The potential energy surface for the oxywater radical cation transformation to hydrogen peroxide radical cation studied by density functional theory and ab initio methods. Are hybrid density functional methods as accurate as coupled-cluster ab initio methods?

Abstract

The G1, G2, G2MP2 ab initio and 12 density functional theory (DFT) methods were applied to study the geometries, energies, activation barriers and vibrational spectra of “distonic” oxywater and hydrogen peroxide radical cations. The results were compared with values previously obtained by coupled-cluster including singlet, doublet and perturbatively treated connected triple excitations. Values obtained with G2 and CCSD ab initio methods are identical. From 12 DFT methods, the three hybrid methods (B3LYP, B3P86 and B3PW91) generate geometries, energies and vibrational spectra that are of the same quality as CCSD or G2. The hybrid DFT computed energy for the H 2OO + [1–12] hydrogen shift to HOOH + is between 33 and 34 kcal mol −1. This is very close to the 32–33 kcal mol −1 computed with G2 and CCSD. The trans hydrogen peroxide radical cation is found to be 7.8 kcal mol −1 more stable than its cis isomer. The value is exactly the same when computed with G2 and CCSD. The hybrid DFT computed vibrational spectra are slightly lower than CCSD values, which are believed to be even closer to the experimental values. Other DFT methods do not perform as well and they are not recommended for use in the computational study of cation radicals.