UNIVERSITY PARK, Pa. — A newly discovered protein naturally houses an unusual binding site that can differentiate between rare earth elements, and researchers at Penn State have made it even better. Rare earth elements are key components used in everything from modern tech to gasoline production. The protein, called LanD, enriches neodymium and praseodymium over other similar rare earth elements (REEs) and has the potential to revolutionize industrial mining, researchers said.
Scientists at Penn State, led by professor of chemistry Joseph Cotruvo, Jr., recently published their LanD discovery in the Proceedings of the National Academy of Sciences.
"Each rare earth element has specific properties that make it useful for different applications, yet they are notoriously difficult to separate from each other," said Cotruvo, who has filed a patent application related to the work. "Current industrial methods are inefficient and require heavy use of toxic chemicals, so a protein-based method for rare earth mining could make this process more efficient, greener, and less expensive."
Close to 80% of the United States' REE supply is imported, according to the United States International Trade Commission. Cotruvo explained that there is plenty of domestic raw material — including recycling old tech and industrial byproducts — to source REEs, but not all REEs are of equal value and application.
A more effective separation approach could help secure a national supply of REEs. The 17 REEs, including 15 metals called "lanthanides," are commonly divided into "light" and "heavy" groups, with the light REEs being far more abundant. Unfortunately, however, the most common light REEs, lanthanum and cerium, have little value, whereas the other light REEs, praseodymium and (in particular) neodymium, are much more valuable. Neodymium is a critical component of permanent magnets used in smartphones and renewable energy machinery like wind turbines, and praseodymium is often combined with neodymium for these applications.
Cotruvo's lab previously identified another protein, LanM , that binds to all REEs with high specificity over any other metal. It does this in a fashion similar to a lock and key mechanism, with the protein being the lock and the REE a key. When the protein binds a REE, it undergoes a change in shape analogous to the key turning in the lock. The LanM proteins studied to date are very good at differentiating between heavy REEs, but they do not do well separating the light REEs, akin to a keyhole that fits a few different keys.
The newly discovered LanD protein, however, has improved separation abilities among the light REEs that are as good as, if not better than, current industry practices, Cotruvo said. With a unique, never-before-seen binding site — where the metal "key" can lock into the protein — LanD's natural REE separation abilities can be engineered to be even more efficient, offering new hope for a greener rare earth element mining industry, he said.
"Current efforts are concentrated towards optimizing REE separation to make it less labor and material intensive," Cotruvo said. "But this organism, Methylobacterium extorquens, a bacterium found abundantly in nature, makes proteins that seem to have already solved the problem."
Methylobacterium extorquens is a species of bacteria known for its ability to grow on one-carbon compounds like methanol, and prefers to use specific REEs, mostly lanthanum and cerium, to support that growth.
When Cotruvo discovered LanM as the first high-affinity, high-specificity REE-binding protein six years ago, it was not clear why LanM needed to bind REEs so tightly in the cell. The discovery of LanD suggested an answer to that question: the two proteins work in tandem, with LanD binding to the lanthanides that the bacterium takes up but doesn't need and delivering them to LanM, where they are sequestered. Those lanthanides, while not important to the bacteria, are the ones that are most important to tech production, Cotruvo said.
"The bacterium can take up a broader group of lanthanides than the small subset that it prefers to use, so it needs a way to prevent those undesirable lanthanides from interfering with the functions of the desirable lanthanides in the cell," Cotruvo said. "LanD and LanM appear to work together to do this sorting, which explains why the previously identified LanM protein is good at lanthanide separations in general."
He added that LanD, with its unique binding site, is much better for the light REEs specifically.
"LanD conveniently binds best to neodymium, which is by far the most valuable of the light REEs," Cotruvo said. "While the naturally occurring LanD protein exhibits a preference for neodymium, we re-engineered it to more effectively be able to extract neodymium from a mixed solution of light REEs, disfavoring the other REEs that are of lesser value."
The researchers found that engineering the LanD binding site allows separations yielding the desired neodymium and praseodymium to become much more effective. In future applications, the researchers said they hope to be able to whittle down the protein size and increase the preference of this binding site even more — and implement it in a larger-scale separation. The site can serve as the starting point for chemists and engineers to develop highly specific proteins to perfect sorting of other tricky-to-separate elements, Cotruvo said.
Furthermore, because LanD and LanM specialize in separation of different REEs, they could be used together in a process to separate complex REE sources like ores.
"The LanD protein is a promising way to improve REE separation practices," he said. "And we're working on making it even better, to pave a path toward more effective, greener rare earth mining."
Paper co-authors include Wyatt Larrinaga and Jonathan Jung, graduate students in chemistry; Chi-Yun Lin, postdoctoral researcher in chemistry; and Amie Boal, professor of chemistry and of biochemistry and molecular biology.
The U.S. National Science Foundation, the DARPA Environmental Microbes as a Bioengineering Resource program, a National Defense Science and Engineering Graduate fellowship, the National Institutes of Health, and the Jane Coffin Childs Memorial Fund for Medical Research funded this work.
