High thermal conductivity graphite nanoplatelet/UHMWPE nanocomposites

Abstract

High thermal conductivity graphite nanoplatelet/ultra-high molecular weight polyethylene (GNPs/UHMWPE) nanocomposites are fabricated via mechanical ball milling followed by a hot-pressing method. The GNPs are located at the interface of the UHMWPE matrix. The thermal conductivity coefficient of the GNPs/UHMWPE nanocomposite is greatly improved to 4.624 W m⁻¹ K⁻¹ with 21.4 vol% GNPs, 9 times higher than that of the original UHMWPE matrix. The significantly high improvement of the thermal conductivity is ascribed to the formation of multidimensional thermally conductive GNPs–GNPs networks, and the GNPs have a strong ability to form continuous thermally conductive networks. The method of cooling-pressing on the machine is more beneficial for the improvement of the thermal conductivity, by increasing the crystallinity of the UHMWPE matrix. Furthermore, the thermal stabilities of the GNPs/UHMWPE nanocomposites are increased with increasing addition of GNPs.

---

Introduction

The integration and miniaturization of microelectronic devices creates greater requirements for advanced microelectronic packaging materials.^1–6 Against this background, high thermal conductivity polymeric composites possess significant potential in advanced microelectronic packaging, which requires good heat dissipation, low thermal expansion and light weight.^7,8

However, the thermal conductivity coefficient of polymeric matrix is much lower than those of metals or ceramic materials. Recent studies have revealed that the incorporation of thermally conductive fillers, such as graphene,^9–12 carbon nanotubes,^13–16 boron nitride nanotubes,¹⁷ silicon nitride,^18–20 silicon carbide^21,22 and graphite,^1,23–25 etc., into polymeric matrix can effectively increase the thermal conductivities of the composites. In our previous work,^26–31 several thermally conductive polymeric composites were fabricated successfully by adding single or hybrid thermally conductive fillers. However, the improvement of thermal conductivities of the composites is often less than expected from previous theoretical design. Furthermore, to fabricate polymeric composites with high thermal conductivities above 4 W m⁻¹ K⁻¹, the addition of thermally conductive fillers beyond 30 vol% is required, which creates a significant challenge in processing behavior, mechanical properties and density.³²

According to the heat-transfer mechanism of lattice vibration, the collision probability of a phonon is smaller and the average free path of a phonon is larger in the crystalline region.³³ Therefore, the crystallinity of polymers strongly affects the thermal conductivity, from 0.2 W m⁻¹ K⁻¹ for amorphous polymers to 0.5 W m⁻¹ K⁻¹ for highly crystalline polymers.³⁴ For this reason, the improvement of crystallinity is emerging as one of the most effective ways to increase the intrinsic thermal conductivity of polymeric matrix.

The more recognized thermally conductive mechanism is that the thermal transport occurs through not only the thermally conductive networks of fillers, but also the polymeric matrix. The thermally conductive percolation behavior plays a major role in the great enhancement of the thermal conductivities.^35,36 To the best of our knowledge, a low percolation threshold can be achieved for polymeric composites with segregated structures, in which thermally conductive fillers are located at the interface of the polymeric matrix instead of being randomly distributed in the polymeric matrix.^37,38

Ultra-high molecular weight polyethylene (UHMWPE) is an engineering thermoplastic, replacing the existing conventional polyethylene, due to its outstanding wear resistance, excellent chemical stability, low friction, good self-lubricating properties and bio-compatibility, as well as surprising impact toughness, etc., which has great applications in the fields of transportation, agriculture, medicine, food, chemicals, textiles, paper making, the aerospace industry and military defense equipment.^39–41 Additionally, owing to their superior diameter/thickness ratio, outstanding physical properties and cost efficiency, graphite nanoplatelets (GNPs) are considered to be effective fillers for fabricating polymeric composites with a relatively higher thermal conductivity as compared to graphene, carbon nanotubes and boron nitride nanotubes.^1,26,42–45 Work by Kalaitzidou and coworkers⁴⁶ has also shown that GNPs can obviously enhance the thermal conductivity of polypropylene (PP). The maximum thermal conductivity value measured for 25 vol% GNPs/PP composites was 6 times higher than that of the original PP.

