With Help From Hydrogen, Spintronics Takes One Step Closer to Digital Logic

Hydrogen ions can turn magnetism on and off in a spintronic device, opening up digital logic applications

3 min read
An illustration shows a side-by-side comparison of how hydrogen (shown as red dots) moves in a material before and after a voltage is applied.
When a voltage is applied within a new spintronic device, hydrogen ions (red dots) move around to change the properties of a nearby magnetic layer (green).
Image: Courtesy of the researchers/MIT News

Spintronics has been in the lexicon of post-CMOS alternatives for so long, it can be easy to forget that there are still significant hurdles to clear in order for it to become the basis for new types of transistors.

In contrast to traditional transistors that operate by stopping and starting the flow of electrons to create binary logic, spintronics exploits the quantum property of spin in electrons. This property of electron spin serves as the basis for magnetism: If the spin of electrons points in one direction, then a material is magnetized.

To exploit this phenomenon in digital logic transistors, most designs have relied on electrons to accumulate at the interface between a metallic magnet and an insulator. Unfortunately, when a voltage is applied to flip the spin in such a design, there’s very little effect.

Now, researchers at MIT have developed a remedy for this problem in a new approach that uses hydrogen ions for spintronics, as opposed to electron accumulation. The result is that the magnetism of the spintronic device can flip by 100 percent, as opposed to a mere 1 percent with the old design. This development marks a significant step toward using spintronics for digital logic applications.

Research in spintronic transistors has focused over the last few years on the electron accumulation design because it can work simply by applying a voltage.

In this kind of device, an electric field is generated at the interface between an electrical insulator and a magnetic layer. Because the magnetic material is only a few atoms thick, the application of an electric field (or the accumulation of charge) can change the electronic structure of the device near the interface of the magnetic field and the insulator.

This provides a way to use a voltage to control the magnetic properties. However, the problem with this approach is that it has a very small effect. The change in magnetic properties is typically only about 1 percent, making the mechanism fundamentally interesting but not practically useful.

Another approach is based on ions instead of electrons. In these designs, oxygen ions have been used to change the magnetic properties of a spintronic device. But because oxygen ions are relatively large, they deform the magnetic material and make the device inoperable in short order.

In new research described in the journal Nature Materials, the MIT team again used ions. But instead of oxygen ions, they used much smaller hydrogen ions. In tests, the MIT group found that using hydrogen ions did not degrade the material over 2,000 cycles.

“By moving ions around in the vicinity of the magnetic material, we are essentially changing the nature of the bonding in that material, or its bonding with an adjacent material,” explained Geoffrey Beach, a professor at MIT and co-author of the research. “In this way, we can completely change its properties.” Beach added that these ionic effects can essentially turn magnetism on and off.

However, it was not as easy as just using the smaller hydrogen ions. The researchers needed to find materials that could readily conduct protons and could be integrated with a magnetic material. 

“Ionic and magnetic materials are typically very different classes of materials,” said Beach. “What we succeeded in doing is bringing together several materials with very different functionalities to make a new kind of device in which those functionalities work in concert.”

This development marks a significant step toward using spintronics for digital logic applications.

In this case, the device simultaneously works to harvest protons from the environment, shuttles them into and out of a solid material, and does so in a way that their presence, or absence, controls the magnetic properties of the material.

Beach concedes that for competitive memory devices, the speed of the response of their device would need to be increased significantly. 

“While we made a big step in this work, faster speeds would still be required for magnetic RAM, for example,” said Beach. “This will require faster proton conductors, which we believe can be reasonably integrated into the existing devices.”

However, Beach believes that for other applications, like neuromorphic computing, speed is not as critical. “What is needed now is to identify the best device architecture to exploit these novel capabilities,” he added.

The Conversation (0)

3 Ways 3D Chip Tech Is Upending Computing

AMD, Graphcore, and Intel show why the industry’s leading edge is going vertical

6 min read
AMD 3D V-Cache

A crop of high-performance processors is showing that the new direction for continuing Moore’s Law is all about up. Each generation of processor needs to perform better than the last, and, at its most basic, that means integrating more logic onto the silicon. But there are two problems: One is that our ability to shrink transistors and the logic and memory blocks they make up is slowing down. The other is that chips have reached their size limits. Photolithography tools can pattern only an area of about 850 square millimeters, which is about the size of a top-of-the-line Nvidia GPU.

For a few years now, developers of systems-on-chips have begun to break up their ever-larger designs into smaller chiplets and link them together inside the same package to effectively increase the silicon area, among other advantages. In CPUs, these links have mostly been so-called 2.5D, where the chiplets are set beside each other and connected using short, dense interconnects. Momentum for this type of integration will likely only grow now that most of the major manufacturers have agreed on a 2.5D chiplet-to-chiplet communications standard.

Keep Reading ↓ Show less