[Wannier] wannier MoS2 bands again

[Aron Szabo]

On 07/02/2015 12:48 AM, Elio Physics wrote:
> Dear all,
>
> I have been trying for quite some time now to reproduce electronic 
> bands of a monolayer of MoS2 bands, but in vain.  The bands I am 
> getting are kind of "wiggly". Anyone can help me on this matter.  the 
> bands ' figure is attached and the .WIN input file is:
>
>  num_wann        =  11
>  num_iter=300
> ! SYSTEM
> exclude_bands:1,2
> search_shells=20
>
> begin unit_cell_cart
> bohr
> 6.0756016508 0.0003917262 0.0000000000
> -3.0374615776 5.2613166712 0.0000000000
> 0.0000000000 0.0000000000 57.9146940176
> end unit_cell_cart
>
> begin atoms_frac
> Mo       0.232815941   0.667313843   0.249637447
> S        0.756568434   0.364728921  -0.264806724
> S        0.756568571   0.364729538   0.764066183
> end atoms_frac
>
> begin Projections
> Mo:d
> S:p
> end Projections
>
>
> begin kpoint_path
> G 0.000000  0.000000  0.000000   M 0.3333333 0.3333333 0.0000000
> M 0.3333333 0.3333333 0.0000000  K 0.5000000 0.0000000 0.0000000
> K  0.5000000 0.0000000 0.0000000 G 0.0000000 0.0000000 0.000000
> end kpoint_path
>
>
> ! KPOINTS
>
> mp_grid : 7 7 7
> begin kpoints
>  0.00000000  0.00000000  0.00000000  2.915452e-03
>   0.00000000  0.00000000  0.14285714  2.915452e-03
>   0.00000000  0.00000000  0.28571429  2.915452e-03
>   0.00000000  0.00000000  0.42857143  2.915452e-03
>   0.00000000  0.00000000  0.57142857  2.915452e-03
>   0.00000000  0.00000000  0.71428571  2.915452e-03
>   0.00000000  0.00000000  0.85714286  2.915452e-03
>   0.00000000  0.14285714  0.00000000  2.915452e-03
>   0.00000000  0.14285714  0.14285714  2.915452e-03
> .
> .
> ..
> ....etc.
>
>
> _______________________________________________
> Wannier mailing list
> Wannier at quantum-espresso.org
> http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier
You should use a denser k point grid in the x-y plane, but you don't 
need more than one point in the z direction, since the third lattice 
vector is quite large. Try a 15x15x1 MP grid (both in the scf 
calculation and in Wannier90). Make sure that the electronic calculation 
converged correctly, and do a band structure calculation with pw.x, to 
see whether these wiggly bands are the results of Wannier90 or the DFT 
calculation. And are you using exactly 13 bands? If not, then you should 
also exclude those bands above 13. And check for the convergence of the 
Wannier functions' spreads. Normally it should be around 2-3 Angstrom 
per function even in the first step, and it should decrease quickly 
below 2 A with these projections. But I think this error is probably the 
result of a non-converged scf simulation. You could also try to increase 
the number of bands in the scf simulation, because depending on the 
algorithm the highest bands might be inaccurate. Then just exclude those 
bands in the .win file.

Best,
Aron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20150702/2c11f3a5/attachment.html>