Covalent Bond

A covalent bond is a primary interatomic chemical linkage formed when two or more atoms share pairs of valence electrons within overlapping spatial wavefunctions, producing an electron probability density distribution between the positively charged atomic nuclei that lowers the total energy of the molecule relative to the isolated constituent atoms. First conceptualized as shared electron pairs by Gilbert N. Lewis in 1916 (and given the name covalence by Irving Langmuir in 1919), the quantum-mechanical origin of the covalent bond was solved in 1927 by Walter Heitler and Fritz London in their treatment of the dihydrogen molecule (H₂).

With bond dissociation enthalpies (*D*°) typically ranging from 150 to 1,072 kJ mol⁻¹ (roughly 60 to 430 times the thermal energy RT ≈ 2.48 kJ mol⁻¹ at 298 K), covalent bonds construct the stable primary backbones of organic molecules, silicate minerals, and biological macromolecules—including the peptide bonds of proteins, the phosphodiester bonds of DNA and RNA, and the glycosidic linkages of carbohydrates.

🧪 Interactive Simulation: Explore covalent σ and π bonding domains and 3D molecular shapes in VSEPR Lab.

The Heitler–London Wavefunction and Exchange Stabilization

Classical electrostatics cannot explain how two neutral hydrogen atoms (each consisting of one proton and one 1*s* electron) bind with a dissociation energy of 436 kJ mol⁻¹ at an equilibrium internuclear separation of r₀ = 0.741 Å (74.1 pm). In 1927, Walter Heitler and Fritz London applied Erwin Schrödinger's wave equation to two hydrogen nuclei (A and B) and two electrons (1 and 2). Because electrons are indistinguishable fermions, assigning electron 1 to atom A (φ_A(1)*φ*_B(2)) is physically indistinguishable from assigning electron 2 to atom A (φ_A(2)*φ*_B(1)). Combining both configurations yields the spatial singlet bonding wavefunction (Ψ₊, partnered with an antisymmetric spin-paired singlet state ↑↓) and the triplet antibonding wavefunction (Ψ₋, partnered with parallel spins ↑↑):

Ψ_±(1, 2) = [ φ_A(1)*φ*_B(2) ± φ_A(2)*φ*_B(1) ] / √[ 2 (1 ± S_AB²) ]

where S_AB = ∫ φ_A(1)*φ*_B(1) d*τ* is the dimensionless spatial overlap integral (0 ≤ S_AB ≤ 1; for H₂ at 0.74 Å, S_AB ≈ 0.75). Evaluating the expectation value of the electronic Hamiltonian (E_± = ⟨*Ψ*_±|*Ĥ*|*Ψ*_±⟩) shows that over 85% of the covalent binding energy of H₂ arises from the quantum-mechanical exchange integral (K)—constructive interference in Ψ₊ delocalizes both electrons across both nuclei simultaneously.

Virial Theorem Analysis: Kinetic Contraction Followed by Potential Drop

Detailed quantum-mechanical analysis by Klaus Ruedenberg using the molecular virial theorem (2⟨*T*⟩ + ⟨*V*⟩ = −*R* (d*E*/d*R*), which reduces to ⟨*T*⟩ = −*E* and ⟨*V*⟩ = 2*E* at the equilibrium bond length d*E*/d*R* = 0) reveals a two-stage physical mechanism for covalent bond formation:

1. Constructive Interference and Electronic Kinetic Energy Lowering: As two atoms approach from infinity, constructive overlap of their atomic wavefunctions (φ_A + φ_B) creates a smoother, more delocalized wavefunction in the internuclear region. Because electronic kinetic energy scales with the squared gradient (curvature) of the wavefunction (T ∝ ∫ |∇*Ψ*|² d*τ*), delocalizing the electron pair between the two nuclei initially lowers the electronic kinetic energy.
2. Orbital Contraction Toward the Nuclei: In response to this relaxed kinetic pressure, the shared electron cloud contracts radially toward both positively charged nuclei, raising electronic kinetic energy back up (so that at equilibrium Δ*T* = −Δ*E* > 0) while producing a twice-as-large drop in electron-nucleus Coulombic potential energy (Δ*V* = 2Δ*E* < 0) that binds the molecule into a deep Morse potential well:

V(r) = D_e [ 1 − exp(−*a*(r − r₀)) ]²

Head-On Sigma Overlap versus Lateral Pi Overlap

Covalent bonds are classified by the rotational symmetry of their overlapping orbitals around the internuclear axis (z-axis):

• Sigma (σ) Bonds: Formed by head-on (end-to-end) coaxial overlap of two s orbitals, an s and a p_z orbital, two p_z orbitals, or two directional orbital hybridization lobes (sp, sp², sp³). A σ bond possesses cylindrical (C_∞v) symmetry around the internuclear axis with zero nodal planes containing the axis, permitting free internal rotation around single bonds at ambient temperatures (e.g., the 12 kJ mol⁻¹ torsional barrier of ethane).
• Pi (π) Bonds: Formed by lateral (side-by-side) overlap of two parallel unhybridized p_x or p_y (or d_xz/d_yz) orbitals above and below a nodal plane that contains the internuclear axis. Because lateral overlap (S_π ≈ 0.27 in C=C at 1.34 Å) is smaller than coaxial head-on overlap (S_σ ≈ 0.65), the π component of a double bond is weaker than the underlying σ component (in ethene, *D*°(C=C) = 614 kJ mol⁻¹ vs. *D*°(C–C) = 348 kJ mol⁻¹, giving a π-bond energy of ~266 kJ mol⁻¹). Rotating one CH₂ terminus of ethene by 90° reduces the p_z–p_z overlap integral to zero (cos 90° = 0), creating a ~264 kJ mol⁻¹ rotational barrier that locks cis (Z) and trans (E) alkene diastereomers in place.

Polar Covalent Bonds, Ionic Character, and Coordinate Linkages

Purely nonpolar covalent bonds occur only when the bonded atoms have identical or nearly identical Pauling electronegativity (Δχ < 0.4, as in H–H, Cl–Cl, O=O, and C–H where Δχ = 2.55 − 2.20 = 0.35). When two covalently bonded atoms differ substantially in electronegativity (0.4 ≤ Δχ ≤ 1.8, as in O–H with Δχ = 1.24, C–O with Δχ = 0.89, or C–F with Δχ = 1.43), the shared electron pair is drawn asymmetrically toward the more electronegative nucleus, creating a polar covalent bond bearing partial atomic charges (δ⁺ and δ⁻) and a bond electric dipole moment. Rather than a sharp binary dichotomy, covalent and ionic bonding form a continuous quantum-mechanical spectrum parameterized by Pauling's empirical fractional ionic character: f_ionic ≈ 1 − exp[−¼(Δχ)²].

When both electrons of the shared pair originate from a lone pair on one donor atom (a Lewis base) and are donated into an empty orbital on an acceptor atom (a Lewis acid), the linkage is termed a coordinate covalent bond (or dative bond). Once formed, a coordinate covalent bond in a symmetric species (such as the four equivalent N–H bonds of the ammonium ion NH₄⁺ formed from NH₃ + H⁺, or the four B–F bonds of BF₄⁻) is physically and spectroscopically indistinguishable from a standard covalent bond.

Breaking a covalent bond A–B requires an energy input equal to the bond dissociation enthalpy (*D*° > 0, an endothermic process), whereas forming a covalent bond releases the same quantity of heat (Δ*H* < 0, exothermic). Cleavage proceeds via one of two electronic pathways:

1. Homolytic Cleavage (Homolysis, A–B → A• + B•): Each departing atom retains one electron of the shared pair, generating two neutral free radicals with unpaired electron spins. Homolysis dominates in the gas phase, under ultraviolet photolysis (such as stratospheric UV-C cleavage of Cl₂ or C–Cl bonds in chlorofluorocarbons), and in radical-SAM or adenosylcobalamin (vitamin B₁₂, Co–C bond *D*° ≈ 130 kJ mol⁻¹) metalloenzyme catalysis.
2. Heterolytic Cleavage (Heterolysis, A–B → A⁺ + :B⁻): The more electronegative atom retains both electrons of the bonding pair as a lone pair, generating a cation and an anion. Although heterolysis requires much more energy than homolysis in the vacuum gas phase (because separating opposite charges q⁺ and q⁻ without dielectric screening costs hundreds of kilojoules per mole), polar protic solvents like liquid water (ε_r = 78.4) and pre-organized enzyme active sites stabilize the nascent ions through hydration and electrostatic complementation, making heterolytic nucleophilic and acid-base pathways the dominant mechanism of cellular metabolism.

Why is the nitrogen-nitrogen triple bond (N≡N, 945 kJ mol⁻¹) nearly six times stronger than the N–N single bond (163 kJ mol⁻¹), whereas the C≡C triple bond (839 kJ mol⁻¹) is only 2.4 times stronger than the C–C single bond (348 kJ mol⁻¹)?

In ethane (H₃C–CH₃), neither carbon atom carries a non-bonding lone pair, so the single C–C σ bond is strong (348 kJ mol⁻¹), and adding two weaker lateral 2*p*–2*p* π bonds in ethyne (HC≡CH, 839 kJ mol⁻¹) increases total bond enthalpy by less than a factor of three. By contrast, in hydrazine (H₂N–NH₂), peroxide (HO–OH, *D*°(O–O) = 146 kJ mol⁻¹), and fluorine (F–F, *D*° = 159 kJ mol⁻¹), the short Period 2 single bond forces non-bonding lone pairs on the two adjacent atoms into close proximity (1.42–1.48 Å), creating severe inter-atomic lone-pair–lone-pair Pauli repulsion that destabilizes the single bond. When two nitrogen atoms form a triple bond in dinitrogen (:N≡N:, 1.098 Å), the lone pairs point outward at 180° on opposite ends of the molecule (zero LP–LP repulsion across the bond) while small 2*p* orbitals achieve exceptionally tight π overlap, making N₂ kinetically inert to most organisms absent the FeMo-cofactor of nitrogenase.

Does breaking the terminal covalent P–O bond in adenosine triphosphate (ATP) "release energy" on its own?

No—breaking any chemical bond, including the terminal phosphoanhydride P–O covalent bond of adenosine triphosphate (ATP), is endothermic (requires an input of energy to pull shared electrons away from positively charged nuclei). The net exergonic free energy released during cellular ATP hydrolysis (ATP⁴⁻ + H₂O → ADP³⁻ + HPO₄²⁻ + H⁺, Δ*G*°′ = −30.5 kJ mol⁻¹ and cellular Δ*G* ≈ −50 to −60 kJ mol⁻¹) arises because the new covalent O–H and P=O bonds formed in the hydrolysis products (ADP and inorganic orthophosphate, P_i), together with relieved electrostatic repulsion among the four negative charges on the polyphosphate tail, greater resonance delocalization in HPO₄²⁻, and stronger hydration shells around the separated products, release more energy than was consumed to cleave the reactant P–O and H–OH bonds.

Why are disulfide bonds (S–S) the primary covalent cross-links used to fold proteins outside the cell, rather than inside the cytosol?

Protein tertiary structure inside the cell relies primarily on non-covalent interactions (hydrophobic effect, hydrogen bonds, and salt bridges) so that chaperones and proteasomes can dynamically remodel polypeptides. Furthermore, the eukaryotic and bacterial cytosol is maintained at a reducing redox potential (E′ ≈ −260 to −290 mV) by millimolar concentrations of reduced glutathione (GSH, at a GSH:GSSG ratio > 50:1) and thioredoxin, which reduce disulfide linkages back to free cysteine thiols (2 R–SH). Inside the oxidizing lumen of the endoplasmic reticulum (E′ ≈ −180 mV) and in the extracellular environment, protein disulfide isomerase (PDI) and Ero1 oxidize paired cysteine thiols into covalent disulfide bonds (R–S–S–R, *D*° = 266 kJ mol⁻¹), locking secreted hormones (such as insulin), digestive enzymes, and antibodies against thermal or proteolytic unfolding in the bloodstream.

1. Lewis, G. N. (1916). "The atom and the molecule." Journal of the American Chemical Society, 38(4), 762–785. https://doi.org/10.1021/ja02261a002
2. Heitler, W., & London, F. (1927). "Wechselwirkung neutraler Atome und homöopolare Bindung nach der Quantenmechanik." Zeitschrift für Physik, 44(6–7), 455–472. https://doi.org/10.1007/BF01397394
3. Ruedenberg, K. (1962). "The physical nature of the chemical bond." Reviews of Modern Physics, 34(2), 326–376. https://doi.org/10.1103/RevModPhys.34.326
4. Pauling, L. (1960). The Nature of the Chemical Bond and the Structure of Molecules and Crystals (3rd ed.). Cornell University Press. ISBN 978-0801403330