Scandium

The Byproduct Problem: Why Scandium Has No Primary Mine Anywhere on Earth

Scandium (Sc, atomic number 21) is the strategic-minerals paradox in its purest form: it is not scarce in the Earth's crust — it is more abundant than lead — but it almost never concentrates into an ore body worth mining for itself. Per the USGS Mineral Commodity Summaries 2026, “scandium was produced exclusively as a byproduct, primarily from nickel and titanium process streams, as well as from previously processed tailings and residues,” with global scandium oxide production capacity exceeding 90 tons in 2025 against actual output of about 80 tons. China was the leading producer, and scandium materials were also produced in the Philippines and shipped to Japan for further processing into oxide.

Bloom Energy, one of the largest industrial consumers of scandium oxide, explains the underlying economics bluntly: scandium is “abundant and widely distributed throughout Earth's crust, more than the amount of lead present on the planet,” but it “seldom exists in deposits rich enough to justify mining solely for scandium” because it is “dispersed across many minerals in very small quantities,” making dedicated mining of enormous ore volumes economically impractical today — “primarily an economics problem, not a total availability problem like gold or platinum” (Bloom Energy, 8 July 2026).

USGS's world resources note confirms the scale mismatch: “resources of scandium are abundant but rarely occur in high concentrations; as a result, economically recoverable scandium was produced mainly as a byproduct,” with resources identified in Australia, Canada, China, Finland, Guinea, Kazakhstan, Madagascar, Norway, the Philippines, Russia, South Africa, Ukraine, and the United States — a wide geographic spread that has, paradoxically, produced almost no commercial mining because none of those occurrences is rich enough to stand alone (USGS MCS 2026).

1. Nickel-cobalt laterite byproduct: the Philippines-to-Japan supply line

The most established byproduct route runs through nickel laterite processing using high-pressure acid leaching (HPAL). Sumitomo Metal Mining (SMM) built a scandium recovery circuit at its Coral Bay Nickel Corporation pilot plant on Palawan Island in 2013, then commissioned a commercial-scale recovery plant at its Taganito HPAL Nickel Corporation operation on Mindanao, shipping the intermediate scandium-oxalate product to its Harima Refinery in Japan for finishing into oxide — a ¥4 billion investment targeting roughly 7.5 tonnes of scandium oxide per year, alongside a long-term sales agreement with a major U.S. fuel-cell manufacturer (Sumitomo Metal Mining, 28 April 2016). Industry commentary notes this plant “increased global supply by around 40%” when commissioned, at a capital cost of roughly $5 million per tonne of annual scandium oxide capacity (Minor Metals Trade Association). As of 2026, SMM was moving to increase Philippine-sourced scandium output by roughly 20% specifically to reduce dependence on China, which the company's own reporting says supplies roughly 80% of world scandium supply (Sumitomo Metal Mining production-increase report, 2026). USGS's own import-source data for the United States confirms the resulting trade pattern: scandium oxide imports were sourced from “Japan, 89%; and China, 11%” over 2021–24, with Philippine-origin material reassigned to Japan in the statistics because the Philippines exports only the unfinished oxalate intermediate, not finished oxide (USGS MCS 2026).

2. Titanium dioxide slag byproduct: Rio Tinto's Quebec breakthrough

Rio Tinto's Iron and Titanium (RTIT) operation in Sorel-Tracy, Quebec, recovers scandium oxide from the waste streams of titanium dioxide feedstock production — a process the company describes as extracting “high-purity scandium directly from the waste streams of titanium dioxide production” at its integrated ilmenite-smelting complex (Rio Tinto, 1 November 2025). This is covered in full in Section 2 below, as it has become the only scaled non-Chinese, non-Russian scandium source in commercial production.

3. Uranium mill residues and phosphate byproducts

Scandium has a decades-long history as a byproduct of uranium processing: as early as the 1960s, the Vitro Chemical Company's uranium solvent-extraction plant discovered scandium contamination in its process streams and developed a fluoride strip system to recover it commercially, producing multi-pound lots of high-purity scandium oxide from what had been simply uranium-refining waste (OneMine/Mining Engineering, 1961). Phosphate-ore processing for uranium and rare-earth byproducts follows a similar path: peer-reviewed work on lean phosphate-ore digestion describes a flowsheet recovering uranium, rare earths, and scandium together, with scandium concentrations of roughly 180 grams per tonne reporting to a ferric residue cake alongside 45% Fe2O3, available for future scandium recovery (IAEA conference paper on uranium/REE/scandium co-recovery).

