found to be 0.189 W (m⁻¹ K⁻¹) at 25 °C and increased with temperature over the temperature range investigated. The thermal conductivity property of AlN/PS composite exhibited the temperature dependences similar to the pure PS. Meanwhile, the thermal conductivity of AlN/PS composite increased with the increasing of AlN content for all samples. The 25 wt% AlN/PS composite had the maximum thermal conductivity (0.418 W (m⁻¹ K⁻¹) at 25 °C), which increased by 121% compared to that of pure PS. The thermal conductivity of the AlN/PS composites with the increasing of AlN content tended to be a higher value with the temperature increasing. And the thermal conductivity of the AlN/PS composites at the same of AlN content tended to be higher value with the temperature increasing. Various mechanisms had been presented to explain the increase of the effective thermal conductivity property of composite containing different nanoparticles, such as interface interaction at the particle/resin interface,⁵⁰ the nature of heat transport in the nanoparticles and the effects of nanoparticle clustering.⁵¹ Based on the experimental results, it was concluded that the three dimensional network formed by AlN/PS composite dominated the thermal conduction mechanism, which led to the composite with excellent thermal conductivity property. Meanwhile, the thermal conduction chains and the prevailing means to conduct thermal diffusion in the large-size filler/resin were the secondary means to conduct thermal diffusion in the composite.

Fig. 1 The thermal conductivity of AlN/PS composites.

Agari's semi-empirical model fitting of the AlN/PS composites

Agari's semi-empirical model can yield better results than the theoretical ones. The logarithmic equation of Agari was written as follows eqn (1):

lg λ_c = V_f × C_f × lg λ_f + (1 − V_f) × lg(C_pλ_p) (1)

where C_p represents the effect of the AlN on the PS structure, i.e. C_p is related to the change of thermal conductivity of the PS matrix, as a consequence of a change of its crystallinity; C_f represents the ability of AlN to form continuous thermally conductive chains and networks, 0 < C_f < 1.

Fig. 2 showed the logarithmic values of the thermal conductivities as a function of the mass fraction of AlN. The parameters of C_p and C_f were calculated to be 1.1867 and 0.723, respectively. The high value of C_p suggested that the AlN could influence the crystallinity of the PS matrix. The low value of C_f suggested that the AlN had a strong ability to form continuous thermal conductive networks, and the formation of thermally conductive networks was much easier with the incorporation of AlN.

Fig. 2 Logarithmic thermal conductivity property of the AlN/PS composite as a function of the AlN mass fraction.

Dielectric properties of the AlN/PS composite

According to the Fig. 3, it could be seen that the dielectric constant of composite with filler content up to 25 wt% remained nearly a constant in the frequency range from 10² Hz to 10⁶ Hz. The results showed the AlN/PS composite with excellent frequency stability. As the filler content increased from 0 to 25 wt%, there was a gradual increase in the dielectric constant, which was due to the higher dielectric constant of AlN. And the fact was that dipole–dipole interaction within the powders was improved. Compared to pure PS system, the lower dielectric constant of the AlN/PS composite could be attributed to the connectivity and the particle size effect. Moreover, the non-polar nature of PS matrix and the polymer molecular chains hindered the contribution of electrical polarization in the composite.⁵² But the dielectric constant of AlN/PS composite decreased when the content of AlN exceeded 25 wt%. The decreasing of dielectric constant of AlN/PS composite could be ascribed to more air voids which could be involved when more filler particles were incorporated.

Fig. 3 The dielectric constant of AlN/PS composite.

It was evident from Fig. 4 that the dielectric loss of the composite increased slowly with increasing the filler content from 0 to 25 wt%, and the dielectric loss of all the composites was below 0.10 at the frequency from 10² Hz to 10⁶ Hz. It was also noted that there was a gradual rise in the loss tangent for each composite, especially the composite filled by 25 wt% AlN with the frequency decreasing. The reason could be explained based on the interfacial polarization (namely, Maxwell–Wagner effect), which would appear when the following inequality existed eqn (2):⁵³

ε_PSσ_AlN ≠ ε_AlNσ_PS (2)

where ε_PS and ε_AlN are the dielectric constant of PS and AlN particles, while σ_AlN and σ_PS are the conductivities of PS and AlN particles, respectively. Because of the difference in the dielectric constant and the conductivity of the matrix and the filler, there was plentiful of accumulation of interfacial charges inside the composite under the applied electric field. As the filler content increased, the interfacial area inside the composite increased gradually and there was a gradual enhancement in the interfacial polarization. Moreover, as the frequency was raised, the interfacial dipoles had less time to orient themselves in the direction of the alternating field.⁵⁴ As a result, as the filler content increased, the dielectric properties of the composite were more and more dependent on the frequency.

