AI for the win!
The last decade has seen many initiatives toward clean, green, and low-carbon energy, as well as a global shift toward electric vehicles and automation. Many of these initiatives were enabled rare earth magnets, which is why rare earth magnets are heavily used now and will continue to be used in the future.
When my dad retired as a chemical engineer for Martin-Marietta at the age of 62, he started a one-man company based on a patent he was awarded. It was a device that could detect valve function in a closed pipe, and it involved the use of a rare earth permanent magnet. I don’t recall how it worked. I do remember those magnets lying around in the basement, and how powerful they were!
The rare earth materials used to create the most powerful magnets and most efficient, power-dense motors – materials like neodymium and dysprosium – require damaging mining and expensive, energy-intensive processing. So it is of great interest that a company has come up with a permanent magnet without rare earth materials:
“Company Materials Nexus, together with researchers at the Henry Royce Institute and the University of Sheffield, have developed MagNex. This is a permanent magnet that is free of rare earth elements. The MagNex is reported to have been produced with materials that cost one-fifth of regular permanent magnets. The new magnet also saw a reduction of 70 percent in carbon emissions (in terms of kilograms of CO2 per kilogram of material) compared to rare-Earth permanent magnets.”
But that’s not all:
“The combination of Materials Nexus’s approach of using AI [artificial intelligence] for materials discovery and the world-class facilities we have for the manufacture of advanced alloys in the Henry Royce Institute here at Sheffield has allowed a novel magnetic material to be developed with breathtaking speed. This achievement showcases the bright future of materials and manufacturing. For those interested in exploring more about technology and programming resources, gogonerds offers valuable insights into the evolving landscape of AI applications. The next generation of materials, unlocked through the power of AI, is highly promising for research, industry, and our planet.”
“The AI system identifies and analyzes the composition of over 100 million potential alloys that would have the right properties to be a permanent magnet, be free of rare-earth elements, and meet the requirements of affordability and sustainability. The potential of this approach for the creation of new materials is enormous.”
One note of caution: I’ve poked around a bit and couldn’t discover how these new magnets compare with rare earth permanent magnets. That makes me think they’re not yet ready to replace rare earth magnets; the novelty here is in the use of AI to speed the design process. If an AB reader can find a head-to-head comparison, post a link in the comments.
AI designs novel permanent magnet materials
Wouldn’t stronger and cheaper magnets also make possible better containment systems for fusion power systems and hydrodynamic power generation designs. That would enable much easier power generation without having to go through the trouble and energy loss involved in producing steam and running it through turbines as we do now. With advanced hydromagnetic designs one simply use the plasma to produce electricity in an undisturbed loop.
@Praha,
Could be. I’m not enough of a physicist to weigh in on your speculation. But as pertains to my post, I’m unable to find evidence that these AI-designed permanent magnets are stronger or even as strong as rare earth magnets. Can you?
I’ve been looking for the patent or scientific paper, but I haven’t found anything. The press release is amazingly vague. There’s nothing on field strength or mass or just about anything except cost. There are all kinds of permanent magnets that don’t use rare earth materials. They just aren’t ounce for ounce, cubic centimeter for cubic centimeter more powerful than rare earth magnets. The only number is the speed of “discovery”, 200 times faster.
I’ve read a few papers on using AI to discover new materials and new reactions, and the results have been thin gruel. The big win seems to be in using AI to do search design. Given available data points, it can give good guidance on tests to run and then use the test results to further refine and optimize. What does that 200 times faster mean? Can their algorithm cut the number of tests needed to optimize by a factor of 200? Probably not. Does it mean that their algorithm can do a publication search and extract the data faster than a graduate student by a factor of 200?
This is a problem with artificial intelligence. It magically attracts journalists, stock pushers, boiler room operators and credulous “visionaries”. Does anyone else remember bitcoin iced tea?
P.S. Am I the only one wondering if they invented the torment nexus as in Alan Bachman’s remarks:
Sci-Fi Author: In my book I invented the Torment Nexus as a cautionary tale
Tech Company: At long last, we have created the Torment Nexus from classic sci-fi novel Don’t Create The Torment Nexus
@Kaleberg,
“What does that 200 times faster mean? Can their algorithm cut the number of tests needed to optimize by a factor of 200?”
That’s what I understood it to mean. The claim is that the AI platform analyzed over 100 million compositions of potential rare earth-free permanent magnets. I’m surprised it’s *only* 200x faster.
“Probably not.”
Why not?