Researchers Set World Record by Forming Stable, Fast-Moving Skyrmions at Room Temperature

When it comes to various advanced technical applications—for example, superconducting wires for magnetic resonance imaging—engineers aim to get rid of electrical resistance and its coexisting synthesis of heat to a considerable extent.

However, it has been observed that a little heat synthesis due to resistance is a desirable property in metallic thin films used for spintronic applications like solid-state computer memory. Likewise, although defects are usually undesired in materials science, they can be helpful in manipulating the formation of magnetic quasi-particles called skyrmions.

Scientists in the team of MIT Professor Geoffrey S.D. Beach and colleagues in California, Germany, Switzerland, and Korea have demonstrated that it is possible to create fast-moving and stable skyrmions in uniquely formulated layered materials at room temperature, setting world records for speed and size. They have reported this in separate papers published in the Nature Nanotechnology and Advanced Materials journals this month, where each paper was featured on the cover of its respective journal.

As part of the study published in Advanced Materials, the scientists developed a wire stacked by with 15 repeating layers of a uniquely fabricated metal alloy formed of platinum (a heavy metal), cobalt-iron-boron (a magnetic material), and magnesium-oxygen. The interface between the cobalt-iron-boron and platinum metal layer in these layered materials forms an ambience where it is possible to create skyrmions by applying electric current pulses traveling along the length of the wire and an external magnetic field perpendicular to the film.

Strikingly, under a magnetic field strength of 20 mT, skyrmions are formed in the wire at room temperature. At temperatures more than 349 K, or 168 °Fahrenheit, the skyrmions are formed without any external magnetic field—an effect brought about by the heating up of the material—and the skyrmions stay stable even when the material is cooled back to room temperature. According to Beach, earlier, outcomes such as this had been observed only at low temperature and with large applied magnetic fields.

Predictable structure

After developing a number of theoretical tools, we now can not only predict the internal skyrmion structure and size, but we also can do a reverse engineering problem, we can say, for instance, we want to have a skyrmion of that size, and we’ll be able to generate the multi-layer, or the material, parameters, that would lead to the size of that skyrmion.

Ivan Lemesh, Graduate Student, Department of Materials Science and Engineering, MIT

Lemesh is the first author of the paper published in Advanced Materials, the co-authors of which include senior author Beach and 17 others.

Spin is a basic property of electrons and points either up or down. A skyrmion is a circular cluster of electrons, where the spins of the electrons are opposite to the spin orientation of adjacent electrons. The skyrmions maintain a clockwise or counter-clockwise direction.

“However, on top of that, we have also discovered that skyrmions in magnetic multilayers develop a complex through-thickness dependent twisted nature,” stated Lemesh during a presentation on his study at the Materials Research Society (MRS) fall meeting in Boston on November 30^th, 2018. Those outcomes were reported in a separate theoretical study in Physical Review B in September 2018.

The present study demonstrates that although the influence of this twisted structure of skyrmions on the potential to calculate the average size of the skyrmion is negligible, it has a considerable impact on their current-induced behavior.

Fundamental limits

In the study published in Nature Nanotechnology, the scientists investigated a different magnetic material, where platinum was layered with a magnetic layer of a gadolinium cobalt alloy, as well as tantalum oxide. They demonstrated that in this material, it is possible to form skyrmions as small as 10 nm and proved that the skyrmions have the ability to move at a rapid speed inside the material.

What we discovered in this paper is that ferromagnets have fundamental limits for the size of the quasi-particle you can make and how fast you can drive them using currents.

Lucas Caretta, Study First Author, Graduate Student, Department of Materials Science and Engineering, MIT

In the case a ferromagnet, like cobalt-iron-boron, adjacent spins are oriented parallel to one another and create a strong directional magnetic moment. The scientists opted to use gadolinium-cobalt—a ferrimagnet—to overcome the basic limitations of ferromagnets. The neighboring spins in gadolinium-cobalt alternate up and down, thus canceling each other out and leading to an overall zero magnetic moment.

“One can engineer a ferrimagnet such that the net magnetization is zero, allowing ultrasmall spin textures, or tune it such that the net angular momentum is zero, enabling ultrafast spin textures. These properties can be engineered by material composition or temperature,” explained Caretta.

In 2017, scientists in Beach’s team and their colleagues experimentally showed that it is possible to develop these quasi-particles at will in particular locations by introducing a specific kind of defect in the magnetic layer.

“You can change the properties of a material by using different local techniques such as ion bombardment, for instance, and by doing that you change its magnetic properties,” stated Lemesh, “and then if you inject a current into the wire, the skyrmion will be born in that location.”

Caretta added that “It was originally discovered with natural defects in the material, then they became engineered defects through the geometry of the wire.”

They employed this technique to form skyrmions in the new Nature Nanotechnology study.

The scientists used X-ray holography to capture images of the skyrmions in the cobalt-gadolinium mixture at room temperature at synchrotron centers in Germany. One of the developers of the X-ray holography method was Felix Büttner, a postdoc in the Beach lab. “It’s one of the only techniques that can allow for such highly resolved images where you make out skyrmions of this size,” stated Caretta.

The size of these skyrmions is as small as 10 nm—the current world record for skyrmions at room temperature. Current-driven domain wall motion of 1.3 km was demonstrated by the scientists with the help of a mechanism that can also be employed to move skyrmions, which also sets a new world record.

