The ABINIT Users Mailing List ( CLOSED )

[Matthieu Verstraete <mjv500@york.ac.uk>]

Hello Alaska,

This is something I've been thinking of as well; basically, for the 
electron-phonon coupling which is coded, you need a metal: low energy 
excitations like phonons can only couple states filled and empty states 
"infinitesimally" close in energy, ie at the Fermi surface (FS). For a 
semiconductor, no states are available, the DOS at Ef is 0, and you should 
get 0 everywhere.

This is not to say the electrons and phonons don't interact, but all the 
normal properties (a2F, phonon linewidths...) are defined as integrals 
over the FS.

The fact that the dielectric matrix is found to be 0 is strange, you are 
probably missing some perturbations or something. Your output will be 
necessary to see this (Netiquette!)

For natifc, no idea: why would you want to set it to 0?

Matthieu

--
================================================================
Dr. Matthieu Verstraete                 mailto:mjv500@york.ac.uk
Dept. of Physics, University of York,     tel: +44 1904 43 22 08
Heslington, YO10 5DD York, United Kingdom fax: +44 1904 43 22 14