In our present work, the method of mechanical ball milling followed by hot-pressing is introduced to fabricate highly thermal conductive GNPs/UHMPWE nanocomposites with segregated structures. The method of cooling-pressing on the machine after hot-pressing is also adopted to further increase the thermal conductivities of the GNPs/UHMPWE nanocomposites by further improving the crystallinity of the UHMWPE matrix. The thermal conductivities and thermal stabilities of the GNPs/UHMWPE nanocomposites are also investigated.

Materials and methods

Materials

Graphite nanoplatelets (GNPs), KNG-180, with a diameter of 40 μm and a superior diameter/thickness ratio of 250, were received from Xiamen Knano Graphene Technology Co. Ltd. (Fujian, China); ultra-high molecular weight polyethylene (UHMWPE), powder, 0.97 g cm⁻³, mean particle size of 60 μm, was supplied by Nanjing Deyuan Science and Technology Co. Ltd. (Jiangsu, China); tetrahydrofuran (THF) and absolute ethanol were both purchased from Tianjin Ganglong Chemical Group Co., Ltd. (Tianjin, China).

Preparation of the GNPs/UHMWPE nanocomposites

Pristine GNPs were firstly immersed in THF and absolute ethanol for 12 h at room temperature for each step to effectively remove other impurities on the surface of the GNPs, followed by storage at 60 °C in a vacuum oven for 24 h. UHMWPE was firstly dried in a vacuum oven at 60 °C for 6 h. The GNPs/UHMWPE nanocomposites were fabricated according to the following procedures: (i) mixing GNPs and UHMWPE matrix using a ball milling machine for 24 h at room temperature, to embed the GNPs at the interface of the UHMWPE matrix; (ii) hot-pressing (195 °C at 10 MPa) to fabricate the GNPs/UHMWPE with segregated structures. Herein, GNPs are located at the interface of UHMWPE. (Schematic diagram in Fig. 1.)

Fig. 1 Schematic diagram of the thermal conductivity mechanism of the GNPs/UHMWPE nanocomposites.

Two methods of “direct cooling-pressing away from the machine” and “cooling-pressing on the machine” were performed after hot-pressing (195 °C at 10 MPa). In the former, the material is firstly hot-pressed at 195 °C at 10 MPa for 15 min, and then the metal mold is directly removed from the press to a room temperature environment quickly under a specific pressure. In the latter, the material is firstly hot-pressed at 195 °C at 10 MPa for 15 min, and then the press is powered off, but the metal mold is still placed on the press under a specific pressure until cooled to room temperature.

Characterization

Differential scanning calorimetry (DSC) analyses of the samples were carried out using a DSC-2910 (TA Corporation, USA) with heating rates of 5 °C min⁻¹, 10 °C min⁻¹ and 20 °C min⁻¹ under a nitrogen atmosphere; the thermal conductivity coefficients of the samples were measured using a Hot Disk instrument (AB Corporation, Sweden), which is based upon a transient technique. The measurements are performed with bulk specimens (20 × 20 × 4 mm³) by putting the sensor (3 mm diameter) between two similar slabs of material. The sensor supplies a heat pulse of 0.03 W for 20 seconds to the sample, the associated change in temperature is recorded, and the thermal conductivity of the individual samples is obtained.²³ Scanning electron microscopy (SEM) images of the sample morphologies were obtained using a VEGA3-LMH (TESCAN Corporation, Czech Republic); thermo-gravimetric analyses (TGA) of the samples were performed using a STA 449F3 thermoanalyzer (Netzsch Group, Germany) in the temperature range of 40–650 °C with a heating rate of 10 °C min⁻¹ under an argon atmosphere.

Results and discussion

Influence of the cooling rate on the crystallization and thermal conductivity

The influence of the cooling rate on the crystallization and thermal conductivity of the UHMWPE matrix is presented in Fig. 2.

