32.19: Part 4 - Biometallury of Rare Earth Metals for Clean Energy

Search Fundamentals of Biochemistry

Learning Goals (ChaptGPT+ 1/16/25)

Learning Goals:

1. Connect biochemical and physical principles to clean energy technologies.

• Explain the basic biophysical mechanisms by which electric motors and generators convert between electrical and mechanical energy, including the role of magnetic fields and current.

2. Describe the chemical properties that make rare earth elements valuable in modern electrotechnologies.

• Identify which rare earth elements (e.g., neodymium, dysprosium, terbium) are most important for permanent magnets used in electric vehicles and wind turbines. 
• Explain how lanthanide electronic structure contributes to strong magnetic properties that outperform many transition metals.

3. Relate metal-binding principles from metalloproteins to rare earth chemistry.

• Define first-sphere (direct metal-ligand) and second-sphere (indirect, e.g., hydrogen bonding) interactions in protein metal binding and describe how these influence metal coordination geometry.
• Compare how biological Ca²⁺-binding motifs (EF-hands) differ from lanthanide binding in terms of coordination number and ligand specificity.

4. Explain how understanding metal-binding coordination and selectivity can inform novel separation strategies.

• Describe how second-sphere interactions and protein quaternary structure can produce highly selective binding of light vs. heavy lanthanides.
• Discuss how engineered or naturally occurring lanthanide-binding proteins (e.g., lanmodulins) can be leveraged in biometallurgical approaches for rare earth separation.

5. Evaluate the broader implications of rare earth extraction and use in the context of climate change and sustainability.

Articulate why rare earth elements are essential for clean energy technologies, including electric vehicles and wind turbines, and how biochemical strategies might improve the sustainability of their extraction and separation.

 

Introduction

To fundamentally address climate change, we must dramatically reduce our reliance on fossil fuels and replace them with clean energy sources such as solar, wind, battery storage, and advanced geothermal energy.  In the US in 2024, 93% of fossil fuels were burned, as shown in Figure \(\PageIndex{1}\) below

sas

Figure \(\PageIndex{1}\):  U.S. fossil fuel consumption by source and sector, 2024 in Quatrillion BTUs.  US Energy Information Administration (accessed 11/31/26 and preserved from removal by US government officials)

The green-highlighted area in the middle chart shows the percentage of fossil fuels that were not burned (e.g., chemical and plastic feedstocks).  The rest, 93%, was burned to power automobiles, heat homes and commercial buildings, support industrial manufacturing that requires heat, and power generators (like gas turbines) to produce electricity.  

Burning fossil fuels for transportation is an incredible waste of energy.  After a century of advances in the internal combustion engine, only about 22% of the energy in the burned fuel is used to power the vehicle. The rest is lost as heat.  Compare this to electric vehicles, where the efficiency is closer to 90%. The long term goal is to power electric vehicles with batteries charged by clean energy.  Residential and commercial heating can be accomplished by heat pumps.  These use electricity to move heat.  They don't have to create heat by burning natural gas, for example. Heat pumps have a coefficient of performance (COP) of 3 (for above-ground models) to 5 (for ground-installed models), meaning they deliver 3-5 units of heat for every 1 unit of electricity consumed. This is an efficiency of 300-500%, which is possible because the heat pump does not create heat; it just moves it. The COP of a natural gas furnace is closer to 1 (100% efficiency).

So, where is the connection to biochemistry and climate change? We need to power everything, not by burning fossil fuels, but through electricity.  We need electric motors and modern wind turbines that rely on extremely strong yet lightweight permanent magnets to efficiently convert electrical energy into motion (motors) or motion into electrical energy (like windmills). Many of the best permanent magnets are made using rare earth elements—especially neodymium (Nd), dysprosium (Dy), and terbium (Tb)—because of how their electrons behave at the atomic level.  We can help purify those metals through biometallurgy, as described in this chapter section.

An introductory physics review -  How electric motors work

See the link below.

Answer

Electric motors rely on the fact that a moving electric charge generates a magnetic field (and vice versa).  Figure \(\PageIndex{2}\) below shows a wire loop (actually a coil of wires in the armature) placed in a permanent magnetic field, with the magnetic field lines pointed from the N to S ends of the permanent magnet (the stator, which is stationary).  The direction of the current (I) is shown in green arrows (in accord with physicists' nomenclature that shows current as the flow of + charge). 

