The ABINIT Users Mailing List ( CLOSED )

[perand@foi.se]

 Dear Forum

I am trying to calculate the (p,V,T) phase diagram of
a few compounds containing rare earths. As these
materials have several degrees of freedom, both internal
and cell shape, which can change a lot under compression,
I have been trying to use ABINIT for the relaxations.
To describe the different atoms I have chosen HGH pseudo
potentials with spin-orbit interaction included.

My problem is that for large compressions I notice a
substantial difference compared to the results from a
very stable and accurate FP-LMTO code (for a fixed
structure, no relaxation in this FP-LMTO). I assume
that this is due to the validity of the PPs and the
choice of rloc (the fitting radius). The PPs I have
made are already extremly hard (ecut > 120 Ha) and
they are pushing the limits with regards to the
recommendations on how to create HGH PPs in the README
in that code.

Are there any references on PPs and pressure or do
anyone in the forum have any experience with this
problem?

Is there some code which can do ionic and cell shape
relaxations without using PPs aviable?

 TIA!

/Per Andersson