A theoretical-electron-density databank using a model of real and virtual spherical atoms

Area of Science:

• Computational chemistry
• Solid-state physics
• Materials science

Background:

• Accurate molecular charge density is crucial for understanding chemical properties.
• Existing multipolar atom modeling requires numerous parameters.
• Developing efficient and transferable charge density models is an ongoing challenge.

Purpose of the Study:

• To propose a novel database for molecular electron density using spherical atoms.
• To offer an alternative to multipolar atom modeling.
• To create a transferable model for computational chemistry applications.

Main Methods:

• Theoretical structure factors were computed using periodic density functional theory.
• Charge density was refined using a model with real spherical atoms and dummy charges.
• A database of transferable spherical atoms was built by averaging parameters from 38 crystal structures.

Main Results:

• The proposed spherical charge modeling requires fewer parameters than multipolar methods.
• The model demonstrates ease of incorporation into molecular modeling software.
• Successful application to urea and the biotin/streptavidin complex validates the method.

Conclusions:

• Spherical charge modeling provides an efficient and parameter-light approach to electron density description.
• The developed database enhances the computation of electrostatic properties.
• This method holds potential for broader applications in molecular modeling and drug design.