.spn file?

[Chengcheng-Xiao]

The spn file contains:

• <\psi_m| S | \psi_n>, where S is the Pauli matrix, the pseudo part.
• <\psi_m| \sigma_{ij} q_{ij} |\beta_i> <\beta_j| |\psi_n>, the augmentation part.

The pseudo part can be easily obtained. For the augmentation part, refer to pw2wannier90.f90 (line 2469) and VASP's PEAD_CALC_OVERLAP and CALC_OVERLAP_GN.

[Chengcheng-Xiao]

Done in 892432cea7af6c56c3b5ed74bff12f74c1b855e3, more tests needed.

Does not work in parallel, because in parallel, bands are distributed. Need to figure out a way to sum them together.

For now, use:

mpirun -np 1 vasp_ncl

with ALGO=None+NELM=1 to skip everything alltogether.

For future: W1_GATHER in wave_high.F doesn't work because it generates wavefun1 instead of wavefuna. Need to take a look at PEAD_CALC_OVERLAP routine in pead.F

[Chengcheng-Xiao]

Correctness checked.

I'll leave the MPI support for the future, close for now.

[kibgboomba]

Thanks for your excellent code! I have a question. Will "ALGO=None NELM=1" affect the accuracy of the spn?

[Chengcheng-Xiao]

@kibgboomba Thanks for reaching out. I've tested ALGO=None on several cases and did not see any accuracy impacts. If you have doubt, run a simple test shouldn't hurt🤔.

[kibgboomba]

@kibgboomba Thanks for reaching out. I've tested ALGO=None on several cases and did not see any accuracy impacts. If you have doubt, run a simple test shouldn't hurt🤔.

Thank you very much!

[kibgboomba]

I have another question. Is the spn file need to be symmetrized like wannier90_hr.dat?

[Chengcheng-Xiao]

@kibgboomba Re-symmetrization is always optional even for the hamiltonian, you should decide yourself🤓