Hybridization and valence bond theory

Intuition [Beginner]

Carbon has the electron configuration $1s^{2}2s^{2}2p^2$. The $2s$ orbital is spherical and holds two paired electrons. The two $2p$ electrons occupy two of the three dumbbell-shaped $p$ orbitals. If carbon used these orbitals directly, it would form only two bonds using its two unpaired $2p$ electrons. Yet carbon forms four equivalent bonds in methane (${\text{CH}}_4$).

The resolution is hybridization. Carbon mixes its one $2s$ and three $2p$ orbitals to form four new orbitals called $s{p}^3$ hybrids. These four orbitals point toward the corners of a tetrahedron, each at $109.5^{\circ }$ from the others. Each hybrid orbital can hold one electron and form one bond. The four C-H bonds in methane are equivalent in length and strength, as observed.

The key idea: atomic orbitals can be mixed (linearly combined) to form new orbitals with different shapes and directions. The number of hybrid orbitals always equals the number of atomic orbitals combined. Mixing one $s$ and one $p$ gives two $sp$ hybrids (linear, $180^{\circ }$). Mixing one $s$ and two $p$ gives three $s{p}^2$ hybrids (trigonal planar, $120^{\circ }$). Mixing one $s$ and three $p$ gives four $s{p}^3$ hybrids (tetrahedral, $109.5^{\circ }$).

Visual [Beginner]

The three hybridization types and their geometries:

Hybridization	Orbitals mixed	Number of hybrids	Angle	Geometry
$sp$	1 s + 1 p	2	$180^{\circ }$	Linear
$s{p}^2$	1 s + 2 p	3	$120^{\circ }$	Trigonal planar
$s{p}^3$	1 s + 3 p	4	$109.5^{\circ }$	Tetrahedral

Bonds formed by hybrid orbitals: a sigma ($\sigma$) bond results from head-on overlap of two orbitals along the bond axis. A pi ($\pi$) bond results from side-on overlap of two unhybridised $p$ orbitals above and below the bond axis. A double bond is one $\sigma$ + one $\pi$; a triple bond is one $\sigma$ + two $\pi$.

Worked example [Beginner]

Methane ${\text{CH}}_4$. Carbon has $2s^{2}2p^2$. One $2s$ electron is promoted to the empty $2p$ orbital, giving $2s^{1}2p_x^{1}2p_y^{1}2p_z^1$. The $2s$ and three $2p$ orbitals mix to form four $s{p}^3$ hybrids, each pointing to a corner of a tetrahedron. Each hybrid overlaps with a hydrogen $1s$ orbital to form a $\sigma$ bond.

Ethylene ${\text{C}}_2{\text{H}}_4$. Each carbon bonds to three other atoms (one C, two H), requiring three hybrid orbitals. Carbon uses $s{p}^2$ hybridization: one $2s$ + two $2p$ orbitals form three $s{p}^2$ hybrids at $120^{\circ }$ in a plane. The remaining unhybridised $p$ orbital is perpendicular to this plane. The C-C bond has one $\sigma$ bond (from $s{p}^2$-$s{p}^2$ overlap) and one $\pi$ bond (from $p$-$p$ side-on overlap). The $\pi$ bond prevents rotation around the C=C double bond, locking the molecule in a planar geometry.

The $\pi$ bond has two lobes of electron density, one above and one below the molecular plane. It is weaker than the $\sigma$ bond because the side-on overlap is less efficient than head-on overlap. This is why double bonds are stronger than single bonds but less than twice as strong.

Acetylene ${\text{C}}_2{\text{H}}_2$. Each carbon bonds to only two other atoms (one C, one H), requiring two hybrid orbitals. Carbon uses $sp$ hybridization: one $2s$ + one $2p$ orbital form two $sp$ hybrids at $180^{\circ }$ along a line. Two $p$ orbitals remain unhybridised on each carbon. The C-C bond has one $\sigma$ bond (from $sp$-$sp$ overlap) and two perpendicular $\pi$ bonds (from two pairs of $p$-$p$ side-on overlap). The molecule is linear. The C-H bond in acetylene is shorter (1.06 A) than in ethylene (1.08 A) or methane (1.09 A) because the $sp$ hybrid has more $s$-character (50% vs 33% vs 25%), pulling the bonding electrons closer to carbon.

Check your understanding [Beginner]

Formal definition [Intermediate+]

Hybridization is the construction of a new orthonormal set of atomic orbitals by taking linear combinations of the standard atomic orbitals $(s,p_x,p_y,p_z,\dots )$ on a single atom. For an atom with $n_s$ $s$-character and $n_p$ $p$-character:

An $s{p}^n$ hybrid orbital (where $n=n_p/n_s$ is the ratio of $p$ to $s$ contribution) has the form

$${\psi }_{s{p}^n}=\frac{1}{\sqrt{1+n}}{\varphi }_s+\sqrt{\frac{n}{1+n}}{\varphi }_p,$$

where ${\varphi }_p$ is a $p$ orbital oriented along the desired direction. The normalisation ensures $⟨\psi |\psi ⟩=1$, and orthonormality between different hybrids in the set is enforced by choosing the $p$-orbital directions appropriately.

