[Wannier] Subject: Question about handling band crossing in Wannier90 for MoS2

[? 雅恬]

Dear Wannier90 developers,
I am using Wannier90 (with VASP) to generate MLWFs for monolayer MoS₂. I think the output files
are good results.
However, after extracting the U matrices and overlaps between neighboring k-points, I find that in regions with band crossings/entanglement, the overlap magnitude drops significantly (from ~1 to <0.2), and the dipole matrix elements (computed by rotating VASP dipole matrices with U) exhibit unphysical jumps (e.g., 0.1 → 0.55 Bohr) at adjacent k-points. The phase after unwrapping remains smooth, suggesting the issue is related to the Wannier gauge discontinuity at crossings.
Could you please advise if there are specific strategies, parameters, or tutorials in Wannier90 to handle such band crossing situations more robustly? Any guidance on achieving a smoother Wannier gauge across k-points would be greatly appreciated.
I can provide input files and figures if needed.
Thank you for your time
Best regards,
Yatian Ning

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20260318/d8484d06/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.rar
Type: application/octet-stream
Size: 1376828 bytes
Desc: test.rar
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20260318/d8484d06/attachment-0001.obj>