The ORBS option is described in the section Getting Started (62.1).
The ATOMS option is described in the section Getting Started (62.1).
The value of the coefficient for the level shift used in the projection-based embedding calculations. The projection-based embedding method is in principle numerically exact in the limit of infinite MU, but in practice very large values of MU will lead to machine-precision errors. Typically, changing MU by 1-2 orders of magnitude relative to the default value should have only negligible impact on the energy.
This option is only relevant if the ATOMS card is present. Any MO having a Mulliken population on an atom specified by the ATOMS card that is greater than CHARGE_THRESHOLD is associated with the active subsystem.
If the ATOMS card is present, as an alternative to defining CHARGE_THRESHOLD, one can instead define N_ORBITALS. In this case, the N_ORBITALS MOs with the largest summed Mulliken population on the atoms specified by the ATOMS card are associated with the active system. This can be particularly useful for ensuring that the number of MOs in the active subsystem remains constant when calculating the potential energy surface of a reaction.
This card specifies the orbital record from which MOs corresponding to a HF-level or DFT-level calculation on the full system are read. These MOs should be localized prior to use by the embedding calculation, as shown in the following example:
If this card is not present, the last calculated HF or DFT MOs are localized using the Pipek-Mezey algorithm and then employed by the embedding calculation. Thus the preceding example is equivalent to:
Level of output. The value print=0 gives standard output. The value print=1 prints the atoms and molecular orbitals in the active region as well as the active and frozen electrons. The value print=2 prints the same output as print=1 along with the localized molecular orbital composition.
The DIRECTA option controls whether the calculation on the active subsystem is performed in integral-direct mode. By default, the calculation on the active subsystem is performed in integral-direct mode only if the DIRECT or GDIRECT directives have been used globally (see section 14.3). Setting DIRECTA=1 causes all calculations on the active subsystem to be performed in integral-direct mode, while setting DIRECTA=0 causes all calculations on the active subsystem to be performed without integral-direct mode, even if the DIRECT or GDIRECT directives have been previously used globally. Setting DIRECTA=0 is necessary when the calculation on the full system is performed using integral-direct mode and the calculation on the active subsystem is performed using a method for which integral-direct mode is not supported, such as CCSD(T). In this case, it is also necessary to use the FDIRECT directive instead of the DIRECT or GDIRECT directives, which forces Molpro to ignore potential method conflicts when enabling integral-direct mode.
An example of this process is provided below:
The EMBED,PROJ command modifies the core Hamiltonian and orbital occupation such that subsequent calculations are performed only on the active subsystem. The EMBED,PROJ,REMOVE command is required to revert to full system calculations after the EMBED,PROJ command. An example is provided below:
The SAVE_ORBITALS command is used in conjunction with REMOVE. If this card is present, orbitals that are generated on the active subsystem are not deleted when calling REMOVE. By default, REMOVE will delete all records and orbitals that were saved by the embedding code. To use the EMBED,PROJ code in input file loops, the SAVE_ORBITALS card must not be present.