no-reciprocal transmission caused by the interface DMI

[Shiwei Meng]

hello everyone

               I am currently simulating a double-layer film structure consisting of a ferromagnetic layer and an antiferromagnetic layer(HM/FM). I aim to achieve the no-reciprocal propagation of spin waves caused by the interface dmi. However, I did not observe this phenomenon in the snapshot images. Here is my code and the snapshot images;

I attempted to change the value of my dind parameter and the direction of the alternating magnetic field application, but none of these efforts yielded the desired outcome.

    I sincerely seek your assistance and am extremely grateful.

setGridSize(1000,20,2)

//setCellSize(5e-9, 5e-9, 2e-9)

setCellSize(3e-9, 3e-9, 3e-9)

OutputFormat = OVF2_TEXT

Msat   = 1.4e5

Aex     = 3.5e-12

Dind   = 0.05e-3

anisU   = vector(1, 0, 0)

//Ku1     = 0.59e6

//alpha   = 2e-4

//alpha   = 0.5

m = uniform(1,0,0)

alpha = 0.5

B_ext = vector(0.05,0,0)//偏置场

relax()

B_ext = vector(0.05,0,0)//偏置场

defregion(11,xrange(1410e-9,1500e-9))//右吸收边界  right  big alpha

alpha.setregion(11,1)

defregion(13,xrange(-1500e-9,-1410e-9))//左吸收边界  left big alpha

alpha.setregion(13,1)

defregion(12,xrange(-1410e-9,1410e-9))//传播区域

alpha.setregion(12,2e-4)

defregion(6,xrange(-3e-9,3e-9))//天线区域

B_ext.setregion(6, vector(0.05,0.009* sin( 2*pi*10e9*t),0))

autosave(m,1e-11)

tableAutosave(1e-11)

run(20e-9)

[Josh Lauzier]

Hi,

I am not an expert with this, but i see a few things:

First, you mention antiferromagnetic coupling, but there is no AFM effect in your script. I am also unsure why a HM would be antiferromagnetic coupled, as they are typically nonmagnetic. Usually the HM is only included implicitly in simulations.

Second, one think I would check is, have you tried looking at the dispersion curve? I'm not sure how obvious the non-reciprocal effect would be just looking at snapshots. It might be rather subtle, especially until you dial in appropriate parameters. (In particular, it should be more noticeable at lower wavenumbers, if I recall. The DMI in this case is linear in k in the dispersion relation)

Last, I'm not sure what range of DMI you have tested, but it may be insufficient. Just skimming the literature for this effect, people seem to use values of the interfacial DMI strength of around 1e-3 up to 3e-3, a factor of 20x larger than what you are using. (Also, I would suggest perhaps copying the material parameters of an existing paper first. Once you are sure on the implementation, then start changing back to yours).

Best,
Josh L.

[IMPACT lut]

Hi

First of all, I would like to express my gratitude for taking the time to reply to my question in your busy schedule. In my description of the problem, I did not provide an accurate account. What I aim to achieve is the  no-reciprocal propagation   caused by iDMi between ferromagnetic/non-magnetic heavy metals. After reviewing the literature, I found that only static magnetic surface spin waves seem to exhibit  no-reciprocal propagation    under the presence of IDMI. I have achieved  no-reciprocal propagation   , but my spin wave waveform seems unstable. I am currently seeking a solution.