Hierarchy Configuration Interaction: Combining Seniority Number and Excitation Degree

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Theoretical Chemistry

Background:

• Configuration Interaction (CI) methods are crucial for accurate electronic structure calculations.
• Excitation-based CI methods face challenges in scaling and efficiency for complex systems.
• A novel partitioning of Hilbert space is needed to improve computational efficiency and accuracy.

Purpose of the Study:

• To introduce and evaluate a novel partitioning of the Hilbert space called hierarchy configuration interaction (hCI).
• To assess the performance of hCI against traditional excitation-based CI methods for molecular dissociation.
• To investigate the impact of orbital optimization on hCI and excitation-based CI methods.

Main Methods:

• Developed hierarchy configuration interaction (hCI) by combining excitation degree and seniority number into a single hierarchy parameter.
• Surveyed the dissociation of multiple molecular systems using hCI and excitation-based CI.
• Investigated the effects of orbital optimization in conjunction with hCI and excitation-based CI.

Main Results:

• hCI generally matches or surpasses the performance of excitation-based CI methods for molecular dissociation.
• Higher orders of hCI and excitation-based CI with orbital optimization offer marginal improvements over calculations with Hartree-Fock orbitals.
• Orbital-optimized CI with single excitations provides a qualitatively correct description of single bond breaking at low computational cost.

Conclusions:

• hCI presents a promising alternative to excitation-based CI methods, offering comparable or superior performance.
• Orbital optimization is not always computationally beneficial for higher-order hCI and excitation-based CI.
• Minimally correlated orbital-optimized CI with single excitations is efficient for describing single bond breaking.