Lanthanide Contraction

The lanthanide contraction (coined in 1925 by the Norwegian geochemist Victor Moritz Goldschmidt as Lanthanidenkontraktion) is the continuous, greater-than-expected decrease in atomic radius and trivalent ionic radius (r(Ln³⁺)) across the 15 elements of the 4*f* lanthanide series from lanthanum (Z = 57) through lutetium (Z = 71). Whereas the metallic radii of neutral lanthanide atoms (barring divalent europium and ytterbium exceptions) decrease by ~14 pm across the row (from 187 pm in La to 173 pm in Lu), the six-coordinate Shannon ionic radii of the closed-subshell-core Ln³⁺ cations ([Xe] 4*f*⁰ to [Xe] 4*f*¹⁴) contract smoothly and monotonically by 17.1 pm (16.6%), from 103.2 pm for La³⁺ down to 86.1 pm for Lu³⁺.

This cumulative radial shrinkage across the 14 f-block elements has two major chemical repercussions across the periodic table: first, within the lanthanide series itself, it reduces the ionic radius of late trivalent lanthanides (Ho³⁺ = 90.1 pm, Er³⁺ = 89.0 pm) to match that of the Group 3 4*d* congener yttrium (Y³⁺ = 90.0 pm), explaining why heavy lanthanides and yttrium co-occur in the same silicate and phosphate minerals (such as xenotime) and require multi-stage solvent extraction to separate; second, immediately following lutetium in Period 6, it cancels out the normal shell expansion expected between the 4*d* (Period 5) and 5*d* (Period 6) transition metal rows, rendering homologous 4*d*/5*d* pairs such as zirconium and hafnium nearly indistinguishable in size and chemical behavior.

---

Poor Radial Shielding of 4*f* Orbitals (s > p > d > f)

As atomic number increases by Δ*Z* = +14 from La³⁺ ([Xe] 4*f*⁰, Z = 57) to Lu³⁺ ([Xe] 4*f*¹⁴, Z = 71), each step adds one proton to the nucleus and one electron to the 4*f* subshell (n = 4, l = 3) while the outermost occupied spatial shells of the Ln³⁺ ion remain the filled 5*s*² 5*p*⁶ xenon-core orbitals (and, in the neutral metal or coordination complex, the 5*d* and 6*s* valence orbitals).

In the radial Schrödinger equation, an orbital with azimuthal quantum number l = 3 experiences an intense centrifugal barrier potential:

V_centrifugal(r) = l(l + 1)*ℏ*² / (2*m*_er²) = 12*ℏ*² / (2*m*_er²)

that keeps 4*f* electron density away from the immediate vicinity of the nucleus (r → 0), while its angular distribution (Y₃^m_l, possessing three planar/conical angular nodes and up to eight directional lobes) is diffuse in angular coordinates. Consequently, the shielding efficiency of subshells within a given shell follows the strict hierarchy:

Shielding ability: s > p > d > f

Although the mean radial maximum (r_max ≈ 30–40 pm) of the compact 4*f* shell lies inside the outer 5*s*²5*p*⁶ subshells (r_max ≈ 60–90 pm), the multi-lobed 4*f* electrons fail to screen one another and fail to screen the outer 5*s*, 5*p*, 5*d*, and 6*s* electrons completely from the +14 increase in nuclear charge (effective screening constant Δ*σ* ≈ 0.69–0.85 per added 4*f* electron). Thus, the effective nuclear charge (Z_eff = Z − σ) experienced by the outer 5*s*²5*p*⁶ and 6*s* shells increases steadily by ~0.15–0.30 units per element across the 14 steps of the lanthanide series, pulling the outer electron cloud progressively inward according to ⟨*r*⟩ ∝ 1/Z_eff.

Contribution of Dirac Relativistic Contraction (10–25%)

High-level comparative Hartree–Fock (non-relativistic) versus Dirac–Hartree–Fock (relativistic) quantum-chemical calculations performed by Pekka Pyykkö, Tom Ziegler, and Michael Dolg demonstrated that poor 4*f* electrostatic shielding accounts for 75–90% of the observed lanthanide contraction, while relativistic orbital contraction accounts for the remaining 10–25%. Because the inner 1*s* and penetrating ns (5*s*, 6*s*) electrons of heavy nuclei (Z = 57–71) travel at relativistic velocities (v/c ≈ Z / 137 ≈ 0.42–0.52), their relativistic mass increases by the Lorentz factor γ = 1 / √[1 − (v/c)²]. Since the Bohr radius scales inversely with electron mass (a₀ = 4π*ε*₀ℏ² / m_ee²), the penetrating 5*s*, 5*p*_1/2, and 6*s* shells contract radially by an additional 3–4 pm across the lanthanide row.

