[Wannier] Error when using wannier_ham.x for graphene

[Giovanni Pizzi]

Dear Yan Wayne,
wannier_ham.x is not part of Wannier90, but of Quantum ESPRESSO, so I 
think that you cannot get any help in this forum regarding the error you 
report.

On the other hand, to get H(k) with Wannier90:
I do not recall a specific flag to print the interpolated H(k) [please 
correct me if I am wrong]; however, H(k) is calculated every time one 
asks,  e.g., for a band structure (but typically it is immediately 
diagonalized and only the eigenvalues are printed).

If you want to obtain H(k), the easiest way that I can think of at the 
moment is:

- calculate the Wannier functions using Wannier90.

- Once you obtain the Wannier functions, do the usual checks to verify 
that you got the correct functions (real WF, good band structure 
interpolation)

- then, I suggest you to give a look at the geninterp module, contained 
in the new postw90.x executable of Wannier90 v. 2.0 (see documentation 
http://www.wannier.org/doc/user_guide.pdf at pages 125-126).

The module calculates the bands (and possibly band derivatives) on any 
list of kpoints specified in input.

I suggest that you first check that you get the correct bands; then, to 
get H(k), you need to slightly modify the source code 
(postw90/geninterp.F90) to print H(k).

The lines to look at are around lines 222-223,

           call fourier_R_to_k(kpt,HH_R,HH,0)
           call utility_diagonalize(HH,num_wann,localeig(:,i),UU)

where the first line calculates the H(k) at the given kpoint kpt, and 
stores it in HH. The next line diagonalizes it.
What you should do is to print HH in between (just be careful because 
the logic around those two lines is a little bit complicated due to the 
need to parallelize the routine).

Best,
Giovanni Pizzi



On 05/06/2014 08:50 AM, jiaxu yan wrote:
> Hi all
> Sorry to bother you again. Still the same problem: How to get the 
> k-space H(k) in Wannier functions basis
> instead of the real space Hamiltonian H(R) given in case_hr.dat? After 
> scf and nscf calculations, I want to obtain the model Hamiltonian 
> using wannier_ham.x for graphene. I have tried different version of 
> pwscf and different potential...But it always failed and one error 
> always occurs:
> * %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>      Error in routine wannier_proj (1):
>      wrong orthogonalization on k-point
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%*
> *     stopping ...
> *
> **
> I am struggling and need help.... Anyone can help me fixing this 
> problem? All related files are attached.
> Wayne, Yan
> Division of Physics and Applied Physics
> School of Physical and Mathematical Sciences
> Nanyang Technological University, Singapore
>
>
>
> _______________________________________________
> Wannier mailing list
> Wannier at quantum-espresso.org
> http://www.democritos.it/mailman/listinfo/wannier


-- 
Giovanni Pizzi
Post-doctoral Research Scientist
EPFL STI IMX THEOS
MXC 340 (Bâtiment MXC)
Station 12
CH-1015 Lausanne (Switzerland)
Phone: +41 21 69 31124

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20140506/45aa320b/attachment.html>