Excluding the synchrotron work, all the study was performed at MIT. “We grow the materials, do the fabrication and characterize the materials here at MIT,” stated Caretta.

Magnetic modeling

These skyrmions are one kind of spin configuration of the spins of electrons in these materials, whereas domain walls, which are the boundary between domains of opposing spin alignment, are another kind. These configurations are called solitons, or spin textures, in the area of spintronics. As skyrmions are a basic characteristic of materials, a complex set of equations taking into account their spin angular momentum, circular size, electronic charge, orbital angular momentum, layer thickness, magnetic strength, and various special physics terms that capture the energy of interactions between neighboring layers and neighboring spins—for instance, the exchange interaction—is needed for the mathematical characterization of their energy of formation and motion.

One of these interactions, termed the Dzyaloshinskii-Moriya interaction (DMI), is of specific importance to the formation of skyrmions and emerges from the interplay between electrons in the magnetic layer and the platinum layer. According to Lemesh, during the DMI, spins get oriented perpendicular to one other, thereby stabilizing the skyrmion. The DMI interaction enables these skyrmions to be topological, resulting in magnificent physics phenomena, rendering them stable, and enabling them to be moved with a current.

The platinum itself is what provides what’s called a spin current which is what drives the spin textures into motion. The spin current provides a torque on the magnetization of the ferro or ferrimagnet adjacent to it, and this torque is what ultimately causes the motion of the spin texture. We’re basically using simple materials to realize complicated phenomena at interfaces.

Lucas Caretta, Study First Author, Graduate Student, Department of Materials Science and Engineering, MIT

In both studies, a blend of micromagnetic and atomistic spin calculations was carried out by the scientists to ascertain the energy needed to form skyrmions as well as to move them.

It turns out that by changing the fraction of a magnetic layer, you can change the average magnetic properties of the whole system, so now we don’t need to go to a different material to generate other properties. You can just dilute the magnetic layer with a spacer layer of different thickness, and you will wind up with different magnetic properties, and that gives you an infinite number of opportunities to fabricate your system.

Ivan Lemesh, Graduate Student, Department of Materials Science and Engineering, MIT

Precise control

“Precise control of creating magnetic skyrmions is a central topic of the field,” stated Jiadong Zang, an assistant professor of physics at the University of New Hampshire, who was not involved in this study, regarding the Advanced Materials paper. “This work has presented a new way of generating zero field skyrmions via current pulse. This is definitely a solid step towards skyrmion manipulations in nanosecond regime.”

Christopher Marrows, a professor of condensed matter physics at the University of Leeds in the United Kingdom, commented on the Nature Nanotechnology paper: “The fact that the skyrmions are so small but can be stabilized at room temperature makes it very significant.”

Marrows, who also was not involved in this study, stated that earlier this year, the Beach team had predicted room-temperature skyrmions in a Scientific Reports paper and noted that the new outcomes are work of the highest quality. “But they made the prediction and real life does not always live up to theoretical expectations, so they deserve all the credit for this breakthrough,” stated Marrows.

Commenting on the Nature Nanotechnology paper, Zang added that “A bottleneck of skyrmion study is to reach a size of smaller than 20 nanometers [the size of state-of-art memory unit], and drive its motion with speed beyond one kilometer per second. Both challenges have been tackled in this seminal work.”

A key innovation is to use ferrimagnet, instead of commonly used ferromagnet, to host skyrmions. This work greatly stimulates the design of skyrmion-based memory and logic devices. This is definitely a star paper in the skyrmion field.

Jiadong Zang, Assistant Professor of Physics, University of New Hampshire

Racetrack systems

Solid-state devices developed using these skyrmions could eventually be used as alternatives to existing magnetic storage hard drives. Streams of magnetic skyrmions can function as bits for computer applications. “In these materials, we can readily pattern magnetic tracks,” Beach stated during a presentation at MRS.

These new outcomes could be used for racetrack memory devices, developed by Stuart Parkin at IBM. A crucial factor in engineering these materials for application in racetrack devices is engineering intentional defects into the material in which skyrmions can be formed, since skyrmions form at places where defects exist in the material.

“One can engineer by putting notches in this type of system,” stated Beach, who also is co-director of the Materials Research Laboratory (MRL) at MIT. When a current pulse is injected into the material, the skyrmions are formed at a notch. “The same current pulse can be used to write and delete,” he stated. According to Beach, these skyrmions are formed at a very rapid pace, in less than one-billionth of 1 second.

To be able to have a practical operating logic or memory racetrack device, you have to write the bit, so that’s what we talk about in creating the magnetic quasi particle, and you have to make sure that the written bit is very small and you have to translate that bit through the material at a very fast rate.

Lucas Caretta, Study First Author, Graduate Student, Department of Materials Science and Engineering, MIT

Marrows, the Leeds professor, added that “Applications in skyrmion-based spintronics, will benefit, although again it’s a bit early to say for sure what will be the winners among the various proposals, which include memories, logic devices, oscillators and neuromorphic devices.”

A challenge that remains is to find the optimal way to read these skyrmion bits. The Beach team has been continuing its study in this area of research, stated Lemesh, who noted that the present difficulty is to find a means for electrically detecting these skyrmions to use them in phones or computers.

“Yea, so you don’t have to take your phone to a synchrotron to read a bit,” stated Caretta. “As a result of some of the work done on ferrimagnets and similar systems called anti-ferromagnets, I think the majority of the field will actually start to shift toward these types of materials because of the huge promise that they hold.”