[Wannier] Wannier90 projection related few queries

[Valerio Vitale]

Dear Anindya,

/1.If the initial projection of a material( suppose silicon with 
projection=sp3) doesn't give the correct wannier band structure in such 
a situation what should we do?/

It very much depends on the material, e.g. insulator/semiconductor or 
metal, the type of calculation you are doing---simple wannierisation or 
disentanglement+wannierisation---as well as convergence with respect to 
the k-points grid. If you are interested in a specific material, it'd be 
helpful if you could give us more information about what you want to do 
and what you have tried so far.

Some obvious options are: 1) Try with random projections. 2) Try the 
same initial projections but with a different orientation (look at the 
user guide for details). 3) Modify the frozen and outer energy windows 
(if any). 4) Try to converge with lower thresholds.

/2.After relaxation, we can check the relaxed coordinate of the atoms of 
our desired unit cell and we can also find the fractional relaxed 
coordinate values of atoms.Can we directly apply these fractional 
coordinate values in the card proj(1),proj(2) etc and can put the 
orbital along the coordinates?/

This should be clear from the user_guide. You can specify any fractional 
co-ordinate in the projection block and attach to it any orbital in the 
list given in the user guide (chapter 3, pag 45-46). The syntax is:

begin projections
f = frac_x , frac_y, frac_z: s (or p,d,sp3, etc.)
.
.
end projections

where the letter "f" instructs the code you are going to specify 
fractional co-odinates and frac_x, frac_y, frac_z are the actual values 
of the x,y,z components of the relaxed co-ordinates of your atom (or any 
other point in the unit cell). This is equivalent to use the atom symbol 
(for example Si), if the atomic co-ordinates in the atoms block are the 
relaxed ones. The syntax in this case is:

begin projections
Si:s (or p,sp3, etc.)
end projections

/3.Is there any other method to compute the correct projection of materials?
/

In the new release of the code there is going to be a new method to 
compute the projections, which doesn't rely on initial guesses, but we 
are still working to make it robust.

Best wishes,

Valerio


On 23/08/18 12:59, Anindya Bose wrote:
> Dear Experts,
> It is always a tough job to obtain correct wannier band structure 
> which is exactly similar with pw band structure.If wannierization is 
> correct,one should obtain correct band structure and band gap also.To 
> the best of my knowledge,convergence of band structure is mostly 
> depend on the projection.Now I would like to know;
> 1.If the initial projection of a material( suppose silicon with 
> projection=sp3) doesn't give the correct wannier band structure in 
> such a situation what should we do?
> 2.After relaxation, we can check the relaxed coordinate of the atoms 
> of our desired unit cell and we can also find the fractional relaxed 
> coordinate values of atoms.Can we directly apply these fractional 
> coordinate values in the card proj(1),proj(2) etc and can put the 
> orbital along the coordinates?
> Will it make the wannier band structure correct?
> 3.Is there any other method to compute the correct projection of 
> materials?
>
> Thanks and regards,
> Anindya Bose
> Research Fellow,
> IIIT Allahabad
>
>
> _______________________________________________
> Wannier mailing list
> Wannier at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/wannier

-- 
Postdoctoral research associate
Imperial College London
Exhibition road
London SW7 2AZ
and
TCM Group, Cavendish Laboratory
19 JJ Thomson Avenue
Cambridge

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20180824/c39fc24c/attachment.html>