PC GAMESS-related discussion club

Alex Granovsky
gran@classic.chem.msu.su

>First of all, I'm a very novice user of PC-GAMESS; I just started using a few days ago.

welcome to the club.


>Yesterday, I was browsing the archive and I found out that if you want to get the singlet and triplet spin contributions to the MP2 correlation energy, you need to do an MP3 calculation. This was in a message from the end of 2006. Is this still a valid statement or this was changed in the code since then?

The situation was not changed since then, however it seems
I overlooked another, more cheaper way how to get them.
More precisely, you can run UHF (note not RHF!) MP2 calculations
for singlet state with $mp2 method=3 $end. Then you can easily
extract singlet and triplet pair contributions to E(2) by
multiplying printed value of

  alpha alpha alpha alpha

contribution to E(2) by the factor of three.

Method=3 is not very appropriate for large scale jobs, however,
if there is a considerable interest to get the separate contributions
of singlet and triplet pairs to E(2), we can consider
implementation of this feature in the standard closed-shell
MP2 code.

The reason why this is not implemented is that these contributions
are seldom used separately, unless you are going to scale them by
the different factors.


>It seems to me that the analytical MP2 frequency calculation hasn't been implemented yet, has it?

Yes they have not yet implemented.

>My third and definitely lame question is: the web page posting is the only way to send messages to this discussion forum?

Yes this is the only way.


Best regards,
Alex Granovsky