What the Fusion Commodity Basket Reveals About Supply Risk

Fusion energy gets discussed as an engineering triumph, a physics milestone, a solution to the climate problem. What almost nobody talks about is the pile of specific metals a fusion plant physically cannot run without.

Consider this: a single company in a single Brazilian town produces enough niobium to supply the entire world’s current demand for the metal, with capacity to spare. Fusion has not even started drawing on that supply yet.

Commercial fusion power depends on five distinct categories of critical minerals, each doing a job that no other material can do. Superconductors confine the plasma. Beryllium multiplies neutrons. Lithium-6 breeds the fuel. Tungsten shields the reactor wall. Specialised steels hold the whole thing together. Supply across this basket is wildly uneven: some materials are everywhere, two are dangerously concentrated, and knowing the difference is what separates a coherent thesis from a bet on a technology timeline.

What follows maps the full fusion commodity basket, identifies where the real supply bottlenecks sit, and gives you a framework for deciding which concentration risks are genuinely investable and which carry more danger than upside.

The five material categories that every fusion reactor must have

The first mistake investors make with fusion materials is treating it as a single-commodity story. It is not a lithium story or a rare-earth story. It is a five-layer supply-chain problem, and every layer has to scale at once.

Think of a fusion reactor as a stack of dependencies. Superconducting magnets, built from niobium-tin (Nb₃Sn) or REBCO (rare-earth barium copper oxide), generate the magnetic fields that hold superheated plasma in place. Without them, the plasma touches the wall and the reaction dies.

Inside the reactor blanket, beryllium acts as a neutron multiplier, amplifying the neutron flux so the plant can breed its own fuel. That fuel comes from lithium-6, which captures neutrons to produce tritium, the hydrogen isotope fusion actually burns.

Facing the plasma directly, tungsten protects the first wall from extreme heat and particle bombardment. And holding everything in place, low-activation structural steels provide the reactor’s skeleton without becoming dangerously radioactive under neutron fire.

The point is that none of these can be traded away. A fusion plant with world-class magnets but no tungsten wall does not run. Neither does one with perfect steels but no lithium-6 to breed fuel. These are co-dependent layers, not competing commodity choices.

Where they differ sharply is in supply concentration. Structural steel feedstock is diffuse and globally available. Niobium, beryllium, and tungsten are the opposite: concentrated in single producers or single countries to a degree that is unusual even by critical-minerals standards. That concentration question is the central investment question this guide addresses.

The Special Competitive Studies Project (SCSP), in its Fusion Supply Chain Report of December 2025, ranks the basket by supply-chain risk. The table below maps each category to its function and that risk rating.

Read this as your map. It stops you over-indexing on whichever fusion material happens to be in the news, and it shows you where to look before you start navigating.

Niobium and the world’s most concentrated mineral supply chain

Here is a supply fact you have probably never seen quoted: one company, CBMM, headquartered in Araxá, Brazil, produces and sells roughly 80-85% of all the niobium on Earth. Not a leading share. Nearly all of it.

The precise figure depends on the source. Sojitz Corporation put it at “roughly 80%” in September 2025, while an earlier unattributed estimate ran to about 85%. Either way, the concentration is extreme. Brazil as a whole accounts for roughly 90% of global niobium supply, with Canada at about 6% and Nigeria under 1%.

Brazil’s niobium reserves are geologically unusual in both scale and grade, concentrated in carbonatite deposits that explain why no credible alternative producer has emerged despite decades of exploration elsewhere in the world.

CBMM’s Araxá operation has capacity of 150,000 tonnes of niobium products per year. In 2023, the entire global market for niobium products was 124,000 tonnes. That single mine can already supply the world with room to spare.

The global supply structure comes down to three producers:

• CBMM (Brazil): the near-monopoly producer, with the Araxá mine and roughly 80-85% of global output
• CMOC: a secondary producer holding part of the remaining supply
• Magris Performance Materials (Canada): the third meaningful producer, behind Brazil’s roughly 90% share

Mining Magazine captured the situation bluntly.

Niobium has “one near-monopoly producer and just three mines in production globally.”

For fusion, niobium matters because niobium-tin (Nb₃Sn) is a primary material in the superconducting magnets that confine plasma. That is a real, non-substitutable demand source. So why is this not a straightforward scarcity trade?

