Are Solvent and Dispersion Effects Crucial in Olefin Polymerization DFT Calculations? Some Insights from Propylene Coordination and Insertion Reactions with Group 3 and 4 Metallocenes

The primary insertion (or 1,2-insertion) of propylene nto (C5Me5)(2)YCH2CH2CH(Me)(2,) as well as the primary and secondary (or 2,1) insertions of propylene into the activated ansa-zirconocene comples [{Ph(H)C-(3,6-tBu(2)Flu)(3-tBu-5-Me-C5H2)}ZrMe](+) were calculated with several DFT methods to find the most adequate methodology for the computation of metallocene-catalyzed olefin polymerization reactions. For the yttrium system, both solvent corrections and dispersion corrections are needed to determine energies of coordination and activation barriers in agreement with experimental data. Dispersion corrections were included directly via the use of specific functionals like B97D and M06 or were added as empirical corrections (GD3BJ) to the B3PW91 calculations. For the zirconocene system, the best method is a combination of B3PW91 with solvent corrections incorporated with SMD continuum model. The dispersion corrections, included via both GD3BJ and M06, tend to overestimate the stabilization of the adducts because of the high steric bulk of the zirconocene system. The addition of dispersion corrections shifts the energy profiles toward lower values but does not affect the relative activation barriers. Implementation of enetropy corrections counterbalances almost perfectly the dispersion corrections. The same observations arise from the study of the C-H activations of propylene induced by the zirconocene complex.