Modern electronic devices are increasingly integrated and miniaturized, which renders them increasingly ubiquitous and wearable. As the power density of electronic devices increases, the quantity of heat generated also increases. These trends have increased the demand for thermal management in electronic devices because heat control affects device performance, stability, and lifetime [1]. Heat dissipation, such as by using materials with high thermal conductivity, is a common option. Given the needs of wearable and portable electronic devices, such materials should be nonfragile, lightweight, inexpensive, and insulating [2]. Polymer-based materials are good options in this context. Although polymers typically exhibit low thermal conductivity, polymer composites of thermally conductive materials exhibit improved thermal conductivity and electrical insulation [3-7]. Representative examples include composites with metal nitride [8, 9], metal oxides [3, 10], carbon nanomaterials [11, 12]. However, a high filler content was often required to improve the thermal conductivity, which impairs the properties of the polymer. It should be also noted that the out-of-plane (through-plane) conductivity of the composites was often lower than that of in-plane, and that often one or the other not was reported [7]. Both conductivities are important for effective heat dissipation in electronic devices.
Epoxy resins are common thermosetting polymers in electrical devices. Such resins offer many advantages: strong adhesion to various materials, excellent corrosion resistance, excellent electrical insulation properties, and easy processing [13]. The thermal conductivity of epoxy resins is as low as ~0.2 W/(m·K) [7]. There have been studies on epoxy-based composites filled with a variety of thermally conductive materials [7, 14-21]. These studies, however, showed that high contents of fillers were required to achieve high conductivity.
Graphene has emerged as a promising two-dimensional nanomaterial in the last 2 decades. It exhibits high thermal conductivity, high mechanical toughness, high gas-barrier performances, and useful optical properties [22-24]. Over the last decade, much work has focused on preparing graphene and graphene nanosheets from graphite [11, 23, 25-30]. However, because of the poor compatibility of graphene with commodity polymers, it is stile challenging to prepare a homogeneous polymer–graphene composite. We previously reported liquid-phase exfoliation of pristine graphite by using poly(3-hexylthiophene-2,5-diyl) (P3HT) as a dispersant [31, 32]. P3HT is a well-studied polythiophene that contains p-conjugated polymers of good thermal conductivity, electrical conductivity, and processability [33, 34]. We prepared a P3HT/graphene complex without loss of electrical conductivity and dispersed it in an organic solvent [31, 32], which implies the potential of P3HT/graphene complexes as thermal conductive fillers. Aluminum nitride (AlN) also has high thermal conductivity and exhibits insulating properties [8, 35], which are useful for thermal management in electric devices. In fact, AlN is already in practical use. In the present study, we prepared epoxy resin films containing P3HT/graphene and AlN, and compared the effects of P3HT/graphene and AlN on the in-plane and out-of-plane conductivities of the epoxy resin.
We prepared the epoxy resin films containing fillers based on a typical curing procedure (see ESI). The thickness of the prepared epoxy films ranged from 200–300 µm. Electrical resistance measurements indicate that the prepared epoxy films exhibited insulating properties (>106 Ω/cm); although pristine graphite flakes, pristine P3HT, and P3HT/graphene complex exhibited electrical conductivity [31].
Fig. 1a–1d shows photos of epoxy resin films containing graphite flakes (Fig. 1a and 1b), 0.1 wt% P3HT (Fig. 1c), and P3HT/graphene complex (Fig. 1d and 1e). In the absence of P3HT, the graphite flakes aggregated in the epoxy resin films (Fig. 1a and 1b). The epoxy resin film containing 0.1 wt% P3HT was slightly colored (Fig. 1c), indicating dispersion of P3HT throughout the epoxy matrix. When we used the P3HT/graphene complex as a filler, the P3HT/graphene complex seemed to be dispersed throughout the epoxy resin films (Fig. 1d and 1e). Carbon nanomaterials tend to aggregate in solvents and in polymer matrices because of strong van der Waals interaction between them [7], which also induced aggregation of graphite flakes in the epoxy resin. Because the complex formation with P3HT improves the dispersity of graphene nanosheets in an organic solvent [31, 32], P3HT also contributed to the dispersion of graphene nanosheets in the epoxy resin (Fig. 1d and 1e). Fig. 1f indicates that 30 wt% AlN was dispersed throughout the epoxy resin film.
