Page 1 of 1

Characterising Excited states in TDDFT Calculation

Posted: 04 Sep 2019, 14:12
by FarNHight
This is very much a Dalton newbie's question, which I fully expect has a straightforward answer. However , I've looked and looked ... but figure I'm missing the obvious.

How do I characterise the excited states returned during **RESPONSE | *QUADRATIC calculations?

By that I mean I want to try to identify/classify them as e.g. 3s Rydberg, pi* Valence etc etc.
I'm not using symmetry (so there is no symmetry classification), so the only distinguishing feature (apart from the requested property moments etc) that I can see output is the excitation energy.

In Gaussian, TDDFT output will show the principal orbital excitations, population analysis can be done for the excited state densities, and <R**2> values can be examined, but I can't see how this kind of characterisation can be achieved in Dalton.

Re: Characterising Excited states in TDDFT Calculation

Posted: 06 Sep 2019, 23:18
by xiongyan21
GAMESS can do this, where the criterion is based on an published article.

Re: Characterising Excited states in TDDFT Calculation

Posted: 07 Sep 2019, 16:42
by xiongyan21
GAMESS can distinguish Rydberg from charge transfer/ valence, where the criterion is based on an published article.

Re: Characterising Excited states in TDDFT Calculation

Posted: 10 Sep 2019, 10:23
by magnus
You can increase print level by adding, e.g.:

Code: Select all

.PRINT
4
under the relevant section in the Dalton input file. I'm not sure what it will give in the *QUADRATIC section, but for excitation energies through the response module, i.e. **RESPONSE & *LINEAR, this will give you orbital contributions and some overlap diagnostics.