[Wien] Slab calculations

[koitzsch]

OK the mixing is already 0.1 but I'll try smaller...

Right now the total energy oscillates:
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.957335
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.961041
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.952793
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.946655
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.941920
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.958858
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.945949
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.950903
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.955180
:ENE  : ********** TOTAL ENERGY IN Ry =        -2403.950114

Thanks a lot for the quick answers!

Christian Koitzsch

PS.: Thorsten, I tried even with 10000 kpoints, but this did not improve the
situations, so I guess it is the mixing


----- Original Message ----- 
From: "Martin Gelfand" <gelfand at lamar.ColoState.EDU>
To: <wien at zeus.theochem.tuwien.ac.at>
Sent: Friday, March 26, 2004 4:48 PM
Subject: Re: [Wien] Slab calculations


> You're right that working with a nonoptimized structure shouldn't
> play a role in convergence of the SCF cycle.
>
> It seems to me that, until you get a handle on the SCF convergence,
> you should reduce Rkmax considerably, maybe down to 6, so that
> the calculations will go faster.  Later on, you can see how properties
> converge as a function of Rkmax.
>
> I'd focus on tweaking parameters in mixer (does reducing the mixing
> parameter or switching to the PRATT scheme make a difference?) and
> lapw2 (try a different efmod).
>
> Regards,
> Martin Gelfand
> Dept of Physics, Colorado State University
>
> On Fri, 26 Mar 2004, koitzsch wrote:
>
> > Hi Everybody,
> >
> > I am trying to do a 6 layer slab calculation of Mg(0001) and I get
energy
> > oscillations in the scf cycle. I already used a Rkmax of 10 and gmax of
14.
> > Does anybody have experience for these kind of calculations? What is the
> > crucial parameter to reach convergence? I have not optimized the
positions,
> > but this should not play a role for convergence, right?
> >
> > Thanks a lot
> >
> > Mg MT sphere: 3
> > unit cell: 25 angstroem
> > vacuum thickness 12 angstroem
> >
> > _______________________________________________
> > Wien mailing list
> > Wien at zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>