Page 1 of 1

"Missing k/q database" by ypp_ph with nqX in Si EPC tutorial

Posted: Sat Nov 20, 2021 4:58 pm
by minyez
Dear all,

Greetings from an absolute newbie to Yambo from China. My question seems related to the post by Antimo (viewtopic.php?p=10781#p10781), but is confronted in the Si tutorial (http://www.yambo-code.org/wiki/index.ph ... n_Coupling). Please kindly remind me of anything I missed to follow the rules of the forum.

I was trying to follow the tutorial to evaluate temperature effect of band gap due to electron-phonon coupling (EPC). The first two parts went smoothly and I was able to create a Eg-Temp figure very close (<10meV diff.) to the one in the tutorial. As the q-point generation was slightly tricky, I continued to go to the 5th part for a regular q-mesh set in phonon/dvscf calculations (ldisp=.true., nq1=nq2=nq3=4). phonon and dvscf calculations finished with no error. In the dvscf/si.save directory, ypp.in is created with `GkkpExpand` uncommented and correct `DBsPath`. Running "ypp_ph" then ended with the error

Code: Select all

 <---> [05.05] Energies & Occupations
 <---> [06] == Electron-Phonon Databases ==
 <---> Missing k/q scattering database

[ERROR] STOP signal received while in[06] == Electron-Phonon Databases ==
[ERROR]Please run a setup with the BSEscatt flag activated
The workflow script, QE and Yambo inputs are tared and attached. ecutwfc is halved for quick trial and error. Is there anything wrong in my inputs or workflow?

Also I note that the q-points in the phonon calculation here is almost the same as the "ypp_ph -k q" && "ypp_ph" in the nscf folder, but there is one difference (No.5).
Is this due to different choice of q-point in the same q-star in QE and Yambo?

Code: Select all

#ypp_ph                                          phonon with nqX
  1   0.000000000   0.000000000   0.000000000      0.000000000   0.000000000   0.000000000
  2   0.125000000   0.125000000  -0.125000000      0.125000000   0.125000000  -0.125000000
  3  -0.250000000  -0.250000000   0.250000000     -0.250000000  -0.250000000   0.250000000
  4   0.250000000   0.000000000   0.000000000      0.250000000   0.000000000   0.000000000
==5   0.375000000   0.125000000  -0.125000000     -0.125000000  -0.375000000   0.375000000==
  6   0.000000000  -0.250000000   0.250000000      0.000000000  -0.250000000   0.250000000
  7  -0.500000000   0.000000000   0.000000000     -0.500000000   0.000000000   0.000000000
  8  -0.500000000   0.250000000   0.000000000     -0.500000000   0.250000000   0.000000000
Thanks!

P.S: Versions of programs:
- Yambo 5.0.4 (GitHub release)
- QE 6.8 (GitHub release)

Re: "Missing k/q database" by ypp_ph with nqX in Si EPC tutorial

Posted: Sun Nov 21, 2021 5:57 pm
by claudio
Dear Hao Zhang

please use yambo_ph for the setup by doing "yambo_ph -i -V all",
uncomment the flag 'BSEscatt' and run it again, then ypp will work.

best
Claudio

Re: "Missing k/q database" by ypp_ph with nqX in Si EPC tutorial

Posted: Mon Nov 22, 2021 4:08 am
by minyez
Dear Claudio,

Thank you very much for the advice. I also notice the update in the EPC tutorial, thanks.

I followed the advice and ran the setup in nscf/si.save (yambo_ph), and did the other steps the same as above.
From the report file, yambo_ph correctly read the `BSEscatt` tag.
However, in dvscf/si.save, ypp_ph step still failed with the same "Missing k/q database" error.

The updated workflow and the record/log file of yambo_ph setup and ypp_ph gkkp are tared and attached.
Appreciated if further advice could be given.

Best Wishes
Min-Ye

Re: "Missing k/q database" by ypp_ph with nqX in Si EPC tutorial

Posted: Tue Nov 23, 2021 9:34 am
by claudio
Dear Min-Ye Zhang

please use the standard procedure, with q-grid generated from the k-grid,
the new one will be available only in yambo 5.0.5 that will be released soon

best
Claudio

Re: "Missing k/q database" by ypp_ph with nqX in Si EPC tutorial

Posted: Tue Nov 23, 2021 12:18 pm
by minyez
Dear Claudio,

Thank you for the reply. I will use the standard procedure.

Regards,
Min-Ye