EAGER: Quantifying the Fluxionality of Surface Oxide Supports through the Integration of Global Optimization Algorithms and Multi-Scale Models for Single-Site Catalysts

In this project a novel catalyst design aimed at reducing pollutants from gasoline- and diesel-fueled vehicles will be investigated. This new design has the potential to reduce the amount of expensive and strategic noble metals, such as platinum and rhodium, required to meet the strictest regulatory standards for carbon monoxide (CO) emissions from tailpipes.

Additionally, the accuracy of computational and artificial intelligence (AI) methods used to predict new catalyst designs will be examined. The outcomes of the research will be incorporated into a computational catalysis educational program for undergraduate and graduate students and integrated into two graduate courses currently taught by the principal investigator.

The project will build on the foundation developed during the lead investigator's decade-long work combining Density Functional Theory (DFT) and Machine Learning (ML) methods. This foundation allows for the exploration of the catalytic properties of the “29” oxide structure, a unique copper oxide film that can be stably grown on a Cu(111) crystal, enabling the development of systems that leverage this structure.

This oxide acts as an ideal support for atomically dispersed precious metals, making it highly effective for catalysis. To design and evaluate the performance of atomically dispersed Pt and Rh atoms on the "29" oxide Cu support, the computational work will be conducted in collaboration with experimental partners. The structure and redox properties of the "29" oxide surface on Cu(111) will also be examined by focusing on dynamic changes (fluxionality) as it is reduced to the "44" oxide structure under reaction conditions.

The "29" and "44" refer to the unit cell size of each oxide structure. A key goal of the study is to reconcile previous models of the "29" and "44" oxide surfaces with recent non-contact atomic force microscopy (nc-AFM) images, which show two distinct stoichiometric arrangements for these structures, as reported in a recent publication (Zhu et al., Journal of the American Chemical Society, 146 (2024), 15887–15896).

Simulated X-ray photoelectron spectroscopy, Scanning Tunneling Microscopy, Normal Incidence X-ray Standing Waves, and Surface X-ray Diffraction data will be correlated with experimental results. New models of the "29" and "44" oxide structures will be generated using global optimization algorithms and compared with the nc-AFM data. Additionally, the cooperative effects between the copper oxide surface and isolated Pt or Rh atoms will be modeled by integrating results from the Atomistic Global Optimization X package with first-principles calculations.

This modeling effort will help identify the most stable structures and enable a better understanding of the oxidation/reduction pathways involved in low-temperature CO oxidation with atomically dispersed metals under realistic reaction conditions.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.