[Wien] about relaxed structure

[Peter Blaha]

You have done the relaxation in a cell without any symmetry.
Of course the relaxation has been stopped at a certain level (forces 
"small"), but this means, that the positions are accurate only
within a certain limit. Thus the "symmetry-dedection programs" (nn, sgroup 
and symmetry) have problems to find your symmetry, because by internal 
limits they sometimes find two atoms to be equivalent (within a given 
accuray), sometimes not.

You should edit your struct file and "symmetrize" it by hand (there are 
also programs in literature like "kplot" which can do that). Please round 
all values to 4-5 significant digits, i.e.
ATOM  -1: X=0.11530009 Y=0.24999925 Z=0.24999840    change to
ATOM  -1: X=0.11530000 Y=0.25000000 Z=0.25000000
 
or
ATOM  -9: X=0.99999924 Y=0.99999970 Z=0.00000064  to
ATOM  -9: X=0.00000000 Y=0.00000000 Z=0.00000000

This will lead to a highly symmetric structure.
 
PS: Most likely, you could have saved lots of time when you would have 
done the relaxation in a higher symmetry. As far as I can see, only the Te 
atoms have the freedom to move a little bit, Zn and Cd are actually fixed 
by symmetry.

>    The quasi-zinc-blende structure of Cd0.5Zn0.5Te containing 16 atoms have been fully relaxed using PORT method. All of the relaxation procedure are going smoothly before I try to use the obtained structure (see attachment)to do init_lapw, but problem in sgroup as well as the following procedure occure. I am very confudsed about it, and eager for your advice. Thanks a lot.
> 
> Sincerely     
> H. Duan
> National Laboratory for Infrared Physics
> Shanghai Institute of Technical Physics
> Chinese Academy of Sciences
> 500 Yutian Road
> Shanghai 200083
> P.R.China.
> duanhe at mail.sitp.ac.cn	
> 


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