Lynas Rare Earths produced its first separated terbium oxide at its Kuantan, Malaysia facility in June 2025, becoming the world's only commercial producer of separated heavy rare earth oxides outside China. The milestone followed first dysprosium production in May and was accompanied by a binding US Department of Defense supply agreement worth approximately $96 million over four years, signed in March 2026. Together, these developments confirm Malaysia as the dominant non-China node for heavy rare earth separation at a moment when Chinese export controls have made alternative supply a strategic imperative.
Introduction
In June 2025, a processing line at Lynas Rare Earths' advanced materials plant in Kuantan, Malaysia produced its first batch of separated terbium oxide. The event was described by the company and independent analysts as the first time terbium had been separated at commercial scale outside China, arriving roughly one month after the same facility achieved first production of dysprosium oxide in May. Together, the two milestones made Lynas the world's only commercial producer of separated heavy rare earth oxides outside China, a status the company has since extended with first samarium oxide production in March 2026, ahead of its original April 2026 target.
Those production firsts did not sit on the open market for long. On March 16, 2026, Lynas announced that its wholly owned subsidiary Lynas USA LLC had signed a binding Letter of Intent with the United States Department of War for rare earth oxide supply, with the US government allocating approximately $96 million to purchase light and heavy rare earth oxide products over a four-year period. The agreement includes a floor price of $110 per kilogram for neodymium-praseodymium oxide, matching a floor price the Pentagon had previously established in its deal with MP Materials.
The sequence of events tells a compressed but consequential story. The first commercial separation of terbium outside China was followed within nine months by a binding government-to-company supply agreement that effectively pre-allocates a substantial share of that new output to strategic defence buyers. For anyone tracking the race to build alternative critical mineral supply chains, the Lynas Malaysia HREE circuit represents the most tangible proof of concept yet that non-Chinese heavy rare earth separation is not merely a policy aspiration but an operational reality.
The HREE Circuit: Commissioning, Capacity, and Feed
The heavy rare earth separation circuit at the Lynas Advanced Materials Plant (LAMP) in the Gebeng Industrial Estate near Kuantan was commissioned during the March quarter of 2025. Lynas had flagged the milestone sequence in its March 2025 Quarterly Report: dysprosium oxide was expected in May, terbium in June. Both targets were met, which is not a trivial achievement for a circuit of this complexity operating without a Chinese technology counterpart.
The initial circuit is rated to process up to 1,500 tonnes of heavy rare earth feedstock per year. Some analysts have noted that if that capacity were focused exclusively on dysprosium and terbium, the output could potentially represent as much as a third of global non-Chinese production of those elements, though the practical product mix will depend on concentrate composition. The feedstock flows from Mt Weld in Western Australia, one of the highest-grade rare earth deposits globally, whose ore reserve grade stands at 400 parts per million dysprosium oxide and 100 parts per million terbium oxide, following a 2024 reserves upgrade that recorded a 92 percent increase in contained dysprosium oxide.
The LAMP facility has been operational since 2012 and is a mature industrial site spanning 100 hectares. It already carries a nameplate NdPr oxide capacity of approximately 10,500 tonnes per year. The addition of the HREE circuit does not replace that light rare earth infrastructure; it sits alongside it, converting the same Mt Weld concentrate into a far wider portfolio of separated products. The plant processes material through three stages: cracking and leaching, solvent extraction, and product finishing, a configuration that positions it to expand its HREE product range as the new expanded facility moves toward commissioning.
CEO Amanda Lacaze, whose departure from the managing director role coincided with the DoD deal announcement in March 2026, framed the terbium milestone in direct commercial terms. "Lynas is delighted to have achieved first production of terbium oxide at the Lynas Malaysia advanced materials plant," she said. "Together with dysprosium oxide, which was first produced on our new heavy rare earths production line in May, Lynas can now supply customers with the two heavy rare earths required for rare earth permanent magnets used in electric motors."
The $96 Million DoD Agreement: Price Floors, Strategic Allocation, and the Texas Complication
The binding Letter of Intent signed on March 16, 2026 between Lynas USA LLC and the US Department of War formalises a four-year supply relationship covering both light and heavy rare earth oxides. The approximately $96 million total is spread across that period, with the NdPr floor price of $110 per kilogram providing Lynas with revenue predictability at a time when spot prices for rare earth oxides have been volatile and, outside China, structurally elevated.
The floor price figure is not incidental. It is the same benchmark the Pentagon attached to its deal with MP Materials, signalling an emerging US government practice of anchoring domestic and allied rare earth procurement at price levels that support commercially viable non-Chinese supply chains, rather than simply buying at prevailing spot prices. Last week, Lynas also agreed an offtake with Japan Australia Rare Earths, which negotiates on behalf of Japanese companies, for at least 5,000 tonnes per year of NdPr oxide until 2038, also at a $110 per kilogram floor. The convergence of US and Japanese procurement strategies around the same price anchor suggests a degree of coordination that goes beyond coincidence.
