SV: DIRAC KR-MCSCF problem

[Hans Jørgen Aagaard Jensen]

Dear Ruairi,

I have now accepted your request to become a member of dirac-users. Unfortunately, we have not been able to find a way to get a notification each time someone wants to become a member of dirac-users, so it can take some time before we discover the request ...

Wrt your problem, you have found a bug. Apparently, nobody has tried to run KRMCSCF in linear mode with an odd number of electrons before!  As a work-around you only need to do the following very small edit :

      

and recompile. (I have tested your input, it works.)

Best, Hans Jørgen.

Fra: Ruairi Graham <rgra...@qub.ac.uk>
Dato: mandag, 10. november 2025 kl. 15.32
Til: Hans Jørgen Aagaard Jensen <h...@sdu.dk>, timo....@irsamc.ups-tlse.fr <timo....@irsamc.ups-tlse.fr>
Emne: DIRAC KR-MCSCF problem

Hi,

For a quick background I am relatively new to computational chemistry. I am using the DIRAC 25 software for my master’s project in university (I go to Queen’s University Belfast), and my aim is to run KR-MRCI(SD) calculation on a linear open shell hetero-nuclear diatomic molecule across many bond lengths. I have applied to access the google Dirac-users group however, I have not yet been accepted. I feel like the problem could potentially be an easy fix I just find myself stuck trying to find a solution. 

I will send my output file here if you would like to inspect it. For reference I am attempting to do HF and MCSCF in the same input. Would it be better to do them separately? And would you know why I keep getting a symmetry error in my calculation?

Thanks,

Ruairi