Highlights
- China controls over 90% of global NdFeB sintered magnet production, creating a critical vulnerability in U.S. military drone supply chains.
- Starting January 1, 2027, DFARS rules extend upstream to raw mining of neodymium, iron, boron, samarium, and cobalt—not just magnet manufacturing.
- Western producers like MP Materials, VAC, and Neo Performance Materials are building capacity, but qualified defense-grade output remains years away.
- Even a few grams of noncompliant rare-earth material in a drone magnet can halt delivery of an entire weapon system.
- The realistic horizon for a redundant, allied rare-earth magnet supply chain spanning mines to finished magnets is 2030 and beyond.
America's military-drone boom has a vulnerability measured not in aircraft, missiles, or microchips, but in grams of magnetic material. General Atomics (opens in a new tab) builds the MQ-9 family (opens in a new tab). Northrop Grumman (opens in a new tab) (NYSE:NOC) produces the MQ-4C Triton (opens in a new tab). AeroVironment (opens in a new tab) (Nasdaq:AVAV) supplies Raven (opens in a new tab), Puma (opens in a new tab), and Switchblade (opens in a new tab) systems. Newer entrants are expanding autonomous aircraft and loitering-munition production.
Follow these systems far enough down the supply chain, however, and many roads eventually reach the same component: the high-performance permanent magnet. That is where China's industrial dominance becomes difficult to avoid. As Rare Earth Exchange® continues to report, China accounts for roughly 90%+ of global NdFeB magnet, alloy, and powder production in 2025. Neo Performance Materials more recently says more than 90% of sintered magnets are produced in China. And on January 1, 2027, the sourcing problem becomes substantially harder.
DFARS Goes All the Way to the Mine
Current DFARS 252.225-7052 restrictions already cover specified magnet-production stages in China, Russia, Iran, and North Korea. Beginning January 1, the rule reaches dramatically further upstream. For neodymium-iron-boron magnets, the restriction encompasses the entire supply chain from mining neodymium, iron, and boron through production of the finished magnet. For samarium-cobalt magnets, it similarly reaches upstream to cobalt and samarium ore or feedstock.
That changes the compliance equation.
A magnet sintered in Ohio from Chinese-origin material does not automatically solve the problem. Neither does a European or Japanese magnet whose covered upstream material originated in China. Contractors must increasingly know not simply where the magnet was manufactured, but where the covered materials originated. There are exceptions—including certain electronic devices, nonavailability determinations, and an important provision for NdFeB magnets manufactured from recycled material when milling of that material and final sintering occur in the United States. But exceptions are not an industrial strategy.
Where the Magnets Hide
Rare-earth exposure varies radically by drone architecture. The MQ-9B, for example, is powered by a Honeywell turboprop (opens in a new tab). Its main propulsion system therefore should not be modeled like a giant electric drone motor. Permanent magnets instead potentially appear throughout electrical generation, actuators, pumps, fans, radar and antenna mechanisms, and EO/IR systems.
Small AeroVironment aircraft such as Raven and Puma are different: electric propulsion makes the propulsion motor itself a direct magnet exposure. Switchblade combines propulsion, actuation, and stabilized electro-optical systems in an extraordinarily weight-constrained package. The relevant chain can therefore run: Drone → motor/actuator/sensor → finished magnet → sintered NdFeB → alloy → Nd/Pr/Dy/Tb metal → separated oxide → concentrate → mine.
Unfortunately, public defense disclosures generally disappear before identifying the magnet manufacturer, grade, or upstream oxide supplier. Exact magnet bills of material for systems such as MQ-9, Triton, Raven, and Switchblade are not publicly available. That opacity becomes a procurement problem when DFARS demands deeper provenance.
How Much Rare Earth Is in the Magnet?
Sintered NdFeB magnets are typically around 30% rare earths by weight in Department of Energy (DOE) reference compositions, principally neodymium and praseodymium. Dysprosium and terbium can be added to increase coercivity—the magnet's resistance to demagnetization—particularly at elevated temperatures.
Grade matters enormously.
