Protein-powered rare earth sorting: a scientific breakthrough

Rare earth elements (REEs) are vital components that power our modern world, from smartphones to renewable energy technologies. Extracting these elements efficiently has long been a challenge, one that requires massive volumes of toxic chemicals, not to mention huge amounts of energy. 

What if there was a better way? A team of researchers at Pennsylvania State University, led by associate professor Joseph Alfred Cotruvo Jr, think they might have found it. The team has successfully engineered a protein that is able to sort and separate REEs quickly and cleanly. 

The team’s research was recently published in the journal Nature – and it is already showing significant potential for use in the mining industry. 

Here, we speak to Cotruvo Jr to delve into the fascinating world of REE extraction and how looking to the natural world could revolutionise the future of the industry.

Joseph Alfred Cotruvo Jr: I am an associate professor in the chemistry department at Penn State [University]. By training, I would say I am a biochemist, a bio-inorganic chemist. So, I have been interested in some time in the roles of metals and biological systems.

Joseph Alfred Cotruvo Jr: ​​​​​​​I feel like I must give a little bit of context, because it is the continuation of work that my lab and our collaborators have been doing since I started my lab, which was in 2016. Over the past five years or so, we have been exploring these naturally occurring lanthanide binding proteins, this family of proteins that we call lanmodulins. 

What this particular study was focused on was a new member of that family and showing how it is able to distinguish between different rare earths, both biochemically and structurally. That mechanism included dimerisation of the protein, which adds a new dimension for differentiation of these metals. Then we utilise this protein to separate heavy and light rare earths; that is the elevator pitch from a more chemistry angle.

Joseph Alfred Cotruvo Jr: ​​​​​​​Just so we are on the same page: the rare earths comprise the lanthanides, of which there are 15. Then there are two other rare earths – yttrium and scandium. We have dealt with all of those rare earths, but this particular paper only talks about the lanthanides. 

Philosophically, the reason why I started working on lanthanide recognition in biology in the first place was that it is a very new idea; it had only been discovered a couple of years before I started my lab. So, there is a lot to be learned. 

What was interesting from a chemical perspective is we know that the organisms, at least those that have been discovered to use lanthanides, only use particular ones, and those are the light lanthanides. In essence, they can carry out at least some sort of separation process, and we know biomolecules in general are quite good at differentiating between very similar-looking metals. 

So, the question is, if we can learn about how biology is differentiating between the rare earths then these are some of the most similar metals to each other on the periodic table. I do absolutely feel that there is the potential for similar ideas being transferred to other valuable metals. 

We have played around with the lanthanide protein; we had another paper that came out in late 2022 that was making it a much better binder for manganese. That data suggests that we might be able to do manganese, cobalt and nickel separations. 

That work is not as far as long as the lanthanide work, but the data is promising. A lot of people, when I talk about this, say: “Oh, proteins, they are so unstable and they are so big, why would you ever use these molecules for these purposes?”. It is true that some proteins are unstable, but not all of them. So, that is not really a concern. 

These molecules are bigger than many of the ligands that are used in mining and other industries, but they also tend to be much, much more selective. I think that selectivity is an important benefit; if you need to use the ligand twice to get really good separation, instead of 20 times or ten times, then even if it is bigger, as long as you can use it multiple times, as long as it is recyclable, it is still potentially valuable. 

The other thing about proteins is you can engineer them very easily. You can make mutations, you can do this in high-throughput fashion and test your ligands. So, you can alter selectivities that way.

Joseph Alfred Cotruvo Jr: ​​​​​​​There are many ways that you could do that. Traditionally, there is biomining, which, I guess, is what we are doing. A lot of people think of biomining as just throwing in natural or engineered organisms into some sort of feedstock – it could be liquid and could be mine tailings. I think there is value in that in certain cases. 

However, we can also be inspired by what those organisms are doing and take the segment of their very complex physiology that is doing the thing that we want. Specifically, the molecules that they make – they could be small molecules, they could be proteins – and use those molecules more efficiently for the extractions and separations that that we are interested in. It is conceptually not all that different from what is already done. It is just using non-traditional ligands.

Joseph Alfred Cotruvo Jr: ​​​​​​​We have filed patent applications from the beginning of working on this project. Some of these applications are just Penn State, some of them are with our collaborators from Lawrence Livermore National Lab, and I certainly give them a lot of credit, as well, for some of the application-driven work. So, those patents are still pending, but many of them have been licensed by a start-up to move the work forward. 

There are still many challenges that we have to solve and scale up, but it is a promising start. From the research side, too, we are continuing to improve. Our focus is really on separations – I feel like we have demonstrated the extraction side, and now there are many engineering problems to be solved – but from the chemistry side, which is my expertise and primary interest, I think we have a really promising start. 

One thing that we are less good at is the separation of adjacent lanthanides. So that has really been our focus, identifying new proteins, engineering proteins that we have already looked at to try to get larger separation factors.