next up previous contents index
Next: Prerequisites Up: DOK Previous: Restrictions:   Contents   Index

CCSD, CCSD(F12) and CCSD(T) calculations


The ricc2 module includes presently since 2010 also an implementation of the full coupled-cluster singles-and-doubles method CCSD, the explicitly correlated CCSD(F12) and CCSD(F12*) variants. CCSD and its correlated variants can be combined with a perturbative correction for connected triples, CCSD(T).% latex2html id marker 52104
\setcounter{footnote}{1}\fnsymbol{footnote}Presently the implementation is restricted to ground state energies. Closed-shell (RHF), unrestricted (UHF) or single determinant restricted (ROHF) open-shell reference wavefunctions can be used, but no excitation energies, gradients or properties are (yet) available for these wavefunction models.

Further limitations:

no MPI parallelization
, calculations at these levels can presently only carried out on a single compute node, only the OpenMP (see Sec. 3.2.2) parallelization is available for calculations at the CCSD and CCSD(T) level.

Please note that calculations with CCSD and methods beyond CCSD require considerably more disc space and core memory than MP2 or CC2 calculations. (See section below for more details and recommendations.)