Scientists from the Chinese Academy of Sciences have identified a new type of rare earth deposit formation in Heilongjiang and Jilin provinces, where freeze-thaw cycles break alkaline granite into loose sand and gravel carrying monazite and xenotime minerals enriched in both light and heavy rare earth elements. The discovery challenges the established 'heavy in south, light in north' geological model and could extend China's low-cost production advantage precisely as Western governments have deployed $9.1 billion in new project financing since the start of 2025. McKinsey projects non-China sources will still supply less than one-fifth of dysprosium and terbium demand by 2035, and this find does nothing to improve those odds.
Introduction
The headline out of Beijing this week is not a policy move, an export licensing decision, or a quota revision. It is a geological paper published in Acta Petrologica Sinica by scientists from the Chinese Academy of Sciences' Institute of Geology and Geophysics and the Heilongjiang Bureau of Geology and Mineral Resources, and it deserves more attention from the investment community than it has received so far.
The CAS-Heilongjiang team has identified a new class of rare earth deposit in northeastern China, specifically in Heilongjiang and Jilin provinces, formed not by the clay-heavy ionic adsorption processes that define China's southern deposits, but by repeated freeze-thaw cycles that break alkaline granite into loose sand and gravel. The rare earth elements in these northern formations are carried in discrete mineral particles, principally monazite and xenotime, rather than adsorbed onto clay surfaces. That structural difference has significant implications for extraction cost, environmental footprint, and the competitive position of every non-Chinese project currently competing for Western government capital.
Building on my analysis of China's midstream processing chokehold in 'The Midstream Is the War' (May 2026), this discovery adds a new upstream variable to an already asymmetric equation. The West is racing to build mine-to-magnet supply chains outside China's control. China just found a new type of deposit that could make its own production cheaper, cleaner, and geographically broader, all at once.
The Science: What the CAS Paper Actually Says
The CAS-Heilongjiang team's central claim is precise and consequential: this find 'could potentially rewrite the heavy in the south, light in the north pattern of rare earth resources in China.' That is not casual language for a Chinese state-affiliated research institution publishing in a peer-reviewed geoscience journal. It is a direct challenge to the foundational assumption that has shaped global rare earth geology for three decades.
The mechanism matters. In southern China, ionic adsorption clay deposits form when rare earth elements leach out of parent rock and bind to clay minerals over geological time. Extracting them requires injecting ammonium sulfate or similar leaching agents into the ground, a process that is expensive, generates significant toxic waste, and typically leaves roughly a quarter of the rare earth content unrecovered. It is also the process that has contaminated significant stretches of southern Chinese countryside, creating ongoing regulatory and reputational pressure even within China's own system.
The northeastern formations work differently. Alkaline granite rocks, subjected to repeated freeze-thaw cycles in Heilongjiang's and Jilin's sub-zero winters, physically disaggregate into loose sand and gravel. The rare earth elements are not adsorbed onto clay; they remain hosted in intact mineral grains, principally monazite, a phosphate mineral, and xenotime, a yttrium-rich phosphate that is one of the primary sources of heavy rare earth elements. Because the elements are in separate, physically identifiable mineral particles rather than dissolved onto clay surfaces, conventional mineral processing techniques, including gravity separation and flotation, become applicable. The team notes that 'mineral dissociation-type deposits show higher total rare earth element concentrations, with significant enrichment in light rare earth elements' relative to southern ionic clay deposits.
Critically, some Jilin samples showed 'particularly high abundances of heavy rare earth elements compared to neighboring regions,' including higher heavy REE values than equivalent sites in Heilongjiang. Xenotime, as a primary carrier, is the key indicator here. This is not just a cerium and neodymium story. If the Jilin heavy REE concentrations hold up at commercial scale, this deposit class has direct relevance to the dysprosium and terbium markets that are currently the single most constrained segment of the entire permanent magnet supply chain.
Price Action and Supply Context: Why the Timing Is Brutal for Western Producers
Let me put some numbers around why this discovery lands with particular force at this specific moment in the market cycle.