---

1. The 4*d* / 5*d* Transition Metal Twin Pairs (Zr/Hf, Nb/Ta, Mo/W)

In moving from the first transition series (3*d*, Period 4) to the second transition series (4*d*, Period 5), adding a new principal shell increases metallic and ionic radii by 13–18 pm (e.g., Ti = 147 pm → Zr = 160 pm; V = 135 pm → Nb = 146 pm; Cr = 128 pm → Mo = 139 pm). However, between the 4*d* row and the 5*d* row (Period 6), the insertion of the 14 lanthanides produces a 17 pm lanthanide contraction that almost exactly cancels the ~16 pm shell expansion from n = 5 to n = 6.

As a result, the 5*d* transition metals possess virtually identical metallic and ionic radii to their 4*d* congeners directly above them:

• Group 4: Zirconium (Zr, r_met = 160 pm, r(Zr⁴⁺) = 72 pm) and Hafnium (Hf, r_met = 158–159 pm, r(Hf⁴⁺) = 71 pm) are the closest chemical twins in the periodic table. Because Zr⁴⁺ and Hf⁴⁺ share the same +4 charge and identical radii, hafnium escaped discovery until 1923 (Dirk Coster and George de Hevesy via X-ray spectroscopy of Norwegian zircons) and must still be separated from nuclear-grade zirconium fuel cladding because Hf has a thermal neutron capture cross-section (104 barns) 580 times larger than Zr (0.185 barns).
• Group 5 & Group 6: Niobium / Tantalum (Nb = 146 pm, Ta = 146 pm; Nb⁵⁺ = 64 pm, Ta⁵⁺ = 64 pm, co-occurring in columbite–tantalite [Fe,Mn][Nb,Ta]2O6) and Molybdenum / Tungsten (Mo = 139 pm, W = 139 pm; Mo⁶⁺ = 59 pm, W⁶⁺ = 60 pm).
• Doubled Density of 5*d* Metals: Because a 5*d* atom packs nearly twice the atomic mass (Δ*A* ≈ +88–95 Da due to 32 extra protons plus ~58 extra neutrons) into the exact same atomic volume as its 4*d* partner, the densities of the 5*d* metals are roughly double those of the 4*d* metals (Zr = 6.52 g cm⁻³ vs. Hf = 13.31 g cm⁻³; Ru = 12.45 g cm⁻³ vs. Os = 22.59 g cm⁻³, the densest naturally occurring element).

2. Gradation of Basicity, Coordination Number, and Rare-Earth Separation

Across the lanthanide series from La³⁺ (103.2 pm) to Lu³⁺ (86.1 pm), the steadily shrinking ionic radius increases the surface charge density (z / r²) and polarizing power of the cation according to Fajans' rules in ionic bonding:

• Decrease in Hydroxide Basicity: Because smaller Lu³⁺ polarizes coordinated OH⁻ ions and H₂O ligands more strongly than larger La³⁺, the covalent character of the Ln–OH bond increases and basicity decreases monotonically from La(OH)₃ (strongest base) to Lu(OH)₃ (weakest base, mildly amphoteric), while the aqueous acidity (hydrolysis constant pK_a) of [Ln(H2O)*n*]3+ increases from La³⁺ (pK_a = 9.0) to Lu³⁺ (pK_a = 7.9).
• Decrease in Aqueous Coordination Number: In aqueous solution, large early lanthanides (La³⁺–Nd³⁺) adopt a 9-coordinate tricapped trigonal prismatic hydration shell [Ln(H2O)9]3+, mid-row lanthanides (Sm³⁺–Eu³⁺) exist in equilibrium between 9 and 8, and smaller late lanthanides (Tb³⁺–Lu³⁺) contract to an 8-coordinate square antiprismatic shell [Ln(H2O)8]3+.
• Chelate Stability and Industrial Separation: Because smaller, higher-charge-density late Ln³⁺ ions bind more tightly to polydentate aminopolycarboxylate chelators (such as EDTA⁴⁻, DTPA⁵⁻) and organophosphorus extractants (such as DEHPA), the log K_f stability constant increases by ~4 orders of magnitude from La³⁺ (log K_EDTA = 15.5) to Lu³⁺ (log K_EDTA = 19.8), providing the thermodynamic basis for industrial counter-current liquid–liquid extraction and ion-exchange chromatography of individual rare-earth elements.

