arXiv · 2611.21790
Multicomponent Adaptive Sampling Configuration Interaction method in openLOWDIN: positronic and quantum nuclear systems
Abstract
We present a multicomponent Selective Configuration Interaction (SCI) method based on the deterministic Adaptive Sampling CI (ASCI) algorithm for computing energies and wavefunction properties of molecular systems containing leptons of variable masses, or quantum nuclei. Near-FCI accuracy is achieved by incorporating second-order Epstein-Nesbet perturbation theory or by dressing the Hamiltonian via Davidson's and co-workers shifted-Bk method. We validate the approach through benchmark positronic (e$^+$LiH, e$^+$OH$^-$, e$^+$Mg) and quantum nuclear systems (HCN), establishing a novel protocol for accurate convergence, based on the iterative adaptation of natural orbitals starting from those obtained by considering only cross-particle CISD excitations, to those obtained from SCI variational space, surpassing the accuracy of previous SCI studies. Overall, we show that the present multi-component SCI framework offers a simple, systematically improvable, and cost-effective alternative to traditional CI approaches.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Jorge Charry, Matteo Barborini. 2026-11-16. Multicomponent Adaptive Sampling Configuration Interaction method in openLOWDIN: positronic and quantum nuclear systems. https://arxiv.science/abs/2611.21790
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.