Computational Mechanistic Investigation of the Phillips Polymerization Catalyst

The Phillips catalyst, chromium oxide supported on amorphous silica, produces half of the world’s annual supply of high-density polyethylene. [1] However, the mechanism by which the catalyst polymerizes ethylene is unknown, because only a small fraction of chromium sites are active and are therefore difficult to detect experimentally. In this study, we used density functional theory to evaluate the feasibility of potential mechanisms. As in previous studies [2], our calculations rule out propagation by either metallacycle ring expansion or alternating carbene/chromacyclobutane intermediates: both mechanisms have activation barriers much greater than what is measured experimentally. However, we discovered two new pathways with low propagation barriers: oxachromacycle ring expansion, and chain growth from a vinylchromium(II) site. Unfortunately, the oxachromacycle sites initiate very slowly, while the vinylchromium(II) sites terminate rapidly; both are therefore unlikely as propagating intermediates in polymer production. We also found that a vinylchromium(III) site recently proposed to arise via direct C-H activation of ethylene [3] also terminates quickly. The only propagation mechanism whose computed kinetics are comparable to the experimental kinetics and which does not terminate to produce oligomers instead of polymer chains is Cossee-Arlman polymerization by a monoalkylchromium(III) site without a neighboring proton source. While the precise nature of the initiation step remains unknown, experiments and calculations based on homogeneous organochromium compounds suggest the active sites may arise via a radical mechanism.

[1] McDaniel, M. P., Adv. Catal., 2010, 53, 123-606.

[2] Espelid, O.; Borve, K. J., J. Catal. 2000, 195, 125-139.

[3] M.P. Conley, M.F. Delley, G. Siddiqi, G. Lapadula, S. Norsic, V. Monteil, O.V. Safonova, C. Coperet, Angew. Chem. Int. Ed. 2014, 53, 1872 –1876.