It’s not about lithium or batteries: the problem driving up the cost of electric cars and wind power might lie in a tiny magnet, and a new AI has already found a way to do without rare earth elements
As the world moves towards cleaner energy solutions, the rising costs of electric vehicles (EVs) and wind power are becoming increasingly concerning. Surprisingly, the issue may not lie in lithium or batteries, but rather in a tiny, yet crucial component: the magnet.
The Role of Rare Earth Elements
A recent study from the University of New Hampshire suggests that artificial intelligence (AI) can significantly expedite the search for magnetic materials that could lessen our reliance on rare earth elements. These elements are vital for the production of permanent magnets used in electric motors and generators.
The International Energy Agency (IEA) emphasizes that rare earth elements are essential for the magnets in EV motors and wind turbines. The U.S. Department of Energy (DOE) also notes that nearly all hybrid and plug-in EVs utilize rare-earth permanent magnets in their traction motors.
The Hidden Costs of Magnet Dependency
This reliance on a narrow supply chain for permanent magnets can have far-reaching implications. A materials problem within the motor can lead to increased factory costs, complicate industrial planning, and ultimately affect the technologies that consumers depend on daily.
AI and the NEMAD Database
To tackle this challenge, researchers created a public database known as NEMAD (New Era of Magnetic Materials Discovery). This database was developed by compiling 100,000 article identifiers from journals such as Elsevier and the American Physical Society. By utilizing large language models and custom parsing tools, the team extracted magnetic, chemical, and structural data, resulting in a comprehensive database with:
- 67,573 entries
- 84 different elements
- 15 material features, including crystal structure, coercivity, and magnetization
One critical aspect of this database is its tracking of Curie temperature—the temperature at which a material loses its magnetic properties. This is particularly important for components that must operate effectively in high-temperature environments, such as EV motors and generators.
Promising Results from AI Models
Lead author Suman Itani stated that the goal is to reduce dependence on rare-earth elements while also lowering costs for electric vehicles and renewable energy systems. The AI models reached an impressive 90% accuracy in classifying materials as ferromagnetic, antiferromagnetic, or non-magnetic. Additionally, the best Curie temperature model achieved an R² of 0.87 with a mean absolute error of 56 kelvin.
Among the most promising candidates identified was GaFe2Co4Si, a predicted ferromagnetic material with a Curie temperature of approximately 1,005 kelvin (about 732 degrees Celsius or 1,350 degrees Fahrenheit). This breakthrough could pave the way for the development of powerful new magnets that do not rely on rare earth elements, potentially reducing costs for clean energy technologies.
The Magnet Problem in Cleaner Technology
Permanent magnets, though small, play a significant role in the functioning of electric vehicles and wind turbines. The IEA indicates that rare earth elements are indispensable for these magnets, and the DOE predicts that internal permanent magnet motors will continue to dominate the electric vehicle market for the next decade due to their high power density and efficiency.
Supply Chain Concerns
However, a troubling supply chain issue looms over this progress. The IEA reports that the market share of the top three refining nations for key energy minerals increased from approximately 82% in 2020 to 86% in 2024. China is the primary contributor to this supply growth, and projections indicate that it will continue to supply around 80% of refined rare earths by 2035.
According to data from the United States Geological Survey (USGS), China accounted for 71% of U.S. imports of rare-earth compounds and metals between 2021 and 2024. This concentration poses a risk; if a major supplier were to be removed from the equation, the remaining rare-earth supply in 2035 would only meet 35 to 40% of the demand.
The Recycling Challenge
Moreover, the USGS indicates that recycling efforts for rare earths from batteries, permanent magnets, and fluorescent lamps currently yield only limited quantities. This bottleneck further complicates the supply chain and increases the urgency for alternative solutions.
A Promising Shortcut, Not a Final Solution
While the findings from the NEMAD database are promising, they do not represent a miracle solution. The compounds identified are still candidates that require experimental verification. Jiadong Zang, a researcher in the field, describes the search for sustainable permanent magnet alternatives as “one of the most difficult challenges in materials science.”
However, the results thus far appear to be more substantial than mere speculation. The study indicates that seven of the screened high-probability compounds have been documented in the literature with experimentally reported magnetic ordering temperatures, lending credibility to the model. The remaining 25 compounds present new experimental targets for researchers to explore.
The Future of Material Science
AI is not replacing traditional laboratory work; rather, it serves as a valuable tool that guides researchers on where to focus their efforts. This approach could potentially be adapted to other fields within materials science, including superconducting, thermoelectric, photovoltaic, and ferroelectric materials.
If even a few of these candidates successfully pass real-world testing, they could alleviate pressure on rare-earth inputs, diversify the supply chain for motors and generators, and provide more flexibility in the clean energy transition.
Conclusion
The study discussed in this article was published in Nature Communications. As the search for sustainable materials continues, the integration of AI into materials science may lead to significant breakthroughs that can transform the landscape of clean energy technology.
Note: The information provided in this article is based on research findings and current industry trends as of October 2023.
