EUVIMEDEUVIMED entdecken
Crossref

Classical-to-DFT multi-fidelity approach for developing machine learning potentials of aqueous systems

Md Abu Bakar Siddique iD, Kendra Coronado iD, Mohammad M Badhon iD, Mirza Galib iD

The Journal of Chemical Physics · 21.11.2026 · 165 (19)

DOI: 10.1063/5.0341814 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 16.11.2026. Verfasst von Md Abu Bakar Siddique und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry, Howard University, Washington, DC 20059, USA. Die Arbeit nennt 65 Literaturangaben. Gefördert von: Office of Science, Office of Advanced Cyberinfrastructure, National Energy Research Scientific Computing Center.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Machine learning interatomic potentials (MLIPs) offer near–ab initio accuracy at a fraction of the cost of conventional electronic structure calculations, but their reliability depends critically on the quality and diversity of training data. Here, we present a classical-to-density functional theory (DFT) multi-fidelity workflow for efficient dataset generation, in which long classical molecular dynamics trajectories are supplemented with selectively DFT-optimized configurations to train neural network potentials. We demonstrate this approach on four systems: bulk water, water at the air–water interface, CaSO4 in bulk water, and CaSO4 at the interface. In all cases, the trained models faithfully reproduce reference DFT properties, validating the workflow across both simple and chemically complex environments. We further probe the extrapolation capabilities of the models, showing that bulk-trained networks can generalize to different temperatures, pressures, and interfacial geometries when datasets include targeted augmented configurations. This multi-fidelity strategy thus enables accurate and transferable MLIPs for complex solvated and interfacial systems, bridging the gap between computational efficiency and first-principles accuracy.

Autor:innen und Institutionen

  • Department of Chemistry, Howard University, Washington, DC 20059, USA.

ORCID-Profile: Md Abu Bakar Siddique, Kendra Coronado, Mohammad M Badhon, Mirza Galib

Förderung

  • Office of Science
  • Office of Advanced Cyberinfrastructure
  • National Energy Research Scientific Computing Center

Zitieren

APA

Siddique, M. A. B., Coronado, K., Badhon, M. M., & Galib, M. (2026). Classical-to-DFT multi-fidelity approach for developing machine learning potentials of aqueous systems. The Journal of Chemical Physics, 165(19). https://doi.org/10.1063/5.0341814

Vancouver

Siddique MAB, Coronado K, Badhon MM, Galib M. Classical-to-DFT multi-fidelity approach for developing machine learning potentials of aqueous systems. The Journal of Chemical Physics. 2026;165(19). doi:10.1063/5.0341814

BibTeX
@article{Siddique2026,
  author = {Siddique, Md Abu Bakar and Coronado, Kendra and Badhon, Mohammad M and Galib, Mirza},
  title = {Classical-to-DFT multi-fidelity approach for developing machine learning potentials of aqueous systems},
  journal = {The Journal of Chemical Physics},
  year = {2026},
  volume = {165},
  number = {19},
  doi = {10.1063/5.0341814}
}
Kostenlos starten →