Embedded mean-field theory.

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Electronic Structure Theory

Background:

• Accurate electronic structure calculations are crucial for understanding chemical phenomena.
• Existing multiscale methods can be limited by parameterization or fixed particle numbers.

Purpose of the Study:

• Introduce and benchmark Embedded Mean-Field Theory (EMFT) for multiscale electronic structure calculations.
• Evaluate EMFT's performance against established methods like ONIOM.

Main Methods:

• Developed EMFT, allowing flexible embedding of one mean-field theory within another.
• Benchmarked EMFT using varying levels of Kohn-Sham theory (PBE, B3LYP/6-31G*, LDA/STO-3G) and density fitting.
• Tested EMFT across diverse chemical problems, including those with complex partitions.

Main Results:

• EMFT demonstrated accurate and stable performance across a wide range of chemical problems.
• The method showed smooth convergence to high-level theory as the active subsystem size increased.
• EMFT performance was comparable or superior to ONIOM, especially for complex electronic structures and partitions.

Conclusions:

• EMFT is a simple, parameter-free, and effective approach for multiscale electronic structure.
• It offers significant advantages over ONIOM for challenging electronic structure problems.
• EMFT presents a promising new direction for multiscale modeling in computational chemistry.