Sulfur reaction with anionic sites

While the effect of cation impurities on the surface chemistry of MgO has been investigated in detail, very little is known about anion substitution. Defect formation and excitation energies for S and Se -doped bulk MgO have been calculated [182,183] but there are no data for the surface. In the bulk it has been estimated that the presence of S or Se impurities result in a outward relaxation of the Mg neighbors of 6% and 8%, respectively [182]. A recent report of the 0 "-S exchange reaction on MgO has been reported [184]. The reaction involves adsorption of CS2 on MgO powders and the subsequent exchange reaction with formation of COS and of S ions probably located at the low coordinated sites. It has been found that the basicity of the MgO surface doped with sulfur ions is drastically modified with respect to that of pure MgO [184]. 

Taking into account the fact that the solvation of ambident anions in the activated complex may differ considerably from that of the free anion, another explanation for the solvent effect on orientation, based on the concept of hard and soft acids and bases (HSAB) [275] (see also Section 3.3.2), seems preferable [366]. In ambident anions, the less electronegative and more polarizable donor atom is usually the softer base, whereas the more electronegative atom is a hard Lewis base. Thus, in enolate ions, the oxygen atom is hard and the carbon atom is soft, in the thiocyanate ion the nitrogen atom is hard and the sulfur atom is soft, etc. The mode of reaction can be predicted from the hardness or softness of the electrophile. In protic solvents, the two nucleophilic sites in the ambident anion must interact with two electrophiles, the protic solvent and the substrate RX, of which the protic solvent is a hard and RX a soft acid. Therefore, in protic solvents it is to be expected that the softer of the two nucleophilic atoms (C versus O, N versus O, S versus N) should react with the softer acid RX. 

Correlation between the catalytic activity in pyridine HY and relative reducibilities of the Ir-Mo sulfide catalysts is demonstrated in Fig. 3. Similar dependence was also obtained for the rate constants of thiophene HDS, as recently shown elsewhere [11]. These relations suggest that the activity of the mbced catalysts is closely related to the amount of hydrogen consumed during the TPR and, obviously, to the number of reducible sulfur surface species. This is in accordance with a generally accepted idea of the formation of anionic sulfur vacancies on the M0S2 phase, which are assumed to be the catalytic sites in the HDS and hydrogenation reactions. 

---