What is a Rare Earth Element?

There are 17 elements that are classified as REEs; these are scandium, ytterium, and the whole Lanthanide Group: lanthanum, cerium, praseodymium, neodymium, promethium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, and lutetium. These are all generally stable and nonradioactive elements, aside from promethium. Yttrium, despite not being a lanthanide, has very similar properties and so is grouped in. Scandium is actually more similar to the ferromagnesian elements - think iron and nickel - but in some circumstances it behaves a lot like an REE, so it gets to join the club.

The reason they’re called “rare earth elements” is due to the circumstances of their discovery. The only place in the world where they could be found in the 19th century was a town called Ytterby in Sweden, thus making them rare, and in French, an oxide of an element is called the “terre” or earth of that element, and these elements were found as oxide minerals. However, since then, we’ve found that some of these elements, such as cerium, are actually more common than lead in the Earth’s crust - and even the rarest ones, like lutetium, are orders of magnitude more common than gold.

REEs tend to have quite high ionic radii, which makes them difficult to fit into minerals when they’re forming when magma cools - in the business, this is a property called incompatibility. This means that as a magma cools down, such as in a magma chamber or some kind of intrusion near the surface, the parts that solidify and form mineral crystals first “reject” the REEs, kinda like unpopular kids not being picked for sports teams in school by team captains, leaving them behind in the remaining liquid rock. This continues until eventually a forming rock is “forced” to pick them up - there’s no other players for the team captains to choose for their teams.

The type of rock that forms at the end of this process is called a pegmatite. You might see the “ite” at the end and conclude that it’s a mineral with some kind of chemical formula, like other minerals like “calcite”, but it’s actually purely a description of texture - pegmatites are defined as being coarse-grained and having large crystals of a certain minimum size. Ytterby, the first mine that REEs were found in was, correspondingly, a granite pegmatite.

The Minerals

The principal minerals that REEs are found in are monazite, bastnaesite, xenotime, and eudialyte. These minerals actually do have chemical formulas - and if you’re curious, two of them are phosphate minerals - but all you need to know as a layman is that these are the main minerals which contain a range of REEs.

Monazite was the first one to be exploited, and typically contains a mixture of Light Rare Earth Elements (LREEs), which are the ones found on the left side of the lanthanides. However, it’s quite a heavy mineral, about twice as dense as a general granite, and resistant to weathering, so it tends to separate out from other minerals in the granite and concentrate into placers - which may be a familiar term if you know about the gold rushes a couple centuries ago.

Bastnaesite also tends to contain LREEs, though it does also have significant amounts of yttrium and small amounts of HREEs (the right side of the lanthanides). In the modern day, bastnaesite has taken monazite’s throne as the primary LREE ore. It’s widespread but doesn’t occur in large quantities, and occurs in a greater variety of rocks than just pegmatites, such as carbonatites (igneous rocks that have a lot of carbonate minerals; imagine melting and resolidifying limestone and chalk).

Xenotime contain a relatively larger proportion of HREEs. It forms in pegmatites and also metamorphic rocks - which is the rock class in between sedimentary and igneous rocks that have been “cooked” to high temperatures but not completely melted and reformed by heat and pressure, giving them interesting properties and appearances that we don’t have time to talk about.

Eudialyte is easily dissolvable in acid and tend to be sources of other elements, like zirconium, but also contain some REEs.

There are numerous other minor minerals that contain one or two REEs, but it is typically not economical at this time to extract them.

What are they used for?

There’s no way that I can be comprehensive nor particularly detailed with this list or we’d be here all day, but I’ll go through each element in turn and give a few interesting things that they’re used in. Feel free to skip this part.

The Balance Problem

I couldn’t find a place to put this section in the main narrative but it’s worth mentioning.

