Active learning speeds critical materials recovery with fewer experiments
Decision-Focused Active Learning for Scale-Aware Critical-Materials Recovery
Artificial IntelligenceComputational Engineering, Finance, and ScienceRobotics
Summary
Choosing the best way to recover valuable materials like rare earths from recycled magnets is tricky because lab results must connect to real-world costs and scales. The authors study an approach called active learning that picks the most promising experiments to run, reducing the total number needed. Their method found top results much faster than traditional methods by smartly adapting based on earlier data. They also examine tradeoffs like purity versus yield in different materials and suggest a way to pick experiments that reduce future decision risks. This approach could help design better and more efficient recovery processes.
What this means in practice
- •For chemical process engineers: Optimize experimental design to find efficient recovery routes for rare earth elements with fewer costly lab tests.
- •For industrial recycling operators: Select recovery process conditions that balance material purity and yield by accounting for scale effects and cost tradeoffs.
Authors
Niranjan Srinivas, Debajyoti Ray, Elias Nakouzi
Abstract
Choosing a recovery process for scale-up requires connecting laboratory results with product requirements, process costs, and scale effects. We analyze records from Pacific Northwest National Laboratory's Computer Intelligence for Critical Element Recovery and Optimization (CICERO) workflow for autonomous selective precipitation. Active learning uses prior results to choose experiments. In a conditional retrospective benchmark with fitted models and recycled neodymium-iron-boron (NdFeB) magnet records, active learning finds the best recorded result with fewer experiments than nonadaptive space filling. Enrichment is the selected rare-earth-to-iron ratio relative to that in the feed. Adaptive policies reach the recorded enrichment maximum by 16 to 24 wells (individual experiments), versus 48. Our two-stage reconstruction ties two adaptive alternatives at 16 wells. Conditional analyses of recycled samarium-cobalt (SmCo) magnets show a Round 2 tradeoff between purity and nominal yield, the recovery fraction calculated from an assumed starting amount - NdFeB Round 1 routes differ in enrichment. Rankings for produced water from oil and gas extraction depend on phase and dilution assumptions requiring confirmation. We propose choosing batches by their expected reduction in downstream Bayes risk: the minimum expected loss among available process decisions under current beliefs. In exploratory simulations, a hybrid that filters candidates has lower estimated loss than the implemented joint search across routes and conditions. Differences involving the synthetic two-stage policy are small relative to estimation uncertainty. We outline a pre-registered prospective test under a shared loss and logging standard, requiring clarified measurements and records, a defined process decision and relevant outputs, credible economic inputs, and validation at the intended scale.