Three independent teams of physicists have unveiled devices that could lead to practical spintronics components of the future. Researchers in the Netherlands have created what they call a “magnon transistor”, whereas a group in China has unveiled their “magnon valve”. Meanwhile in Germany, a team has also demonstrated their own version of a magnon valve. All three devices represent important work towards creating practical spintronics devices that use electron spin to transfer and store information.
Spintronics is attractive as a potential technology because it could solve several important problems facing electronics designers as they try to create ever smaller and more powerful devices. Using the spin of the electron (in addition to its electrical charge) to carry information provides an extra degree of freedom that could lead to smaller devices. At the same time, spin-based devices could be designed to consume much less energy than conventional electronics – making miniaturization easier.
However, creating spintronics based around the electron as the information carrier has its own challenges, so some physicists are keen on exploring alternatives. One possibility is the magnon, which is a collective excitation in a magnetic material. Magnons propagate as waves – flipping spins as they go. They also have particle-like properties, which is why they are called quasiparticles.
Circuits based on magnons have the potential to be much simpler in design than comparable conventional electronics – while at the same time consuming much less energy. But as Andrii Chumak of Germany’s University of Kaiserslautern, who was not involved in the research, explains: “We are still quite far away from realizing this potential”.
In this latest drive to create magnonic devices, a team led by Mathias Kläui of Johannes Gutenberg University of Mainz in Germany and a team led by Xiufeng Han of the Chinese Academy of Sciences in Beijing have showed that a magnon current can be controlled by changing the relative magnetization orientation of two magnetic layers.
Although made from different materials, both devices comprise a sandwich of two magnets separated by a non-magnetic spacer. By aligning the magnetic moments of the “bread” of the sandwich parallel or antiparallel, the researchers managed to increase and decrease the magnon current flowed through their devices – so the devices operated as valves.
“Both [devices] show typical spin valve behavior, and the effects are large so they could in future be used as a non-volatile low-power logic component,” explains Kläui. “But we now need to quantify the modulation of the magnonic spin current transmission in an ideal spin valve geometry.”
Taking a different approach, but still aiming to control magnon current, Bart van Wees of the University of Groningen, the Netherlands, and colleagues altered magnon current using an electrode to change chemical potential in a device they have called a “magnon transistor”. The device consists of a thin rectangle of platinum on top of a larger square of magnetic material. Magnons are generated at one end of the magnet and detected at the other. Then more magnons are pumped into or absorbed from the square depending on the spin polarization of electrons flowing in the platinum strip. By aligning and then oppositely aligning these electron spins with the magnons in the square, the researchers managed to increase and then decrease the magnon current.
This magnon transistor offers two potential benefits compared to the magnon valves: it operates faster than the valves and it should be more useful for creating complex circuits. However, the change in magnon current is much smaller than in the magnon valves. Also, because a spin current is used to modulate the magnon current, the transistor does not offer a low-power advantage over conventional electronics.
The devices could be key steps towards realizing full magnonic devices, but Chumak urges caution for those believing the research signals that magnonic circuits are just around the corner: “My personal feeling is that these papers represent an important step forward, but in fundamental physics only,” he says. “The magnonic signal has to be converted into electric current (in the Dutch device) or to magnetization orientation (in the two other cases) – a serious problem which requires in-depth investigations.”