Spin polarized calculations for a material can give different band structures and energy gaps for up and down spins. However, CASTEP in some cases gives an energy gap value corresponding to an electronic transition in which electron changes its spin (from up to down-spin bands or vice versa). This transition generally is forbidden. Why does CASTEP show this gap? Thanks!
Spin in magnetic materials relates to relative orientation of the magnetic poles of two electrons. When two electrons are forced to come close enough while being in repelling magnetic parallel spin orientation, one of them will switch to the reverse orientation for both of them to fall into least action attractive magnetic spin orientation. Just like when two magnets are being progressively forced closer together while in pole to pole repelling mode.
You might find related information on covalent bounding and Cooper pairs in Sections 10 and 11 of this article:
Article Electromagnetic and Kinematic Mechanics Synchronized in thei...
I am admittedly confused by your question. I believe it depends on how CASTEP takes into account the energy vs momentum relation. If you define the energy gap as the energy difference between the lowest unoccupied to highest occupied state, then you might have that the gap you get is not aware of spin. Since these two states might accommodate electrons with different spin orientation. You shoud try to find a way to inform CASTEP you want to remain in the same spin subspace…
Thank for your reply!
I mean CASTEP shows an energy gap of which electron transition is forbidden. And I think showing this forbidden gap is no physical meaning. So I don't know why CASTEP does so.
CASTEP, as many DFT codes, calculates the band structure and thus evaluates the difference between the last occupied state and the first un-occupied state. The establishment of selection rules to determine if a transition if forbidden or not is another step. You could plot the spin resolved bands or spin resolved DOs, for example, and obtain such information.
Anyway, if you believe this is a bug, I am confident people at CASTEP would be happy to hear your comments and eventually take action. Commenting on a general purpose research website might not be leading to a solution.
Regards,
Roberto
Thank you very much for your comments!
If selection rules were not applied then I can understand.
Best regards,
Hieu
Just to add to the excellent replies by Roberto D'Agosta , you are assuming that a spin-flip transition is forbidden but that is not necessarily the case. CASTEP doesn't know which "experiment" you have in mind for the band-gap, so it cannot know which selection rules apply. If the band-gap measurement involved particles with spin, then a spin flip would be allowed in the transition (accompanied by a corresponding flip in the "measuring" particle).
A similar "rule-breaking" occurs if the valence band maximum and conduction band minimum are at different points in the Brillouin zone. The direct transition cannot occur by light alone (because light has negligible momentum), but CASTEP will still report the energy difference as the band-gap. This is not the optical band-gap, but it *is* the thermal band-gap (at finite temperature, phonons provide a mechanism to change the momentum).
When you ask CASTEP to compute transition probabilities, you always specify what kind of spectrum you wish to compute, for example an "optics" calculation. Once you have specified this, CASTEP *does* know what the appropriate operator is and can apply relevant selection rules.
As we're talking in the context of DFT, I should also remind everyone that the Kohn-Sham band-gap is not actually the band-gap of the material!
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Polymers