Dual Basis Set Approach for Density Functional and Wave Function Embedding Schemes.

Area of Science:

• Computational chemistry
• Quantum chemistry
• Electronic structure theory

Background:

• Computational expenses of Hartree-Fock, Kohn-Sham, and wave function-based correlation methods are significant.
• Embedding calculations are crucial for studying complex chemical systems but are computationally demanding.

Purpose of the Study:

• To propose a dual basis (DB) approach for reducing computational costs in electronic structure calculations.
• To adapt and apply the DB approach specifically for embedding calculations.

Main Methods:

• The proposed dual basis (DB) approach is related to existing DB approximations but tailored for embedding.
• The DB approach is integrated with projector-based embedding theory, Huzinaga-equation, and multilevel local correlation methods.
• Performance is evaluated through benchmark calculations comparing DB methods with mixed-basis approximations.

Main Results:

• The DB approach significantly speeds up embedding calculations when appropriate basis sets are used.
• The DB approach demonstrates superiority over the mixed-basis approximation for chemical processes with significant electronic structure changes.
• Combining DB and mixed-basis approximations offers the most efficient strategy for accelerating embedding calculations.

Conclusions:

• The dual basis (DB) approach is an effective strategy for accelerating computational chemistry calculations, particularly in embedding.
• Integrating DB with mixed-basis approximations overcomes limitations and provides the most efficient acceleration for embedding calculations.