Alignments and network of graphite fillers to improve thermal conductivity of epoxy-based composites

Instead of improving the fillers dispersion in the matrix, some fillers alignments and structured composites were investigated in order to highlight their impact on thermal conductivity. Whereas well dispersed graphite-nanocomposites show some limit to reach high thermal conductivity values (0.84 W m(-1) K-1 at 12 wt.%), 3D-structured composite or Z-pinning samples display much better enhancements of apparent thermal conductivity, reaching 2.1 W m(-1) K-1 at 15 wt.%. Impact of insulating DGEBA interfaces was also investigated in this work. It was demonstrated that only two 4 mu m-DGEBA layers cutting the fibers alignment is enough to bring thermal conductivity back to the value of the non-structured nanocomposite, losing all the positive impact of alignment. Mathematical evaluations helped estimating the through-plane thermal conductivities of the samples, highlighting the negative impact of interfaces, and displaying the major difference between a 3D-network sample and a Z-pinned aligned sample. Whereas the 3D-network sample displays a relatively good improvement of both in-plane and through-plane thermal conductivities, the Z-pinned sample presents a considerable increase of the through-plane thermal conductivity (until 6.8 W m(-1) K-1), but also a negligible effect on the in-plane thermal conductivity. Resulting apparent thermal conductivities of both samples are finally quite comparable and more than doubled compared to non-structured nanocomposites. (C) 2015 Elsevier Ltd. All rights reserved.