DOE's $134 Million Rare-Earth Bet on Industrial Waste: Why the June 2 Selections for Phoenix Tailings and Colorado School of Mines Reset the Domestic REE Supply Chain

The Department of Energy's Office of Critical Minerals and Energy Innovation (CMEI) announced $134 million in selections under the Rare Earth Elements Demonstration Facility Program on June 2, 2026. Two awardees — Phoenix Tailings and the Colorado School of Mines — will build demonstration-scale facilities that produce separated rare-earth oxides and rare-earth metals from industrial waste streams rather than from primary mined ore. Phoenix Tailings was selected for approximately $66 million toward a total project value of $147.8 million, partnered with the Massachusetts Institute of Technology and the University of Minnesota. Colorado School of Mines received the balance of the $134 million for a demonstration facility near the Gramercy alumina refinery in St. John the Baptist Parish, Louisiana, partnered with ElementUSA, Pacific Northwest National Laboratory, Principal Mineral, and Rare Earth Technologies Inc.

This is the third major rare-earth-focused selection from CMEI in 2026, following the $45.7 million FOA 3105 awards in May and the $69 million Critical Minerals and Materials Accelerator cooperative agreements. What sets the June 2 announcement apart is not the dollar size — $134 million is mid-tier in the CMEI portfolio — but the fact that both selected projects sit at demonstration scale rather than at research or pilot scale, and both project teams have committed to processing waste-derived feedstocks rather than primary ore concentrates. CMEI is no longer betting on a future U.S. rare-earth extraction industry the way the FOA 3105 selections suggested; with the June 2 selections, it is betting that the rare earths the country already controls — locked in tailings ponds, alumina refinery byproducts, and end-of-life magnets — are sufficient to demonstrate a commercial separation and refining base inside three years.

Red Mud Is the Sleeper Critical-Materials Story

The Colorado School of Mines selection puts the rare-earth recovery facility at the Gramercy alumina refinery — operating since the 1950s, producing alumina from imported bauxite, and depositing the residual red mud (the alkaline iron-rich slurry left after sodium hydroxide extraction of aluminum) into impoundments that now contain more than 30 million tons of accumulated waste. Red mud is everywhere alumina is produced. The global inventory of accumulated red mud is estimated at roughly 4 billion tons, and globally about 175 million tons of fresh red mud are generated annually. The bauxite ore that produces alumina contains trace concentrations of scandium, neodymium, dysprosium, cerium, and lanthanum — the rare earths concentrate in the residue at modestly higher levels than in the original ore, and the residue is already on the ground at the refinery site, eliminating mining and primary processing cost.

The technical pathway has been demonstrated at bench and small pilot scale by several academic and European industrial teams over the last decade. What CMEI is funding is the integrated demonstration — an end-to-end facility that pulls red mud from the Gramercy impoundment, separates the rare-earth oxides via an integrated hydrometallurgical process, and refines the oxides into rare-earth metals at quantities that downstream magnet and electronics manufacturers can qualify against. The Gramercy site is also strategically located. Louisiana has competitive industrial electricity rates, gulf-port logistics, an established petrochemical and refining workforce, and a federal Opportunity Zone designation that overlaps with the project footprint. ElementUSA is operating the alumina refinery; PNNL contributes process modeling and characterization; Principal Mineral provides downstream metal refining; Rare Earth Technologies Inc. handles the magnet-grade output stream.

The strategic logic is straightforward. The United States imports more than 95 percent of its rare-earth oxide and rare-earth metal supply, with China responsible for more than 90 percent of global rare-earth refining capacity. Standing up a new domestic mine and primary processing chain takes a decade or longer through permitting, capital construction, and offtake qualification. Standing up a separation and refining facility that pulls feedstock from an existing accumulated waste pile takes three years if the technology readiness is sufficient. The June 2 selections bet that it is.

The Phoenix Tailings Selection Targets the Heavy-Rare-Earth Bottleneck

Phoenix Tailings' $66 million selection — part of a $147.8 million total project — addresses a different and arguably more strategic chokepoint: the heavy rare-earth elements. Light rare earths (cerium, lanthanum, neodymium, praseodymium) are the bulk of global production by volume; heavy rare earths (dysprosium, terbium, samarium, gadolinium, holmium, erbium, thulium, ytterbium, lutetium, and yttrium) are produced in much smaller quantities, are concentrated in a small set of ion-adsorption clay deposits in southern China and Myanmar, and command much higher prices because permanent-magnet manufacturers blend small percentages of dysprosium and terbium into neodymium-iron-boron magnets to maintain magnetic performance at elevated temperatures.

