GW bands get tangled up

Run-time issues concerning Yambo that are not covered in the above forums.

Moderators: andrea marini, Daniele Varsano, myrta gruning, Conor Hogan

Ying Chen
Posts: 6
Joined: Thu Jul 02, 2020 9:55 am

GW bands get tangled up

Post by Ying Chen » Mon Nov 16, 2020 8:41 am

Dear Developers,

I am using YAMBO to calculate the quasi-particle band structure of the mos2/wse2 heterojunction. But the calculated GW band structure is non-ideal due to some bands get tangled up. How can I solve this problem? Thank you!

Best regards,
Ying Chen
You do not have the required permissions to view the files attached to this post.
Ying Chen
Academy for Engineering and Technology, Fudan University, Shanghai 200433, China

User avatar
Daniele Varsano
Posts: 2723
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW bands get tangled up

Post by Daniele Varsano » Mon Nov 16, 2020 8:57 am

Dear Ying Chen,
check first if this is a failure of the band interpolation, or if effectively this is consistent with the points you calculated the GW correction.
It may happen that you just need to calculate more occupied bands to recover a correct behaviour.
Next, you can try to interpolate corrections instead of QP energies.
In order to do so, you need first to calculate your DFT band structure with QE, and next indicate theat PW_SAVE directory in your input file,
something like:

Code: Select all

% INTERP_Grid
 24 | 24 | 12 |                          # Interpolation BZ Grid
%
INTERP_Shell_Fac= 20.00000     # The bigger it is a higher number of shells is used
GfnQPdb=  "E < QP/ndb.QP"
CIRCUIT_E_DB_path= "PW_SAVE"    # SAVE obtained from the QE `bands` run (alternative to %BANDS_kpts)
The PW_SAVE directory will contain a SAVE directory generated by p2y.

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
claudio
Posts: 320
Joined: Tue Mar 31, 2009 11:33 pm
Contact:

Re: GW bands get tangled up

Post by claudio » Mon Nov 16, 2020 10:03 am

You can also try to change
the interpolation method

INTERP_mode= "BOLTZ"

and see if it helps

best
Claudio
Claudio Attaccalite
[CNRS/ Aix-Marseille Université/ CINaM laborarory / TSN department
Campus de Luminy – Case 913
13288 MARSEILLE Cedex 09
web site: http://www.attaccalite.com
Freely download scientific books from: http://www.freescience.info

Ying Chen
Posts: 6
Joined: Thu Jul 02, 2020 9:55 am

Re: GW bands get tangled up

Post by Ying Chen » Mon Nov 16, 2020 10:11 am

Dear Dr. Daniele,

Thank you very much for your suggestions. I have checked the process of band interpolation and found an error of "<02s> [INTERPOLATION@E_on_built_in] Errors: Max, Avg, Avg_rel = 1.34691596E-04 9.39816618E-06 1.26988976E-04". The relevant reports are as follows:

<01s> [06] Interpolation tool
<01s> [06.01] Loading special Points for the Unknown lattice
<01s> [WARNING]Unknown lattice unit cell
<01s> [06.02] External/Internal QP corrections
<01s> E<./all_Bz/ndb.QP[ E-E PPA E-E PPA: 27.21138 XG: 235 Xb: 1 80 Scb: 1 100]
<01s> [dE_from_DB-Nearest K] Exact matches [o/o]: 100.0000
<01s> [06.03] Interpolation@work: Circuit
<01s> Number of K-points in the circuit : 72
<01s> [INTERPOLATION] Number of shells: 4432
<02s> [INTERPOLATION@E_on_built_in] Errors: Max, Avg, Avg_rel = 1.34691596E-04 9.39816618E-06 1.26988976E-04
<02s> [06.04] Interpolation@work: Spin and Magnetization
<10s> Spinor Components |########################################| [100%] 07s(E) 07s(X)
<10s> [spin-Nearest K] Exact matches [o/o]: 1.357466
<10s> [spin-Nearest K] Neighbours [user]: 1
<10s> [spin-Nearest K] [real]: 1
<10s> [spin-Nearest K] Interpolations [o/o]: 98.64253
<19s> Magn-factor |########################################| [100%] 09s(E) 09s(X)
<19s> [magnetization-Nearest K] Exact matches [o/o]: 1.357466
<19s> [magnetization-Nearest K] Neighbours [user]: 1
<19s> [magnetization-Nearest K] [real]: 1
<19s> [magnetization-Nearest K] Interpolations [o/o]: 98.64253
<19s> [06.05] Fermi Levels
<19s> Fermi Level (BUILT-in bands) [eV]: 0.3140905
<19s> [06.06] Bands output
<19s> [07] Game Over & Game summary

What does that mean? How can I avoid this error?

Best,
Ying Chen
Ying Chen
Academy for Engineering and Technology, Fudan University, Shanghai 200433, China

User avatar
Daniele Varsano
Posts: 2723
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW bands get tangled up

Post by Daniele Varsano » Mon Nov 16, 2020 10:14 am

Dear Ying Chen,
this is not an error of the calculation, but it indicates the error in the interpolation procedure which is actually small.
I suggest you check carefully if your QP corrections are compatible with the interpolation, if not please see the Caludio's suggestion in changing the inteprolation method. Finally, you can try the method I proposed to you in my first reply.
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/

Ying Chen
Posts: 6
Joined: Thu Jul 02, 2020 9:55 am

Re: GW bands get tangled up

Post by Ying Chen » Tue Nov 17, 2020 3:06 am

Dear Daniele and Claudio,

I have tried to change the interpolation method to "BOLTZ" ,
calculated more occupied bands in the "gw_ppa_all_Bz.in" , and calculated DFT band structure with QE and indicate the "PW_SAVE" which is generated by p2y after calculating the QE band in my input file. But the result of interpolation does not seem to improve.

Best,
Ying Chen
Ying Chen
Academy for Engineering and Technology, Fudan University, Shanghai 200433, China

User avatar
Daniele Varsano
Posts: 2723
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW bands get tangled up

Post by Daniele Varsano » Tue Nov 17, 2020 9:03 am

Dear Ying Chen,
can you post the report file of your GW calculation together with the output of the interpolation?
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/

Ying Chen
Posts: 6
Joined: Thu Jul 02, 2020 9:55 am

Re: GW bands get tangled up

Post by Ying Chen » Tue Nov 17, 2020 9:31 am

Dear Daniele,

I have show the output and report files in the attachments. Thank you very much for your help!

Best,
Ying Chen
You do not have the required permissions to view the files attached to this post.
Ying Chen
Academy for Engineering and Technology, Fudan University, Shanghai 200433, China

User avatar
Daniele Varsano
Posts: 2723
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW bands get tangled up

Post by Daniele Varsano » Tue Nov 17, 2020 9:48 am

Dear Ying Chen,
sorry, but the output you sent does not look like the plot you have shown before.
Please, see plot attached.
bands.png
[attachment=0]

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/

Ying Chen
Posts: 6
Joined: Thu Jul 02, 2020 9:55 am

Re: GW bands get tangled up

Post by Ying Chen » Tue Nov 17, 2020 10:26 am

Dear Daniele,

I am sorry for I just summited the output of another material Ga2O3 because I have tried your method on this material. The eight bands you shown is normal, but the bands with higher energy cross each other,
Image
So is it a normal phenomenon?

For the plot I have shown before, the output and report files is shown in the attachments.

Best,
Ying Chen
You do not have the required permissions to view the files attached to this post.
Ying Chen
Academy for Engineering and Technology, Fudan University, Shanghai 200433, China

Post Reply