Three-Center Four-Electron Bond

A three-center four-electron (3c–4e) bond—also known as the Pimentel–Rundle model—is a quantum-mechanical bonding model in Molecular Orbital theory and Valence Bond theory that explains how heavy main-group p-block elements of Groups 15 through 18 (such as phosphorus, sulfur, chlorine, iodine, and xenon) form stable hypervalent molecules and ions—such as XeF₂, I₃⁻, SF₄, PF₅, and SF₆—without violating the quantum-mechanical octet limit or invoking high-energy outer nd orbital hybridization. Independently formulated in 1951 by George C. Pimentel (at UC Berkeley) and Robert E. Rundle (at Iowa State University), the 3c–4e model combines three colinear p_z (or σ-donor) atomic orbitals across a linear X–A–X triad to generate one bonding molecular orbital (ψ₁), one non-bonding molecular orbital (ψ₂) localized exclusively on the two terminal ligands (X), and one antibonding molecular orbital (ψ₃*).

Populating ψ₁ and ψ₂ with four valence electrons produces a stable linear linkage with a formal covalent bond order of 0.5 per A–X bond, while transferring negative charge onto the electronegative terminal ligands (X^−½–A⁺¹–X^−½) so that the central atom A never exceeds eight valence electrons in its primary coordination shell.

🧪 Interactive Simulation & Curriculum: Inspect the 3D geometries of hypervalent 3c–4e molecules (XeF₂, I₃⁻, SF₄, ClF₃, PF₅, BrF₅, XeF₄, and SF₆) in VSEPR Lab and the VSEPR & Orbital Bonding Mechanics module.

Constructing the Three Molecular Orbitals of a Colinear X–A–X Axis

Consider a linear symmetric three-atom system X₁–A–X₂ aligned along the z-axis, such as xenon difluoride (XeF₂) or the triiodide anion (I₃⁻). The central atom A contributes its valence np_z orbital (φ_A), which has opposite phase signs on its left (−) and right (+) lobes, while the two terminal ligands X₁ and X₂ each contribute a σ-symmetry p_z orbital (φ₁ and φ₂) pointing inward toward A.

Combining these three atomic basis functions (φ₁, φ_A, φ₂) via the Linear Combination of Atomic Orbitals (LCAO) under Hückel/secular theory yields three delocalized σ molecular orbitals:

1. Bonding Molecular Orbital (ψ₁, σ_u symmetry, 0 nodes between atoms):

• ψ₁ = ½ φ₁ + (1/√2) φ_A + ½ φ₂ (Energy E₁ = α + √2 β)
• Both terminal ligand orbitals (φ₁ and φ₂) overlap in-phase with the respective lobes of the central np_z orbital (φ_A), concentrating electron density simultaneously across both A–X₁ and A–X₂ internuclear regions.

1. Non-Bonding Molecular Orbital (ψ₂, σ_g symmetry, 1 nodal plane through central atom A):

• ψ₂ = (1/√2) φ₁ − (1/√2) φ₂ (Energy E₂ = α, exact node at φ_A = 0)
• Because the symmetric combination (φ₁ − φ₂ in p_z coordinates, or g parity) has even (gerade) inversion symmetry whereas the central atom's np_z orbital has odd (ungerade) inversion symmetry, their overlap integral is identically zero. Consequently, the central atom A contributes 0% to ψ₂: the second pair of electrons resides exclusively on the two terminal ligands X₁ and X₂ as a delocalized ligand-centered non-bonding pair.

1. Antibonding Molecular Orbital (ψ₃\*, σ_u\* symmetry, 2 internuclear nodal planes):

• ψ₃* = ½ φ₁ − (1/√2) φ_A + ½ φ₂ (Energy E₃ = α − √2 β)
• Out-of-phase overlap places nodal planes across both A–X₁ and A–X₂ bonds; in a 4-electron system, ψ₃* remains completely unoccupied (LUMO).

Bond Order (0.5), Charge Distribution, and Valence Bond Resonance

In a four-electron system (2 electrons from the central atom's filled np_z² orbital + 1 unpaired electron from each terminal radical X•, totaling 4 electrons), two electrons fill the bonding orbital (ψ₁²) and two electrons fill the ligand-localized non-bonding orbital (ψ₂²). Because only ψ₁ contributes net bonding across the two A–X linkages while ψ₂ is non-bonding:

• Total Bonding Pairs Across X₁–A–X₂: 1 pair shared over 2 bonds ⟹ Bond Order = 0.5 per A–X bond.
• Mulliken Atomic Charge Distribution: The central atom A has electron population 2 × (1/√2)² = 1.00 electron in the p_z manifold (a net formal charge of +1 relative to its initial np_z² pair), while each terminal ligand X₁ and X₂ carries 2 × (½)² [from ψ₁] + 2 × (1/√2)² [from ψ₂] = 0.5 + 1.0 = 1.50 electrons (a net formal charge of −0.5 on each terminal ligand).

