Mary Chan                                   Kumar Vanka                             Cory Pye
 
A Density Functional Study on Ion-Pair Formation in Group 4 Metallocene
and Related Olefin Polymerization Catalysts.
Mary Chan, Kumar Vanka, Cory Pye and Tom Ziegler,
Department of Chemistry, University of Calgary, Calgary, Alberta Canada T2N 1N4.

 
 
 

In addition to their superior performance over traditional Ziegler-Natta type catalysts, metallocene based single-site homogeneous catalysts feature well defined molecular structures that made theoretical investigations into the mechanistic aspects of polymerization possible. The catalyst precursor has the general formula L2MR2 where M is a transitional metal center, L is a ligand and R is an alkyl group. A Lewis acid (A) is used to abstract an alkyl group from the precursor to produce the active cationic catalyst [L2MR+ ]. The accepted mechanism for monomer insertion requires the initial formation of a p-complex between the cationic catalyst and the olefin. However, a number of different species can bind to this cation in the reaction mixture. The present investigation focuses on the equilibria between various types of complexes and ion-pairs formed by the reaction of the active catalyst with the counterion [RA-], the solvent, or the olefin. One example of an olefin complex in shown is Figure 1.
 

The potential energy surface for the formation these complexes and ion-pairs will be described for six catalyst systems with tris(pentafluorophenyl)borane acting as the Lewis acid. The catalysts included in this study are: biscyclopentadienyl systems (1), monocyclopentadienyl systems (2), and constrained geometry systems (3).

The influence of the solvent and the nature of the catalyst precursor on the ion-pair formation equilibria will be discussed, and their implications on the mechanism for polymerization will be presented. The potential energy surfaces were constructed by geometry optimization within the local density approximation, followed by non-local correlation and exchange corrections to the energy terms. Solvation effects were incorporated by single point calculations using the Conductor-like Screening Model, and where appropriate, a single molecule of the solvent was included to model tightly bound solvent-solute interactions.