[Wien] questions about "supercell"

[Lukasz Plucinski]

Hello,

> 1. after I modify the supercell by replacing one or more atoms, should I
> relax the new modified structure using "min_lapw"?

In general yes.

> 2. can I terminate the layers or slabs with any specific atoms I want
> such as H?

People usually use partially charged H to terminate one side of the slab.
I don't know if its possible in WIEN. Certainly using 'normal' H atoms is
possible in WIEN.

> 3. usually what the thickness for lays (slabs) and the vacumm region in x,
> y, z directions are suitable?

If you simulate the surface z is perpendicular to the surface. You are
not 'choosing' x and y - they are defined by the symmetry of the system.
Use smallest possible unit cell because large slabs need a lot of time to
relax and calculate. Some groups use 8 layers for the slab separated by 10
A of vacuum (see PRB 64 035305, 2001). I try to use much more, maybe 20
layers separated by 15 - 20 A of vacuum. However, I relax surface on much
smaller slabs, since it takes too much time.

The bigger the slab the more electrons you have. The more electrons you
have the easier it is to separate between the bulk and surface ones. I
also don't think that e.g. 6th or 7th layer is already a fully 'bulk'
layer. On the other hand what matters is comparison to the experiment.

Regards,

Lukasz