Heavy-rare-earth supply is the binding constraint on U.S. magnet manufacturing — even projects that secure light-rare-earth supply from domestic sources still need heavy-rare-earth inputs that are almost entirely sourced from Chinese refining. Phoenix Tailings' demonstration facility will deploy proprietary separation and reduction technology on a diverse set of domestic feedstocks including mine tailings, refining residues, and end-of-life products. The MIT partnership contributes automation, AI-enabled process controls, and real-time sensing; the University of Minnesota partnership contributes feedstock characterization and validation across a broad range of waste streams. The technical bet is that a single flexible-feedstock facility can be qualified to process whatever waste-derived rare-earth stream a customer brings, rather than being optimized for a single ore body.

What the Selections Tell Researchers and Small Businesses

For grantseekers outside the awarded teams, three signals matter. First, the funded technical pathway is waste recovery, not primary mining. Researchers building proposals for the next CMEI demonstration-scale solicitation should be designing around tailings, residues, recycled materials, and unconventional feedstocks — not around new mine development. The CMEI portfolio is making a deliberate choice to skip the permitting and capital intensity of new primary extraction by going directly to existing accumulated waste.

Second, the partner structure is the model. Both awardees combine a project-leading company (or in the Colorado School of Mines case, a university operating in a company-like role) with national-laboratory characterization support, a research-university automation and process-modeling partner, and a downstream customer or metal-refining partner. CMEI is no longer funding standalone university research at the demonstration-scale level; it is funding integrated teams with credible operational, characterization, automation, and downstream qualification capability. Teams that lack one of those four components should be partnering with awardees rather than competing directly for the next demonstration-scale solicitation.

Third, the geographic and feedstock distribution is informative. Louisiana red mud, mine tailings broadly distributed, end-of-life magnet recycling — these are spread across multiple states with industrial infrastructure, mining histories, or aluminum refining footprints. The next demonstration-scale solicitation is likely to expand into coal byproducts (West Virginia, Pennsylvania, Kentucky have active research programs), into electronic waste streams (the Northeast and Pacific Northwest have established e-waste recycling infrastructure), and into phosphate fertilizer byproducts (Florida and Idaho have the dominant U.S. phosphate processing capacity). Researchers and small businesses with feedstock access in any of these waste streams should be building proposals around demonstrated recovery yields and confirmed downstream offtake.

How June 2 Fits in the Broader CMEI Strategy

CMEI's rare-earth portfolio now spans the full technology-readiness spectrum. The FOA 3105 selections at $45.7 million fund 17 R&D projects and two pilot-scale plants — USA Rare Earth's continuous ion exchange separation pilot in Stillwater, Oklahoma, and Big Blue Technologies' magnesium scale-up in Cheyenne. The Critical Minerals and Materials Accelerator's $69 million funds supply-chain integration work spanning lithium and rare earths. The $500 million critical-minerals battery supply chain initiative targets EV battery materials. The $134 million Rare Earth Elements Demonstration Facility Program is now the demonstration-scale layer.

The aggregate CMEI portfolio is approaching $800 million across these four instruments alone, with an additional $1 billion in forthcoming critical-minerals funding flagged for the FY26 to FY27 cycle. The portfolio is structurally consistent with a multi-administration commitment — the bipartisan consensus on Chinese supply-chain risk is durable enough that the funding pipeline is unlikely to contract substantially even if the administration changes. For researchers and small businesses, this is one of the most predictable areas of federal R&D commitment and one of the few where the demonstration-to-commercial transition is visibly funded.

What to Do Before the Next Solicitation

Three concrete actions follow from the June 2 announcement. First, monitor the Phoenix Tailings and Colorado School of Mines project kickoff materials, milestone schedules, and the inevitable early-stage results — those will define the actual technical pathways that competing teams will need to differentiate against in the next demonstration-scale solicitation. Second, treat the awarded teams' downstream partners (Rare Earth Technologies Inc., Principal Mineral, ElementUSA on the Colorado side; magnet and electronics customers on the Phoenix Tailings side) as candidate offtake partners for separate projects — those companies will publicly signal their qualification standards and capacity constraints, which is exactly the information a competing team needs to write a credible commercial-viability case. Third, look at the CMEI portfolio horizontally — researchers and small businesses with capabilities that match more than one CMEI instrument (e.g., a separation-and-recovery team that could also contribute to a battery-materials integration project) have meaningfully better odds of landing one of the next awards by submitting to the instrument with the best feedstock and partnership match.

Tools like Granted can map a research team's capabilities against the active CMEI instruments, surface the feedstock and partnership patterns that awardees met, and flag the downstream offtake programs that make a critical-minerals processing project commercially defensible before a proposal goes in. The June 2 selections close a chapter on whether the United States will commit to domestic rare-earth separation and refining at scale. The next chapter is about which teams build the second wave of demonstration facilities, and whether the waste-stream bet that Phoenix Tailings and the Colorado School of Mines are executing produces enough separated heavy rare earths to break the magnet supply chokepoint.