USA Rare Earth commissioned a hydrometallurgical demonstration facility in Wheat Ridge, Colorado in June 2026, targeting first separated dysprosium, terbium, and yttrium oxide production in Q3 2026 across three parallel feedstock campaigns. The U.S. Department of Energy has conditionally selected the company for up to $19.3 million under its Critical Materials Innovation program for a continuous ion-exchange separation pilot, part of a $50.5 million total program. The milestone anchors a rapidly assembling mine-to-magnet supply chain spanning Texas, Oklahoma, the United Kingdom, Brazil, and France.
Introduction
On June 15, 2026, USA Rare Earth (Nasdaq: USAR) commissioned a hydrometallurgical demonstration facility in Wheat Ridge, Colorado, and set a target of producing the first separated dysprosium, terbium, and yttrium oxides from that plant before the end of Q3 2026. If that timeline holds, the company will join a very short list of Western enterprises capable of delivering commercial-quality heavy rare earth oxides outside the People's Republic of China.
The significance of that distinction is hard to overstate. As I detailed in "Zero Exports, 25% Surges, and a 99.9% Sample" earlier this month, seven months of zero Chinese dysprosium and terbium exports to Japan have collapsed the fiction that alternative supply was merely inconvenient to secure rather than functionally unavailable. Dysprosium and terbium are not niche industrial curiosities; they are the performance-enabling additions to neodymium-iron-boron permanent magnets that allow those magnets to retain their properties at the elevated temperatures generated inside electric vehicle motors, wind turbine drivetrains, and defense-grade actuators. Without them, next-generation magnet performance degrades sharply.
The Wheat Ridge facility represents one carefully constructed answer to that structural absence. It is not, on its own, a commercial-scale separation plant. It is something arguably more valuable at this stage of the supply chain's development: a fully instrumented, 24-hour continuous demonstration system designed to validate and refine three distinct processing flowsheets simultaneously, generate the engineering data required for a bankable feasibility study, and seed a digital twin program with the U.S. Department of Energy's National Energy Technology Laboratory. First production of separated oxides from recycled magnet scrap was already achieved on July 14, 2026, providing early confirmation that at least one of those flowsheets is performing as designed.
The Wheat Ridge Facility: Design, Infrastructure, and Digital Twin
The Wheat Ridge demonstration plant is built around a fully automated design incorporating multi-stage solvent extraction circuits, live SCADA monitoring across every unit operation, and an on-site analytical laboratory. A team of 28 engineers, scientists, and technicians operates the facility on rotating shifts, running it continuously around the clock. That level of instrumentation is not standard practice at demonstration scale, where batch processing and manual sampling are more common. The decision to operate at near-commercial automation intensity reflects the plant's core purpose: to generate data that can directly feed commercial engineering design rather than simply confirm that a chemistry works in principle.
The facility is explicitly built to digitally and physically simulate USA Rare Earth's future commercial-scale operation. Process data flows in real time into the engineering design of a planned consolidated separation facility intended to handle both magnet swarf and mixed rare earth carbonate, the intermediate product form in which many rare earth concentrates are shipped. Running live operating data through commercial design models, rather than relying on bench-scale proxies, reduces the risk of scale-up surprises that have historically delayed or killed rare earth processing projects.
The digital twin collaboration with DOE's National Energy Technology Laboratory formalizes that data pipeline. Under a Letter of Intent signed in January 2026, NETL will contribute to virtual simulation of the full processing flowsheet, accelerating the path from demonstration to commercial deployment. The Trump administration has made process modeling and digital twin development a priority under its broader critical minerals innovation agenda, and the Wheat Ridge collaboration is among the most technically advanced applications of that approach currently underway in the rare earth sector. For an industry that has repeatedly seen processing projects founder on the gap between laboratory results and commercial reality, the combination of physical demonstration and computational simulation offers a more robust path forward.
Three Feedstock Campaigns: Round Top Ore, Ionic Clay Concentrates, and Magnet Swarf
The Wheat Ridge facility is running three parallel processing campaigns, each targeting a distinct feedstock stream and each designed to validate a different portion of USA Rare Earth's long-term supply chain architecture.
