The 4Rshift file is used to store the repulsive potential used for the
screened structure constants. This has the potential of 4R high up to
a Muffin-Tin radius and it is zero from RMT-Rasa. One point is the choice of
this
MT radius. In a simple system (fcc, bcc etc) the choice is obvious: the true
lattice MT radius. In more complicated structures however where atoms have
different
environments then the choice is more difficult. In principal better space
filling
gives you better screening but you also like to keep things simple and have
only
a single potential for the reference medium. The program can handle multiple
potentials for the screening but the advise is to check which is the smaller
MT sphere you have in your lattice (It is printed out for each atom by the
voronoi program) and decide to choose this if you do not get a very "empty"
lattice
this way. One cost is that you might have to increase the size of the
TB-cluster
to get better screening and this makes the calculation more expensive !
(This treatment is subject to change in next versions since it is a source of
errors rather difficult to find). A small program that constructs this 4Rshift
called "ConstShift.f" exists in the program directory and it should be
used with caution
fcc, bcc, are treated but you can use your own RMT, RWS spheres.
Notice that since the potential is zero for R
RMT, the RWS is not important.
In case of repulsive potentials with different MT spheres or different heights you should include several potentials in the 4Ryshift file and the index CLS will put on each lattice site on of your potentials you have in the file. In case of spin polarized calculation you should have 2 times the same repulsive potential.
After running the voronoi program you have all files to start any calculation with the SKKR program, and it is advised to use the same inputcard file for both voronoi and skkr programs!