The Once-Obscure Metal at the Heart of the Green Energy Trade War

NextFin News - Gallium sits quietly inside the periodic table, the 31st element, a silvery byproduct scraped from aluminium refining that few people had reason to notice. Today it sits at the centre of the world's most important trade war. China, which produces 99% of the world's primary gallium, has turned the metal into a geopolitical lever — and the green energy transition is the battlefield. Gallium nitride power chips now carry the current in electric-vehicle chargers, data-centre power supplies and solar inverters, which means the decarbonisation of transport, computing and electricity all depend on a metal the United States has not produced since 1987.

The confrontation over gallium is not a sideshow to the green energy trade war. It is the war's sharpest illustration of a new rule: in an industrial policy era, leverage migrates down the supply chain until it reaches the smallest, least substitutable input — and then it stays there.

The Valve, Not the Embargo

The sequence of moves is precise and deliberate. In August 2023, China's Ministry of Commerce imposed export licensing on gallium and germanium, apparently in response to American restrictions on advanced semiconductor sales. In December 2024, after Washington tightened chip-export controls again, Beijing escalated: Ministry of Commerce Notice 2024 No. 46, issued December 3, banned in principle all exports of gallium, germanium, antimony and superhard materials to the United States. Shipments to US military users were prohibited outright; civilian exports required licences that were, in practice, rarely granted.

The effect was immediate and measurable. Chinese customs data show exports of unwrought gallium fell 94% in 2025 from 2024 levels, and shipments to the United States effectively disappeared after 2023. Outside China, gallium spot prices tracked by Fastmarkets reached a record $1,850 per kilogram in April 2026 — more than 200% above the start of that year — while Chinese domestic prices held steady. A market that splits along a border is the classic signature of a weaponised supply chain. By late July 2026, the western benchmark price tracked by Argus Media had reached $3,050 per kilogram, roughly nine times the level just before the first controls were announced.

Yet the war did not end there, and this is the point most readers miss. In November 2025, as part of a broader trade truce, China suspended the harshest measures: Announcement No. 70 paused rare-earth controls until 10 November 2026, and Announcement No. 72 suspended the US-specific gallium and germanium ban until 27 November 2026, restoring standard licensing. In September 2026, Beijing said the wider arrangement had been extended to 10 January 2027 — though the specific rare-earth suspension notice still legally expires on 10 November, leaving importers planning against two different dates.

The controls, in other words, are not a permanent embargo. They are a valve, and Beijing keeps its hand on it. That is a more sophisticated instrument than a ban: it extracts concessions and taxes foreign manufacturers without triggering a full break.

A Choke Point Built on Processing, Not Geology

The first thing to understand about gallium is that it is almost never mined. It is recovered as a byproduct of processing bauxite into alumina, and to a lesser extent from zinc ores. That technical fact has a strategic consequence: supply is inflexible. You cannot open a gallium mine in response to a price signal, because gallium does not come in mines of its own. You must build a refinery attached to an aluminium or zinc operation, and you must be willing to run it even when the economics are marginal — which is precisely what Chinese producers did while Western competitors shut their lines down.

The roll call of Western exits is instructive. Germany, Hungary and Kazakhstan all ceased primary gallium production, in 2016, 2015 and 2013 respectively; Ukraine most likely stopped in 2022. The United States has not recovered a gram of primary gallium since 1987. By 2025, China accounted for 99% of worldwide primary low-purity gallium production, with only Japan and Russia remaining as meaningful producers. Global primary output was an estimated 900,000 kilograms in 2025 — under 1,000 tons — while China's reported production capacity vastly exceeds that figure, which means Beijing can flood the market or starve it almost at will.

This is the mechanism that makes export controls so potent. A licensing requirement produces effects comparable to a full trade embargo, at minimal diplomatic cost, because buyers cannot quickly reroute. There are no liquid futures markets for gallium or germanium, so firms cannot hedge financially; they must stockpile physically or sign long-term contracts. And the asymmetry in strategic reserves is stark: the US Department of Defense maintains a germanium stockpile, but gallium is not included in the National Defense Stockpile. The segment of the supply chain most exposed to disruption is the one least protected.

The economic stakes are large enough to matter but small enough to be overlooked. US Geological Survey modelling estimated that a complete one-year restriction of China's gallium exports would reduce US GDP by approximately $3.1 billion, rising to $3.4 billion if both gallium and germanium are restricted. Semiconductor manufacturing absorbs more than 40% of that impact. For a US economy of roughly $28 trillion, $3.4 billion is not catastrophic — but it is concentrated in precisely the industries Washington has identified as strategic, and it arrives at a moment when those industries are being asked to scale at wartime speed.

