ReacNetGenerator: an automatic reaction network generator for reactive molecular dynamics simulations

Area of Science:

• Computational Chemistry
• Chemical Kinetics
• Molecular Dynamics Simulations

Background:

• Reactive molecular dynamics (MD) simulations enable atomic-level study of complex reaction mechanisms.
• Analyzing large-scale MD trajectories with numerous species and pathways presents a significant challenge.

Purpose of the Study:

• To develop an automated method for extracting reaction networks from MD simulation trajectories.
• To overcome the limitations of manual analysis in large-scale reactive MD simulations.

Main Methods:

• Developed the Reaction Network Generator (ReacNetGenerator) for automatic reaction network extraction.
• Species identification from atomic Cartesian coordinates.
• Utilized Hidden Markov Model (HMM) for trajectory noise filtering to improve accuracy.

Main Results:

• Successfully applied ReacNetGenerator to analyze methane and RP-3 surrogate fuel combustion MD trajectories.
• Demonstrated significant advantages in efficiency and accuracy over traditional manual analysis methods.
• Automated extraction of reaction networks without predefined reaction coordinates or elementary steps.

Conclusions:

• ReacNetGenerator effectively automates the analysis of complex reaction networks from reactive MD simulations.
• The method significantly improves efficiency and accuracy for large-scale chemical system studies.
• Enables broader application of reactive MD simulations by simplifying trajectory analysis.