The DoD agreement is also a direct product of complications elsewhere in the Lynas-Pentagon relationship. In 2022, Lynas received a $120 million contract from the Department of Defense to build a commercial heavy rare earths processing facility at Seadrift, Texas; that support was later raised to approximately $258 million under an expenditure-based reimbursement structure. But the Texas facility has encountered persistent difficulties in securing offtake agreements under an environment in which Washington's "America First" industrial policy directed the largest Pentagon rare earth investments toward MP Materials. Lynas acknowledged that the Seadrift project faced "significant uncertainty as to whether construction will proceed," and the March 2026 Letter of Intent is explicitly framed as a modification of the original arrangement to reflect that uncertainty. Discussions for further supply arrangements, including specifically for heavy rare earth oxides, are described as ongoing.
For strategic buyers, the critical point is not the price mechanics but the allocation dynamic. Lacaze captured it in a statement accompanying the DoD announcement: "Through this agreement, the US defence industrial base will continue to have access to light and heavy rare earth oxides that are essential for modern manufacturing." Dysprosium and terbium are both on the list of elements subject to China's April 4, 2025 export controls, and both are essential for the high-performance permanent magnets used in F-35 fighter jets, missile guidance systems, and UAV motors, applications where heat resistance and magnetic durability under extreme conditions are non-negotiable.
China's Export Controls and the Pricing Environment That Makes All of This Matter
The Lynas HREE milestones and the DoD supply agreement did not emerge in a vacuum. They were accelerated and given strategic urgency by China's April 4, 2025 decision to add seven medium and heavy rare earths, including terbium, dysprosium, samarium, gadolinium, lutetium, scandium, and yttrium, to its export control list. The licensing regime established by China's Ministry of Commerce introduced case-by-case evaluation protocols that, in practice, prioritise diplomatic relationships over commercial considerations, turning what had been a commodity market into a tool of strategic resource diplomacy.
The market effects were immediate and severe. Chinese magnet exports fell by 74.3 percent in volume and 76.1 percent in value year on year in May 2025. Exports to Germany dropped by 70 percent; shipments to Japan declined by 84.1 percent. The IEA noted that many carmakers in the United States, Europe, and elsewhere struggled to obtain permanent magnets, with some forced to cut utilisation rates or temporarily shut down factories. For heavy rare earths specifically, BMI reported that pricing outside China surged to a 250 percent premium over Chinese domestic prices in the second half of 2025, following export control escalations in April and October of that year.
China subsequently introduced Article 49 of its 2024 dual-use export regulations in October 2025, which would have required export licenses for foreign-made goods containing at least 0.1 percent Chinese-origin rare earth content. Those October measures were suspended for one year at the APEC summit in Busan on October 30, 2025, but the April 4 controls on dysprosium and terbium remained in force, preserving China's licensing authority over the two elements most critical to Western defence magnet supply chains.
In this environment, heavy rare earth availability has become the central concern for Western industrial buyers, with procurement increasingly resembling strategic sourcing rather than commodity trading. Lynas itself noted in a recent quarterly report that pricing for its heavy rare earth products is expected to carry a premium over Chinese benchmark prices, reflecting what the company described as "high demand" from Western customers. The gap between Chinese domestic prices and export prices for terbium and dysprosium represents a structural feature of the current market rather than a temporary dislocation.
Malaysia's Expanding Role: From Processing Hub to Magnet Ecosystem
Building on my analysis of the Carester-Malaco joint venture in my July 2026 piece on the France-Malaysia rare earth axis, the broader picture in Malaysia is one of deliberate industrial clustering, with multiple actors converging on the country as the optimal non-China location for heavy rare earth midstream activity.
Lynas is not the only company advancing Malaysian rare earth ambitions. Carester and Malaco Mining Group are developing a 13,000-tonne-per-year separation plant in Perak state. Solvay has its own interests in the country's rare earth ecosystem. And Lynas itself has signed a memorandum of understanding with South Korea's JS Link to develop a 3,000-tonne-per-year sintered NdFeB magnet manufacturing facility near the LAMP plant, a joint venture that would move Malaysia from a separation hub into downstream magnet production. Lacaze described it as "leveraging JS Link's technology and expertise in permanent magnet manufacturing to develop an exciting new industry in Malaysia."
Malaysia's positioning reflects both its geological endowment and its industrial infrastructure. The country holds an estimated 16.1 million tonnes of non-radioactive rare earth deposits and is pursuing a mine-to-magnet industrial strategy targeting $3 billion in direct revenue by 2030. The Gebeng Industrial Estate, where LAMP sits, is a mature petrochemical zone with the utilities, logistics, and regulatory familiarity that new processing operations require.
Lynas is also investing in the next generation of its own HREE capacity. The company has announced a new, larger HREE separation facility in Malaysia, estimated to cost approximately RM500 million (around $118 million), self-financed following an equity raise in September 2025. That facility will have the capacity to process up to 5,000 tonnes per year of HRE feedstock, more than three times the initial circuit's capacity, and its initial flowsheet will include separated samarium, gadolinium, dysprosium, terbium, yttrium, and lutetium. The expanded product range is significant: samarium is required for samarium-cobalt permanent magnets used in aerospace and defence applications where NdFeB magnets are unsuitable, and gadolinium has growing applications in medical imaging and next-generation reactor designs.
