First-Principles Based Kinetic Monte Carlo Simulation of Catalytic Systems over

Metal Particles.

Eric Dybeck1 and Matthew Neurock1
(1) Department of Chemical Engineering, University of Virginia, Charlottesville, VA
mn4n@virginia.edu
The design of improved catalytic materials and catalytic processes ultimately requires a more complete understanding of the molecular transformations that occur on realistic catalyst surfaces and how specific sites carry out such transformations under working conditions. While microkinetic models allow one to simulate the wealth of elementary physicochemical steps such as adsorption, desorption, diffusion and surface reactions that control the kinetics, they average over catalyst structure and lose atomic scale resolution and the ability to tie catalyst structure to performance. Kinetic Monte Carlo (KMC) simulations, on the other hand, can maintain the atomic surface structure along with the detailed molecular interactions and transformations that occur at different sites thus providing a direct connection between the catalyst structure and composition to catalyst performance.
Previous Kinetic Monte Carlo simulations carried out over extended single-crystal surfaces have resulted in significant insights into the catalytic transformations, but lack the
ability to simulate the important effects of catalyst structure. Recent theoretical and experimental work has demonstrated that conditions on supported metal nanoparticles can differ markedly from those on idealized flat surfaces. Adsorbates on highly covered metal clusters, for example, undergo lateral relaxation through conforming to the curved surface of the particle, and through expansion of the particle lattice. These relaxation methods are inaccessible in systems
constrained to a flat periodic surface. Metal particles also contain coordinatively unsaturated atoms which exhibit stronger binding properties than atoms in the center of extended surfaces.
We have developed a general three-dimensional Kinetic Monte Carlo simulation algorithm which can in principle follow any complex network of elementary catalytic transformations over metal particles under working catalytic temperatures and pressures. More specifically, the simulations track the elementary adsorption, desorption, diffusion and reaction steps over metal nanoparticles, where the different terrace, edge, and corner sites are included in the active site ensemble. The simulation is generic in nature and can be applied to a range of reactions and particle shapes. This simulation is used herein to examine the oxidation of carbon monoxide and to compare with rigorous catalytic kinetic studies over supported platinum particles recently reported in the literature [1]. The elementary activation barriers and entropies were calculated from ab initio Density Functional Theory calculations carried out at relevant coverages. The simulation results compare favorably with experimental rates of reaction over a range of different pressures and temperatures [1]. The simulations also capture appropriate
particle size effects and demonstrate that the dense CO-adlayer formed at low temperatures and high pressures can confer structure insensitivity to this reaction.
References
[1] Allian, Ayman D., Kazuhiro Takanabe, Kyle L. Fujdala, Xianghong Hao, Timothy J. Truex, Juan Cai, Corneliu Buda, Matthew Neurock, and Enrique Iglesia. "Chemisorption of CO and Mechanism of CO Oxidation on Supported Platinum Nanoclusters." Journal of the American Chemical Society (2011): 110302093443000.
Catalysis and Reaction Engineering Division (20)