Optimally tuned range-separated hybrid van der Waals density functional for molecular binding and quasiparticle...

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Accurate prediction of molecular energies and quasiparticle properties is crucial in computational chemistry.
• Existing van der Waals density functionals (vdW-DFs) have limitations in simultaneously describing short- and long-range interactions.
• Generalized Kohn-Sham (gKS) density functional theory (DFT) requires robust functionals for reliable characterizations.

Purpose of the Study:

• To introduce and evaluate two novel range-separated hybrid (RSH) vdW-DFs: AHBR-mRSH and AHBR-mRSH*.
• To assess the performance of these functionals for total-energy and quasiparticle calculations in molecules.
• To compare the new functionals against a traditional long-range corrected (LRC) vdW-DF, B86R-LRC.

Main Methods:

• Development of two closely related RSH vdW-DFs, AHBR-mRSH and AHBR-mRSH*, within gKS-DFT.
• Definition of the AHBR-mRSH(γ) class with an adjustable inverse length scale γ for exchange contributions.
• Application of the GMTKN55 benchmark suite for evaluating the transferability and accuracy of the functionals.

Main Results:

• AHBR-mRSH, with a fixed γ = 0.106 Bohr⁻¹, accurately predicts molecular energy differences.
• AHBR-mRSH* (optimally tuned) extends these capabilities to quasiparticle characterizations, incorporating nonlocal correlations.
• AHBR-mRSH demonstrates superior performance compared to B86R-LRC for molecular problems, including nucleobases.

Conclusions:

• AHBR-mRSH and AHBR-mRSH* are internally consistent functionals for gKS-DFT, providing simultaneous accuracy for energies and quasiparticles.
• The AHBR-mRSH(γ) class exhibits sufficient transferability, minimizing adverse effects of tuning γ.
• Optimally tuned AHBR-mRSH* functionals show good agreement with experimental and theoretical values for nucleobases like adenine and guanine.