[Wien] Surface Layer Optimization

[Chiung-Yuan Lin]

Dear Prof. Blaha,
  I have 2 more questions:
1. Is there a program in WIEN2k that can do surface relaxation? I don't
seem to find one.
2. If I need to run one SCF cycle for each relaxed surface layer position
by myself, do I try to minimize the total energy in case.scf? The problem
of using forces is that I see the surface and next layers change about the
same amount of forces in the opposite directions. For example, when I
push the surface layer into the slab (all other layer fixed), the force of the
surface layer changes from -169 to -140, and the next layer goes from -22
to -44 (in mRy/a.u.; minus means into the slab). How can I tell which one
is the optimized by looking at those changes?

Thank you
Chiung-Yuan

Peter Blaha wrote:

> You should first optimize theoretically the bulk lattice parameter and use
> this value for the slab.
> Then use the forces and relax the surface layers of your slab. Do not
> change supercell lattice parameters of the slab.