Rare-earth elements (REE) are a highly valued resource in most industrialized countries. Metals with elusive and evocative names such as neodymium, promethium, gadolinium, yttrium, and scandium, among others, belong to this peculiar group of chemical elements. Some of them are relatively rare and not usually found in nature in pure form, but what makes them unique are their physicochemical properties.
Their properties are beyond the reach of other elements on the periodic table, turning them into a valuable resource in many industries in recent decades, especially in the electronics and renewable energy sectors. Manufacturers use them to, for example, produce hybrid and electric car engines, batteries, catalytic converters, lasers, fiber optics, LCD panels, and even wind turbines.
The problem for the U.S., Europe, and surrounding countries is that China dominates the REE industry. According to the U.S. Geological Survey, it has produced more than 90% of these valuable resources for many years. Last December, the Chinese government used its supremacy in this field to restrict the export of some of its REE processing technologies in an apparent attempt to defend its strategic interests against the U.S., Europe, and their allies.
AI to the Rescue
At this point, the U.S., Europe, Japan, and Australia are unwilling to continue their dependence on China for much longer. One strategy on the table is to reduce the use of REE as much as possible by substituting other raw materials. This is challenging to do, but it's possible in some use scenarios, although it usually requires a lot of R&D investment. Tesla, for example, has confirmed that its next generation of electric motors will use REE-free magnets.
Using AI to design, synthesize, and test new permanent magnets that can operate without REE is possible.
But there’s another option. One that is already underway and producing extraordinary results. As it turns out, using AI to design, synthesize, and test new permanent magnets that can operate without REE is possible. This is precisely what the UK company Materials Nexus has done. In just three months, Materials Nexus designed, manufactured, and tested a permanent magnet called MagNex that doesn’t contain REE. The most striking thing is that AI made it possible to speed up this process by a factor of 200, so the savings in time and resources are overwhelming.
“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,” physicist Jonathan Bean, CEO of Materials Nexus, said. But there’s more. According to the company, its MagNex magnet costs only 20% of what it would cost to produce a conventional magnet, and it also reduces the carbon emissions of the material used to make this component by 70%. Sounds pretty good.
Image | GOKLuLe 盧樂
More info | ScienceAlert
View 0 comments