[Pw_forum] phonon symmetry at any point

Paolo Giannozzi giannozz at nest.sns.it
Fri Nov 4 19:30:54 CET 2005


Salut Claire

> Is there a tool (in espresso or somewhere else) that analyzes 
> the symmetry of the eigenvectors of the dynamical matrix ?

in espresso, no; somewhere else, I don't know. Once upon a time
I wrote a small piece of code that managed to recognize the 
irreps of normal modes of C60, but it was a specialized code
for a rather special case

> Can ph.x give this information in the output ?

all you get is the normal modes eigenvectors and frequencies

> Does ph.x use symmetry arguments to simplify the diagonalizing
> of the dynamical matrix ? 

that is, to reduce the matrix to block diagonal form? no, it is 
simply not worth the effort: the matrix is filled and diagonalized 
using standard algorithms that do not take into account any
matrix symmetry (apart from the hermitian character).
Symmetry is used to speed up the calculation of the dynamical
matrix, though, via a rather complex approach that nobody 
understand any longer :-)

> I suppose i should find the sub-group of the new enlarged unit cell
> corresponding to the new q-point and find there which are the new
> symmetry of the modes in this subgroup.  And then finally see how
> they correlate to the symmetries found at gamma point.

it is a typical problem of group theory: you need the irreps of 
the small group of q (i.e. the group of operations tat leave q
unchanged) and their character tables.

> I can do this procedure by hand a few times for high-symmetry 
> qpoints, but for any q-point, i would just like to have a nice code
> helping me to avoid  errors and saving time...
> Do you know of such a tool ?

unfortunately not

Paolo

-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy



More information about the Pw_forum mailing list