Sintering of Pt Nanoparticles via Volatile PtO2: Simulation and Comparison with Experiments

It is a longstanding question whether sintering of platinum under oxidizing conditions is mediated by surface migration of Pt species or through the gas phase, by PtO2(g). Clearly, a rational approach to avoid sintering requires understanding the underlying mechanism. A basic theory for the simulation of ripening through the vapor phase has been derived by Wynblatt and Gjostein. Recent modeling efforts, however, have focused entirely on surface-mediated ripening. In this work, we explicitly model ripening through PtO2(g) and study how oxygen pressure, temperature, and shape of the particle size distribution affect sintering. On the basis of the available data on α-quartz, adsorption of monomeric Pt species on the support is extremely weak and has therefore not been explicitly simulated, while this may be important for more strongly interacting supports. Our simulations clearly show that ripening through the gas phase is predicted to be relevant. Assuming clean Pt particles, sintering is generally overestimated. This can be remedied by explicitly including oxygen coverage effects that lower both surface free energies and the sticking coefficient of PtO2(g). Additionally, mass-transport limitations in the gas phase may play a role. Using a parameterization that accounts for these effects, we can quantitatively reproduce a number of experiments from the literature, including pressure and temperature dependence. This substantiates the hypothesis of ripening via PtO2(g) as an alternative to surface-mediated ripening.