Dysprosium and terbium: the heavy rare earths that decide electric motors

A neodymium magnet is the strongest permanent magnet in commercial use, and it has one weakness: it gets weak when it gets hot. Dysprosium and terbium are what let it hold its strength at the temperature inside an electric car's motor.

The property that matters is coercivity, which is how hard a magnet resists being demagnetized. It is measured in tesla. At room temperature a typical commercial grade holds about 1.2 tesla. At about 160 °C, which is roughly where the motor in an electric car runs, it holds about 0.2.³

That is not enough margin. You put your foot down, the motor's own field pushes back on the magnet, and the magnet permanently loses some of its strength.

Swap a little dysprosium or terbium into the mix and the magnet survives the heat. That is why a traction motor magnet needs them and a fridge magnet does not. Neither does a stereo speaker.

The industry found a way to use far less dysprosium. The hottest motors still need some.

The method is called grain boundary diffusion. Anyone with a materials science background will remember what a misery it is to work through in a classroom. You coat a finished magnet with the heavy element and bake it. The dysprosium or terbium travels along the boundaries between the magnet's grains and forms a thin shell around each one, a bit like chrome plating on a bumper. The expensive element ends up only where it does the work.

It cut the amount of heavy rare earth needed by roughly a factor of ten. A traction motor magnet made the old way, with the dysprosium mixed all through it, could carry up to 11% dysprosium by weight. With diffusion the addition is under 1%.⁴

Then it hit two limits. The coating only reaches about 4 millimeters into the magnet, so a thick magnet does not get the full benefit.⁵ And a magnet that starts with no dysprosium in it at all tops out at about 2.1 tesla of coercivity after diffusion. The hottest applications need about 3 tesla.⁶ For those motors the heavy element still has to be in the magnet from the start.

So the question becomes where the heavy element comes from. The best mines outside China hold very little of it.

Heavy rare earths are a sliver of the rock in most deposits. Digging more of it outside China mostly brings up more of the cheap light elements.

Look at Mount Weld in Western Australia, the best rare earth deposit outside China. Lanthanum and cerium, the two cheapest elements on the board, make up more than 68% of its reserves. Dysprosium and terbium together make up 0.78%.⁷ Lanthanum sells for about $1 a kilogram and terbium for $1,010. Both come out of the same hole. What differs is how much of each the rock holds. You can dig up a mountain and come away with a bucket of terbium, and a magnet that has to survive heat needs what is in that bucket.

America's own mine says the quiet part in its filing. The resource statement for Mountain Pass in California counts cerium, lanthanum, neodymium, praseodymium and samarium. On the heavy rare earths, MP Materials writes: "While other REE, often referred to as HREE, are present in the deposit, they are not accounted for in this estimate due to historic data limitations."⁸ That is after more than 70 years of mining at the site.

The same filing lists what MP Materials could lose if its agreements with the Department of War were impaired. One item is the government's position to "assist and facilitate our sourcing of heavy rare earth feedstock".⁹ The largest rare earth producer in the United States is telling its investors that its heavy feedstock is a question it needs help with.

This is the number that made me want to make the episode. By the count of Adamas Intelligence, China has refined over 99% of the world's dysprosium and terbium in recent years.² I name Adamas because this is a commercial estimate, and readers should know where it comes from.

For magnets as a whole, China made 94% of the world's sintered neodymium magnets in 2024, according to the IEA.¹⁰ That is a monopoly. Over 99% of the ingredients that let a magnet survive heat is something tighter, and you can see it in the price. Why separating the elements is so hard in the first place is covered in are rare earths rare?

On 19 February 2026 a kilogram of dysprosium oxide cost $207 inside China and $1,125 in Europe. That is a premium of 443%. On the same day, the neodymium and praseodymium oxide that makes up the body of the magnet carried a premium of 8%.¹¹

For neodymium, Europe pays 8% more than China, which I read as shipping and paperwork. It is one market. For dysprosium, Europe pays 443% more, because you cannot buy it in China and ship it out without a license. Inside China it is priced like a normal product. Outside China it is priced like something you cannot get.

The license dates from 4 April 2025. That day China put seven elements under export license: samarium, gadolinium, terbium, dysprosium, lutetium, scandium and yttrium.¹² Apart from samarium, that is the heavy end of the table. How the controls widened after that, and when the current pause runs out, is in China's rare earth export controls, a dated timeline.

Inside China, dysprosium is priced like a normal product. Outside China, it is priced like something you cannot get.

China separates almost every gram of the heavy elements, and much of the rock they come from is dug across its border in Myanmar. China has plenty of the light rare earths. Of the heavy ones it runs short, and it gets much of them from Myanmar.

Two clay regions, one of them beyond its government's reach

The ore that carries dysprosium and terbium is weathered clay. It sits in two places: seven provinces of southern China, and Kachin State in northern Myanmar.¹³ The IEA estimates that Myanmar was over 40% of the world's mined dysprosium and terbium in 2024.¹⁴ Myanmar has published no resource statement and no audited production figures, and in the USGS tables its reserves read "not available".¹⁵

The mining is done by informal operators. They drill pipes into a hillside, drip ammonium sulfate solution in for months, and collect what drains out at the bottom in tanks.¹⁶ The USGS counted those tanks from satellite imagery, in a data release published in September 2025. It found 6,655 in Myanmar against 3,089 in China, and it attaches a warning to its own data: only 2 of China's 7 clay provinces were mapped.¹⁷

Those mines do not answer to the government of Myanmar. In October 2024 the Kachin Independence Army took the mining towns of Chipwi and Pangwa and put a levy on exports. Exports were disrupted, and China's imports of rare earths from Myanmar fell sharply until trade recovered in the spring of 2025.¹⁸

The quota China never raised

Then there is the number that reframes the export controls. China sets an annual ceiling on its own mining, split between hard rock and the clay that carries the heavy elements. From 2018 to 2024 the hard rock quota rose 149%. The clay quota has stayed at exactly 19,150 tons since 2018.¹⁹

So China has expanded its rare earth mining enormously and has left its own heavy mining flat for the better part of a decade, while it depends on a Myanmar border region it does not control. The border trouble came first. The April 2025 controls landed on the heavy elements a month after Myanmar's exports were disrupted.

The Royal United Services Institute (RUSI) reads those controls as China "ensuring its domestic access to heavy rare earths" and "restricting unnecessary exports".²⁰ That is their reading, and I find it fairly persuasive.

If you follow this layer, two numbers will tell you whether the heavy grip is loosening before any headline does.

The first is Lynas Rare Earths. Its plant in Malaysia is the only proven separator of dysprosium and terbium outside China, and it produced 62 tons of the two in the financial year to 30 June 2026.²¹ Nobody publishes audited figures for world use, so every share you see is an estimate. Watch whether Lynas breaks 30 tons of dysprosium and terbium in a single quarter. I think that is very doubtful in the near term.

The second is MP Materials. Its resource statement leaves the heavy rare earths out, and its own risk factor ties its heavy feedstock to help from the government. Watch whether the company says out loud where that feedstock will come from.

The companies on both sides of this, with their risks, are in rare earth companies outside China. I hold MP Materials, so read that one with the disclosure below in mind. The full argument, with the filings on screen, is in the video on The Investment Thesis. The question I ask of every layer is where the bottlenecks are and who sits on them. For the heatproof magnet, the answer today is China, with Myanmar's clay behind it.