Unified Differentiable Learning of Electric Response
Abstract
Predicting response of materials to external stimuli is a primary objective of computational materials science. However, current methods are limited to small-scale simulations due to the unfavorable scaling of computational costs. Here, we implement an equivariant machine-learning framework where response properties stem from exact differential relationships between a generalized potential function and applied external fields. Focusing on responses to electric fields, the method predicts electric enthalpy, forces, polarization, Born charges, and polarizability within a unified model enforcing the full set of exact physical constraints, symmetries and conservation laws. Through application to -SiO, we demonstrate that our approach can be used for predicting vibrational and dielectric properties of materials, and for conducting large-scale dynamics under arbitrary electric fields at unprecedented accuracy and scale. We apply our method to ferroelectric BaTiO and capture the temperature-dependence and time evolution of hysteresis, revealing the underlying microscopic mechanisms of nucleation and growth that govern ferroelectric domain switching.
Keywords
Cite
@article{arxiv.2403.17207,
title = {Unified Differentiable Learning of Electric Response},
author = {Stefano Falletta and Andrea Cepellotti and Anders Johansson and Chuin Wei Tan and Albert Musaelian and Cameron J. Owen and Boris Kozinsky},
journal= {arXiv preprint arXiv:2403.17207},
year = {2024}
}
Comments
15 pages, 6 figures