Can a Molpro HF/CASSCF be used to start MRCC EOM?

Hey there,
For an open-shell system, can I start from unconventional ground state orbitals in Molpro (say from a CASSCF), dump reference wave function integrals to MRCC format, then start doing EOM-CC energies/properties/optimizations in the stand alone MRCC?

I feel like this works, and has been implied before in the forum, but I am wondering if anyone has accomplished it in practice. I can't get it to work...

What keywords would be necessary in the MRCC input file? Do I specify the geometry? Will slight numerical issues in the Molpro vs MRCC geometry make everything go haywire?

You can perform EOM-CC energy and transition moment calculations with MCSCF orbitals taken from Molpro. You only need to set the number of states in the Molpro input using the EOM_NSTATES keyword (see Molpro manual). Geometry optimization is not possible.

Thanks Mihály.

One further question: Can you show a real life example of using rest=3 to find a transition energy from ground state to higher root #? I can't follow the manual verbiage below:

The serial number of the roots must be specified in the subsequent line as
< n1 >< n2 >< n3 > . . .
where < ni > is the serial number of the states. The number
of states given here must be equal to nstate or nsing + ntrip.

Kind of at the end of my rope... wondering if there is any way to get a DKH transformed two-e integral file into MRCC and get EOM energies and transition moments.

If I do EOM-CCSD from Molpro with dens=5 or dens=6, I get the error:
Property integrals are missing!
Unable to calculate transition moments!

But this error occurs even before the ground state CCSD calculation begins, so of course there will be no property integrals?

Please see example 2 for keyword rest in the manual.

The problem is that the Molpro interface does not save the dipole moment integrals, so it is not possible to calculate transition moments.