Trans influence and mutual influence of ligands coordinated to a central atom

Of great interest to inorganic, especially transition-metal, chemists for many years has been the effect on a molecular property when one coordinated ligand is replaced by another. In some cases the result is dramatic if the molecular symmetry is reduced and spectroscopic transitions (e.g., IR or UV-vis absorptions) become allowed or vibrations, degenerate in the parent, split apart in energy. Equally obvious, however, are the observations that MX bond properties (length, vibrational force constant, and magnetic resonance parameters of X) are often very sensitive to the nature and number of other ligands coordinated to the metaL2 For example, the ligand lying trans to X often exerts a large effect in square-planar d8 and octahedral d6 complexes (trans influence), usually much larger than any influence of the cis ligand^.^-^ As the number of ligands L coordinated to a central metal atom increases, the M-L vibrational force constants tend to become ~maller.~ In transition-metal carbonyls a similar effect shows up in thefco constants as the number of coordinated carbonyls increases. Here, trans CO groups contribute a much larger effect than cis CO groups.6 In the carbonyl case there is a larger amount of experimental data relevant to this problem, and a recent study shows the effect to be additive to a remarkable degree.7 In hypervalent molecules* such as ClF3 and ClF, the bonds trans to each other are longer than those trans to a vacancy. In PF, the axial bonds are longer than the equatorial ones. Some of these observations have been rationalized over the years in qualitative terms which contain certain recurring themes. We have recast these ideas as three related points. (i) It is energetically unfavorable for two or more ligands to share the same central-atom orbital for T or u bonding. If it is possible for the geometry to adjust in some manner to share two different orbitals, then this alternative is lower in energy. (ii) If some of the ligands must share the same orbital, then the strongest M-L bonds are for the symmetry-equivalent set of linkages for which the ratio, number of central-atom orbitals shared by these ligands/number of ligands, is largest. (iii) If two different ligands must share the same orbital, then an ill-defined (theoretically) differential bond-weaking process occurs. In this paper, using the angular overlap method, we