Crystal structure prediction with nuclear quantum and finite-temperature effects via deep free energy learning

TL;DR

This paper introduces a deep learning-based free energy model that efficiently incorporates nuclear quantum and finite-temperature effects into crystal structure prediction, enabling high-throughput and accurate results.

Contribution

It presents a novel deep free energy (DF) model using a two-level learning workflow that significantly reduces computational cost while accounting for complex physical effects.

Findings

Reproduces stability of known high-pressure hydrides.

Discovers a new thermodynamically stable hydride, LaScH8.

Achieves a 1.72 million-fold cost reduction compared to DFT-based methods.

Abstract

Accurate crystal structure prediction (CSP) requires accounting for finite-temperature and nuclear quantum effects, yet first-principles evaluation of the free energy surface (FES) remains prohibitive for high-throughput searches. We observe that the self-consistent harmonic approximation (SCHA) FES, as a function of nuclear centroid positions, shares the same mathematical structure as a potential-energy surface and can therefore be directly learned by a deep neural network potential. The resulting deep free energy (DF) model, constructed via a two-level concurrent-learning workflow, evaluates free energies, forces, and stresses in a single forward pass. Applied to the La-Sc-H system at 200 GPa and 300 K, DF-based CSP reproduces the stability of the experimentally observed LaH10 and LaSc2H24, and discovers an unreported thermodynamically stable clathrate hydride: P4/mmm LaScH8. Benchmarked on the LaH10 system, the DF model achieves a 1.72*10^6-fold cost reduction relative to DFT-level SSCHA. The DF framework provides a scalable route for incorporating finite-temperature and nuclear quantum effects into high-throughput crystal structure prediction.