Near-Exact CASSCF-Level Geometry Optimization with a Large Active Space using Adaptive Sampling Configuration...

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Theoretical Chemistry

Background:

• Accurate electron correlation is crucial in quantum chemistry but computationally expensive.
• Full Configuration Interaction (FCI) provides exact correlation but is intractable for large systems.
• Selected CI (SCI) offers a computationally feasible approximation to FCI.

Purpose of the Study:

• To develop and implement an efficient computational method for geometry optimization and electronic structure calculations.
• To improve the accuracy of Selected CI (SCI) approximations by incorporating perturbation theory.
• To apply the developed method to complex polycyclic aromatic hydrocarbons and radicals.

Main Methods:

• Adaptive Sampling Configuration Interaction (ASCI) combined with Self-Consistent Field (SCF) calculations (ASCI-SCF).
• Incorporation of second-order perturbation theory (PT2) corrections to ASCI-SCF.
• Implementation of an analytical nuclear gradient algorithm using the Z-vector formalism for ASCI-PT2.
• Extrapolation techniques to recover remaining configuration contributions.

Main Results:

• Optimized geometries and unpaired electron counts for phenalenyl radicals and anthracene achieved near CASSCF accuracy.
• Demonstrated the utility of ASCI-SCF-PT2 for optimizing equilibrium geometries of six-ring-fused polycyclic aromatic hydrocarbons and 4-periacene.
• The method provides accurate electronic structures with reduced computational cost compared to FCI.

Conclusions:

• ASCI-SCF-PT2 offers a computationally efficient and accurate approach for electronic structure calculations.
• The developed analytical gradient method enables reliable geometry optimizations for complex organic molecules.
• This method advances the capability to study large polycyclic aromatic hydrocarbons in quantum chemistry.