[Pw_forum] US

[stefano de gironcoli]

Dear Ashkan Shekaari,
    there is no general rule. it depends on how localized the 
augmentation charges are.
    you should determine this by yourself and do not rely on other 
people suggestions.

    how to determine ecutwfc and ecutrho ?

    it's a 2-step procedure:

  1) Run a convergence test assuming the default ecutrho=4*ecutwfc 
value: that is just specify ecutwfc in the input and leave unspecified 
the ecutrho option. Converge the total energy and monitor stress and 
forces as well to make sure you get the convergence you want.

     This step tells you what is the ecutwfc and ecutrho=4*ecutwfc for 
which the variational freedom of wavefuctions AND the numerical accuracy 
of augmentation-charge integration are BOTH sufficient. The latter is 
usually the harder to get so you should take this value of ecutrho as 
the one needed for your system.

   2) Keeping fixed this value of ecutrho you can now test the 
convergence w.r.t. ecutwfc and see if you can reduce it without 
compromising the convergence of total energy, stress and forces. This 
tells you the value of ecutwfc you should use in your system.

   HTH

stefano


stefano





stefano

On 03/06/2016 09:51, ashkan shekaari wrote:
> Dear experts,
>
> Is it necessary to take the kinetic energy cut-off for the charge 
> density and the potential 8-12 times larger than the energy cut-off 
> for the wave functions for the case of 2D-MoS2?
>
> Doesn't a less value of about 4 times larger than the energy cut-off 
> for the wave functions work?
>
> *--*
> /Best,/
> /Ashkan Shekaari/
> /Plasma Physics Research Center/
> /Science and Research Branch/
> /I A U, 14778-93855 Tehran, Iran./
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160603/e85350d4/attachment.html>