Figure \(\PageIndex{2}\):  Operation of a DC Motor – Learn.  https://www.scienceflip.com.au/subje...netism/learn9/

Since the current is moving in opposite directions near the S and N poles, the magnetic field induced by the current interacting with the external magnetic field produces an upwards force, F_M, on the wires closest to the S pole and a downwards force on the wires closest to the N pole.   The direction of the forces can be determined by the right-hand rule you learned in physics.  Figure \(\PageIndex{3}\) below shows the relative directions of the magnetic field vector B, the current flux J, and the force F.

Figure \(\PageIndex{3}\): https://commons.wikimedia.org/wiki/File:Right_hand_rule_cross_product_F%3DJ%C3%97B.svg

Since the force F pushes the wires near the magnetic poles in different directions, the loop starts rotating clockwise (conversion of electrical to mechanical energy).  However, after half a turn, the forces would start the loop rotating in the opposite direction. For a smooth rotation, the split-ring commutator/brush structure allows the current to change direction through the coil every half-turn, so the coil continues rotating in the same direction.  See this link for more details.  

Here is a link to an animation showing how the opposing forces on each side of the wire loop in the magnetic field cause the loop to rotate, effectively converting electrical energy into mechanical energy.

 

Rare Earth Elements (REE) and Electric Motors

Let's focus on the rare-earth elements (REE) Nd, Tb, and Dy in the lanthanide series (Figure \(\PageIndex{4}\). This series, along with the actinides, lies outside the main blocks (s, p, and d) of the periodic table and is part of the outer transition elements or the f-block.  This contrasts with the inner transition metals of the d-block, including Mn, Fe, and Co. There are 10 elements in each d-block row, corresponding to adding electrons to the five d orbitals of the inner block until all d orbitals are filled with a total of 10 electrons.  The lanthanide and actinide series have 14 elements to accommodate 2 electrons in each of the 7 d orbitals.

Figure \(\PageIndex{4}\):  Periodic Table and Rare Earth Elements (outlined in red rectangle along with Sc and Y).  By Armtuk - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/inde...?curid=2010645

Table \(\PageIndex{1}\) shows the electronic configurations of the neutral Nd, Tb, and Dy elements, compared with that of the magnetic d-block element Fe.

Element	Fe	Nd	Tb	Dy
Electron Configuration ground state element	[Ar]3d^64s^2	[Xe]5s^25p^64f^46s^2	[Xe]5s^25p^64f^96s^2	[Xe]5s^25p^64f^106s^2
3^+ Oxidation state	[Ar]3d^5	[Xe]5s^25p^64f^3	[Xe]5s^25p^64f^8	[Xe]5s^25p^64f^9

Table \(\PageIndex{1}\): electronic configurations of the neutral Nd, Tb, and Dy elements, compared with that of the magnetic d-block element Fe

Some of the d electrons of all these elements are unpaired, so they have stable magnetic moments. The 4f unpaired electrons in Nd, Tb, and Dy are shielded by 5s, 5p, and 6s electrons, so their magnetic moments are shielded and stable.  Alloys containing lanthanides (Nd, Fe, and B) used in magnets can make them lighter, more efficient, and less prone to realignment and demagnetization, making them key for electric motors and wind turbine generators. 

Mining, extracting, and purifying these REE is difficult and environmentally toxic.  Techniques for purifying REE from rocks must rely on differences in the properties of the metals.  Proteins composed of the same amino acids are separated from mixtures based on size, charge, and ligand affinity.  Metals differ in their ionic radii, charge densities, preferred coordination geometries, and ligands.  The RER metals' most common charge state is the same, +3.  Their size decreases somewhat across the row (lanthanide contraction) due to increasing nuclear charge, but it is minor (a decrease of only 0.19  Å across the row.  

A last note.  The REEs are not really rare, as the name implies.  For example, Nd is more abundant than Pb.  The problem is that they are not found in concentrated form but rather are dispersed.  Their chemistries are similar, and the REEs often substitute for other elements in minerals that contain them. The problem of purifying them is not unlike purifying a single protein that looks superficially like all other proteins and that is not highly expressed from the proteome of an organism 

Lanmodulins (LanM): Calmodulin-like, lanthanide-binding proteins

By combining biochemistry and metallurgy into a new field, biometallurgy, less toxic ways are being developed to purify specific REE. Proteins exist in nature that can bind lanthanides with high specificity and affinity.  One such protein family is the lanthanide-binding lanmodulins (analogous to calmodulin, a calcium-binding protein) that bind lanthanides with a 100-million-fold selectivity over Ca²⁺, for example.  These proteins can be engineered to have even higher selectivity and binding affinity.

