Ab initio study and NBO analysis of the allylic rearrangements (hetero Claisen and Cope rearrangements) and decarboxylation reactions (retro carbonyl ene reaction) of allylformate and allyldithioformate

Abstract

MP2/6-311+G **//B3LYP/6-31G **, MP2/6-31G **//B3LYP/6-31G **, MP4/6-31G **//B3LYP/6-31G ** and B3LYP/6-31G **//B3LYP/6-31G ** methods were used to investigate the allylic rearrangement (hetero Claisen and Cope rearrangements) and decarboxylation reactions (retro carbonyl ene reaction) in allylformate ( 1) and its sulfur containing analogue (allyldithioformate, 2). All used methods reveal that the allylic rearrangement barrier heights in compounds 1 and 2 are lower than those of decarboxylation reactions. The calculations based on MP2/6-311+G **//B3LYP/6-31G **, MP2/6-31G **//B3LYP/6-31G **, MP4/6-31G **//B3LYP/6-31G ** and B3LYP/6-31G **//B3LYP/6-31G ** methods show that the allylic rearrangement energy barriers in compound 1 are 44.31, 47.84, 43.07 and 39.10 kcal mol −1, respectively, while in compound 2 these energy barriers are 25.10, 29.09, 27.63 and 23.42 kcal mol −1, respectively. The results also reveal that the allylic rearrangement energy barrier in compound 1 is higher (about 20 kcal mol −1) than in compound 2. In addition, the decarboxylation energy barrier in compound 1 is found to be 47.15, 49.34, 47.55 and 42.96 kcal mol −1, as calculated by MP2/6-311+G **//B3LYP/6-31G **, MP2/6-31G **//B3LYP/6-31G **, MP4/6-31G **//B3LYP/6-31G ** and B3LYP/6-31G **//B3LYP/6-31G ** levels of theory, respectively. Similarly, for compound 2, MP2/6-311+G **//B3LYP/6-31G **, MP2/6-31G **//B3LYP/6-31G **, MP4/6-31G **//B3LYP/6-31G ** and B3LYP/6-31G **//B3LYP/6-31G ** calculations reveal that the decomposition (retro carbonyl ene reaction) energy barriers in compound 2 are 29.90, 31.49, 33.43 and 26.50 kcal mol −1, respectively. In addition, the results show that the HOMO–LUMO energy-gap in the ground state structures in compounds 1 and 2, decrease parallely with the decrease of the rearrangement and decomposition reactions energy barriers from compound 1 to 2. Therefore, these results show that the decomposition and allylic rearrangement processes in compound 2 are faster than in compound 1. Among the used methods in this work, the obtained results show that B3LYP/6-31G ** decomposition energy barrier for compound 1 is in good agreement with the available experimental data. Based on the optimized ground state geometries and using B3LYP/6-31G ** method, the NBO analysis of donor–acceptor (bond–antibond) interactions also reveal that the resonance energies associated with the electronic delocalization from σ 2–3 bonding orbitals to π 4 - 5 * anti-bonding orbitals, increase from compounds 1 to 2. The increase of σ 2 - 3 → π 4 - 5 * resonance energy could decrease the corresponding TS energies of the concerted reactions from compounds 1 to 2, due to the increase of the aromatic character in the TS structures. Also, the decrease of σ 2–3 bonding orbitals occupancies and increase of the π 4 - 5 * anti-bonding orbitals occupancies via σ 2 - 3 → π 4 - 5 * delocalizations, could facilitate the corresponding reactions of compound 2, compared to compound 1.