The Green Energy Connection Runs Deeper Than the Chips

Gallium's reputation until recently was that of a semiconductor oddity — the material inside radio-frequency amplifiers, LEDs and laser diodes. Its new importance comes from power electronics, the invisible hardware that converts and conditions electricity everywhere it is used. According to the US Geological Survey, integrated circuits accounted for 73% of US gallium consumption in 2025 and optoelectronic devices for 26%; the power-electronics applications are the fastest-growing slice of that demand.

An electric vehicle contains gallium in its onboard charger, its DC-DC converter and increasingly in its traction inverter, because gallium nitride devices handle higher voltages and frequencies in a smaller package than silicon. A data centre running AI workloads uses gallium nitride power modules to convert electricity at every stage between the grid and the processor; single-stage 48V-to-1V conversion enabled by the material can cut energy losses dramatically in facilities whose power bills now run into the tens of millions of dollars a year. Solar inverters and wind-turbine power electronics use it for the same reason: less energy wasted as heat means more of the renewable electricity actually reaches the grid.

The irony is deliberate and cuts both ways. The same technologies that Western governments are subsidising to decarbonise the economy — EVs, data centres, renewables — are the ones most exposed to a Chinese-controlled input. Industrial policy in Washington and Brussels assumed that the bottlenecks were batteries and chips; gallium shows that the choke points sit one layer deeper, in the materials that make the enabling components work. A trade war fought over advanced semiconductors has a way of migrating down the stack until it reaches the raw inputs, because that is where leverage is greatest.

Gallium nitride now accounts for roughly 38% of gallium demand by product type, and the gallium nitride device market is projected to grow from about $2.46 billion in 2026 to $7.91 billion by 2034, a compound annual rate above 15%. Every watt of efficiency gained in a data centre or an electric vehicle runs through this metal. That is why the metal's obscurity is ending: it has become an indirect bet on the energy transition itself.

There is a second green-energy front that often gets conflated with gallium but operates differently: rare-earth permanent magnets. Neodymium-iron-boron magnets are the heart of direct-drive offshore wind turbines and many EV motors, and China controls roughly 90% of global magnet manufacturing capacity. In late 2023 and 2024, Beijing restricted the export of rare-earth extraction, separation and magnet-manufacturing technology, banning the transfer of intellectual property and machinery with no exceptions for joint ventures or licensing. The objective is not merely to control what the West can buy today, but what it can build tomorrow.

The two fronts reveal different logics. Gallium is a byproduct bottleneck: hard to substitute quickly, but ultimately expandable if someone is willing to build refineries. Rare-earth magnet technology is a know-how bottleneck: even with ore in hand, the West cannot manufacture high-performance magnets at scale without the process technology China has now placed off-limits. One is a capacity problem; the other is a knowledge problem. Capacity can be built in years; knowledge, when it has been deliberately fenced off, takes longer.

The West's Answer Is Real, but It Is Slow

Faced with this exposure, Western governments and companies have moved with unusual speed — and with unusual realism about how slow the fix will be. The response falls into three categories: recycling, new primary capacity, and substitution.

Recycling is the fastest lever. Neo Performance Materials, a magnet manufacturer, recycles a few dozen tons of high-purity gallium a year from semiconductor scrap at its Ontario plant — the only industrial-scale gallium producer in North America. In the summer of 2023, after China's first controls, the company's vice president of corporate development, Vasileios Tsianos, began fielding calls from officials on both sides of the Atlantic.

"We've spoken to almost everyone" interested in producing gallium outside China, he said.

New primary capacity is starting to appear. Rio Tinto drew the first primary gallium from its Vaudreuil refinery in Quebec in 2025, and is building a pilot plant for 2027 that could scale to 40 tons a year — perhaps 5% to 10% of world supply. In the United States, the Department of Energy announced up to $6 million in September for research into recovering gallium from alumina refining and zinc smelting, aiming to restart domestic primary recovery for the first time in almost 40 years. In November, the Department of War (formerly the Department of Defense), acting under the Defense Production Act, awarded $29.9 million to develop a demonstration facility in Louisiana that would recover gallium and scandium from industrial waste.

These projects are meaningful, but the arithmetic is humbling. Even if every announced project reaches its nameplate capacity, non-Chinese supply by 2030 would still be a fraction of today's Chinese output. Marina Zhang, an associate professor at the University of Technology Sydney who studies critical minerals and geopolitics, estimates that building a genuinely independent rare-earth supply chain could take 15 years and cost dearly.

"It may be very costly, but it could be safe," she said.

Substitution is the third path, and it is already reshaping product design. Some EV makers are developing rare-earth-free motors; Tesla, which switched to rare-earth-based motors in 2017, has said its next-generation motors expected in 2026 will be rare-earth-free again. Valeo's rare-earth-free motor is at least three years from production. For gallium itself, substitution is harder: in many defence and high-frequency applications, no effective substitute exists for gallium arsenide and gallium nitride. Silicon carbide competes in some power-electronics applications, but it is a different material with a different cost structure, not a drop-in replacement.

