Hi,

1) "# td(nstates=10) opt" means optimizing the lowest excited state, evidently this is not what you need.

Assume that the ground state is neutral singlet, to calculate vertical absorption energy under gas phase, firstly use for example "# PBE1PBE/6-31G* opt" with "0 1" to optimize ground state, then based on the resulting geometry, use for example "# PBE1PBE/6-31G* TD(nstates=10)" with "0 1" to carry out TDDFT calculation, then you can directly read vertical absorption energies from the output file.

2) I strongly suggest you use delta-SCF method to optimize T1 geometry, namely use for example "# PBE1PBE/6-31G* opt" with "0 3". This way is much more efficient than the TDDFT way (i.e. # td(triplet,root=1,nstates=10) opt).

Based on the optimized T1 geometry, simply use for example "# PBE1PBE/6-31G* TD(triplet)", then from the output file you will find transition energy between S0-T1, which corresponds to phosphoresence emission energy. (Alternatively, you can do single point calculation respectively with "0 3" and "0 1" then get their difference energy, it corresponds to phosphoresence emission energy computed under delta-SCF method)

Best wishes,

Tian Lu

Hi,

1) "# td(nstates=10) opt" means optimizing the lowest excited state, evidently this is not what you need.

Assume that the ground state is neutral singlet, to calculate vertical absorption energy under gas phase, firstly use for example "# PBE1PBE/6-31G* opt" with "0 1" to optimize ground state, then based on the resulting geometry, use for example "# PBE1PBE/6-31G* TD(nstates=10)" with "0 1" to carry out TDDFT calculation, then you can directly read vertical absorption energies from the output file.

2) I strongly suggest you use delta-SCF method to optimize T1 geometry, namely use for example "# PBE1PBE/6-31G* opt" with "0 3". This way is much more efficient than the TDDFT way (i.e. # td(triplet,root=1,nstates=10) opt).

Based on the optimized T1 geometry, simply use for example "# PBE1PBE/6-31G* TD(triplet)", then from the output file you will find transition energy between S0-T1, which corresponds to phosphoresence emission energy. (Alternatively, you can do single point calculation respectively with "0 3" and "0 1" then get their difference energy, it corresponds to phosphoresence emission energy computed under delta-SCF method)

Best wishes,

Tian Lu

Hi Tian,

Thank you for your reply. In delta-SCF, first I can opt the structure with "0 3" for T1 and then from the opt structure, I calculate single point with "0 1" for S0. The difference in energy is the phosphorescent energy. Is it correct?

- Also "Based on the optimized T1 geometry, simply use for example "# PBE1PBE/6-31G* TD(triplet)"..However I wonder it is " with "0 1" instead? And then we will get the transition energy from S0 to T1... right?

Thank you very much

1 It is correct.

2 When use "# PBE1PBE/6-31G* TD(triplet)", "0 1" should be used together, because the ground state, which is used as reference state for performing TDDFT, is neutral singlet state.

I got it now. I am very appreciate your help. Thank you very much.

Article ChemInform Abstract: The Calculations of Excited-State Prope...

how to run do PES run for excited states ....

pl help

Adsorption spectra are easy. You do exactly how you wrote. To calculate emission spectra use this tutorial

https://docs.google.com/document/d/1f2LbJMfcudxNjUrrchj8_UEKW8HLDPczRiS2m_2QbzA/pub

Dear all,

I would like to ask about an observation I've made.

I'm calculating the absorption spectra of metal complex and I've found that, as more extra states I use in calculation, better it fits. In fact, there are important differences by using nstates=10 or 50 for example.

This situation is somewhat confusing because if I expect transitions around 200nm, by only using 10 states, they should be detected, but the results only give some low transitions around 400nm. While using 50 states, I seeing important transitions at below 300nm. Which is the reason of these results?

Thank you for your time,

Joaquim

Joaquim Mª Rius Bartra When calculating the transitions, TD-DFT usually gives you the results in order of energy, starting from the lowest energy. The more states you calculate, the more transitions you will get with higher energies.

