The convergence of coupled-cluster equations for several cases, CCD, CCSD, CCSDT-n and the full CCSDT is investigated. Comparisons are made between the reduced linear equation (RLE) method for accelerating convergence and simple geometric extrapolation techniques, and between energy and wavefunction
Convergence of a matrix-oriented coupled cluster (doubles) method
β Scribed by Clifford E. Dykstra; Joseph D. Augspurger
- Publisher
- Elsevier Science
- Year
- 1988
- Tongue
- English
- Weight
- 255 KB
- Volume
- 145
- Category
- Article
- ISSN
- 0009-2614
No coin nor oath required. For personal study only.
β¦ Synopsis
The convergence of an iterative matrix-oriented coupled cluster approach is investigated and compared with a more recent implementation. Where the reference wavefunction dominates, the special features of the matrix-oriented implementation result in rapid convergence.
π SIMILAR VOLUMES
With the help of G. Scuseria and H.F. Schaefer III, we have found an improper factor of 2 in non-linear HN T: contributions to T, amplitudes only for the fill CCSDT results
The approximate coupled cluster doubles (ACCD) approximation is tested, both at the infinite-order level and at fourth order of many-body perturbation theory. It is found that in the presence of strong pair coupling, i.e. for most molecules, there is a substantial deviation between the ACCD and the
The direct mversion of the iterative subspace (DIIS) method is implemented into the closed-shell coupled-cluster single-and double-excitation (CCSD! model to improve the convergence of the coupled non-linear CCSD equations. As with self-consistent field and gradient methods the DIE method proves to