Power/Performance Bits: Jan. 21

Two-layer MRAM
Scientists at Tokyo Institute of Technology propose a simpler MRAM construction that could perform faster with less power than conventional memories.

The idea relies on unidirectional spin Hall magnetoresistance (USMR), a spin-related phenomenon that could be used to develop MRAM cells with an extremely simple structure. The spin Hall effect leads to the accumulation of electrons with a certain spin on the lateral sides of a material. The motivation behind this study was that the spin Hall effect, which is particularly strong in topological insulators, can result in a giant USMR by combining a topological insulator with a ferromagnetic semiconductor.

When electrons with the same spin accumulate on the interface between the two materials, due to the spin Hall effect, the spins can be injected to the ferromagnetic layer and flip its magnetization, allowing for write operations. At the same time, the resistance of the composite structure changes with the direction of the magnetization owing to the USMR effect. Because resistance can be measured using an external circuit, this allows for read operations.

The team designed a composite structure comprising a layer of gallium manganese arsenide (GaMnAs, a ferromagnetic semiconductor) and bismuth antimonide (BiSb, a topological insulator). With this combination, they were successful in obtaining a giant USMR ratio of 1.1%. In particular, the results showed that utilizing phenomena called “magnon scattering” and “spin-disorder scattering” in ferromagnetic semiconductors can lead to a giant USMR ratio, making it possible to use this phenomenon in real-world applications.

Plus, rather than the approximately 30 layers needed for conventional MRAM, the new device only needs the two layers. “Further material engineering may further improve the USMR ratio, which is essential for USMR-based MRAMs with an extremely simple structure and fast reading. Our demonstration of an USMR ratio over 1% is an important step toward this goal,” said Pham Nam Hai, an associate professor at Tokyo Tech.

Perfect secrecy
Researchers at King Abdullah University of Science and Technology (KAUST), the University of St Andrews, and the Center for Unconventional Processes of Sciences say they have created an unbreakable encryption technology.

The team created optical chips that enable information to be sent from user to user using a one-time unhackable communication that achieves “perfect secrecy,” allowing confidential data to be protected more securely on public classical communication channels. Their proposed system uses silicon chips that contain complex structures that are irreversibly changed to send information in a one-time key that can never be recreated or intercepted by an attacker.

“This new technique is absolutely unbreakable, as we rigorously demonstrated in our article. It can be used to protect the confidentiality of communications exchanged by users separated by any distance at an ultrafast speed close to the light limit and in inexpensive and electronic compatible optical chips,” said Professor Andrea di Falco of the School of Physics and Astronomy at the University of St. Andrews.

Keys generated by the chip, which unlock each message, are never stored and are not communicated with the message, and they cannot ever be recreated—even by the users themselves—adding extra security.

“With the advent of more powerful and quantum computers, all current encryptions will be broken in very short time, exposing the privacy of our present and, more importantly, past communications. For instance, an attacker can store an encrypted message that is sent today and wait for the right technology to become available to decipher the communication,” said Andrea Fratalocchi, KAUST associate professor of electrical engineering.

The team is currently working on developing commercial applications of the patented technology. They have a fully functional demo and are building user-friendly software for the system.

Self-healing batteries
Researchers at the University of Illinois at Urbana-Champaign developed a solid polymer-based electrolyte for lithium-ion batteries that can self-heal after damage. The material can also be recycled without the use of harsh chemicals or high temperatures.

After multiple charge-discharge cycles, lithium-ion batteries can develop dendrites, thin whiskers of metal that build up and can reduce battery life, cause hotspots and electrical shorts, and sometimes grow large enough to puncture the internal parts of the battery, causing explosive chemical reactions between the electrodes and electrolyte liquids.

Researchers have investigates replacing the liquid electrolyte with a solid ceramic or polymer electrolyte, but many of the potential materials are rigid and brittle resulting in poor electrolyte-to-electrode contact and reduced conductivity.

“Solid ion-conducting polymers are one option for developing nonliquid electrolytes,” said Brian Jing, a materials science and engineering graduate student at Illinois. “But the high-temperature conditions inside a battery can melt most polymers, again resulting in dendrites and failure.”

The researchers developed a network polymer electrolyte in which the cross-link point can undergo exchange reactions and swap polymer strands. In contrast to linear polymers, these networks actually get stiffer upon heating, which can potentially minimize the dendrite problem, the researchers said. Additionally, they can be easily broken down and resolidified into a networked structure after damage, making them recyclable, and they restore conductivity after being damaged because they are self-healing.

“Most polymers require strong acids and high temperatures to break down,” said Christopher Evans, materials science and engineering professor at Illinois. “Our material dissolves in water at room temperature, making it a very energy-efficient and environmentally friendly process.”

The team probed the conductivity of the new material and found its potential as an effective battery electrolyte to be promising, the researchers said, but acknowledge that more work is required before it could be used in batteries that are comparable to what is in use today.