Electron group localization in atoms and molecules.

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Chemical Physics

Background:

• Partitioning atomic and molecular charge densities into chemically meaningful regions is a fundamental challenge in quantum chemistry.
• Existing methods often struggle to define clear, non-overlapping boundaries that align with chemical intuition.

Purpose of the Study:

• To develop a computational tool for objectively determining "good boundaries" for charge density partitioning.
• To leverage information theory and statistical methods to guide the boundary definition process.

Main Methods:

• Minimizing an objective function based on clarity criteria to find optimal boundaries.
• Utilizing the sum of indices of dispersion (ΣD) or mutual information as objective functions.
• Applying elementary statistical methods and information theory principles.

Main Results:

• The developed method successfully partitions charge densities in non-overlapping, chemically significant regions.
• Partitions generated for Lithium (Li) to Radon (Rn) atoms agree well with the Aufbau principle.
• Molecular partitions align effectively with Lewis's electron pairing model.

Conclusions:

• The new method provides a robust and objective approach to charge density partitioning.
• This tool enhances the interpretability of quantum chemical calculations by aligning with established chemical rules.
• The approach offers a promising avenue for further development in computational chemistry and chemical analysis.