I am trying to follow the methods used in this paper (http://arxiv.org/pdf/1208.3023.pdf) to construct the Hamiltonian of a graphene cone, but taking into account the spin-orbit coupling.

The paper constructs the Hamiltonian for a graphene cone with a Haldane mass term of the form $m \tau_z \sigma_z$.

I wish to take into account the spin. So, instead of the mass term having the form $m \tau_z \sigma_z$, it should look like $m \tau_z \sigma_z s_c$, where $s_c = \vec{n} \cdot \vec{s}$ accounts for the spin projection on the surface of the cone. The spin degree of freedom can easily be added to the massless term of the Hamiltonian by putting $s_0$ at the end (since it does not affect it).

(If this were flat graphene, the mass term would be $m \tau_z \sigma_zs_z$ like in the Kane-Mele model.)

The problem is that when I construct my Hamiltonian, I don't know "when" to add the mass term. Do I start from the standard 2D graphene Hamiltonian (with spin), add this mass term, and then use the required formalism to take into account the conical topology?

Or can I already start from the cone's massless Hamiltonian (which is already in block-diagonal form) and simply add the mass term to it?

Ideally I would like to obtain a Hamiltonian which is in block-diagonal form, which is attainable if we neglect the spin (as shown in the paper).


1 Answer 1


Well... I worked through both approaches and got identical results. The mass term played nice and I was able to obtain a block-diagonal, radial Hamiltonian.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.