The ABINIT Users Mailing List ( CLOSED )

[Philippe Ghosez <Philippe.Ghosez@ulg.ac.be>]

> Dear Dr Ghosez,
>
>    I find that the dielectric tensor from abinis run is actually 
> the dielectric tensor for zero frequency. However, in your 
> publications you gave lot of tensor at infinite frequency. How to do 
> such a calculation? Give me some hints, please
>
>    Rongqin Wu

Dear Rongqin Wu:

In the computation of long range forces (as well as for the LO-TO 
splitting), we use indeed \epsilon_\infty that corresponds to the 
ELECTRONIC contribution to \epsilon.

Abinit can compute both :

- \epsilon_\infty : this corresponds to the dielectric constant as 
measured at high frequency (high enough to avoid the ionic 
contribution, ie optical range). It corresponds to the ELECTRONIC 
contribution to epsilon, accessible directly with ABINIT, considering 
the electric field perturbation. What could be confusing is that the 
calculation is artificially done at zero frequency. The result 
corresponds however to epsilon_infty because the ions are kept fixed. 
Only the electronic contribution is considered and it is assumed it 
is nearly independent of the frequency (to be rigorous the computed 
value should be compared to an extrapolation at zero frequency of the 
dielectric constant measured in the optical range).

- epsilon_0 : this corresponds to the dielectric constant as measured 
at low frequency. It corresponds to the electronic contribution 
(epsilon_infty) + an ionic contribution computed from the knowledge 
of the frequencies and IR oscillator strengths of the TO phonon modes 
at Gamma.

Best wishes,

Philippe.


----------------------------------
Philippe GHOSEZ
Universite de Liege
Institut de Physique, Bat. B5
Allee du 6 aout, 17
B- 4000 Sart Tilman
BELGIUM

Phone : ++(32) (0)4-366.36.11
Fax   : ++(32) (0)4-366.29.90
E-mail: Philippe.Ghosez@ulg.ac.be

http://www.ulg.ac.be/phythema
----------------------------------