Exploring the competition between localization and delocalization of the neutral soliton defect in polyenyl chains with the orbital optimized second order opposite spin method.

Theory and implementation of the analytical nuclear gradient is presented for orbital optimized scaled opposite-spin perturbation theory (O2). Evaluation of the O2 analytical gradient scales with the 4th power of molecular size, like the O2 energy. Since the O2 method permits optimization of the orbitals in the presence of wavefunction-based electron correlation, it is suitable for problems where correlation effects determine the competition between localization and delocalization of an odd electron, or hole. One such problem is the description of a neutral soliton defect on an all-trans polyacetylene chain with an odd number of carbon atoms. We show that the results of the O2 method compare well to benchmark values for small polyenyl radicals. O2 is also efficient enough to be applied to longer chains where benchmark coupled cluster methods are unfeasible. For C(41)H(43), unrestricted orbital O2 calculations yield a soliton length of about 9 carbon atoms, while other unrestricted orbital methods such as Hartree-Fock, and the B3LYP and ωB97X-D density functionals, delocalize the soliton defect over the entire chain. The O2 result is about half the width inferred experimentally.