Europe Wants a Critical Minerals Stockpile. What Would It Actually Need to Buy?

Key findings

• A useful European reserve has to be defined by processing stage, purity, and chemical form, not only by mineral name.
• Rare earth exposure becomes more concentrated at separation, processing, and permanent magnets than headline mine import shares suggest.
• Gallium, germanium, and tungsten show how small material volumes can sit beneath much larger industrial systems.
• European conversion assets can matter strategically because stored material only helps if it can be turned into something the next factory can use.

Europe's useful critical-minerals stockpile would not look like equal piles of all 34 materials on the EU's critical list. A reserve designed to cover the supply disruptions Europe has identified would have to mirror the points where European industry has the fewest alternatives. That means paying attention not only to the mineral, but also to its purity, chemical form, processing stage, and the European companies capable of converting it into a usable component.

That distinction became more important on September 16, 2026, when European Commission President Ursula von der Leyen announced plans for a new European Corporation on Critical Raw Materials. She said the EU remains more than 80% dependent on China for many critical materials and about 90% dependent for some rare earths, and said the new corporation would help Europe obtain and stockpile materials needed for electric vehicles, chips, batteries, clean technology and defence.[1][2] The initiative extends the Commission's RESourceEU plan, which already called for joint purchasing, coordinated stockpiling and roughly EUR 3 billion of near-term support for projects intended to diversify raw-material supply.[3]

The investment implication is that the most important nodes are not necessarily the largest mining companies. In several of Europe's most exposed supply chains, the bottleneck sits further downstream, at separation, refining, powder production or component manufacturing. Solvay's rare earth separation plant in France, Neo Performance Materials' magnet operations in Estonia, Umicore's germanium recycling business in Belgium, Sandvik's Austrian tungsten chain and Infineon's gallium nitride manufacturing in Austria illustrate the point. A strategic inventory has much more industrial value when there is a domestic or allied processing route capable of consuming it.

The useful unit of analysis is the material at the processing stage immediately before a European converter or manufacturer would be forced to stop production.

More than 80%

EU dependence on China for many critical materials

European Commission statement, September 2026

About 90%

EU dependence on China for some rare earths

European Commission statement, September 2026

EUR 3 billion

Near-term RESourceEU support for diversification projects

European Commission

More than 90%

EU rare earth magnets sourced from China

European Commission

77%

China share of EU gallium imports by value in 2025

Eurostat

92%

China share of EU magnesium imports in 2025

Eurostat

The stockpile has to be defined by processing stage

The EU's existing Critical Raw Materials Act already contains an important clue about how a serious stockpile would have to work. Member states report not only how many tonnes of strategic materials they hold, but also the chemical form and purity. The Commission's benchmark for a safe stock level is supposed to be expressed as the number of days of average daily net imports that could be covered during a disruption.[4]

That is more sophisticated than simply deciding that Europe needs a certain number of tonnes of lithium, rare earths or tungsten. A tonne of mixed rare earth concentrate cannot automatically replace a tonne of separated neodymium oxide. Lithium ore is not interchangeable with battery-grade lithium hydroxide. Tungsten concentrate, ammonium paratungstate, ferro-tungsten and tungsten carbide powder serve different points in the manufacturing chain.

The trade statistics make this distinction unusually visible. In 2025, China represented 46.8% of EU rare earth imports by physical weight, according to Eurostat.[5] Look only at that number and Europe's dependence can appear substantial but manageable. Look one step further down the chain and the picture changes sharply. The Commission says more than 90% of the EU's rare earth magnets come from China, while its RESourceEU analysis estimated 2025 dependence on a single third country at 95% for rare earth extraction, 100% for rare earth processing and recycling, and 90% for permanent magnets.[3][6]

Extraction

Single-country dependency estimate

95%

Processing and recycling

Single-country dependency estimate

100%

Permanent magnets

Single-country dependency estimate

90%

The figures describe different stages of the rare earth chain. They should not be read as interchangeable measures of the same product market.

