Highlights
- In 2025, China imposed export licensing on heavy rare earth elements, creating significant supply chain challenges for US defense technologies.
- The US defense sector faces a potential critical shortage of dysprosium and terbium, with current stockpiles insufficient to meet annual military demand.
- Domestic and international efforts are underway to diversify rare earth mineral sources.
- There is a 2027 deadline to eliminate Chinese-sourced magnet components.
In April 2025, Beijing tightened its grip on heavy rare earth elements (HREEs), imposing new export licensing on dysprosium, terbium, lutetium, samarium, gadolinium, scandium, and yttrium. These elements are essential for high-performance magnets used in defense and dual-use technologies. While framed as a national security measure, the move was synchronized as a response to U.S President Donald Trump’s Liberation Day (opens in a new tab) announcement and ensuing high tariffs on China. So, following U.S. tariff hikes, the move signaled retaliation in a deepening trade war.
Although not an outright ban, China gained discretion to delay or deny export licenses. Simultaneously, Chinese authorities blacklisted 16 U.S. defense companies, cutting them off from Chinese REE supplies. Prices for dysprosium and terbium surged, and U.S. contractors—from missile makers to radar system suppliers—scrambled to find non-Chinese sources.
Frankly, Dy and Tb are indispensable. They allow neodymium-iron-boron (NdFeB) magnets to perform at high temperatures, critical for fighter jet actuators and missile fin controls. With Beijing tightening the spigot, U.S. defense manufacturers confronted immediate vulnerability.
The 2027 Ban Looms
Even before China’s clampdown, Washington had set a hard deadline. Under the FY2023–24 National Defense Authorization Acts, by January 1, 2027, no magnet delivered to the Pentagon may contain Chinese rare earths at any stage of mining, refining, alloying, or manufacturing. The Defense Federal Acquisition Regulation Supplement (DFARS) (opens in a new tab) phases in restrictions through 2026, with the full ban in 2027. Industry experts widely view the timeline as unrealistic. China still controls about 90% of global refining and an even greater share of magnet production. By 2027, it may remain the only country capable of producing Dy and Tb oxides at an industrial scale.
Many expect a wave of waivers or deadline extensions. Officially, the Pentagon insists the ban will hold, but contingency plans are already in motion. Under conditions of anonymity, Rare Earth Exchanges (REEx) has had multiple discussions supporting such planning.
U.S. Stockpile: Thin Cover
The immediate hedge against supply disruption is the National Defense Stockpile (NDS), managed by the Defense Logistics Agency (opens in a new tab). But decades of drawdowns have left it threadbare.
During the Cold War, Washington maintained a rare earth cushion. In the 1990s, assuming global markets would suffice, the government sold off most inventories. By 2022, according to some sources, the entire U.S. stockpile contained only 203 kilograms of dysprosium metal and 526 kilograms of ferro-dysprosium alloy—together less than one metric ton. There was no meaningful terbium. But this may have changed, and the data is not available without clearances.
For context: a single F-35 contains roughly 427 kilograms of rare earths, while U.S. defense programs consume about 400 tonnes of dysprosium per year, all imported. The current stockpile could cover only weeks of demand.
Recognizing the danger, the Pentagon began quietly replenishing reserves in 2025 under the Defense Production Act (opens in a new tab). Still, officials concede inventories remain inadequate—enough to cushion a short disruption but not sustain operations. Allies like Japan have been more proactive, stockpiling since China’s 2010 embargo. The U.S. is playing catch-up. The FY2024 NDAA ordered studies on adding terbium oxide to the NDS. And in August 2025, DLA sought to acquire up to 30 metric tons of Dy oxide and 10 metric tons of Tb oxide (opens in a new tab), a dramatic increase if realized. Yet even those quantities would cover only a fraction of annual defense demand.
