[Wien] Error - Fermi level was not found for doping

[Gavin Abo]

As mentioned at

http://qe-forge.org/pipermail/pw_forum/2016-May/109991.html

Maybe this happens when VBM, CBM, and Efermi are the same, and Egap is 0.

In your case.outputtrans, are the values like that under the "OUTPUT 
from BANDANA" section?

If so, I think it was mentioned at

http://cms.mpi.univie.ac.at/vasp-forum/viewtopic.php?f=4&t=9660&p=14325#p14325

that increasing lpfac in case.intrans or fixing a problem with the 
case.energy[up/dn] input may have resolved this.

A common problem with case.energy occurs if a file from the WIEN2k scf 
calculation is not used:

http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg11759.html

 From a code perspective, it looks like this error may be coming from 
lines in boltztrap-1.2.5\src\fermiintegrals.F90:

       if (iiter > 999 .or. abs(sumelec-sumelec_target) > 1.d-12) then
         write (*,*) 'Error - Fermi level was not found for doping ', 
idoping, ' at temperature ', temp
       end if

So it looks like the number of iterations (iiter) became too large 
(i.e., greater than 999) during the search for the Efermi or it looks 
like the absolute value of the difference between the sumelec and 
sumelec_target was greater than zero (where the 1.d-12 is used as a 
tolerance to represent zero as floating-point arithmetic is inherently 
inexact such that floating point numbers might not always compare 
properly for exact equality [ 
https://stackoverflow.com/questions/20764911/floating-point-numbers-equality 
, 
https://randomascii.wordpress.com/2012/02/25/comparing-floating-point-numbers-2012-edition/ 
, http://www.lahey.com/float.htm ]).

On 7/30/2017 11:11 AM, chin Sabsu wrote:
> Dear Users,
> After successful attempt of electronic, optical and magnetic 
> properties, I tried to run Boltztrap calculation.
> In an alloyed compound with different doping level concentrations, I 
> encountered below error.
>
> "Error - Fermi level was not found for doping"
>
> Please suggest me:
> What does it mean and how to overcome this situation.
>
>
> Thanks in advance and help will be highly appreciated.
>
> Thank you
> Chin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20170730/985809f0/attachment.html>