Question about orbital-resolved exchange parameters from ABACUS

[chao zhou]

Dear Dr. He Xu,

I hope this message finds you well. I’m currently analyzing the orbital-resolved magnetic exchange parameters (J) obtained from ABACUS, and I greatly appreciate how convenient the workflow is for extracting these values.

However, I’ve encountered a point of confusion regarding the orbital decomposition. In particular, I noticed that the exchange interaction between the Fe s and Fe s orbitals appears surprisingly large. This is puzzling, especially given that the system I’m studying is an insulator and that s orbitals are generally quite delocalized, making their physical contribution to localized exchange somewhat unclear.

While some of this may be attributed to orbital hybridization effects, the interpretation remains ambiguous from a physical standpoint. I’d be grateful for any insights you could share on how to understand such large s–s contributions in this context.

Thank you again for your continued work on these tools and for your time.

Best regards,

Chao Zhou

[Xu He]

Hello, 

This is indeed quite unusual. The exchange between s-s orbital pair for 1NN seems to be the strongest FM exchange. One thing to check is to plot the orbital-resolved DOS, to see if the s orbitals are really strongly spin-polarzied. 

Best regards, 

HeXu