ChemTraYzer - Reaction Models from Molecular Dynamics Simulations


Leonhard, Kai


Kai Leonhard

Junior professor Model-Based Fuel Design


+49 241 80 98174



With the Cluster of Excellence Tailor-Made Fuels from Biomass, the combustion chemistry of unconventional fuels needs to be modeled for designing and optimizing internal combustion engines. Models for describing the combustion chemistry of a fuel often comprise hundreds of species and thousands of reactions. Each of these reactions requires several parameters for describing their kinetics. Parameters of reaction models can be obtained via experiments or computed from first-principles “Ab initio combustion kinetics”, for example quantum mechanics (QM). The high effort required for both of these approaches, however, limits them to the study of few elementary reactions.

  The Figure shows a methane oxidation mechanism Copyright: American Chemical Society

At the cost of accuracy, classical trajectories simulated with a reactive force field allow for studying the chemistry of different species at the same time, while being many orders of magnitudes faster compared to QM. Once the reaction model of a chemical process has been derived from reactive molecular dynamics simulations, important reactions can be identified and updated with QM calculations. This methodology is equally applicable for studying complex reaction networks of polymerization or catalysis processes.

The Figure shows a methane oxidation mechanism obtained using the presented methodology and is reprinted with permission from J. Chem. Theory Comput. 11 (2015), 2517-2524. Copyright 2015 American Chemical Society.

We provide a methodology for obtaining a quantitative model for complex chemical reaction networks which allows to identify, quantify, and evaluate a multitude of reactions from a single reactive molecular dynamics simulation. For a detailed description, please see Döntgen et al. [1]. The Chemical Trajectory AnalYzer (ChemTraYzer) software package is ready to use for post-processing and analyzing ReaxFF [2] trajectories which have been simulated with the LAMMPS software package [3]. The code is accessible as open software under the MIT license via the download link below. A guide to installation and usage is given in the 'readme' file.

Project on

[1] Döntgen et al., J. Chem. Theory Comput. 11 (2015), 2517-2524
[2] van Duin et al., J. Phys. Chem. A 105 (2001), 9396-9409
[3] Plimpton, J. Comp. Phys. 117 (1995), 1-19

Funding: German Research Association, Grant GSC 111