Isoelectronic cluster electron counting

With this question as a guide it became apparent that delocalization in systems such as benzene and allyl must be examined within a broader family of isoelectronic species. This was done in 1984 by Shaik and Bar102 and by Epiotis,105 and in 1985 it was demonstrated by computational means by Shaik and Hiberty.106 Thus, one can, for example, assemble the isoelectronic A clusters of monovalent atoms with different electron counts, e.g., 3 and 4 in Scheme 7. 

The adoption of two different geometries for these isoelectronic Au6Pt clusters is probably due to electronic differences between the carbon monoxide and acetylide ligands, as neither ligand is sterically demanding in these compounds. However, both cluster compounds adopt metal cage geometries that are best described as hemispherical, as expected from their valence electron counts. 

In this article we use the term isoelectronic to refer to species that contain the same total number of valence electrons. It will be necessary, however, to relax this definition for molecules that contain different sub-stitutents, e.g., Me rather than Et. Otherwise, the number of comparisons possible becomes very small. Likewise, some clusters are compared that are only isoelectronic in the sense of skeletal electron count in order to show the limits of the validity of such a separation. 

The electron counting schemes described for boranes can be extended to isoelec-tronic species such as carboranes,22 clusters containing both carbon and boron as framework atoms. The CH+ unit is isoelectronic with BH many compounds are known in which one or more BH groups have been formally replaced by CH+ (or by C, which has the same number of electrons as BH). For example, the replacement of two BH groups in closo-B6H62 with CH+ yields closo-CjB. ... 

According to normal electron counting rules , these complexes are isoelectronic all have 48 cluster valence electrons (CVE). 

It often happens that predictions of the capacity of a cluster to exchange electrons are made simply on the basis of electron-counting rules, sometimes in conjunction with structural features. The aim of this review was to point out how the redox behavior of homo- and heterometallic clusters, even those which are isoelectronic and isostructural, can often be unexpectedly different. 

We have seen what happens in a borane cluster if there are not enough BH fragments to fill the vertices we get a nido structure with an empty vertex. The same is true for transition metal clusters, for example, in Fe5(CO)i5C, the carbon atom, which is not considered as a vertex atom, is encapsulated within the cluster and gives all its four valence electrons to the cluster. The Fe(CO)3 fragment contributes two cluster electrons as it is isoelectronic with Os(CO)3. The total count is therefore (5 x 2) + 4 = 14, and the number of vertices is -1 = 6. This requires the structure shown as 13.18, as is observed for this and the analogous Ru and Os species. 

---