Dear Daniele,
I ran GW calculations on LiF in yambo_3.4.1 and yambo_3.4.2, and obtained gaps with deviations of 0.2 eV (using exactly the same parameters in both yambo) !
Results of yambo_3.4.2 and yambo_4.1.2 are fully consistent.
However yambo_3.4.1 gives a gap which is larger by 0.2 eV relative to yambo_3.4.2 and yambo_4.1.2.
I further add that the results of yambo_3.3.0 and yambo_3.4.1 are fully consistent.
Furthermore, if you set parameters to converged values, deviation increases even more!
Since there are many publications based on yambo_3.4.1 and earlier versions of yambo, i would like to know what is going wrong here eventually.
Please find attached the corresponding data and outputs.
Best
Luca
QP gap inconsistency
Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan
-
luca.montana
- Posts: 27
- Joined: Fri Jun 13, 2014 6:52 pm
QP gap inconsistency
You do not have the required permissions to view the files attached to this post.
Luca Montana
PhD student
University of York, UK
PhD student
University of York, UK
-
Daniele Varsano
- Posts: 4198
- Joined: Tue Mar 17, 2009 2:23 pm
- Contact:
Re: QP gap inconsistency
Dear Luca,
thanks for posting this. We will investigate.
Can you post also the report file for the 3.4.1?
I noticed that the input files are not generated with the same release of the calculations, they are all done with 4.0.4,
if some variable named changed between releases it can cause problems.
Many thanks,
Daniele
thanks for posting this. We will investigate.
Can you post also the report file for the 3.4.1?
I noticed that the input files are not generated with the same release of the calculations, they are all done with 4.0.4,
if some variable named changed between releases it can cause problems.
Many thanks,
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/
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/
-
luca.montana
- Posts: 27
- Joined: Fri Jun 13, 2014 6:52 pm
Re: QP gap inconsistency
Dear Daniele,
Sorry, i forgot to post it.
Attached the report file of yambo_3.4.1.
I paid attention that the variables are consistent within each release.
Best
Luca
Sorry, i forgot to post it.
Attached the report file of yambo_3.4.1.
I paid attention that the variables are consistent within each release.
Best
Luca
You do not have the required permissions to view the files attached to this post.
Luca Montana
PhD student
University of York, UK
PhD student
University of York, UK
-
Daniele Varsano
- Posts: 4198
- Joined: Tue Mar 17, 2009 2:23 pm
- Contact:
Re: QP gap inconsistency
Thanks,
we will have a deep look into that.
Most or all publications are based on plasmon-pole approximations, anyway, we will search carefully the reason for such a discrepancy and let you know.
Best,
Daniele
we will have a deep look into that.
Most or all publications are based on plasmon-pole approximations, anyway, we will search carefully the reason for such a discrepancy and let you know.
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/
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/
-
Davide Sangalli
- Posts: 640
- Joined: Tue May 29, 2012 4:49 pm
- Location: Via Salaria Km 29.3, CP 10, 00016, Monterotondo Stazione, Italy
- Contact:
Re: QP gap inconsistency
Dear Luca,
what happens if you set in both calculations "CGrdSpXd= 0.0000" ?
Best,
D.
what happens if you set in both calculations "CGrdSpXd= 0.0000" ?
Best,
D.
Davide Sangalli, PhD
CNR-ISM, Division of Ultrafast Processes in Materials (FLASHit) and MaX Centre
https://sites.google.com/view/davidesangalli
http://www.max-centre.eu/
CNR-ISM, Division of Ultrafast Processes in Materials (FLASHit) and MaX Centre
https://sites.google.com/view/davidesangalli
http://www.max-centre.eu/
-
luca.montana
- Posts: 27
- Joined: Fri Jun 13, 2014 6:52 pm
Re: QP gap inconsistency
Dear Davide,
I still obtain the 0.2 eV deviation in gap.
CGrdSpXd= 0.0000 changes nothing.
Best
Luca
I still obtain the 0.2 eV deviation in gap.
CGrdSpXd= 0.0000 changes nothing.
Best
Luca
Luca Montana
PhD student
University of York, UK
PhD student
University of York, UK
-
Daniele Varsano
- Posts: 4198
- Joined: Tue Mar 17, 2009 2:23 pm
- Contact:
Re: QP gap inconsistency
Dear Luca,
thanks to Davide he went back in the log file of the yambo commits and he realized that indeed there was a bug in the previous releases that was fixed in 3.4.2.
Fortunately, the bug affected the real-axis implementation only and not the plasmon-pole approximation procedure, where most of the QP calculations are done.
Thanks again for pointing in out.
Best,
Daniele
thanks to Davide he went back in the log file of the yambo commits and he realized that indeed there was a bug in the previous releases that was fixed in 3.4.2.
Fortunately, the bug affected the real-axis implementation only and not the plasmon-pole approximation procedure, where most of the QP calculations are done.
Thanks again for pointing in out.
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/
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/
-
luca.montana
- Posts: 27
- Joined: Fri Jun 13, 2014 6:52 pm
Re: QP gap inconsistency
Dear Daniele,
many thanks for clarification.
Now i need to redo all of my calculations with yambo_4.2.1.
Any way at least i know what is correct.
Best wishes
Luca
many thanks for clarification.
Now i need to redo all of my calculations with yambo_4.2.1.
Any way at least i know what is correct.
Best wishes
Luca
Luca Montana
PhD student
University of York, UK
PhD student
University of York, UK