Effective empirical corrections for basis set superposition error in the def2-SVPD basis: gCP and DFT-C
Autor: | Martin Head-Gordon, Jonathon Witte, Jeffrey B. Neaton |
---|---|
Rok vydání: | 2017 |
Předmět: |
Chemical Physics
010304 chemical physics Basis (linear algebra) Intermolecular force General Physics and Astronomy 010402 general chemistry 01 natural sciences Small set 0104 chemical sciences Set (abstract data type) Engineering Computational chemistry Physical Sciences Chemical Sciences 0103 physical sciences Density functional theory Counterpoise Pairwise comparison Statistical physics Physical and Theoretical Chemistry Basis set Mathematics |
Zdroj: | Witte, J; Neaton, JB; & Head-Gordon, M. (2017). Effective empirical corrections for basis set superposition error in the def2-SVPD basis: GCP and DFT-C. Journal of Chemical Physics, 146(23). doi: 10.1063/1.4986962. UC Berkeley: Retrieved from: http://www.escholarship.org/uc/item/4mp2z7vv The Journal of chemical physics, vol 146, iss 23 |
ISSN: | 1089-7690 |
Popis: | © 2017 Author(s). With the aim of mitigating the basis set error in density functional theory (DFT) calculations employing local basis sets, we herein develop two empirical corrections for basis set superposition error (BSSE) in the def2-SVPD basis, a basis which - when stripped of BSSE - is capable of providing near-complete-basis DFT results for non-covalent interactions. Specifically, we adapt the existing pairwise geometrical counterpoise (gCP) approach to the def2-SVPD basis, and we develop a beyond-pairwise approach, DFT-C, which we parameterize across a small set of intermolecular interactions. Both gCP and DFT-C are evaluated against the traditional Boys-Bernardi counterpoise correction across a set of 3402 non-covalent binding energies and isomerization energies. We find that the DFT-C method represents a significant improvement over gCP, particularly for non-covalently-interacting molecular clusters. Moreover, DFT-C is transferable among density functionals and can be combined with existing functionals - such as B97M-V - to recover large-basis results at a fraction of the cost. |
Databáze: | OpenAIRE |
Externí odkaz: |