[Pw_forum] Fwd: Query in conductance calculation

[N H Joshi]

Dear Alexandre Smogunov and Gabriele Sclauzero,
Thank you so much for helping.

---------- Forwarded message ----------
From: Alex Smogunov <asmogunov at gmail.com>
Date: Thu, Oct 31, 2013 at 6:42 PM
Subject: Re: [Pw_forum] Query in conductance calculation
To: PWSCF Forum <pw_forum at pwscf.org>

Indeed, in the antiparallel case you should do calculations with ikind=2
and have two files for
left and right leads with up and down magnetizations. In the pwcond input
file you should also specify
the bds value, where the scattering region ends up, it will be in the
middle of the cell used in pw.x calculation.
regards,
Alex



2013/10/31 Gabriele Sclauzero <gabriele.sclauzero at epfl.ch>

> Dear Niharika joshi,
>
>     I'm not sure to understand correctly your problem, but I think your
> case is similar to one of the examples on the webpage you mentioned, namely
> "Monatomic Ni wire with a spin reversal". You have to double to unit cell
> (in a specular manner) in the scf calculation of the scattering region,
> then use only the first half in the pwcond calculation.
>
> In both cases there is one "junction", but two "interfaces" (one between
> left lead and scattering region and a second between scattering region and
> right lead).
> The difference is that the leads are the same in the parallel case, while
> they are different in the antiparallel one.
> In both cases, what the code computes is what one expects, the conductance
> for the single junction (i.e., for electrons "going" from the left lead to
> the right one through the scattering region).
>
> HTH
>
> GS
>
>
>
>
>
-- 
Niharika Joshi
Project student,
IISER,Pune
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20131101/a142f443/attachment.html>