Orbital-optimized linearized coupled-cluster doubles with density-fitting and Cholesky decomposition approximations:...

Area of Science:

• Computational chemistry
• Quantum chemistry
• Theoretical chemistry

Background:

• Coupled-cluster methods are essential for accurate electronic structure calculations.
• Orbital-optimized methods improve efficiency and accuracy.
• Density-fitting and Cholesky decomposition approximations reduce computational cost.

Purpose of the Study:

• To present an efficient implementation of density-fitting (DF-OLCCD) and Cholesky decomposition (CD-OLCCD) approximations for the orbital-optimized linearized coupled-cluster double method.
• To compare the computational cost and accuracy of DF-OLCCD and CD-OLCCD with conventional OLCCD, DF-LCCD, MP2, and CCSD methods.

Main Methods:

• Implementation of DF-OLCCD and CD-OLCCD methods.
• Application to a set of alkanes for computational cost comparison.
• Evaluation of barrier heights for hydrogen transfer reactions.
• Assessment of performance for open-shell noncovalent interactions.

Main Results:

• DF-OLCCD and CD-OLCCD show substantially lower computational costs than OLCCD, with over 9-fold reduction for C8H18.
• DF-OLCCD achieves a mean absolute error of 0.9 kcal mol(-1) for hydrogen transfer barrier heights, significantly outperforming DF-LCCD and MP2.
• DF-OLCCD demonstrates superior performance for open-shell noncovalent interactions compared to MP2, DF-LCCD, and CCSD.

Conclusions:

• DF-OLCCD and CD-OLCCD are highly efficient and accurate methods for electronic structure calculations.
• These methods show great promise for studying challenging open-shell and closed-shell molecular systems.
• The implemented approximations offer significant computational savings without compromising accuracy.