[Wannier] Discrepancy in Berry Curvature Values Between kslice and kpath Modules in Wannier90 3.1.0

[Mainak Ghosh]

Hi,

I am currently calculating the Berry curvature using the kslice and 
kpath modules in Wannier90 3.1.0. I noticed a significant discrepancy in 
the results for the same k-point.

For example, for the k-point X with coordinates (0.5, 0.0, 0.0), the 
values obtained from kpath (bands + curv) and kslice (fermi_lines + 
curv) are very different. The k-point density between two points has 
been kept the same in both cases.

While the coordinates of the k-points corresponding to the maximum 
values of the Berry curvature are identical in both modules, the Berry 
curvature values differ substantially -- by roughly 20% to 60%. Ideally, 
the values should be the same at identical k-points.

Could you please clarify why these differences occur? Is there any 
inherent difference in the calculation method or normalization between 
the two modules that could explain this discrepancy?

Thank you for your time and assistance.

Best regards,

Mainak Ghosh

IIT Bombay, India
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20251016/44207622/attachment.html>