Evaluates a new perturbation theory in large electronic configurations, suggesting lower computational costs.
Recently, we have developed a spin-adapted selected configuration interaction (SCI) method for capturing static electron correlation effects in large active orbital spaces and the corresponding nuclear gradients for geometry optimizations. This work reports two methods to incorporate dynamical electron correlation effects on top of the SCI wave function. First, we implemented the strongly contracted variant of N -electron valence state perturbation theory (SC-NEVPT2) including the residual terms emerging from the incompleteness of the SCI wave function. Through utilization of prototyping symmetries, efficient selection procedures and specifically designed parallelization/batching schemes, the steeply scaling computational costs and memory requirements could be alleviated. Second, we developed an approach where the notoriously problematic V ^ a − 1 and V ^ i + 1 perturber classes are treated by Epstein–Nesbet (EN) perturbation theory. The resulting hybrid EN-SC-NEVPT2 method is able to tackle active spaces of up to 30 orbitals with significantly lower computational costs than SC-NEVPT2. A set of test calculations illustrate the performance of the implemented methods.
No takes yet. Share an insight, caveat, or question.
Ugandi et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: