Converting Charge into Spin for Spintronics

It is still early days for spintronic computers, but the possibility of controlling spin electrically instead of magnetically brings them nearer

3 min read
Converting Charge into Spin for Spintronics
Theoretical physicist Jairo Sinova explaining/demonstrating electron spin with a turning top.
Image: Alexander von Humboldt-Stiftlung Foundation

Electronic circuits can only get so small before they’re overwhelmed with heat problems. Encoding bits using the spin of electrons, instead of the usual charge, promises to allow even smaller circuits—but the known processes of flipping electrons’ spins with external magnetic fields are inefficient and require very low temperatures, making such “spintronic” devices impractical.

Now, a team of researchers from Germany, the UK, the Czech Republic, and Japan have found a way to manipulate the spin of electrons using electric fields instead of magnetic ones. Their method, reported in the August issue of Nature Materials, could drastically reduce computers’ energy consumption and lessen heat problems caused by miniaturization.

Electron spin can be visualized as the rotation of an electron in one of two ways, with the rotation axis pointing up or down. Just like the presence or absence of an electric charge represents a bit equaling 1 or 0, a spin pointing up or down can do so as well. Flipping the spin to change a bit requires much less energy than moving charge.  

'You need an effective way to convert charge into spin current in one part of the circuitry that can then be converted again into electric signals in another part.'

In 2000 a team of researchers observed that if current passes through a semiconductor strip, even without the application of a magnetic field, electrons drift to each side of the strip based on their spin. When these electrons move through the semiconductor they experience a sideways force whose direction depends on the spin’s orientation, so the electrons build up at the two sides of the strip and create a voltage across it. Physicists quickly realized that this phenomenon, known as the "spin Hall effect," could be an excellent way to separate electrons according to their spin and that it could be used to process and store bits in future spintronic computers.

"You need an effective way to convert charge into spin current in one part of the circuitry that can then be converted again into electric signals in another part," says Jairo Sinova, a theoretical physicist at the Johannes Gutenberg University Mainz in Germany who participated in the research.

By using gallium arsenide, the semiconductor of lasers and LEDs, the scientists demonstrated efficient conversion at room temperature using an electric field. The achievement exploits the fact that gallium arsenide has two conduction bands that electrons can occupy depending on their momentum.

Illustration: Science Photo Library/Corbis

In the experimental set up [left] a voltage is applied to the GaAs strip. When circularly polarized light is switched on, a voltage appears over the contacts. Electrons passing through the GaAs strip are deflected according to their spin [right].

The lateral force the electrons experience is caused by spin orbit coupling, the interaction of the spin with the electron’s motion; electrons with opposite spins are pushed in opposite directions. For the experiment, the researchers illuminated the semiconductor with circularly polarized light to generate spins aligned in a specific direction. They found that the spin orbit coupling was very weak in the lowest conduction band, but by increasing the voltage they could shift electrons into the higher band where the spin orbit coupling, and consequently the spin Hall effect, was much stronger.

For converting charge into spin one needs a large signal, and for reading out the spins after processing one needs a small signal; Sinova says that they have demonstrated for the first time that “tuning” this conversion is possible by varying the electric field to change the distribution over the two bands.

It is very important to develop a viable spin device, and therefore researchers have to explore material systems in which the spin Hall effect is stronger, says Hidekazu Kurebayashi, a researcher at the London Centre of Nanotechnology (UCL), who directed the experiments. They are now planning to examine other III-V semiconductors for this application. "The advances in semiconductor growth technology allows us to control to a great extent the conduction band structures by selecting elements. There are plenty of exciting opportunities ahead," he says.

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

8 min read
Vertical
A stack of 3 images.  One of a chip, another is a group of chips and a single grey chip.
Intel; Graphcore; AMD
DarkBlue1

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
{"imageShortcodeIds":[]}