4. Red mud (bauxite residue): the largest untapped resource

Every tonne of alumina produced by the Bayer process leaves behind bauxite residue (“red mud”) containing between roughly 15 and 170 milligrams of scandium per kilogram depending on the bauxite source — scandium does not dissolve during Bayer-process leaching and instead concentrates in the residue, with some red muds reaching 60–120 grams per tonne, a concentration high enough that researchers consider ores in the 20–50 g/t range a legitimate scandium resource (Nature Scientific Reports, RWTH Aachen red-mud scandium study). Given the enormous global volume of red mud generated annually by the aluminium industry, peer-reviewed recovery routes — combining electric-arc-furnace smelting to recover iron as pig iron, followed by hydrometallurgical leaching, ion exchange, and selective precipitation — have demonstrated scandium recovery rates as high as 85–94% from red mud in laboratory and pilot settings (Separation and Purification Technology, selective scandium extraction from red mud; Materials, scandium recovery methods review, 2022). Rio Tinto has separately invested in red-mud rehabilitation technology at its alumina operations, though its scandium production (Section 2) draws from titanium slag rather than red mud directly (Rio Tinto, red mud rehabilitation, April 2025).

Rio Tinto's Sorel-Tracy Plant: The First Western Commercial Scandium Source

In January 2021, Rio Tinto announced it would enter the scandium market with a plant at its Rio Tinto Fer et Titane (RTFT) operations in Sorel-Tracy, Quebec — the world's largest titanium dioxide feedstock plant — extracting scandium oxide from the plant's existing waste streams rather than mining new ore (Reuters, 14 January 2021). Production began in 2022, and Rio Tinto states it “becomes the first producer of scandium oxide in North America” (Rio Tinto, 6 May 2022). The company describes the underlying technology as a “breakthrough process to extract and produce high-purity scandium directly from the waste streams of titanium dioxide production” at its integrated Quebec operations, which the company separately calls “the first green scandium plant in the world” (Rio Tinto, RTIT Quebec operations; Rio Tinto, 1 November 2025).

1. The November 2025 Canada Growth Fund expansion

On 1 November 2025, Rio Tinto and the Canada Growth Fund (CGF) — a federal Crown corporation — announced a transaction under which CGF will invest approximately C$25 million (roughly US$18 million) to expand the Sorel-Tracy facility's nameplate capacity to nine tonnes of scandium oxide per annum, up from the demonstration-scale output that has, to date, supplied all of North America's scandium demand (Rio Tinto, 1 November 2025). The deal structure pairs the CGF investment with two separate commercial agreements with the Government of Canada: an offtake agreement under which the government commits to purchase a volume of scandium output, and a marketing-and-storage agreement under which Rio Tinto will market and store scandium on the government's behalf (Rio Tinto, 1 November 2025). Analysis of the deal notes it uses “an innovative equity-like financial royalty structure,” with the roughly C$25 million commitment equating to about C$2.78 million of government investment per tonne of new annual capacity (Canada Growth Fund — Investments).

2. The U.S. Defense Logistics Agency's sole-source contract

Because Rio Tinto's Sorel-Tracy facility is the only scaled North American scandium producer, the U.S. Defense Logistics Agency (DLA) determined Rio Tinto Services, Inc. to be the sole vendor capable of meeting government requirements and issued a sole-source solicitation, SP8000-25-R-0021, in September 2025, for an indefinite-delivery, indefinite-quantity (IDIQ) contract of up to $40 million over five years to acquire 6,384 kilograms of scandium oxide at minimum 99.8% Sc2O3 purity for the National Defense Stockpile (Sweetspot federal contract record, SP8000-25-R-0021). Reuters confirmed the DLA sought to acquire the material because “scandium was predominantly sourced from China until recently. In late 2024, export limitations were placed on scandium by China, disrupting the supply chain and necessitating this procurement,” adding that the DLA's planned 6.4-tonne acquisition equates to roughly 5% of the prior year's total global scandium oxide production of about 40 tonnes (on the DLA's contemporaneous estimate), against installed global capacity of around 80 tonnes (Reuters, 22 September 2025). USGS confirms the structure of the award: “the Defense Logistics Agency announced plans to procure more than 6,000 kilograms of scandium oxide for the National Defense Stockpile from a source in Sorel-Tracy, Quebec, Canada,” over a five-year period with a minimum commitment of $2 million and a potential total value of as much as $40 million (USGS MCS 2026). A May 2025 DLA request for information had drawn responses from four companies, but “only Rio Tinto expressing the capability and capacity to supply scandium oxide” (Shanghai Metals Market, 23 September 2025).

3. Al-Sc master alloy integration and vertical development

Rio Tinto has also trialled combining its Sorel-Tracy scandium oxide output with metal from its own Quebec aluminium smelters to produce small quantities of aluminium-scandium master alloy directly, positioning the company as a potential vertically integrated Al-Sc supplier rather than a pure oxide seller (Fastmarkets, 28 May 2020).

China's April 2025 Export Controls: Scandium Joins the Heavy Rare Earth Licensing Regime

China's Ministry of Commerce (MOFCOM) and General Administration of Customs (GAC) jointly issued Announcement No. 18 of 2025 on 4 April 2025, placing seven categories of medium and heavy rare-earth-related items under export licensing: samarium, gadolinium, terbium, dysprosium, lutetium, scandium, and yttrium. The scandium-specific controls (item 1C907) cover scandium metal, scandium-aluminium alloy, scandium-magnesium alloy, scandium-copper alloy, scandium target materials, scandium oxide and its mixtures, and scandium-containing compounds and their mixtures — effectively every commercial form in which scandium is traded (MOFCOM Announcement No. 18/2025). MOFCOM stated the measures were adopted “to safeguard national security and interests and fulfill international obligations such as non-proliferation,” with a spokesperson later confirming the controlled items “can be employed for both civilian and military purposes” and that regulating their exports “aligns with common international practices” (MOFCOM spokesperson remarks, 10 April 2025). Legal analysis from Holland & Knight confirms the measure is “not an export ban, but rather an export restriction,” requiring exporters to obtain a MOFCOM license for every shipment (Holland & Knight, 4 April 2025).

