EUVIMEDEUVIMED entdecken
Crossref

How accurate are DFT forces? Unexpectedly large uncertainties in molecular datasets

Domantas Kuryla iD, Fabian Berger iD, Gábor Csányi iD, Angelos Michaelides iD

The Journal of Chemical Physics · 14.12.2025 · 163 (22)

DOI: 10.1063/5.0296997 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 10.12.2025. Verfasst von Domantas Kuryla und 3 weiteren Autor:innen. Beteiligte Einrichtung: Yusuf Hamied Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, United Kingdom. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: EPSRC Center for Doctoral Training in Automated Chemical Synthesis Enabled by Digital Molecular Technologies, n-AQUA, Materials Chemistry HEC Consortium.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Training of general-purpose machine learning interatomic potentials (MLIPs) relies on large datasets with properties usually computed with density functional theory (DFT). A prerequisite for accurate MLIPs is that the DFT data are well converged to minimize numerical errors. A possible symptom of errors in DFT force components is nonzero net force. Here, we consider net forces in datasets including SPICE, Transition1x, ANI-1x, ANI-1xbb, AIMNet2, QCML, and OMol25. Several of these datasets suffer from significant nonzero DFT net forces. We also quantify individual force component errors by comparison to recomputed forces using more reliable DFT settings at the same level of theory, and we find significant discrepancies in force components averaging from 1.7 meV/Å in the SPICE dataset to 33.2 meV/Å in the ANI-1x dataset. These findings underscore the importance of well converged DFT data as increasingly accurate MLIP architectures become available.

Autor:innen und Institutionen

  • Yusuf Hamied Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, United Kingdom.
  • Engineering Laboratory, University of Cambridge, Trumpington St. and JJ Thomson Ave., Cambridge, United Kingdom.

ORCID-Profile: Domantas Kuryla, Fabian Berger, Gábor Csányi, Angelos Michaelides

Förderung

  • EPSRC Center for Doctoral Training in Automated Chemical Synthesis Enabled by Digital Molecular Technologies
  • n-AQUA
  • Materials Chemistry HEC Consortium
  • High End Computing Materials Chemistry Consortium
  • Cambridge Service for Data Driven Discovery
  • DiRAC

Zitieren

APA

Kuryla, D., Berger, F., Csányi, G., & Michaelides, A. (2025). How accurate are DFT forces? Unexpectedly large uncertainties in molecular datasets. The Journal of Chemical Physics, 163(22). https://doi.org/10.1063/5.0296997

Vancouver

Kuryla D, Berger F, Csányi G, Michaelides A. How accurate are DFT forces? Unexpectedly large uncertainties in molecular datasets. The Journal of Chemical Physics. 2025;163(22). doi:10.1063/5.0296997

BibTeX
@article{Kuryla2025,
  author = {Kuryla, Domantas and Berger, Fabian and Csányi, Gábor and Michaelides, Angelos},
  title = {How accurate are DFT forces? Unexpectedly large uncertainties in molecular datasets},
  journal = {The Journal of Chemical Physics},
  year = {2025},
  volume = {163},
  number = {22},
  doi = {10.1063/5.0296997}
}
Kostenlos starten →