The catalytic activity of the considered systems is defined by several factors, namely:
(i) The efficiency of detaching oxygen atoms from the sulfur-containing species SOn (n = 1,2,3). The breaking of the S-O bonds may occur at both the substrate and the transition metal cluster. However, the bond-breaking at the substrate is endothermic (and takes about 1.5 eV per bond) while at low-coordinated metal atom of a cluster it is exothermic (with energy gain of about 0.5 eV per bond). This explains why the presence of transition metal clusters is necessary for catalytic activity;
(ii) The ability of the cluster to "clean" itself, i.e., to eliminate oxygen from its surface, in order to regain the catalytically active sites and to continue the process. We found that the clusters of Pd and Pt with the size > 2-3 nm are more efficient in this process (at T > 1,000 K) than the clusters of other TM's considered (Rh, Ir, Ru, and Os);
(iii) The ability of the cluster to keep its size to avoid sintering (that reduces the number of low-coordinated catalytically active sites at the surface of the cluster). We found that the sintering of Rh, Ir, Ru, and Os clusters is significantly suppressed in comparison with the sintering of Pd and Pt clusters of the same size (the individual atoms at the surface of Rh and Ir clusters have a tendency to have higher coordination number, i.e., the detachment of individual atoms from the surface is less likely).
Therefore, the activity of TM nanoparticles is mainly defined by the competing factors (ii) and (iii). At the present, we try to find (experimentally and theoretically) the most optimal combination of the structure, size, and composition of TM nanoparticles, for which the catalytic activity of sulfuric acid decomposition will be the highest.