1. Licensing mechanics and the compliance burden

Every export requires a per-shipment license application including a detailed end-user certificate naming the buyer and end-use, full material specifications and tonnage, and pre-clearance MOFCOM approval; applications tied to military end-users or weapons-of-mass-destruction applications are, in principle, never approved (Pillsbury Law, still-valid MOFCOM/GAC controls summary, 8 November 2025). Industry trade-compliance commentary a year on from the controls describes a licence approval queue “that still averages 47 business days for first-time applicants,” with the underlying legal architecture — the Export Control Law and Regulations on Export Control of Dual-Use Items — unchanged since April 2025 (Sourzi, one-year retrospective, 5 March 2026).

2. Why scandium is different from the other six controlled elements

Trade-compliance analysis highlights scandium as the outlier among the seven April 2025 controlled elements: “scandium, at [an estimated] 68% [Chinese supply share], is the only one of the seven where a genuinely non-Chinese primary supply chain is operable today at commercial tonnage, courtesy of Rio Tinto's Sorel-Tracy titanium slag by-product recovery in Quebec” (Sourzi, 5 March 2026). This is precisely why the U.S. and Canadian government responses (Sections 2 and 4) moved unusually fast relative to other controlled heavy rare earths: an alternative supply chain already existed and could be scaled, unlike dysprosium or terbium magnet feedstock, where Western separation capacity remains negligible.

3. The suspension that did not include scandium

A subsequent, broader round of controls announced by China in October 2025 targeting additional heavy rare earths (holmium, erbium, thulium, europium, ytterbium) and extraterritorial rules for foreign-made products containing Chinese-origin material was suspended for one year following the late-October/early-November 2025 Trump-Xi understanding (China Briefing, 10 November 2025). Critically, that suspension did not touch the original April 2025 Announcement No. 18 controls covering scandium. USGS confirms: “In November, the United States stated that China will issue general licenses for rare-earth exports, effectively eliminating the controls introduced in April. As of December 2025, the April export controls remained in effect, although China began to issue general export licenses to selected exporters” (USGS MCS 2026). Independent reporting as of May 2026 confirms the same: “China's original April 2025 export licensing requirements for seven heavy rare earth elements have never been suspended” and remain “fully in force,” governing “scandium for high-strength aluminum alloys used in aerospace” among other applications (MOFCOM — China Ministry of Commerce (rare earth export controls, April 2025 curbs)).

Why it matters: the April 2025 controls hit at the exact moment Western scandium demand (SOFCs, aerospace alloys, defense programs) was accelerating, and directly catalyzed the wave of U.S. and Canadian government funding actions described in Sections 2 and 4 — all of which explicitly cite reducing China dependence as their rationale.

The U.S. Government Response: NioCorp, Lockheed Martin, and ElementUSA

1. NioCorp's Elk Creek project: the only advanced U.S. primary scandium deposit

NioCorp Developments' Elk Creek Critical Minerals Project in southeast Nebraska is described by the company as the United States' first polymetallic deposit targeting near-term production of scandium, niobium, titanium, and light and heavy magnetic rare-earth oxides (NioCorp, 23 October 2025). On 5 August 2025, the U.S. Department of Defense awarded up to $10 million under DPA Title III to NioCorp subsidiary Elk Creek Resources Corp. to support reserve-upgrade drilling, engineering, feasibility work, and “the establishment of a domestic scandium mine-to-master-alloy supply chain,” funded via a Fiscal Year 2024 Department of War appropriations plus-up championed by Nebraska Senator Deb Fischer (NioCorp, 23 October 2025; Nebraska Public Media, 6 August 2025). NioCorp CEO Mark Smith stated the award would “position the U.S. as a global leader in the mining, processing, and manufacturing of scandium and scandium alloy components for defense and essential civilian technologies” (Reuters, 5 August 2025). Through 31 December 2025, NioCorp had received $6.8 million of the award under milestone-based reimbursement, and by year-end had raised more than $370 million in total 2025 financing, including $360.8 million in equity (NioCorp, 2025 Year in Review Report). NioCorp projects Elk Creek could ultimately produce approximately 100 tonnes per year of scandium oxide — more than the entirety of 2025 global production — and already holds binding long-term offtake agreements covering 12 tonnes of scandium per year, its largest scandium deal to date (Stock Titan, 23 October 2025; LinkedIn, citing Reuters, 5 August 2025).

