Discovery of liquid crystalline polymers with high thermal conductivity using machine learning

Abstract

Next-generation power electronics require efficient heat dissipation management, and molecular design guidelines are needed to develop polymers with high thermal conductivity. Polymer materials have considerably lower thermal conductivity than metals and ceramics due to phonon scattering in the amorphous region. The spontaneous orientation of the molecular chains of liquid crystalline polymers could potentially give rise to high thermal conductivity, but the molecular design of such polymers remains largely empirical. In this study, we developed a machine learning model that predicts with more than 96% accuracy whether liquid crystalline states will form based on the chemical structure of the polymer. By exploring the inverse mapping of this model, we identified a comprehensive set of chemical structures for liquid crystalline polyimides. The polymers were then experimentally synthesized, and the results confirmed that they form liquid crystalline phases, with all polymers exhibiting calculated thermal conductivities within the range of 0.722–1.26 W m⁻¹ K⁻¹.