[Pw_forum] Issues running epsilon.x

[Alexandru Georgescu]

Hello,

I am currently running into issues with the epsilon.x postprocessing code,
qe version:   PWSCF v.6.1 (svn rev. 13369). Namely, I run an AFM NiO
calculation using pwscf.x. I use tetrahedra for occupations, and obtain an
antiferromagnetic insulator with a 2.5eV band gap.

When I run epsilon.x with an input like:

&inputpp

outdir='./scratch'

calculation='eps'

prefix='NiOk'

/

&ENERGY_GRID

 smeartype='gauss'

 intersmear=0.10d0

 intrasmear=0.10d0

 wmax= 30.0d0

 nw=7000

 shift=0.0d0

/


 The program gives me that:


     Fermi energy [eV] is: 12.10491

     The system is a metal...

     metalling system: redefining wmin =     0.001000



I've tried various smearings, including 0 for the two parameters and
consistently got this error. This leads for example to a failed ieps
calculation and wrong results for jdos. It seems to detect the Fermi level
correctly however even though it doesn't detect the band gap.


Alexandru
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170910/41cc24bf/attachment.html>