Fig. 2 The influence of the cooling rate on the crystallization and thermal conductivity of the UHMWPE matrix. (a) Crystallization; (b) thermal conductivity.

With the decrease of the cooling rate, the crystallization temperature, crystallization heat and thermal conductivity are all increased. This reveals that the slower the cooling rate is, the higher the crystallization temperature and crystallization heat are. The reason can be ascribed to a slower cooling rate causing the crystallization of the UHMWPE matrix to be more perfect, improving the final crystallinity of UHMWPE accordingly, which is beneficial for phonon transmission, resulting in the higher thermal conductivity of the UHMWPE matrix.

Thermal conductivities of the GNPs/UHMWPE nanocomposites

The influence of the volume fraction of GNPs on the thermal conductivities of the GNPs/UHMWPE nanocomposites is shown in Fig. 3.

Fig. 3 The influence of volume fraction of GNPs on the thermal conductivities of the GNPs/UHMWPE nanocomposites. (a) Thermal conductivity; (b) thermal conductivity enhancement.

GNPs/UHMWPE nanocomposites exhibit a rapid improvement of the thermal conductivities beyond 4.3 vol% GNPs. The thermal conductivity coefficient of the GNPs/UHMWPE nanocomposite with 21.4 vol% (40 wt%) GNPs is greatly improved to 4.624 W m⁻¹ K⁻¹, 9 times higher than that of the original UHMWPE matrix.

The high intrinsic thermal conductivity of GNPs can offer reasonable explanations for the greater improvement of the thermal conductivities of the GNPs/UHMWPE nanocomposites than those of traditional thermally conductive fillers, and the superior diameter/thickness ratio of GNPs can also result in the greater improvement of the thermal conductivities.

Meanwhile, the thermal conductivities of the GNPs/UHMWPE nanocomposites are strongly dependent on the volume fraction of the GNPs in the UHMWPE matrix. GNPs with a low volume fraction have weak interactions with each other and present a relatively small increase in thermal conductivity. With an increasing volume fraction of GNPs, the interconnectivity of the GNPs–GNPs network is improved obviously, and the probability of the formation of thermally conductive networks is increased, and thus the thermal conductivities of the GNPs/UHMWPE nanocomposites are improved obviously. Moreover, our proposed method can fabricate GNPs/UHMWPE nanocomposites with segregated structures, which easily form thermally conductive networks (Fig. 1). Fig. 4 reveals that the multidimensional thermally conductive GNPs–GNPs networks are formed with the addition of 21.4 vol% (40 wt%) GNPs. Fig. 4(a) shows the original UHMWPE matrix, and Fig. 4(b) shows nanocomposites with 21.4 vol% (40 wt%) GNPs.

Fig. 4 SEM morphologies of (a) the original UHMWPE matrix and (b) GNPs/UHMWPE nanocomposite with 21.4 vol% (40 wt%) GNPs.

As also seen from Fig. 3, for the same volume fraction of GNPs, the GNPs/UHMWPE nanocomposites via cooling-pressing on the machine possess relatively higher thermal conductivities. This is attributed to UHMWPE being able to crystallize more slowly at a lower cooling rate. Therefore, the corresponding crystallinity and crystal perfection of the UHMWPE matrix are both improved, which is beneficial for phonon transmission, resulting in the higher thermal conductivities.

Agari’s semi-empirical model fitting of the GNPs/UHMWPE nanocomposites

Agari’s semi-empirical model can yield better results than the theoretical ones. The logarithmic equation of Agari is written as follows:^47,48

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

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

Fig. 5 shows the logarithmic values of the thermal conductivities as a function of the GNPs volume fraction.

Fig. 5 Logarithmic thermal conductivities of the GNPs/UHMWPE nanocomposites as a function of the GNPs volume fraction.

The parameters of C_p and C_f are calculated to be 1.2851 and 0.761, respectively. The high value of C_p suggests that the GNPs can influence the crystallinity of the UHMWPE matrix. The low value of C_f suggests that the GNPs have a strong ability to form continuous thermally conductive networks, that is, the formation of thermally conductive networks becomes much easier with the incorporation of GNPs.

