(673a) Elucidating Structure-Dependent Activation Energy Trends in Ethane Dehydrogenation

Catalyst deactivation is inevitable as reactions proceed at elevated temperatures; significant economic impacts can be realized by improving catalyst lifetime and reducing the need for regeneration cycles. However, our understanding of the detailed processes underlying deactivation remains limited due to the highly complex nature of catalysts. In particular, the myriad possible arrangements of atoms in a catalyst particle cannot be captured solely by extended surface models. We seek to elucidate the detailed role of catalyst structure in catalyst deactivation, toward the goal of designing catalyst structures with high activity and lifetime. To obtain a deeper understanding of the role of structure, we extend surface-based methods to more realistic models with disordered features representative of nanostructured catalysts under reaction conditions.

Ethane dehydrogenation is one such reaction in which deactivation via coking has been shown to occur at undercoordinated surface sites, particularly on widely studied Pt catalysts. We previously showed relationships between reaction thermochemistry and adsorption site stability, using the adsorption energy of a metal binding site to predict reaction energetics based on rigorous density functional theory (DFT) calculations. In this presentation, we apply DFT to the ethane dehydrogenation network on model transition metal (Pt, Pd, Au) surfaces to obtain generalizable correlations between the activation energies of elementary steps and catalyst structure. Our new results combine the previous model relating molecular adsorption energies and surface stability with well-known Bronsted-Evans-Polanyi relationships to directly link activation energies with surface structure. We thereby show that a complete picture of reaction energetics (thermochemistry and activation energies) can be predicted solely by the coordination environment of an active site, enabling extension to myriad nanoparticle structures with varying features at minimal computational cost. The heterogeneity of surface sites and the interfaces between sites plays a particularly critical role in the design of improved dehydrogenation catalysts.