A transferable quantum mechanical energy model for intermolecular interactions using a single empirical parameter.

Area of Science:

• Computational chemistry
• Materials science
• Solid-state physics

Background:

• Understanding intermolecular interactions is crucial for predicting molecular crystal properties.
• Existing models often require complex parameterization or struggle with accuracy across diverse elements.

Purpose of the Study:

• To introduce a new, simplified single-parameter interaction energy model (CE-1p) for molecular crystals.
• To enhance the accuracy of calculating intermolecular forces, driving forces in crystallization, and elastic properties.

Main Methods:

• Developed the CE-1p model, extending the CrystalExplorer energy model.
• Calibrated the model using density functional theory (DFT) calculations (ωB97M-V/def2-QZVP) on 1157 interactions from 147 crystal structures.
• Incorporated the exchange-hole dipole model (XDM) for dispersion and polarizabilities, and effective core potentials (ECPs) for broad elemental applicability (H to Rn).

Main Results:

• The CE-1p model demonstrates outstanding performance, comparable to state-of-the-art DFT methods.
• Achieved a mean absolute deviation of 3.6 kJ/mol for molecular crystal lattice energies (X23 set).
• Showcased a root mean square deviation of 3.3 kJ/mol for intermolecular interactions (S66x8 benchmark set).

Conclusions:

• The CE-1p model provides a unified and accurate approach for calculating intermolecular interactions in molecular crystals.
• The model's performance validates its utility for predicting crystal properties and offers a valuable alternative to computationally intensive DFT methods.
• Recommendations are provided for its application and comparison with other models like GFN2-xTB.