[Wien] Spacegroup issue

[Peter Blaha]

> I'm running into some trouble with WIEN2k when generating a struct-file for
> alpha-rhombohedral boron (Yes, I know there is a example struct-file for this
> but this is somehow completely wrong).
>
> I'm referring to Spacegroup 166 (R-3m), where I have two cell choices: the
> rhombohedral setup or the hexagonal setup. The rhombohedral would seem
> logical as we only have 12 Atoms in the elementary cell. However, when trying
> to generate a struct-file with a=b=c and alpha=beta=gamma!=90° StructGen
> tells me: "cell parameters are not consistent with space group".
>
> The other possibility would be a hexagonal cell with the drawback of a tripled
> cell volume. Structgen accepts the lattice parameters. But the former two
> six-fold atomic positions (totaling to 12 Atoms) now need to be 18-fold
> according to symmetry. Structgen just generates 12-fold atomic positions.
>
> When trying to import cif-files (both rhombohedral and hexagonal ones) I
> always end up with a mixture of hexagonal lattice parameters and rhombohedral
> atom positions.

Sorry, when I programmed this rhobohedral structure, I had no idea about
cristallography, so I used:

hexagonal lattice parameters    AND   rhobohedral positions!!!

This is the required input, although it is very weared!!!

It is described in the UG (struct file!!!) and was discussed several times
in the ML.

                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------