---

Why do europium (Eu) and ytterbium (Yb) show large upward spikes in the metallic radius plot of neutral lanthanides?

Most lanthanide metals contribute three valence electrons (5*d*¹ 6*s*²) to the conduction band in metallic bonding, leaving a trivalent Ln³⁺ ion core in the crystal lattice. In europium (Z = 63, [Xe] 4*f*⁷ 6*s*²) and ytterbium (Z = 70, [Xe] 4*f*¹⁴ 6*s*²), the strong exchange and spherical stabilization of the half-filled 4*f*⁷ and filled 4*f*¹⁴ subshells locks the seventh and fourteenth f electrons into the localized 4*f* core, so Eu and Yb contribute only two electrons (6*s*²) to the metallic band. Because their lattice cores are divalent (Eu²⁺ and Yb²⁺, analogous to alkaline earth Ba²⁺), their metallic radii jump to 204 pm (Eu) and 194 pm (Yb) with much lower densities and melting points.

What is the actinide contraction, and how does it compare to the lanthanide contraction?

The actinide contraction is the analogous decrease in ionic radii across the 5*f* series from Th³⁺/Ac³⁺ (r(Ac³⁺) = 112 pm) to Lr³⁺ (r(Cf³⁺) = 95 pm). Because 5*f* orbitals have one radial node (n − l − 1 = 5 − 3 − 1 = 1), are more radially extended than nodeless 4*f* orbitals, and experience stronger relativistic Dirac expansion, 5*f* electrons shield one another even more poorly than 4*f* electrons while participating directly in covalent bonding in early actinides (U, Np, Pu, oxidation states +3 to +7).

Why is yttrium (Z = 39) always classified with the heavy lanthanides despite having no f electrons?

Yttrium sits in Period 5 directly above lanthanum in Group 3 ([Kr] 4*d*¹ 5*s*²) and forms exclusively the closed-shell Y³⁺ ([Kr]) cation with a 6-coordinate Shannon ionic radius of 90.0 pm. Because the lanthanide contraction shrinks the 6th-period Ln³⁺ cations from 103.2 pm (La³⁺) down to 86.1 pm (Lu³⁺), the radius Curve crosses 90.0 pm right between holmium (Ho³⁺ = 90.1 pm) and erbium (Er³⁺ = 89.0 pm). Sharing both identical +3 charge and identical ionic radius with Ho³⁺/Er³⁺, Y³⁺ substitutes isomorphously into all heavy lanthanide crystal lattices.

---

1. Goldschmidt, V. M., Barth, T., & Lunde, G. (1925). Geochemische Verteilungsgesetze der Elemente V: Isomorphie und Polymorphie der Sesquioxyde. Die Lanthaniden-Kontraktion und ihre Konsequenzen. Skrifter Norske Videnskaps-Akademi i Oslo, I. Mat.-Naturv. Klasse, No. 7, 1–59. https://doi.org/10.1038/116358a0
2. Seth, M., Dolg, M., Fulde, P., & Schwerdtfeger, P. (1995). "Lanthanide and actinide contractions: Relativistic and shell structure effects." Journal of the American Chemical Society, 117(24), 6597–6598. https://doi.org/10.1021/ja00129a026
3. Küchle, W., Dolg, M., & Stoll, H. (1997). "Ab initio study of the lanthanide and actinide contraction." The Journal of Physical Chemistry A, 101(38), 7128–7133. https://doi.org/10.1021/jp970920c
4. Seitz, M., Oliver, A. G., & Raymond, K. N. (2007). "The lanthanide contraction revisited." Journal of the American Chemical Society, 129(36), 11153–11160. https://doi.org/10.1021/ja072750f