[Pw_forum] band structure, can't I use hybrid functional?
nkxirainbow at gmail.com
Sat Jun 8 15:34:40 CEST 2013
I guess so.
In LDA and GGA functionals, the Hamiltonian is only
related with charge density. Therefore, band structure can be
calculated with nscf.
However, in hybrid functional, the Hamiltonian is not
only related with charge density but also with wavefuncitons.
Therefore, band structure must be calculated with scf, instead of
On Sat, Jun 8, 2013 at 8:15 PM, Davide Tiana <d.tiana at bath.ac.uk> wrote:
> Dear Hui Wang
> thanks for answering.
> I know how to it with vasp, so in QE I had to do exactly the same
> procedure as in vasp?
>> Message: 5
>> Date: Sat, 8 Jun 2013 17:09:31 +0800
>> From: xirainbow <nkxirainbow at gmail.com>
>> Subject: Re: [Pw_forum] band structure, can't I use hybrid functional?
>> To: PWSCF Forum <pw_forum at pwscf.org>
>> <CABBv2WLo+K3peFv5jQRtf0gq33aJtC2TQwpZ3bPe0kppbhAqtQ at mail.gmail.com>
>> Content-Type: text/plain; charset=ISO-8859-1
>> Dear Davide:
>> I have a tip to do band calculation for band with HSE
>> hybrid functional.
>> 1:do scf calculation on a regular k-mesh (for example
>> 4*4*4) and save charge density and wavefunction.
>> 2: write the k mesh by hand: Add additional k points to
>> the regular k-mesh and set the weight of these k point to zero.
>> 3: do scf calculation again.
>> You can find more information at here:
>> It is HSE band calculation procedure using VASP.
>> Hui Wang
>> School of physics, Fudan University, Shanghai, China
> Pw_forum mailing list
> Pw_forum at pwscf.org
School of physics, Fudan University, Shanghai, China
More information about the users