Thermal and photochemical rearrangement of bicyclo[3.1.0]hex-3-en-2-one to the ketonic tautomer of phenol.: Computational evidence for the formation of a diradical rather than a zwitterionic intermediate

The ground state (So) and lowest-energy triplet state (T-1) potential energy surfaces (PESs) concerning the thermal and photochemical rearrangement of bicyclo[3.1.0]hex-3-en-2-one (8) to the ketonic tautomer of phenol (11) have been extensively explored using ab initio CASSCF and CASPT2 calculations with several basis sets. State T-1 is predicted to be a triplet pipi* lying 66.5 kcal/mol above the energy of the So state. On the So PES, the rearrangement of 8 to 11 is predicted to occur via a two-step mechanism where the internal cyclopropane C-C bond is broken first through a high energy transition structure (TS1-S-0), leading to a singlet intermediate (10-S-0) lying 25.0 kcal/mol above the ground state of 8. Subsequently, this intermediate undergoes a 1,2-hydrogen shift to yield 11 by surmounting an energy barrier of only 2.7 kcal/mol at 0 K. The rate-determining step of the global rearrangement is the opening of the three-membered ring in 8, which involves an energy barrier of 41.2 kcal/mol at 0 K. This high energy barrier is consistent with the fact that the thermal rearrangement of umbellulone to thymol is carried out by heating at 280 degreesC. Regarding the photochemical rearangement, our results suggest that the most efficient route from the T, state of 8 to ground state 11 is the essentially barrierless cleavage of the internal cyclopropane C-C bond followed by radiationless decay to the So state PES via intersystem crossing (ISC) at a crossing point (S-0/T-1-1) located at almost the same geometry as TS1-S-0, leading to the formation of 10-S-0 and the subsequent low-barrier 1,2-hydrogen shift. The computed small spin-orbit coupling between the T-1 and So PESs at S-0/T-1-1 (1.2 cm(-1)) suggests that the ISC between these PESs is the rate-determining step of the photochemical rearrangement 8 --> 11. Finally, computational evidence indicates that singlet intermediate 10-S-0 should not be drawn as a zwitterion, but rather as a diradical having a polarized C=O bond.