Thermal properties of the GNPs/UHMWPE nanocomposites

Fig. 6 shows the influence of the mass fraction of GNPs on the melting heat and melting temperature of the GNPs/UHMWPE nanocomposites determined using DSC analysis. The corresponding thermal data are listed in Table 1.

Fig. 6 DSC curves of the original UHMWPE and the GNPs/UHMWPE nanocomposites.

Table 1 Thermal data of the original UHMWPE and the GNPs/UHMWPE nanocomposites from DSC analysis

Samples	Melting heat/(J g^−1)	Melting temperature/°C
Original UHMWPE	158.6	139.1
5 wt% GNPs + UHMWPE	156.4	137.2
10 wt% GNPs + UHMWPE	147.1	136.8
20 wt% GNPs + UHMWPE	149.7	136.5
40 wt% GNPs + UHMWPE	145.9	135.4

---

Both the melting heat and melting temperature of the GNPs/UHMWPE nanocomposites are decreased slightly with increasing addition of GNPs. The corresponding melting heat is decreased from 158.6 J g⁻¹ (original UHMWPE) to 156.4 J g⁻¹ (5 wt% GNPs), 147.1 J g⁻¹ (10 wt% GNPs), 149.7 J g⁻¹ (20 wt% GNPs) and 145.9 J g⁻¹ (40 wt% GNPs). Meanwhile, the corresponding melting temperature is also decreased from 139.1 °C (original UHMWPE) to 137.2 °C (5 wt% GNPs), 136.8 °C (10 wt% GNPs), 136.5 °C (20 wt% GNPs) and 135.4 °C (40 wt% GNPs). This can be attributed to the heterogeneous nucleation of GNPs being able to hinder the homogeneous nucleation of the UHMWPE matrix. Meanwhile, the addition of GNPs can also increase the thickness of crystal plates of the UHMWPE system. A combination of both action above would effectively decrease the heat enthalpy of the GNPs/UHMWPE nanocomposites, decreasing the final crystallinity accordingly.

TGA curves of the original UHMWPE matrix and GNPs/UHMWPE nanocomposites are presented in Fig. 7. The corresponding characteristic thermal data of the original UHMWPE matrix and the GNPs/UHMWPE nanocomposites are listed in Table 2.

Fig. 7 TGA curves of the original UHMWPE and the GNPs/UHMWPE nanocomposites.

Table 2 Thermal data of the original UHMWPE matrix and the GNPs/UHMWPE nanocomposites from TGA analysis

Samples	Temperature/°C		Heat-resistance index^a/°C	Residual mass/%
	T5	T30		Theory	Actual
a Heat resistance index = 0.49[T5 + 0.6(T30 − T5)], T5, T30 is the decomposing temperature at 5%, 30% weight loss, respectively.^28
Original UHMWPE	410.3	449.2	212.5	0	1.49
5 wt% GNPs/UHMWPE	414.0	452.9	214.3	5	4.98
10 wt% GNPs/UHMWPE	417.2	456.1	215.9	10	9.17
20 wt% GNPs/UHMWPE	418.3	457.2	216.4	20	19.15
40 wt% GNPs/UHMWPE	422.7	461.2	218.4	40	38.15

---

The corresponding weight loss temperatures are increased at the same stages with increasing addition of GNPs. The corresponding heat-resistance indices of original UHMWPE and GNPs/UHMWPE nanocomposites are 212.5 °C, 214.3 °C (5 wt% GNPs), 215.9 °C (10 wt% GNPs), 216.4 °C (20 wt% GNPs) and 218.4 °C (40 wt% GNPs), respectively. This suggests that the thermal stabilities of the GNPs/UHMWPE nanocomposites are increased. The reason is that GNPs possess higher heat capacity and thermal conductivity compared to the original UHMWPE. Therefore, GNPs can preferentially absorb the heat, resulting in UHMWPE degrading at higher temperatures. In addition, the actual residual mass value is very close to the theoretical residual mass value, which proves that the GNPs are located uniformly at the interface of the UHMWPE matrix.

Conclusion

GNPs are located at the interface of the UHMWPE matrix via our proposed method of mechanical ball milling followed by hot-pressing. The thermal conductivity coefficient of the GNPs/UHMWPE nanocomposite with 21.4 vol% (40 wt%) GNPs is greatly improved to 4.624 W m⁻¹ K⁻¹, 9 times higher than that of the original UHMWPE matrix. The significantly high improvement of the thermal conductivity is ascribed to the formation of multidimensional thermally conductive GNPs–GNPs networks. Agari’s semi-empirical model fitting reveals that the GNPs have a strong ability to form continuous thermally conductive networks. The method of cooling-pressing on the machine is beneficial for increasing the thermal conductivities of the GNPs/UHMWPE nanocomposites by further improving the crystallinity of UHMWPE matrix. Furthermore, the thermal stabilities of the GNPs/UHMWPE nanocomposites are increased with increasing addition of GNPs.

Acknowledgements

The authors are grateful for the support and funding from the Foundation of Natural Science Foundation of China (no. 51403175 and no. 81400765); Shaanxi National Science Foundation of Shaanxi Province (no. 2014JQ6203); Space Supporting Fund from China Aerospace Science and Industry Corporation (no. 2014-HT-XGD) and the Fundamental Research Funds for the Central Universities (no. 3102015ZY066).

