Subsystem density functional theory with meta generalized gradient approximation exchange-correlation functionals

TL;DR

This paper introduces a Laplacian-level approximation for the Kohn-Sham kinetic energy density, enabling the use of meta-GGA exchange-correlation functionals in subsystem DFT calculations for non-bonded systems, with accuracy comparable to conventional methods.

Contribution

It proposes a novel Laplacian-level approximation for the KED that allows meta-GGA functionals to be used in subsystem DFT, improving applicability and accuracy.

Findings

Density and energy errors are comparable with supermolecular calculations.

The method's accuracy depends mainly on the non-additive kinetic embedding term.

The approach effectively overcomes the KED functional dependency issue.

Abstract

We analyze the methodology and the performance of subsystem density functional theory (DFT) with meta-generalized gradient approximation (meta-GGA) exchange-correlation functionals for non-bonded systems. Meta-GGA functionals depend on the Kohn-Sham kinetic energy density (KED), which is not known as an explicit functional of the density. Therefore, they cannot be directly applied in subsystem DFT calculations. We propose a Laplacian-level approximation to the KED which overcomes the problem and provides a simple and accurate way to apply meta-GGA exchange-correlation functionals in subsystem DFT calculations. The so obtained density and energy errors, with respect to the corresponding supermolecular calculations, are comparable with conventional approaches, depending almost exclusively on the approximations in the non-additive kinetic embedding term. An embedding energy error decomposition explains the accuracy of our method.