[Wannier] Ab initio based Tight binding model

[Jonathan Yates]

On 8 Apr 2010, at 14:20, Elie Moujaes wrote:

> Thanks again for the clarification. I was not aware of this difference in scale thing. Of course by just looking at the plots, you can figure out that there is a misplacement because the energies are the same but the x axis positions are not. Is there anywhere I can read more about this scale difference because I looked at the output of both W90 and PWSCF and found that the poisitions of the atoms in cartesian coordinates are the same so I am not quite sure where does it show that there is a difference in the scale of the x axis in both methods.

 Let me describe a simplified version of the problem you are trying to understand: You have two kpoints. By convention we express the kpoints in terms of the reciprocal lattice vectors. However, to make a bandstructure plot we  want to know the distance between the two kpoints (a suitable unit would be inverse angstroms). If the cell is orthorhombic the conversion is trivial - for a general cell it's also a simple problem but might involve more math. I suggestion reading Chapter 4 of Richard Martin's book 'Electronic Structure'. Note the discussion starting at the end of page 83 regarding finding lengths in real and reciprocal space. He introduces the real and recip space metric - if you look in the wannier90 source code (plot.F90) you'll find we adopted a similar notation.

 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/