The ABINIT Users Mailing List ( CLOSED )

[verstraete@pcpm.ucl.ac.be]

It looks like your analysis is already pretty far into the problem. I've 
never actually seen the gradient in the vacuum, but it is the problem.

> So my questions are:
> 1) Can Abinit handle polar slabs?
> 2) Does Abinit have, or plan to, incorporate such a dipole layer in the 
> vacuum?
> (or do I have to tackle it myself...?!)
People have talked about it, but no one has done anything. There are
other, more elegant and systematic screening techniques to make things
converge. Someone (who?) was also working on a cutoff of the Hartree
potential in the direction perpendicular to the slab, which should have
beneficial effects as well. If you do tackle it, we'll plant a tree in
your name.

> 3) Has anyone managed to converge geometry for polar slabs, 
>  without just making the slab symmetric? (this would make my slabs too 
> large)
Yes a few times, but the vacuum has to be huge and the polarity not too 
important.

> 4) Has anyone any further suggestions or tricks for converging such
> structures,
> other than whats in the FAQ, etc.? Am I simply missing something?
> It was suggested to me that it might be a problem with the pseudo...
> 5) Where does the "sensitivity to the vacuum thickness" come from?
> Would a molecular dynamics calculation help?
I don't know exactly where it comes from, but it looks oscillatory: even
for very large vacuums, a small difference in acell can make the
difference between converging and not. I often have problems converging
the self-consistency as well. 

My most robust combination is ionmov 2 iscf 3, tolvrs 1.e-5 or 1.0e-6...
(if you can afford it): this assures that the potential is converged,
which is much more difficult than just converging the density-filled part
which contributes to the energy and forces. The potential in the vacuum is
hard to converge because of the low density: it has almost no incidence on
the energy, and hence no restoring force.  Getting resvrs "low" is
typically 2 times longer than getting resetot or resfor "low".

Making a thinner but symmetric slab would also be cleaner. Do you really 
need 11 layers?

Matthieu

-- 
===================================================================
ATTENTION!   PHONE NUMBER HAS CHANGED!!!    010 47 3359   !!!

Matthieu Verstraete               mailto:verstraete@pcpm.ucl.ac.be 
PCPM, Boltzmann, pl. Croix du Sud, 1            tel: 010/ 47 33 59
B-1348 Louvain-la-Neuve  Belgium                fax: 010/ 47 34 52