Comparison of gas phase and condensed phase SNi reactions. The competitive four- and five-centre cyclisations of the 3,4-epoxybutoxide anion. A joint experimental and theoretical study

Abstract

Ab initio calculations [at the MP2-Fc/6-31+G(d) level] indicate that the 3,4-epoxybutoxide anion should undergo competitive S_Ni cyclisations (through four- and five-membered transition states) to yield the (M – H)^– ions of oxetan-2-ylmethanol and tetrahydrofuran-3-ol respectively. The barriers to the transition states are comparable (ca. 70 kJ mol^–1) for each process, and the latter product is the more stable by 82 kJ mol^–1 at the level of theory indicated. Gas phase studies of the 3,4-epoxybutoxide anion excited by collisional activation are in accord with this scenario, and show, in addition, that deprotonated 2-oxetanylmethanol can convert to the starting material. Base treatment of 2-(oxiran-2-yl)ethan-1-ol (3,4-epoxybutan-1-ol) in two different solvent systems [10% aqueous sodium hydroxide and sodium hydride–tetrahydrofuran (both at reflux)] yields the same two products observed in the gas phase studies. However, deprotonated tetrahydrofuran-3-ol is the kinetic product in both solvent systems.