Coupled-Cluster x2c(eamf) for first order properties in separate steps

[juan...@gmail.com]

Dear Dirac users,

I'm trying to perform a coupled-cluster x2c(eamf) calculation in separate steps and yet I couldn't do it.

I've tried following the steps proposed in the manual (https://www.diracprogram.org/doc/release-23/tutorials/restart/coupled_cluster.html) with no success.

Then I tried copying and putting more files from scratch in each step. I'm attaching the outputs. In the ones labeled by steps, I got/put

• On step 1 (get files):
  AOMOMAT X2CMAT eamfX2C_F4c xamfi-2c soc-contributions CHECKPOINT.h5 OPERATORS.h5
• On step 2 (put files):
  CHECKPOINT.h5 amfPCEC.h5 OPERATORS.h5
• On step 2 (get files):
  CHECKPOINT.h5 MDCINT MRCONEE MDPROP
• On step 3 (put files):
  CHECKPOINT.h5 amfPCEC.h5 OPERATORS.h5 MDCINT MRCONEE MDPROP

But as you can see, I get different final values for EFG when performing the calculation in one step. 

Thanks in advance,
Juan J. Aucar

[juan...@gmail.com]

Dear Dirac experts,

I'm facing again a problem similar to the one I pointed out in this regard.
I'd like to perform x2c(eamf) calculations for first order properties in three diferent steps:

1) Perform the atomic calculations
2)  Get the WF and perform the corresponding 4c->2c picture change transformations
3) Calculate first order properties (let's say at CC level)

The problem is that when I try the third step I find no way to avoid the 4c->2c picture change transformations (that I already got in the second step), and they imply much time for me, as I want to do an analysis on the size of the active space (so that I need to repeat step 3 several times).

I can see that on src/dirac/dirone.F there is a call to " x2c_main()" that it's always done, wheter I put or not files as X2CMAT in the scratch folder. I also tried with the keyword .skippc under *X2C. 

Any help would be appreaciated

Best regards,

Juan J.