Hydrogen Bond Benchmark: Focal-Point Analysis and Assessment of DFT Functionals

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• Accurate description of hydrogen bonds is crucial in chemistry and biology.
• Density functional approximations (DFAs) are widely used but their performance varies.
• A systematic benchmark is needed to guide the selection of appropriate DFAs.

Purpose of the Study:

• To systematically evaluate the performance of 60 density functionals for hydrogen bond description.
• To identify the best-performing functionals for various systems, from small complexes to larger molecules.
• To provide reference data for future computational studies involving hydrogen bonds.

Main Methods:

• Hierarchical, convergent ab initio benchmark study.
• Focal point analyses (FPA) extrapolated to the ab initio limit using high-level wave function methods (up to CCSDT(Q) and CCSD(T)).
• Evaluation of 60 density functionals, including dispersion-corrected variants, against accurate reference data.

Main Results:

• The meta-hybrid M06-2X demonstrated the best overall performance for hydrogen bond energies and geometries.
• Dispersion-corrected GGAs, BLYP-D3(BJ) and BLYP-D4, also provided accurate results and are cost-effective.
• FPA hydrogen-bond energies were converged to within a few tenths of a kcal mol⁻¹.

Conclusions:

• M06-2X is recommended as a top-performing functional for hydrogen bond studies.
• BLYP-D3(BJ) and BLYP-D4 offer accurate and efficient alternatives for large-scale calculations.
• This benchmark provides valuable guidance for selecting DFAs in computational chemistry research.