Intrinsic thermal conductivity of ZrC from low to ultra-high temperatures: A critical revisit

Abstract

Current phonon transport theory based on ground-state calculations has been successful in predicting thermal conductivity at room and medium temperatures but may misrepresent behavior at high temperatures. In this work, we predict the thermal conductivity ($\kappa$) of ZrC including electronic and phonon contributions from 300 K to 3500 K, by including high-order phonon scattering, lattice expansion, temperature-dependent (TD) harmonic and anharmonic force constants, and inter-band phonon conduction by using first principles. For the phonon transport, we find that four-phonon scattering significantly reduces the phonon thermal conductivity ($\kappa_{ph}$), e.g., by $\sim$60% and $\sim$75% at 2500 K and 3500 K, respectively. After including four-phonon scattering and all other factors, $\kappa_{ph}$ shows a $\sim$T$^{-1.5}$ rather than $\sim$T$^{-1}$ dependence. The contribution from inter-band (Wigner) phonon conduction is small, even at ultra-high temperatures. The temperature dependence of anharmonic force constants decreases the phonon scattering cross-section at elevated temperatures and increases the $\kappa_{ph}$ significantly (by 52% at 3500 K). For the electronic thermal transport, we find that it is sensitive to and can be changed by 20% by the TD lattice constants. The Lorenz number varies from 1.6 to 3.3$\times$10$^{-8}$ W$\cdot\Omega\cdot$K$^{-2}$ at different temperatures. The theoretical prediction in the literature overpredicts $\kappa_{ph}$ (e.g., $\sim$28%) and underpredicts the $\kappa_{el}$ (e.g., $\sim$38%), resulting in an overall underprediction of $\kappa$ ($\sim$26% at 1500 K). The impacts of grain size and defects are found strong, and no reported experimental data has reached the intrinsic theoretical thermal conductivity of ZrC yet.