The ABINIT Users Mailing List ( CLOSED )

[Valerio Olevano <valerio.olevano@polytechnique.fr>]

Dear Zyli,
the term "plasfrq" was an unfortunate choice
to indicate the main parameter of the plasmon-pole
model as implemented in the present version of the
GW code.
In this version the screening is calculated at frequency 0,
and at a second, pure imeginary, frequency
i E_0
Then the quantities \tilde{\omega} and \Omega
are calculated from the screning \epsilon at the two frequencies.
the keyword plasfrq indicates exactly the value E_0.
One reasonable choice of plasfrq could be the plasma frequency of the 
system,
where the system should present a peak in \epsilon^-1.
The code calculates at a certain point (check in the output)
the plasma frequency using the classic
formula starting from the average density of the system.
However, if you move a little from this value, the final GW correction 
result
should be stable.
It's a good idea to verify the stability of the GW correction result by 
varying
plasfrq around that value.
If large variations are observed, the use of the plasmonpole
model in that system could be discutable.

best wishes