Magnetic Resonance


LR circuit coupled to YIG

Nonlinear Studies/ Suhl Instability


So far, our focus has been on YIG and doped YIG spheres.  YIG stands for Yttrium Iron Garnet.  Its crystal structure is indicated in the following diagram of a unit cell. Note the two non-equivalent sites for the iron ions.

YIG is a ferrimagnetic insulator. It is ideal for microwave engineering applications because of its large room-tempertaure magnetization and sharp magnetic resonance line.


YIG Resonance Frequencies

What determines the frequency of a magnetic resonance in a ferromagnet like YIG? The first thing that come to mind is the external magnetic field applied, but also intrinsic properties of the magnet: the effective field that the spins feel due to spin-spin interactions. In this study we showed explicitly that the resonance frequency is sensitive also to what is around the magnet. It has been long known that cavities can shift the intrinsic frequencies of magnets, but here we see that simply the presence of a loop of wire changes the measured spin resonance frequency.


The basic idea is shown in the following figure (see also this paper). We have a YIG sphere inside both a static magnetic field (horizontal) as well as an alternating magnetic field (at microwave frequencies, vertically). The YIG crystal is then surrounded by a wire loop terminated by a resistor.

The experimental setup is shown in the next figure. The YIG sphere sits on a patterned PC board inside the poles of a electromagnet. Microwaves are produced by mixing the LO output of a spectrum analyzer with a fixed frequency, these are then channeled to the PC board and create the alternating field at the crystal. The transmitted microwave signal is measured at the spectrum analyzer.

The experimental results are shown below. The solid line represents the YIG resonance without a copper loop around it, the dotted trace is for a large loop (1.75 mm radius), and the dot-dashed line is for a small loop (0.825 mm radius). Note that the YIG sphere has a radius of 0.45 mm. We observe that the resonance frequency shifts up in the presence of the loop. The frequency shift is very reproducable and in good agreement with theoretical considerations (see this paper for further details).


Nonlinear Studies / Suhl Instability

(under construction)