The performance of heterogeneous catalysts featuring metal nanoclusters on oxide supports can be significantly impacted by interactions between the metal and the support. Charge transfer between the metal and the support can alter the oxidation state of the metal, which in turn will change the activity of reaction sites on the metal cluster. Charge transfer can also influence the metal’s strength of adhesion to the surface, which will affect metal sintering rates and cluster size distributions. Thus, identifying physical descriptors for predicting the extent of these interactions will enable researchers to tune the overall activity, selectivity, and stability of many common catalytic materials. The underlying phenomena that control these interactions arise from the physical properties of the metal and the support, as well as from the properties of surface-adsorbed species originating from the reaction environment. In this work, we identify physical descriptors for predicting the interplay of these properties by employing density functional theory (DFT) together with screening tools derived from statistical learning (SL). We use DFT to generate adsorption energy and charge transfer data for a range of metal-support pairs in the presence of adsorbates derived from typical catalytic reaction environments. Using this data, we apply SL to scan the descriptor feature-space for the properties of the metal, support, and adsorbate that most significantly influence the extent of the metal-support interaction (i.e., adhesion strength and charge transfer). This methodology provides predictive models for designing and modifying supports to enhance catalyst performance via metal-support interactions.

Identifying physical descriptors for predicting metal-support interactions in catalysis

2019 Spring National Meeting

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