The Counter-Thesis: A Jab, Not a Haymaker

The strongest argument against the alarm is that the controls have not, so far, actually strangled Western industry. Analysis of US customs data by the Peterson Institute for International Economics found that China's shares of total US imports of gallium, germanium and graphite had not moved much after the first controls, because importers rerouted shipments through intermediaries, drew on inventories and found alternative suppliers. The institute concluded that Chinese export controls on critical minerals "aren't starving the United States—at least so far" — closer to a jab than a haymaker.

There is truth in this. The November 2025 suspension demonstrated that the controls are a negotiating lever as much as an economic weapon, and that Beijing is willing to ease them when broader trade talks demand it. The price spike, while dramatic, has not translated into visible production stoppages in Western EV or semiconductor plants — at least not yet. And gallium is a small market: global primary production is under 1,000 tons a year, worth a few hundred million dollars at pre-control prices. By the scale of the chip war, where hundreds of billions of dollars of equipment sales are at stake, gallium looks almost trivial.

But this reading mistakes the absence of collapse for the absence of leverage. The reason Western plants have not shut is precisely that they stockpiled, rerouted and paid the premium — costs that do not show up as outages but do show up as margin pressure and higher capital tied up in inventory. The bifurcated price — $3,050 per kilogram in the West versus steady prices inside China — is itself the weapon working as intended: it taxes Western manufacturers without triggering a casus belli. The controls are designed to be painful enough to extract concessions and slow foreign capacity build-out, but not painful enough to force a full break. That is not weakness. It is calibration.

The counter-thesis also underestimates the signalling effect on investment. Every announcement of a new Western refinery or recycling plant carries an implicit admission that the market cannot be trusted to clear on its own. When governments must subsidise the first gram of domestic gallium in 40 years, the era of cost-minimised global supply chains for critical inputs is over — regardless of whether the current suspension holds.

What to Watch: Three Scenarios

The central judgment is this: gallium's price spike is cyclical and will ease as non-Chinese capacity comes online, but the fragmentation of the supply chain is structural and will not revert. The two must be separated, because they point in opposite directions for investors and policymakers. Cyclical means the $3,050-per-kilogram western price is not a permanent plateau — Rio Tinto's Quebec output, Neo's recycling expansion and the US demonstration facilities will add supply, and if the suspension is renewed in 2027 the premium will compress. Structural means the market that emerges will be a different market: regionalised, subsidised, and priced for resilience rather than efficiency.

Three scenarios frame the path ahead. In the base case, Beijing issues a further suspension in late 2026 or early 2027, the western premium compresses toward $1,000-$1,500 per kilogram, and non-Chinese capacity creeps toward 5% of global output by 2028 — enough to matter politically, not enough to restore the old market. In the upside case for supply security, the January 2027 extension holds and a numbered notice clarifies the rare-earth suspension, allowing announced projects to reach financial close; western capacity reaches double-digit share by the end of the decade and the premium largely disappears. In the downside case, the 10 November notice lapses without replacement, exports to the United States remain near zero through 2027, and the western premium retests its record highs while every green-energy project prices in a two-supply-chain world.

The falsifying test for the structural thesis is specific and observable. If non-Chinese primary gallium capacity reaches more than 10% of global output by 2028 and the western price premium narrows below 20% of Chinese prices, then the choke point is closing and the decoupling narrative has been overstated. If instead Chinese exports to the United States remain near zero through 2027 and western capacity stalls below 5%, the re-fragmentation is confirmed and every green-energy supply chain will be priced accordingly.

Who benefits and who is exposed is clear. Beneficiaries are the few companies with non-Chinese gallium or recycling capacity — Neo Performance Materials in North America, Rio Tinto in Quebec, and any firm that can recover gallium from alumina red mud or zinc residues at scale. Also advantaged are rare-earth producers outside China, such as Lynas and MP Materials, and makers of rare-earth-free motor technology. The exposed are Western EV makers, data-centre operators and solar developers whose efficiency roadmaps assume uninterrupted gallium nitride supply — and, more subtly, the policymakers who have promised to electrify everything on a timeline that assumes inputs will arrive on schedule.

The deeper lesson of gallium is uncomfortable for both sides. For the West, it is that industrial policy discovered the bottleneck only after the bottleneck had already been weaponised. For China, it is that leverage is self-limiting: every use of the valve accelerates the construction of alternatives, until one day the valve controls a smaller and smaller share of the flow. Gallium will not decide the green energy trade war. But it has already decided one thing: the transition will be built on two supply chains, not one, and both will cost more than the models assumed.

Explore more exclusive insights at nextfin.ai.