Atomic Radius

Atomic radius is a quantitative measure of the spatial size of an atom, defined as the characteristic distance from the center of the atomic nucleus to the outer boundary of its surrounding electron configuration cloud, typically expressed in picometers (1 pm = 10⁻¹² m) or ångströms (1 Å = 100 pm = 10⁻¹⁰ m). In quantum mechanics, the radial wavefunction R_n,*l*(r) of a bound electron decays with an exponential tail e^{−*Z*_effr / (na₀)} that approaches zero only asymptotically as r → ∞. Because an isolated atom lacks a sharp physical outer surface, its "radius" cannot be measured directly on a single free atom; instead, atomic radii are defined operationally as half the equilibrium internuclear distance (d_A–A / 2) between identical or known atoms in molecules, metals, or crystals measured by X-ray, electron, or neutron diffraction, or theoretically as the radius of maximum radial probability density (r_max) in self-consistent Dirac–Fock wavefunctions.

Across the periodic table, empirical covalent radii of ground-state atoms span nearly an order of magnitude—from 31 pm in helium and hydrogen up to 244–260 pm in cesium and francium. Atomic radius governs steric hindrance, bond lengths, crystal packing density, ionization energy, electron affinity, and polarizability.

---

Because the internuclear distance between two atoms depends strongly on the nature of the chemical interaction between them, chemists employ four distinct operational scales of atomic size:

1. Covalent Radius (r_cov): Defined as half the internuclear distance (d_A–A / 2) between two identical atoms joined by a homonuclear single covalent bond (e.g., in Cl₂, d_Cl–Cl = 198 pm, so r_cov(Cl) = 99 pm; in diamond, d_C–C = 154 pm, so r_cov(C) = 77 pm). For elements that do not form homonuclear single bonds, single-bond covalent radii are derived additively from heteronuclear bond lengths (d_A–B ≈ r_cov(A) + r_cov(B), corrected by the Schomaker–Stevenson equation d_A–B = r_A + r_B − *c*|*χ*_A − χ_B| for bond polarity). Double-bond and triple-bond covalent radii (e.g., C=C at 67 pm and C≡C at 60 pm) are 10–22% shorter due to lateral π-orbital overlap and increased s-character under orbital hybridization.
2. Metallic Radius (r_met): Defined for metallic elements as half the nearest-neighbor internuclear distance between adjacent metal cores in the crystalline metallic lattice at 298 K (e.g., in sodium metal, d_Na–Na = 372 pm, giving r_met(Na) = 186 pm). Because atomic size increases slightly with coordination number (CN), Goldschmidt metallic radii normalize all values to an ideal close-packed 12-coordinate FCC/HCP lattice (multiplying 8-coordinate BCC radii by 1/0.97).
3. Van der Waals Radius (r_vdW): Defined by Arnold Bondi (1964) as half the minimum equilibrium internuclear contact distance between two non-bonded atoms of the same element in adjacent molecules or in condensed noble gas solids held together only by van der Waals force and London dispersion force interactions. For any given nonmetal, the van der Waals radius is always substantially larger than the single-bond covalent radius (e.g., for chlorine, r_vdW = 175 pm vs. r_cov = 99 pm), because covalent bonding requires interpenetration of valence orbitals whereas van der Waals contact occurs where Pauli exclusion repulsion balances dispersion attraction:

r_vdW > r_met > r_cov (single) > r_cov (double) > r_cov (triple)

1. Shannon–Prewitt Ionic Radius (r_ion): Derived by partitioning experimental cation–anion internuclear distances (d_M–X = r₊ + r₋) in crystalline ionic bonding oxides and fluorides using high-resolution X-ray electron-density maps (assigning 6-coordinate O²⁻ = 140 pm and F⁻ = 133 pm).

---

Contraction Across a Period: Effective Nuclear Charge (Z_eff)

From the radial Schrödinger equation, the expectation value of an orbital's radius ⟨*r*⟩_n,*l* scales approximately as:

⟨*r*⟩_n,*l* ≈ (a₀ / 2*Z*_eff) × [3*n*² − l(l + 1)]

where a₀ = 52.918 pm is the Bohr radius and Z_eff = Z − σ is the effective nuclear charge after subtracting the Slater screening constant (σ). According to John C. Slater's empirical rules (1930):

• Electrons in higher principal shells (n + 1, ...) contribute 0 to σ.
• Each other electron in the same (ns, np) valence group shields the test electron by only 0.35 (0.30 for 1*s*).
• Each electron in the underlying (n − 1) shell shields an ns or np electron by 0.85, and each electron in (n − 2) or deeper shells shields by 1.00.

Consequently, as one traverses from left to right across a main-group period (e.g., Period 2 from Li [Z = 3] to F [Z = 9], or Period 3 from Na [Z = 11] to Cl [Z = 17]), each step adds one proton (+1 to Z) and one valence electron to the same principal shell (+0.35 to σ). The net effective nuclear charge climbs by Δ*Z*_eff = +0.65 per column (from Z_eff = 1.30 in Li to 5.20 in F), pulling the entire valence shell inward and shrinking the single-bond covalent radius continuously from 128 pm (Li) → 96 pm (Be) → 84 pm (B) → 76 pm (C) → 71 pm (N) → 66 pm (O) → 57 pm (F).

