Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

Re: MCSCF state tracking question

Alex Granovsky

Dear Thomas,

you need to adjust NSTATE in CI procedure to capture additional
CI states and set NTRACK to some value greater or equal than 3
and less or equal to the overall number of captured CI states
of desired multiplicity.

By default, MCSCF state tracking works using NSTATE previously remapped
states (the initial map is identity). Therefore, if state #3 was at some point remapped
to state #5, it still will be tracked even with NSTATE=3.

However, the tracking itself is cheap, so why not to include
some additional states into tracking procedure?


On Mon Jan 31 '11 1:14am, Thomas wrote
>Dear fellow Firefly-ers,

>I have a question about the MCSCF state tracking function. I would like to perform PES scans for a certain molecule on the S0, S1 and S2 surfaces using CASSCF with state averaging over the lowest three roots. However, I expect that root switching with higher lying roots might take place. How should I set the $MCSCF NTRACK value? Should I specify NTRACK=3 in order to track only these three lowest roots, or is it necessary to also include the higher roots that could be involved in switching (and thus specify NTRACK=4 or higher)? Thanks in advance for your help.
>Kind regards,

[ This message was edited on Mon Jan 31 '11 at 1:28pm by the author ]

[ Previous ] [ Next ] [ Index ]           Mon Jan 31 '11 1:28pm
[ Reply ] [ Edit ] [ Delete ]           This message read 1109 times