How to compute phosphorescence rates

[ethan]

Can you please provide me a quick tutorial on how to compute phosphorescence rates (T1 => S0)? Also which is the best method to compute it? How does the input look like?

[magnus]

I'm don't know if such a tutorial exists. Also I cannot help you with the choice of best method. I'd recommend that you do a literature search for that. With respect to the input, you need a structure like this (where of course the ... in the wavefunction part has to be changed to something reasonable):

Code: Select all

**DALTON INPUT
.RUN RESPONS
**INTEGRALS
.SPIN-ORBIT
**WAVE FUNCTIONS
...
**RESPONS
*QUADRATIC
.PHOSPHORESENCE
.ROOTS
 4
**END OF DALTON INPUT

The specification of number of roots assumes that you do not use point-group symmetry or that your molecule has no symmetry.

[taylor]

Another aspect to consider is what sort of molecules and electronic states you are interested in. If the systems are composed of lighter atoms, like a typical organic molecule, then you should consider the AMFI approximation for dealing with the spin-orbit terms (this is described in the manual). AMFI is much less computationally demanding than the full (one- and two-electron) spin-orbit treatment, and our own investigations (among many others) confirm that for typical excited states of small organic systems there is essentially no difference between the results obtained with AMFI and those with the full operator.

On the other hand, if you are interested in systems involving significantly heavier atoms it may be that AMFI is less appropriate. If you are concerned about this you could try to find a suitable model system which you can use to calibrate AMFI against the full operator.

Best regards
Pete