The first campaign processes ore from USA Rare Earth's Round Top deposit in Hudspeth County, Texas. Round Top is a heavy rare earth-enriched rhyolite containing a 20-element resource with significant dysprosium, terbium, yttrium, and uranium credits. The hydrometallurgical flowsheet being tested at Wheat Ridge is designed to produce separated oxides of all four elements, along with hafnium and zirconium. Results from this campaign are expected to underpin a Round Top Definitive Feasibility Study targeted for completion in Q4 2026 and publication in Q1 2027. Fluor Corporation and WSP Global have already been selected as EPCM partners for the Round Top build-out, and the company is targeting commercial production at the mine by late 2028 at an extraction rate of up to 40,000 metric tons per day.
The second campaign tests a mixed rare earth carbonate feedstock sourced from third-party ionic clay operations, including material from the Pela Ema mine in Goiás, Brazil. Ion-adsorption clay deposits are the world's primary commercial source of heavy rare earth elements, and their geological chemistry is fundamentally different from hard-rock mineralization like Round Top. Where hard-rock deposits tend to concentrate light rare earths such as cerium and lanthanum, ionic clay deposits yield the scarcer and more strategically critical heavy variants: dysprosium and terbium. Pela Ema, which began production in 2024 after more than $1.1 billion in cumulative investment, is projected to produce approximately 6,400 metric tons of total rare earth oxide annually by 2027, with ambitions to supply more than half of non-Chinese heavy rare earth output. USA Rare Earth is in the process of acquiring Serra Verde Group, the operator of Pela Ema, which would bring that feedstock stream directly into the company's supply chain.
The third campaign addresses magnet swarf recycling, and it has already produced its first validated output. On July 14, 2026, USA Rare Earth announced that the Wheat Ridge facility had produced commercial-grade dysprosium oxide and neodymium-praseodymium oxide from NdFeB magnet scrap sourced from its own Stillwater, Oklahoma manufacturing operation. Magnet machining and finishing generates fine metallic scrap, called swarf, that contains recoverable concentrations of the same rare earth elements embedded in the finished magnets. The company estimates that swarf recovery could supply up to 30 percent of its future magnetic rare earth oxide feedstock requirements, and the Wheat Ridge team expects that validating the swarf flowsheet will also lay the groundwork for processing end-of-life magnets as an additional commercial feedstock stream. The oxides produced at Wheat Ridge are expected to be shipped to Less Common Metals, USA Rare Earth's subsidiary in Cheshire, United Kingdom, for qualification and conversion into rare earth metals and strip cast alloy.
The $19.3 Million DOE Award and the Case for Continuous Ion Exchange
On May 21, 2026, the U.S. Department of Energy conditionally selected USA Rare Earth for up to $19.3 million in funding under its Critical Materials Innovation, Efficiency and Alternatives program. The award supports construction and operation of a pilot-scale continuous ion-exchange rare earth element separation plant. The total program value is approximately $50.5 million, with the remaining $31.2 million to be funded through private capital. The conditional selection is subject to the successful conclusion of negotiations and execution of a definitive funding agreement, and it does not constitute a binding financial commitment at this stage.
The technology at the center of the award is meaningfully different from the solvent extraction circuits that dominate most existing rare earth processing infrastructure globally. Conventional solvent extraction requires hundreds of sequential chemical mixing and settling stages, generating large volumes of organic solvent waste and demanding significant physical footprint. Continuous ion exchange replaces that approach with specialized resins that selectively capture individual rare earth elements based on differences in ionic radius and charge behavior, automating the adsorption and elution cycles that batch ion exchange performs manually and one step at a time. The result is a more compact plant footprint, lower reagent consumption per unit of output, and a reduced environmental impact profile relative to conventional solvent extraction at comparable throughput.
For a country attempting to build rare earth separation capacity from a near-standing start, those advantages are not marginal. Processing rare earth ores into separated oxides is chemically intensive work that generates significant waste streams and demands a workforce with specialized expertise that has substantially atrophied in the United States over the past three decades. The DOE's decision to fund a continuous ion-exchange pilot reflects a broader federal judgment, consistent with the priorities I examined in the DOE's $75 million coal-feedstock award earlier this month, that reinventing process chemistry is at least as important as simply locating new ore deposits.
The market registered the award's significance immediately. USAR stock jumped 13 percent to $22.57 following the announcement, extending a year-to-date gain that reached nearly 90 percent at that point. The company's market capitalization stood at approximately $1.37 billion as of the finalization of its CHIPS Act agreements in June 2026, supported by total committed capital of roughly $3.5 billion across DOE, Department of Commerce, and private sources.
