Researchers have discovered a new way to create and control magnetic information signals using sound waves. The technique could help electronic devices use less power and generate less heat.
Today’s electronics work by moving electric charges through circuits. This movement wastes energy as heat. As devices become smaller and faster, reducing this energy loss becomes increasingly important.
Instead of moving electric charge, the researchers use magnons — tiny waves of magnetic activity inside a material — to carry information. Magnons can travel with much lower energy loss than electrons.
The team from the Institute of Nano Science and Technology (INST), Mohali, showed theoretically that surface acoustic waves (sound waves that travel along the surface of a material) can create tiny distortions in a magnetic material. These distortions act like forces that push magnons and generate a spin current. This offers a new way to control magnetic information without relying on conventional electric currents.
The researchers built a theoretical model of an ultrathin magnetic material placed on a piezoelectric substrate. They found that sound waves travelling through the substrate can control the movement of magnons in the magnetic layer.
This work suggests that sound waves could become a practical tool for controlling magnetic signals in future low-power electronics, offering an alternative to conventional charge-based circuits.
Memory more suited for AI
Researchers in Germany have developed a new type of computer memory that could help make future electronic devices faster, more energy-efficient and better suited for artificial intelligence applications. The technology, developed jointly by researchers at Fraunhofer IPMS and semiconductor manufacturer GlobalFoundries, is based on ferroelectric random-access memory (FRAM). Unlike conventional memory devices, which may lose stored information when power is switched off, FRAM retains data permanently while consuming very little energy.
The new memory technology uses a material called ferroelectric hafnium oxide. Information is stored by shifting atoms within the material’s crystal structure, creating tiny changes in electrical polarisation. Because this process requires very little power, the memory can operate below one volt and switch states within billionths of a second.
A key achievement of the project was integrating the new memory technology into GlobalFoundries’ existing chip manufacturing process. This means the technology can potentially be produced on an industrial scale rather than remaining confined to the laboratory.
According to the researchers, the memory technology is particularly attractive for battery-powered devices, autonomous sensors and “edge AI” applications, where artificial intelligence runs directly on a device instead of relying on remote data centres. Lower power consumption could allow smart devices to process more information locally while extending battery life.
The researchers say the development combines an ultra low-power chip platform with a memory technology that is both fast and durable. They believe it could help support the next generation of electronics for applications ranging from industrial automation and automotive systems to medical devices and AI-enabled products.
Published on June 15, 2026
