Orbital-optimized linearized coupled-cluster doubles with density-fitting and Cholesky decomposition approximations: an efficient implementation

Abstract

An efficient implementation of the orbital-optimized linearized coupled-cluster double method with the density-fitting (DF-OLCCD) and Cholesky decomposition (CD-OLCCD) approximations is presented. The DF-OLCCD and CD-OLCCD methods are applied to a set of alkanes to compare the computational cost with the conventional orbital-optimized linearized coupled-cluster doubles (OLCCD) [U. Bozkaya and C. D. Sherrill, J. Chem. Phys., 2013, 139, 054104]. Our results demonstrate that the DF-OLCCD method provides substantially lower computational costs than OLCCD, and there are more than 9-fold reductions in the computational time for the largest member of the alkane set (C₈H₁₈). For barrier heights of hydrogen transfer reaction energies, the DF-OLCCD method again exhibits a substantially better performance than DF-LCCD, providing a mean absolute error of 0.9 kcal mol⁻¹, which is 7 times lower than that of DF-LCCD (6.2 kcal mol⁻¹), and compared to MP2 (9.6 kcal mol⁻¹) there is a more than 10-fold reduction in errors. Furthermore, the MAE value of DF-OLCCD is also lower than that of CCSD (1.2 kcal mol⁻¹). For open-shell noncovalent interactions, the performance of DF-OLCCD is significantly better than that of MP2, DF-LCCD, and CCSD. Overall, the present application results indicate that the DF-OLCCD and CD-OLCCD methods are very promising for challenging open-shell systems as well as closed-shell molecular systems.