Nonlocal Subsystem Density Functional Theory.

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Materials Science

Background:

• Subsystem Density Functional Theory (DFT) offers a divide-and-conquer approach to reduce computational costs in electronic structure simulations.
• The accuracy of subsystem DFT relies on the nonadditive kinetic energy functional (NAKE).
• Existing NAKEs are limited to semilocal approximations, restricting large-scale simulations to weakly interacting systems.

Purpose of the Study:

• To introduce and evaluate fully nonlocal NAKEs for subsystem DFT simulations for the first time.
• To improve the accuracy of interaction energies and electron densities in large-scale molecular and materials simulations.
• To address the over-attractive interaction energy curves often produced by semilocal NAKEs.

Main Methods:

• Development and implementation of fully nonlocal NAKEs within the subsystem DFT framework.
• Benchmark analysis using the S22-5 test set to compare performance against traditional GGA NAKEs.
• Evaluation of computed interaction energies and electron densities, particularly for systems with overlapping electron densities.

Main Results:

• Nonlocal NAKEs demonstrate considerable improvements in computed interaction energies and electron densities compared to GGA NAKEs.
• Performance enhancements are particularly notable with increasing intersubsystem electron density overlap.
• The study successfully resolves the issue of overly attractive interaction energy curves associated with GGA NAKEs.

Conclusions:

• Fully nonlocal NAKEs represent a significant advancement in subsystem DFT, enabling more accurate simulations of molecular and material systems.
• This new methodology overcomes limitations of semilocal approximations, expanding the applicability of large-scale electronic structure calculations.
• The improved accuracy in interaction energies paves the way for more reliable predictions in computational chemistry and materials science.