Dear Robert,

                  Thanks for your e-mail !. I have calculated graphene bands 
with p2 orbitals in a supercell so i know that it is not so much 
computationally expensive. In this case i would need to use only the pz 
orbitals to reduce the tight binding hamiltonian so i would whish to improve 
the bands calculations  that correspond to the pz orbitals by means of some 
parameter of wannier90 or some calculation tecnich.
Thanks in advance, Eduardo.

De: Robert Peyton Cline <>
Enviado: jueves, 26 de septiembre de 2019 16:14
Para: Eduardo Perassi <>
Cc: <>
Asunto: Re: [Wannier] (no subject)

Dear Eduardo,

In my experience when dealing with graphene, the wannierization works best when 
you include the sp2 orbitals with the pz orbitals.  This allows you to use a 
frozen window on all the states up to the Dirac point, and then your outer 
window can include any conduction bands that significantly contribute to the 
pi/pi* bands.  For a supercell like yours and on a remote computer cluster, 
this shouldn't be that computationally expensive.

Please refer to page 1430 in this 
paper<> (figure 6 
and related text) for this procedure for a single cell of graphene.

Peyton Cline
5th Year PhD Student
Eaves Group

On Thu, Sep 26, 2019 at 1:09 PM Eduardo Perassi 
<<>> wrote:
Hi, I am using wannier90 to reproduce some bands of a big unit cell of graphene 
with 24 carbons atoms. I am using pz orbitals and I did not include sp2 
orbitals so I am trying to reproduce only 24 bands. I attached to this email a 
picture of the bands. The image shows the bands which are affected by a 
"ripple" so, is it possible to remove the ripple by increasing the number of k 
point or any other parameter?  I will appreciate any help!, thanks.
Wannier mailing list<>
Wannier mailing list

Reply via email to