Research Bits: Aug. 23

Algae-powered microprocessor

Engineers from the University of Cambridge, Arm Research, Scottish Association for Marine Science, and Norwegian University of Science and Technology used a widespread species of blue-green algae to power an Arm Cortex M0+ microprocessor continuously for over a year.

The algae, Synechocystis, is non-toxic and harvests energy from photosynthesis. The tiny electrical current this generates then interacts with an aluminum electrode and is used to power the microprocessor.

About the size of an AA battery, the device operated in a domestic environment and semi-outdoor conditions under natural light and associated temperature fluctuations. “We were impressed by how consistently the system worked over a long period of time – we thought it might stop after a few weeks but it just kept going,” said Paolo Bombelli of the University of Cambridge’s Department of Biochemistry.

The algae does not need feeding and even was able to continue producing power during periods of darkness, which the researchers attribute to the algae processing some of its food when there’s no light, and this continues to generate an electrical current.

The researchers say it could be most useful for low-power IoT devices in off-grid situations or remote locations.

“The growing Internet of Things needs an increasing amount of power, and we think this will have to come from systems that can generate energy, rather than simply store it like batteries,” said Christopher Howe, a professor in the University of Cambridge’s Department of Biochemistry. “Our photosynthetic device doesn’t run down the way a battery does because it’s continually using light as the energy source.”

Neon qubits

Researchers from Argonne National Laboratory, Florida State University, University of Chicago, Washington University in St. Louis, Lawrence Berkeley National Laboratory, and Massachusetts Institute of Technology designed a new qubit platform using solid neon.

The process of creating the qubit involves freezing neon gas into a solid at very low temperatures, spraying electrons from a light bulb onto the solid, and trapping a single electron there. Neon is an inert element and does not interact with other elements. This allows the qubit to have a long coherence time.

“Because of this inertness, solid neon can serve as the cleanest possible solid in a vacuum to host and protect any qubits from being disrupted,” said Dafei Jin, an Argonne scientist.

The platform uses a chip-scale microwave resonator made out of a superconductor. “The microwave resonator crucially provides a way to read out the state of the qubit,” said Kater Murch, professor of physics in Arts & Sciences at Washington University in St. Louis. “It concentrates the interaction between the qubit and microwave signal. This allows us to make measurements telling how well the qubit works.”

Electrons from a heated light filament (top) land on solid neon (red block), where a single electron (represented as a wave function in blue) is trapped and manipulated by a superconducting quantum circuit (bottom patterned chip). (Image by Dafei Jin/Argonne National Laboratory.)

“With this platform, we achieved, for the first time ever, strong coupling between a single electron in a near-vacuum environment and a single microwave photon in the resonator,” said Xianjing Zhou, a postdoctoral appointee at Argonne. ​“This opens up the possibility to use microwave photons to control each electron qubit and link many of them in a quantum processor.”

“At the moment, the qubit we have is based on the motion of the electron,” Murch said. “We call this a charge qubit, but one of the exciting future prospects will be to convert this into a spin qubit, relating to the spin of the electron. This should make the qubit much less sensitive to its environment, increasing the quality of the qubit by orders of magnitude.”

“Thanks to the relative simplicity of the electron-on-neon platform, it should lend itself to easy manufacture at low cost,” Jin added. ​“It would appear an ideal qubit may be on the horizon.”

“Our qubits are actually as good as ones that people have been developing for 20 years,” said David Schuster, physics professor at the University of Chicago. ​“This is only our first series of experiments. Our qubit platform is nowhere near optimized. We will continue improving the coherence times. And because the operation speed of this qubit platform is extremely fast, only several nanoseconds, the promise to scale it up to many entangled qubits is significant.”

Nanomagnet AI

Researchers from Imperial College London, Kyushu University, Los Alamos National Laboratory, and University College London used a network of nanomagnets to perform artificial intelligence-like processing, using the magnets themselves to process and store data. The nanomagnets can be used for ‘time-series prediction’ tasks, such as predicting and regulating insulin levels in diabetic patients.

Applying a magnetic field to a network of nanomagnets changes the state of the magnets based on the properties of the input field, but also on the states of surrounding magnets, according to the researchers. They were then able to design a technique to count the number of magnets in each state once the field has passed through, giving the ‘answer.’

“We’ve been trying to crack the problem of how to input data, ask a question, and get an answer out of magnetic computing for a long time. Now we’ve proven it can be done, it paves the way for getting rid of the computer software that does the energy-intensive simulation,” said Jack Gartside of Imperial College London.

Nanomagnets however don’t rely on the physical transport of particles like electrons, but instead process and transfer information in the form of a ‘magnon’ wave, where each magnet affects the state of neighboring magnets, the researchers said. This results in higher energy efficiency, and processing and storage of information can be done together.

Will Branford of Imperial College London added, “It has been a long-term goal to realize computer hardware inspired by the software algorithms of Sherrington and Kirkpatrick. It was not possible using the spins on atoms in conventional magnets, but by scaling up the spins into nanopatterned arrays we have been able to achieve the necessary control and readout.”

The team will next teach the system using real-world data, such as ECG signals, and hope to make it into a real computing device.