Credit: ©2008 Wiley

Bidentate ligands are ubiquitous in organometallic chemistry. They are popular because when coordinated to a metal a fairly rigid structure is generated, and it is thus easier to understand how small changes in ligand structure affect reactivity. Ligands that bond through phosphorus and nitrogen have been used successfully in catalytic reactions including allylic substitutions, olefin oligomerization and asymmetric hydrogenation.

While investigating the properties of one particular type of P,N-ligand — pyridyl-N-phosphinoimines — Philip Dyer and co-workers from the University of Durham, discovered1 one that unexpectedly existed in equilibrium between two possible structures in solution. The two structures resulted from a tautomerization between an open form of the ligand (phosphinoimine) and a closed form (iminophosphorane), and the position of the equilibrium was strongly dependent on the size of the ortho-substituent of the pyridine ring. This discovery prompted a further investigation of how the reactivity of the ligand would be affected. Reaction with elemental selenium and complexation of rhodium(I) salt both proceeded through the open form, whereas complexation with a hard Lewis acid occurred with the closed form. At this stage the closed form of the ligand had reacted as expected for an iminophosphorane, but when reacted with an electon-deficient alkyne, Dyer and co-workers observed a cycloaddition across a C=C double bond, rather than with the P=N double bond as expected.

Computational studies indicate that there is only a very small energy barrier between the open and closed form of the ligand. Therefore, very small changes in the reagents determine in which of three ways the ligand reacts — as either an iminophosphorane, a P,N-chelate, or a dihydropyridine. A key factor in the observation of this reactivity was the use of bulky substituents on the phosphorus — a methodology that is expected to assist in the observation and isolation of other unusual reactive species.