Scientists have developed the world’s strongest iron-based superconducting magnet with the help of artificial intelligence; This could be a breakthrough for affordable MRI machines and the future of electric transportation.

Superconducting magnets are capable of producing very strong, stable magnetic fields without requiring large amounts of energy. This means it can be used in a range of technologies, including MRI machines, which require a strong magnetic field to produce clear 3D images of soft tissues. They can also be used in next-generation transportation, including the SCMaglev train system in Japan.

However, superconductors currently in use mostly take the form of large coils of niobium-tin superconducting wire. Devices that use them must comply with this size, which may limit their use.

In the article published in NPG Asian SuppliesResearchers from King’s College London and Japan have used machine learning (ML) to produce a cheap and powerful iron-based superconducting magnet, paving the way for widespread and affordable use of the technology.

Dr Mark Ainslie, from the Royal Department of Engineering, collaborated on this study with researchers from Tokyo University of Agriculture and Technology, Japan Science and Technology Agency, National Institute of Materials Science and Kyushu University.

Dr Mark Ainslie said: “Superconducting magnets are the foundation of the future. Not only are they used to image cancer with MRI machines, they are also vital to electric planes and nuclear fusion.

“However, the materials and technologies required to create traditional copper wire superconductors are generally expensive, resulting in limited market penetration. Using them en masse as a magnet that does not lose magnetism upon magnetization can result in a smaller footprint compared to heavier coils of wire, however, bulk copper-based superconductors can take weeks to produce.

“Using artificial intelligence (AI), we have created a cost-effective and scalable alternative to iron that is much easier to machine and opens the door to smaller and lighter devices. The first iron-based superconductors were created more than 10 years ago, but the magnetic fields they produced were not strong or stable enough for widespread use.

“Although superconducting magnets still need to be cooled to very low temperatures to work effectively, our process lays the foundation for manufacturers to make them fast and powerful enough for industrial applications, meaning more MRI machines at less cost.

“By reducing the need for large amounts of superconducting wire in MRI machines, we can also create a new generation of smaller units that can be placed in the GP’s office, increasing accessibility, rather than requiring large hospital facilities.”

MRI machines have strict requirements for the strength and stability of the magnetic field created by their magnets to ensure patient safety and image quality. The researchers’ prototype is the first bulk iron-based superconductor to meet these requirements.

Using a new machine learning system called BOXVIA, ​​scientists have developed a framework that could make creating superconductors in the laboratory faster than ever before.

Learning from researchers’ attempts to improve the superconducting properties of magnets by changing parameters such as heating and time during the manufacturing process, BOXVIA detects patterns that improve performance and fine-tunes parameter changes to create the optimal design. It usually takes months for researchers to design each magnet and test its properties to optimize them for different scenarios, but this new software significantly shortens that time.

The researchers also found that superconducting magnets developed with this ML system had a different structure at the microscopic level than those made without BOXVIA, ​​with larger iron-based crystals in the magnet structure.

The structure of the AI-generated samples differed from the high-performance samples created by humans. These samples had a wide range of iron-based crystal sizes, as opposed to the uniform structure traditionally preferred by researchers.

The team’s next task is to find out how this never-before-seen nanostructure contributes to extraordinary superconducting properties that will lead to even stronger magnets in the future.