The Feedstock Problem Nobody Solved: Ucore’s RapidSX Just Hit 99.5% NdPr From Two Different Ores

Ucore Rare Metals (TSXV: UCU | OTCQX: UURAF) reported Thursday that its RapidSX™ platform produced 99.5% purity NdPr from two structurally different feedstocks — a result that speaks less to chemistry than to whether America can build a rare earth supply chain that does not depend on a single mine.

Washington has spent the better part of a decade talking about rare earth independence as a mining problem. It is not. The United States already pulls rare earth ore out of the ground. What it has struggled to do — what China spent thirty years perfecting and now uses as leverage — is separate that ore into the individual oxides that go into permanent magnets, guidance systems, and electric drivetrains.

Separation is the chokepoint. And separation plants have a habit of being built for one specific ore body, which means they inherit that ore body’s risk.

That is the frame for the announcement Ucore Rare Metals put out Thursday morning, and it is why a purity milestone deserves more attention than a purity milestone usually gets.

What Ucore Actually Reported

Ucore said its RapidSX™ platform produced neodymium-praseodymium at 99.5% purity from two distinct feedstock sources at the company’s Commercialization and Demonstration Facility (CDF) in Kingston, Ontario.

The first run processed roughly two tonnes of Vietnamese mixed rare earth oxide derived from ionic clay, completed in the second quarter of 2026. The second run processed approximately five tonnes of North American mixed rare earth carbonate derived from bastnaesite, completed in early August.

Ionic clay and bastnaesite are not variations on a theme. They differ in mineralogy, in element distribution, and in the impurities they carry. A separation circuit tuned for one does not automatically handle the other.

Ucore has now logged more than 7,400 cumulative hours of RapidSX operation in a simulated commercial environment at Kingston, and cites over 16,000 comparative test results supporting its claim that the platform delivers recovery rates and product purities equal to or better than conventional solvent extraction — with faster throughput and a smaller equipment footprint.

Mike Schrider, Ucore’s Vice President and Chief Operating Officer, tied the Kingston work directly to the company’s commercial build: “Ucore is executing a very deliberate plan for the development of its first RapidSX production ‘Machine A’ at the Louisiana SMC. The work done at the CDF directly supports and validates the SMC engineering program.”

Schrider added that Kingston “will continue to serve as a co-commissioning, technology development and training center.”

Why Feedstock Flexibility Is The Whole Thesis

Here is the distinction that matters for anyone underwriting this name.

If RapidSX only worked on one ore type, Ucore’s Louisiana Strategic Metals Complex would be a bet on securing a single reliable supplier — and every delay, renegotiation, or geopolitical disruption at that supplier would land squarely on Ucore’s income statement.

If RapidSX processes whatever arrives at the gate, the Louisiana SMC stops being a mining-adjacent project and starts being infrastructure. A toll processor. A facility that Western miners, allied governments, and magnet manufacturers route material through because it is the separation capacity that exists outside China.

Thursday’s result is a data point in favor of the second interpretation. Two ores, two chemistries, same 99.5% output.

It is a data point, not proof. Two feedstocks at demonstration scale is not the same as commercial throughput across a dozen supply agreements. But it is the right kind of data point, and it arrived on schedule.

The Catalyst Chain Behind It

The announcement does not stand alone, which is what separates it from the average micro-cap milestone release.

Ucore executed a US$18.4 million award with the U.S. Department of Defense to fund initial construction of rare earth processing capacity in Louisiana. The company has since taken delivery of DPAS-rated long-lead equipment at the SMC site — a Defense Production Act priority rating that moves Ucore’s orders ahead of commercial buyers in the manufacturing queue.

In June, Ucore produced NdPr oxide and shipped qualification samples to major rare earth magnet manufacturers, putting its material in front of the customers who would eventually sign offtake.

And on August 13, the company closed a C$69 million public offering.

That last item deserves emphasis in a sector where promising micro-caps routinely stall out on financing. A company that just raised C$69 million is not, in the near term, a dilution story waiting to happen.

What Ucore Is Building

The Louisiana Strategic Metals Complex is designed to process both heavy and light rare earths, producing NdPr, samarium, gadolinium, terbium, dysprosium, and yttrium. Ucore has signaled subsequent SMC facilities in Canada and Alaska, and separately controls 100% of the Bokan-Dotson Ridge heavy rare earth project on Prince of Wales Island in Southeast Alaska.

Bokan matters as optionality rather than near-term production: a domestic heavy rare earth resource sitting behind a domestic heavy rare earth separation facility is the vertically integrated outcome the Pentagon has been funding toward.

What Investors Should Watch

Thursday’s release did not include a commissioning date for the Louisiana SMC. It did not convert June’s qualification samples into announced offtake agreements. And one of the two demonstration feedstocks was Vietnamese — a reminder that feedstock flexibility and feedstock security are separate problems, and Ucore has made more visible progress on the first than the second.

Ucore’s own forward-looking disclosures name the dependencies plainly: completion of engineering studies, timely equipment delivery, sufficient financing, permits and approvals, successful scale-up from demonstration to commercial operation, availability of qualified third-party feedstock, and successful customer qualification and offtake.

The next real tell is not another purity number. It is a signed offtake and a date for Machine A.