Login / Signup

An efficient algorithm for capturing quantum effects in classical reactive scattering: application to D + H+3 → H 2 D + + H.

Matthew BraunsteinLaurent Bonnet
Published in: Physical chemistry chemical physics : PCCP (2022)
Motivated by a recent semiclassical analysis of chemical reaction thresholds [Bonnet et al. , J. Chem. Phys. , 2022, 157, 094114], we present an efficient algorithm for including zero-point energy (ZPE) effects in classical reactive scattering. The algorithm is an extension of the quasi-classical trajectory (QCT) Gaussian binning method. We apply it to the astrophysically important D + H+3 reaction, where there are significant quantum effects and where application of other methods is problematic [Braunstein et al. , Phys. Chem. Chem. Phys. , 2022, 24 , 5489]. The rate constants computed with the new, general algorithm closely match recent Ring Polymer Molecular Dynamics (RPMD) [Bulut et al. , J. Phys. Chem. A , 2019, 123 , 8766] and experimentally derived [Bowen et al. , J. Chem. Phys. , 2021, 154 , 084307] ones spanning ∼4 orders of magnitude from 70 to 1500 K.
Keyphrases
  • molecular dynamics
  • machine learning
  • deep learning
  • neural network
  • monte carlo