[Pw_forum] problem with symmetry recognition

[giuseppe.mattioli at mlib.ism.cnr.it]

Dear all
I was thinking again about keeping the space group symmetry fixed  
along a vc-relax calculation. Actually, I did not directly ask my main  
question...:-) Is there a way to put the above constraint which I am  
not aware of?
Yours

Giuseppe

ISM-CNR
Italy

Quoting Paolo Giannozzi <giannozz at democritos.it>:

>
> On Jul 10, 2011, at 12:10 , pascal boulet wrote:
>
>> Following Stefano's comment I need to provide ALL the atomic
>> positions,
>> which I did not do! I gave only the positions in the primitive unit
>> cell (UC).
>
> what you call "primitive", I would call "irreducible" (like the
> irreducible
> segment of the Brillouin Zone)
>
>> But then, if I provide all the positions (I guess in the full UC)
>
> what you call "full", I would call "primitive"
>
>> why should QE fill the others?
>
> which others?
>
>> In addition, I did some calculations on Mg2Si (FCC, 12 atoms in the
>> full UC)
>
> antifluorite structure? what you call "full" here, I would call
> "conventional"
>
>> by providing only the 3 atoms of the primitive UC. QE found 48
>> symmetry
>> elements, which is correct, and the optimized cell parameters
>> seemed OK.
>> So, still, is it wrong?
>
> no, because you provided all the needed atoms (3 in the unit cell of the
> FCC lattice in the [anti]fluorite structure)
>
>> BTW, I also noticed something that looks strange to me. The
>> coordinates
>> reported by QE in the output are not those I give in the input: X
>> and Y
>> are the same but not Z which is scaled by c/a.
>
> nothing strange: atomic positions in input are in crystal coodinates:
>
>> ATOMIC_POSITIONS crystal
>
> atomic positions in output are in xyz coordinates, divided by a:
>
>>    Cartesian axes
>>
>>      site n.     atom                  positions (alat units)
>
> P.
> ---
> Paolo Giannozzi, Dept of Chemistry&Physics&Environment,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>