China controls approximately 69.2% of global rare earth mining output and roughly 90% of downstream refining and separation capacity. Its share of NdFeB permanent magnet production sits at 94%. For dysprosium and terbium specifically, China maintains a near-monopoly on separation capacity, with Benchmark Mineral Intelligence and CRU Group both estimating that China plus Myanmar will account for approximately 80% of global dysprosium and terbium supply through 2031. McKinsey's projection is even starker for the longer term: non-China sources will supply less than 20% of dysprosium and terbium demand by 2035, against a global magnetic REE demand stack that the same firm projects will triple from 59,000 tonnes in 2022 to 176,000 tonnes by 2035, implying a potential 60,000-tonne shortfall if diversification falls short.
Against that backdrop, Western governments and private capital have deployed $6.3 billion in non-China REE project investment in 2025 and a further $2.8 billion in Q1 2026 alone, per Benchmark Mineral Intelligence. The U.S. Department of Defense committed $400 million in direct equity to MP Materials and extended a $150 million loan for heavy rare earth separation capacity at Mountain Pass. USA Rare Earth closed a $1.5 billion PIPE in January 2026. The DOE issued a $19.3 million grant to USA Rare Earth for continuous ion-exchange separation technology, a development I covered in 'The Midstream Is the War.' Australia has 89 active REE projects competing for capital, with Iluka's Eneabba refinery and Arafura's Nolans project expected to contribute roughly 23,000 tonnes of refined oxides by 2028.
The investment thesis underpinning all of that capital is, at its core, a bet on a sustained structural price premium for non-Chinese supply. China's April 2025 export controls on seven heavy rare earth elements, which remain fully operative as of May 31, 2026, drove European dysprosium prices to as high as six times their Chinese domestic equivalent. That basis, six times, is the number that makes the economics of high-cost Western separation projects pencil. The CAS discovery is a direct threat to that basis. If China can produce heavy REEs from northeastern freeze-thaw deposits at lower cost than its own southern ionic clay operations, the floor under ex-China price premiums becomes less stable over a 5- to 10-year horizon, precisely the window in which most Western projects are targeting first production.
Institutional Activity: How the Market Should Read a Geological Paper
A geological paper does not move spot prices this week. Preliminary resource estimates for the Heilongjiang and Jilin formations are not expected until Q3 2026 at the earliest, and the path from academic publication to commercial mine involves exploration drilling, resource classification under JORC or equivalent standards, feasibility studies, permitting, and construction, a timeline measured in years, not months. Traders positioning in terbium or dysprosium derivatives on the back of this headline alone are getting ahead of the data.
However, institutional investors with 3- to 5-year rare earth equity exposure should be stress-testing their long book against this scenario: China's northeastern deposits prove commercially viable by 2029 to 2030, adding a lower-cost, lower-environmental-footprint source of both light and heavy REEs to a Chinese production base that already sets global price architecture. The counterpressure that scenario creates for non-Chinese project IRRs is not trivial.
The relevant comparison is China's 2022 playbook. When Beijing increased rare earth processing by 25% to suppress global market prices, the explicit objective was to make foreign producers uneconomic. That tactic ended Molycorp. MP Materials' $110 per kilogram NdPr price floor, negotiated with the Pentagon and described by MP Executive Vice President Matthew Sloustcher to Congress as taking China's 'price manipulation off the table,' exists precisely because the market has seen this film before. The floor provides protection for MP. It does not provide protection for the dozens of junior miners and explorers who have no offtake agreement, no government equity sponsor, and no price support mechanism.
Neha Mukherjee of Benchmark Mineral Intelligence is correct that 'the heavier the rare earths, the higher the cost of refining,' and that China executes that refining 'at a very low cost and very efficiently.' A deposit class that is cheaper to extract than ionic clay, combined with China's existing refining dominance, is a compounding cost advantage, not a simple one. Rowena Smith of Australian Strategic Materials may be right that near-term producers 'should be sitting comfortably' given current supply tightness, but the long-dated forward curve for projects targeting 2030-plus production dates is what requires reassessment.
The Diversification Race: Structural Headwinds That Predate This Discovery
It is important to calibrate the CAS discovery against a diversification effort that was already running behind schedule before this paper was published. The structural headwinds are not created by the discovery; they are amplified by it.
