An Experimental and DFT Workflow for Active Site Characterisation of Single Atom Catalysts using Core-Level Spectroscopy and ORR Microkinetic Modelling

Typically, no single active site dominates an experimental sample, hence representing the active site structure as a distribution of individual sites offers a more realistic interpretation. Accordingly, the primary focus of this work is to present a workflow that identifies active sites and determines their respective distribution and activity within an experimental sample. Here, we demonstrate such a workflow on palladium/nitrogen-doped-graphene single-atom (PdNC) catalysts for the electrochemical oxygen reduction reaction (ORR). The presented active site characterisation approach couples ex situ core-level spectroscopy and microkinetic modelling to relate atomic-scale structures to an experimentally observed sample. The workflow proceeds by screening structures based on their agreement between calculated density functional theory (DFT) and experimental X-ray absorption near edge structure (XANES) spectra. Each acceptable atomic-scale active site is fitted with a Gaussian function, and the peak centre is assigned as the calculated DFT absolute X-ray photoelectron binding energy. The Gaussian parameters are then optimized such that the sum of all calculated DFT peaks reproduces the experimental spectrum of interest. Subsequently, the peak areas are then taken as a ratio of each other, thus representing a distribution of active sites. This workflow reveals that the analysed experimental PdNC material consists of a site distribution dominated by PdN$_4$C$_{10}$ bulk and nitrogen enriched bulk PdN$_4$C$_{10}$ active sites. A microkinetic model then asserts that the calculated distribution remains consistent between core-level spectroscopy measurements and operation under ORR conditions.