Bifunctional catalysts like gold supported on oxides, carbides, and nitrides can activate inert molecules like CO2 and enable their further conversion to C1 species. The catalytic properties of these supported metal catalysts are highly tunable through changes in the support composition, support termination, or by introducing dopants/vacancies. Yet, generalized design principles that establish reactivity and stability trends for supported gold catalysts are still lacking. Using first principles density functional theory, we address two questions. First, how to determine the adsorption energies of catalytic descriptors/probe molecules like CO for generic gold/support heterostructures? Second, can we predict the adhesion energy (a reflection of thermodynamic stability) of supported gold systems? We systematize reactivity trends of Au-support heterostructures using site-specific scaling relations. These scaling relations (accuracies of 0.1 eV) correlate the binding energy of CO* to the intrinsic stability of the metal active site, as the coordination number of the gold site is varied. The active site stabilites can, in principle, be predicted using coordination-based models. The slope of these scaling relations indicates how the binding energy of CO* changes as the nuclearity of the Au site is modified, thus directly reflecting structure sensitivity. Across a wide-range of oxides, carbides, nitrides, and sulfides, we show that the structure sensitivity is inversely proportional to the charged state of the gold site. More positively (negatively) charged gold sites have greater (lower) structure sensitivity. By modulating the charge transfer, we can re-balance the structure sensitivity of supported gold catalysts, thus increasing the density of interfacial sites that contribute to catalytic turnovers. The scaling relations show that gold-support interactions are limited to two gold layers proximal to the support, with active sites located beyond this region being minimally influenced by the support. To determine the stability of gold-support heterostructures, we build a physics-based model that predicts the adhesion energy using physico-chemical properties like the electronegativity, work function of the support, and strain, with accuracies of 0.01 eV/A2. Using the CO2 reduction reaction as a probe reaction, we demonstrate how by tuning metal-support interactions and the nuclearity of gold atoms, we can shift the selectivity from CO/HCOOH to CH4.

Generalized design principles for determining the stability and reactivity of supported gold catalysts

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.