Assessment of empirical interatomic potential to predict thermal conductivity in ThO₂ and UO₂

Abstract

Computing vibrational properties of crystals in the presence of complex defects often necessitates the use of (semi-)empirical potentials, which are typically not well characterized for perfect crystals. Here we explore the efficacy of a commonly used embedded-atomempirical interatomic potential for the U x Th1−x O2 system, to compute phonon dispersion, lifetime, and branch specific thermal conductivity. Our approach for ThO2 involves using lattice dynamics and the linearized Boltzmann transport equation to calculate phonon transport properties based on second and third order force constants derived from the empirical potential and from first-principles calculations. For UO2, to circumvent the accuracy issues associated with first-principles treatments of strong electronic correlations, we compare results derived from the empirical interatomic potential to previous experimental results. It is found that the empirical potential can reasonably capture the dispersion of acoustic branches, but exhibits significant discrepancies for the optical branches, leading to overestimation of phonon lifetime and thermal conductivity. The branch specific conductivity also differs significantly with either first-principles based results (ThO2) or experimental measurements (UO2). These findings suggest that the empirical potential needs to be further optimized for robust prediction of thermal conductivity both in perfect crystals and in the presence of complex defects.

Improving empirical interatomic potentials for predicting thermophysical properties by using an irreducible derivatives approach: The case of thorium dioxide

“…The non-analytical correction of the Born charge effect was also included in the phonon calculations to account for LO-TO splitting. For SCAN calculations, the macroscopic dielectric constant and Born effective charges were κ = 4.482, Z ⋆ T h = 5.317, and Z ⋆ O = −2.673; for EIP calculations, the particle charges in EIP were taken as Born effective charges (Z ⋆ T h = q T h and Z ⋆ O = q O ), and the dielectric constants were fitted to reproduce the phonon frequency of the highest longitudinal optical mode (denoted LO2 [38]) at the Γ point. The fitted dielectric constants were ǫ = 0.86 for the CRG and ǫ = 0.90 for the PW, respectively.…”

“…Multiple studies have applied MD to simulate thermal transport of pure ThO 2 and ThO 2 with various defects [34][35][36][37], by using the CRG potential, an EIP developed by Cooper et al [20], fitted to experimental measurements. Recently, Jin et al [38] carefully assessed the accuracy of phonon property predictions using DFT-BTE and the CRG potential, showing that the CRG exhibits disagreement with the experimental phonon spectra of optical branches as well as with thermal conductivity. By training a new EIP for ThO 2 and validating the EIP through comparison to phonon properties, we aim to both demonstrate our EIP training approach using IDs, and to support future MD studies on thermal transport in ThO 2 .…”

Thermal Energy Transport in Oxide Nuclear Fuel

“…The next level of approximation is directly computing the cubic phonon interactions from first principles, computing the phonon lifetimes from perturbation theory, and then using the RTA. This has been executed for ThO2 using DFT [151,152] and UO2 using DFT+U [27], where the latter demonstrated rather substantial deviation from experiment. The next level of approximation goes beyond the RTA and fully solves the BTE, which has been executed for ThO2 [153,154].…”

“…Recently, the CRG [923] potential was rigorously accessed for accurate predictions of phonon dispersion, lifetime, and branch specific conductivity for UO2 and ThO2 [152]. It was found that the CRG potential captures the dispersion of the acoustic branches well, but exhibits significant discrepancies for the optical branches, leading to an overestimate of phonon lifetime and conductivity for both materials.…”

“…632 Recently, the CRG potential was rigorously assessed for accurate predictions of phonon dispersion, lifetime, and branchspecific conductivity for UO 2 and ThO 2 . 152 It was found that the CRG potential captures the dispersion of the acoustic branches well but exhibits significant discrepancies for the optical branches, leading to an overestimate of the phonon lifetime and conductivity for both materials. Similarly, it has been found that several UO 2 potentials accurately capture the dispersion of the acoustic branches but that inaccuracies in the optical branches result in a poor description of thermal properties.…”