The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

Dear Timothy French,

timothy.french@yale.edu wrote:
> Hello all I am a PhD student in chemistry attempting to use ABINIT to 
> generate calculated IR
> spectra of crystalline materials.  I can get the phonon frequencies (a la 
> tutorial 5.5), but is
> there any way to get their IR intensities?  For anyone familiar with 
> Gaussian, I am basically
> looking for the output of an "freq" calculation--something with which a stick 
> spectrum can be
> created.  Would the "anaddb" utility help with that?  I've searched through 
> all of the
> documentation on the website (tutorials, help file, "new user guide", past 
> posts on this site,
> etc.), but to no avail.  Any help or leads in the right direction would be greatly appreciated. 
> Thanks in advance -Tim

The ANADDB will be of valuable help. Set dieflag to 1,
ask for a range of frequencies (nfreq). See
http://www.abinit.org/ABINIT/Infos_v4.3/anaddb_help.html .

In the output of ANADDB, you will see a list of
frequency-dependent dielectric tensor values, as well as
the predicted reflectivity. The formula implemented
are described in Phys. Rev. B 55, 10355 (1997),
see Eqs.(57) and (58). Anharmonicity effects are not taken
into account (and cannot be, with only linear response), so that
there is no damping of the reflectivity, and the absorption
spectrum would simply be made of Dirac delta's at the TO phonon
frequencies, with strength corresponding to the oscillator strength,
see Eq.(58). This might be the "stick spectrum" that you
are looking for. As mentioned in the previous forum mail, the
atomic unit for oscillator strengths is erroneous in present versions
of ABINIT.

Have fun ...
Xavier