A generalization of the charge equilibration method for nonmetallic materials.

Area of Science:

• Computational chemistry
• Molecular modeling
• Quantum chemistry

Background:

• Accurate effective atomic charges are essential for constructing reliable interatomic potentials.
• Existing charge equilibration methods have limitations in reproducing molecular electrostatic fields.

Purpose of the Study:

• To develop a new charge equilibration method for calculating effective atomic charges.
• To improve the accuracy of electrostatic potential predictions in molecules.

Main Methods:

• Introduced a novel charge equilibration approach using 'split charges' for covalent neighbors.
• Incorporated atomic properties (electronegativity, hardness) and bond-dependent properties into semiempirical fit parameters.
• Applied the methodology to molecules containing silicon, carbon, oxygen, and hydrogen.

Main Results:

• Achieved over twofold improvement in predicting effective charges from electrostatic potential surfaces compared to previous methods.
• The new method encompasses two previously separate approaches as limiting cases.
• A minor 30% accuracy increase was observed for Mulliken charge predictions, with potential for improvement.

Conclusions:

• The proposed 'split charge' method offers a significant advancement in calculating accurate effective atomic charges.
• This method enhances the construction of interatomic potentials, particularly for electrostatic potential surface predictions.
• The approach provides a unified framework for charge equilibration, improving molecular electrostatic field representation.