Open Shell Systems

Questions and doubts about the spin-dependent version of Yambo.

Moderators: Davide Sangalli, andrea marini, Conor Hogan, myrta gruning

Open Shell Systems

Postby vitoversace » Thu Mar 13, 2014 4:10 pm

Hi all,

Some Questions :
1) Is Yambo capable of calculating open shell periodic systems ( open shell solids ) with doublet spin state, too ?
2) Does Yambo produces in the output file explicitly the singlet and triplet excitation energies for, let say, closed shell periodic systems or should one read it from the data ?
3) Does Yambo give the exciton probability density or distribution of bulk materials ?

With best wishes
Vito
Vito Versace
PhD student
University of York
vitoversace@yahoo.com
vitoversace
 
Posts: 58
Joined: Thu Mar 13, 2014 3:43 pm

Re: Open Shell Systems

Postby Daniele Varsano » Thu Mar 13, 2014 4:39 pm

Dear Vito,
please fill your signature with your affiliation, it is a rule of the forum.

1) Is Yambo capable of calculating open shell periodic systems ( open shell solids ) with doublet spin state, too ?


Yes, but me personally cannot help much on that.

2) Does Yambo produces in the output file explicitly the singlet and triplet excitation energies for, let say, closed shell periodic systems or should one read it from the data ?


Yes, in the BSE formulation you can calculated separately singlet and triplet excitations. At the moment the triplet calculations it is straightforward by input in the 3.3.0 release. It is also in the 3.4.0 release but it need a simple patch I can send you, or more easily I will insert in the repository very soon.

3) Does Yambo give the exciton probability density or distribution of bulk materials ?

Yes, this is done by post processing using ypp, see examples in the tutorials.

In general, the documentation you find in the code web page and the tutorial can help you. Also a search in the forum can be useful.
Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 2027
Joined: Tue Mar 17, 2009 2:23 pm

Re: Open Shell Systems

Postby vitoversace » Thu Mar 13, 2014 8:47 pm

Dear Daniele,

Thanks for your answers.
1) Would you please tell me what are the specific keywords for singlet and triplet exciation energies in the BSE input file to run it separately, as i could n't find any thing related to it in tutorials. (I am using 3.3.0 version)
2) Are the gw approximations sophisticated like in Abinit or is it only the simplest version G0W0 implemented so far ?
3) Does BSE take care of the full spin structure if the spin-orbit interaction (coupling) is of the same order of magnitude as the electron-hole interaction ?
4) Do you think that there is a relationship between exciton probability distribution and optical absorption PEAKS ? (are these peaks to be seen in exciton density in some way, too ?)

Bests
Last edited by vitoversace on Fri Mar 14, 2014 3:46 am, edited 1 time in total.
Vito Versace
PhD student
University of York
vitoversace@yahoo.com
vitoversace
 
Posts: 58
Joined: Thu Mar 13, 2014 3:43 pm

Re: Open Shell Systems

Postby Daniele Varsano » Thu Mar 13, 2014 10:20 pm

Dear Vito.
1) Would you please tell me what are the specific keywords for singlet and triplet exciation energies in the BSE input file to run it separately, as i could n't find any thing related to it in tutorials.


For the singlet, please follow the Documentation. You can either use the full coupling matrix activating
Code: Select all
BSKmod= "SEX" 
BSEmod = coupling
WehCpl

or discard the coupling (Tamm-Dancoff approximation) is suitable.

About the triplet as I told you, the structure of the code changed from the 3.3 to 3.4.
In order to calculate triplet excitation, I will update the code soon, in the meanwhile you can apply the patch attached in the post by:
Code: Select all
patch -p0 < patch.tex

and set in the input file:
Code: Select all
BSKmod= "TRIPLET" 

even if this not appear in the legend. This should work, even if it has not been extensively tested and this is the reason why I still not have inserted it in the source.

2) Are the gw approximations sophisticated like in Abinit or is it only the simplest version G0W0 implemented so far ?


I do not know how GW in Abinit it is now, it is a long time I did not have run the code. Anyway in the GPL version of Yambo only G0W0 is supported, Self conistency is presente but in the development version only.

3) Does BSE take care of the full spin structure if the spin-orbit interaction (coupling) is of the same order of magnitude as the electron-hole interaction ?

Unfortunately I cannot help you on that without the risk to tell you something wrong, people who coded it can answer you giving you details.
4) Do you think that there is a relationship between exciton probability distribution and optical absorption PEAKS ? (are these peaks to be seen in exciton density in some way, too ?)

There is a relation between exciton wavefunction localization and binding energy of the energy (see for example PRL 101 133002 (2008). Anyway it is not clear to me what exactly you mean by exciton density, so I cannot answer you in a clear way.

About the signature, you can add it in your profile once for all.

Best,
Daniele
You do not have the required permissions to view the files attached to this post.
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 2027
Joined: Tue Mar 17, 2009 2:23 pm

Re: Open Shell Systems

Postby vitoversace » Thu Mar 13, 2014 11:41 pm

Dear Daniele many many thanks for your prompt answers.

1 ) Some one was discussing with me,saying, that probably one can see singlet excitation energies in electronic density or even in spin density
but i believe that these singlet or triplet excitations could be probably seen in squared of exciton wave function in some way or at least there COULD be a relationship between squared of exciton wave function and excitation energies
which one sees in optical spectrum as peaks. For instance in LiF the first peak located at 12.8 eV in optical absorption spectrum is a spin singlet excitation state and one can at least try to verify whether this also could be seen in a way in
the plotted squared of exciton wave function. Any way this is just an idea to find evidences to see the same physics in other possible quantities.

