(256d) Elucidation of Non-Catalytic Ethylene Polymerization Reactions through Computational Study

Ethylene is a highly important industrial precursor chemical that is used primarily in the synthesis of various polymers. Ethylene, whether obtained directly from natural gas sources or synthesized from other hydrocarbons, is transported primarily through pipelines under high pressure conditions.¹ In order to mitigate ethylene polymerization within pipelines during transport, it is essential to understand the mechanisms by which the polymerization occurs. While ample research effort has been applied to understanding the catalytic formation of various oligomers from ethylene,^2,3 both using homogeneous and heterogeneous catalysts, knowledge of how such oligomerization may occur in the absence of catalysts is still lacking.

Owing to the intense focus on catalytic ethylene polymerization, coupled with a scarcity in recent quantum chemical investigations into the topic, the mechanisms of ethylene polymerization are still not well understood. In this work we employ density functional theory calculations to elucidate the pathways by which oligomers are formed and to better understand the temperature and pressure dependence of the product distributions. We begin by modeling and comparing several bimolecular reactions of ethylene, identifying those which may serve as initiators for subsequent chain growth reactions. From these initiation reactions we identify two key routes for the generation of biradical species, one through homolytic cleavage of cyclobutane, and the other through cleavage of cyclobutene. Lastly, we explore how, through intramolecular rearrangements, intermolecular hydrogen abstractions, and free-radical polymerization reactions, these biradical species facilitate both the formation of oligomers and the generation of new radical species through more facile routes than available during the initiation period. The results of this work can help guide transport conditions in order to minimize oligomer formation during ethylene transport in pipelines.

All calculations were conducted in the gas-phase using the M062X meta-hybrid functional with the Def2-TZVP basis set of Alrichs and co-workers. The UltraFine integration grid and default optimization convergence criteria were used throughout. Dispersion was included in the form of Grimme’s D3 empirical dispersion correction without any damping. Contributions to the enthalpy of the system from frequency modes below 100 cm-1 were adjusted using the quasi-harmonic (QH) correction of Head-Gordon. Contributions to the entropy from the same small vibrational modes were adjusted using the QH method of Grimme. Both methods were used as provided in the GoodVibes software. GoodVibes was also used to scale all vibrational frequencies by a factor of 0.971, as recommended by the work of Donald Truhlar at the University of Minnesota.

References:

1 J. M. Bielicki, R. S. Middleton, J. S. Levine and P. Stauffer, Energy Procedia, 2014, 63, 7215–7224.

2 J. Malinowski, D. Jacewicz, B. Gawdzik and J. Drzeżdżon, Sci. Rep., 2020, 10, 2–7.

3 R. Y. Brogaard and U. Olsbye, ACS Catal., 2016, 6, 1205–1214.