[Pw_forum] matdyn file output

[Stefano Baroni]

On Aug 26, 2009, at 10:47 PM, Denny Jackson wrote:

> Hello,

Hi Denny:

>
> I am having trouble understanding the output from ph.x in the matdyn  
> file.  I am looking at Barium Titanate in the cubic phase, but with  
> broken symmetry and then relaxed.
> I'm trying to compare the linear response frequencies from ph.x to  
> frozen phonon frequencies.  From what I understand, the diagonalized  
> dynamical matrix should give displacements of each atom.  So in the  
> output:
>
> omega( 4) =       3.763989 [THz] =     125.553988 [cm-1]
>  (  0.097362  0.000000 -0.116603  0.000000 -0.288197  0.000000 )
>  ( -0.066349  0.000000  0.077204  0.000000  0.200853  0.000000 )
>  ( -0.169282  0.000000  0.204883  0.000000  0.421687  0.000000 )
>  ( -0.159112  0.000000  0.218980  0.000000  0.474827  0.000000 )
>  ( -0.184271  0.000000  0.191581  0.000000  0.474679  0.000000 )
>
> each column gives the real and imaginary components of the  
> displacements in cartesian coordinates.  I could not find what units  
> these values have, first of all.  I assume they are unitless and are  
> a fraction of the lattice parameter.

the answer is, "does not matter". vibrational mode are identified by  
the direction and the relative magnitude of each atom's displacement.  
the absolute magnitude is a matter of convention. hope it is clear why  
it has to be so. if not, please revert to us and I'll try to explain.

> Also, each column should add up to zero to preserve symmetry, should  
> it not?  For higher frequencies in this calculation the columns do  
> indeed add up to zero, but for lower frequencies (and negative  
> frequencies) this is not the case.  Does this indicate a problem  
> with my pw calculation or something else?

not sure what you mean by "preserve the symmetry". what is sure is  
that zone-center acoustic modes correspond to a rigid translation of  
the whole crystal. for these modes, the atomic displacements certainly  
do not sum to zero. the other modes have to be orthogonal to acoustic  
ones, and that's why their displacements sum to zero. beware, though,  
that orthogonality holds with respect to the "vibrational metric"  
which is proportional to a diagonal matrix with the atomic masses on  
the diagonal ...

> Is there a resource for good documentation on the matdyn file I have  
> overlooked?

not that I know, but a good textbook on lattice dynamics (such as Born  
and Huang or Botteger) and/or molecular vibrations may help clarify  
some of the general concepts you need to master to understand what  
ph.x does ...

Stefano B



---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
http://stefano.baroni.me [+39] 040 3787 406 (tel) -528 (fax) /  
stefanobaroni (skype)

La morale est une logique de l'action comme la logique est une morale  
de la pensée - Jean Piaget

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20090827/ff1ad340/attachment.html>