Localized and delocalized states of a diamine cation: A critical test of wave function methods

TL;DR

This study evaluates various quantum chemistry methods for their ability to accurately predict localized versus delocalized electronic states in a specific cation, highlighting the importance of dynamic correlation effects.

Contribution

The paper provides a comprehensive comparison of multiple electronic structure methods on a benchmark system, revealing how dynamic correlation influences state localization predictions.

Findings

CCSD predicts a localized state, while CCSD(T) does not.

Large active-space CASSCF predicts localization, NEVPT2 does not.

MRCI+Q results align closely with experimental data.

Abstract

The relative stability of a localized and delocalized electronic state in the same molecule, the N,N' -Dimethylpiperazine cation, is calculated at various levels of theory up to multireference configuration interaction (MRCI+Q). This system has received a great deal of attention because of recent experimental studies of corresponding Rydberg states of the molecule and the failure of most density functional approximations to produce a metastable localized state. A cut through the energy surface involving two dihedral angles is generated at the level of MRCI+Q as well as Hartree-Fock (HF), M\"oller-Plesset second order perturbation theory (MP2), coupled cluster theory with and without perturbative triple excitations (CCSD and CCSD(T)) and complete active space self-consistent field calculations with and without perturbative correction (CASSCF and NEVPT2). Remarkably, while CCSD produces a localized state, CCSD(T) does not, and similarly, large active-space CASSCF does while NEVPT2 does not. The inclusion of dynamic correlation in a perturbative way thus adversely affects the accuracy of the calculation, most notably for CCSD(T), the 'golden standard'. The MRCI+Q results are in close correspondence with the experimental results, as well as CAS(19,20) DMRG-CASSCF calculations. The results presented here establish a benchmark system for the study of electronic state localization.