The ionic radius of Ca²⁺ for an octahedral (6 coordination) complex is 100 pm (1.00 Å).  The radii for the lanthanides for octahedral geometries vary from 1.03 Å for La³⁺ at the beginning of the series to 0.89 Å for Er³⁺ at the end, with Nd³⁺, Tb³⁺, and Dy³⁺ at 0.98, 0.92, and 0.91 Å, respectively.  This suggests that separating and purifying lanthanides based on size would be very difficult. 

Given their size and charge, lanthanide ions can bind to Ca²⁺-binding proteins.  However, their +3 charge makes electrostatic interactions at normal Ca²⁺ binding sites stronger, which can decrease the dissociation rates of lanthanide ions and decrease the protein's conformational flexibility when bound to Nd³⁺, for example.  This, in turn, could alter Ca²⁺ signaling pathways.

Like the well-known calcium-binding protein calmodulin (see 28.7: Calcium Signaling), lanmodulin contains EF-hand motifs (helix-loop-helix structures, see Chapter 4.4) that have coordinating ligands that bind Ca²⁺ ions. 

In calmodulin, these loops contain Asp and Glu side chains that coordinate the Ca²⁺ ion, with a coordination number of 7, in a distorted pentagonal bipyramidal geometry, allowing higher affinity for the large Ca²⁺ ion and conformational flexibility. The EF hand is disordered before Ca²⁺ binding.  Figure \(\PageIndex{5}\) shows an interactive iCn3D model of a Ca²⁺ ion bound in an EF hand motif of human calmodulin with interacting amino acids labeled.

Figure \(\PageIndex{5}\): Bound calcium ion and interacting amino acids in human calmodulin (1cll) (Copyright; author via source).
Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?bCf4mtNbk4kjkCHw6

A review on metal-ligand geometry and bonding in proteins:  Ca²⁺, Mg²⁺ and Zn²⁺ 

If you are interested, click the link below 

Answer

Table \(\PageIndex{5}\) shows the radii and coordination state of biologically common divalent cations.

Ion	Typical CN	Ionic radius (Å)	Binding examples
Mg^2^+	6 (rigid)	0.72	ATP,
Zn^2^+	4–6	0.60–0.74	Zn fingers bind DNA
Ca^2^+	7–8 (flexible)	1.06–1.12	calcium-binding proteins

In introductory chemistry classes, you learned about crystal lattices, packing of ions, and unit cells.  For example, the NaCl crystal can be envisaged as a face-centered cubic (FCC) lattice of the larger Cl- ions, with the small cation Na⁺ occupying the lattice holes.  Since the table, the coordination number and packing of the cations depend on their size, a small review of packing holes is helpful. 

First, let's start with this example: the packing of circles in 2D, or equivalently, a top-down view of spheres packing in one layer.  Figure \(\PageIndex{6}\) below shows two ways to do so.  In the left two images showing square packing, the red spheres are not packed as closely as in the right image, which shows hexagonal closest packing, which represents the closest packing possible.  Note that the vacant spaces or holes in the left images are larger than in the right, closest-packed images.  Note also that in the right hexagonal packing image, each internal sphere has 6 nearest neighbors, whereas in the left image, it has only 4.  These represent the coordination numbers of the spheres. 

Figure \(\PageIndex{6}\): Packing of spheres in 2D

The formula under the middle figure shows how to calculate the radius r of the smaller cation sphere that fits into the hole formed by the 4 red anions with radius R. It's a simple application of the Pythagorean theorem.  There could be a range of cation sizes depending on the anion size.

We are not interested in extended lattices for the interaction of metal ions with proteins.  However, the relative size of the anion ligands compared to the cation and the coordination number and hole size available for the cation are biologically relevant.  Figure \(\PageIndex{7}\) below shows the relative sizes of holes for tetrahedral, octahedral, and cubic packing of spheres in 3D.  Again, the cation (red sphere) sits in the hole.  

