[gpaw-users] Problem with workfunction calculation

[Jens Jørgen Mortensen]

Den 14-05-2012 13:46, Fredric Granberg skrev:
> Hi!
>
> I'm doing the exercise on calculating the workfunction for a aluminium
> surface on the homepage. It works fine with LDA also when I increase
> the vacuum layer. The problem arises when I use PBE and a vacuum layer
> bigger than or equal to 12.5A. When I run the work_function.py there
> appears some noise on both sides in the plot where the averaged
> effective potential is printed out. This will give a much higher
> workfunction than the one where there isn't any noise. The only thing
> changed from the original file on the homepage is the vacuum thickness
> and the xc='PBE'. Are there any parameters that should be changed to
> get rid of this problem? I also tried the same script with copper and
> got exactly the same problem.

I think this noise comes from the way libxc handles small gradients and 
small densities for GGA's.  For this reason it is better to use the 
electrostatic potential in the vacuum 
region(v=calc.hamiltonian.vHt_g*Hartree).  See this tutorial:

   https://wiki.fysik.dtu.dk/gpaw/tutorials/dipole_correction/dipole.html


Jens Jørgen

> Best regards
> -----------------------------
> Fredric Granberg
>
> fredric.granberg at helsinki.fi
> -----------------------------
>
>
>
>
>
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users