The “balance problem” is essentially just the observation that because not all REEs are used equally - for example, the big superstar is neodymium, with others of similar abundance not used quite as much - and because ores containing REEs often will contain a mixture of lots of different REEs, then you inevitably end up with a situation where you’re stockpiling the less used elements in larger and larger quantities as the most used ones get traded away. The price of stockpiling these excess REEs goes up and up and the market price for them goes down as supply increases, and because a mining company must (under capitalist logic) make a profit, this introduces a pressure that makes them not want to mine as much ore (due to stockpiling costs) but simultaneously need to mine more ore (because often the minor elements are what the world economy needs more of), so it’s difficult to achieve an overall balanced market.

There are five proposed solutions to the problem:

1. Find uses for the elements that aren’t used very much.

2. Find substitutions for elements that are used a lot.

3. Mine ores outside of the big three of monazite, bastnaesite, and xenotime that contain only one or two REEs (but this is, under capitalism, less economically feasible, and even not under capitalism, still introduces big environmental concerns to other places)

4. Recycle REEs more.

5. Reduce their use as much as possible by using them more efficiently.

Part 2.

While I cover the uses of REEs more broadly in the introduction, it’s briefly worth mentioning them to give a context to upcoming events in the story. Several REEs have had their time to shine as their properties were uncovered and applied in technologies. I’ve mentioned europium already and its use in CRTs, though initially it had no use and was stockpiled. Lanthanum was used in the optical glass industry in the 1960s. Praseodymium and neodymium were used to make alloys for coloring glass. Samarium, like europium, initially had no uses and was stockpiled until it was discovered you could make powerful magnets by combining it with cobalt.

Magnetic technology is today the dominant use of REEs. In fact, I would wager that for most people the only REE they’ve heard of is neodymium, for its use in magnets (The famous “neodymium magnet” is actually usually an alloy of neodymium, iron, and boron, but I’ll continue calling it a “neodymium magnet” for simplicity.) Magnets made with REEs are stronger than those made of your more traditional magnetic materials like iron, and can be made much, much smaller, which is necessary in civilian technologies like computers and also military technologies like in missiles and aircraft.

In the 1980s, General Motors and Hitachi simultaneously developed neodymium magnets and fought a battle of patents, eventually deciding to create two different designs of magnets with each company taking one. In 1986, GM developed a division to produce their magnets called Magnequench. In 1996, two Chinese corporations joined forces with an American investment firm to acquire Magnequench, and the United States was like “Alright, so long as you keep Magnequench in America for a minimum of five years.”

Five years and one day later, in 2002, Magnequench’s entire operation in the United States disappeared, all equipment vanished, all employees were laid off, and it moved to China. At the time, nobody particularly cared - but it eventually dawned on the Americans that a big mistake had been made. In 1998, 90% of global magnet production was in the West, largely the US and Japan, though Europe had a relatively minor role too. In 1996, China was outputting just 2600 tons of sintered magnets, the type that Hitachi took. By 2007, China was outputting 80,000 tons.

Remember Molycorp, the owner of the Mountain Pass mine in California? In 1978, Unocal, the Californian oil business, purchased Molycorp, and production expanded over the decades, eventually producing neodymium in the late 1980s. In 2005, the China National Offshore Oil Corporation (CNOOC) submitted an $18.5 billion bid for Unocal, outbidding Chevron. A media and political frenzy commenced, and eventually CNOOC’s bid was withdrawn and the company acquired by Chevron. A close call for the US rare earth industry.

Another area of rare earth production at this time was Australia. The Lynas Corporation, an Australian company, planned to build a mine at Mount Weld in western Australia, but construction was suspended due to funding problems. In 2009, a Chinese corporation proposed that it would invest about $250 million in return for a 51% stake in Lynas. After much delaying and bartering on the part of the Australian government, including to try and reduce China’s stake below 50%, the Chinese withdrew their offer. Instead, they invested in another Australian rare earth developer, called Arafura Resources, based in the Northern Territory of Australia, which hasn’t yet begun production (though I don't think they have a majority stake).

