Talk:Andreev reflection
![]() | This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||
|
This article would profit a lot from a nice figure I think. If anybody with artistic talent reads this, please add one to the page. I imagine a picture of an NS-interface with trajectories of electron, hole and maybe an indication of the superconducting density of states or the gap.
What is still missing are maybe some formulae for transmission/reflection amplitudes. And of course everything could be explained more understandably. Bamse 15:23, 25 July 2006 (UTC)
I've drawn a simple illustration and added it to the article. Improvements are of course welcome. —Ilmari Karonen (talk) 23:16, 17 August 2006 (UTC)
Spin of the hole
[edit]The spin of the hole is opposite to the electron; I've corrected this in the text, but the figure still needs correction Brienanni 19:17, 2 December 2006 (UTC)
- I don't think that is correct. In order to form a Cooper pair we need two electrons with opposite spin. One of the electrons is the incoming electron with spin up (see figure). The second electron (with spin down) is taken from the left side and appears as a hole. This hole is thus a missing spin-down electron, and has effectively spin-up! Bamse 15:25, 3 December 2006 (UTC)
To resolve this issue, which is really a matter of definition, I have rephrased the text to refer to the spin bands: an electron in the spin-up band is converted into a hole in the spin-down band. I have also mentioned the suppression of Andreev reflection in a ferromagnet, if only one of the two spin bands is populated. I still think that the figure should be adjusted to illustrate the different spin bands of electron and hole, but I gladly leave that to the artist. Brienanni 20:15, 4 December 2006 (UTC)
- Thank you. Now it is very clear. As for the figure, I contact the artist. Or maybe it is better to omit spin at all in the figure and have a different angle of incidence instead-could be interesting. Bamse 01:40, 5 December 2006 (UTC)
- I've updated the illustration to flip the spin of the hole and to adjust the incidence angle. I agree that the issue of the spin of the hole is confusing either way; nominally the hole must of course have the same spin as the incoming electron, in order to conserve total spin, but this obscures the fact that it occupies a different spin band. Incidentally, the issue could be easier to explain if we had a proper article behind that link; I'll leave its creation to someone who actually knows something about physics. (We do, it turns out, have an article on "Spin (band)", but that's not quite the same thing...) —Ilmari Karonen (talk) 06:59, 12 December 2006 (UTC)
Have rewritten a nicer, more detailed discussion of it and added a bit about CAR (phd interest!)- will return to add citations to papers/books. Wikicommons CAR diagram would be good!
With regards to above diagram discussion, the retroreflected hole is of opposite spin to the incident electron — simplistically AR takes a conduction electron from the normal metal up spin band and one from the down spin band (which is promoted into the conduction band from the down spin valence band giving a down spin hole) which then can form a Cooper pair. (think the full mechanics are a bit beyond what wikipedia needs!) The hole has the exact opposite momentum, wavevector and velocity (at the Fermi energy) as the incident electron - so the diagram is a reasonable view of what is going on; might be better to have the incident electrons perpendicular to the interface though. Jw2034 01:48 21 December 2006 (UTC)
Hello, anyone else thinking that Andreev-bound states in superconducting/semiconducting contacts should be mentioned, as well as the resulting subharmonic gap structure? Maybe there should be only a reference and a -more detailed- article about the BTK-model? Any suggestions? —Preceding unsigned comment added by 78.54.137.99 (talk) 17:38, 18 January 2010 (UTC)
- I agree, multiple Andreev reflections, subharmonic gap structure, etc all belong here.bamse (talk) 21:28, 18 January 2010 (UTC)