Building New Supply
As REEx chronicles, Washington and its partners are racing to diversify sources and scale processing capacity. Key initiatives include:
| Company | Model | Summary |
|---|---|---|
| MP Materials (opens in a new tab) | Mine-to-Magnet | In July 2025, DoD took a 15% equity stake via a $400 million investment and extended a $150 million loan for heavy REE separation. MP Materials plans a “Stage II+” circuit for Dy/Tb and a tenfold expansion in magnet production. A 10,000-ton per year Texas magnet plant is slated for 2027–28, matching U.S. demand. Until then, only a small Fort Worth facility produces a few hundred tons annually—less than 1% of global output. |
| Lynas Rare Earths (opens in a new tab) | Texas Plant | With $120 million in DoD funding, Lynas is building a U.S. heavy REE separation facility, expected online in 2026. This will be the first meaningful non-Chinese source of refined Dy and Tb, though capacity will be modest at start. |
| USA Rare Earth | Round Top, Texas | In early 2025, the company produced its first lab-scale sample of Dy oxide at 99.1% purity, a technical milestone. Commercial output is years away, but the project demonstrates potential for a domestic HREE source. |
| Recycling and Substitutes | MP Materials/Apple deal as example | Apple’s $500 million deal with MP Materials commits to magnets made from 100% recycled rare earths, with a new recycling facility at Mountain Pass. While consumer electronics use less Dy/Tb, this builds technical capacity for closed-loop supply. Meanwhile, U.S. agencies fund R&D into magnet alternatives like iron-nitride compounds and ferrites, though mission-critical adoption is years off. |
*Note others include ReElement Technologies, (opens in a new tab) which just secured a couple of million from DoD, plus a $20 million financing. This firm, partially owned by publicly traded American Resources Corporation (opens in a new tab), focuses on the ever-important mid-market. Mid-stream processing remains a bottleneck in both America and Europe.
Even with these efforts, supply gaps will persist. Analysts expect the U.S. will not achieve reliable heavy rare earth independence until the late 2020s, leaving a risky transition period. And REEx suggests even the late 2020s scenario presupposes near flawless execution.
Crisis Scenario: Total Cutoff
If China cut off heavy rare earth exports tomorrow, the U.S. would face stark choices:
| Possible Options | Summary |
|---|---|
| Tapping the Stockpile | Dy and Tb reserves could be rationed to the most critical systems, like missiles. But with only hundreds of kilograms available, supplies would vanish within months |
| Allied Supply | Australia’s Lynas and Japan’s reserves would help, but global non-Chinese output is minuscule. Terbium in particular is scarce. Emergency coordination among allies could mitigate, not solve, the shortfall. |
| Redesign and Substitution | Engineers could shift some systems to samarium-cobalt magnets, which tolerate heat without Dy/Tb. Yet Sm-Co magnets are weaker and present their own supply risks. Recycling and thrift—recovering Dy/Tb from waste and end-of-life products—would be accelerated. |
| Emergency Procurement | Some experts urge DoD to buy and stockpile finished alloys and magnets now, even if sourced from China, to build a buffer. Politically fraught, such a move underscores the urgency. In extremis, Washington could even negotiate a temporary truce with Beijing to ensure flows, though this would come with concessions. |
The imbalance is hard to overstate. In 2024, China turned out roughly 300,000 tons of NdFeB magnets while the United States produced almost none. Beijing also controls about 99% of global heavy–rare-earth processing—the dysprosium and terbium steps that make high-temperature magnets possible. Again, the very inputs our collective defense requires. In a crisis, China could simply prioritize its own military and industry, and the U.S. and its allies would be left scrambling for substitutes, waivers, or favors---for years.
This isn’t a comfortable story in Washington, but it’s past time to tell it. For years, political, corporate, and academic leaders rationalized the offshoring of critical mineral and rare earth element refining as efficient and inevitable. It wasn’t; it was fragile. The result is a defense and industrial base tethered to a rival’s choke point. Rebuilding will take real money, real timelines, and real stockpiles—not talking points. Yes, President Trump has done more than the previous administrations, but not nearly enough. The question isn’t whether the dependency exists. It’s whether we’ll confront it with the urgency it demands.
Conclusion
The U.S. is racing the clock to build a secure heavy rare earth supply chain. The 2027 ban on Chinese magnets looms, but meeting it remains improbable without waivers. The National Defense Stockpile offers little more than weeks of cover, though planned acquisitions of tens of tons of Dy and Tb oxides mark a step in the right direction.
Domestic projects—from MP Materials’ expanded capacity to Lynas’s Texas plant—signal serious investment, while recycling and R&D offer longer-term promise. Yet until these initiatives deliver volume, the vulnerability is acute.
The events of 2025 underscored that rare earths are central to modern military power. Deterrence runs on Dy, Tb, and other critical elements. Stockpile buildups, accelerated projects, and international coordination are now urgent priorities. In effect, Washington must treat heavy rare earths like strategic oil: build reserves, secure allies, and innovate substitutes.
For now, the U.S. heavy rare earth stockpile is a small insurance policy—one policymakers are scrambling to enlarge before the next shock. Whether that proves enough to bridge the gap until mines, refineries, and factories online remain the trillion-dollar question.
©!-- /wp:paragraph -->
0 Comments