QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Dipankar Saha on October 11, 2016, 20:21

Title: Absorption Coeff.
Post by: Dipankar Saha on October 11, 2016, 20:21
Hi,
Can we directly plot the absorption coeff.  v/s.  energy (eV) ? / Do we need to use the "OpticalSpectrum" object for this purpose... ?

Regards_
Dipankar
Title: Re: Absorption Coeff.
Post by: Petr Khomyakov on October 12, 2016, 08:17
To plot absorption coefficient vs energy, one should first calculate it, and this is what OpticalSpectrum analizer essentially does by calculating the imaginary part of the dielectric function related to the absorption coefficient, see http://docs.quantumwise.com/tutorials/optical/optical.html.
Title: Re: Absorption Coeff.
Post by: Dipankar Saha on October 12, 2016, 14:55
Thanks a lot for the reply ... :)

Best_
Dipankar
Title: Re: Absorption Coeff.
Post by: Dipankar Saha on October 18, 2016, 09:16
Hi,

Is it possible to obtain the optical spectrum (or, Absorption v/s. Energy)...  with/ with-out electron-hole interactions ??

Best_
Dipankar
Title: Re: Absorption Coeff.
Post by: Petr Khomyakov on October 18, 2016, 10:39
To account for two particle excitations in the optical spectrum, one should adopt a many-body approach that goes beyond the ground state density functional approach implemented in ATK.   
Title: Re: Absorption Coeff.
Post by: Dipankar Saha on October 18, 2016, 12:43
Okay, I understand... :) 
..................

 Now, for example the dielectric constant is not isotropic ...i.e.,  exx =eyy , but not equals to ezz ... then how do you calculate "extinction coefficient" (rather, e1 and e2)?

Besides,  while finding absorption coefficient of any wide band S.C. ...how critical is the selection of the parameters e.g. , broadening, bands_below_fermi_level,  and  bands_above_fermi_level ??

Thanks and Regards_
Dipankar
 
Title: Re: Absorption Coeff.
Post by: Petr Khomyakov on October 18, 2016, 14:12
One should adopt sufficient number of bands to accurately calculate the absorption coefficient.  The actual numbers depend on the considered system and required accuracy.