Refining materials discovery with AI

How an innovative AI-based materials platform is redefining material discovery.

Typically based on a trial-and-error approach, materials discovery has traditionally been a slow and resource-intensive process. However, the emergence of artificial intelligence (AI) is turning this on its head, prompting an evolution of research and development techniques for producing novel materials for net-zero technologies.

Materials Nexus, a deep-tech company using AI to design novel materials, is an active player in this field. The company’s most recent breakthrough centres around the creation of MagNex, a new rare earth-free permanent magnet, the identification of which was powered by Material Nexus’ AI-based materials discovery platform.

“AI is expected to play a crucial role in developing future materials by revolutionising the traditional materials discovery process,” says Dr Jonathan Bean, Materials Nexus’ CEO. “By enabling a shift from a trial-and-error approach to a design-focused approach, AI can significantly accelerate R&D. It improves R&D techniques by analysing vast datasets and predicting material properties with high accuracy, minimising the need for expensive, time-consuming and resource-intensive experiments. This makes the process more cost-effective, less wasteful and environmentally sustainable.”

AI HAS THE ANSWER

The company’s unique technology combines AI with quantum mechanics to accurately and rapidly predict novel, high-performing and sustainable materials. The platform is designed to fast-track the discovery and development of such materials to deliver efficiency and performance benefits to green technologies like wind turbines and electric vehicles (EVs), which still rely heavily on the mining of rare-earth materials and precious metals.

“Unlike traditional trial-and-error methods, this platform enables an intent-based design process, analysing the entire periodic table to identify and model materials from the quantum level to the bulk scale,” Bean explains. “The platform utilises quantum mechanics for composition prediction and models processing and synthesis techniques, allowing for the rapid identification and production of high-performance materials in months rather than decades. This allows us to not only identify, but also physically produce high-performance materials accurately, significantly speeding up the R&D process.”

INTRODUCING MAGNEX

Permanent magnets are essential for manufacturing across many vital industries, including EVs, wind turbines, robotics and drones. Demand for rare earth magnets is expected to outpace supply in the coming years, with the use of rare earth magnets set to increase up to tenfold by 2030 in the EV industry alone. However, the rare earth metals required to produce these magnets are vulnerable to several supply chain issues.

To address this, Materials Nexus has used its AI platform to design and create a new rare earth-free permanent magnet, MagNex. Working in partnership with the Henry Royce Institute and the University of Sheffield, the company deployed its AI platform to identify and analyse over 100 million compositions of rare earth-free permanent magnet candidates that address industry challenges, such as supply chain security, cost, performance and environmental issues. In just three months, the partners were able to design, synthesise and test MagNex – 200 times faster than the development timescale for the current industry standard permanent magnet. MagNex can also be produced at 20% of the materials cost and a 70% reduction in material carbon emissions compared to rare earth element magnets currently on the market.

“MagNex is a significant milestone in the use of AI to design materials of the future which are cheaper, higher-performing and more sustainable than existing options,” says Bean. “AI-powered materials design will impact not only magnetics but also the entire field of materials science – we have now identified a scalable method for designing new materials for all kinds of industrial needs. Our platform has already attracted widespread interest for various products with applications that include semiconductors, catalysts and coatings.”

Going forwards, Bean says the company will continue to develop MagNex for industrial applications. “MagNex is a demonstration of our platform’s ability to predict a novel functional material based on desired variables,” he explains. “Next steps will be the continued collaboration with industry partners to develop the magnetic material. The platform will continue to support this effort as we look at processing techniques. Further development of the platform will be done simultaneously, building on capabilities across multiple material areas such as semiconductors, superconductors and green hydrogen technologies.”

For more information visit: www.materialsnexus.com