[Wannier] Problem with disentanglement

[Christoph Wolf(신소재공학과)]

Dear all!

I am facing some problems with disentanglement using QE6.0

My system is a cubic perovskite CsPbBr3. I calculated nscf with 44 bands (22 valence and equal number of unoccupied bands). I first calculated the pdos to identify the contributions of orbitals to the bands (fatband representation); the output is as below:

[cid:image003.png at 01D2A75A.0B80EEA0]
Here, I chose Pb(6s,6p)+Br(4p).
As test-case I calculate the MLWF of the Pb d-band (located below -12 eV) excluding all bands 6-44

The resulting MLWF look nicely like   d-orbitals so I think it is all right there. Now I want to get the same for the bands shown above; as they are not separated well I want to take the “disentanglement approach” but I think I am doing it wrong;

Using this as input:

num_wann  12
num_bands 12

begin projections
Pb:p !1x3=3
Br:p ! 3x3=9
end projections

dis_win_min -15.000
dis_win_max  15.000
dis_froz_min -2.000
dis_froz_max  5.000

I unfortunately get

Wanted band  : 1 found band  : 13
Wanted kpoint: 2 found kpoint: 1
A common cause of this error is using the wrong
number of bands. Check your input files.
If your pseudopotentials have shallow core states remember
to account for these electrons.

Now I tried to move the frozen window a bit but that didn’t change; I also tried to exclude other bands by the exclude_bands keyword excluding all but the above but it didn’t work either.

I am relatively new to wannier90 so I appreciate any help!

Thank you in advance!

Chris

Materials Science and Engineering
POSTECH university, Korea

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20170327/429333b9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.png
Type: image/png
Size: 65787 bytes
Desc: image003.png
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20170327/429333b9/attachment.png>