Brain-inspired magnetic waves can power next-generation computing without freezing

Swedish researchers show precise control of magnetic oscillation, which could lead to room temperature devices comparable to quantum computers

For the future of computing, researchers have shown a technology that utilizes magnetic ripples to transmit and process information, which has the potential to open the door to energy-efficient computers that can run at room temperature without the need for extreme cooling like quantum machines.

The findings, published in Natural Physics in January, show how nanoscale devices can synchronize in the form of rotating Hall nano-oscillators, ultimately leading to powerful computing systems solving complex problems while consuming only a fraction of the energy required by traditional processors.

“With spin waves, we’re closer to building efficient, low-power computing systems that can solve real-world problems,” said Akash Kumar, lead author of the study at the University of Gothenburg.

This study represents a significant advance in the field of rotational type, which utilizes the quantum spin properties of electrons, not just their charges. With conventional electronic devices moving electrons through circuits, SpinTronics manipulates the magnetism of the material at the nanoscale level.

Browse magnetic waves

The researchers demonstrated control of what they call “spin waves”—applied in magnetization of materials with specific phase and energy properties. These waves can be generated by applying magnetic fields, currents and voltages to special nano-thin layer magnetic material layers.

What makes this breakthrough particularly important is the ability of researchers to accurately control the phase relationship between two oscillators. They showed for the first time that rotational waves can mediate the “phase” and “phase” relationships between oscillators. More importantly, they can adjust this relationship by adjusting various external factors: magnetic field strength, current, applied gate voltage, or distance between oscillators.

This phase control is crucial because it creates binary states – the basic foundation of digital computing. However, unlike traditional computers that use transistors to switch between state and off state, these oscillator networks may solve certain complex problems more efficiently.

Room temperature advantages

The study has a particular correlation with a class of professional computing systems called ISING machines, which aim to solve specific types of optimization problems in which it is more important to find the best approximation than to calculate the exact answer.

Many AI systems today rely on these approximations, but when running on traditional computers, the calculation consumes a lot of energy. Quantum computers provide an alternative, but usually require cooling to a temperature close to absolute zero to operate.

The spin-wave method demonstrated by the Gothenburg team at room temperature works at room temperature, which is a great practical advantage that can make the technology more accessible and versatile.

After the initial success of the oscillator pair, researchers are now working to scale up. “Researchers at the University of Gothenburg are now building a network of hundreds of thousands of oscillators to develop the next generation of Ising machines,” the research team said.

Technical details

The study involved fabricating equipment with nanoshrinkage (only 150 nanometers wide) in layers of magnetic materials. The team used a bunch of tungsten, cobalt iron and magnesium oxide (W/COFEB/MGO), which exhibited what is called perpendicular magnetic anisotropy, which means that the magnetization naturally points to the membrane plane.

As the current shrinks through these nanos, the spin Hall effect in the tungsten layer produces a spin polarization current that can stimulate automatic oscillation in the magnetolayer above. These oscillations produce propagating spin waves that can fuse adjacent oscillators together.

The researchers directly visualize and measure the phase relationship between oscillators using electrical measurements and a specialized microscopy technique called phase-resolved microfocus.

Through careful experiments, they demonstrated that the phase difference between the two oscillators can be adjusted by adjusting the drive current to nearly 0° (completely relative) to 150° (almost opposite phase), which actually converts the coupling from positive to negative to negative and returns.

Beyond computing

These implications are more than just building new computers. According to Kumar, “Spintronics has the potential to impact many different fields, from artificial intelligence and machine learning to telecommunications and financial systems. The ability to control and manipulate nanoscale spin waves may lead to the development of more powerful, more efficient sensors and even high-frequency inventory trading machines.”

This technology is expected to be integrated into the prior art. “Since the oscillators operate at room temperature and have a nanoscale footprint, these devices can easily adapt to larger systems and can also be used in smaller devices such as cell phones,” Kumar added.

At present, the technology is still in the research stage and its practical application is still the same. However, with the limitations of traditional computing, miniaturization and energy efficiency, these alternative approaches are attracting increasing attention from scholars and industries.

By demonstrating precise control of the coupling between these nano-osters, the Gothenburg team is a major step towards leveraging magnetic wave physics for next-generation computing systems that may one day solve today’s most powerful machines while consuming a small portion of energy, problems that are even today’s most powerful machines.