Sorry for a long question.
What I want to do is find out parity eigenvalues of wave functions (with
SOC turned on) at 8 time reversal invariant points for space groups with
non-symmporphic group operations. I've tried a few workaround without luck.
x irrep does not work for k-points on the surface of the BZ. So I tried to
plot the wave function using lapw7. But lapw7 does not support SOC. I then
tried copying case.vectorso to case.vector. It worked. lapw7 prints out
wave function in case.psink. So I suppose this is the spin-up part of the
wave function. (My calculation is non-spin polarized with SOC). As a test
case, I calculated the parity using x irrep and then plotted large
components of the wave function using lapw7. It seems that the resulting
wave functions do not have the parity expected from the output of x irrep.
My main question is as follows,
In a non-spin polarized calculation, the spin up and spin down parts of the
wave function are stored in case.vectorso and case.vectorsodn (assuming
that only lapw basis is used). Therefore, it should be alright to just copy
case.vectorso to case.vector and run lapw7, right? (If I'm only interested
in the wave function of spin up channel).
Wien mailing list
SEARCH the MAILING-LIST at: