Using Diffusion Maps to Analyze Reaction Dynamics for a Hydrogen Combustion Benchmark Dataset

Taehee Ko,Joseph Heindel,Xingyi Guan,Teresa Head-Gordon,David Williams-Young,Chao Yang

2023-04-19

Abstract:We use local diffusion maps to assess the quality of two types of collective variables (CVs) for a recently published hydrogen combustion benchmark dataset~\cite{guan2022benchmark} that contains ab initio molecular dynamics trajectories and normal modes along minimum energy paths. This approach was recently advocated in~\cite{tlldiffmap20} for assessing CVs and analyzing reactions modeled by classical molecular dynamics simulations. We report the effectiveness of this approach to molecular systems modeled by quantum ab initio molecular dynamics. In addition to assessing the quality of CVs, we also use global diffusion maps to perform committor analysis as proposed in~\cite{tlldiffmap20}. We show that the committor function obtained from the global diffusion map allows us to identify transition regions of interest in several hydrogen combustion reaction channels.

Chemical Physics,Computational Physics

What problem does this paper attempt to address?

The paper primarily aims to address the following issues:

1. **Evaluating the quality of Collective Variables (CVs)**: By using the local diffusion map method to assess the effectiveness of two types of collective variables (i.e., internal coordinates and variables obtained from principal component analysis) for a recently published hydrogen combustion benchmark dataset.

2. **Analyzing reaction kinetics**: Utilizing global diffusion maps for committor analysis to identify transition regions in several hydrogen combustion reaction channels.

Specifically, the researchers focus on how to effectively identify collective variables that can describe the main dynamic pathways of molecular system conformational changes and chemical transformations. The background issues mentioned in the paper include:
- For some simple molecular systems (such as alanine dipeptide), intuitive collective variables (such as dihedral angles) perform well;
- For complex chemical or material systems, in condensed phases or involving chemical reactions, identifying relevant collective variables is not obvious;
- Chemical reactions involving the formation and breaking of chemical bonds require quantum mechanical treatment, which can be very time-consuming or computationally expensive, and the obtained collective variables may not be intuitive for complex reaction mechanisms.

The researchers used a recently advocated method—diffusion maps—to evaluate the quality of collective variables and analyze reactions modeled by classical molecular dynamics simulations. Additionally, they demonstrated how to use global diffusion maps to compute committor functions, thereby identifying transition regions in different hydrogen combustion reaction channels.

In summary, the core objective of this paper is to validate the effectiveness of the diffusion map technique in evaluating the quality of collective variables and analyzing reaction kinetics based on quantum mechanics first-principles molecular dynamics simulations.