Mechanism on lattice thermal conductivity of carbon-vacancy and porous medium entropy ceramics

Abstract

High-entropy ceramics with vacancy and pores show abnormal lattice thermal conductivity with temperature. In this work, we conduct the machine learning interatomic potentials in combination with the classical molecular dynamics to study the mechanism of lattice thermal conductivity in (NbTaZr)C medium entropy ceramics with different carbon vacancies and porous defects. The trained neuroevolution potentials excellently reproduce the ab initio calculations. Results indicate that both vacancy and pore can enhance the phonon scattering at low- and middle-frequency ranges, reduce the phonon lifetime, shift from acoustic to defect scattering, and make the phonon vibration modes localized. Further, they decrease the temperature dependence of lattice thermal conductivity. We find that propagation-type phonons dominate the thermal conductivity in the perfect structure, whereas diffusion-type phonons become predominant in carbon-vacancy and porous (NbTaZr)C.