Figure \(\PageIndex{7}\): relative sizes of holes for tetrahedral, octahedral, and cubic packing of spheres in 3D

The cubic hole on the right, with a coordination number of 8 (nearest neighbors, shown not touching in these figures), is the largest.  This makes sense since the red cation can interact with a large number (8) of blue anions, so sufficient space must be available for this to happen.  In contrast, for the red cation in the tetrahedral hole (left), only 4 ligand anions can make close contact, making the hole the smallest.  The octahedral hole, with a coordination number of 6, can accommodate six blue anions, so the hole is intermediate in size. 

Figure \(\PageIndex{7}\) below shows, for an additional example, the packing of NaCl in a salt crystal (left and right bottom images).  The red Cl^- anions pack in the corners and faces of the unit cell (left image), making up the lattice.  The black Na⁺ ion is packed in the very center of the large unit cell and at all the edges (bottom right).  As shown in the bottom-right image, the Na⁺ ions are packing into octahedral holes.

Figure \(\PageIndex{8}\): Packing of cations in octahedral and tetrahedral hole

The top-right image shows only part of a unit cell, with red ions at the corners and faces.  This does not occur in NaCl but rather in CaF₂ as shown in Figure \(\PageIndex{9}\) below.  In flourite, CaF_2, the Ca²⁺ ions (purple, 1.12 Å, coordination number 8), which you can see if you duplicate a unit cell, occupy the faces and edges of the unit cell, creating the unit cell shape (blue lines), while the F^- ions (green, 1.31 Å, coordination number 4) occupy every internal tetrahedral hole. 

Figure \(\PageIndex{9}\): ChemDraw 3D.  https://www.chemtube3d.com/_fluoritefinal/

Now back to the biological ions.  Although Mg²⁺ and Zn²⁺ are similar in size, Mg²⁺ is a "hard" (non-polarizable) d⁰ ion that favors octahedral coordination for electrostatic (ion - ion) stabilization. It has no d electrons that would allow ligand field stabilization (extra stability gained when metal complex d-orbitals split in the presence of ligands aligned with or between the coordinate axes).  The octahedral geometry allows the six ligands to approach the metal ion as closely as possible while minimizing electrostatic repulsion.

In contrast, Zn²⁺ is a d¹⁰ ion that has no vacant d orbitals to contribute to molecular orbitals and covalent bond formation.  Since the orbitals are filled, no ligand field stabilization of the d orbitals by the ligands is possible (since stabilizing energy gained by electrons in lower-energy orbitals is exactly cancelled by the destabilizing energy of electrons in higher-energy orbitals). However, some electron sharing can occur with the empty 4s and 4p orbitals, which are close in energy.   Although Zn²⁺ and Ca²⁺ have the same charge, the zinc ion is smaller, given the much greater nuclear charge.  This leads to a tetrahedral geometry, placing the ligands as close as possible to the charge center.  Hence, Zn²⁺-ligand interactions are generally tetrahedral with some covalent bond character.  This geometry also minimizes electron-electron repulsion from the d electrons when the ligands are packed in an octahedral arrangement.

In contrast, Ca²⁺ is not found in tetrahedral or octahedral complexes.  It's not a transition metal and has no d or f electrons to form covalent-like bonds.  Using crystal field theory, there is no real ligand-field stabilization that provides orbital stabilization.  Rather, the bonding and stabilization are entirely ionic (electrostatic).  More ligands can fit around the larger Ca²⁺ ion, increasing its flexibility and its ability to adopt distorted geometries.

 

LanM differs from calmodulin (CAM) in several ways.  In the absence of Ca²⁺, CAM retains a significant alpha-helical structure and is "primed" (significantly pre-folded) to bind Ca²⁺.  In contrast, apoLanM is mostly disordered, and no PDB structures exist for it.   Circular dichroism measurements show little helical structure in the apo-state.  The AlphaFold-predicted structure is much closer to the metal-ion-bound form since AlphaFold is biased toward predicting functional states that are more prevalent in structural databases.

The dissociation constant for Ca²⁺and CAM varies for each site from about 1 μM to 0.1 μM, while for LanM, the dissociation constant of lanthanide ion binding is closer to 1 pM (much tighter).  In addition, LanM is much more selective for lanthanides than for Ca²⁺.

