extracting full dielectric tensor

Anything regarding the post-processing utility (e.g. excitonic wavefunction analysis) is dealt with in this forum.

Moderators: Davide Sangalli, Daniele Varsano, andrea marini

Post Reply
michele guerrini
Posts: 11
Joined: Thu Mar 28, 2019 10:53 am

extracting full dielectric tensor

Post by michele guerrini » Wed Feb 12, 2020 12:40 pm

Dear developers, I would like to ask if it is possible to obtain the full dielectric tensor tensor (or the its imaginary part) after having performed three calculations with independent polarizations. Thank you!
Dr. Michele Guerrini
Humboldt-Universität zu Berlin
Institut für Physik und IRIS Adlershof
Zum Grossen Windkanal 6, 12489 Berlin, Germany
e-mail: guerrini@physik.hu-berlin.de
Skype: merlin__88

User avatar
Daniele Varsano
Posts: 2447
Joined: Tue Mar 17, 2009 2:23 pm

Re: extracting full dielectric tensor

Post by Daniele Varsano » Wed Feb 12, 2020 2:06 pm

Dear Michele,
I think it should be possible by using yambo_kerr.
Maybe others can tell you more on that.
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale

User avatar
Davide Sangalli
Posts: 344
Joined: Tue May 29, 2012 4:49 pm
Location: Via Salaria Km 29.3, CP 10, 00016, Monterotondo Stazione, Italy

Re: extracting full dielectric tensor

Post by Davide Sangalli » Wed Feb 12, 2020 6:12 pm

Dear Michele,
the standard version of yambo can compute the diagonal elements of the dielectric tensors. Playing with them, i.e. selecting different directions in input (you would need 6 in total), there are ways to reconstruct also the off-diagonal components of (the symmetric part of) the dielectric tensor. You can check in the literature for equations.

For the off diagonal components you can also use yambo_kerr, as mentioned by Daniele. Chosen a direction "d" in the "xy" plane in input, it will give epsilon_dd(w) and epsilon_{dd'}(w) where d' is a direction rotated by 90 degrees with respect to the original chosen. However it works only in the xy plane. Thus you need to rotate your sample (or change the code) to get also off diagonal terms which involve "z".
Notice that this approach, at variance with the previous one, would give also the anti-symmetric part of the dielectric tensor if the material has any.

Hoep it helps.
Kind Regards,
Davide Sangalli, PhD
CNR-ISM, Division of Ultrafast Processes in Materials (FLASHit) and MaX Centre

Post Reply