A Machine Learning Approach for Rate Constants. III. Application to the Cl(²P) + CH₄ → CH₃ + HCl Reaction

Abstract

The temperature dependence of the thermal rate constant for the reaction Cl( 3 P) + CH 4 → HCl + CH 3 is calculated using a Gaussian Process machine learning (ML) approach to train on and predict thermal rate constants over a large temperature range. Following procedures developed in two previous reports, we use a training data set of approximately 40 reaction/potential surface combinations, each of which is used to calculate the corresponding database of rate constant at approximately eight temperatures. For the current application, we train on the entire data set and then predict the temperature dependence of the title reaction employing a "split" data set for correction at low and high temperatures to capture both tunneling and recrossing. The results are an improvement on recent RPMD calculations compared to accurate quantum ones, using the same high-level ab initio potential energy surface. Both tunneling at low temperatures and significant recrossing at high temperatures are observed to influence the rate constants. The recrossing effects, which are not described by TST and even sophisticated tunneling corrections, do appear in experiment at temperatures above around 600 K. The ML results describe these effects and in fact merge at 600 K with RPMD results (which can describe recrossing), and both are close to experiment at the highest experimental temperatures. These results are in accord with a recent high-level experiment−theory study of this reaction.

Neural network potentials for chemistry: concepts, applications and prospects

“…Reactions considered included the Cl + HCl H-atom exchange reaction (in 1d and 3d), the H 2 +OH / H + H 2 O and for O + CH 4 / OH + CH 3 which was investigated more in-depth in a separate study. 225 The results for reactions not used in the learning procedure indicate that it is possible to obtain thermal rates close to those from explicit quantum simulations or trajectory-based quantum calculations (ring polymer MD). 226…”

Predicting Rate Constants of Alkane Cracking Reactions Using Machine Learning

“…Meanwhile, some attention were also paid to directly predict rate constants by machine learning. − Houston et al utilized Gaussian process regression to fit thermal rate constants for a collection of 13 gas-phase bimolecular reactions across a large temperature range. ,, The reactions were characterized by three parameters: Eckart tunneling, skew angle, and reactant symmetric stretch vibrational frequency. The predicted rate constants averaged over the 39 test reactions exhibited an accuracy within 80% of the precise quantum rate constants.…”

Parametrically Managed Activation Function for Fitting a Neural Network Potential with Physical Behavior Enforced by a Low-Dimensional Potential

“…Application of machine learning to chemical and physical problems is a growing field with applications ranging from discovery of new molecules and materials to improving existing theoretical models. − The present article is about using machine learning to represent potential energy surfaces for small molecules and chemical reactions. − …”