[Wannier] About disentanglement keywords in wannier90.win

[xiangru kong]

Dear all,

I am seeking help for the parameters in wannier90.win.

The 1st one I am confusing is num_wann and num_bands.
In my opinon now, if i want to focus on all the valence bands only for the
insulators , the num_wann and num_bands should be the same, for this is the
isolated bands. However, if I want to include some conduction bands, the
num_bands should be larger than num_bands for this is the case entangled
bands. Is it right?

The 2nd is the disentanglement parameters.

The user guide tells us that the disentanglement keywords will be actived
if  num_wann < num_bands, but if the num_wann=num_bands, does the keywords
in disentanglement still works well ?
More I want to know is how to set correctly the dis_win_max(min) and
dis_froze_max(min)?

the 3th
how can I make sure the results by wannier90 is good ? Is comparing the
band structures by wannier90 with DFT a good way?

the 4th

if I have atom orbitals projected like s, px, py, pz, what is the
difference between projected WFs (num_iter=0) and MLWFs (num_iter=1000 as
an example)?




All the best to all
Xiangru Kong
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20160521/9a587b22/attachment.html>