The ABINIT Users Mailing List ( CLOSED )

[mmikami@rc.m-kagaku.co.jp]

Hello,

First of all,  please make the most use of "Subject".
(Not a few readers might skip (or read later) the message,
 just because it has no subject ...)

Then you may want to dig in the "forum" archive ...
http://www.abinit.org/wws/info/forum
(How to dig ? Please be aware of the key-word-search system.)

For example, read this:
http://www.abinit.org/wws/arc/forum/2003-02/msg00042.html
(NB: ifc was renamed as anaddb ... Why ? Please check "FAQ" page.
 It might be a good idea to read it through...)

For references, you may notice the sources in the ABINIT package,
only if you look into the "Infos" directory:
- ~ABINIT/Infos/bibliography.htm
- ~ABINIT/Infos/acknowledgments.htm
Here the specific references are cited from the Acknowledgments page:
(http://www.abinit.org/ABINIT/Infos_v4.0/acknowledgments.htm )
-------------------------------------------------
# B.5. Many ingredients needed for the calculations of responses to
atomic displacements or homogeneous electric fields (dynamical
matrices, effective charges and dielectric constants), as well as
the Fourier interpolation implemented in the 'anaddb' code are
described in
[6] X. Gonze, Phys. Rev. B55, 10337 (1997) "First-principles responses 
of solids to atomic displacements and homogeneous electric fields: 
implementation of a conjugate-gradient algorithm"
[7] X. Gonze and C. Lee, Phys. Rev. B55, 10355 (1997) "Dynamical 
matrices, Born effective charges, dielectric permittivity tensors, and 
interatomic force constants from density-functional perturbation 
theory".
-------------------------------------------------
(And, the refereces therein, if needed, of course...)

About the unit conversion, how about try the following command
at the top directory where your ABINIT tar.gz ball is extracted:
        grep -i "Thz" Src*/*.f
(This is just the way as mentioned in the FAQ ...)
And you will notice that how the conversion is perfomed
in Src_6response/prtph3.f.
The parameters are defined in Src_defs/defs_basis.f.
Then you will find the answer to your question.
(approximate answer : e.g.  http://infrared.als.lbl.gov/IRunits.html )

By the way, you might be welcome to post summaries
(i.e. finally what you got from replies on Forum ML)
to your previous questions ... Not a few readers might be interested.
And also, summaries would be appreciated for future update of the FAQ.
(In this way, you could contribute to the ABINIT ... "Dank u" in 
advance.)

Regards,
Masayoshi

On 2003.5.14, at 00:03 Asia/Tokyo, M.G.A.Tijssens wrote:

> Hello everybody,
>
> I am trying to calculate eigenfrequencies and -modes from the
> dynamical matrices I obtained from abinit. However I cannot make
> the numbers match. Simply using D.u = \omega^2 u for the eigenvalues
> and then freq=\omega/(2\pi) gives results inconsistent with the 
> solution
> obtained with ifc.
>
> So, can anyone explain to me (or point me to a source of information)
> how the eigenfrequencies and -modes in ifc are determined
> from the dynamical matrix obtained from a RF-calculation with abinit?
>
> And can anyone tell me how one gets from eigenvalues from the dynamical
> matrix to eigenfrequencies in Thz and cm^{-1} (what 'factors' are used
> to transform the units)?
>
> Thanks,
> Martin Tijssens
> --
> Dr. M.G.A. Tijssens               T: +31 15 278 9552
> Delft University of Technology    F: +31 15 261 1465
> Dept. of Aerospace Engineering    E: M.G.A.Tijssens@lr.tudelft.nl
> Postbus 5058                      O: 1.54
> 2600 GB  Delft                    U:
> http://www.em.lr.tudelft.nl/~tijssens
> The Netherlands