[QE-developers] Eigenvalue difference between 'scf' and 'bands' calculations

[Jung-Hoon Lee]

Hi Paolo, 

Hope you are safe and well. 

Thanks for your response and solutions! It seems the ordering of G-vectors is very sensitive to the lattice symmetry.  

Cheers,
Jung-Hoon
--
Jung-Hoon Lee, Ph. D.

Senior Research Scientist
Computational Science Research Center
Korea Institute of Science and Technology (KIST)

Email: jhlee84 at kist.re.kr
https://sites.google.com/site/junghoonleedavid/ <https://sites.google.com/site/junghoonleedavid/>

> 25 Sep 2020 22:31, Paolo Giannozzi <p.giannozzi at gmail.com> 작성:
> 
> Hi, I think I have found the problem, but the solution is not straightforward. Some explanations here:  https://gitlab.com/QEF/q-e/-/issues/230 <https://gitlab.com/QEF/q-e/-/issues/230>
> If the correct value of "ibrav" works for you, keep using it and ignore the problem. Solving it will take some time and effort. Thank you for reporting this
> 
> Paolo
> 
> On Thu, Sep 24, 2020 at 4:38 AM Jung-Hoon Lee <jhlee.david at gmail.com <mailto:jhlee.david at gmail.com>> wrote:
> Dear Paolo, 
> 
> Thanks for your response. 
> 
> When I change the cell parameter setting from ibrav =0 with CELL_PARAMETERS to ibrav = -5 with A,B,C,cosAB,cosBC,and cosAC, the problem is solved. It seems my previous cell parameter setting (ibrav=0 with CELL_PARAMETERS) is not enough to provide the correct bravais lattice information to QE code. Maybe not enough number of digits? But, I don’t still understand why this different setting gives such a large eigenvalue difference. 
> 
> Cheers,
> Jung-Hoon
>  
> --
> Jung-Hoon Lee, Ph. D.
> 
> Senior Research Scientist
> Computational Science Research Center
> Korea Institute of Science and Technology (KIST)
> 
> Email: jhlee84 at kist.re.kr <mailto:jhlee84 at kist.re.kr>
> https://sites.google.com/site/junghoonleedavid/ <https://sites.google.com/site/junghoonleedavid/>
> 
>> 2020. 9. 24. 02:48, Paolo Giannozzi <p.giannozzi at gmail.com <mailto:p.giannozzi at gmail.com>> 작성:
>> 
>> Very strange. There is clearly something not right going on: notice the "negative charge", that should not exist for norm-conserving PP's.
>> 
>> Paolo
>> 
>> On Wed, Sep 23, 2020 at 10:08 AM Jung-Hoon Lee <jhlee.david at gmail.com <mailto:jhlee.david at gmail.com>> wrote:
>> Dear Developers, 
>> 
>> I’ve been trying to compute the band structure of UIO66 MOF. Interestingly, however, the computed eigenvalues obtained from ‘bands’ calculation are very different from those obtained from ’scf’ calculation as attached. Since the band gap is at the gamma point I’ve compared eigenvalues at the gamma point. The band gap difference at the gamma point is about about 2.4 eV. This is very strange…could you take a look at input and out files? 
>> 
>> Thank you, in advance, for your help. 
>> 
>> Cheers,
>> Jung-Hoon
>> 
>> 
>> --
>> Jung-Hoon Lee, Ph. D.
>> 
>> Senior Research Scientist
>> Computational Science Research Center
>> Korea Institute of Science and Technology (KIST)
>> 
>> Email: jhlee84 at kist.re.kr <mailto:jhlee84 at kist.re.kr>
>> https://sites.google.com/site/junghoonleedavid/ <https://sites.google.com/site/junghoonleedavid/>
>> _______________________________________________
>> developers mailing list
>> developers at lists.quantum-espresso.org <mailto:developers at lists.quantum-espresso.org>
>> https://lists.quantum-espresso.org/mailman/listinfo/developers <https://lists.quantum-espresso.org/mailman/listinfo/developers>
>> 
>> 
>> -- 
>> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>> Phone +39-0432-558216, fax +39-0432-558222
>> 
> 
> 
> 
> -- 
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20200925/6225dffa/attachment-0001.html>