[Wien] symmetry and *in2

[Peter Blaha]

Doing a calculation with less symmetry operations should be possible, as
long as you have enough SO to transform equivalent atoms into each other.

However, I don't see any reason to do this (except in special cases for
ELNES,...)
Reducing the SO should increase the k-mesh, because you calculate some
k-points twice (or more often) which MUST have identical eigenvalues.

In addition the K-vector list of the fourierseries increase, because
equivalent k-vectors become nonequivalent.

If you do get different E-tot then something is not correct. As far as I
could see from a quick check (sgroup) the 4 symmetryoperations should be
correct.

Probably this needs a more systematic investigation from the very
beginning (first iteration: compare eigenvalues of 2 and 4-calc, compare
equivalent k-points,....)

> We calculated 5 layers Pd(113) surface by wien2k code.
>
> At first, we used the attachment 4_symmetry.struct
> file. "x symmetry" generated 4 symmetry. But if we
> accepted all these initial files and continued SCF,
> :DIS is converged well but the total energy is not
> reasonable (a little bit too small). After we checked
> the symmetry operation, we thought the following two
> symmetry operations are not right.
>
> -1 0 0 0.0000000
> -1 1 0 0.0000000
>  0 0-1 0.0000000
>
>  1 0 0 0.0000000
>  1-1 0 0.0000000
>  0 0 1 0.0000000
>
> We simply kicked the two symmetry operations out
> structure file, left the other right ones in this
> file. Then we got a new structure file
> (2_symmetry.struct). "x symmetry" generated WARINGs in
> the 2_symmetry.outputs file.
>
>  WARNING !!!!!!
> nsym found by symmetry differs from iord read in
> struct   4   2
>
> If we didn't care of this warning, "init_lapw" can be
> finished without any other errors.
>
> But from the two different structure file with
> different symmetry operations, we get the same *in2
> file. Do you think it is right? Or can I simply change
> symmetry operation in structure file and run
> "init_lapw" without any other changes?
>
> (By the way, if we used 2_symmetry.struct file to
> generate initial file without caring of WARNING in
> symmetry, the total energy from SCF seemed good).
>
> Regards
> Zhang
>
>
>
>
> __________________________________
> Do you Yahoo!?
> SBC Yahoo! - Internet access at a great low price.
> http://promo.yahoo.com/sbc/


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