2. Lockheed Martin Skunk Works: scandium alloy prototypes for fighter jets

On 23 October 2025, NioCorp announced a Pentagon-funded joint development agreement with Lockheed Martin's Skunk Works advanced-development division “to support the development of a scandium-based defense technology,” designed to produce “prototype components of aluminum-scandium alloy that will expand the capabilities of modern fighter aircraft” (NioCorp, 23 October 2025). Lockheed Martin's OJ Sanchez, vice president and general manager of Skunk Works, said the partnership would “connect domestic critical materials mining through the value-chain to advanced materials to provide advanced capabilities for the warfighter” (Stock Titan, 23 October 2025). Investor commentary connects this to the F-35 and F-22 airframes specifically, citing scandium's potential to reduce reliance on titanium parts and mitigate galvanic corrosion between dissimilar metals in structural joints (Investor analysis, October 2025), while trade press has also linked scandium-aluminium alloys to sixth-generation fighter development discussions, though no additional technical detail on the program has been publicly released (NioCorp, 23 October 2025).

3. ElementUSA's $29.9 million gallium-and-scandium recovery facility

On 20 November 2025, the Department of War announced a $29.9 million DPA Title III award to ElementUS Minerals, LLC (ElementUSA) to enable a demonstration facility in Gramercy, Louisiana, that will separate and purify gallium and scandium from existing industrial waste streams (U.S. Department of War, 20 November 2025). USGS confirms “part of this award was for the development of a demonstration facility to separate and purify scandium from existing industrial waste” (USGS MCS 2026). Aviation Week described the award as intended to develop “a domestic supply chain of gallium and scandium, two key critical minerals for military applications” (Aviation Week, 21 November 2025).

4. Additional U.S. refining capacity under development

Beyond the Elk Creek and ElementUSA projects, USGS notes the United States had small-scale scandium-metal refining capacity operating in Ames, Iowa, and Tolleson, Arizona, in 2025, with additional capacity under development in Urbana, Illinois, and at the Elk Creek project in Nebraska — though domestically, “scandium was not commercially mined or recovered in 2025” (USGS MCS 2026). The U.S. net import reliance for scandium stood at 100% of apparent consumption for every year from 2021 through the 2025 estimate, with no government stockpile of scandium existing prior to the DLA's 2025 procurement initiative (USGS MCS 2026).

Aluminium-Scandium Alloys: From Soviet MiG Fighters to Airbus Additive Manufacturing

1. The metallurgy: why scandium works where other additives don't

Scandium's effectiveness comes from forming coherent Al3Sc intermetallic precipitates that refine grain structure and suppress recrystallization at elevated temperature. A technical overview from alloy producer KBM Affilips describes the resulting benefits as “high strength, ductility, weld-ability, improved corrosion resistance,” supplied commercially as a 2%-scandium, 98%-aluminium master alloy that downstream producers dilute into sheet and plate containing 0.2–0.4% scandium (AZoM/KBM Affilips, Al-Sc overview). Peer-reviewed metallurgical literature confirms scandium “enhances not only mechanical and thermal properties, but also increases corrosion resistance and weldability; and reduces the hot cracking susceptibility” of aluminium alloys (Iranian Journal of Materials Science and Engineering). Zirconium co-addition, forming Al3(Sc,Zr) core-shell structures, further raises the coarsening temperature above 400°C, extending the alloy's useful range into turbine and hypersonic structural applications (PatSnap Eureka, Al-Sc alloy defense material overview).

2. Cold War origins and the Soviet MiG program

Al-Sc alloys were first developed and deployed at scale by the Soviet military, which used scandium-containing aluminium alloys — sometimes termed “migram” — in the Mikoyan-Gurevich MiG-21 and MiG-29 fighter jets (Aluminium–scandium alloys, technical reference). A materials-science review confirms “the development of Al-Sc alloys first flourished in the Soviet Union, where military demand was the main driving force,” noting that at the time of the Soviet Union's collapse, scandium alloys were “on the verge of major application in MIG-29 fighters” because of their strength and weight advantages over Al-Mg and Al-Li alternatives (Reinforced Aluminum materials review).

3. Airbus, Scalmalloy®, and additive manufacturing

The modern Western commercialization path runs through additive manufacturing. APWorks, a subsidiary of the Airbus Group, developed Scalmalloy®, described by Fastmarkets as “a high-performance scandium-aluminium magnesium alloy designed for the additive manufacturing of high-strength aerospace structures,” developed in cooperation with Airbus's research and development division (Fastmarkets, 28 May 2020). Fastmarkets reports major aircraft producers “have been assessing the adoption of aluminium-scandium alloys to reduce aircraft weight by 15–20%” (Fastmarkets, 28 May 2020). Technical suppliers note Scalmalloy® achieves “mechanical properties that surpass traditional aerospace aluminum alloys like 6061, 7075, and 2024,” is exclusively printed via laser powder-bed fusion to ≥99.5% part density, and has been used by Airbus to produce structural brackets for commercial aircraft (Neway 3D Printing, Scalmalloy® technical overview).

