For more than a few excited states, you should probably find the potential energy surfaces (PES) of the molecule in question using some method that includes electronic correlation( MP2 might work but coupled cluster would be better) . For each excited state, the optimized structure will correspond to the minimum energy for a particular state's PES. You will need to calculate the zero point energy vibrational correction. Note that some states will not be bound and thus will not have a optimized structure ( The optimized structure would correspond to separated fragments of the molecule).

Thanks for you comments.

Actually i am interested to calculate vibrationally resolved electronic specta using TD-DFT.

Hello Mohammad,

your state of interest is determined by the root argument in TD=(Root=n) where n revers to the specific state. There is an example on http://www.gaussian.com/g_tech/g_ur/k_scrf.htm in the example section for "Emission (Fluorescence) from First Excited State (n→π*) of Acetaldehyde" which you can modify easily for your needs.

Best of luck

Thanks Marcel

if you are interested in calculating the vibrationally resolved electronic spectra you have to calculate frequencies when you are optimizing the structure

opt freq CIS(nstates=20) or for opt freq td=(direct, nstates=20) if you have got the negative frequency then the obtained structure are not a stable structure.

Hi... look for the gaussian documentation of obtaining Franck-Condon factors. From that you can construct vibrationally resolved optical spectra. Its mentioned in the "Freq" keywords section: "ELECTRONIC EXCITATION ANALYSIS OPTIONS". There is, to my knowledge, no easy way to obtain operator expectation values bewtween different electronic TDDFT states. But a detailed study of the IOps may reveal some prospects.

Dear Mohammad,

Have you find the answer for yourself. I am having the same question. Do you mind to share your experience.

I have a closed-shell system and look to optimize the first excited triplet state. I was wondering between three aproach:

1. Optimize ground state (charge 0, spinmultiplicity 1) --> otp + freq td=(nstates=6, triplet, root=1 ) UB3LYP

2. Optimize ground state (charge 0, spin 1) -> otp + freq td=(nstates=6, triplet, root=1 ) UB3LYP

3. Optimize "ground state" (charge 0, spin 3) -> otp + freq td=(nstates=6, triplet, root=1 ) UB3LYP

Many thanks!

Is it possible to obtain the bond length, angle and dihedral angle of the excited state by TD-DFT method?

Hi Mer:

If you want the first excited triplet there is an easy way. A triplet state is technically an excited state, which is defined as any stable electronic state above the ground state. Just calculate the compound of interest in the triplet state using an unrestritced wavefunction. That's your excited state. The nice thing will be that you will also get orbital energies for your excited state this way, which TDDFT will not give you as it uses the ground-state orbitals for its analysis. To check that this works you can also optimize the structure using TDDFT and you should find the structures the same.

Best,

Peter

Dear Partha:

Yes. Just optimize the structure with root=X and you will have the output of the excited state. Best,

Peter

One more point. Why do need the vibrationally resolved spectra. 0-0 transitions are all the rage these days. Is that why? If you are interested in an absorption process, you do not need frequencies to get the right answer. The zero point corrections will be small compared to the excitation energy. The reason 0-0 is being used is that DFT experts want agreement with experiment to promote their little functionals. Its easier to get this agreement if you are calculating a structure near the minimum energy as this is where the functionals have been parametrized. A vibrationally and electronically excited state is a weird beast with strange orbitals.

Peter

For Absorption Process one don't need to optimize excited state. For emission process one need to optimize the excited state. As Peter said 0-0 energy becomes popular even without their necessity. But to confirm optimization geometry as a minima or saddle point, Hessian is required.

Peter Damian Jarowski ,

As per your reply;

" If you want ........hnically an excited state, which is defined as any stable electronic state above the ground state. Just calculate the compound of interest in the triplet state using an unrestricted wave function. That's your excited state. The nic...is way, which TDDFT will not give you as it uses the ground-state orbitals for its analysis. To check that this works you can also optimize the structure using TDDFT and you should find the structures the same."

I need any related paper to give a reference in my paper please help me.

Regards

Hussain

# opt freq td=(nstates=n, root = m) b3lyp/6-31g

-----------------------------------------------------------------------

m=desired state, b3lyp: as example for exchange-correlation and 6-31g: BS, you can change according to your needs.

Whenever you optimize an excited state always check for frequency too, it helps in identifying whether the resulted optimized structure is real or some TS or intermediate. It will take longer time than the only Optimization step however it will save you from lot of trouble in your foreseeable tasks.

Thanks

Ajay Khanna

These options will first optimize the molecule in the mth excited state (find minimum energy potential surface) and then perform frequency calculations for that mth excited state.

What if i am interested in electronic groujnd state excited vibrational states?

by using DFT Method and 6.31 G base site and b3lyp fonction.

how to calculate radio active rate or life time in gaussian 16 software via TDDFT calculation?