[Wien] Improving vacuum basis set

[georg@chem.au.dk]

In principle a planwave basis set should be complete.  I doen't have much
experience with surfaces so I can only give some untried suggestions

1) your GMAX (from in2) is to small? I guess a too low GMAX could give some
artifact in the density (and especially in the xc potential)

2) Your supercell is to small. It's a bit fishy that your artifact's at z=.5

3) An artifact of the functional. GGAs are not really perfect for the tail of
the wave function. 
I must admit I don't believe this possibility myself. I guess it would have been
found before. I guess you have also tried different xc functionals (I remember
a question about B88). I've seen that the vanLeuwen GGA, which was constructed
to have a 1/r potential, is in LAPW0. Maybe this could be interesting for you
(I tried it once. It didn't work "out of the box" and I was to lazy to look
into it)

 Best wishes
   Georg


Quoting Laurence Marks <L-marks at northwestern.edu>:

> Not in fact an atom, rather the electron density. I'm getting some
> relatively flat, small density (~1e-4 via lapw5/rhoplot) far from the
> surface, which appears to be stabilized by it's exchange-correllation
> energy (vxc potential around -0.1 at z=0.5). I'm not sure if it's
> real, or whether the basis-set for this type of density might not be
> poor.
> 
> On 10/23/06, georg at chem.au.dk <georg at chem.au.dk> wrote:
> > Do you mean an atom some distance from (weakly bonded to) the surface?
> >
> > One thing is that WIEN sets the linearization energy to 0.30 because the
> > fermienergy is "usually" around 0.5 Ry in densely packed solids, so 0.3Ry
> can
> > be expected to be somewhere in the valence bands.
> >
> > If you have an atom the upper most state will have a negative energy so
> often
> > the lineraization energy needs to be changed. As the bands are flat
> sometimes
> > the energies have to be set very precisely. Maybe the -in1new option can
> help
> > you.
> >
> >  Best wishes Georg
> > Quoting Laurence Marks <L-marks at northwestern.edu>:
> >
> > > I am running in to some issues with electrons at some distance from a
> > > surface, and I suspect that I might not have the best basis set for
> > > these in a conventional calculation. I'm wondering if anyone has
> > > thought about this problem and come up with some ideas to try.
> > >
> > > --
> > > Laurence Marks
> > > Department of Materials Science and Engineering
> > > MSE Rm 2036 Cook Hall
> > > 2220 N Campus Drive
> > > Northwestern University
> > > Evanston, IL 60208, USA
> > > Tel: (847) 491-3996 Fax: (847) 491-7820
> > > email: L-marks at northwestern dot edu
> > > http://www.numis.northwestern.edu
> > > _______________________________________________
> > > Wien mailing list
> > > Wien at zeus.theochem.tuwien.ac.at
> > > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> > >
> >
> >
> > --
> > Georg Madsen
> > Department of Chemistry
> > Aarhus University
> > DK-8000 Århus C
> > Denmark
> > tlf (+45) 89423885
> > http://www.chem.au.dk/~webuorg/gm
> > _______________________________________________
> > Wien mailing list
> > Wien at zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >
> 
> 
> -- 
> Laurence Marks
> Department of Materials Science and Engineering
> MSE Rm 2036 Cook Hall
> 2220 N Campus Drive
> Northwestern University
> Evanston, IL 60208, USA
> Tel: (847) 491-3996 Fax: (847) 491-7820
> email: L-marks at northwestern dot edu
> http://www.numis.northwestern.edu
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> 


-- 
Georg Madsen
Department of Chemistry
Aarhus University
DK-8000 Århus C
Denmark
tlf (+45) 89423885
http://www.chem.au.dk/~webuorg/gm