Figure \(\PageIndex{10}\) below shows an interactive iCn3D model of the Methylorubrum extorquens AM1 lanmodulin (LanM) with neodymium (III) bound (8FNS). Note that there is no long helix separating the two ion-binding domains.

Figure \(\PageIndex{10}\):  Methylorubrum extorquens AM1 lanmodulin (LanM) with neodymium (III) bound (8FNS).  (Copyright; author via source). Click the image for a popup or use this external link: https://www.ncbi.nlm.nih.gov/Structu...4bac17f798558e

Figure \(\PageIndex{11}\) below is an interactive iCn3D model showing the alignment of calcium-bound human calmodulin (1CLL) with Methylorubrum extorquens AM1 lanmodulin (Mex-LanM) with neodymium (III) bound (8FNS).  The first structure that appears is a partially aligned PDB structure.  Use the "a" key to toggle back and forth between the aligned human calmodulin (1CLL) and the aligned M. extorquens lanmodulin (8FNS). Select OK to the warning that the structures can't be aligned with VAST.

Figure \(\PageIndex{11}\): Alignment of calcium-bound human calmodulin (1CLL) with Methylorubrum extorquens AM1 lanmodulin (Mex-LanM) with neodymium (III) bound (8FNS).  (Copyright; author via source). Click the image for a popup or use this external link: https://www.ncbi.nlm.nih.gov/Structu...190c41a1e1e63d

In lanmodulin, the long separating helix is absent, but it has two sets of EF hands, each with two bound Nd^III ions, which bind using the same ion-binding motifs.

Landmodulins have been used to isolate REEs and separate them from each other (e.g., Nd from Dy).  A variant of LanM from Hansschlegelia quercus (Hans-LanM) has been shown to be more effective at binding, separating, and purifying REEs than Mex-LanM.  Hans-LanM undergoes a metal-sensitive dimerization:

2 Hans-LanM ↔ (Hans-LanM)₂

The dimerization depends on the size of the lanthanide ion.  A larger lanthanide, such as Nd, promotes dimerization of Hans-LanM by 100-fold compared to the smaller ion Dy. This suggests that only a few picometer-scale differences can trigger conformational changes in the protein, leading to dimerization. The equilibrium can be changed by subtle changes in conditions, leading to the dissociation of Nd³⁺ (a lighter RE or LRE, 98 pm) and Dy³⁺ (a heavier RE or HRE, 91 pm) at different times, allowing selective purification of different lanthanide metal ions. A Ld³⁺/Dy³⁺ mixture can be separated on a Hans-LanM column to > 98% purity.

Figure \(\PageIndex{12}\) below shows differences in Hans-LanM (monomer) and Mex-LanM (forms dimers) in structure and lanthanide binding selectivity.  

Figure \(\PageIndex{12}\): Differences in Hans-LanM (monomer) and Mex-LanM (forms dimers) in structure and lanthanide binding selectivityMattocks, J.A., Jung, J.J., Lin, CY. et al. Enhanced rare-earth separation with a metal-sensitive lanmodulin dimer. Nature 618, 87–93 (2023). https://doi.org/10.1038/s41586-023-05945-5.  Creative Commons Attribution 4.0 International License.  http://creativecommons.org/licenses/by/4.0/.

Panel c: Circular dichroism spectra from a representative titration of Hans-LanM with La^III, showing the metal-associated conformational response increasing helicity; apoprotein is bold black, La^III-saturated protein is bold red. (See Chapter 3.4 for a review of CD).  Interpretation: This shows that apo-Hans-LanM is mostly disordered, with little helical content, as evidenced by the lack of peaks between 210 and 230 nm, including at 222 nm, characteristic of alpha helices.

Panel d: Circular dichroism titration of Hans-LanM with La^III (103 pm), Nd^III (98 pm), and Dy^III (91 pm) (pH 5.0). Each point represents the mean ± s.d. from three independent experiments. Interpretation:   "Binding of La^III and Nd^III (lighter RE metals, and larger) to Hans-LanM increases the molar ellipticity CD signal at 222 nm by 2.3-fold.  The K_d,app values are similar, 68 and 91 pM, respectively. At least one of the Dy^III-binding sites is very weakly responsive (K_d,app > 0.3 µM).