Mt Weld's reserve base supports the long-term ambition. The mineral resource stands at 106.6 million tonnes at 4.12 percent total rare earth oxide, with ore reserves of 32 million tonnes at 6.44 percent TREO. Those reserves support a mine life of more than 35 years at 7,200 tonnes per year of NdPr oxide production capacity, and more than 20 years at the 12,000-tonne-per-year expansion capacity currently under construction.
Structural Limitations and What the Milestones Do Not Yet Resolve
The Lynas HREE milestones are genuinely significant, but a complete picture requires acknowledging the constraints that remain. Mt Weld's heavy rare earth endowment, while impressive by global standards, sets a ceiling on what can be produced from that feedstock. The deposit's ore reserve grade of 400 parts per million dysprosium oxide and 100 parts per million terbium oxide is respectable for a light-rare-earth-dominant deposit, but it does not approach the HREE concentrations found in Chinese ionic clay deposits in Jiangxi province, which have historically provided the bulk of global dysprosium and terbium supply. Even a perfectly operating 5,000-tonne-per-year HREE separation circuit fed from Mt Weld concentrate will produce considerably less dysprosium and terbium than the nameplate suggests once actual feed grades are applied.
China still controls approximately 85 percent of global rare earth processing capacity, and the West's challenge is not merely to find separation capacity outside China but to replicate the full value chain, including metals production, alloy manufacturing, and magnet fabrication, at commercial scale. The Lynas-JS Link magnet manufacturing MoU is a step in that direction, but it remains a memorandum, not an operating facility. The Texas Seadrift plant, which would have brought heavy rare earth processing onto US soil, remains in a state of unresolved uncertainty despite years of investment and negotiation.
The regulatory dimension in Malaysia also warrants attention. Lynas's Permanent Disposal Facility for radioactive processing residues is reported to be 72 percent complete and targeted for completion by end-2026, but it must meet strict requirements from Malaysia's Atomic Energy Licensing Board, including detailed engineering plans and ensuring residue stays below 1 becquerel per gram. Lynas has also committed 1 percent of its revenue to local research and development as part of its Malaysian operating license conditions. These are manageable obligations for a company of Lynas's scale, but they are real constraints on operational flexibility.
None of these limitations diminish what the June 2025 terbium milestone represents. For thirty years, commercial-scale separation of heavy rare earth oxides was a Chinese monopoly. That monopoly is now broken, even if the alternative capacity remains small relative to Chinese output. The question that the next three to five years will answer is whether the current wave of investment, spanning Lynas, Carester, USA Rare Earth (whose Wheat Ridge hydromet demonstration plant I covered in July 2026), and others, can accumulate enough non-Chinese separation and processing capacity to provide genuine market resilience before the next geopolitical shock forces a crisis.
Conclusion: The Race to Allocate What Has Not Yet Been Produced
The most striking feature of the Lynas HREE story is not the production milestone itself but the speed with which new non-Chinese heavy rare earth output is being allocated before it exists at scale. The DoD supply agreement was signed in March 2026, nine months after first terbium production and at a point when the expanded 5,000-tonne-per-year HREE facility had been announced but not yet built. The Japanese offtake agreement covering at least 5,000 tonnes per year of NdPr oxide until 2038 was signed around the same time. Strategic buyers are not waiting for spot market availability; they are contracting now for supply that will come online over the next several years.
This dynamic is a direct consequence of China's export control posture. When Beijing added terbium and dysprosium to its export control list in April 2025, it did not merely disrupt a commodity market; it converted every tonne of non-Chinese heavy rare earth output into a strategic asset with a queue of government-backed buyers. The $110 per kilogram NdPr floor price in both the DoD and Japanese offtake agreements is a mechanism for keeping that queue stable: it gives Lynas the revenue certainty to invest in expansion while giving buyers the supply certainty they need to plan defence procurement and industrial strategy.
The Malaysia cluster, encompassing Lynas's operational HREE circuits, the forthcoming 5,000-tonne-per-year expansion, Carester and Malaco's Perak separation plant, and the prospective JS Link magnet facility, is the most advanced concentration of non-Chinese rare earth midstream infrastructure currently being assembled anywhere in the world. It is not sufficient on its own to replace Chinese supply, and the limitations of Mt Weld's HREE grades are real. But it is a working foundation for a supply chain architecture that, a decade ago, did not exist outside of government white papers.
For the rare earth sector, the Lynas terbium milestone marks the end of the beginning. The first commercial separation outside China has been achieved, the strategic buyers have signed agreements, and the next round of capacity investment is underway. The question is no longer whether non-Chinese heavy rare earth separation is technically possible; it is whether the pace of investment can keep up with the pace of demand from a defence and clean energy industrial base that is no longer willing to depend on a single supplier.