[QE-users] Doubling the k points - Reg

[rekha sharma]

Hii,
I am not expert on QE but I can guide you here.

How you can say that 4 4 4 0 0 0 is fit for your scf? In the same way you
can decide for nscf.

I usually do a convergence test for k-mesh with 2X+2, 2X+4,2X+6 and so on,
where X is my starting mesh size, and see whether my required properties
are meeting the convergence criteria, i.e. 10^-3, 10^4,.... ect, Ry.
If your system is complex (many atoms in the primitive cell) then you can
check for energy per atom (above energy/number of atoms in primitive cell).

How to get the value of difference in energy:
(grep ! Nscf_2X+2.out) - (grep ! nscf_2X+4.out) and so on.


I could be wrong and expert can correct me too.

Best wishes
Rekha




On Fri, Jul 3, 2020, 16:33 singaravelan T R <trsingaravelan27 at gmail.com>
wrote:

> Dear sir,
> For example say :Si. For that I have done the proper convergence test and
> my convergence says using my brain I am not doing blindly sir . It is 4 4 4
> 0 0 0 and  is fit for the scf calculation.
> *My question is: For nscf calculation will I need to double the k points
> or not?*
> *Or there any procedure (like optimization of k points in scf calculation)
> for fixing the number of k points for nscf calculation too.*
> *If it depends on material then please guide me in this regard.*
> With thanks,
>
>
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200703/38e83491/attachment.html>