Distinct Landé Factors in Magnetic Sublattices

[Aman Khosla]

Hi,

I am simulating FeCoGd ferrimagnets using Mumax Plus and would like to know if it is possible to assign different gyromagnetic ratios (γ) or Landé g-factors to each magnetic sublattice separately. If Mumax plus supports this feature, please let me know.  
Thanks 

[Ian Lateur]

Hi,

No, mumax+ assumes γ to be a constant fact of the universe. I guess you could rescale all other parameters accordingly?

Ian

[Aman Khosla]

Hi Ian,

Thank you for your response. I understand that MuMax Plus assumes γ to be a constant. Would it be possible to rescale the other parameters accordingly to account for the gyromagnetic ratio of Gd and FeCo? If so, could you please guide me on how to implement this adjustment in the simulation?
Please see the image below for clarity.  

Image-1 from https://www.nature.com/articles/

image-2 from  https://doi.org/10.7566/JPSJ.90.081001

Best regards,
Aman Khosla

[Josh Lauzier]

Hi,

If you don't need specific features in mumax+, there is a mumax3 fork that has this feature (and it will be in mumax3 with the 3.11) release. You can find it here: https://github.com/mumax/3/pull/315 . It allows to set γ region-wise.

However, if you're interested in ferrimagnetic sublattices, you would have to use the custom fields feature to add the relevant interactions manually. I don't know of an example script for ferrimagnets, but there is one for AFMs you can find here by de Clercq et al. It should be relatively straightforward to tweak it for ferrimagnetic behavior instead, though.

Best,

Josh L.

[Ian Lateur]

Hi,

Seeing as the gyromagnetic ratio simply scales the torque in the LLG equation, rescaling all relevant effective fields should have the same effect. For example the exchange field scales linearly with Aex. I'm not 100% sure this will work for everything though, especially demagnetization between different sublattices, and this might get very confusing.

If I find some extra time, I might implement the mumax3 pull request in mumax+, but this might open pandora's box, so no promises.

Ian

Op zo 19 jan 2025 om 07:08 schreef Josh Lauzier <joshl...@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/mumax2/8eb498dc-2662-491a-983b-d3bdc3849291n%40googlegroups.com.

[Lars Moreels]

Hi all

The gyromagnetic ratio in mumax+ has now been promoted to the Parameter class in PR#119, making it possible to set a different value in different cells/regions/sublattices/timesteps... This feature will be included in the main branch of the code in the next version release.

Lars