Meta-generalized gradient approximation: explanation of a realistic nonempirical density functional

Area of Science:

• Quantum Chemistry
• Condensed Matter Physics
• Computational Physics

Background:

• The development of accurate density functional approximations (DFAs) is crucial for computational materials science and quantum chemistry.
• Existing generalized gradient approximations (GGAs) have limitations in describing certain electronic properties.
• The meta-generalized gradient approximation (meta-GGA) offers a potential improvement by incorporating more information about the electron density.

Purpose of the Study:

• To introduce and detail the construction of the nonempirical Tao, Perdew, Staroverov, and Scuseria (TPSS) meta-GGA functional.
• To investigate the performance of the TPSS functional for various properties across different systems.
• To explore the implications of exact constraints and potential improvements in density functional theory.

Main Methods:

• Construction of a nonempirical meta-GGA functional based on exact constraints.
• Testing the TPSS functional on molecules, solids, and solid surfaces.
• Analysis of exchange and correlation energies, shell-removal energies, and atomization energies.
• Investigation of nonlocality using the enhancement factor F(xc).

Main Results:

• The TPSS meta-GGA demonstrates encouraging uniform accuracy for diverse properties and systems.
• Accurate energies are obtained under uniform coordinate scaling to low-density or strong-interaction limits.
• Non-self-consistent calculations using TPSS yield results comparable to self-consistent GGA calculations for molecular properties.

Conclusions:

• The TPSS meta-GGA is a promising development in density functional approximations.
• Satisfying additional exact constraints could lead to further advancements in density functional theory.
• The functional shows good performance for a range of chemical and physical applications.