Diligating Tripodal Amido-Phosphine Ligands: the Effect of a Proximal Antipodal Early Transition Metal on Phosphine Donor Ability in a Building Block for Heterometallic Complexes

The ligand precursors P(CH2NH-3,5-(CF3)2C6H3)3 (1a), P(CH2NHPh)3 (1b), and P(CH2NH-3,5-Me2C6H3)3 (1c), react
with the reagents Ti(NMe2)4 and tBuNTa(NEt2)3 to generate metal complexes of the type P(CH2NArR)3TiNMe2
(2a−c) and P(CH2NArR)3TaNtBu (3a−c) (where ArR = 3,5-(CF3)2C6H3, Ph, and 3,5-Me2C6H3). Due to ring strain,
the phosphine lone pair cannot chelate and is available to bind a second metal, and this feature can be utilized
to synthesize heterometallic polynuclear complexes. The 31P chemical shifts observed upon complexation of the
early transition metals to the amido donors are large and in the opposite direction expected for the increased
C−P−C bond angles in these complexes; these unusual shifts are due to P−Ti and P−Ta distances that are
significantly shorter than the sum of van der Waals radii. The reaction of 2c with Ni(CO)4 produces at first the
bimetallic complex (CO)3Ni[P(CH2N-3,5-Me2C6H3)3TiNMe2] (4c), which gradually converts to the trimetallic complex
(CO)2Ni[P(CH2N-3,5-Me2C6H3)3TiNMe2]2 (5c). The effect of the complexation of Ti and Ta fragments on the donor
ability of the phosphine ligand was determined by the preparation of the bis-phosphine complexes trans-L2Rh(CO)Cl, (where L = 1a−c, 2a−c, and 3a−c) prepared by the reaction of the appropriate phosphine with [Rh(CO)2(μ-Cl)]2, and a measurement of the resultant CO stretching frequencies. Surprisingly, the complexes with the larger
C−P−C angles are significantly poorer donors. Density functional theory calculations were performed to determine
what factors affect the donor ability of the phosphine and if through-space interactions might play an important role
in the observed electronic properties.