(504f) Balancing Reactivity and Stability in Metal Nanoparticle and Alumina Support Systems Via Redox Reactions: A Multiscale Computational and Experimental Approach

Pt nanoparticles (NPs) supported on γ-Al₂O₃ (Pt/γ-Al₂O₃) serve in catalytic applications ranging from methanol reformation to CO oxidation, benefiting from the large surface area and strong metal-support interaction of γ-Al₂O₃ capable of stabilizing NPs. Additionally, Pt/γ-Al₂O₃ catalyst systems benefit from the high catalytic reactivity of small, nanoscale metal Pt NPs, the high surface-to-volume ratios of which enable many available sites for catalytic reactions and bonding to supports. However, under the reaction conditions – such as high (>600 °C) temperatures or highly oxidizing environments – frequently employed in industrial catalytic applications, Pt NPs are affected by issues such as poor long-term thermal stability and catalytic deactivation from Pt oxide formation. Solutions attempting to address these issues, such as the introduction of a reducing H₂ environment to remove Pt oxides, can change Pt NP morphology and reduce the effective surface area of the Pt/γ-Al₂O₃ interface, weakening the metal-support interaction and further destabilizing the catalyst system. In contrast, the addition of lower pressure oxidative pretreatments of O₂ or H₂O to Pt/γ-Al₂O₃ systems can further stabilize Pt/γ-Al₂O₃ via Pt-O and Pt-OH interfacial bonding. Thus, for Pt/γ-Al₂O₃ systems, catalysts can be rationally designed to be both stable and reactive by balancing the presence of reducing and oxidizing environments.

Though both O vacancies on oxide supports and metal-metal or metal-oxide bonding can stabilize metal-support catalyst systems, comparisons between experimental and computational Energy Electron Loss Spectroscopy (EELS) results show only metal-metal and metal-oxide bonding are found to be stable at the Pt/γ-Al₂O₃ interface. Therefore, computational models will seek to find temperature dependent conditions under which O₂ and H₂ environments can stabilize Pt NPs without forming metal oxide layers on them, as well as determining whether differently composed Pd and Rh NPs observe the same redox driven trade-off. In this study, energetic and structural data achieved using Density Functional Theory (DFT) modeling of M/γ-Al₂O₃ (M = Pt, Pd, Rh) systems will be applied to a multiscale Reactive Force Field (RFF) optimization approach, which will parameterize RFFs capable of performing Molecular Dynamics (MD) simulations. RFF MD simulations will then be completed under different temperatures, H₂/O₂ coverages, metal NP sizes, and other considerations to determine the redox reaction conditions under which the smallest possible metal NPs can retain catalytic activity while staying stabilized on γ-Al₂O₃ supports.