This is exactly what you see: with few calculated states, you get those with the lowest energy, i.e. longest absorption wavelength, first.

Dear Dr Simon Janich ,

Thank you very much for your answer. It's the fist time I've performing this type of calculations with large molecules and I didn't expect such situation and I prefer to be sure.

Best regards,

Joaquim

First you have to determine excited state geometry using suitable functional, TDDFT. Once you get the output file, look for emission data ( singlet/triplet states).

Write 0 3 in the input file for triplet optimization. it will generate triplet excited state geometry. now calculate single point corresponding to the triplet excited state geometry.

Thao P. Nguyen Joaquim Mª Rius Bartra Simon Janich Wojciech Kołodziejczyk Anand Bharadvaja Fouad N. Ajeel

Thanks for your helpful answers

Dear Dr. Tian Lu

How to generate "fluorescence spectra"? Is it different from the calculated "phosphorescence spectra"?

Thank You

Permono Adi Putro Yes, but you should optimize S1 instead of T1, and then use S0-S1 excitation energy and oscillator strength to simulate the spectrum.

Dear Dr. Tian Lu

What are the keywords and multiplicity used to generate “fluorescence spectra”? please, tell me about its step-by-step using the Gaussian09 program

Thank You

Permono Adi Putro Here I assume that ground state of your system is singlet, the Kasha's rule is satisifed, and vibronic effect is ignored.

First, optimize your system using keywords like # PBE1PBE/6-311G* opt TD. This task will simultaneously calculate three excited states but only optimize the geometry for the first excited state (S1 state). Then, boot up Multiwfn (http://sobereva.com/multiwfn) and load the TDDFT output file, then input the following commands:

11 // Plotting spectra

3 // UV-Vis

20 // Modify oscillator strengths

2-3 // Choose S2 and S3 states

0 // Set oscillator strengths to zero (because emission is from S1 state)

0 // Plot spectrum

The current graph shown on screen is just the fluorescence spectrum. You can then use various options in the present interface to change plotting setting to improve the plotting effect. See Section 4.11 of Multiwfn manual for examples and illustrations of spectrum plotting.

Dear Dr. Tian Lu

Thank you very much for your answer. I will try it.

Best regards,

Permono

Tian Lu

I am trying to calculate the absorption and emission spectra with g09 in solvent methanol. I have

1. optimized the ground state

2. on the optimized geometry, I ran PBE1PBE/6-31G* TD(nstates=10) density=current this shows the absorption spectra

3. I ran TD optimization using # td(singlet,root=1,nstates=10) opt. I am getting a spectra. Is it the emission?

Prasanta Bandyopadhyay Using density=current is completely redundant if you do not need to study wavefunction of excited state (e.g. obtaining atomic charges and dipole moment for the excited state chosen by "root"), this keyword will request Gaussian to calculate relaxed wavefunction of excited state and thus significantly increase computational cost.

The spectrum directly obtained based on the output file of step 3 is not emission spectrum. It is important to understand that according to Kasha's rule, the emission comes from S1 state, therefore you need to set oscillator strenghts of all other excited states to zero before plotting spectrum.

Tian Lu , I have some question about dft calculation

1-when my calculation is finished (normal termination of Gaussian) the 6 bonds between the cobalt atoms and the oxygen atoms are destroyed (octhedral CoO6) I use the basis set LANL2DZ only.

why these bands are missing from the structure how i can resolve this problem

I am recognizing for your help

Jamai Nabil

Bonds are sometime not shown by GView, have you checked the distance between the atoms? are they acceptable? If yes, you do not need to care about the bond shown or not by GV.

Prasanta Bandyopadhyay thanks for your answer, my questions after the calculations finished normal termination there is somes bands missing Co-O after we open the file log to see my crystal structure it's normal ?? in gaussian ?? these results is acceptable in journal ? how i can checked the distance between the atoms