[QE-developers] Matrix Elements of the Momentum Operator

[Lorenzo Paulatto]

would it make sense to set Ef to the HOMO (which is always computed in 
insulators) in this case ?

CALL get_homo_lumo (ehomo, elumo)

On 23/01/2020 10:01, Paolo Giannozzi wrote:
> Thank you for reporting this. As far as I know, during a band 
> calculation the Fermi energy is not computed. It is however read from 
> file (the one produced by the previous scf calculation) and re-written 
> to file at the end of the band calculation. A zero (not set) Fermi 
> energy means that it wasn't computed during the scf calculation. This 
> happens if you do not use smearing.
> 
> There was however a bug with units: Ef was written to file in Hartree 
> a.u. but not converted to Rydberg a.u. when read. This was fixed a few 
> days ago.
> 
> Paolo
> 
> 
> 
> On Thu, Jan 23, 2020 at 7:30 AM Rafal Korlacki <rkorlacki2 at unl.edu 
> <mailto:rkorlacki2 at unl.edu>> wrote:
> 
>     Hello,____
> 
>     __ __
> 
>     I believe there is a bug in the routine write_p_avg in all the
>     versions 6 of QE. Specifically, in the following fragment: ____
> 
>     __ __
> 
>           !____
> 
>           !   Compute the number of occupated bands at this k point____
> 
>           !____
> 
>           DO ibnd = 1, nbnd____
> 
>              IF (et (ibnd, ik)<=ef) nbnd_occ = ibnd____
> 
>           ENDDO____
> 
>     __ __
> 
>     It seems that the variable ef=0 rather than whatever the right
>     number is, and the code gets confused about which bands are occupied
>     and which are not. If I replace ef with a number corresponding to an
>     energy somewhere within the bandgap for a particular material, the
>     output is consistent with earlier versions of QE, like 5.3.0.
>     Otherwise the code calculates overlaps between the wrong band pairs,
>     the format of the output varies between k-points, etc. I tested this
>     behavior for QE 6.2 to 6.5.____
> 
>     __ __
> 
>     Best regards,____
> 
>     __ __
> 
>     Rafal Korlacki____
> 
>     University of Nebraska-Lincoln____
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     __ __
> 
>     _______________________________________________
>     developers mailing list
>     developers at lists.quantum-espresso.org
>     <mailto:developers at lists.quantum-espresso.org>
>     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
> 
> 
> _______________________________________________
> developers mailing list
> developers at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/developers
> 

-- 
Lorenzo Paulatto - Paris