Recycling Industry

NSF Puts $6 Million Behind Microbial Rare-Earth Recovery From Coal Waste

NSF's $6 million EPSCoR award pairs extremophile bacteria with AI-run bioreactors to recover rare earths from coal refuse, targeting a modular biorefinery demo by August 2030.

NSF Awards $6 Million EPSCoR Grant to University of Alaska Consortium for AI and Microbial Rare-Earth Recovery
NSF Awards $6 Million EPSCoR Grant to University of Alaska Consortium for AI and Microbial Rare-Earth Recoverykhoinguyenfoto / pixabay

Waypoints

  1. NSF awarded $6 million over four years, running through August 31, 2030, to a UAA-led consortium across Alaska, Wyoming and Montana.

  2. UAF executes nearly $913,000 of the award for bioreactor design and bioleaching kinetics.

  3. The U.S. imports more than 70 percent of its processed rare earths from foreign suppliers.

  4. Microbial processing could compress hundreds of solvent extraction stages into four or five operations.

  5. A modular biorefinery demonstration unit processing multi-kilogram coal refuse batches is planned for the final phase through August 2030.

A $6 million, four-year NSF grant launched in November 2026 will pay an Alaska-led consortium to extract rare-earth elements from coal refuse using bacteria and machine learning — with a modular biorefinery demonstration targeted for August 2030.

The U.S. National Science Foundation announced the formal launch of the Focused EPSCoR Collaborations award, titled "Transformative Bioprocessing and AI-Driven Frameworks for Sustainable Critical Mineral Recovery," anchored at the University of Alaska Anchorage (UAA). The project runs through August 31, 2030 and sits inside NSF's $39.9 million national EPSCoR Research Infrastructure Improvement portfolio, spanning jurisdictions in Alaska, Wyoming and Montana.

For practitioners, this is not another lab study. It's a funded attempt to compress a flowsheet — and to prove techno-economics on a waste stream the U.S. already owns in the hundreds of millions of tons.

Who's doing what, and with how much?

The consortium splits work across four public research institutions, each pairing academic labs with mineral processing capabilities:

  • University of Alaska Anchorage — lead awardee. Dr. Brandon Briggs, professor of biological sciences, is lead principal investigator, backed by Dr. Subhabrata Dev (chemical engineering) and Dr. Masoumeh Heidari Kapourchali (electrical engineering), with Dr. Mohammad Heidari Kapourchali. UAA handles microbial bioprocess discovery and predictive AI models.
  • University of Alaska Fairbanks — nearly $913,000 of the total award, led by Dr. Srijan Aggarwal of the Institute of Northern Engineering, covering bioreactor design, bioleaching kinetics and water-quality lifecycle controls.
  • University of Wyoming — Tyler Brown leads coal refuse characterisation, mineral composition analysis and techno-economic modelling through the School of Energy Resources.
  • Montana Technological University — Dr. Robin Bullock and Dr. Courtney Young oversee hydrometallurgical validation, downstream precipitation and pilot materials engineering.

What does the biology replace?

Conventional rare-earth processing — for elements like neodymium, dysprosium, terbium, europium and yttrium — leans on open-pit mining, energy-intensive roasting, aggressive acid leaching and hundreds of solvent extraction stages that generate substantial hazardous wastewater volumes.

The consortium's platform instead deploys naturally occurring bacteria, including extremophile strains harvested from Alaska's cold glacial environments and electrochemically active Shewanella oneidensis. The microbes release organic acids and transfer electrons across mineral surfaces on powdered coal refuse and fly ash, selectively dissolving targeted lanthanides. Dr. Briggs said the team anticipates condensing traditional recovery sequences into four or five streamlined operations.

The AI layer attacks the classic bioleaching bottleneck: kinetic variability as you scale from bench flasks to throughput. Sensors tracking pH, oxidation-reduction potential, dissolved oxygen and metal ions feed telemetry into reinforcement-learning and deep neural network models that forecast bioreactor performance and flag suboptimal feedstock — letting researchers adjust nutrient feeds dynamically instead of running months of empirical trials.

Why coal waste, why now?

The U.S. imports more than 70 percent of its processed rare earths from foreign suppliers, exposing EV motors, wind turbines, telecoms and defense avionics to export restrictions and supply shocks. Domestic ore grades are often too low for conventional mining to pencil out.

Coal refuse and fly ash offer a feedstock alternative that's already sitting in surface impoundments across Alaska, Wyoming's Powder River Basin and Appalachian basins — a century's accumulation of preparation waste and combustion residuals. Reprocessing it produces a domestic feedstock while remediating legacy waste sites, without opening new open-cast mines.

Tyler Brown told Wyoming Public Media that microbial biorecovery has already proven economical at micro scales; the consortium's job now is establishing commercial viability and material characterisation on what would otherwise remain industrial waste.

What happens next?

The pathway is phased. The first 18 months cover sampling and cataloguing coal refuse and ash from Alaskan mines and Wyoming coal basins — spectroscopic profiling, particle-size distribution and metagenomic sequencing of native bacterial communities. Through 2028, researchers build continuous-flow benchtop bioreactors with edge-computing AI control loops tuned to maximise extraction kinetics while cutting reagent consumption. The final two years target a modular biorefinery demonstration unit processing multi-kilogram batches on site, plus quantitative techno-economic lifecycle assessments for commercial mining and energy utility partners.

Briggs called the award the culmination of a decade of research to develop biotechnology capable of scaling into a viable multi-state enterprise. Aggarwal noted no single institution could assemble the required breadth alone.

The milestone to watch: whether that multi-kilogram demonstration unit delivers a lifecycle cost number that mining and utility partners will actually sign against — before the money stops in August 2030.

via casrai.org (Source)

Share this article:

More from Elena Vasquez

Elena Vasquez

Show full bio

Senior reporter covering technology and product news in Circular economy and the environmental industry at Circular Wire.

575 articles

Nearby routes

« Previous article