Thermal transport properties of two-dimensional boron dichalcogenides from a first-principles and machine learning approach

Abstract

The investigation of thermal transport is crucial to the thermal management of modern electronic devices. To obtain the thermal conductivity through solution of Boltzmann transport equation, the calculations of anharmonic interatomic force constants have a high computation cost based on the current method of single-point density functional theory force calculations. The recent suggested machine learning interatomic potentials (MLIPs) method can avoid the huge computation demands. In this work, we study the thermal conductivity of 2D MoS2-like H-B2Ⅵ2 (Ⅵ = S, Se and Te) with the combination of MLIPs and phonon Boltzmann transport equation. The room temperature thermal conductivity of H-B2S2 can reach up to 336 Wm-1K-1, obviously larger than that of H-B2Se2 and H-B2Te2. It is mainly derived from the difference of phonon group velocity. By substituting the different chalcogen elements in the second sublayer, H-B2ⅥⅥ' have lower thermal conductivity comparing with that of H-B2Ⅵ2. The room-temperature thermal conductivity of B2STe is only 11% of that for H-B2S2. It is explained from the comparing phonon group velocity and phonon relaxation time. The MLIPs is proved to be an efficient method to study the thermal conductivity of materials and H-B2S2-based nanodevices have excellent thermal conduction.