4. Defense drones, UAVs, and next-generation fighter programs

Al-Sc alloys, and Scalmalloy® in particular, are increasingly specified for unmanned aerial vehicle (UAV) structures where fatigue resistance and strength-to-weight ratio dominate design choices. A metal-additive-manufacturing industry overview notes Scalmalloy® is “often preferred” over standard AlSi10Mg for defense UAV applications including “drone swarms,” citing “superior strength and fatigue life” for airframes operating under “extreme durability and resilience” requirements (MET3DP, drone-arm additive manufacturing materials comparison, April 2025). A 2025 DLA/DoD Small Business Innovation Research solicitation titled “Advancing Scandium Use in Metal Alloys for U.S. Weapon System Production and Sustainment” explicitly listed potential applications spanning “missiles, aircraft, space launch vehicles, satellites, solider systems, military ground vehicles, marine applications, and other weapon systems (e.g., small arms and artillery),” and specifically flagged “the high weldability of Al-Sc alloy” and its suitability “as a powder or wire in additive manufacturing” (Minor Metals Trade Association, citing DLA SBIR solicitation text). As detailed in Section 4, the Lockheed Martin Skunk Works program targets aluminium-scandium alloy prototype components for “modern fighter aircraft,” with investor commentary specifically naming the F-35 and F-22 as reference platforms (NioCorp, 23 October 2025).

5. Sporting goods and non-aerospace niche uses

USGS lists sporting goods among scandium's minor uses (USGS MCS 2026), and Al-Sc alloys have been commercialized for baseball and softball bats, bicycle frames, lacrosse sticks, and tent poles, where a materials review found scandium-alloy bicycle frames delivered “a 12% reduction in weight, a 50% increase in yield strength, and a 24% improvement in fatigue life over the best-selling aluminum bicycle” (Reinforced Aluminum materials review). Smith & Wesson has also produced firearm frames from scandium-alloy castings paired with titanium or carbon-steel cylinders (Aluminium–scandium alloys, technical reference).

Why adoption remains niche despite the performance gains: KBM Affilips states plainly that “the current price of scandium prevents a wider application in the aluminium industry” (AZoM/KBM Affilips) — the central constraint explored in Section 6.

Prices: A Thin, Volatile Market With No Formal Exchange Benchmark

Source: USGS MCS 2026, citing Asian Metal Ltd. ex-works China price assessments. Notably, USGS's benchmark shows scandium oxide prices trending down from 2021 through 2025 even as China tightened export controls in April 2025 — reflecting that the USGS/Asian Metal series tracks the domestic Chinese ex-works price, which is largely insulated from the licensing friction faced by foreign buyers.

1. The China-domestic vs. export-market price gap

A wide gap separates China's internal ex-works price from what non-Chinese buyers actually pay once licensing delays and scarcity premiums are included. Trade-compliance research from March 2026 estimated a landed cost of roughly $488,000 per tonne ($488/kg) for scandium-related HS codes into the United States, inclusive of a 5.5% MFN duty, a 25% Section 301 List 3 tariff, and a further 10% reciprocal-tariff layer added in November 2025, alongside a licence-queue-adjusted lead time of roughly 68 business days from purchase order to vessel departure (Sourzi, 5 March 2026). Separately, ScrapMonster's June 2026 assessments quoted Chinese ex-works scandium metal (99.99% min) at roughly $3,142/kg, scandium metal (99.999% min) at roughly $5,213/kg, and scandium oxide (99.99% min) at roughly $663/kg — broadly consistent with the USGS figures but reflecting intraday volatility of several percentage points week to week (ScrapMonster, scandium price data, 19 June 2026).

2. Western retail premiums and the 2026 price spike

Independent market-data compilations reported far higher figures for smaller, non-Chinese-sourced lots in 2026. One specialist market tracker listed scandium oxide at approximately $9,705/kg in Western retail markets as of March 2026, alongside Chinese bulk metal pricing of $4,500–$5,250/kg requiring a roughly four-month licensing process with no guarantee of approval, and noted that laboratory-grade scandium in small quantities (e.g., 2-gram lots from major chemical suppliers) reached approximately $261 per gram — a roughly 50-times markup over Chinese bulk pricing (Lanthanides.io Strategic Materials Ledger, 27 March 2026). The same source estimated global Sc2O3 production at approximately 40 tonnes per year against estimated global demand of approximately 117 tonnes per year — a supply deficit consistent with the scarcity premium observed at the retail level, though the higher demand figure sits above USGS's independently estimated 60-tonne global consumption figure for 2025, illustrating the wide uncertainty band in scandium market-size estimates (Lanthanides.io, 27 March 2026; USGS MCS 2026). Separately, Shanghai Metals Market's industrial benchmark showed scandium metal at $3,748.07/kg as of 1 July 2026, up 12.8% from $3,321.91/kg at the prior monthly benchmark — evidence of continuing month-to-month volatility even within the Chinese domestic market (Rare Earth Mining News, citing Shanghai Metals Market, 1 July 2026).

