Rare Earth Oxides, Metals and Magnets: What Can Actually Be Held When Supply Control Sits in China

The common mistake is to treat “rare earths” as one physical commodity. They are not. A sealed drum of neodymium oxide, a dysprosium-iron alloy, a samarium-cobalt magnet and a laboratory quantity of promethium have entirely different industrial uses, transport profiles, export-control exposure and resale channels. The relevant question is not whether a rare-earth element can exist in physical form, but whether a specific lot is identifiable, properly specified, lawfully movable and acceptable to an industrial counterparty.

Rare earth metals comprise 17 elements: scandium, yttrium and the 15 lanthanides from lanthanum to lutetium. In commercial practice, they are often grouped as light rare earths and heavy rare earths, although scandium and yttrium are frequently treated as separate categories while remaining on rare-earth trade and policy lists. China is the central control point not simply because of mining output, but because separation, refining, metal-making and permanent-magnet manufacturing remain concentrated there.

Key Points

  • Rare earth oxides are the most common industrially traded physical form; they are powders with purity, moisture and impurity specifications rather than fungible bullion products.
  • Dysprosium, terbium, samarium, scandium and yttrium require heightened export-control review because China’s April 2025 licensing regime covers those elements and related products.
  • Metals, alloys and permanent magnets can be physically held, but each creates a different custody and resale proposition. A magnet is not equivalent to elemental or oxide inventory.
  • Promethium has no normal private physical market: it is radioactive, extremely scarce and used principally in research and niche applications.
  • A credible holding is evidenced by a lot-specific assay or certificate of analysis, chain-of-custody records, origin and customs documents, storage records and a contractually clear title position.

The Rare Earth Elements List: The Differences That Matter

“Light” and “heavy” are commercially useful shorthand, not a substitute for a specification. Light rare earths such as lanthanum, cerium, praseodymium and neodymium account for large-volume catalyst, polishing and magnet demand. The more supply-sensitive heavy rare earths include dysprosium and terbium, which are used to maintain permanent-magnet performance at elevated temperatures, and samarium, the basis of samarium-cobalt, or SmCo, magnets used in demanding aerospace and defence applications.

For investors and industrial buyers alike, the category label matters less than the exact material form. Neodymium oxide and neodymium metal are not interchangeable claims. Nor is a neodymium-iron-boron, or NdFeB, magnet equivalent to a lot of oxide feedstock. The table below distinguishes broad physical forms from practical marketability. “Physical inventory possible” means that oxide, metal, compound or alloy form exists and can be held in principle. It does not mean there is a deep private resale market, an exchange warrant system or a standardised global assay convention.

Element Category Main industrial uses Supply, control point and Chinese licensing status Physical holder status
Scandium (Sc) Separate / heavy Aluminium-scandium alloys, fuel cells, lighting Thin, concentrated market; licensed under the active April 2025 controls Oxide, metal and alloy exist
Yttrium (Y) Separate / heavy Phosphors, ceramics, lasers Concentrated separation chain; licensed under the active April 2025 controls Oxide and metal exist
Lanthanum (La) Light Catalysts, battery alloys, optical glass China-dominant midstream chain; not in the April 2025 baseline list cited here Oxide and metal exist
Cerium (Ce) Light Polishing, catalysts, metallurgy China-dominant midstream chain; not in the April 2025 baseline list cited here Oxide and metal exist
Praseodymium (Pr) Light NdPr magnets, alloys, glass China-centric magnet-feedstock separation; not in the April 2025 baseline list cited here Oxide and metal exist
Neodymium (Nd) Light Permanent magnets, motors, turbines China-centric magnet-feedstock separation; not in the April 2025 baseline list cited here Oxide and metal exist
Promethium (Pm) Light / radioactive Research and niche instruments No meaningful commercial bulk market No normal private market
Samarium (Sm) Heavy SmCo magnets, control applications, alloys Strategic magnet input; licensed under the active April 2025 controls Oxide, metal and alloy exist
Europium (Eu) Heavy Phosphors, anti-counterfeit markers Specialty concentrated chain; added in the October 2025 package, which is suspended through 10 November 2026 Oxide and compounds exist
Gadolinium (Gd) Heavy MRI, magnetic materials, shielding Specialty materials chain; licensed under the active April 2025 controls Oxide and metal exist
Terbium (Tb) Heavy Magnet coercivity, phosphors Supply-sensitive magnet input; licensed under the active April 2025 controls Oxide and metal exist
Dysprosium (Dy) Heavy High-temperature magnets, motors Supply-sensitive magnet input; licensed under the active April 2025 controls Oxide and metal exist
Holmium (Ho) Heavy Lasers, alloys, magnetic materials Specialty thin market; added in the October 2025 package, which is suspended through 10 November 2026 Oxide and metal exist
Erbium (Er) Heavy Fibre-optic amplifiers, lasers Specialty optics chain; added in the October 2025 package, which is suspended through 10 November 2026 Oxide and metal exist
Thulium (Tm) Heavy Lasers and X-ray sources Specialty thin market; added in the October 2025 package, which is suspended through 10 November 2026 Oxide and metal exist
Ytterbium (Yb) Heavy Lasers, alloys, optics Specialty thin market; added in the October 2025 package, which is suspended through 10 November 2026 Oxide and metal exist
Lutetium (Lu) Heavy PET detectors, catalysts, optics Concentrated niche supply; licensed under the active April 2025 controls Oxide and metal exist

