[Wannier] Ab initio based Tight binding model

[Jonathan Yates]

On 8 Apr 2010, at 00:21, Elie Moujaes wrote:

> Thanks very much, as usual, for the clarification. I am actually trying to compare the band structure of both methods. When I have plotted the two curves (DFT based) and (ab initio TB based using WANNIER90), I realized that the K-point is displaced i.e, in both methods you get the same shape of the curves but in one of them they are at an earlier point (The K point is not the same in both). I looked at the both files and checked positions of atoms and positions of K and M points. I realised that in the .WIN file, the positions of the atoms were still in terms of the lattice vectors whereas they should be in cartesioan coordinates ( either angstroms or bohrs) so I did this correction but it did not seem to have solved the discrepancy. I will attach the ps file just to give you an idea of what i am talking about. Could you please, whenever possible, have a look at teh graph and tell what could have gone wrong.

The answer is pretty straight-forward:
  Think about what the x-axis should be - in the wannier90 plot it is the distance along the recip space path (in inverse Angstroms). For the pwscf bands you've used a different scale.

 Either figure out what the correct x-axis should be (I think in this case the missing factor is 2*pi/a=2.46 but in general it is more complicated), or, if you have used the same kpoints for both calculations you can just take the x-axis scale from the wannier90 calculation.

 Yours
  Jonathan


-- 
Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK
tel: +44 (0)1865 612797                http://users.ox.ac.uk/~oums0549/