Gaussian Charge-Based Electrostatic Embedding Scheme for Solid-State Excited-State Modeling

Area of Science:

• Computational Chemistry
• Materials Science
• Solid-State Physics

Background:

• Electrostatic embedding schemes are crucial for modeling excited-state properties of crystalline materials affordably and accurately.
• Traditional point-charge (PC) formalisms can suffer from numerical instabilities and unphysical electron density behavior.

Purpose of the Study:

• Introduce and validate a Gaussian charge-based (GC) electrostatic embedding scheme for solid-state excited-state calculations.
• Address limitations of PC formalisms by offering a more physically sound approach.

Main Methods:

• Developed a GC electrostatic embedding scheme utilizing analytical Ewald lattice summations.
• Validated the scheme by comparing electrostatic potentials from PC and GC formalisms across 530 crystalline structures.
• Proposed a criterion based on interatomic distance to optimize the Gaussian width parameter.

Main Results:

• Achieved excellent agreement between PC and GC electrostatic potentials with an optimized Gaussian width parameter.
• Demonstrated that GC embedding avoids the excessive electron density contraction seen with PC methods.
• Obtained excitation energies for crystalline imidazole using GC embedding that agree well with GW-BSE and hybrid calculations.

Conclusions:

• The proposed GC electrostatic embedding scheme is a robust and physically sound alternative to PC models for excited-state calculations in solids.
• GC embedding offers a promising framework for future methodological advancements and practical applications in embedded excited-state calculations.