Question about q-grid shift in wfnq calculations

[박기명(일반대학원 물리학과)]

Dear GW users,

While working on an orthorhombic structure, I noticed that different q-grid shifts lead to different results. The lattice parameters are approximately 7.5, 8.0, and 11.5 (ang).

In a semiconductor calculation to evaluate the band gap, I first applied a q-shift of 0.001 only along the z-axis. For verification, I then repeated the calculation with a shift of (0.001, 0.001, 0.001) along the x, y, and z axes. Comparing the two calculations, I found a systematic difference of about +0.1 eV across all k-points. My understanding is that results should be numerically insensitive to such a tiny shift, so this ~0.1eV discrepancy is larger than expected. (I know that slight shift is to avoid the gamma point singularity.)

Could you help me understand the numerical/physical reasons for this discrepancy and advise on which q-grid (and shift) settings would be more appropriate?

Best,

GiMyung Park

[Zhenglu Li]

Dear GiMyung,

Your results indicate that the dielectric function is anisotropic in your material system.  In this case, one can find an "average" representative direction to set up the small q.

Best,

Zhenglu

---

From: 박기명(일반대학원 물리학과) <dus...@yonsei.ac.kr>
Sent: Sunday, October 26, 2025 11:08 PM
To: BerkeleyGW Help <he...@berkeleygw.org>
Subject: Question about q-grid shift in wfnq calculations

 

Dear GW users, While working on an orthorhombic structure, I noticed that different q-grid shifts lead to different results. The lattice parameters are approximately 7. 5, 8. 0, and 11. 5 (ang). In a semiconductor calculation to evaluate the band

--
You received this message because you are subscribed to the Google Groups "BerkeleyGW Help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to help+uns...@berkeleygw.org.
To view this discussion visit https://groups.google.com/a/berkeleygw.org/d/msgid/help/01ee9a01-8a1c-44e7-b7fb-9f3f47dc5dean%40berkeleygw.org.