[Wien] Whether or not can we predict band gap with scissors operator?

[Rocquefelte]

Dear Joey Davis,

Your question is curious in some points because it contains a part of 
the answer (so it looks like a discussion more than a question).
However, I can try to give you my point of view on this topic. As soon 
as we are using an adjustable parameter (scissor operator for the band 
gap) no prediction can be done on the quantity we are correcting. 
However, by modifying this gap value with a scissor operator, prediction 
of related quantity (dielectric function, refractive index ...) can be 
done.
So in my point of view, the comparison of DFT+SO (SO: Scissors Operator) 
optical data with the experiment is definitely not a proper way to 
define the experimental band gap.
The experimental band gap must be determined from experiments only (or 
appropriate treatment of the so-obtained data).

Another important point is the difference between the fundamental band 
gap and the optical band gap. Depending on the experiments you are 
looking you will access to a different band gap (electronic or optical 
gap).

If you really wants to do optical band gap prediction, you should look 
at method that goes beyond DFT (GW, hybrid approaches, exact exchange, 
...). However, I could insist on one last point, in the different 
systems I have treated I have observed that usually the scissor operator 
we are using is nearly the same (or the same) for a given chemical 
formula. Then one possibility is to predict the gap variation more than 
the absolute gap value as a function of structure variation. More 
generally, DFT alone does not allow to predict the absolute value of the 
band gap but relative data can be obtained in some cases.

With my Best Regards

Xavier Rocquefelte



Joey Davis a écrit :
> Dear WIEN users:
>  
>    I am trying to calculate the optical and electronic properties of a 
> compound. I only have the structure and optical conductivity spectra 
> got from experiment.
>  
> You know that the DFT cannot give the accurate band gap. Scissors 
> operator is   introduced, that is, the conduction bands are shifted. 
> Then the spectra can be similar to the experiment.
>  
> Now, we got the band gap (Eg_1 ) from calculation. With scissors 
> operator (Eg_2 ), the shape of the mainly peaks in the optical 
> conductivity spectra agrees well with the spectra of experiment. Can 
> we declare the real band gap (the band gap from experiment) is equal 
> to Eg_1 +Eg_2 ?  Can we regard that this is a method to get the band 
> gap in our compound?
>  
> Can anyone tell me, or give me some references about the question?
>  
> Many thanks!
>  
> Joey Davis
>  
>
> ------------------------------------------------------------------------
> Be a PS3 game guru.
> Get your game face on with the latest PS3 news and previews at Yahoo! 
> Games. <http://us.rd.yahoo.com/evt=49936/*http://videogames.yahoo.com>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20070719/0a64c018/attachment.html