Structural Correlations
Bond orders and bond lengthsThroughout the preceding chapters of this book we have reported and in some cases discussed the lengths of the multiple bonds between metal atoms. We have reported in tabular form most of the M-M distances that have been determined crystallographically. It is now time to make some general comments on these distances in relation to our understanding of the M-M bonds.It is a general qualitative rule in chemistry that bond lengths and bond orders are inversely related. In some limited areas, particularly with the fi rst-row elements carbon, nitrogen, and oxygen, quantitative relationships expressing bond length as a single-valued function of bond order are well known. However, these quantitative relationships are based heavily on faith, as well as fact. Bond lengths are facts; bond orders are not. In the process of inferring a bond order from a bond length one is often getting out no more than what was put in to begin with.We shall not further digress into a discussion of the philosophical ambiguities of these quantitative bond order-bond length correlations because they have proven to be useful, within their own sphere of application. However, there is no a priori reason to expect that similar procedures will (or will not!) work in the very different realm of metal-to-metal bonds. Experience is the only test, and experience thus far has shown that M-M bonds cannot usefully be treated in such a way.In the realm of M-M bonds it is best to invoke only a qualitative inverse relationship between bond order and bond length and to defi ne bond order only in a qualitative or ordinal way. In this book we have used the term M-M bond order only to indicate how many electron pairs are believed, on the basis of essentially qualitative considerations, to play a signifi cant part in holding the pair of metal atoms together. We condemn as foolish and hopeless any effort to associate a unique, precise, quantitative bond order with each and every M-M internuclear distance.In principle, M-M bond lengths, like others, should be amenable to treatment by the method of molecular mechanics, and some efforts [1][2][3][4] (hampered by a dearth of necessary experimental data) to do this have been reported. In general the results are encouraging and can account for