Source: Altsets presentation of European Commission RESourceEU estimates cited in this article

Copy link

Lithium produces the same problem in another form. Eurostat found that Chile supplied 70% of EU lithium carbonate imports by value in 2025.[7] The Commission's broader RESourceEU analysis placed Europe's single-country lithium dependency at 89% in 2025.[6] Those figures are measuring different parts and definitions of the chain. That is precisely why "stockpiling lithium" is not a complete policy or investment concept. The relevant question is which lithium product a European cathode or cell plant would actually run out of first.

For investors, this means headline mining concentration can understate downstream exposure. It also means a new mine does not necessarily remove a bottleneck. Mining capacity can expand while separation, purification or component production remains concentrated somewhere else.

Rare earths are the clearest example of a component-level bottleneck

Rare earths combine several features that make the European dependency unusually difficult to manage: concentrated processing, concentrated magnet manufacturing, applications with limited short-term substitution, and relatively small quantities of material embedded in much higher-value products. China's 2025 export restrictions demonstrated how quickly that structure can reach factory floors. European automotive supplier plants halted production during the permitting delays, while Mercedes-Benz and other manufacturers worked to protect inventories and investigate alternative supplies.[8]

The important corporate network therefore begins before the automaker.

At La Rochelle in France, Solvay began commercial production in 2025 of rare earth materials intended for permanent magnets. The company describes the site as the largest plant outside China capable of separating all rare earth materials. Its new line targets magnet-grade materials used in electric vehicle motors, wind turbines, advanced electronics and defence systems.[9]

The next conversion stage is also beginning to exist inside the EU. Neo Performance Materials announced on September 14, 2026 that its Narva, Estonia permanent magnet facility had entered commercial production and was shipping sintered rare earth magnets to a Tier 1 electric vehicle traction-motor customer.[10] Neo also has a multi-year agreement with Bosch under which production capacity at the European plant is reserved for Bosch's mobility and energy-efficient motor applications.[11]

That creates a developing European chain from separated rare earth material to magnets to motors. It is still small relative to the established Asian ecosystem, but it changes what inventory can accomplish. Magnet-grade neodymium and praseodymium compounds, and the heavy rare earths dysprosium and terbium used in high-performance magnets, can now potentially feed European conversion capacity rather than merely sit in storage.

1. 01
   Stockpiled feedstock
   Material in a form a European processor can accept
2. 02
   Separation
   Example capability: Solvay, France
3. 03
   Permanent magnets
   Example capability: Neo Performance Materials, Estonia
4. 04
   Motors and turbines
   Downstream applications include mobility and wind power

This is a stage diagram, not a disclosed Solvay-to-Neo-to-Bosch commercial chain. The companies are examples of capabilities described in the article.

Source: Altsets analysis of company disclosures cited in this article

Copy link

The downstream exposure extends well beyond cars. Siemens Gamesa, part of Siemens Energy, has been diversifying permanent-magnet procurement for wind turbines. In 2025 it disclosed an agreement with Japan's TDK and said it was discussing European manufacturing with Chinese magnet suppliers as another way to reduce supply risk.[12] Siemens Energy subsequently reported that it had increased resilience in permanent magnets through diversified supply chains outside China and work to reduce rare earth intensity.

This is why the rare earth problem cannot be reduced to tonnes of mined material. The interruption that matters to a Bosch motor line or a Siemens Gamesa turbine program can occur between oxide, alloy, magnet and finished motor. A reserve of upstream concentrate provides limited protection if the separation or magnet-making step remains constrained.

There is also an inventory-design problem. Permanent magnets are manufactured to detailed specifications, while the Commission has already noted that changing material specifications can make long-duration corporate inventories difficult to manage.[3] That favors a combination of standardized upstream material that can be rotated through European processors and, where specifications permit, inventories closer to the component stage.

Gallium, germanium and tungsten turn small markets into large industrial risks

Some of Europe's most consequential stockpile candidates are not large commodity markets at all. Gallium and germanium matter precisely because tiny material volumes sit beneath much larger semiconductor, communications and defence markets.

Eurostat found that China supplied 77% of EU gallium imports by value in 2025. Imports had fallen sharply from their 2022 level following Chinese export restrictions.[7] The International Energy Agency estimates that China accounts for more than 90% of global refining for gallium and identifies gallium among the minerals with the highest supply-risk exposure because of concentration, strategic uses and limited alternatives.[13]

Europe is simultaneously building more technology that consumes gallium. Infineon is ramping 300 millimeter gallium nitride semiconductor technology at its Villach plant in Austria. The company is positioning GaN power devices for markets including AI data centres, renewable energy, mobility and industrial electronics.[14] That creates an important asymmetry. Europe can own advanced semiconductor manufacturing capability while remaining exposed to a much earlier material-processing bottleneck.

For that network, generic gallium inventory is less informative than access to the purity and chemical form required by compound-semiconductor production. The financial attraction of stockpiling a material such as gallium is also different from copper or lithium. A comparatively small physical reserve can represent substantial disruption coverage for a high-value manufacturing chain.

Germanium has a similar shape. The Commission estimated a 45% single-country dependency in 2025 but believes projects already identified under RESourceEU could eliminate that particular single-country dependency by 2030.[6] Whether those projects arrive on schedule matters because germanium feeds fibre optics, infrared optics, electronics and other strategic applications.

Umicore provides an example of what diversification can look like without opening a new mine. The Belgian group says more than half of the feed entering its germanium business comes from recycling. It refines germanium-bearing material into high-purity germanium compounds, metal and crystals, and two of its Belgian germanium projects were selected as strategic projects under the Critical Raw Materials Act.[15] In this case, recyclable material and production scrap are effectively part of the supply-security network. Material already inside Europe can be repeatedly converted back into usable feedstock.

Tungsten adds another layer. China accounted for 68% of EU ferro-tungsten imports by value in 2025.[7] Yet Europe already contains an important integrated tungsten operation. Wolfram Bergbau und Hütten, part of Sandvik, mines scheelite at Mittersill in Austria and processes its own, imported and recycled material at St. Martin im Sulmtal into ammonium paratungstate, tungsten oxides, tungsten metal and tungsten carbide powders.[16]

Those powders feed applications across tooling, mining, automotive, aerospace, electronics and other industrial markets. The presence of an operating European converter again changes the relevant question. Stocking only ferro-tungsten would not necessarily address a shortage facing a carbide-powder customer. Feed material that can move through an existing European refinery has a different strategic value from an inventory that cannot be readily converted.

Gallium, germanium and tungsten therefore reveal a broader rule in the dependency network: physical scarcity and economic importance are not proportional to tonnage. An industrial system can be highly exposed to a material that represents an almost invisible share of the finished product's cost.

Company	Location	Relevant capability
Solvay	France	Rare earth separation and magnet-grade materials
Neo Performance Materials	Estonia	Sintered rare earth permanent magnets
Umicore	Belgium	Germanium recycling and high-purity refining
Wolfram Bergbau und Hütten	Austria	Tungsten intermediates, metal and carbide powders
Infineon Technologies	Austria	300 millimeter gallium nitride semiconductor manufacturing

The companies shown are examples, not a complete European critical-minerals supply chain.

Source: Altsets synthesis of company disclosures cited in this article

Copy link

Graphite and magnesium are larger inventory problems

Battery-grade graphite belongs in the same discussion, but its scale and processing chain are different. China represented 43% of EU natural graphite imports by value in 2025, with Madagascar and Brazil providing another 18% and 13% respectively.[7] The import number alone again understates the processing issue. The IEA estimates that China's share of global graphite refining exceeds 90%.[13]

A battery plant does not simply require mined graphite. It requires processed anode material meeting tight particle-size, purity, coating and performance specifications. As Europe tries to localize battery manufacturing, an inventory based on unprocessed natural graphite could leave the downstream processing bottleneck untouched.

