Computational Thermodynamic and Kinetic Analysis of Platinum Particle Redispersion in Zeolites

Zeolites, which are microporous aluminosilicate structures, offer an elegant solution to prevent the sintering and dispersion of metal catalysts. However, under different reaction conditions, catalysts may deactivate due to particle restructuring. Recent studies demonstrate that oxidizing conditions may reversibly transform nanoparticles encapsulated in, or on the external surface of, zeolites to smaller particles or single metal cations, but the lack of molecular level knowledge in thermodynamics and kinetics limits exploiting the full potential of interconversion for practical applications. In this study, we computationally analyze the conditions under which the reversible interconversion of Pt-exchanged CHA zeolites is feasible using density functional theory (DFT), first-principle thermodynamic modeling, and kinetic Monte Carlo. Our results establish that Pt metal particles, under oxidizing conditions, compete to convert into Pt-oxides or disintegrate into Pt²⁺ cations that populate aluminum pairs located in the six-membered rings of CHA zeolites. We developed and compared thermodynamic models using supported and unsupported particle energy calculations, both of which predict that the temperature for interconversion increases with particle size and that the addition of water promotes agglomeration and the reduction of the Pt species to Pt metal. Additionally, both models are consistent with the complete interconversion at conditions reported in literature, implicating kinetic effects in differentiating the validity of our two models. Our kinetic model, which describes ion-exchange as Ostwald Ripening with atom-trapping sites, suggests that the unsupported model best represents the system. Our unsupported thermodynamic model for other zeolites is consistent with experimental interconversion observations.