Where Supply Control Is Exercised

Mine origin alone does not determine deliverability. Rare-earth ore and mixed concentrates must be separated into individual oxides or compounds; those products may then be converted into metals, alloys or magnets. Each stage can create a different processor, jurisdiction, customs classification and export-control issue. For physical holders, the separation and refining history can be more consequential than the location of the mine.

China’s 4 April 2025 export-control package established a licensing regime for samarium, gadolinium, terbium, dysprosium, lutetium, scandium and yttrium. The controlled scope extends beyond elemental material to associated metals, oxides, alloys, compounds and downstream products, including samarium-cobalt magnets and neodymium-iron-boron magnets containing terbium or dysprosium. The practical distinction is important: a lot may be outside China physically yet still require careful legal analysis before export or re-export if its documentation, origin or product form connects it to the controlled chain.

A subsequent package announced in October 2025 added holmium, erbium, thulium, europium and ytterbium, together with broader measures that were later suspended. That suspension runs through 10 November 2026 and applies to the October 2025 expansion, not to the original April 2025 baseline. As of September 2026, the April 2025 regime remains the active core control cited in trade reporting, while the October additions remain paused.

Trade has not stopped. China exported 10,468.3 metric tons of rare earths in January and February 2026, and Reuters later reported that China was working to streamline rare-earth export licences in December 2025. The operative condition is regulated trade through licensing and end-use scrutiny, not unrestricted movement.

Oxide, Metal, Alloy or Magnet: Four Different Physical Claims

Oxides are generally the clearest physical form for industrial rare-earth inventory. Neodymium oxide, praseodymium oxide, dysprosium oxide, terbium oxide, samarium oxide, gadolinium oxide, scandium oxide and yttrium oxide are recognisable industrial feedstocks. Their condition is not incidental. Moisture content, particle size, purity expressed as rare-earth oxide content, and specified impurities can determine whether a processor will accept the lot.

Metals exist for numerous elements, including neodymium, praseodymium, samarium, dysprosium, terbium, gadolinium, yttrium, scandium and lutetium. But an ingot or metal piece is not automatically easier to sell than oxide. Metal grade, oxidation condition, packaging atmosphere, dimensions and the buyer’s conversion process matter. A metal certificate describing only an element name and nominal weight is inadequate where the intended industrial buyer requires a defined chemical specification.

Alloys are closer to downstream production. NdPr alloy, SmCo alloy and dysprosium-iron or terbium-iron alloys can be physically held, but their value depends on composition and suitability for a named industrial process. They require a certificate of analysis that identifies each relevant constituent, not merely a generic reference to “rare-earth alloy.”

