Supra Elemental Recovery announced on September 30 that it has won a contract from the U.S. Department of Defense's Defense Logistics Agency (DLA) to research the recovery of high-purity scandium from North American industrial byproducts. The award is a Phase I Small Business Innovation Research (SBIR) contract, which supports research and development at small businesses. Under it, the company will evaluate whether its separation technology can be applied to scandium-bearing feedstocks.
Scandium is used in aerospace alloys, fuel cells, and other applications, yet the United States depends on imports for its supply. Supra, a spinout from the University of Texas at Austin, is testing a technology that separates scandium from domestic industrial byproducts and turns it into material that manufacturers can use.
The company's approach is to capture a target element in a 3D-printed, sponge-like cartridge and release it later. However, the award announcement only marks the start of research into whether the technology can be applied to scandium-bearing feedstocks. It does not demonstrate recovery rates or purity at production scale.
The key question is whether a technology that combines chemistry for selectively capturing specific ions with a porous structure that lets liquid flow through efficiently can handle the complex feedstocks found in real plants. The answer will help determine whether a new domestic source of scandium can be created.
The U.S. import dependence isn't about a lack of resources
According to the U.S. Geological Survey's Mineral Commodity Summaries 2026, there was no commercial mining or recovery of scandium in the United States in 2025, and net import reliance as a share of apparent consumption was 100%. Small-scale scandium metal refining facilities exist, but that is separate from the ability to recover scandium from domestic resources at commercial scale.
Scandium is used in aluminum alloys for aerospace components and other applications, and in solid oxide fuel cells for power generation.
Supply is difficult not because scandium is scarce on Earth. The USGS explains that, while it is widely distributed as a resource, it rarely occurs in high concentrations, and most of the scandium that can be recovered economically is obtained as a byproduct of producing other metals. Global production likewise depends on nickel and titanium processing, and on recovery from tailings and residues that have already been processed.
What is needed, then, is a technology that can selectively extract small amounts of scandium mixed in with large quantities of other components. A feedstock that merely contains scandium does not become a material customers can buy. One must also judge whether it is worth recovering commercially, including the cost of concentration and refining.
Japan is part of this supply chain. According to shipment records the USGS used to identify the sources of U.S. scandium oxide imports from 2021 to 2024, Japan accounted for 89% and China for 11%.
However, because scandium has no dedicated trade code, scandium contained in higher-value intermediate materials and finished products is not captured in these statistics. Nor can the figure be taken as the share of origin for all scandium used in the United States.
The same USGS document also describes a flow in which scandium-bearing material obtained in the Philippines is sent to Japan and processed into oxide. The country that recovers a resource and the country that refines it are not necessarily the same.
What Supra aims to do is build, within the United States, the step in this supply chain that separates the target element from byproducts.
Selectively recovering elements with 3D-printed porous cartridges
In the process Supra outlined when it launched on February 3, industrial waste is first dissolved, and the resulting liquid is passed through reusable cartridges. The cartridges selectively adsorb the target element, which is then released and collected.
The system is designed to separate multiple elements in sequence. It is not a filter that sifts elements out of solid waste passed through it directly.
The University of Texas at Austin's explanation says the 3D-printed porous material works like a sponge, using water or alcohol to recover metals, and combines the advantages of solvent extraction and ion exchange.
The important point is that metal ions are not sorted by pore size alone. The technology uses both a chemical mechanism that captures the target ion and a porous structure that lets liquid reach the interior of the material.
A useful guide to this thinking is research involving Supra's technology inventors, Jonathan Sessler and Zachariah Page, among others. In "3D-Printed Porous Supramolecular Sorbents for Cobalt Recycling", published in 2024 in the peer-reviewed Journal of the American Chemical Society by Keldy S. Mason and colleagues, the team tested a similar idea with cobalt.
According to the publicly available author manuscript, the researchers anchored molecular receptors that interact with metal ions onto an acrylic polymer backbone and shaped the material into a porous form. They combined 3D printing, which cures resin with light, with phase separation of the material during polymerization, building in both fine pores and the overall shape of the part.
By adjusting the surface area in contact with the liquid and the distance ions must travel inside the material, they could change how much the material could adsorb and how quickly.
To recover the adsorbed metal, the force holding it must also be switchable. In this study, cobalt chloride was adsorbed in ethanol and then released by moving the material into water. The change in solvent polarity alters the interactions between the ions and the surrounding solvent.
If the molecular receptors are fixed to a solid material, the cartridge material can be reused after the metal is removed.
In reuse tests, adsorption and release were each carried out for 12 hours and repeated five times, with no statistically significant change in performance. The results were averages of three independent trials.
Experiments separating cobalt and lithium were also conducted, but they used simulated leachate in which salts of both were dissolved in ethanol. They do not show results from continuously processing real scandium-bearing industrial byproducts, nor long-term durability in a plant.
Basic demonstration, feedstock evaluation, and commercial operation are separate stages
Recovering scandium from industrial byproducts is already a commercial business in Canada.
Rio Tinto says it first produced high-purity scandium oxide in May 2022 at its commercial-scale demonstration facility in Sorel-Tracy, using a byproduct stream from titanium dioxide manufacturing in Quebec. According to the company, no new mining is needed for this recovery.
The technology stages shown in currently available materials fall into three groups: a basic demonstration with cobalt, Supra's evaluation of applicability to scandium feedstocks, and Rio Tinto's commercial-scale recovery from byproducts.
| Public source / date | Subject and stage confirmed | Limits when judging supply capacity |
|---|---|---|
| Mason et al. paper, 2024 | Basic demonstration of cobalt salt adsorption and release, including five reuse cycles | Not confirmed to use the same material composition as the current scandium cartridges; does not show performance at scandium production scale |
| Supra award announcement, September 30, 2026 | Phase I research to evaluate whether the separation technology can be applied to North American industrial byproducts | No figures on achievable purity, recovery rate, or throughput have been released |
| Rio Tinto operations description, first production May 2022 | Recovery of scandium oxide from a titanium dioxide manufacturing byproduct stream | An operating example using a different technology; does not support Supra's performance |
This classification does not compare the recovery rates or costs of different technologies under the same conditions. That a business supplying scandium from byproducts already exists, and that Supra's new separation material works in practice on a particular feedstock, must be verified separately.
The USGS statement that there was no commercial recovery in the United States in 2025 is also consistent with commercial-scale recovery taking place in Canada.
Moreover, recovering scandium oxide does not mean it can be used directly in aircraft alloys. Further steps are needed to process it into metal or alloy and meet the specifications for each application.
Supra's research targets the separation and recovery step at the upstream end of that supply chain.
Real feedstocks and continuous operation will determine how close domestic supply is
According to the DLA's program description, SBIR Phase I evaluates the feasibility of a technology, Phase II advances prototyping and proof of concept for processes or products, and Phase III aims at commercial use. This contract sits at the first of those stages.
Supra has not disclosed the contract value or details of the feedstocks it will use in the evaluation. The fact of Defense Department support alone does not allow a prediction of when commercial production might begin.
In February, the company presented initial results showing selectivity and processing speed up to 100 times those of conventional methods, and announced that it had raised $2 million in pre-seed funding.
However, the announcement does not make clear which methods and feedstock conditions the "100 times" figure is compared against, or what the absolute performance values are. It is an early-stage assessment presented by the company and does not mean scandium output would rise 100-fold. The $2 million raised is also separate from the value of this research contract.
To judge whether the technology is practical in a plant, one must confirm from which feedstocks, how much scandium can be recovered, and to what level of purity it can ultimately be raised.
Even at high purity, a process that misses much of the scandium in the feedstock is less valuable as a source of supply. Conversely, even with a high recovery rate, if large amounts of other elements get mixed in, additional refining steps will be needed afterward. Both purity and recovery rate must be evaluated for each feedstock actually used.
Furthermore, commercial viability is not determined by cartridge performance alone. Pretreatment to convert industrial byproducts into a liquid suited to separation also costs money. Operating costs will vary with how far solvents can be reused and whether the cartridges maintain performance when exposed to various impurities.
Whether the adsorption and release cycles shown in published research can be carried over into a process that repeats them while continuously flowing liquid in an actual plant will be a key condition for commercialization.
What this DLA support tests is whether the fact that scandium is present in U.S. industrial byproducts can be turned into a recovery process that is actually usable.
If Supra can stabilize purity and recovery rate with real industrial byproducts and show costs that include pretreatment and cartridge reuse, its technology could become an option for extracting new critical minerals from existing plants and supplementing U.S. scandium supply.