Size consistency and counterpoise correction in explicitly correlated calculations of interaction energies and interaction-induced properties
Micha{\l} Lesiuk, Bogumi{\l} Jeziorski

TL;DR
This paper introduces a method to achieve size-consistent interaction energies in explicitly correlated calculations by relaxing the Pauli principle, enabling more accurate interaction energy and property computations at large intermonomer distances.
Contribution
It presents a novel approach to obtain size-consistent interaction energies by relaxing the Pauli principle and optimizing basis functions, improving accuracy in explicitly correlated calculations.
Findings
Method yields interaction energies that vanish at large distances.
Counterpoise correction effectively reduces basis set superposition error.
Demonstrates improved accuracy in hydrogen atom interactions.
Abstract
Explicitly correlated calculations of interaction energies with wave functions that include all interparticle distances have suffered so far from the lack of size-consistency resulting from the difficulty to define monomer energies corresponding to the applied dimer basis. As a consequence it has not been possible to obtain interaction energies vanishing at infinite intermonomer distance . This has dramatically reduced the accuracy of calculations at distances where the error in the dimer energy was comparable with the interaction energy itself. The same problem occurs in calculations of interaction-induced properties. In this communication we show how to circumvent this difficulty and obtain interaction energies or interaction-induced properties that vanish at large . This is achieved by relaxing the Pauli principle in the diagonalization of the Hamiltonian of noninteracting…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