Because of the nanoscale of graphene, we observed epoxy resin films containing graphite flakes and P3HT/graphene complex by transmission electron microscopy (TEM) (Fig. 2a and b). These observations are consistent with the results in Figs. 1e and 1f. Because the AlN particles were micrometers in diameter, we used scanning electron microscopy (SEM) for observing AlN and a cross-section of the epoxy resin film containing 30 wt% AlN (Fig. 2c and 2d). Unaltered AlN particles were evident in the epoxy resin film. Fig. 2d also indicates aggregation and connection of AlN particles in the epoxy resin film, probably because of the high content of AlN. Fig. S1a shows X-ray computed tomography (CT) images of an epoxy resin film containing 30 wt% AlN. White dots indicate the AlN particles in the film. X-ray CT analysis is nondestructive, further indicating the homogeneous dispersion of AlN particles in the film. X-ray CT analysis did not detect the P3HT/graphene complex because of the lack of a differential X-ray absorption between the P3HT/graphene complex and epoxy resin (Fig. S1b).
Fig. 3 shows thermogravimetric analysis (TGA) profiles of epoxy resin films containing various types of filler. Thermal decomposition of pristine epoxy resin started at ca. 230°C. Decomposition of an epoxy resin containing P3HT and P3HT/graphene complex started at ca. 170°C, whereas pristine P3HT and graphite flakes were as thermally stable as or much more stable than pristine epoxy resin. Adding P3HT and P3HT/graphene complex might affect polymerization of the epoxy monomers, which would decrease the thermal stability of the epoxy resin. The TGA profile of the epoxy resin containing 30 wt% AlN was similar to that of pristine epoxy resin, indicating negligible interactions between the epoxy resin and AlN.
We measured the in-plane and out-of-plane thermal conductivities of epoxy resin films containing fillers (Fig. 4). The epoxy resin film (without any fillers) exhibited a thermal conductivity of 0.08 W/(m·K), both in-plane and out-of-plane, comparable with previous reports [36-38]. Adding graphite flakes increased the in-plane and out-of-plane thermal conductivities (Fig. 4a). Upon adding 0. 1 wt% graphite flakes to the epoxy resin, the in-plane thermal conductivity was maximal among the films with varying the graphite flake content. The out-of-plane thermal conductivity was constant for graphite flakes >0.1 wt% and was less than the in-plane thermal conductivity, which indicates surface segregation of graphite flakes in the epoxy resin. Upon adding P3HT to the epoxy resin film, the in-plane thermal conductivity also increased (Fig. 4b). The P3HT content of 0.05 wt% and 0.9 wt% corresponded to the highest thermal conductivity in experiments that varied the P3HT content. Although adding P3HT also increased the out-of-plane thermal conductivity, the out-of-plane conductivity was constant for P3HT >0.025 wt%.
When we added a P3HT/graphene complex to the epoxy resin, the in-plane thermal conductivity fluctuated in accordance with the complex content (Fig. 4c). Adding 0.1 wt% P3HT/graphene complex gave the highest in-plane thermal conductivity [0.99 W/(m·K)], whereas 0.9 wt% P3HT/graphene complex resulted in 0.53 W/(m·K). The in-plane thermal conductivity did not increase in proportion to the content of graphite flakes, P3HT, and P3HT/graphene complex. The out-of-plane thermal conductivity increased with increasing content of P3HT/graphene complex. The P3HT/graphene complex content of 0.9 wt% afforded 0.58 W/(m·K), which is almost the same as the corresponding in-plane thermal conductivity. The coincidence of the out-of-plane and in-plane thermal conductivities at 0.9 wt% P3HT/graphene complex indicates isotropic distribution of the P3HT/graphene complex in the epoxy resin.
Fig. 4d shows the effect of AlN on the thermal conductivity of the epoxy resin film. Both in-plane and out-of-plane thermal conductivities increased in proportion to the AlN content, which is in good agreement with previous reports [16, 39]. The in-plane and out-of-plane thermal conductivities were almost equal to one another, implying isotropic distribution of AlN in the epoxy resin. These results are consistent with the X-ray CT results. An AlN content of 40 wt% only resulted in thermal conductivity of ca. 0.5 W/(m·K), whereas a P3HT/graphene content of only 0.9 wt% corresponded to a thermal conductivity of >0.5 W/(m·K).
In summary, we prepared epoxy resin composites (filled with a P3HT/graphene complex and AlN) that exhibited insulating properties. We evaluated the out-of-plane and in-plane thermal conductivities. Adding a small quantity of P3HT/graphene complex (<1 wt%) increased the out-of-plane and in-plane thermal conductivities of the epoxy resin. Adding AlN also increased the thermal conductivities, but required ≤40 wt% to obtain a thermal conductivity comparable with that of a 0.9 wt% P3HT/graphene complex. These results reveal the potential of P3HT/graphene complexes as fillers for improving the thermal conductivity of epoxy resins without affecting their insulating properties.