Heterogeneous catalysts in industrial processes usually take the advantage of loading catalytic species on high specific surface area, which are normally amorphous in nature. The heterogeneity of the support contributes another complexity level to the catalysis system, besides the supported clusters’ fluxionality and the influence of the reactants. Nowadays, it is still a huge challenge to experimentally characterize in-situ amorphous catalysts, and quantitatively study the structure-reactivity relationship. In this lecture, we take two steps from first-principles simulations. First, we explore the supported catalyst structures on amorphous solids by a designed global optimization method. Based on an established amorphous alumina model, grand canonical sampling method is used to explore the structures of a supported Cu4 oxide cluster in both configurational and compositional spaces. In the meanwhile, by considering different spatial regions of the supports, the influence of the support heterogeneity is also considered. By simulating the experimental characterizations, like infrared, X-ray photoelectron spectroscopy and XANES etc., we benchmarked the interpretability of averaged experimental spectra on amorphous catalyst. In the second step, the heterogeneity of amorphous supports is addressed by statistical model. We design a structural descriptor, which links the amorphous surface structures to their energetic stabilities. A descriptor based local stability provides the estimation of their occurrence probability in a real amorphous surface. In the end, we can describe the catalyst from a descriptor based statistics model. In conclusion, from first-principles based structure sampling methods and statistical analysis of amorphous support, we bring a renewed understanding to heterogeneous catalysts in atomistic details and evaluate their macroscopic performance.

Amorphous catalysts from global optimization and statistical analysis

2019 Fall National Meeting

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