fluorescence wavelength by CP2K

[Victor Volkov]

Dear developers:

good morning/evening.

Using pyrimidine, I try to understand how to compute 0'-0 fluorescence energy (wavelength).

In gaussian, this relatively well-established series of three calculations:

1) B3LYP/6-31+G(d,p) Opt
2) B3LYP/6-31+G(d,p) TD=(nstates=xx)
3) B3LYP/6-31+G(d,p) Opt TD=(nstates=xx,root=state_of_interest)
Emission energy is the difference Eex - Eexgr, where
Eex is the energy of the excited state: print in the string "TD-KS".
Eexgr is the ground state energy of the excited state geometry: print for the string "HF=-".

Here, I attach current CP2K results for pyrimidine structural optimization

in the first excited state, while accounting the excitation set to include three states.

Upon every optimization cycle, the output file reports four energies, for example:

1) Total energy:                                               -45.00011553431995

 !--------------------------- Excited State Energy ----------------------------!
2) Excitation Energy [Hartree]                                     0.1408310694
3) Total Energy [Hartree]                                           -44.8592844650

4)  ENERGY| Total FORCE_EVAL ( QS ) energy [a.u.]:   -44.859271658539249

Would you comment what are these?

Why there are three TOTALS?

Next, in the input file I instruct

&MOTION
  &GEO_OPT
    MAX_DR    1.0E-03
    MAX_FORCE 1.0E-03
    RMS_DR    1.0E-03    
    RMS_FORCE 1.0E-03
    OPTIMIZER CG

but, it seems, the output file does not report on the convergence criteria.

Does TDDFPT require additional instruction that such data would be printed

or this is out of the scope of the current package?

Finally, I see that upon the start:

-                            Excitation analysis                              -
 -------------------------------------------------------------------------------
        State             Occupied              Virtual             Excitation
        number             orbital              orbital             amplitude
 -------------------------------------------------------------------------------
             1   3.83221 eV
                                15                   16               0.995900
                                14                   17              -0.072640
             2   4.07218 eV
                                15                   17               0.989483
                                14                   16              -0.140638
             3   5.17410 eV
                                14                   16              -0.982529
                                15                   17              -0.137943
                                15                   18               0.080624
                                13                   16              -0.068511
 -------------------------------------------------------------------------------

while later, upon optimization:

-                            Excitation analysis                              -
 -------------------------------------------------------------------------------
        State             Occupied              Virtual             Excitation
        number             orbital              orbital             amplitude
 -------------------------------------------------------------------------------
             1   3.01565 eV
                                15                   16              -0.997677
             2   3.59856 eV
                                15                   17               0.989952
                                14                   16              -0.137983
             3   4.63370 eV
                                14                   16              -0.985780
                                15                   17              -0.135202
                                15                   18              -0.077828
 -------------------------------------------------------------------------------

Since the latter excitation energy  of 3.01565 eV is significantly lower than the initial one of 3.83221 eV, should the difference of about 0.8eV correspond to the Stokes shift,

and the 3.01565 eV would be the desired 0'-0 fluorescence energy?

Thank you.

I hope/wish the users would find my questions valuable.

With best wishes.

Victor

[Jürg Hutter]

Hi

yes, all your assumptions are correct. At the end of the excited state geometry optimization
you will get the energies
ground state Eg
excitation Eex this is the fluorescence energy
total energy Eg+Exc (this was optimized)

Eg+Exc - (ground state energy optimized) is the adiabatic excitation energy

and

excitation energy at ground state optimized - Eex is the Stokes shift.

Some other remarks:

these energies are different

3) Total Energy [Hartree] -44.8592844650
4) ENERGY| Total FORCE_EVAL ( QS ) energy [a.u.]: -44.859271658539249

because your SCF convergence is not tight enough.
If you use Diagonalization epsscf should be 10E-8 or even smaller.
But for this system I would use OT as it is much faster and then 1E-6 is good enugh.

You are using the CG geometry optimizer. Apparently then you don't get detailed output
on the progress of the optimization. If you use the default optimizers the usual
gradient breakdown will be printed. There is nothing special about TDDFT here.

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of Victor Volkov <volk...@gmail.com>
Sent: Sunday, November 27, 2022 1:22 AM
To: cp2k
Subject: [CP2K:18094] fluorescence wavelength by CP2K

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/e9b50a33-7672-4afe-8d9f-a589b2b690d1n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/e9b50a33-7672-4afe-8d9f-a589b2b690d1n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Victor Volkov]

Dear Fr. Hutter:

good evening.

Thank you for the kind explanations.

With best regards,

Victor

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZR0P278MB0759712D8A545287A5A4D9AA9F159%40ZR0P278MB0759.CHEP278.PROD.OUTLOOK.COM.