Page 2 of 2

Re: Electron self-energy in polar materials - sanity check?

Posted: Wed Feb 28, 2018 4:17 am
by chrisewolf
A small update, changing the ASR in matdyn.x does not affect the phonons from ph.x >q2r.x>matdyn.x; the dispersions are very comparable and look "OK"

https://drive.google.com/open?id=1XppGC ... p5NEB7uCb-

Re: Electron self-energy in polar materials - sanity check?

Posted: Sat Mar 03, 2018 12:47 pm
by chrisewolf
long story short;

when run with ibrav=0 and CELL_PARAMETERS set it produced output as before; when ibrav=2 is used to define the cell it works.

the resulting spectral function looks fairly nice (not sure if that is converge but values seem reasonable);

https://drive.google.com/open?id=1DzIls ... vnx5N2GRcV

a similar calculation on MgO was published here: https://journals.aps.org/prb/pdf/10.110 ... .92.085137

Thanks everyone for their patience!

Best,
Chris

Re: Electron self-energy in polar materials - sanity check?

Posted: Sat Mar 03, 2018 4:16 pm
by carla.verdi
Hi Chris,

Thanks for letting us know, we will try and fix the bug and for now warn not to use ibrav 0!

Carla

Re: Electron self-energy in polar materials - sanity check?

Posted: Mon Mar 05, 2018 10:50 am
by sponce
Hello,

There was indeed an issue with ibrav=0 in the past. However, it should have been corrected.

In readmat_shuffle2.f90 there should be specific call when ibrav ==0.

What is your setting ?

- which version of QE/EPW (if it is an old one, could you confirm that the problem persists with the most recent one)
- crystal or simple sum rule
- with or without SOC

If you have a very small (fast/unconverged) example that reproduce the issue, it would be fantastic if you can send it to me.

Thanks,
Samuel

Re: Electron self-energy in polar materials - sanity check?

Posted: Mon Mar 05, 2018 3:15 pm
by chrisewolf
Hi Samuel,

I am using the epw version bundled with qe-6.2.1;

Code: Select all

rogram EPW v.4.3.0 (svn rev. 14038) starts on  6Mar2018 at  0: 7:53 

     This program is part of the open-source Quantum ESPRESSO suite
     for quantum simulation of materials; please cite


I will send you the input files tomorrow to your institutional email tomorrow, office hours are already over here in Korea. But if I remember correctly the only thing I changed was ibrav=0 to ibrav=2 (cif2cell has the habit to make input files with ibrav=0 I think);

- no SOC
- ASR simple
- identical phonon dispersions with ph.x, q2r and matdyn but not with epw.

HTH!

Best,
Chris

Re: Electron self-energy in polar materials - sanity check?

Posted: Tue Mar 13, 2018 12:31 am
by carla.verdi
Hi Chris,

Just a clarification - when you set ibrav=2 do you also obtain the correct interpolated
phonon dispersion within EPW (simple asr)?

Best
Carla

Re: Electron self-energy in polar materials - sanity check?

Posted: Sat Apr 14, 2018 6:51 pm
by chrisewolf
Hi Carla,

sorry for the late reply but the answer is "yes". Seems with an ibrav that is NOT 0 it works well! Great example in the School2018 material, was fun to work through that as well!

Best,
Chris