× If you have problems during the execution of MRCC, please attach the output with an adequate description of your case as well as the followings:
  • the way mrcc was invoked
  • the way build.mrcc was invoked
  • the output of build.mrcc
  • compiler version (for example: ifort -V, gfortran -v)
  • blas/lapack versions
  • as well as gcc and glibc versions

This information really helps us during troubleshooting :)

Not recognizing symmetry

More
5 months 1 week ago #989 by bkwx97
Dear developers. I am using the Feb 2020 binary for this computation. I'm attempting to obtain an LCCSD(T) energy. I'm finding that MRCC isn't recognizing the Cs symmetry. I have even lowered the symmetry tolerance and it is saying it is C3 and the computational group is C1.


Attached is my input.
Attachments:

Please Log in or Create an account to join the conversation.

More
5 months 1 week ago #990 by kallay
Replied by kallay on topic Not recognizing symmetry
Something is wrong with your input file as there are only 61 atoms in the geometry specification instead of 62.

Best regards,
Mihaly Kallay

Please Log in or Create an account to join the conversation.

More
5 months 1 week ago #991 by bkwx97
Replied by bkwx97 on topic Not recognizing symmetry
I just saw that. That was a typo. I did rerun it with 61 and it still recognizes it as c3 and the computational group as c1.

Please Log in or Create an account to join the conversation.

More
5 months 1 week ago #992 by nagypeter
Replied by nagypeter on topic Not recognizing symmetry
Dear bkwx97,

your structure does not appear to be sufficiently symmetric, other programs cannot recognize its full C3v symmetry either.

After symmetrizing your structure to C3v, the proper Cs computational point group is recognized by MRCC. Thus you should also symmetrize your structure and reoptimize it keeping C3v as much as possible.

Generally, increasing gtol to such extreme values is not a good idea, as enforcing symmerty constraints for a non-symmetric structure can introduce a sizable error.

Note, that the SCF and orbital localization codes will not constrain the orbitals according to C3v symmetry (only to Cs in SCF), thus you need tight thresholds for scf to have a chance at symmetric local orbitals. Because of that localcorrsymm is not a back-box feature and there is no guarantee that the LNO-CC code will get sufficiently symmetric orbitals.
You can play around with the lccsymtol keyword which sets the tolerance for symmetry recognition for the local MOs in the LNO-CC calculation, but again, the introduced local error on enforcing insufficient symmetry will increase with 10^-lccsymtol. lccsymtol=cctol=5 is the default.

Using the following worked for me for the C3v symmetrized structure:
lccsymtol=4
itol=12
scfdtol=8

Best wishes,
Peter

Please Log in or Create an account to join the conversation.

Time to create page: 0.022 seconds
Powered by Kunena Forum