In 2009, the United States began a Government Accountability Office study to determine the risks of dependency on China. In September 2010, a maritime incident between Japan and China led to REE exports to Japan being de facto cut off. To do so officially would violate WTO rules, so it could not be de jure. Here is Naked Capitalism’s own reporting back in 2010 about this issue. (for those unfamiliar with the website, it is source of news and analysis that the News Megathread often uses):

Two aspects of the ban that don’t get the mention they warrant. First is that most US rare earth purchase go through Japan (Japan produces intermediary and finished goods using Chinese rare earths). Second, the ban is not across the board, but merely on raw materials. China is willing to sell manufactured goods with rare earths as significant content. Thus this change is also designed to try to use the rare earths near monopolies to force foreign buyers to more manufactured products from China, rather than simply raw materials.

One has to wonder whether the restriction was something China planned to do regardless, and the diplomatic dispute provided a useful cover.

Chinese officials have previously argued that the status quo was harmful to their reserves and their environment, with Chao Ning in 1996 saying:

“China cannot afford to continue to carry the burden of supplying the world, from a strategic, environment and economic point of view,” Chao said…

“China is not the only country that has these deposits, but it has been carrying the lion’s share of the supply in more than a decade, at the cost of quickly depleting its own resources and hurting its environment,” Chao said.

By 2009, despite China controlling over 90% of REE production, it had only 30% of global reserves, as other sources were found in other countries. So on the one hand, given what we’ve already seen are the tremendous consequences of REE mining on the environment, it is understandable that China would feel exploited by the rest of the world, especially as the reason China had this near-monopoly at all was due to low prices due to low labor costs and so on. On the other hand, given Deng and other Chinese leaders’ comments, it appeared they seemed quite fine with bearing this burden for the geopolitical importance it gave to China - and they seemed quite happy to continue investing in foreign companies to ship their REE minerals to China for processing. I would need to do more research than my free time allows to give a good answer to this conundrum, but it seems plausible that both of these lines of reasoning were true, and like how we see in Middle Eastern oil-producing states today, there appears to both be a genuine understanding that this situation of massive oil production cannot continue indefinitely if the state is to survive in the long term as reserves will eventually run out, but actually acting on that reality is difficult due to the massive potential profits and the geopolitical role of OPEC.

Anyway. Japan was China’s largest market for REEs, needed for high-tech goods. Japan released a detained Chinese fishing captain and backed down, humiliating the Japanese, but the broader damage to China was done - the prices for REEs surged around the world. Again, I think Yves at Naked Capitalism says it all:

Even if China wins this round, this is an extraordinarily heavy-handed and short-sighted move. China is telling the world loud and clear that it is an unreliable partner. China is conducting this ban apparently without issuing formal regulations, which would subject it to WTO sanctions. But for it to think this action won’t lead to retaliation is naive, when political pressures abroad make China an easy target.

And it was rather counterproductive. Whatever China’s motivation was, it simply did not have the strength in 2010 to compel other states to its whim in any meaningful capacity. Worldwide interest in restarting old, and developing new mines increased as the prices of REEs did. To be clear, the interest did not start from zero - several countries, like the US, Japan, Australia, Canada, South Africa, and Kazakhstan were already working on REE mining leading up to 2010. But China’s attempt to flex its muscle lead to a concerted effort against it. By 2013, Molycorp and Lynas were delivering REEs to global markets. Rhodia in Europe joined the REE blitz and hundreds of companies around the world raised money for new mining projects. In fact, part of the reason why I’m able to write this essay in the first place is because of the reports that were written during this very frenzy.

Needless to say, when looking through material writing this essay, the Western sources get very smug around this point. A typical sentence from one of these reports might as well be boiled down to “We’re showing those communists what-for! You can’t get away with manipulating global markets for your own domestic gain!” without even a shred of self-awareness. But you may have noticed that China’s share of world REE markets has not been brought to shambles. By my count, China made up something like 70% of REE mine production in 2022, with the next largest competitor, the USA, at 14%; and god knows how much foreign ore is processed in China. And that is because the story is not yet over.

China quickly ended its unofficial embargo to Japan, but the damage was done. Regardless, China began raising its export duties from 10% to 15% and then 25% in 2011, in order to retain them for their own market. The same happened to ferro-alloys, containing more than 10% REEs. Chinese exports of REEs subsequently dropped, and prices as much as quintupled; this whole saga is called the REE Crisis.