The July 2022 issue of IEEE Spectrum is here!

Close bar

For First Time, Researchers Demonstrate Heat and Sound Are Magnetic

A magnet can reduce the amount of heat flowing through a semiconductor by 12 percent

2 min read
For First Time, Researchers Demonstrate Heat and Sound Are Magnetic
Photo: Ohio State University

Earlier this month, we reported on research demonstrating that heat propagates as a wave through graphene rather than as vibrations of atoms the way it does in 3-D materials.  In 3-D materials, the collective state of those vibrating atoms is known as phonons.

For the first time, researchers at Ohio State University (OSU) have demonstrated that acoustic phonons, which can carry both heat and sound, have magnetic properties that allow them to be manipulated with magnetism. 

In research published in the journal Nature Materials, the OSU researchers applied a magnetic field equivalent to that inside a magnetic resonance imaging (MRI) device (in this case, the magnet was reported to be fairly powerful at seven Tesla). They discovered that they could reduce the amount of heat flowing through a semiconductor by 12 percent.

“This adds a new dimension to our understanding of acoustic waves,” said Joseph Heremans, professor of mechanical engineering at Ohio State, in a press release. “We’ve shown that we can steer heat magnetically. With a strong enough magnetic field, we should be able to steer sound waves, too.”

Before anyone starts thinking about the discovery’s applicability to heat management in computers, they should keep in mind that the semiconductor had to be kept at temperatures very close to absolute zero (specifically, -268 degrees Celsius) in order for the researchers to measure the movements of the phonons.

In fact, it was the complexity of taking the measurements that had prevented researchers from recognizing the magnetic properties of phonons previously. In order to take thermal measurements at such a low temperature, Hyungyu Jin, a postdoctoral researcher and lead author of the study, used the semiconductor indium antimonide and shaped it into a lopsided tuning fork in which one arm was 4 millimeters wide and the other was 1 mm wide. Then he placed a heater at the base of each arm.

At normal temperatures, the ability of the material to transfer heat would be solely dependent on the kind of atoms in the material. But near absolute zero, the ability of the material to transfer heat can be determined by the physical size of the material. In this case, the difference in the sizes of the fork arms was significant. Phonons more easily filled the wider arm.

“Imagine that the tuning fork is a track, and the phonons flowing up from the base are runners on the track,” explained Heremans in the press release. “The runners who take the narrow side of the fork barely have enough room to squeeze through, and they keep bumping into the walls of the track, which slows them down. The runners who take the wider track can run faster, because they have lots of room.”

Eventually they all end up at their respective finish lines. But the track’s geometry determines just how quickly. 

With this understanding, Jin was able to compare the temperature changes in the two fork arms. He first took the measurements without a magnet and then with one. With the magnet on, the heat flow through the larger arm slowed down by 12 percent.

Now that the researchers have measured magnetism’s effect on heat, they want to move on to see if they can use it to deflect sound waves.

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":[]}