Friday 21 March 2025
Scientists have made significant progress in searching for stable and low-energy phases of cerium-based intermetallic compounds, which could potentially be used as new permanent magnets. These materials are being researched because they offer a more abundant and cheaper alternative to rare earth elements like neodymium and dysprosium, which are currently used in many high-tech devices.
The researchers employed a machine learning (ML) framework integrated with first-principles calculations to explore the compositional and structural landscape of cerium- cobalt-copper ternary compounds. This approach enabled them to efficiently screen for promising candidates, significantly accelerating the materials discovery process. The team predicted five stable compounds, along with hundreds of low-energy phases that could be potentially metastable.
The ML+DFT (Density Functional Theory) predicted low-energy phases are characterized by unique crystal structures and compositions. Some of these phases have already been experimentally confirmed, while others remain to be verified. The researchers also identified several Co-rich compounds that exhibit high magnetizations, making them potential candidates for future permanent magnet applications.
One notable finding is the discovery of a new class of cerium-based alloys that can be tailored to possess specific magnetic properties. These alloys have the potential to replace traditional rare earth magnets in various applications, including electric motors, generators, and wind turbines. The ML+DFT approach has also revealed the existence of previously unknown phases with unique electronic structures, which could lead to innovative devices and technologies.
The search for new permanent magnet materials is driven by the need for more sustainable and cost-effective alternatives. Cerium-based intermetallic compounds offer a promising route towards this goal, as they are relatively abundant and can be synthesized using existing manufacturing processes. The discovery of new phases with desirable magnetic properties will require further experimental verification and optimization.
The research team’s findings have significant implications for the development of advanced technologies that rely on permanent magnets. As the demand for sustainable energy solutions continues to grow, the need for efficient and cost-effective magnet materials becomes increasingly pressing. The discovery of cerium-based alloys with high magnetizations could play a crucial role in meeting this challenge.
In addition to their potential applications in technology, these materials also offer opportunities for scientific exploration. The unique electronic structures and magnetic properties of cerium-based intermetallic compounds can provide insights into the fundamental physics of magnetism and superconductivity.
The study’s findings demonstrate the power of combining machine learning with first-principles calculations in accelerating the discovery of new materials.
Cite this article: “Unlocking New Permanent Magnet Materials through Machine Learning and First-Principles Calculations”, The Science Archive, 2025.
Cerium, Magnets, Machine Learning, Intermetallic Compounds, Rare Earth Elements, Permanent Magnets, Magnetization, Electronic Structures, Superconductivity, Materials Science
