Accessing the W
Masoud Mansouri
Dear BGW developers,
I'd like to access the COHSEX W(q, ω=0) explicitly for post-processing. Given W(q,0) = ε⁻¹(q,0) · v(q+G), I tried to reconstruct it using ε⁻¹ from eps0mat.h5 (q≈0) and epsmat.h5 (q≠0), and the Coulomb v (vcoul.dat). However, I am running into difficulties below:
eps0mat.h5 and epsmat.h5 are defined on different effective G-vector subspaces, with special treatment of the head and wings at small q.
The Coulomb kernel is treated with q-shift around q0, making a naive mapping between ε⁻¹ and v.
Although vcoul information appears under /eps_header/gspace/ in the HDF5 files, I am not sure if this is sufficient to reconstruct the same W that is used internally in sigma.x.
Given these issues, my questions are:
Is there a code-level option in BGW to directly store W?
If not, is there an internal or home-made code to help me to reconstruct W that is guaranteed to be consistent with sigma.x (including truncation, head corrections, and G-space mappings)?
Any guidance on this issue would be very helpful.
Best regards,
Masoud