3. Historical price levels and the SOFC-era baseline

Bloom Energy's own technical literature from the early-2020s cited scandia (scandium oxide) prices “sold between US$1,400 to US$2,000 per kilogram in 99.9% form,” with worldwide annual scandium production estimated at under 2,000 kilograms in that period, most of it sourced from “limited former Soviet-era stockpiles” (IOSR Journal of Electronics and Communication Engineering, Bloom Energy technology review). This confirms that, even accounting for methodology differences across sources, scandium oxide has traded consistently in a roughly $500–$2,500/kg band across purity grades for well over a decade, with Al-Sc master alloy consistently priced in the low tens of dollars per kilogram given its dilution to only 2% scandium content.

4. No formal exchange listing or futures market

Unlike copper, aluminium, or nickel, scandium has no LME contract, no LBMA-style fixing, and no exchange-traded futures product. Price discovery runs entirely through bilateral producer/trader quotes and specialist price-reporting agencies such as Asian Metal, Shanghai Metals Market, and ScrapMonster, with Fastmarkets covering scandium only within its broader rare-earths methodology rather than as a standalone benchmark (Fastmarkets, rare earths methodology). This structural thinness is a direct consequence of the tiny absolute market size — global annual scandium oxide production of 60–80 tonnes is worth on the order of only $40–80 million at USGS-benchmark prices, an order of magnitude too small to support exchange-traded derivatives.

Solid Oxide Fuel Cells: Bloom Energy and the Largest Single Demand Pool for Scandium

USGS identifies solid oxide fuel cells (SOFCs) used in large-scale power generation and backup power for critical infrastructure as one of scandium's two principal global uses in 2025, alongside aerospace alloys (USGS MCS 2026). Bloom Energy, the leading commercial SOFC manufacturer, explains the mechanism directly: at the heart of every Bloom fuel cell is a ceramic electrolyte layer whose job is to let oxygen pass through as efficiently as possible so it can react with methane or hydrogen to generate electricity; Bloom enhances ultra-pure zirconium oxide with a small amount of scandium oxide as a dopant, and this “seemingly small change” lets the fuel cells “perform better, last longer, and use less fuel than earlier generations,” requiring “significantly fewer layers of fuel cells to generate the same amount of electrical power” (Bloom Energy, 8 July 2026).

1. Scandium-stabilized zirconia and conductivity performance

Academic technology reviews describe the specific ceramic used as scandia-stabilized zirconia (ScSZ), which “has a higher conductivity than at lower temperatures... when used as an electrolyte in SOFC applications,” providing greater efficiency and reliability compared with the more common yttria-stabilized zirconia used in earlier fuel-cell generations (IOSR Journal of Electronics and Communication Engineering).

2. Bloom's supply-chain diversification claims

Bloom's July 2026 technical blog post — updated the same day to further clarify its sourcing position — states explicitly that it “does not rely on primary mining,” and instead developed “proprietary processes to recover scandium oxide from existing industrial exhaust streams / tailings / process streams” associated with titanium, nickel, cobalt, and uranium processing that is already occurring globally at massive scale — more than 1.2 million metric tons of zirconium oxide and roughly 10 million metric tons of titanium ore processed annually worldwide, generating byproduct streams from which “several hundred tons of scandium oxide can be produced annually” (Bloom Energy, 8 July 2026). Bloom states that more than half of global titanium processing occurs outside China, with substantial titanium mining in Canada and Australia and high-purity zirconium oxide production concentrated in Japan and the United States using ore mined mostly in Australia and South Africa. The company explicitly asserts: “we believe that our current diversified global supply chain of scandium oxide can support up to 25 GW per year of production capacity, and that we are not dependent on China for scandium oxide” (Bloom Energy, 8 July 2026). Bloom maintains strict confidentiality around its specific supplier relationships, sourcing volumes, and procurement strategies, describing this sourcing diversification as a competitive advantage in its own right.

3. Historical scale and the AI-driven demand surge

Earlier technical literature on the Bloom Energy Server noted that “current annual worldwide production of scandium is improving to around 40 tons per year since 2022, up from the 15–20 tons annually in previous years,” and that “most of the 5,000 kilograms used annually is sourced from Soviet-era stockpiles” — illustrating how thin the global market was even a few years before Bloom's aggressive fuel-cell scale-up (Bloom Energy Server, technical reference). Bloom's stated requirement of roughly 130–150 kilograms of scandium per gigawatt of fuel-cell production capacity implies that reaching its claimed 25 GW/year supply-chain ceiling would require on the order of 3,250–3,750 kilograms — several tonnes — of scandium oxide per year, a material share of total global byproduct supply, underscoring why Bloom has invested so heavily in proprietary, diversified sourcing technology rather than relying on spot-market purchases (Bloom Energy Server, technical reference). Reddit-aggregated investor commentary on NioCorp separately references a $5 billion Bloom Energy partnership with Brookfield to power AI data-center factories, illustrating the scale of new SOFC demand growth that AI-driven power buildouts are expected to bring to the scandium market (Investor commentary, October 2025).