Mine to Magnet: The Integrated Platform Taking Shape
The Wheat Ridge hydromet plant and the DOE ion-exchange award are best understood not as standalone projects but as two components of a supply chain architecture that USA Rare Earth has been assembling at notable speed over the past 18 months.
The downstream anchor of that chain is already operational. USA Rare Earth commissioned Phase 1a of its commercial sintered NdFeB magnet production line in Stillwater, Oklahoma on March 26, 2026, using core equipment originally purchased from Hitachi Metals America in 2020 and extensively recommissioned. Phase 1a is ramping toward a run-rate capacity of 600 metric tons per year by the end of Q4 2026, with Phase 1b expected to bring total Stillwater capacity to 1,200 metric tons per year in Q1 2027. At full build-out, Stillwater would rank among the largest NdFeB magnet plants outside China by volume, in a global market estimated at 200,000 to 220,000 metric tons per year where China accounts for roughly 85 to 90 percent of output.
The July 14 swarf recycling result is particularly significant in the context of Stillwater. The facility generates magnet swarf as a byproduct of normal machining operations, and the Wheat Ridge validation means that scrap can now flow back upstream for oxide recovery and re-entry into the production cycle. CEO Barbara Humpton described the dynamic as closing the loop between downstream magnet manufacturing and upstream separation, which is precisely the kind of circularity that reduces a supply chain's vulnerability to external feedstock disruptions.
The metals and alloys layer is handled by Less Common Metals in Cheshire, UK, acquired in October 2025. LCM is being expanded to 3,000 metric tons per year of metal and alloy capacity by the end of 2026 to meet rising demand from third-party magnet manufacturers in aerospace, semiconductor, mobility, and consumer electronics. In parallel, LCM plans to build a 3,750-metric-ton-per-year rare earth metal and alloy plant alongside Caremag's processing facility in Lacq, France, creating a European production node for the broader platform.
The upstream layer spans Round Top in Texas and, pending acquisition completion, Pela Ema in Brazil. Together they provide two geologically distinct heavy rare earth feedstock streams: hard-rock rhyolite from Texas and ionic clay concentrate from Goiás. Having both in the portfolio hedges against the geological and logistical risks inherent in depending on any single deposit type, and it gives the Wheat Ridge separation facility the feedstock diversity required to maintain continuous operations across market cycles.
Senior Vice President of Mining and Processing Dr. Alex Moyes put the challenge plainly: very few companies outside China have proven they can produce separated oxides of neodymium-praseodymium, dysprosium, terbium, and yttrium at commercial quality. The work at Wheat Ridge, he said, is intended to convert proven chemistry into bankable feasibility studies and move the company closer to producing the rare earth materials that American industry depends on from mine to magnet. The July 14 swarf result is the first public evidence that the conversion is proceeding on schedule.
Conclusion: A Demonstration Plant Doing Commercial Work
Demonstration facilities in the mining and minerals processing sector have a long and sometimes inglorious history of generating data that never travels far enough downstream to change anything. What distinguishes the Wheat Ridge plant from that pattern is the deliberate design choice to make demonstration and commercial engineering simultaneous rather than sequential. Live operating data feeds commercial flowsheet design in real time. The digital twin under development with NETL will allow virtual testing of process modifications before they are implemented in physical equipment. Three feedstock campaigns running in parallel will produce a range of engineering datasets broad enough to support a feasibility study for a specific mine while also validating toll-processing and recycling economics for third-party material.
The $19.3 million DOE conditional award validates the technical approach at the federal level and provides partial funding for the continuous ion-exchange pilot that represents the next step beyond solvent extraction demonstration. With $3.5 billion in total committed capital, CHIPS Act loan capacity, and a magnet plant already in commercial production in Oklahoma, USA Rare Earth has assembled the financial and operational foundation to move from demonstration to commercial scale on a timeline that was essentially inconceivable for a domestic rare earth company as recently as 2023.
The heavy rare earth supply crisis documented across recent price data and Japanese import statistics is not a problem that resolves itself through market signals alone. It requires the physical construction of processing infrastructure that does not currently exist outside China at meaningful scale. The Wheat Ridge facility, modest in throughput but sophisticated in instrumentation and purpose, is one of the clearest examples currently operating of what that construction process actually looks like in practice: not a single transformative announcement, but a series of disciplined engineering milestones accumulating toward a capability that the Western supply chain has lacked for decades.