Re: Anti Spin-Hall Effect.

[Jose Luis Lado Villanueva]

Hi Denis

that is pretty interesting!

First I want to point out that I almost know nothing about the valley Hall. The fact that vacancy can give rise to valley dependent scattering is actually true (Luis is working on this, he is able to explain this way better!), it can be simply understood by keeping in mind that a vacancy generates an anomalous velocity of different signs in each valley, so that electrons in each valley will tend to scatter to different directions. However the direction of the scattering depends on in which sublattice is the hydrogen atom (Luis's result), so that you will have a net Hall effect if you manage to create a net imbalance between both sublattices (either by adatoms or by substrate), which if you are doing things at random, seems quite unlikely right? Maybe I am missing something...

Jose

[Денис Александрович Бандурин]

Hi guys,

Most of you remember 3 articles from Singapore where they observed Spin Hall Effect in graphene by hydrogenation, substrate and gold adatoms.

Few recent articles suggest that the observed non-locality is due to valley hall effect or anything else.

1. http://journals.aps.org/prb/pdf/10.1103/PhysRevB.91.165412

2. http://arxiv.org/ftp/arxiv/papers/1504/1504.05785.pdf

Dear theoreticians, what do you think, can local defects result in valley-dependent skew scattering and under which conditions?

Thanks!

Denis.

P.S. Jose Luis, I totally agree with you. Local breaking of inversion symmetry will result in non-zero Berry curvature (different for the K and K' valleys) and modify group velocity of charge carriers from different valleys. But in order to observe effect globally there should be order in these defects which is hard to believe to be possible to achieve for the time being. But the order is not important for spin hall? Is that correct?

[Jose Luis Lado Villanueva]

Yes, that's right, for the quantum spin Hall the sublattice is not important, because the effect of the hydrogen adatom is to buckle the graphene locally, so that new channels of SOC arise when electrons pass near the modified structure. So, the spin Hall is triggered by a local change in geometry around the adatom, which increases SOC, and it is independent on the sublattice. On the other hand, for the valley Hall, the effect is triggered by the sublattice imbalance, so it is critical where the atom is.

A easy way to understand it, is that the spin Hall response is even under inversion symmetry, whereas the valley Hall is odd, so that if you have "average inversion symmetry" (which is what you get is you throw atoms at random), the valley Hall is identically zero, whereas the spin Hall can be nonzero. Only if you manage to break inversion symmetry globally, you will get a valley Hall response (which would be for example in the bilayer with electric field).