[Pw_forum] HELP !!! How to know the DOS out file right or wrong ?!

[Nga Do]

​
Dear QE users,

I am a new QE user and trying to calculate the electron concentration of
pure ZnO and some different kind of defects in same system from DOS file.
But the problem is I am not sure that the DOS file that I calculated is
right.
Please give me some advices and suggestions !!!

These are my following steps to calculate DOS ( the attached files below)
1) scf calculation
2) none scf calculation
3) calculate DOS. When I plot the DOS I did shift the Fermi level (9.5021
eV) (I take it from scf.out file ) to zero ( "F_zno_1.dos" file. ) .

Mesh of k-points in scf calculation is 4x4x3. Firstly, I keep the same mesh
of k-points for the n-scf calculation.
I set "occupations = 'fixed'" to find the gap first, and then set
"occupations = 'tetrahedra' to the DOS step.
But when I checked the DOS file, I found that there were still states
between highest occupied ( 8.8056 eV) and lowest unoccupied level  (9.6268
eV).
I plot the DOS and can see where is the gap, but why there are states in
the gap ?!
I timed double the mesh of k-points for n-scf calculation 8x8x6 and the
results did not change much. I think the gap from the Fermi level to the
bottom of Conduction band should not have any states in it.
Please check my calculations and give me some advices about this issue!?
Here is the link to attached files, please take a look:
https://drive.google.com/open?id=0BwnHIyCcZ3jbb0c4eDBXZldOc1E

Best regards,
Nga​

-- 
Nga Do
Ho Chi Minh City Institute of Physics
Third floor, Building 1, Mac Dinh Chi street, Ben Nghe Ward, District 1, Ho
Chi Minh City, Vietnam
website: www.vast.ac.vn
Mobile: +84 938 672 414
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20171001/8633c440/attachment.html>