Long-range-corrected Rung 3.5 density functional approximations.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Density functional theory (DFT) approximations are crucial for electronic structure calculations.
• Existing Rung 3.5 functionals, while flexible, struggle with accuracy in atomic cores and density tails.
• Semilocal approximations and exact Hartree-Fock (HF) exchange have limitations that Rung 3.5 aims to bridge.

Purpose of the Study:

• To develop new Rung 3.5 functionals that address limitations in atomic cores and density tails.
• To introduce range-separated admixture of HF exchange into Rung 3.5 approximations.
• To provide a framework for more flexible range-separated Rung 3.5 approximations in DFT.

Main Methods:

• Development of three novel Rung 3.5 functionals: LRC-ωΠLDA, SLC-ΠLDA, and LRC-ωΠLDA-AC.
• Implementation of a new Rung 3.5 scheme capable of analytic fourth derivatives.
• Testing the functionals against established benchmarks including atomization energies, reaction barriers, and electronic properties.

Main Results:

• LRC-ωΠLDA and SLC-ΠLDA show an 8-fold improvement in atomization energies and reaction barriers over full-range ΠLDA.
• LRC-ωΠLDA-AC achieves accuracy comparable to standard long-range corrected schemes like LC-ωPBE.
• New functionals provide more accurate highest occupied orbital energies and correctly describe complex electronic phenomena, including charge-transfer complexes and defect-induced spin distributions.

Conclusions:

• The developed range-separated Rung 3.5 functionals significantly enhance the accuracy of DFT calculations.
• These functionals effectively address the limitations of previous approximations in specific electronic regions.
• The study establishes a versatile framework for future development of advanced DFT approximations.