Back to feed
News Story
DeepTech深科技
1 sources

USA Rare Earth partners with Pasqal and Riven Systems to use quantum machine learning and automated lab for rare earth extraction molecules

USA Rare Earth announced a collaboration with French quantum computing company Pasqal and industrial AI company Riven Systems to use quantum machine learning and an automated laboratory to find new rare earth extraction molecules, aiming to improve separation selectivity and reduce steps and energy consumption. The project is still in R&D, with Pasqal first benchmarking quantum machine learning models against classical models, and Riven's automated lab conducting thousands of experiments to train models. Final candidates will be validated with real rare earth feedstocks at USA Rare Earth's Colorado facility.

SynthePulse Insight · AI deep readingMembers

Can Quantum Computing Untie the Deadlock in Rare Earth Separation?

Version 1 · 1 source

USA Rare Earth, Pasqal and Riven Systems have announced a collaboration to try to find new rare earth extraction molecules using quantum machine learning plus automated laboratories. But for now the three parties are only committing to benchmark comparisons, not claiming the problem is solved.

  • The real bottleneck in the rare earth supply chain is midstream separation and refining, not mining; lanthanide elements have similar chemical properties and require extensive repeated extraction.
  • Division of labor: Riven uses self-driving laboratories to run thousands of automated experiments and produce data, Pasqal uses neutral-atom quantum processors to test quantum machine learning, and USA Rare Earth provides real feedstock for validation.
  • Pasqal's current first task is to benchmark quantum models against classical models; there is no evidence yet that the quantum approach is superior.
Open section navigationThe Real Bottleneck Is Not the Mine, but Separation

The Real Bottleneck Is Not the Mine, but Separation

The hardest part of the rare earth industry is not exploration and mining. After ore is beneficiated and given preliminary processing, what you get is a mixture in which multiple rare earth elements coexist, and it must go through tedious, precise chemical separation to produce high-purity single rare earth oxides such as neodymium, praseodymium, dysprosium and terbium. Because lanthanide elements have very similar chemical properties and ionic radii, industry often requires extensive repeated extraction to separate them one by one, which makes separation one of the links with the highest technical and capital barriers in the rare earth supply chain.

This background explains why "having a mine" does not equal "having a complete rare earth supply chain." Even if mixed rare earth feedstock can be obtained, if the different elements cannot be separated stably and efficiently, downstream metal, alloy and permanent magnet production will still be constrained by midstream processing capacity. This is especially true for heavy rare earths such as dysprosium, terbium and yttrium: suppliers able to reliably achieve high-purity separation remain highly concentrated in Asia.

USA Rare Earth is said to be developing the Round Top rare earth project in Texas while building its own separation and permanent magnet material capabilities. This provides a clear industrial landing point for the collaboration: if the technology works, it must ultimately land in real processing flows.

Free for now

Read the full analysis

4 more sections of analysis, plus the full takeaway

Loading

Credibility boundary

The information in this article comes from DeepTech's report on the three-party collaboration, and its core facts can be traced to GlobeNewswire and Pasqal press releases. Note: the scale of Riven's automated experiments and Pasqal's quantum machine learning tests are both company claims and have not been independently verified; there is currently no evidence supporting the idea that quantum methods are superior to classical methods.

Primary report

DeepTech深科技

Primary source