A cleaner route to purifying rare earth elements

Rare earth elements like lanthanum, neodymium, and dysprosium are used to build the electric motor in your car, the LED lights in your house, and the MRI machine at your doctor’s office. But first, they have to be mined and separated from each other. Historically, that purification has been a difficult, costly, process, relying on huge amounts of toxic chemicals.

Now, researchers in the lab of Assoc. Prof. Chong Liu at the University of Chicago Pritzker School of Molecular Engineering (UChicago PME), working with colleagues at Northwestern University and Argonne National Lab, have discovered a cleaner method to separate rare earth elements from each other.

The new approach relies on a layered form of manganese oxide—a mineral material with the right size layers to allow ions to slip in and out and to differentiate rare earth elements.

“This is the first time that people have used electrochemical intercalation and harnessed the structural characteristics to separate similar lanthanides, which are intrinsically very hard to separate,” said Liu, senior author of the new study, which published in Nature Chemical Engineering. “What’s also valuable is that we provided a lot of new understanding of how rare earth ions are interacting with this material and how we can manipulate it to better selectivity.”

“This kind of separation is competitive with other rare earth separation methods, but it’s done in water, without organic solvents,” said George Schatz, professor of chemistry at Northwestern University and a co-author of the study. “That’s a difference that could actually matter at manufacturing scale.”

This is the first time that people have used electrochemical intercalation and harnessed the structural characteristics to separate similar lanthanides, which are intrinsically very hard to separate.

Assoc. Prof. Chong Liu, senior author of the study

Squeezing elements through channels

The 17 rare earth elements—including the 15 lanthanides, plus scandium and yttrium—rarely occur alone. They’re almost always mined together and chemically they’re nearly identical, with only tiny differences in ion size and acidity differentiating each one. Pulling them apart typically requires custom-built molecules and large amounts of acid, which is used to strip each element off those molecules. 

“Rare earths always come mixed together, whether they’re in an ore or in a waste stream, and separating them from each other is a second, very challenging step even after you’ve pulled them away from everything else,” said UChicago PME graduate student Jiadong Liu, a co-first author of the new paper. 

Chong Liu and her colleagues knew that one of the differences between rare earth ions was the size of the water shell surrounding each one when they are dissolved in solution. Lighter rare earths like lanthanum have larger first water shell, while heavier rare earths like dysprosium have a smaller first shell. 

Taking advantage of that size difference, Chong Liu’s group engineered manganese oxide so that the gaps between its stacked layers were only a few water molecules wide. Then, they squeezed raw mixtures of rare earth elements inside. 

The approach divided the elements into two groups. Heavier lanthanides with smaller water shells stuck in the channels more tightly. Lighter lanthanides with larger shells pushed the layers apart, loosening their grip. 

To confirm what was happening at a molecular level, the UChicago PME team collaborated with Schatz’s group at Northwestern to run quantum mechanical simulations using a method called density functional theory, which predicts how atoms arrange themselves and interact based on the underlying physics. They also worked with Argonne scientists to obtain experimental X-ray data. 

“It was incredibly rewarding to see how closely our density functional theory calculations matched the synchrotron X-ray measurements,” said co-first author Woo Cheol Jeon, who conducted the research as a postdoctoral researcher in George Schatz’s lab at Northwestern. “The calculations let us see, atom by atom, how each rare earth element arranges its hydration shell inside the confined channel, which experiments couldn’t resolve directly.”

Fine-tuning the purification

While the new process separated the heaviest and lightest rare earths, some of the most useful elements still behaved too similarly to separate. To fine-tune the purification so that it could differentiate similar pairs of rare earth elements, the research team used an electric current and added magnesium ions. 

The magnesium acted as a scaffold, holding the manganese oxide channels to their designed spacing, even when rare earth elements tried to expand it—an effect called pinning. Now, rare earth ions showed differences in binding to the layered material even when they were extremely similar. 

“Even elements that behave almost identically will still try to expand the material to make room for their water molecules,” said Siqi Zou, PhD'24, co-first author of the study and former UChicago PME graduate student. “By pinning the channel so it can’t expand at all, we forced that small difference in behavior to become a much bigger difference in how strongly each element binds.”

With the addition of magnesium, the enrichment of neodymium over lanthanum jumped from a 1.6-fold difference to a 5.4-fold difference. After two cycles of purification, researchers could get a neodymium sample that was 97% pure. Similar improvements were seen for other rare earth elements. 

This kind of separation is competitive with other rare earth separation methods, but it’s done in water, without organic solvents. That’s a difference that could actually matter at manufacturing scale.

George Schatz, professor of chemistry at Northwestern University and a co-author of the study

A cleaner future

The new purification method could ultimately point toward different ways of thinking about where rare earth processing happens, the researchers said. 

“Right now, rare earth ores get mined all over the world, but almost all of them end up being sent overseas for processing,” said Schatz, co-author of the study. “A method like this, that just uses water and electricity instead of organic solvents, is the kind of technology that could actually change how and where that processing gets done.”

The method, however, isn’t yet ready to be scaled up to replace industrial purification. Still, it points to a broader design principle: tuning a channel’s width can determine which ions a material prefers, even for ions that differ by a fraction of an angstrom. 

Chong Liu’s group is now testing the approach against more of the 15 lanthanides, while Schatz is refining his computational models, aiming to explain the pinning effect more quantitatively. He is also using the same modeling approach developed for this work to study other materials. 

Citation: “Pinning Angstrom-size solid ionic channel for the separation of rare earth elements,” Zou et al, Nature Chemical Engineering, July 21, 2026, DOI: 10.1038/s44286-026-00418-8