Expansion Down a Group and d-Block / f-Block Contractions

Moving down a column (group) of the periodic table (e.g., Group 1 from Li → Na → K → Rb → Cs), each step increments the principal quantum number n by +1, adding a new outer radial shell with an extra spherical radial node (n − l − 1). Although the nuclear charge Z increases by +8, +18, or +32, the newly added inner core shells screen the nucleus almost completely, so the n² numerator term dominates ⟨*r*⟩ and atomic radius increases down the group (covalent radii: Li 128 pm < Na 166 pm < K 203 pm < Rb 220 pm < Cs 244 pm).

Two major structural contractions modify these vertical trends in the lower half of the periodic table:

1. The d-Block (Scandide) Contraction: Between Calcium (Z = 20) and Gallium (Z = 31) in Period 4, ten electrons are added to the 3*d* subshell (Sc–Zn). Because diffuse, multi-lobed 3*d* orbitals shield the outer 4*s* and 4*p* electrons less completely than s or p core shells, gallium (r_cov = 122 pm) has a slightly smaller covalent radius than aluminum directly above it (r_cov = 121–126 pm; metallic radius Ga 135 pm vs. Al 143 pm).
2. The Lanthanide Contraction: Across the 6th period between lanthanum (Z = 57) and lutetium (Z = 71), 14 electrons populate the deeply buried 4*f* subshell, whose poor radial shielding—reinforced by relativistic 6*s* contraction—shrinks the 5*d* transition metal elements so that hafnium (Z = 72, r_met = 159 pm) is virtually identical in radius to 4*d* zirconium (Z = 40, r_met = 160 pm).

Cation Contraction, Anion Expansion, and Isoelectronic Series

• Cations are always smaller than their parent neutral atoms (r(Mⁿ⁺) < r(M)): Removing valence electrons often empties the outermost principal shell n completely (e.g., Na [186 pm] → Na⁺ [102 pm]) and reduces inter-electronic repulsion (σ), increasing Z_eff on the remaining electrons.
• Anions are always larger than their parent neutral atoms (r(Xⁿ⁻) > r(X)): Adding electrons increases electron–electron repulsion (σ rises, Z_eff drops), expanding the valence cloud (e.g., F [57 pm] → F⁻ [133 pm]; Cl [102 pm] → Cl⁻ [181 pm]).
• Isoelectronic Series: Across a series of ions possessing the exact same number of electrons (e.g., the 10-electron [Ne] series: O²⁻ [140 pm] > F⁻ [133 pm] > Na⁺ [102 pm] > Mg²⁺ [72 pm] > Al³⁺ [53.5 pm]), ionic radius decreases monotonically with increasing nuclear charge Z (+8 → +13).

---

Why do transition metal atomic radii decrease at the start of a d-series, flatten out in the middle, and increase slightly at the end?

Across the 3*d* series from Sc (162 pm) to Cr/Mn/Fe/Co/Ni (125–128 pm), adding protons while filling the inner 3*d* subshell increases Z_eff and contracts the metallic radius, reinforced by strengthening metallic bonding as bonding d-band states fill. Toward the end of the row in copper (128 pm) and zinc (134 pm), the 3*d*¹⁰ subshell becomes completely filled; enhanced inter-electronic d–d repulsion inside the crowded 3*d* shell plus population of antibonding d-band states causes the interatomic spacing to expand slightly.

How does the spin state of a transition metal ion affect its ionic radius?

In an octahedral coordination complex, the five d orbitals split into lower-energy non-bonding t_2g orbitals (pointing between the ligands) and higher-energy σ-antibonding e_g\* orbitals (pointing directly at the ligands) according to crystal field theory. For d⁴–d⁷ ions such as Fe²⁺ (d⁶), a high-spin state (t_2g⁴e_g\*²) places two electrons in the σ-antibonding e_g\* orbitals, lengthening metal–ligand bonds and yielding a significantly larger Shannon ionic radius (78 pm) than the low-spin state (t_2g⁶e_g\*⁰, 61 pm).

Why are noble gas radii often listed as larger than the halogens immediately preceding them?

Because lighter noble gas atoms (He, Ne, Ar) do not form stable homonuclear covalent bonds, tables that mix operational definitions compare the covalent radius of fluorine (57 pm) or chlorine (102 pm) against the non-bonded van der Waals radius of neon (154 pm) or argon (188 pm). When compared on the same quantum-mechanical scale (calculated self-consistent field r_max of the valence shell) or comparing van der Waals radii (F 147 pm vs. Ne 154 pm; Cl 175 pm vs. Ar 188 pm), the calculated valence shell of neon is actually smaller than fluorine's (r_max = 38 pm for Ne vs. 42 pm for F) due to higher Z_eff.

---

1. Cordero, B., Gómez, V., Platero-Prats, A. E., Revés, M., Echeverría, J., Cremades, E., Barragán, F., & Alvarez, S. (2008). "Covalent radii revisited." Dalton Transactions, 2008(21), 2832–2838. https://doi.org/10.1039/B801115J
2. Shannon, R. D. (1976). "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides." Acta Crystallographica Section A, 32(5), 751–767. https://doi.org/10.1107/S0567739476001551
3. Bondi, A. (1964). "van der Waals volumes and radii." The Journal of Physical Chemistry, 68(3), 441–451. https://doi.org/10.1021/j100785a001
4. Rahm, M., Hoffmann, R., & Ashcroft, N. W. (2016). "Atomic and ionic radii of elements 1–96." Chemistry – A European Journal, 22(41), 14625–14632. https://doi.org/10.1002/chem.201602949