[QE-users] phonon calculation : Vibrational frequency with no-intensity

[Lorenzo Paulatto]

> In my organometallic molecule /compound , TMA (trimethylealuminum ), as 
> there is metal Al at the centre, I cannot put epsil=true, the code stops 
> (as its defined only for insulators).
> 
> Effective charges are calculated only when both trans and epsil are true.

You can compute the value of epsilon in a metal without all the hassle 
of density functional theory, it is infinite. If you use a smearing to 
define the occupation of the electronic bands, the code will assume that 
it is metallic.

If your system is actually metallic, there is nothing you can do. If it 
is not metallic, either you use fixed occupations from the beginning, or 
you follow the scf calculation (with smearing) with non-scf one (with 
fixed occupations).

hth


-- 
Lorenzo Paulatto - Paris