The Active Surface Species in Alkali-Catalyzed Carbon Gasification: Phenolate (COM) Groups vs Clusters (Particles)

Abstract

The activities and catalytic roles of different species of potassium in the gasification reactions of graphite by H 2O and CO 2 are investigated. TEM techniques are used to measure the rates of monolayer edge recession (uncatalyzed and COK catalyzed) and rates of monolayer channeling (catalyzed by particles) on the basal plane of graphite reacting with 21 Torr H 2O at 700°C. The turnover frequencies for carbon gasification are: 0.08 s −1 (uncatalyzed), 0.15 s −1 (catalyzed by COK groups), and 7.8 s −1 (catalyzed by particles). Thus the particles have a high activity, whereas the COK groups have only a small activity. TGA and literature results using mixtures of carbon and alkali salts show a sigmoidal dependence of gasification rates on catalyst loading. This is the result of catalyst dispersion and competition between the COK groups and alkali particles. A CNDO semi-empirical molecular orbital calculation is performed on model graphite substrates with -O and -OK groups bonded to the {10 1 l} zigzag face and {11 2 l} armchair face. On the zigzag face, the carbon atom bridging two COK groups gains a large negative charge (−0.486) and hence is a favorable site for binding an O atom. The surface CC bonds in this structure are substantially weakened by adding O atoms on the bridging C atoms, leading to CO release. The O atoms are supplied by the dissociation of H 2O and CO 2. The possible reason for the alkali particles being more active than the COM (where M = alkali) groups is that the particles can dissociate H 2O and CO 2 at higher rates, by providing either more active sites or higher activities. The CNDO results also predict that the COK groups have an inhibiting effect on the armchair face; an inhibiting effect has indeed been observed earlier in our laboratory.