Back to top
Low-grade feedstocks, such as coal byproducts, acid mine drainage, and electronic waste (e-waste), are riddled with materials ripe for reuse, particularly rare-earth elements (REEs). REEs—the 15 lanthanides, plus scandium and yttrium—are irreplaceable elements in mission-relevant applications for national security devices and clean energy technologies. The materials are also used in consumer products such as smartphones, LEDs, and digital camera lenses. Ironically, REEs are abundant, but they do not form concentrated ores in nature, rendering their purification from other raw materials difficult. Their similar chemistries also incline them to occur as a mixture of REEs, making individual ones hard to separate.
Only one operational REE extraction mine exists in the United States, placing the need for creativity in utilizing alternative sources. A historical approach to extracting REEs from low-grade secondary feedstocks, involving solvent extraction with chemically synthesized ligands, is expensive, inefficient, and environmentally damaging. The United States lacks control over facets of the REE supply chain because of this long-standing insufficiency in separating them, and the high price of externally sourced REEs, which come predominantly from China, is a source of vulnerability. A domestic supply is required to ensure that national security needs can be reliably met.
In 2014, microbes that use lanthanides as an essential nutrient were discovered, introducing the relevance of biology to rare-earth chemistry and separation. These microbes secrete ligands that enable them to gather and take in lanthanides from the environment. Many studies have since sought to understand and exploit this mechanism for biotechnology applications. The REE-binding biological ligand lanmodulin (LanM) from the bacteria Methylorubrum extorquens (Mex-LanM) is, to date, the most promising, sustainable alternative for REE extraction and separation. The application of the protein LanM for this purpose is an example of biomining, a term broadly used to describe the use of biology in mining operations. Livermore has been at the forefront of significant REE biomining advancements with the goal of improving the economics of a large-scale, domestic REE supply chain. Through collaborative research efforts, Livermore scientists are enhancing LanM’s selectivity and prospecting new REE-binding proteins at scales currently achievable only at the Laboratory.
A More Economical Approach
A main barrier to a domestic supply of REEs is the economic impracticality of existing extraction and separation methods, especially when targeting low-grade ores. For most industrial applications, high-purity individual REEs are required, but typical solvent extraction cannot produce such pure products without tens to hundreds of steps. The legacy process also fails to economically exploit low-grade feedstocks, an essential domestic REE source, particularly for the rarer, heavy REEs such as dysprosium.
Livermore researchers have utilized the exceptional selectivity of the LanM proteins to significantly reduce the number of separation steps, while working with feedstocks as dilute as one or two percent REEs. “We are taking advantage of the selectivity of biology to try to do things in a more streamlined, less energetically and chemically demanding, and less capital-intensive process,” says Dan Park, Livermore staff scientist and group leader for Systems and Synthetic Biology (SSB).
Ziye Dong, SSB staff scientist, has pioneered column-based separation to leverage LanM’s REE adsorption behavior. In this process, columns contain a resin of LanM immobilized onto polymer microbeads. Low-grade REE feedstock leachate flows through the columns, the REEs present in the feedstock bind to the resin on the way down, and non-REE metal ions flow through to the bottom without interacting with LanM. To be economically feasible, the resin must be reusable for thousands of cycles. LanM resin has exhibited high stability for reuse, and its production is inexpensive and readily scalable. Dong’s current column work supports up to 100 milliliters of resin and is poised to be scaled up significantly. The process has been successfully applied in several separation efforts by the SSB team to date.
An early success using the column-based method for REE separation was spearheaded by Dong; Park; Yongqin Jiao, deputy division leader for the Biosciences and Biotechnology Division; and Gauthier Deblonde, a staff scientist in the Nuclear and Chemical Sciences Division. This collaboration was funded by the Critical Materials Innovation Hub (CMI). CMI, a Department of Energy (DOE) energy innovation hub led by Ames National Laboratory, seeks to advance technologies that make better domestic use of materials with unreliable supply chains. With Pennsylvania State University (PSU) collaborators, the Livermore team achieved 99.9 percent purity for both dysprosium and neodymium from a mixed feed in just two cycles of acid desorption. Acid desorption involves the flow of low-pH solution through columns of already REE-bound resin to encourage specific REEs to detach from LanM. This early breakthrough demonstrated LanM’s effectiveness in isolating REEs from a common pair present in recycled rare-earth magnets, encouraging further research seeking LanM variants that could separate other groups and, potentially, all 17 REEs.
Improving Selectivity
After establishing the LanM resin method, the team now seeks to hone its abilities for additional separation. Together with CMI collaborators Joseph Cotruvo and Joseph Mattocks of PSU, Park, Dong, and Jeremy Seidel, an SSB postdoctoral researcher, developed a framework for further separation of REEs. PSU contributed purified LanM for use in the Livermore team’s columns and played an integral part in developing the separation process. After REEs were suspended in LanM columns, the team examined the use of different elution solutions including acid, malonate, and citrate to selectively desorb REEs along a gradient. Elution yielded fractions enriched in scandium and yttrium, as well as groups of light, middle, and heavy lanthanides, bringing the team closer to achieving individual REE separation. This result also advances the goal of widespread domestic production of individual REEs from various domestic feedstock sources.