Fig. 4 The dielectric loss factor of AlN/PS composite.

Thermal properties of the AlN/PS composites

Fig. 5 showed the influence of the mass fraction of AlN on the melting heat and melting temperature of the AlN/PS nanocomposite determined using DSC analysis. The corresponding thermal data were listed in Table 1.

Fig. 5 DSC curves of the pure PS matrix and the AlN/PS composite.

Table 1 Thermal data of the pure PS and the AlN/PS composite from DSC analysis

Samples	Melting heat/J g^−1	Melting temperature/°C
Pure PS	78.7	101.5
5 wt% AlN/PS	81.2	102.1
10 wt% AlN/PS	82.5	101.9
15 wt% AlN/PS	82.3	102.6
20 wt% AlN/PS	83.1	102.5
25 wt% AlN/PS	83.2	102.7

---

Both the melting heat and melting temperature of the AlN/PS nanocomposite were increased slightly with the increasing addition of AlN. The corresponding melting heat was increased from 78.7 J g⁻¹ (pure PS) to 81.2 J g⁻¹ (5 wt% AlN), 82.5 J g⁻¹ (10 wt% AlN), 82.3 J g⁻¹ (15 wt% AlN), 83.1 J g⁻¹ (20 wt% AlN) and 83.2 J g⁻¹ (25 wt% AlN). And the corresponding melting temperature was also increased from 101.5 °C (pure PS) to 102.1 °C (5 wt% AlN), 101.9 °C (10 wt% AlN), 102.6 °C (15 wt% AlN), 102.5 °C (20 wt% AlN) and 102.7 °C (25 wt% AlN). The results could be attributed to the heterogeneous nucleation of AlN which was able to hinder the homogeneous nucleation of the PS matrix. Meanwhile, the addition of AlN in the composite could also decrease the thickness of crystal plates of the PS system. A combination of action above would effectively decrease the heat enthalpy of the AlN/PS nanocomposite and reduce the relative degree of crystallinity.

TGA curves of pure PS and the AlN/PS composite were presented in Fig. 6. And the corresponding characteristic thermal data of pure PS and the AlN/PS composite were listed in Table 2. From Fig. 6 and Table 2, the corresponding weight loss temperatures of the AlN/PS composite with the increasing addition of AlN were increased at the same weight loss stage. Meantime, the corresponding heat resisting index of the AlN/PS composite was also improved. It revealed that the thermal stability property of the AlN/PS composite with the increasing addition of AlN was improved. The enhancement of thermal stability property of the AlN/PS composite could be ascribed to two reasons:⁵⁵ (i) AlN had higher thermal conductivity and higher heat capacity than PS system, which could cause it to preferably absorb the heat. So this could result in PS chains starting to degrade at a higher temperature. (ii) The AlN filler in the composite could act as barriers, retarding the formation and volatile byproducts during pyrolysis. Meanwhile, the thermal motion of PS segments near the surface of AlN could be restricted because of the physical interlock.

Fig. 6 TGA curves of the pure PS matrix and the AlN/PS composite.

Table 2 Thermal data of pure PS matrix and AlN/PS composite from TGA analysis

Samples	Temperature/°C		Heat resistance index^a/°C
	T5	T30
a Heat resistance index = 0.49[T5 + 0.6(T30 − T5)], T5, T30 is the decomposing temperature at 5%, 30% weight loss, respectively.^56
Pure PS	306	408	180
5 wt% AlN/PS	385	411	196
10 wt% AlN/PS	386	412	197
15 wt% AlN/PS	387	414	198
20 wt% AlN/PS	381	409	195
25 wt% AlN/PS	386	415	198

---

Mechanical properties of AlN/PS composite

The mechanical properties of AlN/PS particle composite were measured, and the results were shown in Fig. 7. The tensile strength of AlN/PS composite increased obviously with the increasing of AlN content when it was no more than 5 wt%, and then decreased with the further increasing of the AlN content. First, compared to pure PS system, the tensile strength of 5 wt% AlN/PS composite was up to 12%, indicating a better dispersion of particles and stronger interfacial interaction between the nanoparticles and the matrix.⁵⁷ Then the decrease of the tensile strength of composite was due to not only the decrease in the deformation area of the matrix upon increasing the fraction of the filler but also the lack of adhesion between PS and AlN. In addition, the AlN particles with low aspect ratio were generally not support stresses transferred from the polymer and thus weaken the tensile strength of composite. However, the elongations at break of the composite showed a similar tendency, and the elongations at break of pure PS system was less than 15.1%. Moreover, traditionally the dispersion of particles in polymeric material had been proven difficult and frequently led to the phase separation and agglomeration.⁵⁸ In some composite, there was a critical weight fraction at which the aggregation occurred and the tensile strength went down. With respect to the PS-based composite, the change in tensile strength at 5 wt% AlN/PS composite was believed to be related mainly to the change in the dispersion of nanoparticles, namely the change from the uniform dispersion to clustering resulting in the occurrence of excessive agglomeration and bad agglomeration. It was noteworthy that the tensile strength of the PS-based composite filled with inorganic particle was larger than that of the PS-based composite filled with micro-scale particles.⁵⁷

Fig. 7 Tensile strength and elongation at break of pure PS and AlN/PS composite.

Microstructure of AlN particles and AlN/PS composite

Fig. 8 showed the SEM morphologies of impact fractures for the AlN/PS composite. It could be seen that a small amount of AlN was dispersed uniformly inner the PS matrix (Fig. 8b and c), which revealed that the introduction of powder processing technology. The powder processing technology could effectively improve the uniform dispersion of AlN. It was noted that the fracture surfaces of 0, 5 wt% and 10 wt% AlN/PS composite showed uniform dispersion and distribution of particles throughout PS matrix, which suggested that the appropriate amount of AlN had well compatibility with PS matrix. However, when the content of AlN was above 5 wt%, as shown in Fig. 8d, large clusters in the matrix appeared due to the aggregation of AlN at high concentration, which could lead to more rapid crack initiation and impact failure. And with the increasing addition of AlN, there was some connectivity of AlN–AlN inner PS matrix in small regions (Fig. 8d). With the further increasing addition of AlN, AlN could pack tightly to contact each other, and the corresponding thermal conductive channels of AlN–AlN could be effectively generated (Fig. 8e and f).

Fig. 8 SEM images of the pure PS matrix and the AlN/PS composites (a) 0 wt% AlN, (b) 5 wt% AlN, (c) 10 wt% AlN, (d) 15 wt% AlN, (e) 20 wt% AlN, (f) 25 wt% AlN.

Conclusion

Thermal conductive property of the AlN/PS composite increased with the increasing addition of AlN. And the addition of 25 wt% AlN/PS composite improved the thermal conductive property. Compared to the pure PS (0.189 W (m⁻¹ K⁻¹)), the value of the thermal conductive property of 25 wt% AlN/PS composite was up to 0.418 W (m⁻¹ K⁻¹). SEM analysis showed that the AlN could pack tightly to contact each other, and the corresponding thermal conductive channels of AlN–AlN could be generated with the further addition of AlN. Agari model fitting of the AlN/PS composite revealed that the AlN could disperse more uniformly and the thermal conductive channels of AlN–AlN were more easily formed inner the PS matrix. The thermal stabilities of the AlN/PS composite were also increased with the increasing addition of AlN. As the content of filler increased, the tensile strength of composite decreased. The obtained composite with 25 wt% AlN showed a low dielectric constant of 3.58 and a low dielectric loss tangent of 0.0036 at 1 MHz, which was still remain at relatively low levels. In addition, the dielectric constant of the composites showed weak frequency dependence from 10² Hz to 10⁶ Hz.

Acknowledgements

This work was financially supported by the National Science Foundation of China (No. 51407134), China Postdoctoral Science Special Foundation (No. 2015T81028), China Postdoctoral Science Foundation (No. 2016M590619, No. 2016M592138), Natural Science Basic Research Plan in Shaanxi Province of China (No. 2015JM5215), Special Scientific Research Program of Shaanxi Provincial Department of Education (No. 16JK1043), Key Project of Baoji University of Arts and Sciences (No. ZK16072), Baoji Engineering Technology Research Center for Ultrafast Optics and New Materials (No. 2015CXNL-1-3). A Feng is supported by scholarship from the China Scholarship of Council (No. 201508615058). The authors also thank their colleagues in their laboratory for their support.

References

1. J. W. Gu, Z. Y. Lv, Y. L. Wu, R. X. Zhao, L. D. Tian and Q. Y. Zhang, Composites, Part A, 2015, 79, 8–13.
2. Z. Wang, S. Li and Z. Wu, Compos. Sci. Technol., 2015, 112, 50–57.
3. G. Hatui, P. Bhattacharya, S. Sahoo, S. Dhibar and C. K. Das, Composites, Part A, 2014, 56, 181–191.
4. J. W. Gu, X. T. Yang, C. M. Li and K. C. Kou, Ind. Eng. Chem. Res., 2016, 55, 10941–10946.
5. W. Zhou, C. F. Wang, T. Ai, K. Wu, F. J. Zhao and H. Z. Gu, Composites, Part A, 2009, 40, 830–836.
6. N. Song, J. Yang, P. Ding, S. Tang and L. Shi, Composites, Part A, 2015, 73, 232–241.
7. W. Zhou, Q. Chen, X. Sui, L. Dong and Z. Wang, Composites, Part A, 2015, 71, 184–191.
8. G. Wu, Y. Cheng, Q. Xie, Z. Jia and F. Xiang, Mater. Lett., 2015, 144, 157–160.
9. E. Afsharmanesh, H. K. Maleh, A. Pahlavan and J. Vahedi, J. Mol. Liq., 2013, 181, 8–13.
10. X. Y. Huang, S. Wang, M. Zhu, K. Yang, P. K. Jiang, Y. Bando, D. Golberg and C. Y. Zhi, Nanotechnology, 2014, 26, 015705.
11. Z. Nenova, T. Nenov, S. Kozhukharov and M. Machkova, Mater. Sci., 2015, 21(2), 191–197.
12. R. Fleming, T. Pawlowski, A. Ammala, P. Casey and K. Lawrence, IEEE Trans. Dielectr. Electr. Insul., 2005, 12, 745–753.
13. Y. Q. Li, Y. Yang, S. Y. Fu, X. Y. Yi, L. C. Wang and H. D. Chen, J. Phys. Chem. C, 2008, 112, 18616–18622.
14. G. Pezzotti, I. Kamada and S. Miki, J. Eur. Ceram. Soc., 2000, 20, 1197–1203.
15. C. Zhi, Y. Bando, C. Tang, S. Honda, H. Kuwahara and D. Golberg, J. Mater. Res., 2006, 21, 2794–2800.
16. W. Park, J. Hu, L. A. Jauregui, X. Ruan and Y. P. Chen, Appl. Phys. Lett., 2014, 104, 113101–113104.
17. L. A. Fredin, Z. Li, M. A. Ratner, M. T. Lanagan and T. J. Marks, Adv. Mater., 2012, 24, 5946–5953.
18. F. M. Du, R. C. Scogna, W. Zhou, S. Brand, J. E. Fischer and K. I. Winey, Macromolecules, 2004, 37, 9048–9055.
19. S. Rimdusit and H. Ishida, Polymer, 2000, 41, 7941–7949.
20. G. L. Wu, K. C. Kou, L. H. Zhuo, Y. Q. Wang and J. Q. Zhang, Thermochim. Acta, 2013, 559, 86–91.
21. Y. Wang, K. Kou and G. Wu, Polymer, 2015, 77, 354–360.
22. J. Gu, C. Liang, J. Dang, W. Dong and Q. Zhang, RSC Adv., 2016, 6(42), 35809–35814.
23. H. Gao, D. Yorifuji, Z. H. Jiang and S. Ando, Polymer, 2014, 55, 2848–2855.
24. Y. Wang, K. Kou, G. Wu, A. Feng and L. Zhuo, RSC. Adv., 2015, 5, 58821–58831.
25. G. Wu, H. Wu and K. Wang, RSC Adv., 2016, 6, 58069–58076.
26. W. C. Wang, R. H. Vora, E. T. Kang, K. G. Neoh, C. K. Ong and L. F. Chen, Adv. Mater., 2004, 16(1), 54–57.
27. H. Gu, C. Ma, J. Gu, J. Guo, X. Yan, J. Huang, Q. Zhang and Z. Guo, J. Mater. Chem. C, 2016, 4, 5890–5906.
28. G. Wu, Y. Cheng, Q. Xie and C. Liu, J. Polym. Res., 2014, 21, 615–621.
29. J. Gu, X. Yang, Z. Lv, N. Li, C. Liang and Q. Zhang, Int. J. Heat Mass Transfer, 2016, 92, 15–22.
30. G. L. Wu, Y. H. Cheng, K. K. Wang and A. Feng, J. Mater. Sci.: Mater. Electron., 2016, 27(6), 5592–5598.
31. X. Y. Huang and P. K. Jiang, Adv. Mater., 2015, 27(3), 546–554.
32. G. Wu, K. Kou, N. Li, H. Shi and M. Chao, J. Appl. Polym. Sci., 2013, 128, 1164–1169.
33. Y. Wang, K. Kou, W. Zhao, G. Wu and F. Han, RSC Adv., 2015, 5, 99313–99321.
34. C. Pan, K. Kou, Q. Jia, Y. Zhang, Y. Wang, G. Wu and A. Feng, J. Mater. Sci.: Mater. Electron., 2016, 27, 11909–11916.
35. C. Pan, K. Kou, G. Wu and Y. Zhang, J. Mater. Sci.: Mater. Electron., 2016, 27, 286–292.
36. F. L. Riley, J. Am. Ceram. Soc., 2008, 83, 245–265.
37. H. He, R. L. Fu and Y. Shen, Compos. Sci. Technol., 2007, 67, 2493–2499.
38. W. Y. Zhou, S. H. Qi and Q. L. An, Mater. Res. Bull., 2007, 42, 1863–1873.
39. X. L. Zeng, S. H. Yu and R. Sun, J. Appl. Polym. Sci., 2013, 128, 1353–1359.
40. T. L. Zhou, X. Wang and M. Y. Gu, Polymer, 2008, 49, 4666–4672.
41. M. L. McGrath, S. Richard and H. Saskia, Polymer, 2008, 49, 999–1014.
42. S. Kim, S. Ando and X. G. Wang, RSC Adv., 2015, 5, 40046–40054.
43. S. Kim, S. Ando and X. G. Wang, RSC Adv., 2015, 5, 98419–98428.
44. S. Kim, X. Y. Wang, S. Ando and X. G. Wang, RSC Adv., 2014, 4, 27267–27276.
45. G. Wu, Y. Cheng, Z. Wang, K. Wang and A. Feng, J. Mater. Sci.: Mater. Electron., 2016 DOI:10.1007/s10854-016-5560-8.
46. S. C. Liu, H. Q. Zhang, B. Zhang, X. Zhong, J. Ma and F. Ye, Ceram. Int., 2016, 42(6), 7253–7258.
47. T. Zakrzewski and P. Boguslawski, J. Alloys Compd., 2016, 664, 565–579.
48. S. Y. Wu, Y. L. Huang, C. C. M. Ma, S. M. Yuen, C. C. Teng and S. Y. Yang, Composites, Part A, 2011, 42, 1573–1583.
49. Y. C. Zhou, H. Wang, L. Wang, K. Yu, Z. Lin, L. He and Y. Y. Bai, Mater. Sci. Eng., B, 2012, 177, 892–896.
50. H. Okubo, IEEE Trans. Dielectr. Electr. Insul., 2012, 19, 733–754.
51. J. W. Shang, Y. H. Zhang, L. Yu, B. Shen, F. Z. Lv and P. K. Chu, Mater. Chem. Phys., 2012, 134, 867–874.
52. G. X. Hu, F. Gao, J. Kong, S. J. Yang, Q. Q. Zhang, Z. T. Liu, Y. Zhang and H. J. Sun, J. Alloys Compd., 2015, 619, 686–692.
53. Z. M. Dang, H. P. Xu and H. Y. Wang, Appl. Phys. Lett., 2007, 90, 012901–012906.
54. S. Thomas, V. N. Deepu, P. Mohanan and M. T. Sebastian, J. Am. Ceram. Soc., 2008, 91, 1971–1975.
55. A. Luyt, J. Molefi and H. Krump, Polym. Degrad. Stab., 2006, 91, 1629–1636.
56. J. Gu, N. Li, L. Tian, Z. Lv and Q. Zhang, RSC Adv., 2015, 5, 36334–36339.
57. Y. C. Chen, H. C. Lin and Y. D. Lee, J. Polym. Res., 2003, 10, 247–258.
58. M. E. Mackay, A. Tuteja, P. M. Duxbury, C. J. Hawker, B. V. Horn, Z. B. Guan, G. H. Chen and R. S. Krishnan, Science, 2006, 311, 1740–1743.

---