(265g) Using Reaction Mechanism Generator to Investigate the Catalytic Synthesis of MeOH

Determining and predicting the elementary steps that occur at a catalyst surface is an active area of research, where experimental and computational data are often limited. For complex systems, identifying every species and reaction is difficult and time consuming; determining respective thermodynamic properties and rate expressions is even more challenging. Automatic mechanism generation offers a practical solution by estimating possible species’ structures, thermodynamics, and kinetics. One such tool is Reaction Mechanism Generator (RMG), an open-source software originating from MIT.

This work leverages RMG to (re)create a well-studied system: methanol synthesis from hydrogen (H2), carbon monoxide (CO), and carbon dioxide on Cu(111). RMG constructs detailed models comprised of elementary steps using known data and informed estimates of thermodynamic and kinetic properties. Species’ thermodynamic values are determined by Benson’s Group Additivity method, and linear scaling relations are used for estimating binding energies on different metals. New species and reactions are determined by systematically applying reaction templates, arranged in reaction families (over 50 for the gas phase and almost 20 for surface reactions). Kinetic estimates are derived from rate rules for reactions that fit a similar template. Species are selected and pathways explored using a flux-based algorithm.

In addition to updates to RMG to account for coverage dependent effects, generating the methanol synthesis mechanism required expanding and updating reaction families in RMG’s database. Reaction rates were collected from the literature, and improved with plane-wave density functional theory (DFT) calculations to better determine binding energies and barrier heights. The resulting model is compared to prior computational and experimental studies. The uncertainty and the sensitivities with respect to key species are also evaluated, to examine the reliability of RMG and our model.

This work was supported by the U.S. DOE, Office of Science, BES, CSGB Division, as part of the CCS Program (Award Number: 0000232253).