In Pauling–Coulson Valence Bond (VB) theory, this 3c–4e MO description is completely equivalent to a 50:50 resonance hybrid between two octet-compliant ionic-covalent canonical structures:

[ X₁—A⁺   :X₂⁻ ] ⟷ [ X₁:^{−   +}A—X₂ ]

The orbital topology of the non-bonding HOMO (ψ₂ = (φ₁ − φ₂)/√2) directly explains three hallmark experimental observations of main-group inorganic chemistry:

1. Why Only Electronegative Ligands (F, O, Cl) Stabilize Hypervalent Compounds: Because the two HOMO electrons in ψ₂ reside exclusively on the two terminal ligands X₁ and X₂ (giving each terminal atom a −0.5 partial charge), thermodynamic stability requires the terminal ligands X to have a higher electronegativity than the central atom A (χ_X > χ_A, so α_X lies low in energy) while the central atom A must have a relatively low ionization energy (as in heavy Group 15–18 elements P, S, Cl, Br, I, Xe). This explains why PF₅, SF₄, SF₆, ClF₃, and XeF₂ are stable isolable compounds, whereas perhydrogenated hypervalent hydrides such as PH₅, SH₄, and SH₆ do not exist as stable ground-state minimums (hydrogen, χ = 2.20, is not electronegative enough to stabilize the ligand-localized ψ₂² negative charge against spontaneous reductive elimination of H₂).
2. Why Axial Bonds Are Longer and Weaker Than Equatorial Bonds in SN = 5 (PF₅, PCl₅, SF₄): In trigonal bipyramidal molecules such as phosphorus pentafluoride (PF₅) and phosphorus pentachloride (PCl₅), the three equatorial ligands bond via conventional localized sp² two-center two-electron (2c–2e) covalent σ bonds (bond order = 1.0), whereas the two colinear 180° axial ligands bond via a single 3c–4e bond using the central phosphorus 3*p*_z orbital (bond order = 0.5). Gas-phase electron diffraction confirms that the axial bonds are systematically longer and weaker than the equatorial bonds:

• PF₅: r(P–F_ax) = 1.577 Å vs. r(P–F_eq) = 1.534 Å
• PCl₅: r(P–Cl_ax) = 2.14 Å vs. r(P–Cl_eq) = 2.02 Å
• SF₄: r(S–F_ax) = 1.646 Å vs. r(S–F_eq) = 1.545 Å
• ClF₃: r(Cl–F_ax) = 1.698 Å vs. r(Cl–F_eq) = 1.598 Å

1. Why Electronegative Substituents Prefer Axial Sites (Bent's Rule in SN = 5): In mixed phosphoranes such as PCl₂F₃ or CH₃PF₄, the most electronegative fluorine ligands always occupy the axial sites because the axial 3c–4e ψ₂ HOMO places −0.5 charge directly onto the axial atoms.

The electron-rich 3c–4e hypervalent bond (found on the right side of the p-block, Groups 15–18) is the mirror image of the three-center two-electron (3c–2e) electron-deficient bond formulated by H. Christopher Longuet-Higgins (1949) and William N. Lipscomb (1976 Nobel Prize) for boron hydrides (such as diborane, B₂H₆, and the trihydrogen cation H₃⁺):

• In a 3c–2e bond (such as the two bridging B–H–B "banana bonds" in diborane B₂H₆), three atomic orbitals combine to form ψ₁ (bonding), ψ₂ (non-bonding), and ψ₃* (antibonding), but only two electrons are available. Both electrons occupy the lowest bonding orbital (ψ₁²) while ψ₂ and ψ₃* remain empty, favoring an acute or bent triangular geometry (such as equilateral D_3h H₃⁺ or curved B–H–B bridges) that maximizes simultaneous three-way overlap.
• In a 3c–4e bond, four electrons populate both ψ₁² and ψ₂²; because ψ₂ has an antibonding node between the two terminal ligands X₁ and X₂, bending the X₁–A–X₂ angle away from 180° brings the out-of-phase lobes of ψ₂ together and raises its energy, which explains why 3c–4e triads strongly prefer 180° colinear geometry.

How does the 3c–4e model relate to the S_N2 transition state and enzymatic phosphoryl transfer?