The three standard cases:

$sp$ hybridization. One $s$ + one $p$ orbital give two hybrids at $180^{\circ }$:

$${\psi }_1=\frac{1}{\sqrt{2}}({\varphi }_s+{\varphi }_{p_z}),{\psi }_2=\frac{1}{\sqrt{2}}({\varphi }_s-{\varphi }_{p_z}).$$

Two $p$ orbitals ($p_x$, $p_y$) remain unhybridised. Used in linear molecules (e.g., BeH${}_2$, C${}_2$H${}_2$).

$s{p}^2$ hybridization. One $s$ + two $p$ orbitals give three hybrids at $120^{\circ }$ in a plane:

$${\psi }_1=\frac{1}{\sqrt{3}}{\varphi }_s+\sqrt{\frac{2}{3}}{\varphi }_{p_1},{\psi }_2=\frac{1}{\sqrt{3}}{\varphi }_s+\sqrt{\frac{2}{3}}{\varphi }_{p_2},{\psi }_3=\frac{1}{\sqrt{3}}{\varphi }_s+\sqrt{\frac{2}{3}}{\varphi }_{p_3},$$

where the ${\varphi }_{p_i}$ are $p$ orbitals directed $120^{\circ }$ apart in the $xy$-plane. One $p$ orbital remains unhybridised (perpendicular to the plane). Used in trigonal planar molecules (e.g., BF${}_3$, C${}_2$H${}_4$).

$s{p}^3$ hybridization. One $s$ + three $p$ orbitals give four hybrids at $109.5^{\circ }$:

$${\psi }_i=\frac{1}{2}{\varphi }_s+\frac{\sqrt{3}}{2}{\varphi }_{p_i},i=1,2,3,4,$$

where the four ${\varphi }_{p_i}$ are directed toward tetrahedral corners. Used in tetrahedral molecules (e.g., CH${}_4$, NH${}_3$, H${}_2$O).

Valence bond theory

Valence bond (VB) theory describes a covalent bond as the overlap of two half-filled atomic orbitals, one from each bonded atom. The bond wave function for a diatomic molecule AB is

$${\Psi }_{\text{VB}}={\varphi }_A(1){\varphi }_B(2)+{\varphi }_A(2){\varphi }_B(1),$$

where the two terms ensure that the electrons are indistinguishable (symmetric spatial function) and the spin part is antisymmetric (singlet: one spin up, one spin down). The Heitler-London treatment of H${}_2$ (1927) showed that this exchange term lowers the energy relative to two separated atoms, providing the first quantum-mechanical explanation of the covalent bond.

Overlap integral. The strength of a covalent bond is proportional to the overlap between the bonding orbitals:

$$S_{AB}=\int {\varphi }_A^{\ast }(\mathbf{r}){\varphi }_B(\mathbf{r})d\mathbf{r}.$$

$\sigma$ bonds (head-on overlap) have larger $S$ than $\pi$ bonds (side-on overlap) because the orbital lobes overlap more extensively along the internuclear axis. This is why $\sigma$ bonds are stronger than $\pi$ bonds.

Counterexamples to common slips

• "Hybridization is a physical process that atoms undergo when they form bonds." Hybridization is a change of basis in the space of atomic orbitals, not a dynamical event. The total electronic wave function is invariant under this basis change. Atoms do not "decide" to hybridize; the model is a bookkeeping tool for predicting geometry.

• "The superscript in $s{p}^3$ means the $p$ orbital is three times bigger than the $s$ orbital." The superscript indicates the ratio of $p$-character to $s$-character. An $s{p}^3$ hybrid has 25% $s$-character and 75% $p$-character (ratio 3:1). The contributions are by orbital composition, not by size.

• "An $s{p}^2$ carbon always has bond angles of exactly $120^{\circ }$." The $120^{\circ }$ angle holds only when all substituents are equivalent. In propene, the angles around the $s{p}^2$ carbons deviate from $120^{\circ }$ because the C=C bond, C-H bond, and C-C single bond have different steric and electronic demands.

• "Lone pairs do not count when determining hybridization." The steric number (which determines hybridization type) counts all electron domains: bonds and lone pairs. In water, oxygen has four electron domains (two bonds + two lone pairs), giving $s{p}^3$ hybridization and an approximately tetrahedral electron-pair geometry, even though the molecular shape is bent.

