Optimization of a Range-Separated UW12 Hybrid Functional

08 October 2021, Version 1
This content is a preprint and has not undergone peer review at the time of posting.

Abstract

In a previous paper we presented a new hybrid functional B-LYP-osUW12-D3(BJ) containing the Unsöld-w12 (UW12) hybrid correlation model. In this paper we present a new 15-parameter range-separated hybrid density functional using a power series expansion together with UW12 correlation. This functional is optimised using the survival of the fittest strategy developed for the ωB97X-V functional, fitted to data from the Main Group Chemistry Database (MGCDB84). In addition we optimize a standard hybrid and double hybrid using the same method. We show that our fully self-consistent UW12 hybrid functional WM21-D3(BJ) outperforms both of these functionals and other range-separated hybrid functionals.

Keywords

DFT
Electronic Structure
Density Functional Theory
Hybrid density functionals
Hybrid correlation

Supplementary materials

Title
Description
Actions
Title
Supporting Information
Description
Includes a summary of the main-group chemistry database (MGCDB84) used in this work. As well as the errors in the training data as the functional is re-optimised using self-consistent orbitals. Also included are errors in the testing data for both the initial non-self-consistently optimised functionals and the final functionals.
Actions
Title
Functional data
Description
Tables detailing the individual energy contributions for each component used to optimise the UW12 functionals for all datasets and molecules included in the paper. In addition to calculated energies for the final self-consistent functionals and a summary of functional coefficients.
Actions

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.