Predicting hot electrons free energies from ground-state data. (arXiv:2205.05591v1 [cond-mat.mtrl-sci])

Machine-learning potentials are usually trained on the ground-state,
Born-Oppenheimer energy surface, which depends exclusively on the atomic
positions and not on the simulation temperature. This disregards the effect of
thermally-excited electrons, that is important in metals, and essential to the
description of warm dense matter. An accurate physical description of these
effects requires that the nuclei move on a temperature-dependent electronic
free energy. We propose a method to obtain machine-learning predictions of this
free energy at an arbitrary electron temperature …

arxiv data free state