[Wien] Slab calculations

[Robert Laskowski]

hi,

The highest symmetry the less k-points in the IBZ. 

For the slab use only 1 k-point in the z direction. 

Lower RKMAX for the first geometry minimization.

If you have more then one core available, use them.

regards

Robert

On Tuesday, August 31, 2010 16:30:13 Bartosz Slomski wrote:
> Dear users,
> 
> in order to safe computational time I would like to aks you about a good
> strategy to perform slab calculations.
> The goal is to calculate Pb slabs with different lattice constants
> taking SOI into account because of the high Z.
> 
> So far I generated a slab of 20 bohr thickness seperated by 30 bohr of
> vaccum (using octave) and run the initialization procedure.
> 
> My first question is about the initialization of the structure. Should I
> allow to lower the number of inequivalent atoms lets say from 24 to 3
> and how many k-points should I at leaset choose.
> 
> 
> In my first calculation I ran the SFC cylcle with 100 k points and
> without inversion symmetry. It takes nearly 3 hours per one cylcle
> without spin-orbit.
> 
> After that I would like to include spin-orbit and run the SFC cyle again.
> 
> Thanks for every hint on this topic.
> 
> With best regards,
> 
> Bartosz

-- 
Dr Robert Laskowski
Vienna University of Technology,  Institute of Materials Chemistry, 
Getreidemarkt 9/165-TC, A-1060 Vienna, Austria
tel. +43 1 58801 15675               Fax  +43 1 58801 15698