Metal Halide Lamps and Other Legacy Electronics Uses

1. The scandium-sodium iodide lamp chemistry

Metal halide lamps generate light by vaporizing metal halide salts inside a high-pressure discharge arc tube; academic literature describing “the family of metal halide lamps” explains that “the halides of sodium, scandium and thorium achieve higher luminous efficacies, up to a maximum of 110 lm/W,” although their color-rendering quality does not reach that of lamps using rare-earth halides, with scandium-sodium lamps producing color temperatures ranging between approximately 3500K and 5000K (Bernhard Kuhl and Alexander Dobrusski, Sage Journals, 1975). Patent literature confirms the specific commercial formulation: metal halide lamps “with iodides of sodium, lithium and scandium” and, separately, “scandium halide and alkali metal halide discharge lamp” designs describing scandium triiodide (ScI3) as a core fill compound for high-color-rendering-index, constant-color-output lamps (Canadian Patent CA1102390A; U.S. Patent 5,225,733; Canadian Patent CA2292091A1, high-CRI metal halide lamp with constant color).

2. Why this use is now legacy rather than growth

USGS's current commodity chapter does not list lighting among scandium's principal 2025 uses, instead naming aerospace alloys, other alloys, and solid oxide fuel cells as the primary global applications, with electronics and alloys for military equipment and sporting goods as minor uses (USGS MCS 2026). This reflects the broader displacement of metal halide lighting by LED technology across commercial and industrial lighting markets over the past decade, which has structurally shrunk scandium iodide demand even as aerospace and SOFC demand has grown. Scandium triiodide remains commercially referenced in chemical databases and lighting-industry technical literature, but is no longer treated by USGS as a demand driver requiring separate quantification (Scandium triiodide, technical reference; Metal-halide lamp, technical reference).

3. Other minor electronics and alloy uses

USGS's substitutes note for scandium states that “titanium and aluminum high-strength alloys as well as carbon-fiber materials may substitute in high-performance scandium-alloy applications,” but cautions that “in some applications that rely on scandium's unique properties, substitution is not possible” (USGS MCS 2026) — a category that includes SOFC electrolyte doping (Section 7), where no other dopant matches scandia-stabilized zirconia's conductivity-to-cost profile at commercial scale, per Bloom Energy's own disclosure that it has developed alternative materials but “scandium oxide remains the best choice for the current fuel cell platform” (Bloom Energy, 8 July 2026).

Australia's Two Primary Scandium Hopefuls: Nyngan and Clean TeQ Sunrise

1. Nyngan Scandium Project: the world's first scandium-only mine design

Scandium International Mining Corp. (TSX: SCY) has spent over a decade advancing the Nyngan Scandium Project, located approximately 500 kilometres northwest of Sydney, New South Wales, as “the world's first scandium-only mine development project” (Scandium International Mining Corp., Nyngan project page). The company's Definitive Feasibility Study, completed in May 2016, projects average production of 37,690 kilograms (37.7 tonnes) of scandium oxide per year at grades of 98.0–99.9% purity, generating after-tax cumulative cash flow of US$629 million over a 20-year project life, with an NPV at a 10% discount rate of US$177 million (Scandium International Mining Corp., Nyngan detailed review). A revised NI 43-101 resource estimate lists 16.9 million tonnes of measured and indicated resource at an average grade of 235 parts per million scandium, with a higher-grade mineral reserve of 1.43 million tonnes at 409 ppm scandium supporting the feasibility-study production plan (Scandium International Mining Corp., investor presentation, 19 March 2026).

2. The nine-year mining-license saga and its October 2025 resolution

Nyngan's regulatory path illustrates how difficult even a well-studied scandium project has been to permit. The original Mining Lease (ML 531) application was filed in 2017 following a 2016 NSW Development Consent, but the project's mining-lease status lapsed and required renewal; a replacement ML 1792 was granted in 2019 covering only 370 hectares of company-owned surface land, insufficient to cover the full deposit (Scandium International Mining Corp., MD&A, 31 March 2025). On 7 October 2025, the NSW Department of Primary Industries and Regional Development issued a new Mining Lease, ML 1893, covering the entire deposit — “concluding an application review process that lasted over nine years” — granted for an initial 21-year term through October 2046 (Scandium International Mining Corp., press release, 7 October 2025). USGS separately confirms: “In October, the Australian Government granted a mining license to a company for its Nyngan scandium project in New South Wales, Australia” (USGS MCS 2026). Company materials now describe Nyngan as “fully shovel-ready,” with the mining lease, feasibility study, and infrastructure access (water, power, site access) all secured, alongside a combined potential capacity across development phases exceeding 100,000 kilograms per year of scandium oxide and a mine life exceeding 40 years (Scandium International Mining Corp., investor presentation, 19 March 2026). The project still lacks a public Final Investment Decision or disclosed full construction financing as of mid-2026.

