Back to Table of Contents

45664-AC5
Alkane C-H Bond Breaking at Catalytic Metal Surfaces: Theory Development Coupled with Effusive Molecular Beam Experiments

Ian Harrison, University of Virginia

The focus of this work has been to assess the range of applicability of a microcanonical unimolecular rate theory (MURT) model for activated dissociative chemisorption and to extract reactive transition state characteristics from the analysis of varied thermal equilibrium, nonequilibrium, and quantum-state resolved experiments. Earlier work had concentrated on the analysis and simulation of CH₄ and SiH₄ dissociative chemisorption on several surfaces because this reactive step can limit the rates of natural gas reforming and the chemical vapor deposition of silicon. Over the course of this project, we first demonstrated that (i) the dissociative chemisorption and associative desorption dynamics of H₂ on Cu(111) and (ii) the CO₂ dissociative chemisorption and CO oxidation dynamics on Rh(111) can be well modeled with the MURT, even at quantum state resolved levels of detail. For both systems, treating rotation as a spectator to the reaction dynamics was a good approximation at thermally accessible energies. The surface plays a crucial role in the activated dissociative chemisorption of H₂ and CO₂ under thermal equilibrium conditions where phonons are calculated to contribute 30-40% of the energy required to surmount the activation barrier for dissociation. MURT analysis of experiments indicates that the threshold energy for activated dissociative chemisorption of H₂ on Cu(111) is E₀ = 62 kJ/mol, whereas E₀ = 73 kJ/mol for CO₂ on Rh(111). Analysis of the CO₂ product angular yields and translational and rovibrational energy distributions from CO oxidation on Rh(111) indicate that an indirect reaction mechanism dominates under reactive molecular beam conditions. It is likely that a thermalized intermediate involving chemisorbed O (e.g., a carbonate species) facilitates the indirect CO oxidation pathway. The MURT was also used to calculate thermal dissociative sticking coefficients, S_T, for CH₄ on Ir(111), Pt(111), Ru(0001), and Ni(100) based on analysis of non-equilibrium supersonic molecular beam experiments. Interestingly, the calculated S_T values for the flat single crystal surfaces were 2-4 orders of magnitudes greater than values experimentally derived for the analogous supported nanometal catalysts used for methane reforming. This is a surprising result because methane reforming rates are structure sensitive and increase by a factor of ~5 as the nanocatalyst diameter shrinks to ~2 nm. The high curvature nanocatalysts should more abundantly expose the surface steps that are often believed to be more catalytically active than the terrace sites that predominate on the low index flat metal surfaces. In this instance, it seems that the surface science experiments show that the surface metal atoms on the working nanocatalysts are mostly inactive (covered with C?) and relatively few sites are able to turnover. Moving to larger molecules, we concentrated on measuring dissociative sticking coefficients, S(T_g, T_s), for alkanes on Pt(111) as a function of the impinging gas temperature, T_g, and surface temperature, T_s, using an effusive molecular beam. Methane, ethane, and propane studies were completed and experimental procedures were improved. Dissociative chemisorption is activated for all these alkanes but the apparent efficacy to promote reaction for energy in the molecular degrees of freedom monotonically declines as the alkane complexity increases. Hence, energy transfer to the surface increases with molecular complexity (n.b., thermalized trapping mediated chemisorption was not observed). Theoretical analysis of the new propane results is on-going but the experiments clearly point to the need for detailed experimental studies of the larger alkanes where the reactive chemistry is intimately coupled to gas-surface energy transfer.

Back to top