Highly thermally conductive polymer composite enhanced by two-level adjustable boron nitride network with leaf venation structure

Highlights

• The composite with high thermal conductivity is obtained.

• The primary and secondary BN networks construct efficient phonon conduction channels.

• The polydopamine interface greatly reduces interfacial phonon scattering.

• The composite has excellent electrical insulation.

Abstract

Thermally conductive polymer-based composites are extensively used in many fields as thermal control materials. Their thermal conductivity can be effectively improved via the construction of a 3D thermal conduction network. However, multiple 3D networks have low density and lack elasticity and flexibility, leading to suboptimal thermal conductivity. In this study, a composite with high thermal conductivity is obtained by building a two-level adjustable boron nitride (BN) network with leaf venation structure in an epoxy resin matrix, and the density and orientation of the network are controlled by compression. The primary and secondary BN networks construct efficient phonon conduction channels. Moreover, the polydopamine interface between the thermally conductive network and substrate greatly reduces interfacial phonon scattering. The in-plane and cross-plane thermal conductivities of the composite at 35.9 wt% BN loading reach 10.20 and 4.95 W m⁻¹K⁻¹, respectively. And the composite has excellent electrical insulation, all making it promising for the thermal management of electronic equipment and thermal interface material in application prospects, such as the soft robotics, flexible smart devices, and aerospace.

Introduction

The power density of electronic components has rapidly increased with the development of semiconductor technology, leading to heat accumulation [[1], [2], [3]]. Therefore, the thermal management of modern electronic devices have become a major concern. In recent years, thermally conductive filler-reinforced polymer composites have been used for thermal management because of their light weight, excellent processing performance, and corrosion resistance [[4], [5], [6]]. However, it is still a challenge to considerably improve their thermal conductivity (κ) because of the large amount of interfacial thermal resistance [7,8].

High tensile strength can improve the stability of thermal conductive materials, and promote the multi field and application of materials. Therefore, controlling the filler structure and achieving high strength and high thermal conductivity at low filler content is a challenge in the research of thermal conductive materials. Moreover, studies have been conducted to reduce the interfacial thermal resistance between the matrix and thermally conductive fillers, and building a continuous heat conduction network of fillers, which can be prepared by the template method [[9], [10], [11], [12], [13], [14], [15]], self-assembly [16,17], and freeze-drying [[18], [19], [20], [21], [22], [23]], was reported to be the most effective method to reduce the interfacial thermal resistance. For example, Cheng et al. [24] prepared a composite using aligned graphene nanosheets filled with a continuous network of graphene foam. The composite shows the highest thermal conductivity enhancement ever reported at room temperature (8100%) at a graphene loading of 6.2 vol%. However, the mechanical strength and machinability of materials are poor. Fu et al. [25] prepared a synergistic segregated thermal conductive polystyrene composite with graphene nanoplates (GPNs) and multiwalled carbon nanotubes (MWCNTs) double network. The improvement in κ is primarily attributed to the synergistic effect of GPNs and MWCNTs. For the freeze drying method, Wu et al. [26] prepared an epoxy composite containing a 3D network of nanofibrillated cellulose (NFC)-assisted unidirectional freeze drying of a boron nitride nanosheets (BNNSs) slurry. The composite possesses a high cross-plane κ of 1.56 W m⁻¹K⁻¹ at an extremely low BNNSs loading of 4.4 vol%. The thermal conductivity and mechanical properties of the composites have been significantly improved under low filler.

Compared with the traditional blending process, although the construction of the thermal conduction network considerably improves the κ of polymer, the κ of the composite is still relatively low because of the relatively low density of the thermal conduction network. Because of the large viscosity of the polymer matrix, it is difficult to increase the density of the prefabricated network by reducing the pore size because small pores cannot be filled, reducing both the density and κ [27]. To solve this limitation, we previously reported a double heat conduction network of graphene and melamine with a high κ and hyperelasticity [28]. The density of the network can be adjusted by compression before curing the polymer matrix. The heat conduction of polymer composites can be considered as two-stage phonon transmission. The first stage, phonons transmitting from the matrix to the heat conduction network, is closely connected to the polymer and network, which is critical for improving the κ of the composite [29]. Although the compressed high-density continuous thermal conduction network provides abundant phonon transfer paths and promotes thermal conductivity, the average distance is considerably large, limiting the improvement of κ.

Herein, because of the leaf vein network with high-efficiency mass transfer, we proposed a two-level adjustable hexagonal boron nitride (hBN) network with leaf venation structure to enhance the κ of epoxy (EP) composite. The primary thermal conduction network melamine-formaldehyde sponge @ boron nitride nanosheets (MS@BNNSs) was developed by coating melamine-formaldehyde sponge (MS) with BNNSs. The secondary thermal conduction network was formed by adding the dopamine-treated hBN (hBN@PDA) into the EP matrix. Polymer composites (MS@BN/EP) were obtained by impregnating hBN@PDA/EP into MS@BNNSs, followed by deformation and cross-linking. The primary and secondary hBN networks guarantee the efficient transmission of heat through the polymer matrix, and the hyperelasticity facilitates the control of the network structure and the κ of the composites. Both in or cross-plane κ of the composite reached 10.20 and 4.95 W m⁻¹K⁻¹, respectively, at a hBN loading of 35.9 wt%. The high κ results mainly from the formation of a high-density phonon transmission pathway provided by the two-level BN network. Moreover, hBN@PDA can effectively connect the filler with EP, leading to good interfacial compatibility and reduced phonon interfacial scattering. Owing to the excellent thermal conduction and electrical insulation performance, the composite holds considerable potential as an integrated circuit board for thermal management in electronic devices.