Long-range electrostatics for machine learning interatomic potentials is easier than we thought
The Journal of Chemical Physics · 14.02.2026 · 164 (6)
DOI: 10.1063/5.0316886 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 12.02.2026. Verfasst von Dongjin Kim und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, UC Berkeley, Berkeley, California 94720, USA. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 101 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, National Institute of General Madical Sciences of the National Institutes of Heath.
Abstract (Originalsprache)
The lack of long-range electrostatics is a key limitation of modern machine learning interatomic potentials (MLIPs), hindering reliable applications to interfaces, charge-transfer reactions, polar and ionic materials, and biomolecules. In this Perspective, we distill two design principles behind the Latent Ewald Summation framework, which can capture long-range interactions, charges, and electrical response just by learning from standard energy and force training data: (i) use a Coulomb functional form with environment-dependent charges to capture electrostatic interactions, and (ii) avoid explicit training on ambiguous density functional theory partial charges. When both principles are satisfied, substantial flexibility remains: essentially any short-range MLIP can be augmented; charge equilibration schemes can be added when desired; dipoles and Born effective charges can be inferred or fine-tuned; and charge/spin-state embeddings or tensorial targets can be further incorporated. We also discuss current limitations and open challenges. Together, these minimal, physics-guided design rules suggest that incorporating long-range electrostatics into MLIPs is simpler and perhaps more broadly applicable than is commonly assumed.
Autor:innen und Institutionen
- Department of Chemistry, UC Berkeley, Berkeley, California 94720, USA.
- Chemical Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA.
- Bakar Institute of Digital Materials for the Planet, UC Berkeley, Berkeley, California 94720, USA.
- The Institute of Science and Technology Austria, Am Campus 1, 3400 Klosterneuburg, Austria.
ORCID-Profile: Dongjin Kim, Bingqing Cheng
Förderung
- NIGMS NIH HHS
- National Institute of General Madical Sciences of the National Institutes of Heath
Zitieren
APA
Kim, D., & Cheng, B. (2026). Long-range electrostatics for machine learning interatomic potentials is easier than we thought. The Journal of Chemical Physics, 164(6). https://doi.org/10.1063/5.0316886
Vancouver
Kim D, Cheng B. Long-range electrostatics for machine learning interatomic potentials is easier than we thought. The Journal of Chemical Physics. 2026;164(6). doi:10.1063/5.0316886
BibTeX
@article{Kim2026,
author = {Kim, Dongjin and Cheng, Bingqing},
title = {Long-range electrostatics for machine learning interatomic potentials is easier than we thought},
journal = {The Journal of Chemical Physics},
year = {2026},
volume = {164},
number = {6},
doi = {10.1063/5.0316886}
}
EUVIMED