2 ) Would you please tell me who can help me concerning open shell systems and BSE implementation in Yambo ???
3 ) Do you accidentally know which program can calculate spin densities of excited states in solids ?
4 ) Do you know how to generate a KSS file format for a quasi particle file QPS obtained from Abinit in order to be read by yambo ?

Best wishes
Vito Versace
PhD student
University of York
vitoversace@yahoo.com
vitoversace
 
Posts: 58
Joined: Thu Mar 13, 2014 3:43 pm

Re: Open Shell Systems

Postby Daniele Varsano » Fri Mar 14, 2014 10:06 am

Dear VIto,

1 ) Some one was discussing with me,saying, that probably one can see singlet excitation energies in electronic density or even in spin density
but i believe that these singlet or triplet excitations could be probably seen in squared of exciton wave function in some way or at least there COULD be a relationship between squared of exciton wave function and excitation energies
which one sees in optical spectrum as peaks. For instance in LiF the first peak located at 12.8 eV in optical absorption spectrum is a spin singlet excitation state and one can at least try to verify whether this also could be seen in a way in
the plotted squared of exciton wave function. Any way this is just an idea to find evidences to see the same physics in other possible quantities.


Do not have a clear answer on what you mean by "see". In order to look at the response densities you first have to know which electron-hole pairs participate in that excitation, so you need first to solve your BSE/TDDFT/RPA equation.

2 ) Would you please tell me who can help me concerning open shell systems and BSE implementation in Yambo ???

People implemented it read the forum, they will answer you as soon as they can. We try to avoid personal email as in the forum all people can read the answer, we avoid to provide the same answer to many people, and this save us a lot of time.
3 ) Do you accidentally know which program can calculate spin densities of excited states in solids ?

Not really, the TDDFT version of Quantum Espresso gives you the response density, I'm not sure that spin-resolved calculation are implemented: check it. Anyway this at TDDFT level and not BSE. I think that ecxitonic wavefunction ca provide you similar information.
4 ) Do you know how to generate a KSS file format for a quasi particle file QPS obtained from Abinit in order to be read by yambo ?

Yambo reads KSS format for the KS electronic structure. QPs are read from previous yambo calculation. Anyway, if you want to use QP from Abinit, or any other code, the yambo postprocessing ypp, permits you to generate a user defined QP file that can be read by yambo.
See here the manual: it is the ypp -q option.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 2027
Joined: Tue Mar 17, 2009 2:23 pm

Re: Open Shell Systems

Postby vitoversace » Fri Mar 14, 2014 4:10 pm

Dear Daniele,

Unfortunately it does n't work really.
I have a QPS file from abinit. Self consistency only on energies (gwcalctyp = 10).

ypp -q tgw1o_DS3_QPS produces nothing !

Bests
You do not have the required permissions to view the files attached to this post.
Vito Versace
PhD student
University of York
vitoversace@yahoo.com
vitoversace
 
Posts: 58
Joined: Thu Mar 13, 2014 3:43 pm

Re: Open Shell Systems

Postby Daniele Varsano » Fri Mar 14, 2014 4:20 pm

Dear Vito,
this is not the right way to use it.
By typing ypp -q
an input file is produced, you have to edit by hand inserting the QP corrections you calculated.
Next running ypp
a database wit the QP energy is produced to be used for next calculations.

Yambo does not know anything about QPs file form abinit.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 2027
Joined: Tue Mar 17, 2009 2:23 pm

Re: Open Shell Systems

Postby vitoversace » Fri Mar 14, 2014 5:08 pm

Dear Daniele many thanks, it works now.

But it reads only the first 10 bands and i can n't change it to higher numbers !
What do these numbers in second row mean ( -1| -1| -1| -1| 0.0| 0.0| 0.0|) ???

%QPkrange # [QPDB]QPDB energy correction( kp1| kp2| bnd1| bnd2| E-Eo[eV]| Img(E)[eV]| Z|)
-1| -1| -1| -1| 0.0| 0.0| 0.0|
%

Bests
You do not have the required permissions to view the files attached to this post.
Vito Versace
PhD student
University of York
vitoversace@yahoo.com
vitoversace
 
Posts: 58
Joined: Thu Mar 13, 2014 3:43 pm

Re: Open Shell Systems

Postby Daniele Varsano » Fri Mar 14, 2014 5:24 pm

Dear Vito,
that line tells you how to build the input file.
Here goes an example:
Code: Select all
qpdb                         # [R] Quasi-particle database manager
%QPkrange            # [QPDB]QPDB energy correction( kp1| kp2| bnd1| bnd2| E-Eo[eV]| Img(E)[eV]| Z|)
  1|  1|  1|  1|   -1.361170|    0.464390|    0.575880|
  1|  1|  2|  2|   -1.290040|    0.129796|    0.573814|
  1|  1|  3|  3|   -1.229511|    0.135846|    0.564947|
  1|  1|  4|  4|   -1.336101|    0.116444|    0.587400|
....
....
%


so insert mantaining this format:
the k point index in the first two column, the band index (see the correspondence between index and k-points in the yambo setup report) in the third and fourth, the correction it should be your deltaE, the Img(E)
and the renormalization factor should be irrelevant for your needs, anyway your output reports the Z (some of them are unphysical!) you can plug it, and put wathever in the Img(E) if you do not have it.
You need to maintain the syntax: each value spearated by | and close the list with %.

Best,
Daniele

PS: Please remember that yambo and abinit are interfaced in the sense that you can read the KSS DFT file using a2y: all what comes next it is not compatible.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/
User avatar
Daniele Varsano
 
Posts: 2027
Joined: Tue Mar 17, 2009 2:23 pm

Next

Return to Spin-dependent and spinorial calculations

Who is online

Users browsing this forum: No registered users and 1 guest

cron