Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Xiaoyang, Wang, Yinan, Zhao, Wenbo, Liu, Hanyu, Xie, Hao, Wang, Lei, Wang, Han
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2604.20230
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910156956106752
author Wang, Xiaoyang
Wang, Yinan
Zhao, Wenbo
Liu, Hanyu
Xie, Hao
Wang, Lei
Wang, Han
author_facet Wang, Xiaoyang
Wang, Yinan
Zhao, Wenbo
Liu, Hanyu
Xie, Hao
Wang, Lei
Wang, Han
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_20230
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Crystal structure prediction with nuclear quantum and finite-temperature effects via deep free energy learning
Wang, Xiaoyang
Wang, Yinan
Zhao, Wenbo
Liu, Hanyu
Xie, Hao
Wang, Lei
Wang, Han
Materials Science
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.
title Crystal structure prediction with nuclear quantum and finite-temperature effects via deep free energy learning
topic Materials Science
url https://arxiv.org/abs/2604.20230