To move closer to individual REE separation, Livermore is now working in collaboration with American Rare Earths (ARE), a major supplier of low-grade REE material in the form of raw ores and enriched intermediates, as part of the Defense Advanced Research Projects Agency’s (DARPA) Environmental Microbes as a BioEngineering Resource (EMBER) program. EMBER is a collaborative program that aims to improve the specificity and selectivity of microbes and biomolecules for REE separation.
ARE funnels their material to a University of Kentucky team that concentrates the REE-bearing material and discards the gangue (commercially valueless) material. The university team leaches the REEs into a leachate solution ready for later column separation by Livermore. The Livermore team’s goal is to improve selectivity such that the separation of eight more individual REEs, specifically the less abundant, heavy REEs, may be achieved from this feedstock. Park observes that separation of lanthanum, yttrium, and cerium in particular show promise. Their separation would also demonstrate LanM resin’s effectiveness in separating REEs from a more extensive range of feedstocks.
Although not rare earths, actinides are another group of interest in LanM research with potential biomining implications. Actinides are radioactive and high-value elements with applications in cancer medicine and nuclear forensics—the investigation of nuclear materials to identify their source and history. “Actinide work was a natural development from the initial work on rare-earth elements, since we are a National Nuclear Security Administration lab, one of the rare institutions where biologists and actinide chemists can innovate together,” says Deblonde. As reported in Science Advances and Journal of the American Chemical Society, Deblonde and PSU collaborators realized that LanM has even higher selectivity for actinides over REEs. In research funded by DOE, the team demonstrated that LanM is extremely selective for actinium, a radioactive element and emerging material for targeted alpha therapy anti-cancer treatments. The team was the first to measure the affinity of a protein for actinium, noting that LanM’s large appetite for actinium highlights its potential medical use. Deblonde notes that the application of LanM for actinide separation could become widely adopted before REE biomining, which may improve the speed and cost at which REE biomining is implemented.
Bigger and Better
Most work in biomining has used the Mex-LanM variant, which currently lacks the selectivity to separate all 17 REEs. However, Livermore has demonstrated capabilities to scale the biomining effort, notably in identifying LanM variants with higher selectivity and expanding the repertoire of biomining proteins. In the face of this challenging endeavor, Jiao is enthusiastic about the team’s ongoing impact. “Even if we find just a few specialized protein variants, we add a lot to the field compared to the existing, chemically synthesized ligands,” she says. “Our specificity is already so much higher, and we have decreased the number of extraction steps by a lot.” An example is the variant Hans-LanM, a dimer with distinct metal coordination strategies that make it more selective between neodymium and dysprosium than Mex-LanM. Hans-LanM’s selectivity was revealed through CMI efforts by Dong, Park, and Christina Kang-Yun, an SSB postdoctoral researcher, in collaboration with PSU.
The expansion effort continues via the DARPA EMBER program. Patrick Diep, SSB postdoctoral researcher, has developed a high-throughput screening capability to test more than 800 LanM variants in miniaturized columns for their selectivity toward different REEs. Simultaneously, Seidel uses computational modeling to assist with scale-up and to optimize process conditions. Unlike traditional industry ligands, each LanM protein can bind multiple REEs at once. Discovering or engineering LanM variants that are specific for individual REEs requires an understanding of the protein’s sequence-to-function relationship and their thermodynamic stability with all possible REE combinations, an intense, information-gathering task made possible with high-throughput laboratory automation and high-performance computing. As more selective variants are identified, Diep and Seidel’s work will enable prioritization of the most promising variants for further testing and scale-up.
As Livermore stands at the cusp of analyzing data for patterns that appear in variants of higher REE selectivity and better binding performance, the stage is set for engineering proteins capable of separating individual REEs more effectively. “Compared to other protein design projects, we are pushing the limit,” says Jiao. “Lanthanides are so alike in their chemical properties. Finding unique proteins specific to individual lanthanides is a tall order, and I see that as the most challenging and most exciting part of our work.”
Livermore’s biomining-focused collaborations pave the way to a future of secure, domestic REE production to meet national security, clean energy, and other technological needs. Livermore’s team has unparalleled potential to unlock individualized separation for all 17 REEs and the demonstrated ability to make use of previously untapped low-grade REE feedstocks. Nevertheless, changing a traditional industry such as mining is no small feat; implementing biomining technology on a large scale will take time. Deblonde acknowledges that while the work to advance biomining for industry use represents a long-term investment, Livermore’s research is on the best track. “LanM can be economically and technologically feasible, and the work this group has been doing is fantastic and really promising,” he says. Park adds, “I can’t overstate the importance of the expertise we have at Livermore.”
—Lilly Ackerman
For further information contact Dan Park (925) 423-7508 (park36 [at] llnl.gov (park36[at]llnl[dot]gov)).
This article is approved for Public Release, Distribution Unlimited
