[Wannier] Help: Optimising low lying bands for transport calculations

[Giovanni Pizzi]

Dear Jesse,

>
> Is there any way I can treat them (the lowest two bands) 
> specifically/specially? If I'm not mistaken, when specifying a frozen 
> window it must encompass the entire band which as you can see leads to 
> something like 9 bands being in the frozen window not just the two I'm 
> interested in.

I'm not sure to understand what you mean, but you don't need to freeze 
the whole band; a suggestion I have is to set the frozen window at 
(roughly) the Fermi energy, and see if you can converge.
Check instead to choose a outer window large enough to include the whole 
bands (or anyway bands of similar symmetry, even if they are 
disconnected due to anticrossing).

Best,
Giovanni


-- 
Giovanni Pizzi
Post-doctoral Research Scientist
EPFL STI IMX THEOS
MXC 340 (Bâtiment MXC)
Station 12
CH-1015 Lausanne (Switzerland)
Phone: +41 21 69 31124




On 07/17/2015 01:24 AM, Jesse Vaitkus wrote:
> Hello Wannier mailing list,
>
> I am trying to use Wannier90 to replicate the low lying conduction 
> bands of a very entangled set of bands but I haven't been able to make 
> much progress. I'd like to use them for conduction properties and as 
> such I'd like to get them as correct about that point as possible.
>
> As you can see below, the Fermi energy (black line) sits low in the 
> conduction band and as such I only care about bands in the local 
> vicinity; unfortunately the bands are very intertwined (there are 
> several hundred more above the visible axis).
>
> DFT calculations were made using a 6x1x1 grid as the amount of padding 
> in the y,z directions should mean that Y and Z are being folded to 
> zero; the corresponding image goes from Gamma to X.
>
> Is there any way I can treat them (the lowest two bands) 
> specifically/specially? If I'm not mistaken, when specifying a frozen 
> window it must encompass the entire band which as you can see leads to 
> something like 9 bands being in the frozen window not just the two I'm 
> interested in.
>
> In any case, I have done this and the runs do not converge nor do the 
> WF centres make much sense. I would provide some of the output 
> bands/centres but they change greatly from case to case and often 
> don't make any sense at all.
>
> If you think Wannier functions are not suited to this particular 
> problem, I am open to suggestions for possible alternatives.
>
> Inline images 1
>
> Cheers,
> Jesse Vaitkus
>
>
> -- 
>
> ————————————————————————————————
> Mr. Jesse Vaitkus MAIP
> BAppSc(Phys) Hons, BAppSc(Nano)
>
> PhD candidate
> Room 14.6.02
>
> Chemical and Quantum Physics
> School of Applied Sciences
> RMIT University
> Victoria 3001
> Australia
>
> jesse.vaitkus at rmit.edu.au <mailto:jesse.vaitkus at rmit.edu.au>
> ————————————————————————————————
>
>
> _______________________________________________
> Wannier mailing list
> Wannier at quantum-espresso.org
> http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20150717/e31e17c2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 10892 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20150717/e31e17c2/attachment.png>