[Wannier] Exiting....something wrong: Too many nearest neighbours!

[Jonathan Yates]

On 15 Mar 2010, at 19:22, Elie Moujaes wrote:

> Dear All,
>  
> i am still executing the wannier90.x command for graphene and I used a better 5x5x1 mp_grid but i keep on getting the error:
>  
> EXITING.... SOMETHING WRONG :TOO MANY NAREST NEIGHBOURS.

Elie,

 I'm afraid my answer is exactly the same as to your previous email:

 For wannier90 the kpoints need to form a regular grid in the full reciprocal unit cell - no symmetry is used.

from your win file
  0.00000000  0.00000000  0.00000000 
  0.00000000  0.25000000  0.00000000 
  0.00000000  0.50000000  0.00000000
  0.00000000  0.75000000  0.00000000 
  0.00000000  1.00000000  0.00000000

while this is a regular grid, you'll notice that (0.0 0.0 0.0) and (0.0 1.0 0.0) are actually the same kpoint. Use kmesh.pl to generate the kpoint grid - it's much simpler.

 Maybe I can repeat the suggestion from my previous reply more strongly. Run example10 (graphite) - check it works and you understand it (it is also discussed in the wannier90 CPC article). Remove one plane of atoms, and repeat. That should give you a working graphene input to play with. 
 The input file you attached will have problems even if you fix the kpoint grid - note that graphene does not have a isolated set of valence bands, so you will need to use the disentanglement procedure.

 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/