Automatic Mechanism Generation Using Rmg, Linear Scaling Relationships, and Sensitivity Analyses Applied to the Catalytic Partial Oxidation of Methane

While kinetic parameters for surface reactions can be estimated using DFT calculations, scaling relations, and machine learning algorithms, the construction of microkinetic models requires knowledge of all possible reaction pathways. The recently improved Reaction Mechanism Generator (RMG) [1] can build the complete reaction network from user-defined initial conditions. It can propose elementary reactions and species by applying reaction templates and recipes; estimate thermo-kinetic parameters using built-in databases of DFT calculations, functional-group decision trees, and simple rules of thumb; solve governing rate equations; and decide which reaction pathways should be investigated further.

Using linear scaling relations [2], RMG can estimate adsorbate thermochemistry on a range of hypothetical metal surfaces and generate detailed microkinetic models without a priori assumptions. By running simulations with sensitivity analysis, we can determine the rate limiting step on each metal surface and plot a “volcano surface” for the degree of rate control of each surface reaction as a function of elemental binding energies. We demonstrate this new tool on the catalytic partial oxidation of methane [3].

This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award #0000232253, as part of the Computational Chemical Sciences Program.

[1] Goldsmith, C.F., West, R.H. J. Phys. Chem. C. 21 (18), 9970–9981 (2017)

[2] Abild-Pedersen, F., Greeley, J., Studt, F., Rossmeisl, J., Munter, T. R., Moses, P. G., Skúlason, E., Bligaard, T., Norskov, J. K. Phys. Rev. Lett. 99 (1), 016105, (2017)

[3] Horn, R., K. A. Williams, N. J. Degenstein, A. Bitsch-Larsen, D. Dalle Nogare, S. A. Tupy, and L. D. Schmidt. J. Catal. 249, no. 2 (2007): 380-393.