Panel e: Comparison of K_d,app values (pH 5.0) for Mex-LanM (black) and Hans-LanM (red), plotted versus ionic radius. Mean ± s.e.m. from three independent experiments.  Interpretation: Mex-LanM shows only a modest preference for LREs (about fivefold), Hans-LanM discriminates more strongly between LREs and HREs than does Mex-LanM, with the HRE complexes exhibiting lower affinity, lesser cooperativity, and a lesser primary conformational change."

The effective ionic radii used are the Shannon radii (not those found in chemistry textbooks mentioned above).  These radii are often used for oxygen-rich sites in proteins and ions with very high coordination numbers (e.g., 9 or higher, common for lanthanide-binding sites). Most common tables assume octahedral geometry.

Shannon ionic radii can also be influenced by other factors involved in ligand binding.   Two sets of interaction regions influence ion binding:

• The first coordination sphere involves atoms or groups that directly bind the metal ion.
• The second coordination sphere involves atoms or groups that do not bind the metal directly but interact with first-sphere ligands, typically through hydrogen bonds, ion-ion interactions, or hydrophobic interactions, thereby altering the local environment.  For example, steric interactions could force a ligand to bind in a mono- vs. bidentate mode.

In biological metal ion interactions with proteins (as an example), the first shell ligands (carboxylates, lone pairs from carbonyl oxygens, or water) are generally fixed. However, the geometry and rigidity can be fine-tuned using weak interactions from adjacent main-chain and side-chain atoms. Hence, selectivity (ion-binding preference) can be changed through second-sphere interactions without changing the first-sphere ions.

Does the apparent molecular weight of the protein (i.e. dimierization state) depend on which lanthanide ion is bound? Figure \(\PageIndex{13}\) below shows that lighter transition (larger) metal ions (3+), such as La³⁺ and Nd³⁺, promote dimerization.  The graph shows the apparent molecular weight of the protein vs the effective radius of the ions.  Two techniques, Size-Exclusion Chromatography-Multi-Angle Light Scattering (MALS) and Size-Exclusion Chromatography (SEC), were used to determine the apparent molecular weight. 

Figure \(\PageIndex{13}\): A dimerization equilibrium sensitive to LRE versus HRE or non-RE coordination.  Mattocks, J.A., Jung, J.J., Lin, CY. et al. ibid.

Panel a: Apparent molecular weight of Hans-LanM complexes with REs as determined by analytical SEC (red lines) or SEC–MALS (black dashed line). Each individual data point is the result of a single experiment

Interpretation:  In the presence of three equivalents of La³⁺, Hans-LanM elutes from a size-exclusion chromatography (SEC) column not at the expected molecular weight (MW) of 11.9 kDa but instead at 27.8 kDa, suggestive of a dimer. Starting gradually after Nd³⁺ but sharply at Gd³⁺, the apparent MW decreases towards that expected for a monomer.

How might the dimer formation promote binding and ion selectivity?  First, let's examine the structure of the ion-bound dimer. Figure \(\PageIndex{14}\) below is an interactive iCn3D model of the lanthanide-induced dimer of Hansschlegelia quercus lanmodulin (LanM) (8DQ2).

Figure \(\PageIndex{14}\): Lanthanide-induced dimer of Hansschlegelia quercus lanmodulin (LanM) (8DQ2).  (Copyright; author via source). Click the image for a popup or use this external link: https://www.ncbi.nlm.nih.gov/Structu...c47f8242b77f18

The iCn3D shows a significant burial of surface area (600 Ų) mediated by hydrophobic and polar contacts. Figure \(\PageIndex{15}\) shows a closeup of the dimer interface and a zoomed-out representation of interactions in the interface. From these structures, we can discern the protein side chains and backbone interactions involved in first- and second-sphere interactions.

Figure \(\PageIndex{15}\): A dimerization equilibrium sensitive to LRE versus HRE or non-RE coordination.  J.A., Jung, J.J., Lin, CY. et al., ibid.

Panel c: Detailed view of the dimer interface near EF3 of chain A (blue cartoon). Arg100 from chain C (light blue cartoon) anchors a hydrogen-bonding network involving Asp93 of chain A and two EF3 La^III ligands (Glu91 and Asp85). These interactions constitute the sole polar contacts at the dimer interface, providing a means to control the radius of the lanthanide-binding site at EF3.  Panel d: Schematic of the interactions at the dimer interface. Red dashed lines indicate hydrogen-bonding interactions and grey dashed lines indicate hydrophobic contacts.

