General Learning of the Electric Response of Inorganic Materials

TL;DR

This paper introduces MACE-Field, a novel field-aware interatomic potential that accurately predicts dielectric properties and enables finite-field simulations for inorganic materials, enhancing the capabilities of pretrained models.

Contribution

The work develops MACE-Field, a field-aware extension of the MACE model, capable of predicting dielectric properties and performing finite-field simulations with minimal retraining.

Findings

MACE-Field accurately predicts polarisation and Born effective charges.

It reproduces dielectric spectra and hysteresis in inorganic solids.

The model benchmarks well against existing methods like Allegro-pol and DFPT.

Abstract

We present MACE-Field, a field-aware $O(3)$-equivariant interatomic potential that provides a compact, derivative-consistent route to dielectric properties (such as polarisation $\mathbf P$, Born effective charges $Z^*$ and polarisability $\boldsymbol\alpha$) and finite-field simulations across chemistry for inorganic solids. MACE-Field preserves the standard MACE readout and can inherit existing MACE foundation weights, turning pretrained models into field-aware ones with minimal change. To demonstrate, we fine-tune MACE-MP-0 on multiple heads covering BECs and polarisabilities ($\sim$6k MP dielectrics spanning 81 elements), polarisations (2.5k MP nonpolar-to-polar polarisation branches), and energies, forces, and stresses (10,000 structure-replay set from MPtraj), resulting in a field-aware foundation model, MACE-Field-MP-0. We show that MACE-Field can evaluate polarisation branches and spontaneous polarisations, predict $Z^*$ and dielectric constants across diverse chemistries, and reproduce finite-field MD simulations, such as BaTiO$_3$ polarisation hysteresis and the IR/Raman and dielectric spectra of $\alpha$-quartz, benchmarking against Allegro-pol and DFPT.