A coarse-grained modeling scheme to characterize thermal transport properties in thermoplastic polymers

Abstract

Though all-atomistic (AA) molecular dynamics (MD) simulations have been effectively employed to develop structure-to-property relationships in thermal transport phenomena, the modeling of structures with millions of atoms pertinent to the real microstructures of highly conductive condensed matter is rarely attempted. Herein, we present a novel coarse-grained (CG) modeling scheme for predicting the thermal conductivity of an amorphous polymer using nylon 6 as a representative thermoplastic, based on non-equilibrium molecular dynamics (NEMD) simulations. To accurately describe the thermal transport through the primary and secondary bonds in any structural conformation of nylon 6, the CG potential parameters were systematically optimized using a particle swarm optimization (PSO) algorithm to reproduce the thermal conduction of a single chain as well as bulk amorphous state. The validity and performance of the newly developed CG potential parameters were primarily confirmed from the compatibility with the reference AA model in terms of two basic structure-property relationships: the domain-size effect on thermal conductivity of a single chain and strain-dependent thermal conductivity of bulk amorphous state. Finally, we calculated the vibrational properties of the amorphous state and found that the CG model could describe the low-frequency vibrational motion that was primarily observed in the AA model.