TDDFT/GW for Nanocluster

[anjalisingh]

Hi All,

I am interested in calculating first exciton binding energy for Mn doped ZnSe nanocluster.
I am confused about should I use TDDFT or GW+BSE calculations to get that?
I look at old tutorial of Yambo for SiH4 molecule whose link is given below
http://www.yambo-code.org/tutorials/SiH4/index.php.
This example shows poor agreement of calculated absorption spectra with experimental one.
Should I use TDDFT or GW+BSE for getting first exciton binding energy?
And new tutorials (http://www.yambo-code.org/wiki/index.ph ... =Main_Page) are quite well documented but I did not get any example of TDDFT on Yambo wiki page.
Can I get example of TDDFT on wiki page or somewhere else which has good documentation.

Warm Regards,
Anjali
JNCASR, Bangalore
India

[Daniele Varsano]

Dear Anjali,
if you are interested in excitonic effects and binding energies you should use GW/BSE methods.
In yambo the TDDFT is implemented in LDA or GGA fashion which does not capture excitonic effects.
The fact that GW/BSE spectrum does not match the experiment in SiH4 it is due surely to poor convergence (tutorial are meant to run in few minutes), and also because comparison with experiments has to be made with care as many effects are neglected (enviroments, vibrionic effects etc.).
In literature there are several GW/BSE calculations in finite systems (molecules, clusters) which present satisfactory agreement.
I think that GW/BSE is the best you can do (beside quantum chemistry calculations that can be too much expensive) even if it is not guarantee you can match the experimental absorption.

Best,

Daniele