Because CBMM already has more capacity than the market needs. With 150,000 tonnes of capacity against 124,000 tonnes of 2023 demand, and CBMM selling 92,000 tonnes of ferroniobium equivalent that year, the near-term picture is one of headroom, not squeeze. CBMM is also expanding into battery materials, targeting 20,000 tonnes of niobium oxide capacity by 2030, which adds supply-side complexity rather than tightness.

What this tells you is that niobium is not a near-term scarcity story. The investable thesis here is strategic optionality on a producer that becomes indispensable as fusion scales, not a price spike waiting to happen. That distinction matters for how you would size and time any exposure.

Nb₃Sn and REBCO: why the fusion magnet market is not a winner-takes-all bet

There are two superconductor technologies in play, and they do not compete head-to-head in the way you might expect.

Large tokamak projects such as ITER built their toroidal field coils around Nb₃Sn conductors cooled to roughly 4.5 K. Nb₃Sn is mature, cheaper per unit of field, and available from multiple industrial suppliers in large volumes, which suits big, mechanically robust coils where reliability dominates.

REBCO is the higher-performance alternative. Commonwealth Fusion Systems and MIT’s Plasma Science and Fusion Center have demonstrated REBCO magnets exceeding 20 T on coil for the SPARC concept, enabling far more compact machines. The trade-off is cost and manufacturing maturity: REBCO is significantly more expensive and harder to fabricate at scale.

Many designs are therefore hybrid: Nb₃Sn for large, lower-field coils where cost and robustness win, REBCO for the highest-field regions. Niobium demand from fusion is real, but it grows alongside REBCO demand rather than being replaced by it. Both materials sit in the basket together.

What makes beryllium the fusion basket’s most constrained Western supply

Niobium’s concentration sits in one friendly country. Beryllium’s is more precarious: the entire Western primary supply chain runs through a single mine, sitting across a geopolitical fault line from the only other meaningful production base.

Beryllium’s job in fusion is neutron multiplication. Inside the reactor blanket, it amplifies neutron flux so the plant can breed enough tritium to fuel itself. That links beryllium directly to whether a fusion plant can be a net fuel producer, which is the difference between a working power station and an expensive experiment.

Materion operates the only primary beryllium facility outside China, at Spor Mountain, Utah. Its position comes down to three data points:

• Location: Spor Mountain, Utah, the lone Western ore-to-alloy beryllium chain
• Output: 250-270 tonnes of concentrate per year
• Global primary share: approximately 65-70% (Persistence Market Research, August 2026)

The rest of primary capacity is concentrated in China. So a Western fusion supply chain for beryllium is, quite literally, one facility deep.

The market itself is small and tightly held. Persistence Market Research valued global beryllium at US$248.2 million in 2026, projected to reach US$396.0 million by 2033, a 6.9% CAGR. Mordor Intelligence tracks volume at 362.29 tonnes in 2026, rising to 432.58 tonnes by 2031, a slower 3.61% CAGR. (A separate Fact.MR estimate put the 2025 market at US$124.6 million using a different methodology, though that figure is unverified.)

The global beryllium market is characterised as “highly consolidated.”

The gap between a 3.61% volume growth rate and a 6.9% value growth rate is the detail to sit with. It tells you pricing power is already embedded in the market structure. Value is rising faster than volume, which means even modest new demand from fusion could move prices disproportionately, and price is the metric that matters for your exposure. With a single-facility Western chain, beryllium is a chokepoint any fusion materials thesis has to reckon with directly.

For investors wanting to move from the strategic map to specific equity exposure, our dedicated guide to tungsten and beryllium stocks examines which listed producers carry genuine investable concentration risk versus those whose valuations already price in the scarcity narrative.

The rest of the basket: lithium-6, tungsten, and why structural steel is a different problem

You now know the two most investable materials in detail. The map is not complete without the other three, because two of them carry serious strategic risk that is easy to misread as commodity upside.

Here they are, ranked by near-term investability:

1. Tungsten (watchlist): a genuine geopolitical concentration item worth tracking, but a well-understood one
2. Lithium-6 (capability gap): a missing industrial process, not a commodity trade, and a signal of future policy activity
3. Structural steels (engineering problem): a materials-science challenge to be solved in the lab and factory, not a supply squeeze

Tungsten is the plasma-facing material, chosen for its very high melting point, low tritium retention, and low sputtering. The problem is that China supplies roughly 80% of global tungsten, which earned it a Medium-High risk rating from the SCSP for fusion applications. That concentration pattern is familiar from the rare-earth and broader critical-mineral space, which means the playbook for watching it already exists.

