Power/Performance Bits: July 14

5G switches
Researchers from the University of Texas at Austin and University of Lille built a new radio frequency switch that could save power in 5G devices when not actively jumping between different networks and spectrum frequencies.

“It has become clear that the existing switches consume significant amounts of power, and that power consumed is useless power,” said Deji Akinwande, a professor in the Cockrell School of Engineering’s Department of Electrical and Computer Engineering at UT Austin. “The switch we have developed is more than 50 times more energy efficient compared to what is used today. It can transmit an HDTV stream at a 100 gigahertz frequency, and that is unheard of in broadband switch technology.”

Instead of remaining on and drawing processing power, the new switches stay off, saving battery life for other processes, unless they are actively helping a device jump between networks. They have also shown the ability to transmit data well above the baseline for 5G-level speeds. It is capable of functioning across the spectrum from the low-end gigahertz frequencies to high-end terahertz frequencies.

A diagram of the UT Austin team’s switch showing two gold electrodes with a layer of hBN in between. (Source: Cockrell School of Engineering/UT Austin)

The team’s switches use hexagonal boron nitride, a nanomaterial from the same family as graphene. The structure of the switch involves a single layer of boron and nitrogen atoms in a honeycomb pattern sandwiched between a pair of gold electrodes. Hexagonal boron nitride is the thinnest known insulator with a thickness of 0.33 nanometers.

Beyond smartphones, the team sees potential for the switch in satellite systems, smart radios, and IoT devices. “Radio-frequency switches are pervasive in military communication, connectivity and radar systems,” said Dr. Pani Varanasi, division chief, materials science program at the U.S. Army Research Office, which helped fund the work. “These new switches could provide large performance advantage compared to existing components and can enable longer battery life for mobile communication, and advanced reconfigurable systems.”

Sound storage
Researchers from the University of Sydney, Max Planck Institute for the Science of Light, University of Southern Denmark, and Australian National University propose a way of using sound waves to store and transfer information chips receive from fiber optic cables.

“As demand for high bandwidth information systems increase, we want to get ahead of the curve to ensure we can invent devices that don’t overheat, have low energy costs and reduce the emission of greenhouse gases,” said Dr. Moritz Merklein from the Eggleton Research Group in the School of Physics and Sydney Nano.

However, information transferred from fiber optic cables onto chips in the form of sound waves decays in nanoseconds, which is not long enough to do anything useful.

“What we have done is use carefully timed synchronized pulses of light to reinforce the sound waves on-chip,” said Dr. Birgit Stiller, of the Max Planck Institute for the Science of Light. “We have shown for the first time that refreshing these phonons is possible and that information can therefore be stored and processed for a much longer time.”

Using carefully timed pulses of light, the researchers were able to extend the lifetime of the information stored in sound waves on the chip by 300%, from 10 nanoseconds to 40 nanoseconds.

“Acoustic waves on chips are a promising way to store and transfer information,” said Stiller. “So far, such storage was fundamentally limited by the lifetime of the sound waves. Refreshing the acoustic waves allows us to overcome this constraint.”

“We plan to use this method to extend how long the information remains on-chip,” noted Merklein.

Christian Wolff, associate professor at the University of Southern Denmark, added, “Theoretically, this concept can be extended to the microsecond regime.”

ML for batteries
Researchers at the Department of Energy’s SLAC National Accelerator Laboratory, Stanford University, European Synchrotron Radiation Facility, Virginia Tech, Chinese Academy of Sciences, and Purdue University turned computer vision algorithms to the problem of lithium-ion battery degradation.

The team focused on a Li-ion battery’s nickel-manganese-cobalt (NMC) cathode. In the cathode, NMC particles are held together by a conductive carbon matrix, which has been suspected of contributing to performance decline as particles of it break apart.

However, standard computer vision algorithms that use the boundaries between light and dark to define objects didn’t work well for the X-ray tomography data of the NMC cathode. When differentiating between several small NMC particles stuck together and a single large but partially fractured one, the fractured one would appear to have clean breaks.

The researchers turned to an algorithm used for hierarchical objects which, like a jigsaw puzzle, create a complete entity from individual pieces. With training, this algorithm was able to distinguish different kinds of particles and thus develop a three-dimensional picture of how NMC particles, whether large or small, fractured or not, break away from the cathode.

Ultimately, the results showed that particles detaching from the carbon matrix really do contribute significantly to a battery’s decline, under typical consumer electronics conditions.

While large NMC particles are more likely to become damaged and break away, quite a few smaller particles break away, too, and overall, there’s more variation in the way small particles behave, said Yijin Liu, a staff scientist at SLAC. That’s important because researchers had generally assumed that by making battery particles smaller, they could make longer-lasting batteries – something the new study suggests might not be so straightforward, Liu said.