FM/AFM interfacial exchange coupling in mumaxplus

[Smith John]

Hi everyone, I am trying to model a FM/AFM   bilayer in mumaxplus.

However, I am not sure whether mumaxplus currently supports explicit nearest-neighbor exchange coupling between two different magnet objects, for example: 

 

Fe = Ferromagnet(world, ...) CoO = Antiferromagnet(world, ...)

with an interfacial exchange term between the bottom Fe cells and the top CoO sublattice cells.

So my questions are: Is there currently a supported way to define exchange coupling between a Ferromagnet and an Antiferromagnet across their interface? If not, is the recommended approach to approximate the FM/AFM exchange bias by applying an effective local bias field to the FM interface layer, representing pinned uncompensated AFM interface spins? Is there any existing example or suggested pattern for modeling exchange bias / FM-AFM bilayers such as CoO/Fe or Co/CoO in mumaxplus?  

Any advice on the intended or recommended way to handle this would be greatly appreciated.

[Ian Lateur]

Hi,

This is indeed not explicitly supported. However, it is doable by making a single antiferromagnet of 2 layers and setting msat of one sublattice of one layer to 0. You can find an example in this discussions reply.

If you're expecting one of the layers to be static, you can always calculate/estimate what the effective field would be and set it as a bias field, like you suggested.

Hope this helps!

Ian

Op vr 8 mei 2026 om 07:50 schreef Smith John <bigot...@gmail.com>:

--
You received this message because you are subscribed to the Google Groups "mumax2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mumax2+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/mumax2/0b3f64f9-7051-4f1a-bca9-5239c3da9c1bn%40googlegroups.com.

[Smith John]

Thanks a lot! The discussion link you shared is very helpful.