Regioselectivity in oxidation of vitamin E - SIBL model versus Mills-Nixon effect

ORGN 11

Thomas Rosenau, thomas.rosenau@boku.ac.at1, Amnon Stanger, stanger@tx.technion.ac.il2, and Paul Kosma, paul.kosma@boku.ac.at1. (1) Department of Chemistry, University of Natural Resources and Applied Life Sciences, Muthgasse 18, A - 1190 Vienna, Austria, (2) The Lise-Meitner-Minerva Center for Computational Quantum Chemistry, Technion - Israel Institute of Technology, 32000 Haifa, Israel
The regioselectivity of the oxidation of alpha-tocopherol (the main component of vitamin E) to an ortho-quinone methide (oQM) has been “explained” in the literature mostly by the ill-defined term “Mills-Nixon effect”. In this paper we describe the preparation of eleven alpha-tocopherol model compounds, different from each other by the sum of annulation angels, keeping the electronic factors unchanged. These compounds underwent oxidation, forming the two regioisomeric (“up” and “down”) oQMs which were trapped by vinyl methyl ether. The regioselectivity of the oQM formation changed smoothly as a function of angular strain, but not of the annulated ring size, as predicted by the Mills-Nixon effect. This is best explained by the SIBL (strain-induced bond localization) model. The combination of experimental and theoretical facts (computations at the B3LYP/6-31G* theoretical level) finally allowed conclusively addressing the almost centurylong debate about the “Mills-Nixon effect” (the regioselectivity issue) in oxidations of vitamin E-type antioxidants, and provided evidence that the highest energy intermediate in the oxidation of vitamin E is a phenolate-benzyl cation.