I am studying Giant Magnetoresistance (GMR) and I'm having difficulty reconciling why the GMR effect is isotropic; the resistance of the material drastically drops when exposed to a North oriented magnetic field, and equally so for a South oriented magnetic field. GMR: Resistance vs External H-Field GMR: Ferromagnetic and Non-Magnetic layers What I know is that, in a typical GMR multilayer material, a very thin non-magnetic metal (called the conductor layer, or spacer layer) is sandwiched between two ferromagnetic metal layers. The two ferromagnetic layers are manufactured such that their electrons have spins that are anti-parallel to each other. Also, one layer is "pinned" such that an external magnetic field won't change the magnetic moment of it.

This causes an electron in the conduction layer to experience strong scattering effects, which means the material is highly resistive.

When exposed to an external magnetic field (North or South), the "unpinned" layer's electron spins becomes oriented to it, such that both the ferromagnetic layers have parallel magnetic moments.

This causes an electron in the conduction layer to experience less scattering effects, meaning the material is suddenly much less resistive.

This is the GMR effect.

What I don't understand:

How come both a North or South magnetic field equally cause the same parallel orientation (with respect to the pinned ferromagnetic layer) of the unpinned ferromagnetic layer's electron spins?

I would expect one field direction (say, North) to cause the orientation to be parallel, and the other (say, South) to cause the orientation to be more anti-parallel (or unchanged).

Edit: Perhaps I'm confusing spin valves with generic GMR effects. See the below image from this excellent video: https://youtu.be/cID4fKraWkE?t=1143

Multilayer vs spin valve

  • $\begingroup$ But is there a pinned layer in this case? $\endgroup$ – user137289 Apr 21 '18 at 8:10
  • $\begingroup$ Yes, at least in the papers I've read regarding spin valves and GMR magnetometers, one the ferromagnetic layers is always "pinned" $\endgroup$ – InfiniteZoom Apr 21 '18 at 8:13
  • $\begingroup$ So, what article is the source for the measurement you show here? I repeat my question: is there a pinned layer in this case? Anyway, the field is enormous here, way larger than typical pinning coercivities. $\endgroup$ – user137289 Apr 21 '18 at 9:15
  • $\begingroup$ I lifted those images from the Wikipedia on GMR en.wikipedia.org/wiki/Giant_magnetoresistance however it appears I'm getting confused about general GMR effects in a multi-layer configuration and a spin valve configuration. Updated the question $\endgroup$ – InfiniteZoom Apr 21 '18 at 9:43

Your edit clarifies. In a spin valve with pinning, it is a device that is supposed to detect the small fields from bits on a magnetic recording material. The magnetization of the bottom layer is pinned. The top layer is weakly coupled to this bottom layer. The field from a bit can switch the magnetization direction of the top layer, and this affects the resistance of the device.

But a somewhat stronger field will overcome the pinning. This spin valve tutorial shows graphs of the resistivity as a function of the field. At fields larger than a few hundred oersted, both layers are always parallel and the device is in the low resistance state. It is outside the rage of the graph in the Youtube screenshot, not relevant for the function of the device. But for a large negative field, the resistance should be low again.

The GMR curves in the post are for magnetic multiplayers with thin chromium spacers that give a strong antiferromagnetic coupling. It takes 10 kG (1 tesla) to overcome this antiferromagnetic coupling.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.