Low-activation structural steels are a different animal entirely. The challenge is not that one country controls the feedstock. It is that specialised reduced-activation ferritic-martensitic steels, engineered to avoid producing long-lived radioactive isotopes under neutron bombardment, do not yet exist at industrial scale. That is an engineering and scale-up problem, not a commodity trade.

Lithium-6 enrichment: a capability gap, not a commodity shortfall

Lithium-6 is not the lithium in your EV battery. It is a specific isotope that has to be separated out of natural lithium through enrichment, and it is what captures neutrons inside the reactor to breed tritium fuel.

Fusion lithium demand is not simply additive to the EV battery market; the isotopic specificity of lithium-6 means it draws on a separate enrichment pathway, creating a parallel demand stream that conventional lithium price indices do not capture.

The strategically alarming part, per the SCSP’s December 2025 report, is that the United States currently has no commercial lithium-6 enrichment capacity at all. Global enrichment infrastructure is described as limited and opaque.

This is structurally different from a niobium or beryllium problem. There is no mine to buy into, no producer to back. The bottleneck is a missing industrial process, and no amount of price signal alone conjures an enrichment plant. Filling this gap will take deliberate government or private investment, which is why you should read lithium-6 as a policy signal rather than a near-term commodity opportunity.

Supply concentration as an investment thesis: what commodity history actually teaches

It is tempting to see extreme concentration and conclude “buy.” History says slow down. The energy-transition commodities of the last decade offer a consistent pattern, and it is more instructive than any single fusion forecast.

Look at how three of them played out:

• Lithium: early concentration in South American brines and Australian hard rock delivered strong returns, then rapid capacity build-out and forecast convergence triggered sharp corrections. Concentration bought early optionality, not permanent pricing power.
• Cobalt: concern over Democratic Republic of Congo concentration pushed battery makers toward nickel-rich, cobalt-lean chemistries. High prices and strategic worry accelerated substitution and capped long-run demand.
• Rare earths: China’s 2010 export restrictions spiked prices and triggered a wave of non-Chinese projects, which later normalised prices. Extreme single-country concentration rarely survives once a material becomes politically strategic.

The through-line is uncomfortable for a simple scarcity thesis. Concentration and strategic importance create attractive early-stage windows, but capacity responses, substitution research, and policy interventions consistently compress returns for anyone holding through the full cycle.

Now apply that to niobium and beryllium. Both are genuine early-stage strategic concentration plays. But CBMM’s overcapacity, 150,000 tonnes against a 124,000-tonne 2023 market, argues against a near-term squeeze. And beryllium’s growth trajectory, somewhere between a 3.61% volume CAGR and a 6.9% value CAGR, is steady rather than explosive.

Mining Magazine flagged the paradox directly.

Niobium “might be too scarce for its own good.”

That warning captures the trap. The reliable returns from fusion commodity exposure come from identifying the early window before demand consensus forms and capacity responses begin, not from assuming concentration guarantees perpetual pricing power. For tungsten and lithium-6, the SCSP framing adds a further layer: both are strategic enough to invite export controls, stockpiling, or price-distorting policy, which cuts both ways for an investor.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. These statements are speculative and subject to change based on market developments.

What the basket tells serious investors right now

Read as a whole, the fusion commodity basket is a thesis-positioning exercise, not a momentum trade. Build the framework now and you are ready to act when procurement signals arrive, rather than chasing them afterward.

Sort the basket into two buckets. Niobium, through CBMM’s strategic positioning, and beryllium, through Materion’s monopoly Western chain, are the currently investable concentration plays. Lithium-6 and tungsten are policy-watch items, not commodity trades. And structural steel is neither: it is an engineering and procurement challenge that industrial policy will solve, not market pricing.

The thesis moves from strategic optionality to near-term catalyst only when specific signals appear. These are the ones to watch:

• Commercial fusion reactor procurement announcements, particularly from the CFS/SPARC and ITER pathways whose magnet decisions will generate the first real demand signals
• A government lithium-6 enrichment programme launch, which would confirm the capability gap is being addressed
• Chinese tungsten export restrictions, signalling the basket is becoming politically live
• New beryllium mine development outside the US or China, which would ease the single-facility chokepoint

Track those, and you will know when the fusion commodity thesis shifts from speculative to actionable. Until then, the work is understanding the map, not trading it.

For investors ready to move from framework to portfolio construction, our full explainer on the fusion minerals investment case maps the specific entry signals, position sizing logic, and catalyst timelines that separate early-stage optionality from speculative overpay.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.