Re: [Kwant] Reflectionless contacts

Hello Anton, So, my contacts are exactly of the same structure and have the same TB model defined as the scattering region. I was more concerned about the semi-finiteness of the leads, and wanted to make sure that I was not just defining two finite regions at the edges of the scattering region. But from what I see leads actually do take shape functions and not necessarily just translation vectors. Regards, Eleni Quoting Anton Akhmerov :

Hi Eleni,

It's a hard problem, and usually we reduce reflection by creating a smooth interface between the scattering region and the contact. As far as measuring the quality of such interface goes, I think the most practical approach is to replace the scattering region with another lead with dispersion mimicking that of the scattering region as closely as possible.

Best, Anton

On Sun, Apr 22, 2018 at 12:13 PM, wrote:

...

Hello everyone,

I have defined two semi-infinite contacts in my system as follows:

-------------------------------------------------------------------- def lead_shape(p): x, y, z = p return -10 < x < 10 and -6 < y < 6 and 0 < z < 30

sym_lead = kwant.TranslationalSymmetry(lattice.vec((-6, 0, 0))) lead = kwant.Builder(sym_lead) lead[lattice.shape(lead_shape, (0, 0, 15))] = 0 --------------------------------------------------------------------

Is there an easy way to calculate any reflections from the lead back into the conductor? I mean, ideally, I would like all electrons entering the contacts to stay there.

Regards

Eleni

-- Dr. Eleni Chatzikyriakou Computational Physics lab Aristotle University of Thessaloniki elchatz@auth.gr - tel:+30 2310 998109

-- Dr. Eleni Chatzikyriakou Computational Physics lab Aristotle University of Thessaloniki elchatz@auth.gr - tel:+30 2310 998109