Lynas Rare Earths became the first non-Chinese producer to commercially separate heavy rare earths at its Malaysia facility in May and June 2025, a genuine milestone. But its combined quarterly output of dysprosium and terbium runs approximately 8 metric tons, against a global annual demand measured in thousands of tonnes. MP Materials produced a record 2,599 metric tons of NdPr in 2025, a 101% year-over-year increase, but the company remains unprofitable at current scale. Japan, which has invested in rare earth resilience for 15 years following China's 2010 embargo, still sources 76% of its rare earth supply from China. As Baskaran of CSIS noted, Japan 'still remains highly impacted by Chinese export controls' despite a decade and a half of policy effort. That data point is worth sitting with.
Michel Van Hoey of McKinsey summarized the structural reality bluntly: 'Meaningful diversification will take longer than many anticipate.' Mick McMullen, speaking at the U.S. Capital Access Forum in Singapore on March 10, 2026, was even more direct: 'China has been at this for more than 30 years,' and 'I'm not sure how long it takes to solve it, and whether it can get done in one administration term.' The Griffith University 2026 policy study reinforces the same point, finding that efforts focusing only on new mines, without parallel investment in separation, refining, and magnet-making, are structurally insufficient to reduce dependence.
The CAS discovery does not change the 30-year head start. It potentially extends it. China already controls 92% of global praseodymium-neodymium oxide refining capacity and 98 to 99% of dysprosium and terbium separation. A new, lower-cost deposit class in the northeast adds supply optionality to an entity that already dominates the downstream. The IEA estimates that $6.5 trillion in annual economic activity outside China is at risk from full reimplementation of REE export controls. That number does not shrink because China found new deposits. In the near term, it stays the same. In the medium term, if the northeastern deposits validate at scale, China's ability to sustain price pressure on competing projects for longer periods becomes more credible, not less.
The Investment Case: Key Levels and Positioning Framework
The trade thesis here is not binary. This discovery does not invalidate the Western diversification investment case in the near term, and it does not make the long-term case stronger. It introduces a specific, time-bounded risk that should be reflected in discount rates and project valuations for non-Chinese REE equity.
For near-term producers with secured offtake and government price support, principally MP Materials with its Pentagon $110 per kilogram NdPr floor and Apple supply agreement, the northeastern discovery is a background variable, not an immediate catalyst. The price floor insulates MP from the kind of market flooding that ended Molycorp. Investors long MP should monitor Q3 2026 preliminary resource estimates from the CAS-Heilongjiang team for any indication of deposit scale; a large-scale validation with commercially viable heavy REE grades would be the data point that warrants a reassessment of the long-dated NdPr forward curve.
For pre-production and exploration stage companies without offtake or government backstops, the risk-reward calculus shifts more materially. The 30% projected global supply shortfall by 2035 remains the primary bull case for the sector, but that shortfall assumes a specific trajectory for Chinese supply expansion. A new, lower-cost deposit class that produces both light and heavy REEs more efficiently and with lower environmental liability directly challenges the supply deficit math. Junior miners in this category should be pricing that uncertainty into capital allocation decisions now, before Q3 resource estimates arrive.
On price levels: terbium oxide near $890 per kilogram, a figure I referenced in 'The Midstream Is the War,' and dysprosium trading at up to six times its Chinese domestic equivalent in European markets represent the kind of basis that makes Western separation economics viable. Watch $700 per kilogram on terbium oxide as a threshold level; a sustained break below that level on renewed Chinese supply signals, whether from northeastern deposits or expanded southern quotas, would begin to pressure the project-level economics of the pipeline. The November 10, 2026 deadline for China's suspended second wave of export controls covering holmium, erbium, thulium, europium, and ytterbium remains the most important near-term binary event for the entire sector, and it now sits alongside a geological wildcard that the market had not priced two weeks ago.
The bottom line is this: China just told the market that its rare earth resource base is larger, geologically more diverse, and potentially cheaper to exploit than previously understood. That is not priced into the 5- to 10-year supply models that underpin current Western project valuations. It should be.