Phosphines and related ligands (PE3) are some of the most versatile and well-studied in the chemist’s arsenal. The vast majority of studies on these ligands have focused on their interactions with low valent transition metals. From these investigations, it is now widely accepted that there are two major electronic interactions between the PE3 ligand and a low valent metal: a σ-donor bond between the metal and phosphorus and a π-acceptor interaction, where the phosphorus accepts electron density from the metal. The π-acceptor interaction is a form of negative hyperconjugation, which becomes more significant for electron-withdrawing E groups, such as those in phosphites, P(OR)3. In this study, the bonding between a high valent transition metal, chromium(VI), and a variety of different PE3 ligands was interrogated. The Lewis acidity of the NCr(NiPr2)2+ fragment was quantified using the Gutmann method, and the acceptor number was found to be 100, similar to highly Lewis acidic SbCl5. Using calculations (DFT, Natural Resonance Theory) it was found that rarely discussed resonance forms were important for some phosphines on d0 metal centers. These alternative resonance forms included negative hyperconjugative forms that appear when aryl or heteroatom substituents are placed on the phosphorus atom, leading to increased electron density at the Lewis acidic metal center. Attempts to explore the consequences of the computational results experimentally with the [NCr(NiPr2)2PE3]+X− system were undertaken, however, due to experimental difficulties with ion pairing and solvent effects, experimental quantification of donor ability across a diverse series of phosphines was not possible. However, trialkylphosphines were fit to a simple 2-parameter model, which showed that both steric and electronic effects were extremely important in the system. While we could not reliably quantify the observation, it was noted that aryl and heteroatom substituted PE3, P(OEt)3 and PMePh2, for example, have faster amide rotation at lower temperatures than many of the trialkylphosphines. Despite experimental complications, these studies suggest that bonding to high valent metals could be more complex than simple lone pair donation, and it appears that several resonance forms may be of importance when d0 metals bond to aryl- and heteroatom substituted phosphines.
Read full abstract