Structure-Aware Design and Percolation Modeling of Segregated Polymer Composites for Advanced Electronic Thermal Management

Segregated polymer composites achieve high electrical and thermal conductivity by confining fillers at interparticle interfaces, but their performance is limited by micro-voids and by models that neglect their three-dimensional structure. This presentation integrates two studies establishing a structure-aware design and modeling framework for segregated composites. First, polypropylene composites containing graphene nanoplatelets or hexagonal boron nitride were engineered using a low-melting terpolymer to fill interfacial micro-voids while preserving excluded-volume-driven networks. Micro-computed tomography and finite element analysis were incorporated into new electrical and thermal percolation models, directly linking excluded volume, micro-void fraction, and network connectivity to transport properties. The optimized GNP and hBN composites increased maximum filler loading by 113.3% and 130.9%, respectively, while improving thermal conductivity by 68.1% and 53.5% without sacrificing mechanical integrity. Building on this concept, amine-functionalized hBN was produced by sequential Ar/NH3 plasma treatment and incorporated into epoxy-based segregated thermal interface materials. Covalent bonding at the hBN–epoxy interface improved filler localization, suppressed micro-voids, reduced thermal resistance, and formed continuous heat-transfer pathways. The optimized composite reached 6.5 W m−1 K−1 at 30 vol% hBN and showed lower thermal resistance than a commercial TIM in GPU and CPU demonstrations. An excluded-volume-informed Foygel model coupled with FEM simulations accurately described the thermal percolation behavior. Together, these studies show that simultaneous control of excluded volume, micro-voids, and interfacial bonding provides a general strategy for high-performance multifunctional polymer composites for electronic thermal management.