Ab initio investigation of the ground X2A′ [ X2A1] and low-lying excited electronic states of C 2B

Abstract

Ab initio single- and multi-reference approaches, i.e. restricted open-shell Hartree–Fock (ROHF), coupled-cluster singles-and-doubles with approximate triples correction [CCSD(T)], complete active space self-consistent field (CASSCF), 2nd-order CAS perturbation theory (CASPT2), multi-reference singles-and-doubles configuration-interaction (MRSDCI), and density-functional B3LYP – in the cc-pVQZ basis set – have been employed to investigate the structure of the ground X 2 A′ [ X 2 A 1] state of the triatomic C 2B. We found that the ground state of the cyclic C 2B is prone to symmetry breaking along the asymmetric stretching coordinate. Single-reference highly correlated CCSD(T)/aug-cc-pVQZ method (as well as ‘static’-correlated CASSCF, and B3LYP) predicts two equivalent global minima of the adiabatic potential energy function, with the barrier between the minima of about 20 cm −1 at C 2v geometry. Wavefunctions almost free from SB are obtained using the MRSDCI approach based on the CASSCF and state-averaged (SA) CASSCF wavefunctions. Highly correlated multi-reference full-valence (FV) CASPT2 model exactly predicts a C 2v equilibrium geometry; however, the highest level FV-CAS-MRSDCI/aug-cc-pVQZ method predicts lower-symmetry structure. We found that this effect is not artifactual as stated in Leonard et al. (2000) [5]. SB arises from a pseudo-Jahn–Teller effect (PJTE), the non-adiabatic interaction with the first excited 1 2 B 2 state, which has the same symmetry as the asymmetric stretching coordinate in the ground state. Several low-lying excited electronic states have been optimized using SA-CAS-MRSDCI/aug-cc-pVTZ procedure. The first excited doublet is 1 2 B 2 (0.95 eV), and the lowest-lying quartet state 1 4 B 2 (1.74 eV). The vertical excitation spectra and transition matrix elements for six low-lying doublet and four quartet states were calculated using full valence SA-CAS-MRSDCI/aug-cc-pVTZ procedure.