Analytical gradients of the state-average complete active space self-consistent field method with density fitting.

Area of Science:

• Computational chemistry
• Quantum chemistry
• Molecular modeling

Background:

• State-averaged complete active space self-consistent field (SA-CASSCF) is a crucial method for studying electronic structures of molecules with strong electron correlation.
• Calculating gradients for SA-CASSCF is computationally intensive, limiting its application to larger systems.
• Existing methods face challenges in scaling and computational prefactors, hindering efficiency.

Purpose of the Study:

• To develop and present an efficient implementation of state-averaged complete active space self-consistent field (SA-CASSCF) gradients using density fitting (DF).
• To demonstrate the performance and scalability of the new algorithm on various molecular systems.
• To reduce the computational cost of SA-CASSCF gradient calculations without compromising accuracy.

Main Methods:

• Implementation of density fitting (DF) approximation within the SA-CASSCF gradient calculation framework.
• Utilizing linear response theory for efficient calculation of gradients.
• Testing the algorithm on a range of molecular systems, including a large iron-Heme b complex.

Main Results:

• The density fitting approach significantly reduces both the computational scaling and prefactors involved in SA-CASSCF gradient calculations.
• Demonstrated timing reductions in linear response calculations and gradient formation compared to conventional methods.
• Successfully computed the largest SA-CASSCF gradient to date for a 79-atom iron-Heme b complex.
• Achieved a typical reduction in overall geometry optimization cost by more than one order of magnitude.
• Confirmed negligible loss in accuracy compared to traditional methods.

Conclusions:

• The presented density fitting implementation offers a substantial efficiency improvement for SA-CASSCF gradient calculations.
• This method enables the study of larger and more complex molecular systems previously inaccessible.
• The significant reduction in computational cost makes geometry optimizations more feasible and efficient.