Thiolate-protected gold nanoclusters, Aum(SR)n have attracted scientific attention as novel functional nanomaterials due to their excellent stability and unique physical and chemical properties that distinguish them from their bulk counterpart. We discuss our results of the Au/TiO2 interface. Strong interactions between Au nanoparticles and Ti species at the edge/corner sites of anatase are favorable for the formation of stable oxygen vacancies; thereby enhancing catalysis. Furthermore, we also explore the structural, electronic and catalytic properties of bimetallic nanoparticles of the form Au25-xAgx(SR)18. Due to the combinatorial enormity of the number of different alloyed structures, we choose random configurations corresponding to each alloying level and energetically optimize their structures. These random configurations are training sets for machine-learning algorithms. Our goal is to understand the correlations between features of the random configurations and reactivity or adsorption. We will report on the properties of the lowest energy structures which enable us to determine the most important features that can be analyzed for training machine-learning algorithms. For instance, nanoalloys with one Ag at the center and the rest in the outer shell of the Au13 kernel are less energetically favorable than the ones with all the Ag atoms occupying the surface of the Au13 kernel. Experimental results showing that catalytic oxidation of CO is adversely affected due to Ag alloying. We provide qualitative and quantitative evidence to explain this reduction of the catalytic activity using Fukui functions and average adsorption energies, respectively.

Machine-learning approaches to rationally designed bimetallic nanoparticles and nanoparticles at the interface

256th National Meeting (2018)

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