[Pw_forum] dynmat.f90 and raman configuration

[José Gadelha da Silva Filho]

Dear QE users,

I encountered similar questions on forum, but I would like to understand
precisely the raman configuration defined in the dynmat.f90 code.

In the 90° geometry, the laser beam direction and the axis of the
collection lens are at 90° to each other, but in the 180° scattering
geometry (also called back scattering mode) they are coincident.

I am doing the calculation in a periodic system and I read the paper from
 Michele Lazzeri and Francesco Mauri (PRL, 90, 2003), which describe the
implementation of this calculation in pwscf. In the same paper (at
reference 20) they use the backscattering configuration on simulated raman
spectrum.

However, the dynmat.f90 code write the raman intenisty as

(45.d0*alpha**2 + 7.0d0*beta2)*amu_ry

with

alpha = (raman(1,1,nu) + raman(2,2,nu) + raman(3,3,nu))/3.d0

Beta2 = ( (raman(1,1,nu) - raman(2,2,nu))**2 + &
                 (raman(1,1,nu) - raman(3,3,nu))**2 + &
                 (raman(2,2,nu) - raman(3,3,nu))**2 + 6.d0 * &
          (raman(1,2,nu)**2 + raman(1,3,nu)**2 + raman(2,3,nu)**2) )/2.d0

These expressions are described in PRB 54, 7830 (1996)  for a 90° geometry.
In the same paper we can find "Further, the direction of the incident beam,
the polarization direction of this beam, and the direction of observation
are perpendicular to each other".

So, I would like to know if the dynmat.f90 is really doing the calculation
in 90° geometry? Please, someone else can show me If I misunderstand some
point?

Sorry for the long question.

Best,

-- 
José Gadelha da Silva Filho

Aluno de Doutorado do Curso de Pós-Graduação em Física
da Universidade Federal do Ceará.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150921/f8db9460/attachment.html>