The familiar number in grades such as N42 relates to maximum energy product. Suffixes including M, H, SH, UH, EH, and AH denote progressively higher coercivity/temperature families. DOE reference compositions illustrate the materials consequence: an H-class composition contains roughly 2.8% Dy, SH about 4.2%, and UH about 6.5%; older EH/AH reference compositions can reach roughly 8.5–11%.
Thus one kilogram of an SH reference magnet could contain roughly 258 grams Nd/Pr and 42 grams Dy. Modern grain-boundary diffusion can substantially reduce heavy-rare-earth requirements, making these reference values unsuitable as a universal contemporary specification.
And no credible public evidence establishes that a particular Switchblade uses, for example, N42SH rather than another grade. Claims of exact military-drone magnet grades or dysprosium content should therefore be treated skeptically without supplier documentation.
Neo offers a useful scale comparison. Its teardown of a commercial drone estimated roughly 10–60 grams of sintered NdFeB, plus several grams of bonded NdFeB. Applying a 30% reference rare-earth loading suggests only about 3–18 grams of rare earths in those sintered magnets. That sounds insignificant—until several grams of noncompliant material prevent delivery of an entire weapon system.
The Non-China Magnet Buildout
The Western supply chain is no longer starting from zero. MP Materials (opens in a new tab) (NYSE:MP) produced its first NdFeB magnets on commercial equipment in Texas in 2025 and is developing its much larger 10X campus in Northlake. Commissioning is scheduled to begin in 2028, with MP ultimately targeting approximately 10,000 tonnes of total U.S. annual magnet capacity.
VAC's eVAC (opens in a new tab) (merging with Energy Fuels) facility in South Carolina began commercial magnet production in 2025 and has roughly 2,000 tonnes per year of nameplate capacity.
Neo Performance Materials (opens in a new tab) (TSX.NEO) has built a 2,000-tonne-per-year first-phase sintered-magnet facility in Estonia, with mass production scheduled to begin in 2026 and expansion toward 5,000 tonnes contemplated.
USA Rare Earth (opens in a new tab) (Nasdaq:USAR) is commissioning U.S. magnet production. Lynas Rare Earths (opens in a new tab) (ASX.LYC) provides an important non-Chinese upstream source of separated rare earths. Energy Fuels (opens in a new tab) (NYSEAmerican:UUUU), acquiring VAC, is developing a broader U.S.-allied strategy spanning separated oxides, metals/alloys, and magnets. Advanced Magnet Lab (opens in a new tab) (AML), a firm profiled via the REEx podcast (opens in a new tab), reports multiple collaborations with the emerging American drone supply chain. Noveon Magnetics (opens in a new tab) also is reportedly collaborating with the drone supply chain.
These developments matter—but nameplate capacity is not qualified capacity. A new factory must achieve yield, consistency, coercivity, geometry, and coating requirements; establish traceable feedstock; qualify with motor and actuator manufacturers; and ultimately be approved within defense supply chains.
High-coercivity Dy/Tb-bearing magnets present an especially difficult bottleneck. That makes 2028 an inflection point, not an independence date. The more credible horizon for multiple mines, separators, metallizers, alloy producers, and qualified magnet manufacturers operating with meaningful redundancy is 2030 and beyond.
The Pentagon's immediate challenge, therefore, is not replacing China's entire magnet industry by January 1.
It is identifying which magnets can stop which weapon systems. Every critical magnet part number needs traceable provenance. Buyers need to qualify grades rather than simply reserve tonnes. Recycling should be exploited where the DFARS exception applies. Engineers should reduce Dy/Tb intensity through grain-boundary diffusion, thermal management, and optimized magnet grades. And procurement must integrate U.S. and allied mines, separators, metallizers, and magnet producers rather than wait for an entirely domestic ecosystem.
The rare-earth content of an individual drone may be surprisingly small.
Its strategic importance is not.
In the next phase of drone warfare, a few grams of the wrong magnet could become the weakest link in a multibillion-dollar defense supply chain.
0 Comments
No replies yet
Loading new replies...
Moderator
Join the full discussion at the Rare Earth Exchanges Forum →