Unveiling the thermal transport mechanism in compressed plastic crystals assisted by deep potential

TL;DR

This study uses molecular dynamics with deep neural network potentials to show that applying 9% compressive strain significantly increases the thermal conductivity of a plastic crystal, aiding its use in thermal management.

Contribution

It demonstrates that strain engineering can enhance thermal transport in plastic crystals, providing a new approach for their application in solid-state cooling.

Findings

Thermal conductivity increased sixfold under 9% strain.

Enhanced group velocity and reduced phonon scattering drive the conductivity increase.

Volume compression within 0-1 THz frequency range is key to the mechanism.

Abstract

The unique properties of plastic crystals highlight their potential for use in solid-state refrigeration. However, their practical applications are limited by thermal hysteresis due to low thermal conductivity. In this study, the effect of compressive strain on the thermal transport properties of plastic crystal [(CH3)4N][FeCl4] was investigated using molecular dynamic simulation with a deep neural network potential. It is found that a 9% strain along [001] direction enhances thermal conductivity sixfold. The underlying mechanisms are analyzed through vibrational density of states, spectral energy densities, and mean square displacements. The enhancement in thermal conductivity is primarily due to increased group velocity and reduced phonon scattering, driven by volume compression within the 0-1 THz. These findings offer theoretical insights for the practical application of plastic crystals in thermal management systems.