Journal of Molecular Catalysis A-Chemical, Vol.163, No.1-2, 189-204, 2000
Catalytic reaction rates in thermodynamically non-ideal systems
Chemical reactions reflect the universal tendency of systems to approach equilibrium. The dynamics towards equilibrium, reflected in rates of chemical reactions, are therefore influenced only by thermodynamic properties, such as reaction affinity and the chemical potential, activity, or fugacity of reactants and products. Reaction rates depend on concentrations only in ideal reaction mixtures, because here, concentration appears in the defining equations for all relevant thermodynamic properties. Catalytic reactions in gas-liquid-solid systems involve molecules solvated in a non-ideal environment and reacting on surfaces. Transition state treatments show that such reactions on surfaces detect the presence and identity of a liquid phase only when a liquid solvates kinetically relevant adsorbed intermediates and activated complexes or when its presence prevents gas-liquid equilibrium by imposing transport restrictions. Chemisorption energies are much larger than typical intermolecular interactions in liquids; therefore, inert liquids rarely influence the structure or reactivity of chemisorbed reactants and activated complexes. However, solvent effects become possible on heterogeneous catalysts when adsorption or desorption steps are rate-determining or kinetically relevant. Here, the reaction coordinate involves molecules in solution, and the corresponding activated complexes can become solvated by the surrounding fluid phase. In the special case of identical solvation of a reactant and an activated complex, a fortuitous cancellation of activity coefficients leads to reaction rates that depend on concentration rather than the thermodynamic activity of reactants. This stringent requirement makes concentration-driven reaction rates unusual exceptions to the general case of chemical reaction rates that depend on the thermodynamic activity of reactants and products. We have used transition state treatments of reaction rates in non-ideal systems to explain observed solvent effects for cyclohexene hydrogenation on Pt and Pd catalysts. A dihydrogen dissociative-adsorption rate-determining step on Pt leads to solvation of the kinetically relevant activated complex. Its activity coefficient and that for dissolved H-2 cancel. As a result, the hydrogenation rate on Pt depends on H-2 concentration in the Liquid phase. On Pd, the rate-determining step involves chemisorbed species that are not influenced by the solvent; the reaction rate depends only on H-2 partial pressure in the gas phase and not on the nature of the liquid. A similar treatment shows that the presence of liquid products in three-phase Fischer-Tropsch synthesis reactors cannot increase the rate of olefin readsorption, unless the liquid introduces transport restrictions that prevent rapid removal of olefins from catalyst pores. Higher solubility of larger olefins cannot account for enhanced readsorption. In fact, increasing solubility either has no effect or, under certain circumstances, increases the propensity for desorption rather than readsorption. Finally, we show mechanistic implications of the dependence of paraffin cracking rates on intrazeolite paraffin activities or concentrations. For example, for dependence on the latter, we involve a molecule in a precursor state that is solvated by the environment within zeolite channels to the same extent as a paraffin absorbed within the structure. Since zeolites introduce non-ideality to a reaction system, experimental observations need to be rationalized using the approach we have proposed.
Keywords:catalytic reactions;non-ideal systems;transition state theory;cyclohexene hydrogenation;zeolite catalysis;Fischer-Tropsch synthesis