Interpretation: Dimer interactions occur largely between side chains contributed by the core helices α1 (between EF1 and EF2) and α2 (between EF3 and EF4; Supplementary Fig. 9). Residues at the dimer interface make direct contact with only one of the four metal-binding sites, EF3; three residues of EF3 in each monomer form a hydrogen-bonding network with Arg100 of the other monomer, suggesting that occupancy and coordination geometry at this site may control oligomeric state.  Arg100 is clearly involved in second sphere interactions.

Figure \(\PageIndex{16}\) below shows both first sphere and second sphere interactions participating in the binding of La³⁺ and Nd³⁺ to Hans-LanM and Mex-LanM.

Figure \(\PageIndex{16}\): Hans-LanM uses an extended hydrogen-bonding network to control lanthanide selectivity. Mattocks, J.A., Jung, J.J., Lin, CY. et al. ibid.

Panel a: Zoomed-in views of EF2 (left) and EF3 (right) in La^III–Hans-LanM. La^III ions are shown as green spheres. Coordination bonds and hydrogen bonds are shown as dashed lines. Residues contributed by chain A are shown in blue, and those contributed by chain C (in the case of EF3) are shown in light blue. Inset: overlay of La^III–Hans-LanM (blue and light blue) with Dy^III–Hans-LanM (grey), showing the carboxylate shift of Glu91 from bidentate (La) to monodentate (Dy). Coordination and hydrogen bonds (dashed lines) are shown only for the Dy case.

Interpretation:   A monodentate Asn (N₁ position), four bidentate Asp or Glu residues (D₃, D₅, E₉, and E₁₂), and a backbone carbonyl (T₇ or S₇) complete the first coordination sphere in EF1–3. No solvent is observed.  The subscript numbers 1, 3, 5,... are used to label the metal-ion-interacting ligands, but they are not clearly defined in the paper.

Panel b: Representative metal-binding site (EF3) in Nd^III–Mex-LanM. Nd^III ion is shown as an aqua sphere. Solvent molecules are shown as red spheres

Interpretation:  The metal binding site differs in Mex-LanM and Hans-Lan M.  For Nd³⁺-bound Mex-LanM, the EF1-3 are nine-coordinate (only 8 coordinates are directly seen in Panel B however) and EF4 is ten-coordinate (since one Glu is bidentate) . All 4 EF hands each contain 2 bound water molecules, which are absent in Hans-LanM. 

The Nd³⁺ Mex-LanMbinding sites are similar to the seven-coordinate Ca²⁺-binding sites of calmodulin but have a bidentate Glu and one or more water as ligands. Yet the LanM shows much greater selectivity (10⁸x) for lanthanides compared to Ca²⁺, likely due to second-sphere interactions.  

Hence, Hans-LanM dimerizes in a process that depends on REE, with the lighter ones (larger radii) strongly shifting the equilibrium toward the dimer form. Metal-controlled protein-protein interactions are key to the selectivity of binding lanthanide ions, so it's not just the metal-ligand interactions alone that determine binding.

Finally, look again at the role of R100 as a second-sphere ligand.  It forms hydrogen bonds with Asp 83, 93, and Glu 91, helping to form an extended hydrogen-bond network that stabilizes the dimer and the EF3 lanthanide-binding site.  A single mutation of this amino acid to lysine (R100K) weakens dimerization and destabilizes LRE without affecting HRE.  This shows that dimerization is the source of enhanced selectivity

Here is a summary directly from J.A., Jung, J.J., Lin, CY. et al., ibid.)  ""Biochemical and structural characterization of Hans-LanM’s mechanism of metal-sensitive dimerization provides a new, allosteric mechanism for LRE versus HRE selectivity in biology, extending concepts in dimer-dependent metal recognition recently emerging from synthetic lanthanide complexes and engineered transition metal-binding proteins and showing that these principles are hard-wired into nature. Our work also shows that dimerization strength, and thus metal selectivity, can be rationally modulated. Hans-LanM evolved LRE-selective dimerization at physiological protein concentrations closer to those in our biochemical assays (10–20 µM) rather than those on the column (about 3 mM). Therefore, leveraging dimerization in a separation process would be assisted by shifting dimerization sensitivity to the higher concentration regime, such as by tuning hydrophobic interactions at the dimerization interface. Furthermore, our studies establish that LanMs with as low as 33% identity are easily predicted yet have useful differences in metal selectivity; further mining of this diversity may reveal yet additional mechanisms for tuning RE separations. Finally, the solvent-excluded coordination spheres of Hans-LanM should outperform Mex-LanM in RE/actinide separation, luminescence-based sensing and stabilization of hydrolysis-prone ions. Continued characterization of the coordination and supramolecular principles of biological f-element recognition will inspire design of ligands with higher RE versus RE selectivities and their implementation in new RE separation processes."

