How to fix a space group for a given supercell ?

Message

ypipon · #1 Post by **ypipon** » Fri Jun 12, 2009 10:49 am

Hi,

We have currently a very simple (I assume) problem.
We try to create vacancies and divacancies in a 64-atoms supercell (we remove one or two atms in the POSCAR file). The problem is that when VASP runs, it does not recognize the primitive CFC 8-atoms cell and consequently is very low and give a false space group.

How can we resolve this problem ?

Thank you for your help.

alex · #2 Post by **alex** » Fri Jun 19, 2009 12:23 pm

Hi,

this you have to do on your own. A change to the primitive cell is not implemented (like in CRYSTAL).

Alex

panda · #3 Post by **panda** » Tue Jul 07, 2009 4:33 am

Yes, unfortunately you must calculate the lattice positions in the primitive cell. You can consult the International Tables of Crystallography for the symmetry operations for your respective space group and selective remove atoms from the cell by hand based on symmetry constraints. You might find some of these resources of use as well: http://mt.seas.upenn.edu/Archive/Graphics/A/

centauris · #4 Post by **centauris** » Sat Jul 11, 2009 6:06 pm

you got to remove all the atoms that occupy the same wyckoff positions. I don't think your strategy for vacancy is applicable for cubic system unless you use a large unit cell.