Every bimolecular nucleophilic substitution (S_N2) reaction at carbon (Nu:⁻ + R₃C–X → [Nu···CR₃···X]^−‡ → Nu–CR₃ + :X⁻) and every associative phosphoryl-transfer reaction catalyzed by kinases, ATP synthase, and DNA polymerases passes through a pentacoordinate trigonal bipyramidal (SN = 5) transition state or phosphorane intermediate. Along the 180° apical reaction coordinate, the attacking nucleophile (Nu, such as a 3′-OH or H₂O oxygen), the central electrophilic atom (carbon's 2*p*_z or phosphorus's 3*p*_z orbital), and the departing leaving group (ADP), form a classic colinear three-center four-electron (3c–4e) bond. Because the central carbon atom in Period 2 is small and sterically crowded, its pentacoordinate 3c–4e species is a saddle-point transition state, whereas the larger Period 3 phosphorus atom can accommodate five ligands closely enough for pentacoordinate phosphoranes (or enzyme-bound PO₃⁻ trigonal bipyramids stabilized by Mg²⁺ and positively charged arginine/lysine side chains) to exist as high-energy intermediates or tight transition states.

Why does solid elemental iodine (I₂) dissolve poorly in pure water but dissolve readily in aqueous potassium iodide (KI)?

Molecular iodine (I₂) is a nonpolar homonuclear diatomic molecule (μ = 0 D) held together in the solid state by London dispersion forces; it cannot disrupt the strong hydrogen-bonding network of pure water, limiting its aqueous solubility at 298 K to 1.3 × 10⁻³ mol L⁻¹ (0.33 g L⁻¹). When solid I₂ is added to an aqueous solution of potassium iodide (KI), the nucleophilic iodide anion (:I⁻) donates a 5*p*_z lone pair into the vacant *σ*\*_5*p* LUMO of I₂ to form the linear, symmetrical triiodide anion (I₃⁻, AX₂E₃) held together by a 3c–4e bond (I₂ + I⁻ ⇌ I₃⁻, K_eq ≈ 700 M⁻¹ at 298 K). Because I₃⁻ is a charged ion with −0.5 charge localized on each terminal iodine atom, it is strongly solvated by water dipoles, increasing apparent iodine solubility by orders of magnitude (the basis of Lugol's iodine and starch-iodine histology staining, where linear polyiodide I₃⁻/I₅⁻ chains thread inside the α-1,4 helical cavity of amylose).

Why are all five fluorine atoms in PF₅ equivalent by ¹⁹F NMR spectroscopy at room temperature if axial and equatorial bonds have different lengths?

Gas-phase electron diffraction (which operates on an instantaneous ~10⁻¹⁸ s vibrational timescale) resolves two distinct P–F bond lengths in PF₅: two longer 3c–4e axial bonds (1.577 Å) and three shorter 2c–2e equatorial bonds (1.534 Å). However, ¹⁹F nuclear magnetic resonance (NMR) spectroscopy operates on a much slower millisecond (~10⁻³ to 10⁻⁶ s) timescale and shows only one sharp doublet for all five fluorine atoms even down to −100 °C. As explained by R. Stephen Berry in 1960 (Berry pseudorotation), the two axial fluorines and two of the equatorial fluorines undergo a concerted low-barrier scissoring vibration (passing through a C_4v square pyramidal transition state with E_a ≈ 15 kJ mol⁻¹ around the remaining equatorial "pivot" bond) that swaps the axial 3c–4e pair and two equatorial 2c–2e pairs more than 10⁵ times per second at room temperature, averaging their magnetic environments on the NMR timescale.

1. Pimentel, G. C. (1951). "The bonding of trihalide and bifluoride ions by the molecular orbital method." The Journal of Chemical Physics, 19(4), 446–448. https://doi.org/10.1063/1.1748245
2. Hach, R. J., & Rundle, R. E. (1951). "The structure of tetramethylammonium pentaiodide." Journal of the American Chemical Society, 73(9), 4321–4324. https://doi.org/10.1021/ja01153a086
3. Berry, R. S. (1960). "Correlation of rates of intramolecular tunneling processes, with application to some Group V compounds." The Journal of Chemical Physics, 32(3), 933–938. https://doi.org/10.1063/1.1730820
4. Reed, A. E., & Schleyer, P. v. R. (1990). "Chemical bonding in hypervalent molecules. The dominance of ionic bonding and negative hyperconjugation over d-orbital participation." Journal of the American Chemical Society, 112(4), 1434–1445. https://doi.org/10.1021/ja00160a022