(192bi) Addressing Discrepancies in Hydrogen Abstraction By Ooh Radical Via Automatic Transition State Theory Calculations

Combustion of fuels is a ubiquitous process that is prevalent in many aspects of our lives, and understanding combustion allows us to smartly design new fuels and better engines. Detailed kinetic modeling is an imperative tool to thoroughly understand combustion, and the construction of these models requires the determination of thousands of kinetic and thermodynamic parameters through experimentation, electronic structure calculations, or estimations. For combustion models, reactions are initialized by the abstraction of a hydrogen by one of 6 radical molecules (HO·, H·, HOO·, O·, H₃C· and O₂) making hydrogen abstraction reactions key for combustion. For the scope of this project, we chose to focus on hydrogen abstraction via hydroperoxyl radical (HOO·), and address these discrepancies.

Given that a variety of methods exist to calculate or estimate reaction kinetics, discrepancies often arise and need to be addressed. Two common methods currently exist to determine reaction kinetics: rate rules, and transition state theory (TST) calculations. Rate rules can determine reaction kinetics quickly, but they can be misapplied, causing incorrectly calculated kinetics. On the other hand, TST calculations rely on quantum chemistry calculations to determine kinetics. These calculations are often more accurate than rate rules, but they require large amounts of computation time, a good initial guess for the transition state geometry, and human input.

To eliminate the need for human input, we have developed an Automated TST method (AutoTST) [1] a module included in the Reaction Mechanism Generator (RMG) software [2] which automatically estimates transition state geometries using a group contribution method. AutoTST, utilized in this work, optimizes the transition state using a quantum chemistry package such as Gaussian [3] and is integrated with a canonical transition state calculator, CanTherm [4].

From 72 detailed kinetic models collected from recent literature, 331 reactions were identified as possible hydrogen abstraction reactions via HOO· radical. Using algorithms in RMG, the reactions that successfully matched the bimolecular hydrogen abstraction reaction template were fed through the AutoTST method. At the time of wring, 228 reaction kinetics have been calculated. Literature values for these reaction rates were then compared against AutoTST calculated rates to validate and observe discrepancies in existing models and the AutoTST algorithm. Future works include calculation of the existing 447 identified HOO· reactions, continued validation of existing models, and refinement of the AutoTST method.

[1] P. L. Bhoorasingh and R. H. West, Transition state geometry prediction using molecular group contributions, Phys. Chem. Chem. Phys. 17 (2015) 32173–32182.

[2] C. W. Gao, J. W. Allen, W. H. Green, and R. H. West, Reaction Mechanism Generator: Automatic construction of chemical kinetic mechanisms, Comput. Phys. Commun. 203 (2016) 212 - 225. DOI: 10.1016/j.cpc.2016.02.013.

[3] M. J. Frisch et al., Gaussian 09, Gaussian, Inc., Wallingford, CT, 2009.

[4] J. W. Allen and W. H. Green, CanTherm: Open-source software for thermodynamics and kinetics, Included in: Reaction Mechanism Generator, v2.0.0. 2016, URL: http://reactionmechanismgenerator.github.io.