ABSTRACT
This has been attempted in various ways which were recently reviewed by Golodets [60]. The most direct method consists of calorimetric measurements of the heat of adsorption <1s of oxygen on the oxide surface. However, the value of % strongly depends on both the coverage and the pretreatment of tne oxide. Sazonov and colleagues [65] determined chemisorption heat of oxygen on the oxides of period IV transition metals as a function of the amount of oxygen removed from the surface. In these experiments the stoichiometric surface of the oxide was at first treated with a small amount of CO and then thoroughly evacuated. From the amount of desorbed C02 the degree of surface deoxygenation was calculated and expressed as 8, the percentage of oxygen monolayer removed. In most of the investigated oxides (C0304, V2°5, NiO, CuO, Mn02) .9s increased with the deoxygenation of the surface. It remained practically constant in the case of Fe203 and in that of ZnO exhibited a small increase after about 0.25% of monolayer of oxygen was removed and then remained constant. It could be shown [60] on the example of nickel oxide that ~ values obtained at 327°C for the surfaces of a given 8 were also dependent on the time which elapsed between prereduction and chemisorption. With the prolongation of this time the 9:s value decreased. This was interpreted as the result of partial regeneration of the surface due to the diffusion of oxygen from the deeper layers of the crystallites. This effect was stronger with increasing deoxygenation but vanished for 8 approaching o. The values of 90s obtained for 8 = 0 reproduced well and could be taken as the measure of oxygen bond strength.
