Replacing a boron vertex in a [B(n)H(n)](2-) cluster with a smaller atom, e.g. carbon, results in the distance from that atom to the polyhedral centroid being shorter. This forms the basis of a simple but very effective method of distinguishing between B and C atoms in (hetero)carboranes that have been studied crystallographically, the Vertex-to-Centroid Distance (VCD) method. Examples of well-characterised icosahedral and sub-icosahedral closo carboranes, nido carboranes, icosahedral metallacarboranes and supraicosahedral metallacarboranes are used to illustrate the generality of the VCD method. In this process a number of structures are identified in which the method suggests B/C disorder not previously recognised and some structures in which it appears that a cage C atom has been wrongly assigned. The largest sub-group of heterocarboranes is the family of 3,1,2-MC(2)B(9) compounds, and for these species consideration of Exopolyhedral Ligand Orientation (ELO) can sometimes be used to quickly suggest when a literature structure is suspect in terms of cage C atom positioning. The crystal structure of one such compound, 3,3-NO(3)-3-PPh(3)-3,1,2-closo-RhC(2)B(9)H(11), has been redetermined and the correct location of the cage C atoms established.

13-vertex indenyl cobaltacarboranes with 4,1,6-, 4,1,10- and 4,1,2-CoC(2)B(10) architectures have been synthesised by reduction of the corresponding closo carborane and metallation with an {(eta-C(9)H(7))Co} fragment. Variants of the 4,1,6-isomer were prepared with no, one and two methyl groups on cage C atoms, whilst 4,1,2-species were obtained both with two methyl groups and a trimethylene tether on the cage C atoms. Thermolysis of the 4,1,6-isomers yielded the corresponding 4,1,8-isomers, which in turn were converted to 4,1,12-isomers by thermolysis at higher temperatures. Alternatively relatively mild heating of the 4,1,10-isomer led to the 4,1,12-isomer directly. Products were characterised by mass spectrometry, (1)H and (11)B NMR spectroscopies and, in most cases, elemental analysis, and nine compounds were studied crystallographically. The 4,1,6-, 4,1,8-, 4,1,10- and 4,1,12- species have docosahedral cages whilst the 4,1,2-species are henicosahedral. In the structural studies attention focused on the orientation of the indenyl ligand with respect to the carborane ligand since this affords experimental information on the metal-cage bonding through the structural indenyl effect. There is a general tendency for the indenyl ligand to adopt orientations in which the ring junction C atoms lie trans to cage B atoms. In cases where the orientation is not compromised by the presence of a non-H substituent on the face of the carborane there is generally good agreement between the experimental orientation and that computed by DFT calculations for the related naphthalene ferracarboranes (eta-C(10)H(8))FeC(2)B(10)H(12). The presence of C-methyl substituents in the indenyl cobaltacarboranes tends to override this preference except in the case of 1,6-Me(2)-4-(eta-C(9)H(7))-4,1,6-closo-CoC(2)B(10)H(10) where the indenyl ligand instead is forced to incline away from the cage methyl groups. In DCM solution the 4,1,6-, 4,1,8-, 4,1,10- and 4,1,12- isomers of (eta-C(9)H(7))CoC(2)B(10)H(12) exhibit two, stepwise, 1-electron reductions assigned to Co(III)/Co(II)/Co(I) couples at less negative potentials than those of the corresponding Cp compounds. Moreover these reductions are easier for those isomers (4,1,6- and 4,1,10-) in which there are two cage C atoms in the carborane face to which the metal atom is bound. By spectroelectrochemical and EPR measurements it is concluded that the reductions of these indenyl cobaltacarboranes are largely metal-based.

Three isomers of [(Cp*Ru)2 C2 B10 H12 ], the first examples of 14-vertex heteroboranes containing 14-skeletal electron pairs, have been synthesized by the direct electrophilic insertion of a {Cp*Ru(+) } fragment into the anion [4-Cp*-4,1,6-RuC2 B10 H12 ](-) . All three compounds have the same unique polyhedral structure having an approximate Cs symmetry and featuring a four-atom trapezoidal face. X-ray diffraction studies could confidently identify only one of the two cage C atoms in each structure. The other C atom position has been established by a combination of i) best fitting of computed and experimental (11) B and (1) H NMR chemical shifts, and ii) consideration of the lowest computed energy for series of isomers studied by DFT calculations. In all three isomers, one cage C atom occupies a degree-4 vertex on the short parallel edge of the trapezium.

Reduction of the tethered carborane 1,2-μ-(CH(2)SiMe(2)CH(2))-1,2-closo-C(2)B(10)H(10) followed by metallation with {CpCo} or {(p-cymene)Ru} fragments affords both C,C'-dimethyl 4,1,2-MC(2)B(10) and 4,1,6-MC(2)B(10) species. DFT calculations indicate that the barriers to isomerisation of both 4-Cp-4,1,2-closo-CoC(2)B(10)H(12) and 4-(η-C(6)H(6))-4,1,2-closo-RuC(2)B(10)H(12) to their respective 4,1,6-isomers are too high for this to be the origin of the unexpected formation of 4,1,6-MC(2)B(10) products (in marked contrast to the related isomerisation of 1,2-closo-C(2)B(11)H(13) to 1,6-closo-C(2)B(11)H(13)), and, indeed, the 4,1,2-species are recovered unchanged from refluxing toluene. Equally, the DFT-calculated profile for the isomerisation of [7,8-nido-C(2)B(10)H(12)](2-) to [7,9-nido-C(2)B(10)H(12)](2-) suggests that the unexpected formation of 4,1,6-metallacarboranes is unlikely to result from isomerisation of a reduced (nido) carborane following desilylation. Instead, the source of the 4,1,6-MC(2)B(10) compounds is traced to desilylation of 1,2-μ-(CH(2)SiMe(2)CH(2))-1,2-closo-C(2)B(10)H(10) by Li or Na prior to reduction. The supraicosahedral metallacarboranes 1,8-Me(2)-4-Cp-4,1,8-closo-CoC(2)B(10)H(10), 1,12-Me(2)-4-Cp-4,1,12-closo-CoC(2)B(10)H(10) and 1,12-Me(2)-4-(p-cymene)-4,1,12-closo-RuC(2)B(10)H(10) are also reported with all new species characterised both spectroscopically and crystallographically.

No abstract