3. Clean TeQ Sunrise: scandium as a byproduct of one of the world's largest nickel-cobalt laterite deposits

The second major Australian scandium project, Clean TeQ Sunrise (formerly Clean TeQ Holdings, now advanced under Sunrise Energy Metals), is located near Fifield, roughly 350 kilometres west of Sydney, and is described by the company as “one of the largest cobalt deposits outside of Africa, and one of the largest and highest-grade accumulations of scandium ever discovered” (Clean TeQ, GlobeNewswire, 27 August 2020). The Sunrise deposit is fully permitted, with a Development Consent originally granted under the NSW Environmental Planning and Assessment Act in 2001 and Mining Lease ML 1770, covering 2,195 hectares, held by wholly owned subsidiary Scandium21 Pty Ltd (Clean TeQ Sunrise Definitive Feasibility Study, 2018). The project uses Clean TeQ's proprietary Clean-iX® resin-in-pulp continuous ion-exchange technology to recover nickel, cobalt, and scandium together from a single laterite orebody, with engineering literature citing a hydrometallurgical plant design targeting roughly 28 tonnes of Sc2O3 per year from the Sunrise resource (RWTH Aachen, Advances on Scandium Recovery Beyond State of the Art, 2018). In January 2026, Clean TeQ Water secured an engineering contract with Sunrise Energy Metals specifically to advance the Syerston (Sunrise) scandium project further, indicating continued technical progress toward development (Clean TeQ Water, 27 January 2026).

Why both projects matter beyond Australia: Nyngan and Clean TeQ Sunrise are commonly cited together as the only two primary (non-byproduct-of-an-unrelated-metal) scandium deposits anywhere in the Western world at an advanced permitting stage, making them structurally different from Rio Tinto's Sorel-Tracy (a byproduct of titanium slag) and NioCorp's Elk Creek (a polymetallic niobium-titanium-rare-earth deposit where scandium is one of several co-products).

Forward Look 2026–2030: Can Supply Keep Pace With Aerospace, Defense, and SOFC Demand?

1. The capacity pipeline through 2030

Summing publicly disclosed targets: Rio Tinto's Sorel-Tracy expansion to 9 tonnes/year (Section 2), NioCorp's Elk Creek target of approximately 100 tonnes/year (Section 4), Nyngan's phased target exceeding 100 tonnes/year combined capacity (Section 9), and Clean TeQ Sunrise's roughly 28 tonnes/year design (Section 9) together imply a theoretical non-Chinese capacity approaching or exceeding 200 tonnes per year if every project reaches full announced scale — more than double USGS's estimated 2025 global production of about 80 tonnes. However, none of these projects has reached that scale today: Sorel-Tracy is mid-expansion, Elk Creek and ElementUSA remain in early DPA-funded development, and both Australian projects are permitted but pre-construction (USGS MCS 2026). Independent market analysis projects non-Chinese production is “not expected at scale until 2027 to 2029” (Lanthanides.io Strategic Materials Ledger, 27 March 2026).

2. Demand-side growth scenarios

Peer-reviewed demand modeling published in 2023 projected scandium oxide consumption reaching approximately 38 tonnes annually by 2030 under a business-as-usual scenario, driven overwhelmingly by SOFCs (about 36 of the 38 tonnes), with an alternative “additional 10% uptake” scenario — incorporating lightweighting in internal combustion engines and electric vehicles — projecting total consumption rising to a much higher figure, representing a 92–160% increase over the base case (Journal of Cleaner Production, exploring global supply and demand of scandium oxide in 2030). These academic projections, published before the 2025 surge in AI-driven power demand and the Pentagon's aerospace-alloy programs, likely understate current demand trajectories: Bloom Energy alone now states its supply chain is designed to support up to 25 GW/year of fuel-cell capacity (Section 7), and NioCorp's binding offtake book already includes a 12-tonne-per-year scandium contract (Section 4) — figures that, if realized, would materially exceed the 2023 academic base-case demand estimate for the entire market.

3. Substitution limits and the ceiling on demand destruction

USGS's substitutes assessment for scandium is unusually narrow among critical minerals: titanium and aluminum high-strength alloys, and carbon-fiber materials, may substitute in some high-performance scandium-alloy applications, but “in some applications that rely on scandium's unique properties, substitution is not possible” (USGS MCS 2026). Bloom Energy's own disclosure reinforces this for SOFCs specifically: the company has developed alternative dopant materials but “scandium oxide remains the best choice for the current fuel cell platform” (Bloom Energy, 8 July 2026). This combination — limited substitution plus rapidly growing defense, aerospace, and clean-power demand — is the structural reason multiple governments moved simultaneously in 2025 to fund alternative supply, rather than waiting for the market to solve scarcity through substitution alone.

4. Key risks through 2030

Three risks dominate the outlook. First, geopolitical: MOFCOM Announcement No. 18/2025 remains in force with no indication of a scandium-specific carve-out, and China's roughly 80% share of global supply (per Sumitomo Metal Mining's own reporting) means licensing friction alone can constrain Western buyers even without an outright ban (Sumitomo Metal Mining production report, 2026). Second, execution: every non-Chinese project profiled in Sections 2, 4, and 9 has taken far longer to reach production than initially planned — Nyngan's mining license alone took over nine years to finalize, and NioCorp's Elk Creek remains reliant on further EXIM Bank debt financing beyond its DPA award (Scandium International Mining Corp., 7 October 2025; Nebraska Public Media, 6 August 2025). Third, price: because the market is thin and unbenchmarked (Section 6), any single new large-scale producer coming online has the potential to move prices sharply in either direction, which in turn affects the economics of every other project trying to reach financing.