Note: Joseph Cotruvo, the principal investigator on the LanM dimerization study, won the 2026 ASBMB Mildred Cohn Young Investigator Award 

Lanmodulin can be covalently attached to chromatography resins to form an affinity column for the separation and purification of REE.  They can be extracted from electronic and mining waste, for example. Much has been said about the need for the US to begin mining and processing REE to avoid supply chain bottlenecks and for national security, since China has the largest underground resources and processing technologies.  In fact, the extraction and purification technology was developed mostly in the West but was effectively abandoned here, with extraction carried out in China (which has the most abundant underground REE resources), in part to avoid the environmental damage associated with it. 50-70% of the REE extracted from the ground ends up in discarded, unused waste. Less than 1% of the REEs in retired items such as jets and ships, and discarded electrons is recycled.  Burned fossil fuels are lost forever, and continued extraction is needed to resupply them. In contrast, Li and other metal ions in discarded batteries, and the REEs in mining and manufacturing wastes are still present and, if reprocessed and purified, could meet many of our needs for those metals without mining new resources.

The Critical Minerals Data Explorer from the International Energy Agency (IEA) provides demand projection results under various energy scenarios and technology evolution trends. Users can look up total demand and supply for key minerals (copper, cobalt, lithium, nickel, graphite and rare earth elements) and projected mineral demand in the clean energy sector by technology and commodity under different scenarios and technology cases, including 11 alternative cases modeling the impact of different technologies or consumer behaviors.

Summary

Modern clean-energy technologies rely heavily on rare earth elements (REEs), particularly the lanthanides, because of their unique electronic and magnetic properties. Elements such as neodymium, dysprosium, and terbium are essential components of high-performance permanent magnets, which are central to the operation of electric vehicle motors, wind-turbine generators, and other energy-efficient electromechanical systems. These magnets enable compact designs with high power density and efficiency, directly supporting the transition away from fossil-fuel-based energy systems.

The exceptional functionality of rare earth magnets arises from the electronic structure of lanthanide ions, especially their partially filled 4f orbitals. These orbitals are shielded from the environment, producing strong and stable magnetic moments that differ fundamentally from those of transition metals. Small differences in ionic radius across the lanthanide series nevertheless lead to meaningful changes in material properties, making specific rare earth elements more suitable for particular technological roles.

From a biochemical perspective, rare earth elements present a paradox: they are chemically similar to biologically abundant metal ions such as Ca²⁺, yet living systems can discriminate between them with remarkable selectivity. This selectivity arises from coordination chemistry principles, including high coordination numbers, oxygen-rich ligand environments, and the influence of second-sphere interactions that fine-tune metal binding without direct metal–ligand contact. These principles are familiar from metalloproteins and metal-binding motifs such as EF-hands, providing a conceptual bridge between biochemistry and materials science.

The chapter highlights how insights from biochemistry can inform sustainable approaches to rare earth separation and recycling, a major bottleneck in clean-energy deployment. Conventional rare earth extraction is environmentally damaging and energy intensive, largely because of the chemical similarity of lanthanide ions. In contrast, biologically inspired strategies—such as the use of selective metal-binding proteins—demonstrate how molecular recognition, cooperativity, and subtle structural effects can be exploited to achieve separations that are difficult using traditional chemical methods.

Overall, rare earth elements exemplify how molecular-level chemical properties scale up to global technological and environmental consequences. Understanding their coordination chemistry, electronic structure, and interactions with biological molecules equips biochemists to contribute meaningfully to challenges in climate change, energy sustainability, and resource management. This chapter underscores the expanding role of biochemistry beyond the cell, into the design and optimization of technologies critical for a low-carbon future.