Structure and function of real catalysts

Abstract

This paper addresses issues such as the relation between the physical and chemical nature of active centers, activation of reactant molecules, mechanisms of catalytic reactions and distribution of promoters in real catalysts which are invariably multicomponent multiphase systems. Interactions among the catalyst components and phases often result in the formation of compounds and defects that do not exist in the separate components. Although such interactions give rise to a great variety of morphologic, chemical, and electronic properties of the real catalysts, they can be controlled, mostly by careful preparation and doping procedures, to the degree that the catalyst displays reproducible activity and selectivity. Most of the examples given in this paper are based on the results from the author's laboratory concerning the activity and selectivity of copper-based catalysts for the hydrogenation of carbon monoxide and carbon dioxide, the water gas shift reaction, and some reactions of amines. Evidence is presented for the interactions between the copper and zinc oxide then analyzed in terms of defect equilibria using models and constants partially known from the literature and adapted for the present system. It is shown on the basis of boundary layer theory for small particles that charge transfer between copper metal particles, also present in the system, and the zinc oxide phase is insignificant. On the other hand, electron equilibria between the solute copper species and the zinc oxide matrix are dictated primarily by intrinsic ionization Cu 0 → Cu + + e - and oxygen vacancy formation. Optical absorption spectroscopy strongly corroborates the theoretical contention that a significant portion of the copper is in electron deficient state, and there is some evidence based on Auger spectroscopy for the presence of Cu + species on the catalyst surface. It is reiterated, as has been proposed before in our earlier work, that these copper species activate substrates such as carbon monoxide or unsaturated hydrocarbons through back-bonding of the copper d-orbitals into the π ∗ orbitals of the substrates. In a paper by D.L. Roberts and G.L. Griffin at this Symposium, additional evidence is presented that the same finely dispersed Cu species are the chemisorption and activation sites for hydrogen. Some significant mechanistic features of carbon monoxide hydrogenation are demonstrated by the enhancement of methanol synthesis rates and carbon-carbon bond formation in the presence of alkali promoters. The nature and concentration of the alkali ions on the catalyst surface determine the outcome of the carbon monoxide hydrogenations in the following way: (i) of all the alkali and alkaline earth promoters, cesium displays the most pronounced effects; (ii) at high temperatures and low hydrogen-to-carbon monoxide ratios, maximum amount of n-propanol and 2-methyl-propanol is observed in the product over the Cs/Cu/ZnO catalysts, consistent with the function of the alkali as base catalysts in aldol condensation of aldehydic or enolic surface intermediates; (iii) at low temperatures and high hydrogen-to-carbon monoxide ratios, cesium enhances methanol synthesis as well as water gas shift rates in water- and CO 2-free synthesis gas, retards the methanol synthesis rate in synthesis gas containing intermediate amounts of water, primarily due to loss of surface area upon cesium doping, and again accelerates the synthesis in water-rich synthesis gas. These latter effects point to a mechanism in which the rate of formation of surface formate is enhanced by cesium in water-free synthesis gas and a rapid removal of surface hydroxyls free sites that activate hydrogen in water-rich synthesis gas. The role of Group VIII metals as promoters of the Cu/ZnO catalysts for low alcohol and hydrocarbon synthesis is represented by the effects of small additions of iron. Product composition is intermediate between that in methanol and Fischer-Tropsch syntheses, with significant amount of low alcohols formed. Characterization of the 1%Fe/Cu/ZnO catalyst by analytical electron microscopy reveals two forms of iron, a fine dispersion in the Cu/ZnO catalyst and metallic particles, suggesting that the alcohol and hydrocarbon products are formed over two different parts of the catalyst. In the concluding remarks it is contended that in many systems the catalytic activity and selectivity is primarily controlled by chemical entities, defects with respect to the pure components, that are induced and stabilized by solid state reactions involving a significant part of the bulk of the catalyst particles. As the various spectroscopic techniques as well as the tools of high resolution and analytical electron microscopy are used in an increasingly refined way for the analysis of real catalysts, it seems inevitable that the traditional art of catalyst preparation will take advantage of the new understanding of catalyst structure and will become a branch of materials science firmly footed in solid state chemistry and physics.