Automatic purpose-driven basis set truncation for time-dependent Hartree-Fock and density-functional theory.

Area of Science:

• Computational Chemistry
• Theoretical Chemistry
• Quantum Mechanics

Background:

• Real-time time-dependent density-functional theory (RT-TDDFT) and linear response time-dependent density-functional theory (LR-TDDFT) are crucial for simulating electronic spectra.
• Current TDDFT calculations often use basis sets optimized for ground-state properties, potentially limiting accuracy and efficiency for excited-state simulations.

Purpose of the Study:

• To develop a systematic and robust scheme for truncating atomic orbital (AO) basis sets in TDDFT and TD Hartree-Fock (TDHF) calculations.
• To enhance the computational efficiency of electronic spectra simulations without significant loss of accuracy.

Main Methods:

• Proposed a novel scheme to systematically truncate AO basis sets for TDDFT and TDHF.
• Tested the truncated basis sets using both LR-TDDFT and RT-TDDFT, as well as RT-TDHF methods.
• The procedure involves a brief RT calculation and a simple modification to the basis set file.

Main Results:

• Achieved computational acceleration of up to an order of magnitude in electronic spectra simulations.
• Observed shifts in excitation energies generally within 0.2 eV compared to full basis set calculations.
• Demonstrated the applicability of the method across different TDDFT/TDHF approaches and quantum chemistry packages.

Conclusions:

• The proposed AO basis set truncation scheme offers a significant reduction in computational cost for TDDFT and TDHF calculations.
• This method provides a practical approach to accelerate electronic spectra simulations while maintaining high accuracy.
• The findings offer insights into basis set design for specific electronic excitation calculations.