Formulation and Implementation of Density Functional Embedding Theory Using Products of Basis Functions

Area of Science:

• Computational chemistry
• Quantum chemistry
• Materials science

Background:

• Density functional embedding theory (DFET) is crucial for accurate electronic structure calculations.
• Representing the embedding potential efficiently is key to scaling DFET to larger systems.
• Current methods face challenges in treating different types of atomic calculations and transferring potentials.

Purpose of the Study:

• To develop a novel representation of the embedding potential using products of atomic orbital basis functions.
• To enable seamless integration of pseudopotential and all-electron calculations within DFET.
• To introduce a cost-reduction strategy for basis sets and potentials.

Main Methods:

• Developed a formalism for embedding potential representation based on atomic orbital basis functions.
• Implemented the scheme for condensed-phase and molecular systems using Gaussian and plane-wave formalisms.
• Proposed a cost-reduction technique utilizing population analysis.

Main Results:

• The new formalism allows pseudopotential and all-electron calculations to be treated uniformly.
• The embedding potential can be transferred efficiently in a compact matrix form.
• Computational scaling for embedding potential optimization is comparable to hybrid density functional theory but with a reduced prefactor.

Conclusions:

• The developed method provides an efficient and versatile approach for density functional embedding theory.
• The cost-reduction strategy enables large-scale applications to extended systems.
• Successful testing on proton-transfer reactions and surface adsorption systems validates the approach.