Key theorem with proof [Intermediate+]

Theorem (Angular separation of $s{p}^n$ hybrids). The $N$ equivalent $s{p}^{(N-1)}$ hybrid orbitals formed from one $s$ and $(N-1)$ $p$ orbitals point to the vertices of a regular $(N-1)$-simplex centred at the nucleus. For $N=2,3,4$:

• $N=2$: two $sp$ hybrids at $180^{\circ }$ (linear).
• $N=3$: three $s{p}^2$ hybrids at $120^{\circ }$ (trigonal planar).
• $N=4$: four $s{p}^3$ hybrids at $\arccos (-1/3)\approx 109.5^{\circ }$ (tetrahedral).

Proof. The $N$ hybrid orbitals ${\psi }_1,\dots ,{\psi }_N$ are orthonormal, each containing an equal fraction $1/N$ of $s$-character and $(N-1)/N$ of $p$-character. Write ${\psi }_i=(1/\sqrt{N}){\varphi }_s+\sqrt{(N-1)/N}{\hat{p}}_i$ where ${\hat{p}}_i$ is a unit $p$-type orbital in direction ${\hat{r}}_i$.

Orthonormality requires $⟨{\psi }_i|{\psi }_j⟩=0$ for $i\ne j$:

$$\frac{1}{N}+\frac{N-1}{N}⟨{\hat{p}}_i|{\hat{p}}_j⟩=0.$$

Since $⟨{\hat{p}}_i|{\hat{p}}_j⟩={\hat{r}}_i\cdot {\hat{r}}_j$ for real $p$ orbitals, this gives

$${\hat{r}}_i\cdot {\hat{r}}_j=-\frac{1}{N-1},i\ne j.$$

For $N=2$: ${\hat{r}}_1\cdot {\hat{r}}_2=-1$, so $\theta =180^{\circ }$.

For $N=3$: ${\hat{r}}_i\cdot {\hat{r}}_j=-1/2$, so $\theta =120^{\circ }$.

For $N=4$: ${\hat{r}}_i\cdot {\hat{r}}_j=-1/3$, so $\theta =\arccos (-1/3)\approx 109.47^{\circ }$. $\square$

The tetrahedral angle $\arccos (-1/3)$ is a consequence of the orthonormality requirement on four equivalent hybrid orbitals. It is not an approximation or a fitting parameter.

Bridge. This angular-separation result builds toward 15.01.01, where the tetrahedral geometry of $s{p}^3$ carbon directly produces chirality and the planar geometry of $s{p}^2$ carbon produces E/Z isomerism. The foundational reason that organic stereochemistry exists as a subject is that four equivalent $s{p}^3$ hybrids cannot point in the same hemisphere, forcing substituents into a non-planar arrangement. The tetrahedral angle appears again in 14.12.01 as the geometric parameter that determines the coupling constants and vibrational frequencies observed in NMR and IR spectroscopy. This is exactly the structural constraint that underlies every prediction of three-dimensional molecular shape from Lewis structures.

Exercises [Intermediate+]

Lean formalization [Intermediate+]

Mathlib has no formalisation of hybridization or valence bond theory. Available infrastructure:

• Mathlib.LinearAlgebra.Basis: orthonormal bases and basis construction, directly applicable to the linear-algebra of orbital mixing.
• Mathlib.Analysis.InnerProductSpace: inner product spaces for computing overlap integrals.
• Mathlib.Geometry.Matrix: rotation matrices for orienting hybrid orbital sets.

A formalisation path would be: define atomic orbitals as elements of a finite-dimensional real inner product space; define hybrid orbitals as orthonormal linear combinations; prove the angular separation theorem (key theorem above); prove Coulson's direction theorem (Master tier below); and define sigma/pi bonding as overlap integrals along specified axes. This is a substantial modelling effort with no current Mathlib support. lean_status: none; no lean_module ships with this unit.

Hybridization beyond the ideal types [Master]

The ideal $sp$, $s{p}^2$, and $s{p}^3$ categories describe molecules where all substituents on the central atom are equivalent. Most real molecules have inequivalent substituents, and the hybridization adjusts accordingly. The hybrid orbitals directed toward different substituents then have different $s$- and $p$-character fractions. This is not a failure of the hybridization model; it is a prediction of the model once the orthonormality constraint is applied to non-equivalent directions.

Coulson's direction theorem. The most general relationship between $s$-character and bond angle was given by Coulson ^{[Coulson 1937]}. Let ${\alpha }_i$ denote the $s$-character fraction in hybrid orbital ${\psi }_i$ and ${\theta }_{ij}$ the angle between the directions of ${\psi }_i$ and ${\psi }_j$. If ${\psi }_i$ and ${\psi }_j$ are orthogonal, then:

$$\cos {\theta }_{ij}=-\sqrt{\frac{{\alpha }_i{\alpha }_j}{(1-{\alpha }_i)(1-{\alpha }_j)}}.$$

This reduces to the Key theorem for equivalent hybrids (${\alpha }_i={\alpha }_j=1/N$) and provides a quantitative tool for predicting how bond angles change when one substituent is replaced. In methanol (${\text{CH}}_3\text{OH}$), the oxygen directs more $s$-character into the lone pairs and more $p$-character into the O-H and O-C bonds, widening the H-O-C angle slightly beyond the $109.5^{\circ }$ predicted by equivalent $s{p}^3$.