The Commission's own RESourceEU estimates reinforce that distinction. It assessed Europe's 2025 single-country dependency at 41% for graphite, 63% for cobalt, 89% for lithium, 41% for manganese and 29% for nickel. Projects already in the pipeline could reduce all of those figures by 2030, but not eliminate them.[6]

Those battery materials should not be treated as a single stockpiling category. Lithium carbonate imported by the EU is already sourced primarily from Chile rather than China. Nickel supply is less concentrated at the EU level than gallium or magnesium. Cobalt has significant geographic concentration but also growing recycling potential. Graphite stands out because the processing stage remains exceptionally concentrated even when mine supply appears more diverse.

Magnesium is different again. It is a broad industrial input rather than a specialized battery material, but its European concentration is extreme. China supplied 92% of EU magnesium imports in 2025.[7] Magnesium is used across automotive, aerospace, electronics and metallurgical applications, so a disruption can propagate through many otherwise unrelated manufacturers.

Its economics are less convenient than those of gallium or germanium. A meaningful magnesium reserve requires much more physical material and storage capacity. That does not make the exposure less real. It means the capital and logistics required to cover a given number of import days are materially different.

Copper illustrates the opposite end of the spectrum. It is strategically important and Europe is a large net importer, but its import origins are considerably more diversified. In 2025 no single country supplied more than 25% of EU copper imports by value. Chile supplied 23%, the Democratic Republic of Congo 17% and Brazil 15%.[7] EU imports also run into millions of tonnes annually.

That makes copper a different resilience problem from gallium or rare earth magnets. A disruption to copper supply would be economically significant, but the market has multiple major origins and a stockpile large enough to cover substantial consumption would require enormous working capital and storage. The same "critical material" label therefore captures very different types of risk.

There are also concentrated dependencies outside China. Brazil supplied 83% of EU ferro-niobium imports in 2025, Turkey supplied 73% of borates, Chile supplied 70% of lithium carbonates and Tajikistan supplied 52% of the antimony products covered by Eurostat's analysis.[7] A concentration statistic is not a geopolitical risk score. These relationships have different political, commercial and logistical characteristics. They do show that European supply-chain resilience is broader than simply reducing Chinese import share.

Magnesium from China

92%

Ferro-niobium from Brazil

83%

Gallium from China

77%

Borates from Turkey

73%

Lithium carbonates from Chile

70%

Antimony products from Tajikistan

52%

These are product-specific import shares, not a common geopolitical risk score. Product definitions and processing stages differ.

Source: Altsets presentation of Eurostat data cited in this article

Copy link

The investable bottleneck may be the converter, not the mine

For investors, the new stockpiling initiative matters because it could affect companies in several different ways without changing the underlying geology.

The first is utilization. A European processor with qualified capacity can become more valuable to the supply chain when buyers place a premium on material that can be processed outside the dominant country. Solvay in rare earth separation, Neo in magnets, Umicore in germanium recycling and Sandvik's Wolfram business in tungsten are examples of assets located between mine production and the final manufacturer. Their strategic relevance comes partly from what they can convert, not simply what they own underground.

The second is working capital. The Commission acknowledged in RESourceEU that many companies purchase critical materials only as needed because stockpiling ties up financing, requires storage and can create specification risk.[3] A public or coordinated inventory mechanism can move part of that burden away from individual manufacturers. But it can also shift demand forward, particularly in small markets where government purchases are large relative to normal commercial volumes.

The third is project finance. The EU has already selected 60 strategic critical-mineral projects from its first round, including 47 inside the bloc and 13 outside it. By September 2026, however, developers from 23 selected projects had publicly warned that financing, market access and slow implementation threatened their ability to proceed. Reuters reported that some projects had already been suspended and that French lithium developer Viridian Lithium had collapsed.[17]

That makes procurement relevant even before a stockpile is released. A credible buyer or long-term offtake contract can help a processing project demonstrate future demand. Conversely, announcing a reserve without specifying purchasing volumes, material specifications or release rules does little by itself to make a new refinery financeable.

The fourth is substitution. Stockpiling protects the existing production system, while substitution changes it. Siemens Gamesa has worked on reducing heavy rare earth content in magnets. Automakers and suppliers have explored motor designs that require fewer or no rare earth permanent magnets. Recycling at Umicore and Wolfram reduces the amount of primary material that must cross Europe's border. Each of these mechanisms changes future stockpile requirements.

That is why the European Corporation announced in September 2026 should be viewed by investors as one layer of a larger industrial network, not a substitute for mining, processing, recycling or redesign. A warehouse can bridge a disruption. It cannot manufacture the missing intermediate product.

Conclusion

Europe's critical-minerals exposure is not one problem and therefore cannot be measured by one import-dependence number.

Rare earths are exposed most severely at separation and permanent magnets. Gallium and germanium are small-volume, high-purity bottlenecks embedded in much larger semiconductor, communications and defence systems. Tungsten connects mining, refining, powders and industrial tooling. Graphite demonstrates how apparently diversified mine supply can still feed into highly concentrated processing. Magnesium combines extreme supplier concentration with much larger physical consumption. Copper is critical but has a broader supplier base and an entirely different storage problem.

Europe now has pieces of an alternative conversion network: Solvay can separate magnet-grade rare earths in France, Neo is commercially producing magnets in Estonia, Umicore can recover and refine germanium in Belgium, Wolfram can turn domestic and imported tungsten feed into advanced powders in Austria, and Infineon is expanding gallium nitride semiconductor manufacturing in Villach. The effectiveness of any future stockpile will depend on whether its inventories can actually flow through assets like these during a disruption.

That is also where the investment signal is most interesting. The scarcity value in critical minerals does not necessarily sit with whoever owns the largest deposit. In a fragmented supply chain, it can sit with the company controlling the conversion step that turns an available raw material into something the next factory can use.

Methodology and limitations

This analysis combines EU legislation and policy documents, Eurostat trade data, IEA supply-concentration research, recent reporting and company disclosures. Import shares are not treated as complete measures of supply-chain dependence. Trade statistics can cover different product codes and processing stages, which is why figures for a material may differ between Eurostat and broader European Commission dependency estimates.

Company examples identify publicly disclosed positions in relevant value chains. They do not imply that the companies named represent the complete European supply chain, that a particular upstream material is technically interchangeable with every company's current feedstock, or that a supply disruption would produce a specific financial or stock-price response.

No Altsets relationship percentages were used because the supplied examples do not directly quantify the critical-mineral relationships examined here.

Sources

1. 2026 State of the Union Address by President von der Leyen, European External Action Service, September 16, 2026, https://www.eeas.europa.eu/delegations/moldova/2026-state-union-address-president-von-der-leyen_en

2. EU will use all tools to cut China trade deficit, von der Leyen says, Reuters, September 16, 2026, https://www.reuters.com/world/china/eu-will-use-all-tools-cut-china-trade-deficit-von-der-leyen-says-2026-09-16/

3. New measures to secure raw materials and strengthen the EU's economic security, European Commission, December 3, 2025, https://commission.europa.eu/news-and-media/news/new-measures-secure-raw-materials-and-strengthen-eus-economic-security-2025-12-03_en

4. Regulation (EU) 2024/1252 establishing a framework for ensuring a secure and sustainable supply of critical raw materials, EUR-Lex, April 11, 2024, https://eur-lex.europa.eu/eli/reg/2024/1252/oj

5. EU trade in rare earth elements increased in 2025, Eurostat, June 29, 2026, https://ec.europa.eu/eurostat/en/web/products-eurostat-news/w/ddn-20260629-1

6. RESourceEU Action Plan factsheet, European Commission, December 2025, https://single-market-economy.ec.europa.eu/document/download/e9ac2181-0dc7-4e61-a964-ba0a39c2aea8_en

7. International trade in critical raw materials, Eurostat Statistics Explained, data extracted June 2026 and published July 2026, https://ec.europa.eu/eurostat/statistics-explained/SEPDF/cache/118506.pdf

