8.9 `cpf`

The CPF program generates SDCI, CPF[42], MCPF[43] or ACPF[44], wavefunctions from one reference configuration.

The CPF program is a modification to a CPF program written by P. E. M. Siegbahn and M. Blomberg (Institute of Physics, Stockholm University, Sweden).

The program is based on the Direct CI method[45], with the coupling coefficients generated by the Graphical Unitary Group Approach[46]–[47] (See program description for GUGA). CPF generates natural orbitals that can be fed into the property program to evaluate certain one electron properties. Also, the natural orbitals can be used for Iterative Natural Orbital calculations.

8.9.0.1 Orbital subspaces

The orbital space is divided into the following subspaces: Frozen, Inactive, Active, Secondary, and Deleted orbitals. Within each symmetry type, they follow this order. Their meaning is the same as explained in the GUGA and MOTRA sections, except that, in this case, there is only a single reference configuration. Therefore, the active orbitals in this case are usually only open shells, if any. Since explicit handling of orbitals is taken care of at the integral transformation step, program MOTRA, orbital spaces are not specified in the input, except when orbitals are frozen or deleted by the CPF program, rather than by MOTRA (which should normally be avoided).

8.9.1 Dependencies

The CPF program needs the coupling coefficients generated by the program GUGA and the transformed one and two electron integrals from the program MOTRA .

8.9.2 Files

8.9.2.1 Input files

The CPF program need the coupling coefficients generated by GUGA and the transformed integrals from MOTRA.

CPF will use the following input files: ONEINT, RUNFILE, CIGUGA, TRAINT, TRAONE (for more information see ). and CPFVECT (for restarted calculations).

8.9.2.2 Output files

CPF generates an two output files:

File Contents
CPFORB The natural orbitals from the CPF functional.
CPFVECT The CI expansion coefficients. These may be used for restarting an unconverged calculation.

8.9.3 Input

8.9.4 `CPF`

This section describes the input to the CPF program in the MOLCAS program system. The input for each module is preceded by its name like:
&CPF

8.9.4.1 Optional keywords

Keyword	Meaning
TITLe	Followed by a title line
SDCI	Specifies that a SDCI calculation is to be performed. No additional input is required. Only one of the choices SDCI, CPF, MCPF or ACPF should be chosen.
CPF	Specifies that a CPF calculation is to be performed. Only one of the choices SDCI, CPF, MCPF or ACPF should be chosen.
MCPF	Specifies that a Modified CPF calculation is to be performed. This option is in fact the default choice. Only one of the choices SDCI, CPF, MCPF or ACPF should be chosen.
ACPF	Specifies that an Average CPF calculation is to be performed. Only one of the choices SDCI, CPF, MCPF or ACPF should be chosen.
RESTart	Restart the calculation from a previous calculation.
THRPr	Threshold for printout of the wavefunction. All configurations with a coefficient greater than this threshold are printed in the final printout. The default is 0.05.
ECONvergence	Energy convergence threshold. The update procedure is repeated until the energy difference between the last two iterations is less than this threshold. The default is 1.0e-8.
PRINt	Print level of the program. Default is 5.
MAXIterations	Maximum number of iterations in the update procedure. Default 20. The maximum value of this parameter is 75.
FROZen	Specify the number of orbitals to be frozen in addition to the orbitals frozen in the integral transformation. Default is 0 in all symmetries.
DELEted	Specify the number of orbitals to be deleted in addition to the orbitals deleted in the integral transformation. Default is 0 in all symmetries.
LOW	Specifies that this is a low spin case, i.e. the spin is less than the maximum possible with the number of open shells in the calculation. See Refs. [42,43]. This requires special considerations.
MAXPulay	Maximum number of iterations in the initial stage. After that, DIIS extrapolation will be used. Default is 6.
LEVShift	Levelshift in the update procedure. Default is 0.3.