EUVIMEDEUVIMED entdecken
Crossref

Accurate machine learning of rate coefficients for state-to-state transitions in molecular collisions

Darin E Mihalik iD, R Wang, B H Yang iD, P C Stancil, T J Price iD, R C Forrey iD, N Balakrishnan iD, R V Krems iD

The Journal of Chemical Physics · 14.01.2025 · 162 (2)

DOI: 10.1063/5.0242182 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 14.01.2025. Verfasst von Darin E Mihalik und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Physics and Astronomy and the Center for Simulational Physics, University of Georgia, Athens, Georgia 30602, USA. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: National Aeronautics and Space Administration, National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present an algorithm that combines quantum scattering calculations with probabilistic machine-learning models to predict quantum dynamics rate coefficients for a large number of state-to-state transitions in molecule–molecule collisions much faster than with direct solutions of the Schrödinger equation. By utilizing the predictive power of Gaussian process regression with kernels, optimized to make accurate predictions outside of the input parameter space, the present strategy reduces the computational cost by about 75%, with an accuracy within 5%. Our method uses temperature dependences of rate coefficients for transitions from the isolated states of initial rotational angular momentum j, determined via explicit calculations, to predict the temperature dependences of rate coefficients for other values of j. The approach, demonstrated here for rovibrational transitions of SiO due to thermal collisions with H2, uses different prediction models and is thus adaptive to various time and accuracy requirements. The procedure outlined in this work can be used to extend multiple inelastic molecular collision databases without exponentially large computational resources required for conventional rigorous quantum dynamics calculations.

Autor:innen und Institutionen

  • Department of Physics and Astronomy and the Center for Simulational Physics, University of Georgia, Athens, Georgia 30602, USA.
  • Department of Natural Sciences, SUNY Suffolk, Brentwood, New York 11717, USA.
  • Department of Physics, Penn State University, Berks Campus, Reading, Pennsylvania 19610, USA.
  • Department of Chemistry and Biochemistry, University of Nevada, Las Vegas, Nevada 89154, USA.
  • Department of Chemistry, University of British Columbia, Vancouver, British Columbia V6T 1Z1, Canada.

ORCID-Profile: Darin E Mihalik, B H Yang, T J Price, R C Forrey, N Balakrishnan, R V Krems

Förderung

  • National Aeronautics and Space Administration
  • National Science Foundation

Zitieren

APA

Mihalik, D. E., Wang, R., Yang, B. H., Stancil, P. C., Price, T. J., Forrey, R. C., Balakrishnan, N., & Krems, R. V. (2025). Accurate machine learning of rate coefficients for state-to-state transitions in molecular collisions. The Journal of Chemical Physics, 162(2). https://doi.org/10.1063/5.0242182

Vancouver

Mihalik DE, Wang R, Yang BH, Stancil PC, Price TJ, Forrey RC, et al. Accurate machine learning of rate coefficients for state-to-state transitions in molecular collisions. The Journal of Chemical Physics. 2025;162(2). doi:10.1063/5.0242182

BibTeX
@article{Mihalik2025,
  author = {Mihalik, Darin E and Wang, R and Yang, B H and Stancil, P C and Price, T J and Forrey, R C and Balakrishnan, N and Krems, R V},
  title = {Accurate machine learning of rate coefficients for state-to-state transitions in molecular collisions},
  journal = {The Journal of Chemical Physics},
  year = {2025},
  volume = {162},
  number = {2},
  doi = {10.1063/5.0242182}
}
Kostenlos starten →