Permanent magnets are tangible and may be commercially tradeable, but they are finished or semi-finished components rather than elemental inventory. Their value incorporates geometry, coating, magnetic grade and manufacturing quality. Their regulatory treatment can also be more complex where they contain controlled terbium, dysprosium or samarium inputs. A container of magnets is not a simple substitute for an allocated lot of oxide or metal.

Salts and compounds also exist as physical forms, but the market is usually more specialised. They may be entirely legitimate industrial products while remaining less straightforward for a private holder than a well-specified oxide or metal lot. As ever in rare earth metals, physical existence and practical marketability are not the same thing.

The Documentation That Separates Inventory from an Unsupported Claim

There is no universal bullion-style assay standard for rare earth metals. Serious evaluation therefore begins with the lot, rather than a generic product description. In practice, a serious evidential package includes a supplier-issued certificate of analysis identifying the exact chemical form; net and gross weight; lot or batch identifier; rare-earth oxide purity or metal purity; moisture content where relevant; and the impurity schedule used in the sale specification.

Custody evidence matters equally. Sealed bags, lined drums, tamper-evident labels, warehouse intake records, inventory reconciliations and photographs can establish that the tested material is the material in storage. Where title is held through a custodian or warehouse operator, the storage agreement needs to identify whether the lot is allocated to a named owner or merely represented by an undivided contractual claim. Rare-earth powders are not interchangeable merely because they share an element name.

  • Origin and trade record: commercial invoice, packing list, customs declaration, commodity classification and any export licence or licence determination.
  • Quality record: certificate of analysis, sampling method, laboratory identity, specification sheet and material safety documentation.
  • Custody record: warehouse receipt, lot location, seal numbers, inventory statement, insurance confirmation and chain-of-custody transfers.
  • Resale record: evidence that the stated form, purity and packaging correspond to a recognisable industrial off-take channel, rather than an unsupported claim of “liquidity.”

Failure Modes Specific to Rare-Earth Holdings

The first failure mode is a mismatch between assay and deliverable form. A document may state “99.5% neodymium oxide” without identifying moisture, other rare-earth impurities or the specific lot. That may not satisfy a processor whose production line requires a narrower impurity profile. The second is contamination or packaging degradation: oxide powders can absorb moisture, while some metals require protective packaging to preserve condition.

The third is custody ambiguity. A stock statement showing aggregate kilograms at a warehouse does not establish segregated title to a numbered drum or sealed container. The fourth is export-control ambiguity. A material may be physically stored in a third country while its proposed onward movement implicates Chinese-origin controlled inputs, regulated alloy forms or controlled magnet products. The fifth is thin-market substitution: a niche oxide such as holmium, thulium, ytterbium or lutetium may be physically real but lack a broad and continuous private resale channel.

A sixth failure mode is category confusion. “Rare earths” sounds like a single asset class, but in practice it covers catalyst materials, phosphor inputs, optical materials, alloy additions and magnet feedstocks with very different buyers. The relevant off-take universe for cerium oxide is not the same as the off-take universe for scandium metal or lutetium oxide. That difference matters when an owner eventually needs to move material, not merely hold it.

What a Defensible Rare-Earth Holding Looks Like

A defensible claim to physical rare earth inventory identifies an element and chemical form precisely, links that material to a numbered and tested lot, records where and how it is stored, establishes allocated title, and preserves the trade and origin documentation needed for lawful movement. It also distinguishes a liquid industrial feedstock such as a specified oxide from a niche compound, a fabricated magnet or a non-commercial material such as promethium.

In other words, rare earths can be held physically, but only in the specific sense that identifiable industrial lots can be held physically. For private holders, family offices and wealth managers, the practical questions are specification, custody, lawful movability and industrial acceptability. Those attributes are not administrative detail. In rare earth metals, they are part of the commodity itself.

Sources

Conclusion

Rare earths should be analysed as specific lots, forms and process stages rather than as a single physical commodity. The core market implications are straightforward: policy risk sits most clearly in the controlled heavy elements and certain magnet products, while commercial risk sits in documentation, specification and end-user acceptance. For any physical holder, what matters is not only what the material is, but whether it remains verifiable, movable and saleable when tested by the market.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *