Description of the procedure to be used to do a positron lifetime calculation:

We have coded two formalisms for calculating the positron lifetime calculation. The main difference between the two
approaches is that in one case we consider that the electrons are not perturbed by the presence of the positron.
This is almost correct for a positron in a perfect bulk material. But this approximation fails when defects exist in
the material. For example, the positron will be trapped by a vacancy. Then locally its density will be important and
though will influence the electrons density.
In the following we will concentrate on the first formalism only.

First of all, one have to keep in mind that two different pseudopotentials have to be used to calculate
the GS of the positron. The first one for the GS calculation of the electrons and the second one for the GS
calculation of the positron. For the first one, there is only one restriction : no NLCC is allowed.
The second one has to be of the FHI type. Moreover it must contain at its
end, the All-electrons core density generated with FHI. At least this particular pseudopotential must verify
lmax=lloc=0 and pspxc=0 (no self-interaction terms must be taken
into account for the positron).


The calculation is done in two steps.
The first one is the GS calculation for the electrons. The only specific thing to
do is to set prtden and prtvha to 1 in the input file. This will create the
associated files _DEN and _VHA which are used as input files for the GS calcuation
of the positron.

The second step is the GS calculation of the positron and subsequently its
lifetime. Only two new variables are to be added in the imput file : "positron=1"
and "ixcpositron". The later one corresponds to the type of electron-positron
correlation that is used. A proper calculation must be done using one band and
only the gamma point in the BZ.

The reader can find an example in the directory Test_v4 (t95 and t96 tests).