References

1. C. Min, D. M. Yu, J. Y. Cao, G. L. Wang and L. H. Feng, Carbon, 2013, 55, 116–125.
2. Y. C. Zhang, K. Dai, J. H. Tang, X. Ji and Z. M. Li, Mater. Lett., 2010, 64, 1430–1432.
3. K. Kim, M. Kim and J. Kim, Compos. Sci. Technol, 2014, 103, 72–77.
4. K. M. F. Shahil and A. A. Balandin, Nano Lett., 2012, 12, 861–867.
5. L. M. Gao, T.-W. Chou, E. T. Thostenson, A. Godara, Z. G. Zhang and L. Mezzo, Carbon, 2010, 48, 2649–2651.
6. T. Luo and J. R. Lloyd, Adv. Funct. Mater., 2012, 22, 2495–2502.
7. Q. Liang, X. Yao, W. Wang, Y. Liu and C. P. Wong, ACS Nano, 2011, 5, 2392–2401.
8. M. D. Losego, M. E. Grady, N. R. Sottos, D. G. Cahill and P. V. Braun, Nat. Mater., 2012, 11, 502–506.
9. S. H. Song, K. H. Park, B. H. Kim, Y. W. Choi, G. H. Jun and D. J. Lee, Adv. Mater., 2013, 25, 732–737.
10. H. Im and J. Kim, Carbon, 2012, 50, 5429–5440.
11. Y. X. Fu, Z. X. He, D. C. Mo and S. S. Lu, Int. J. Therm. Sci., 2014, 86, 276–283.
12. C. Heo and J.-H. Chang, Solid State Sci., 2014, 24, 6–14.
13. V. Datsyuk, S. Trotsenko and S. Reich, Carbon, 2013, 52, 605–608.
14. C. H. Liu and S. S. Fan, Appl. Phys. Lett., 2005, 86, 123106–123109.
15. S. Y. Kwon, I. I. M. Kwon, Y.-G. Kim, S. Lee and Y.-S. Seo, Carbon, 2013, 55, 285–290.
16. S. Y. Pak, H. M. Kim, S. Y. Kim and J. R. Youn, Carbon, 2011, 50, 4830–4838.
17. X. Y. Huang, C. Y. Zhi, P. K. Jiang, D. Golberg, Y. Bando and T. Tanaka, Adv. Funct. Mater., 2013, 23, 1824–1831.
18. Y. Zhu, K. X. Chen and F. Y. Kang, Solid State Commun., 2013, 158, 46–50.
19. H. He, R. L. Fu, Y. Shen, Y. C. Han and X. F. Song, Compos. Sci. Technol., 2007, 67, 2493–2499.
20. W. Y. Zhou, C. F. Wang, T. Ai, K. Wu, F. J. Zhao and H. Z. Gu, Composites, Part A, 2009, 40, 830–836.
21. T. L. Zhou, X. Wang, X. H. Liu and D. S. Xinog, Carbon, 2010, 48, 1171–1176.
22. K. Yang and M. Y. Gu, Composites, Part A, 2010, 41, 215–221.
23. H. M. Tu and L. Ye, Polym. Adv. Technol., 2009, 20, 21–27.
24. N. Vitorino, J. C. Abrantes and J. R. Frade, Mater. Lett., 2013, 92, 100–103.
25. Z. Li, W. G. Sun, G. Wang and Z. G. Wu, Sol. Energy Mater. Sol. Cells, 2014, 128, 447–455.
26. J. W. Gu, J. J. Du, J. Dang, W. C. Geng, S. H. Hu and Q. Y. Zhang, RSC Adv., 2014, 4, 22101–22105.
27. J. W. Gu, C. Xie, H. L. Li, J. Dang and Q. Y. Zhang, Polym. Compos., 2014, 35, 1087–1092.
28. J. W. Gu, Q. Y. Zhang, J. Dang, C. J. Yin and S. J. Chen, J. Appl. Polym. Sci., 2012, 124, 132–137.
29. J. W. Gu, Q. Y. Zhang, J. Dang and C. Xie, Polym. Adv. Technol., 2012, 23, 1025–1028.
30. J. W. Gu, Q. Y. Zhang, J. Dang, J. P. Zhang and Z. Y. Yang, Polym. Eng. Sci., 2009, 49, 1030–1034.
31. J. W. Gu, Q. Y. Zhang, J. Dang, J. P. Zhang and S. J. Chen, Polym. Bull., 2009, 62, 689–697.
32. Z. Han and A. Fina, Prog. Polym. Sci., 2011, 36, 914–944.
33. X. D. Yang, S. H. Wu and Y. H. Song, Adhesion, 2012, 33, 75–79.
34. M. Hu, D. Yu and J. Wei, Polym. Test., 2007, 26, 333–337.
35. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney and E. A. Stach, et al., Nature, 2006, 42, 282–286.
36. N. Shenogina, S. Shenogin, L. Xue and P. Keblinski, Appl. Phys. Lett., 2005, 87, 133106.
37. C. Zhang, C. A. Ma, P. Wang and M. Sumita, Carbon, 2005, 43, 2544–2553.
38. J. F. Gao, Z. M. Li, Q. J. Meng and Q. Yang, Mater. Lett., 2008, 62, 3530–3532.
39. M. Sattari, M. R. Naimi-Jamal and A. Khavandi, Polym. Test., 2014, 38, 26–34.
40. D. I. Chukov, A. A. Stepashkin, M. V. Gorshenkov, V. V. Tcherdyntsev and S. D. Kaloshkin, J. Alloys Compd., 2014, 586, S459–S463.
41. J. H. Hong, Z. J. Pan, M. Yao and X. Y. Zhang, Synth. Met., 2014, 193, 117–124.
42. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth and T. Stauber, Science, 2008, 320, 1308.
43. Q. Liu, Z. F. Liu, X. Y. Zhang, L. Y. Yang, N. Zhang and G. L. Pan, et al., Adv. Funct. Mater., 2009, 19, 894–904.
44. A. B. Kuzmenko, E. F. van Heumen, F. Carbone and D. van der Marel, Phys. Rev. Lett., 2008, 100, 117401.
45. L. Y. Zhang and G. H. Chen, Mater. Rev., 2011, 25, 85–88.
46. K. Kalaitzidou, H. Fukushima and L. T. Drzal, Carbon, 2007, 45, 1446–1452.
47. Y. Agari, A. Ueda, M. Tanaka and S. Nagai, J. Appl. Polym. Sci., 1999, 40, 929–941.
48. Y. Agari, A. Ueda and S. Nagai, J. Appl. Polym. Sci., 1991, 43, 1117–1124.

---