(378a) Generalized Design Principles for Designing Metal/Support Interfaces with Active Site Specificity

Metal/support interfaces contain active sites with catalytic properties that are distinctive from either the metal or the support considered separately. Recent advances have elucidated the presence of bifunctional interfacial sites for transition metals supported on 2D and 3D carbides, oxides, nitrides, and sulphides. Catalytic properties of these metal/interfaces can be tuned by changing the composition/termination of the support, dopants, or oxygen vacancies. Such interfaces have been designed through chemical because of a lack of generalized design principles. Herein, we present generalized principles for determining active-site-specific reactivity trends for metal/support interfaces.

For a generic metal-support heterostructure, we show that a family of scaling relations succinctly predict reactivity descriptors at every active site located at different proximities to the interface. These scaling relations correlate the binding energy of a descriptor (e.g., CO* and OH*) with the intrinsic stability of the metal active site. This intrinsic stability is, in turn, estimated using coordination based models. Each scaling relation unites active sites of different coordination environments ranging from adatoms to monolayered films. The slopes of these site-specific scaling relations reflect structure sensitivity; namely how strongly reaction energies depend on the local structure of the site. This structure sensitivity is dictated by the charged state of metal sites, such that more (less) negatively charged metal sites are less (more) structure sensitive. We show that these interfacial effects are strongest at metal active sites closest to the interface and having fewest number of nearest neighbours, and become negligible at sites located more than two lattice units from the interface. These design principles are used to engineer metal/support heterostructures for CO₂ reduction and oxygen reduction. The histogram in the figure shows that charge transfer shift over 30 Au-heterostructures into the selectivity window for CO₂ to CH₄, a region exclusively occupied by unsupported Cu-based catalysts.