Second-Order Orbital Optimization with Large Active Spaces Using Adaptive Sampling Configuration Interaction (ASCI)...

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Theoretical Chemistry

Background:

• Selected configuration interaction (SCI) methods offer accurate electronic structure calculations.
• Embedding SCI in a mean field yields molecular orbitals comparable to complete active space self-consistent field (CASSCF) methods.

Purpose of the Study:

• Implement analytical gradient theory for the single-state adaptive sampling CI (ASCI) SCF method.
• Enable molecular geometry optimization using ASCI-SCF.

Main Methods:

• Developed analytical gradient theory for ASCI-SCF.
• Combined augmented Hessian (AH) and Werner-Meyer-Knowles (WMK) second-order orbital optimization with ASCI-SCF for tight convergence.
• Tested algorithms for orbital and geometry optimizations.

Main Results:

• Achieved accurate molecular geometry optimization for large active spaces using approximate analytical gradients.
• Demonstrated successful geometry optimizations for polyacenes and periacenes.
• Analyzed the geometric dependence of singlet ASCI wave function characteristics.

Conclusions:

• The implemented analytical gradient theory for ASCI-SCF is effective for molecular geometry optimization.
• The combination of AH and WMK methods ensures the tight convergence required for accurate gradients.
• This work facilitates the study of larger and more complex molecular systems in quantum chemistry.