Coherent and incoherent phonon transport in graphene/h-BN superlattice: A machine learning potential

Currently, numerous studies demonstrate two mechanisms of coherent and incoherent phonon transport in superlattices. The graphene/hexagonal boron nitride (h-BN) superlattice has attracted significant attention because of its excellent thermal, electrical, and mechanical properties. In this study, we construct an accurate neuroevolution potential (NEP) model and conduct molecular dynamics (MD) simulations using the GPUMD package to investigate phonon transport behavior in the graphene/h-BN superlattice. We find that with increasing periods, phonons transition from coherent to incoherent transport, with their intersection corresponding to the lowest thermal conductivity. Additionally, we discuss phonon transport under varying temperature conditions. We also explore phonon transport in aperiodic superlattices to further illustrate the impact of structural disorder. This study enhances our understanding of phonon transport in superlattices and has the potential to broaden the application of graphene/h-BN superlattices in thermal management and electronic devices.