[Wien] Question!

[Peter Blaha]

> Actually, I am cacluting a polymer with 70 atoms in the unit cell. I have
> faced 2 questions as below:
>
> 1. I always get WIEN97 stopped after LAPW0. The reason stated in dayfiel
> is "lapw1  -c   (15:43:19) Filesize limit exceeded". Then, I found that
> the ".vec" file is larger than 2GB (the filesize limit of Redhat linux).
> Is there any ways to reduce the size of ".vec" file?

How many k-points do you use ?
For a polymer with 70 atoms ONE !!! (or maybe a few more if it is
conducting,....) should be sufficient (at least for scf calculation).

> 2. If my target is the bandstructure of my molecule but I am not sure the
> geometry of my molecule. Is it I should perform both "Optimize internal
> param." and "Optimize Vol. or c/a" ? If it is so, which one should I
> perform first?

Most likely the cell parameter of that polymer are known with reasonable
precision, while the internal coordinates are not.
Thus I would do a minimization of the atomic positions (min_lapw)
using the experimental cell parameter.


                                      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/
--------------------------------------------------------------------------