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 :)

MO guess from other program

  • jskennyu
  • Topic Author
  • Offline
  • New Member
  • New Member
More
6 years 2 months ago #488 by jskennyu
MO guess from other program was created by jskennyu
Dear Prof. Kallay,

I would like to learn the possibility to generate or read SCF initial guess from other program. We are doing a radical system and its SCF wavefunction is very difficult to converge. If I have already got a converged UHF wavefunction from other programs using orbital rotation, and have printed out the MO coefficients, is there a possibility to format those coefficients into the format that MRCC can read, then either feed it as the initial guess for UHF calculation using MRCC, or just proceed to the post-HF calculations?

I checked the MOCOEF file generated after MRCC calculation that could be used with scfiguess=mo in the next calculation, however, that file seems to be a binary file and I don't know how to deal with it.

Thank you in advance for your kind response.

Sincerely,
Jen-Shiang
(Kenny)

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

  • kallay
  • Offline
  • Administrator
  • Administrator
  • Mihaly Kallay
More
6 years 2 months ago #489 by kallay
Replied by kallay on topic MO guess from other program
Dear Jen-Shiang,
It is a complicated task. First, you should take into account that the order of AOs in the two programs are probably different, and you should permute the MO coefficients accordingly. To get the ordering used in mrcc set verbosity=3, and you will find the order at the beginning of the output. Second, you should write a routine that saves the MO coefficients in the required format. For that purpose you should call the wrtmo from file combin.f. Third, you should run the program with scfiguess=mo. You cannot skip the SCF since the Fock-matrix must be generated.

Best regards,
Mihaly Kallay

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

  • jskennyu
  • Topic Author
  • Offline
  • New Member
  • New Member
More
6 years 2 months ago #490 by jskennyu
Replied by jskennyu on topic MO guess from other program
Dear Prof. Kallay,

Thank you for your explanation. I checked the MOLDEN file and it might be easier to start over there.

There is another question that might or might not related to this. I also saw a MOLDEN.perm file after SCF calculation and it seems that orbital permutation might be possible in MRCC. My question is, can I just permute the orbitals generated from a high-spin calculation then do orbital rotation just like that in MOLPRO, which might be able to converge the UHF calculation for the radicals?

Thanks a lot for your advise.

Jen-Shiang

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

More
6 years 2 months ago #492 by SBehnle
Replied by SBehnle on topic MO guess from other program
When you're using MOLPRO anyway, why don't you use the MOLPRO->MRCC interface?

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

  • kallay
  • Offline
  • Administrator
  • Administrator
  • Mihaly Kallay
More
6 years 2 months ago #493 by kallay
Replied by kallay on topic MO guess from other program
Dear Jen-Shiang,
The MOLDEN.perm file contains the permutation which converts the MO coefficients to the order required by molden.
I don't know exactly what molpro does, but as explained below, you can easily read the MO coefficients and interchange the MOs or rotate them.

Best regards,
Mihaly Kallay
The following user(s) said Thank You: jskennyu

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

Time to create page: 0.041 seconds
Powered by Kunena Forum