The UHF potential energy surfaces of linear polyenes in the lowest excited triplet state

Triplet State Z/E-Photoisomerizations of Polyenes: A Comparison of ab Initio and Density Functional Methods

“…The multireference character of the T 1 wave functions of HT1 and HT3 is low since the determinant composed of the HOMO−LUMO excitation makes up ∼85% of the wave function at the CASSCF(6,6)/6-31G(d,p) level in both isomers (Table ). This is in contradiction to what was recently stated by Kikuchi and co-workers, , who claimed that planar polyenes with odd numbers of CC bonds must be described as “resonating biradicals” in their T 1 states and that multireference methods are needed when computing such states. It should be mentioned that Roos and co-workers recently found that E -stilbene in the T 1 state is described to 76% by the determinant corresponding to the HOMO−LUMO excitation .…”

“…Nevertheless, UHF together with the STO-3G minimal basis set was recently utilized by Kikuchi and co-workers to investigate twisting around various CC bonds in a range of polyenes. , It was suggested that the triplet state of planar polyenes in some cases must be described by multiconfigurational SCF methods due to a “resonating biradical” character. However, an investigation of the multireference character of the T 1 state of BD gives at hand that there is merely little such character along the twisting of the CC bonds.…”

“…When calculating HT1 in the T 1 state at the UHF/STO-3G level, as recently done by Kikuchi and coworkers, , we obtained a spin-contaminated wave function with 〈 Ŝ 2 〉 = 2.6, i.e., far from the ideal value of 2. Remarkably, there is no instability of the UHF/STO-3G T 1 wave function of HT1 , but since it is contaminated by higher multiplets, it corresponds to a state that should be described as a partial tetraradical.…”

“…From computations it has been found that Z / E -isomerization of the central CC bond in HT indeed is easier than isomerization of a terminal CC bond. The most recent study was made by Kikuchi and coworkers at the UHF/STO-3G level. , Calculations at more sophisticated levels of theory were performed by Orlandi and co-workers . At the CASSCF(6,6)/6-31G(d,p) level the vertical S 0 → T 1 excitation is well described, whereas ROHF gives a value that deviates by 9.0 kcal/mol from experiment.…”