Electrostatic Potential Derived Atomic Charges for Periodic Systems Using a Modified Error Functional

Area of Science:

• Solid-state Chemistry
• Computational Materials Science
• Quantum Chemistry

Background:

• Accurate atomic charges are crucial for simulating materials properties.
• Conventional electrostatic potential (ESP) fitting methods fail for crystalline solids due to ill-defined potentials.
• Existing methods struggle to provide reliable atomic charges for periodic systems.

Purpose of the Study:

• To introduce a novel method, REPEAT, for deriving ESP-based atomic charges in crystalline solids.
• To address the limitations of molecular ESP fitting in periodic systems.
• To enable accurate charge generation for simulations of microporous and nanoporous materials.

Main Methods:

• Developed a new error functional focusing on relative potential differences, circumventing the constant offset issue in periodic systems.
• Formally demonstrated the convergence of the new functional to conventional molecular ESP functionals for large simulation boxes.
• Validated the REPEAT method through calculations on isolated molecules and crystalline siliceous sodalite.

Main Results:

• The REPEAT method yields atomic charges in good agreement with established molecular methods for isolated molecules.
• For siliceous sodalite, REPEAT produces chemically intuitive and consistent charges, unlike conventional molecular approaches.
• Demonstrated the method's applicability to various microporous materials, showing consistency with experimental and fragment-based charges.

Conclusions:

• The REPEAT method provides a robust and reliable approach for generating ESP-derived atomic charges in crystalline solids.
• This method overcomes fundamental issues with applying molecular charge derivation techniques to periodic systems.
• REPEAT facilitates accurate simulations of zeolites, metal-organic frameworks, and other nanoporous materials by providing reliable partial atomic charges.