Auxiliary Density Matrix Methods for Hartree-Fock Exchange Calculations

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Hartree-Fock exchange (HFX) calculations are computationally intensive for large systems.
• High-quality basis sets further increase the computational cost of HFX.
• Efficient methods are needed to incorporate HFX in large-scale simulations.

Purpose of the Study:

• To develop and validate an efficient method for calculating Hartree-Fock exchange (HFX).
• To demonstrate the accuracy and performance of Auxiliary Density Matrix Methods (ADMM) for HFX.
• To investigate the impact of HFX on the structure of liquid water and solvated proteins.

Main Methods:

• Development of several schemes to derive an auxiliary density matrix from a high-quality density matrix.
• Incorporation of a correction based on generalized gradient approximations for HFX accuracy.
• Application of ADMM in Born-Oppenheimer molecular dynamics simulations for liquid water (64 molecules, 300 ps).
• Performance evaluation on a solvated protein (Rubredoxin) system.

Main Results:

• Auxiliary Density Matrix Methods (ADMM) achieve excellent performance and good accuracy for HFX calculations.
• ADMM integrates seamlessly into existing HFX codes and supports linear scaling implementations.
• Simulations of liquid water show the effect of HFX on its structure.
• ADMM demonstrated a 20-fold speedup over standard HFX for a solvated protein calculation.

Conclusions:

• ADMM offers a computationally efficient and accurate approach for including HFX in large system calculations.
• The method is suitable for large-scale quantum chemistry simulations, including molecular dynamics.
• ADMM significantly reduces the computational cost of HFX, enabling studies previously intractable.