Ceilidh ... Re^5: Difference between excited state and ground state density

Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly

We are NATO-free zone

Re^5: Difference between excited state and ground state density

Dear Michal,

Yes, there will be three $cube groups in the punch file, first for KS
density, second for unrelaxed TDDFT density for the state of interest,
and third for relaxed TDDFT density. Each $cube group will contain
a description for which density it was computed.

All the best,
Alex

On Thu Jun 16 '16 11:09am, Michal Krompiec wrote
------------------------------------------------
>Dear Alex,
>Thanks, that's clear. So is it possible to generate a CUBE with relaxed TDDFT density itself? If I specify the state of interest in $tddft and $dft tdprp=.t. $end $ELDENS IEDEN=1 $END $CUBE CUBE=.T. $END, will the relaxed TDDFT density be punched?
>Best regards,
>Michal

>On Wed Jun 15 '16 9:14pm, Alex Granovsky wrote
>----------------------------------------------
>>Dear Michal,

>>unfortunately this still should produce the difference between KS
>>density and unrelaxed TDDFT density. To obtain relaxed TDDFT density,
>>it is necessary to perform additional computations besides TDDFT
>>energy calculations.

>>Kind regards,
>>Alex
>>
>>
>>
>>
>>On Mon Jun 13 '16 3:53pm, Michal Krompiec wrote
>>-----------------------------------------------
>>>Dear Alex,
>>>Thanks. So if I generate the difference density in a TDDFT restart job (restarting from a converged TDDFT calculation, using $TDDFT RDTDVC=.t. $END ), this should be the difference between the input TDDFT density and the KS ground state density, right?
>>>Best regards,
>>>Michal

>>>On Sun Jun 12 '16 10:33am, Alex Granovsky wrote
>>>-----------------------------------------------
>>>>Dear Michal,

>>>>I'm sorry for delayed reply.

>>>>The exact answer to your question depends on computational method.
>>>>
>>>>
>>>>The general comment is that if you add the following option to your
>>>>input file,

>>>>

 $eldens diffd=.t. $end

>>>>(This option is documented here:

>>>>then Firefly will automatically compute and punch the differences
>>>>between two consecutive densities. This may require setting proper
>>>>iroot or istate keyword depending on particular computational method.

>>>>Unfortunately this does not work well for TDDFT as normally one is
>>>>interested in difference of relaxed density and KS density while
>>>>Firefly would computes two differences:
>>>>one between non-relaxed TDDFT density and KS densities and second
>>>>between relaxed and non-relaxed TDDFT densities. To make this work
>>>>one needs to specify proper nstate and istate in $tddft group and
>>>>adds tdprp=.t. to $dft group.

>>>>As to second part of your question, I'd suggest you to use gOpenMol
>>>>to manipulate cubes and plt files. First, gOpenMol distribution
>>>>contains gamess2plt utility that was developed to extract $CUBE
>>>>information from the PC GAMESS/Firefly punch files and convert them to
>>>>plt files. Second, there is an utility called contman that can be used
>>>>to add or subtract two plt files. Finally, plt files can be visualised
>>>>by either gOpenMol or by (the better way) Chimera and probably by some
>>>>other visualizer. In addition, powerful general purpose vilualizers
>>>>like VisIt and ParaView accepts cube files and can manipulate them.
>>>>
>>>>
>>>>All the best,
>>>>Alex
>>>>
>>>>
>>>>
>>>>
>>>>On Wed Jun 8 '16 4:09pm, Michal Krompiec wrote
>>>>----------------------------------------------
>>>>>Hi, how can I generate a CUBE file for the difference in electron density between a chosen excited state and the ground state? If it is not possible directly, can I subtract the respective CUBEs with some external software? (other than ChemCraft)

Thu Jun 16 '16 2:10pm

This message read 509 times