[Wien] "X-FORCE for atom 1 is not zero as required by symmetry"

[foyevtsova at th.physik.uni-frankfurt.de]

Dear Peter,

with your recipe the problem is indeed fixed.

It turns out after all that my in2 was wrong as a result of producing and
running a series of struct files with varying degree of BiO6-octahedra
tilting. I performed initialization for a structure with no tilting and
then ran scf for a sequence of structures with increasing tilting without
reinitializing (with same in2). Now I see that initialization gives
different in2 in the case of tilting and no tilting, probably because of
the lower symmetry in the former case, even though nominally the space
group of the unit cell stays the same. I've got the original structure
from a colleague and so do not know about the origin of the rot matrices.

Thank you for your help and time.

Best,
Kateryna


>> 1) I am using version 13.1. Still, min_lapw has changed the space group
>> number from 14 to 4, which I only noticed after Gerhard had pointed this
>> out. The regular scf cycle was done with the correct space group in
>> .struct and produced "X-FORCE for atom 1 is not zero as required by
>> symmetry".
>
> The spacegroup-label in the struct file is NOT used at all, except in the
> structgenerator of w2web.
> In other words, you can put there anything and as long as you do not load
> such a struct file into structgen at w2web and press SAVE/ACCEPT CHANGES,
> there
> is no effect-
>
> Nevertheless, I'll modify the structgenerator, such that it writes a blank
> in between #of atoms and SG number, as this is very confusing.
>
>> 3) If possible, I'd like to keep this space group or, at least, the unit
>> cell dimensions for further analysis.
>>
>> So, can I conclude that it is not possible to optimize the position of
>> the
>> first atom using this space group/unit cell dimensions because of not
>> accepting the space group suggested by sgroup? Perhaps, I should try
>> reducing the symmetry to P1...
>
> It should be possible. I did not check it very carefully, but to my
> understanding
> you do NOT need to follow the suggestion of sgroup and can keep the unit
> cell as
> desired. (I still do not know who wrote these local rotation matrices into
> case.struct ???)
>
> Please transfer the struct file to a new directory.
>
> Put the number of symmetry operations to ZERO in case.struct
> (use an editor, not "structgen" to avoid any problems with the SG-label).
>
> Then reinitialize again and run mini again.
>
>
> --
> -----------------------------------------
> Peter Blaha
> Inst. Materials Chemistry, TU Vienna
> Getreidemarkt 9, A-1060 Vienna, Austria
> Tel: +43-1-5880115671
> Fax: +43-1-5880115698
> email: pblaha at theochem.tuwien.ac.at
> -----------------------------------------
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>