Benchmarking the Effective Fragment Potential Dispersion Correction on the S22 Test Set

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Modeling dispersion interactions in Density Functional Theory (DFT) often relies on empirical parameters, limiting accuracy.
• Accurate calculation of non-covalent interactions is crucial for understanding molecular systems.

Purpose of the Study:

• To compare the accuracy of empirical dispersion corrections with a first-principles derived effective fragment potential (EFP) dispersion correction.
• To evaluate these methods for calculating interaction energies and intermolecular distances in molecular dimers.

Main Methods:

• Calculated equilibrium interaction energies and intermolecular distances for the S22 test set dimers.
• Employed Density Functional Theory with Dispersion (DFT-D) and Hartree-Fock with Dispersion (HF-D) methods.
• Utilized coupled cluster CCSD(T) at the complete basis set (CBS) limit as the reference method.

Main Results:

• The HF-D(EFP) method accurately predicted dimerization energies and distances for hydrogen-bonded systems without empirical parameters.
• B3LYP-D(EFP) showed comparable or superior performance to other DFT-D and HF-D methods for dispersion-dominant and mixed systems.
• The first-principles derived -D(EFP) correction proved effective across various system types.

Conclusions:

• The first-principles derived -D(EFP) correction offers a reliable, non-empirical alternative to traditional empirical dispersion corrections.
• HF-D(EFP) is accurate for hydrogen-bonded systems, while B3LYP-D(EFP) is suitable for dispersion-dominant and mixed systems.