Symmetries

[gonze]

From Davide Sangalli :

Dear forum,
I've a question on symmetries.
I'm running abinit on some cnt and in particular the (5,5) and the (8,8)

********************************************************
For the (5,5) in the log files abinit says:
symanal : COMMENT -
The Bravais lattice determined only from the primitive
vectors, bravais(1)= 6, is more symmetric
than the real one, iholohedry= 2, obtained by taking into
account the atomic positions.
symspgr : the symmetry operation no. 1 is the identity
symspgr : the symmetry operation no. 2 is an inversion
symaxes : the symmetry operation no. 3 is a 2_1-axis
symplanes : the symmetry operation no. 4 is a mirror plane
symspgr : spgroup= 11 P2_1/m (=C2h^2)

but then reports 2 symmetries in the report file:
symmorphi 0
symrel 1 0 0 0 1 0 0 0 1 1 0 0 0 1 0 0 0 -1
toldfe 1.00000000E-06 Hartree
***************************************************************

On the other hand for the (8,8) in the log abinit says:
symanal : COMMENT -
The Bravais lattice determined only from the primitive
vectors, bravais(1)= 6, is more symmetric
than the real one, iholohedry= 2, obtained by taking into
account the atomic positions.
symspgr : the symmetry operation no. 1 is the identity
symspgr : the symmetry operation no. 2 is an inversion
symaxes : the symmetry operation no. 3 is a 2-axis
symplanes : the symmetry operation no. 4 is a mirror plane
symspgr : spgroup= 10 P2/m (=C2h^1)

and then reports 4 symmetries in the report file:
symmorphi 0
symrel 1 0 0 0 1 0 0 0 1 -1 0 0 0 -1 0 0 0 -1
-1 0 0 0 -1 0 0 0 1 1 0 0 0 1 0 0 0 -1
toldfe 1.00000000E-06 Hartree
******************************************************************

I do not understand what is going on.
Shouldn't there be 4 symmetries in both ?

Regards,
Davide


Dear Davide,

Yes, strange.
Well, the treatment of the case where
the holohedry determined from the lattice only, and
from the lattice and atomic positions, differ, was problematic
in ABINITv5. These difficulties have been satisfactorily addressed,
I think, in ABINITv6...
You have a branch on the repository, could you give a try with
ABINITv6.0.1 ?

Best regards,
Xavier

[mverstra]

See my mail reply: using symmorphi 0 eliminates the operations which contain a translation, which is probably the case (?) for the 5,5 tube and the inversion symmetry... Is this the case Davide?

Matthieu

[davide.sangalli.82]

Dear Matthieu,
I think you are right, for the (5,5) tube, some of the symmetries should take into account a translation.
If with symmorphi 0 abinit eliminates the translations everything seems fine to me.

Thanks a lot,
Davide