8. China's rare earth export curbs hit the auto industry worldwide, Reuters, June 4, 2025, https://www.reuters.com/business/autos-transportation/some-european-auto-supplier-plants-shut-down-after-chinas-rare-earth-curbs-2025-06-04/

9. Solvay advances European rare earths production through capacity expansion, Solvay, April 8, 2025, https://www.solvay.com/en/press-release/solvay-advances-european-rare-earths-production-through-capacity-expansion

10. Neo Performance Materials Advances to Commercial Production at its European Magnet Facility, Neo Performance Materials, September 14, 2026, https://www.neomaterials.com/neo-performance-materials-advances-to-commercial-production-at-its-european-magnet-facility/

11. Neo Performance Materials Reports Third Quarter 2025 Results, Neo Performance Materials, 2025, https://www.neomaterials.com/neo-performance-materials-reports-third-quarter-2025-results/

12. Siemens Gamesa, Chinese magnet suppliers discuss European production, COO says, Reuters, June 26, 2025, https://www.reuters.com/sustainability/climate-energy/siemens-gamesa-chinese-magnet-suppliers-discuss-european-production-coo-says-2025-06-26/

13. Global Critical Minerals Outlook 2026: Executive summary, International Energy Agency, 2026, https://www.iea.org/reports/global-critical-minerals-outlook-2026/executive-summary

14. Infineon advances on 300-millimeter GaN manufacturing roadmap, Infineon Technologies, July 2, 2025, https://www.infineon.com/regional/austria/press-release/2025/infineo-advances-on-300mm-gan-manufacturing-roadmap

15. EU's selection of Umicore's germanium projects underline leadership in germanium recycling, Umicore, March 26, 2025, https://www.umicore.com/en/media/newsroom/eu-selection-of-umicore-germanium-projects/

16. Tungsten products plant in St. Martin, Austria, Wolfram Bergbau und Hütten, https://www.wolfram.at/en/about-us/sustainability/st.-martin-im-sulmtal

17. Some critical mineral firms chosen by EU flag project liquidity concerns, Reuters, September 8, 2026, https://www.reuters.com/world/china/some-critical-mineral-firms-chosen-by-eu-flag-project-liquidity-concerns-2026-09-08/

How to Cite This

For research inquiries or data access: press@altsets.com

Go Deeper

Altsets Research

Get new Altsets research when we publish.

Original supply-chain research, market casefiles, and new data findings.

Email addressSubscribe

Research updates only. Unsubscribe any time.

Sources

• 2026 State of the Union Address by President von der Leyen
• Reuters: EU will use all tools to cut China trade deficit, von der Leyen says
• European Commission: New measures to secure raw materials and strengthen the EU's economic security
• Regulation (EU) 2024/1252
• Eurostat: EU trade in rare earth elements increased in 2025
• European Commission: RESourceEU Action Plan factsheet
• Eurostat: International trade in critical raw materials
• Reuters: China's rare earth export curbs hit the auto industry worldwide
• Solvay: European rare earths production capacity expansion
• Neo Performance Materials: European magnet facility enters commercial production
• Neo Performance Materials: Third Quarter 2025 Results
• Reuters: Siemens Gamesa and European magnet production
• IEA: Global Critical Minerals Outlook 2026
• Infineon: 300-millimeter GaN manufacturing roadmap
• Umicore: EU selection of germanium projects
• Wolfram Bergbau und Hütten: Tungsten products plant in St. Martin
• Reuters: Critical mineral firms flag project liquidity concerns

Methodology

Read the methodology for this research.

Related research

• What Happens to AI Infrastructure If China's Mineral Truce Ends?

• South Korea Is Looking to Central Asia for Critical Minerals. The Real Opportunity Is Processing.

• The Rare Earth Problem Is Not the Mine. It Is the Magnet

• Gallium: The Semiconductor Material Nobody Talks About Until China Stops Shipping It