Abstract
Almost all empirical parameterizations of dispersion corrections in DFT use only energy errors, thereby mixing functional and density-driven errors. We introduce density and dispersion-corrected DFT (D2C-DFT), a dual-calibration approach that accounts for density delocalization errors when parametrizing dispersion interactions. We simply exclude density-sensitive reactions from the training data. We find a significant reduction in both errors and variation among several semilocal functionals and their global hybrids when tailored dispersion corrections are employed with Hartree-Fock densities.
Supplementary materials
Title
Supporting Information
Description
Dispersion parameters for 14 D2C-DFT and DFT-D4, Full table of across 14 function-
als on five datasets, D2C-B3LYP parameterization process, Parameterization process using
density-sensitive reactions, Abbreviation used for GMTKN55, Description of subsets within
GMTKN55, Simplified view on parameterization strategy, Performance on GMTKN55 for
14 functionals, Distribution of weighted errors for 14 functionals across GMTKN55, Per-
formance on WATER27 for 14 functionals, Performance on water hexamers and 20-mers,
Evaluation of r2SCAN correction variants using the Báuza dataset, Transferability of disper-
sion parameters for r2SCAN variants.
Actions
Title
Raw data of calculations in GMTKN55 database
Description
Raw data of calculations in GMTKN55 database
Actions