Abstract
A series of piano-stool complexes of the cymantrene family (cymantrene = Mn(eta(5)-C5H5)(CO)(3), 1) undergoes facile replacement of a carbonyl ligand by P(OPh)(3) when oxidized by one-electron in CH2Cl2/[NBu4][B(C6F5)(4)]. Data on the previously characterized complexes 1, Mn(eta(5)-C5H4NH2)(CO)(3) (3) and Mn(eta(5)-C5Me5)(CO)(3) (6) have been supplemented by cyclic voltammetry (CV) and IR spectroscopy on Mn(eta(5)-C5H4Me)(CO)(3) (2), Mn(eta(5)-C5H4I)(CO)(3) (4), and Mn(eta(5)-C5H4C(O)H)(CO)(3) (5). The substitution rates, determined by digital simulations of CV scans, ranged from 4 M-1 s(-1) for 6 (+) to 3 x 10(5) M-1 s(-1) for 5 (+). In general, a more strongly donating cyclopentadienyl substituent slows down the CO substitution rate. For mono-cyclopentadienyl substituted complexes, the logarithm of k(sub) is shown to increase linearly with either the weighted average of the CO stretching frequencies or the E-1/2 value of the redox process. An exception to this generalization is the amine-substituted complex 3, for which the CO-substitution rate is higher than predicted by its E-1/2 potential. The substitution rate of the pentamethylated Cp complex 6 (+) is slowed by about an order of magnitude owing to steric effects. The efficacy of this method to predict the CO-substitution rate of a cymantrene-tagged molecule was tested with a cymamtrene-derivatized diarylethene complex, 7. The measured P(OPh)(3)-for-CO substitution rate of 3.7 x 10(2) M-1 s(-1) for 7 (+) was very close to that predicted by the E-1/2 value of 7. A ligand electronic parameter, E-L, of 0.62 was determined for the triphenylphosphite ligand. These studies build on the previous CO substitution-rate analyses by Sweigart and others.