Login / Signup

Investigation of the Exchange-Correlation Potentials of Functionals Based on the Adiabatic Connection Interpolation.

Eduardo FabianoSzymon ŚmigaSara GiarrussoTimothy J DaasFabio Della SalaIreneusz GrabowskiPaola Gori-Giorgi
Published in: Journal of chemical theory and computation (2019)
We have studied the correlation potentials produced by various adiabatic connection models (ACMs) for several atoms and molecules. The results have been compared to accurate reference potentials (coupled cluster and quantum Monte Carlo results) as well as to state-of-the-art ab initio DFT approaches. We have found that all the ACMs yield correlation potentials that exhibit a correct behavior, quite resembling scaled second-order Görling-Levy (GL2) potentials and including most of the physically meaningful features of the accurate reference data. The behavior and contribution of the strong-interaction limit potentials have also been investigated and discussed.
Keyphrases
  • monte carlo
  • electronic health record
  • machine learning
  • big data
  • molecular docking
  • artificial intelligence