Bent's rule. Henry Bent formulated a qualitative principle in 1961 ^{[Bent 1961]} that predicts the direction of these hybridization adjustments: atomic $s$-character concentrates in orbitals directed toward electropositive substituents (or lone pairs), and $p$-character concentrates in orbitals directed toward electronegative substituents.

The physical basis is that $s$ orbitals are more compact (lower energy, closer to the nucleus) than $p$ orbitals. Placing more $s$-character in a lone-pair orbital localises the lone pair closer to the nucleus, lowering its energy. Placing more $p$-character in a bond to an electronegative atom (which already pulls electron density toward itself) compensates by keeping the bonding pair in a more extended orbital that reaches toward the electronegative partner.

A quantitative consequence: the H-C-H bond angle in substituted methanes varies with the substituent. In ${\text{CH}}_2{\text{F}}_2$, the F-C-F angle ($108.3^{\circ }$) is less than the ideal tetrahedral $109.5^{\circ }$, while the H-C-H angle ($112.6^{\circ }$) is greater. The C-F bonds have more $p$-character (concentrated toward the electronegative F), and the C-H bonds have more $s$-character, widening the H-C-H angle. Coulson's formula reproduces these angles from the measured $s$-character fractions.

The anomeric effect. In molecules containing the O-C-O fragment (glycosides, acetals, dimethoxymethane), the lone pairs on one oxygen preferentially adopt an antiperiplanar orientation relative to the adjacent C-O bond. This stereochemical preference, called the anomeric effect, arises from hyperconjugative donation of the oxygen lone pair into the ${\sigma }^{\ast }$ orbital of the antiperiplanar C-O bond.

The hybridization picture makes this concrete. The oxygen lone pair occupies a hybrid orbital with high $p$-character (by Bent's rule, since the oxygen also bonds to an electronegative partner). This $p$-rich orbital has lobes extending away from the nucleus, making it geometrically suited for overlap with the ${\sigma }^{\ast }$ orbital when antiperiplanar. The $n_O\to {\sigma }_{C-O}^{\ast }$ donation stabilises the gauche conformation over the anti conformation, which is the opposite of what steric considerations would predict.

The anomeric effect has significant stereochemical consequences in carbohydrate chemistry. In the pyranose form of glucose, the anomeric hydroxyl group preferentially adopts the axial orientation (the $\alpha$-anomer) in the gas phase, despite the equatorial orientation ($\beta$-anomer) being sterically favoured. The axial preference arises because the ring oxygen lone pair is antiperiplanar to the axial C-OH bond, maximising $n_O\to {\sigma }_{C-OH}^{\ast }$ hyperconjugation. In aqueous solution, the anomeric equilibrium reverses (favouring the equatorial $\beta$-anomer) because solvation and steric effects overcome the anomeric stabilisation. The energy difference between anomers is typically 3-8 kJ/mol, a range consistent with a hyperconjugative mechanism rather than a steric one.

Walsh diagrams. A. D. Walsh (1953) ^{[Walsh 1953]} introduced correlation diagrams that predict molecular geometry from the number of valence electrons. A Walsh diagram plots the orbital energies of a molecule as a function of a geometric parameter (typically a bond angle). For AH${}_2$ molecules, the diagram shows that molecules with 4 valence electrons (BeH${}_2$) adopt linear geometry while those with 8 (H${}_2$O) adopt bent geometry.

The mechanism: as the H-A-H angle decreases from $180^{\circ }$, the in-plane hybrid orbital on A (built from $s$ and $p_x$) is destabilised because the increased bending forces electron density into a smaller angular region, raising Pauli repulsion. Meanwhile, the out-of-plane $p_y$ orbital is unaffected. When the $p_y$ orbital is occupied (6 or more electrons), the molecule gains nothing by staying linear and bends to improve orbital overlap with the hydrogen atoms. The crossover point determines the equilibrium geometry.

Walsh diagrams generalise to larger molecules (CH${}_4$, NH${}_3$, H${}_2$O all on the same diagram parametrised by the number of electron pairs) and provide a molecular-orbital rationale for the VSEPR predictions that hybridization explains through a valence-bond lens. The two approaches are complementary: hybridization explains geometry through localised orbital directions, Walsh diagrams explain it through delocalised orbital energies.

For the AH${}_2$ series specifically, the orbital ordering at linear geometry ($180^{\circ }$) is: ${\sigma }_g(1s_A+combination)<{\sigma }_u<{\pi }_u(=p_y,p_z\text{ on A})<{\sigma }_g^{\ast }$. As the molecule bends, the ${\pi }_u$ orbital (which is doubly degenerate at $180^{\circ }$) splits into two components: the in-plane component mixes with the ${\sigma }_u$ orbital and is stabilised by bending, while the out-of-plane $p_y$ component is unaffected. For 4 valence electrons (BeH${}_2$), only ${\sigma }_g$ and ${\sigma }_u$ are occupied; bending raises the ${\sigma }_u$ energy and the molecule stays linear. For 8 valence electrons (H${}_2$O), the ${\pi }_u$-derived orbitals are occupied; the stabilisation upon bending drives the molecule to a bent equilibrium at $\sim 104^{\circ }$.

Bond length, bond strength, and $s$-character. The fraction of $s$-character in a hybrid orbital directly affects the bond length and bond strength of the resulting bond. An $sp$ hybrid (50% $s$-character) is more compact than an $s{p}^2$ hybrid (33% $s$-character), which is more compact than an $s{p}^3$ hybrid (25% $s$-character). The increased $s$-character pulls the electron density closer to the nucleus, shortening the bond and increasing the effective nuclear charge felt by the bonding electrons.

This produces systematic trends. C-H bond lengths decrease with increasing $s$-character: $s{p}^3$ C-H $\approx 1.09$ A, $s{p}^2$ C-H $\approx 1.08$ A, $sp$ C-H $\approx 1.06$ A. The corresponding bond dissociation energies increase: $s{p}^3$ C-H $\approx 439$ kJ/mol, $s{p}^2$ C-H $\approx 464$ kJ/mol, $sp$ C-H $\approx 506$ kJ/mol. The same trend applies to C-C bonds: a bond between two $sp$ carbons ($\sim 1.20$ A) is shorter than one between two $s{p}^2$ carbons ($\sim 1.34$ A), which is shorter than one between two $s{p}^3$ carbons ($\sim 1.54$ A). The $s$-character of the hybrids forming the bond determines the radial extent of the orbital overlap.

Rehybridization during reactions. During a chemical reaction, the hybridization at a carbon atom changes as bonds form and break. In the addition of HBr to ethylene, each $s{p}^2$ carbon of the double bond becomes $s{p}^3$ as the $\pi$ bond breaks and two new $\sigma$ bonds form (one C-H, one C-Br). The geometry at each carbon changes from trigonal planar ($120^{\circ }$) to tetrahedral ($109.5^{\circ }$), a process called rehybridization.

The energy cost of rehybridization is part of the activation barrier. Converting an $s{p}^2$ carbon to $s{p}^3$ requires moving from a lower-energy $s$-rich hybrid (33% $s$) to a higher-energy $s$-poor hybrid (25% $s$), since $p$-character raises the orbital energy. In reactions where the transition state involves partial rehybridization, the Hammond postulate predicts that the transition-state geometry resembles the hybridization state closer in energy. For exothermic additions (like HBr to ethylene), the transition state is early and the carbon retains substantial $s{p}^2$ character. For endothermic processes, the transition state is late and the carbon has moved closer to $s{p}^3$.

This rehybridization picture connects to the kinetic isotope effects used in mechanistic studies. When a carbon rehybridizes from $s{p}^3$ toward $s{p}^2$ in a rate-determining step, the out-of-plane bending vibration softens, lowering the zero-point energy for C-H more than for C-D (because the lighter hydrogen has a higher vibrational frequency). The result is a normal secondary kinetic isotope effect ($k_H/k_D>1$), typically 1.10-1.25 per $\alpha$-deuterium. This diagnostic appears in SN1 substitution (where the reacting carbon goes from $s{p}^3$ to $s{p}^2$ in the carbocation intermediate) and in E1 and E2 elimination reactions.

Hybridization is a model, not a process. The orbitals on a carbon atom in methane are not literally $s{p}^3$ hybrids that were "made" by mixing. The total electronic wave function is the same regardless of whether it is expressed in terms of $s$ and $p$ orbitals or $s{p}^3$ hybrids -- they are related by a unitary (orthogonal) transformation of the basis. Hybridization is a choice of basis that makes the bonding description simpler, not a physical event.

The isoelectronic principle. Molecules with the same number of valence electrons often adopt similar geometries because the same hybridization pattern is favoured. The series ${\text{CH}}_4$, ${\text{NH}}_3$, ${\text{H}}_2\text{O}$, $\text{HF}$, and $\text{Ne}$ all have 8 valence electrons around the central atom and all use $s{p}^3$ hybridization. The geometry changes (tetrahedral, trigonal pyramidal, bent) because lone pairs replace bonding pairs, but the underlying orbital framework is the same. Similarly, ${\text{BH}}_3$ and ${\text{CH}}_3^{+}$ both have 6 valence electrons and are trigonal planar ($s{p}^2$), and ${\text{BeH}}_2$ and ${\text{HgCl}}_2$ both have 4 valence electrons and are linear ($sp$).

The predictive power of this principle extends across the periodic table. ${\text{CCl}}_4$, ${\text{SiCl}}_4$, ${\text{GeCl}}_4$, and ${\text{SnCl}}_4$ are all tetrahedral $s{p}^3$ molecules despite spanning four periods. The hybridization type is determined by the steric number (electron domain count), not by the identity of the central atom. What changes across the periodic table is the quantitative s-character distribution (Bent's rule applies differently depending on the electronegativity of the central atom), not the qualitative hybridization category.

Valence bond theory: limits and extensions [Master]

Hyperconjugation. The stability ordering of carbocations ($3^{\circ }>2^{\circ }>1^{\circ }>{\text{CH}}_3^{+}$) cannot be explained by inductive electron donation alone. The additional stabilisation comes from hyperconjugation: donation of electron density from adjacent C-H $\sigma$ bonds into the empty $p$ orbital of the carbocation. Each C-H bond that is antiperiplanar to the empty orbital contributes a small resonance stabilisation. A tertiary carbocation has nine such C-H bonds; a primary carbocation has three.

Hyperconjugation is a valence-bond concept: it describes the interaction in terms of overlap between a filled bonding orbital on one fragment and an empty orbital on another, with the resonance hybrid lowering the total energy. Molecular orbital theory describes the same interaction as a mixing of the ${\sigma }_{C-H}$ orbital into the empty $p$ orbital, producing a delocalised combination. The energy lowering is identical in both descriptions, but the VB picture makes the origin in specific bond rotations transparent.

The same mechanism explains the preference for the staggered conformation of ethane. In the staggered conformation, each C-H bond on one carbon is antiperiplanar to a C-H bond on the other carbon, maximising hyperconjugative donation from ${\sigma }_{C-H}$ to ${\sigma }_{C-H^{\prime }}^{\ast }$. The eclipsed conformation places donors and acceptors out of alignment, losing this stabilisation. The rotational barrier of ethane ($\sim 12\text{kJ/mol}$) is the energy cost of disrupting the hyperconjugative network.

Hyperconjugation also explains the increased stability of alkenes with more alkyl substitution (tetrasubstituted > trisubstituted > disubstituted > monosubstituted > unsubstituted ethylene). Each additional alkyl group provides additional C-H or C-C $\sigma$ bonds that can donate into the ${\pi }^{\ast }$ orbital of the double bond, lowering the total energy. The stabilisation per C-H hyperconjugative interaction is approximately 4-8 kJ/mol, accumulating to substantial preference for more substituted double bonds. This is the electronic component of Zaitsev's rule in elimination reactions: the more substituted alkene is both more stable (thermodynamic control) and formed faster (kinetic control when the transition state is product-like).

Charge-shift bonding. Shaik and Hiberty identified a class of bonds where the primary stabilisation comes not from covalent sharing (the Heitler-London mechanism) but from the resonance energy between covalent and ionic configurations ^{[Shaik & Hiberty 2008]}. In the VB framework, the total bond energy is:

$$E_{\text{bond}}=E_{\text{covalent}}+E_{\text{ionic}}+E_{\text{resonance}},$$

where $E_{\text{resonance}}$ is the stabilisation from mixing the covalent ($\text{A}\cdot \cdot \text{B}$) and ionic (${\text{A}}^{+}{\text{B}}^{-}$, ${\text{A}}^{-}{\text{B}}^{+}$) structures. For homonuclear diatomics like F${}_2$, the covalent structure is actually repulsive at equilibrium distance; the bond exists entirely because of the resonance energy between covalent and ionic structures. This is charge-shift bonding, distinct from both covalent and ionic bonding.

The F${}_2$ bond ($\sim 159\text{kJ/mol}$) is a charge-shift bond. The Heitler-London covalent contribution at equilibrium is approximately zero; the ionic contribution is small; the resonance between them provides the full bond energy. In contrast, H${}_2$ is a classical covalent bond where the covalent structure is strongly attractive and resonance provides a modest additional stabilisation.

Charge-shift bonding has implications for hybridization: in bonds with substantial charge-shift character, the hybridization description of directional orbital overlap captures the geometry correctly but may overstate the covalent-sharing contribution to the bond energy. The bond direction comes from orbital overlap; the bond strength may come from resonance.

$d$-orbital participation and hypervalency. The $s{p}^{3}d$ and $s{p}^3{d}^2$ hybridizations invoked for ${\text{PCl}}_5$ and ${\text{SF}}_6$ are controversial. Quantum-chemical calculations show that $d$-orbital participation in bonding for third-period elements is small (contribution less than 5% to the bond wave function). The expanded octet is better described by ionic-covalent resonance with charge-shift bonding or by MO theory with 3-centre-4-electron bonds. The $s{p}^3{d}^2$ label is a useful mnemonic but not an accurate description of the electronic structure.

The 3-centre-4-electron bond model (Rundle and Pimentel, 1950s) describes hypervalent bonding without invoking $d$ orbitals. In the linear F-Xe-F fragment of XeF${}_2$, three molecular orbitals form from the three $p_z$ orbitals (one on each atom): a bonding ${\sigma }_g$, a nonbonding ${\sigma }_u$, and an antibonding ${\sigma }_g^{\ast }$. Four electrons occupy the bonding and nonbonding orbitals, producing a net bond order of 1 shared across two Xe-F bonds (0.5 each). This description reproduces the geometry and bond lengths without any $d$-orbital character.

The same 3c-4e model applies to ${\text{SF}}_6$ (six equivalent S-F bonds, each a 3c-4e interaction involving the sulfur $3p$ orbital and two fluorine $2p$ orbitals in an octahedral arrangement) and to the axial bonds of ${\text{PCl}}_5$ (the equatorial bonds are conventional 2c-2e covalent bonds using $s{p}^2$ hybrids on phosphorus). The distinction between axial and equatorial bonds in trigonal bipyramidal molecules -- axial bonds are longer and weaker -- follows naturally from the 3c-4e description: the axial bond order of 0.5 is less than the equatorial bond order of 1.0.

VB theory vs. MO theory: quantitative comparison. Valence bond theory builds molecular wave functions from localised atomic orbitals (bond pairs). Molecular orbital theory builds them from delocalised orbitals spanning the entire molecule. Both are valid quantum-mechanical descriptions; both are approximations to the exact solution of the electronic Schrodinger equation.

The Heitler-London treatment of H${}_2$ gives a bond energy of 3.14 eV (experimental: 4.75 eV). Adding ionic terms (${\text{H}}^{+}{\text{H}}^{-}$ configurations) in the covalent-ionic resonance correction improves this to 4.03 eV. The Hartree-Fock MO treatment gives 3.64 eV. Both miss the full experimental value because neither includes electron correlation. Configuration interaction (mixing excited states into the VB wave function) or post-Hartree-Fock methods recover the remaining energy.

For molecules with strong localisation (most organic compounds), VB theory gives an intuitive and accurate picture. For delocalised systems (conjugated $\pi$ systems, aromatic rings, transition-metal complexes), MO theory is more natural and more accurate. The modern view, championed by Shaik and Hiberty, is that both models are complementary tools for the same underlying quantum mechanics, each providing insight that the other obscures.

Synthesis. The foundational reason that hybridization predicts molecular geometry with such reliability is that orthonormality of the hybrid set forces specific angular relationships between bonding directions, and this is exactly the constraint that Coulson's theorem makes quantitative for non-ideal cases. The central insight of modern VB theory is that the covalent-ionic resonance energy -- not just electron pair sharing -- is often the dominant bonding contribution, generalising the classical Heitler-London picture to include charge-shift bonding as a distinct bonding mechanism. Putting these together with Bent's rule and hyperconjugation provides a unified framework: orbital direction from orthonormality, $s$-character distribution from electronegativity differences, stabilisation from hyperconjugative donation and resonance. The bridge is between the static geometry predicted by hybridization and the dynamic stabilisation mechanisms that determine bond energies. This framework appears again in 14.12.01, where hybridization-dependent C-H stretching frequencies ($sp$: 3300 cm${}^{-1}$, $s{p}^2$: 3100 cm${}^{-1}$, $s{p}^3$: 2900 cm${}^{-1}$) serve as spectroscopic fingerprints for the $s$-character distribution that Bent's rule predicts. The pattern builds toward 15.01.01, where the three-dimensional molecular shapes imposed by hybridization become the substrate for stereochemistry: $s{p}^3$ tetrahedra produce chirality, $s{p}^2$ planes produce E/Z isomerism, and the deviations from ideal angles predicted by Coulson's theorem determine conformational preferences.

Full proof set [Master]

Proposition (Coulson's direction theorem). Let ${\psi }_i=\sqrt{{\alpha }_i}{\varphi }_s+\sqrt{1-{\alpha }_i}{\hat{p}}_i$ and ${\psi }_j=\sqrt{{\alpha }_j}{\varphi }_s+\sqrt{1-{\alpha }_j}{\hat{p}}_j$ be two normalised hybrid orbitals on the same atom, where ${\alpha }_i,{\alpha }_j\in (0,1)$ are the $s$-character fractions and ${\hat{p}}_i,{\hat{p}}_j$ are unit $p$-type orbital contributions directed along unit vectors ${\hat{r}}_i,{\hat{r}}_j$. If ${\psi }_i\perp {\psi }_j$, then

$$\cos {\theta }_{ij}=-\sqrt{\frac{{\alpha }_i{\alpha }_j}{(1-{\alpha }_i)(1-{\alpha }_j)}},$$

where ${\theta }_{ij}$ is the angle between ${\hat{r}}_i$ and ${\hat{r}}_j$.

Proof. Orthogonality requires $⟨{\psi }_i|{\psi }_j⟩=0$. Expanding the inner product:

$$0=\sqrt{{\alpha }_i{\alpha }_j}⟨{\varphi }_s|{\varphi }_s⟩+\sqrt{(1-{\alpha }_i)(1-{\alpha }_j)}⟨{\hat{p}}_i|{\hat{p}}_j⟩+\text{cross terms}.$$

The cross terms vanish because $s$ and $p$ orbitals on the same atom are orthogonal: $⟨{\varphi }_s|{\hat{p}}_i⟩=⟨{\varphi }_s|{\hat{p}}_j⟩=0$. Also $⟨{\varphi }_s|{\varphi }_s⟩=1$ and $⟨{\hat{p}}_i|{\hat{p}}_j⟩={\hat{r}}_i\cdot {\hat{r}}_j=\cos {\theta }_{ij}$ for real orbitals. Substituting:

$$0=\sqrt{{\alpha }_i{\alpha }_j}+\sqrt{(1-{\alpha }_i)(1-{\alpha }_j)}\cos {\theta }_{ij}.$$

Solving for $\cos {\theta }_{ij}$:

$$\cos {\theta }_{ij}=-\frac{\sqrt{{\alpha }_i{\alpha }_j}}{\sqrt{(1-{\alpha }_i)(1-{\alpha }_j)}}=-\sqrt{\frac{{\alpha }_i{\alpha }_j}{(1-{\alpha }_i)(1-{\alpha }_j)}}.\square$$

Corollary (Equivalent hybrids). For $N$ equivalent hybrids, each with $\alpha =1/N$, Coulson's theorem recovers $\cos \theta =-1/(N-1)$, the result of the Angular Separation Theorem.

Proof. Substitute ${\alpha }_i={\alpha }_j=1/N$:

$$\cos \theta =-\sqrt{\frac{(1/N)(1/N)}{(1-1/N)(1-1/N)}}=-\frac{1/N}{(N-1)/N}=-\frac{1}{N-1}.\square$$

Proposition (Conservation of $s$-character). On an atom forming $N$ hybrid orbitals from one $s$ orbital, the $s$-character fractions satisfy ${\sum }_{i=1}^N{\alpha }_i=1$.

Proof. Each hybrid has the form ${\psi }_i=\sqrt{{\alpha }_i}{\varphi }_s+(\text{terms orthogonal to }{\varphi }_s)$. The completeness of the orthonormal hybrid set as a basis for the subspace spanned by the original orbitals requires that the total $s$-orbital character is preserved. Formally, since $\{{\psi }_i\}$ is an orthonormal set spanning the same subspace as $\{{\varphi }_s,{\varphi }_{p_1},\dots \}$, the projection of ${\varphi }_s$ onto the hybrid basis satisfies:

$$1=⟨{\varphi }_s|{\varphi }_s⟩=\sum _{i=1}^N|⟨{\varphi }_s|{\psi }_i⟩{|}^2=\sum _{i=1}^N{\alpha }_i.$$

The last equality uses $⟨{\varphi }_s|{\psi }_i⟩=\sqrt{{\alpha }_i}$. $\square$

This conservation law constrains the possible $s$-character distributions. In water, the two lone-pair hybrids and two bonding hybrids must share the total $s$-character of 1.000. If the bonding hybrids each have $\alpha \approx 0.20$ (as calculated in Exercise 6), the lone-pair hybrids must each have $\alpha \approx 0.30$. The deviation from the ideal $0.25$ for all four reflects Bent's rule in quantitative form.

Connections [Master]

• Lewis structures and VSEPR 14.02.01. The steric number determined from a Lewis structure selects the hybridization type: four electron domains require $s{p}^3$, three require $s{p}^2$, two require $sp$. VSEPR predicts the electron-pair geometry; hybridization provides the orbital-level mechanism for why that geometry arises. The two descriptions are complementary: VSEPR is the electron-counting shortcut, hybridization is the quantum-mechanical rationale.

• Atomic orbitals 14.01.01. The shapes, energies, and quantum numbers of $s$, $p$, $d$, and $f$ orbitals provide the building blocks that hybridization mixes. The radial extent and angular distribution of each atomic orbital determine the directionality and overlap efficiency of the resulting hybrids. Without the concrete shapes of $s$ (spherical) and $p$ (dumbbell) orbitals, the concept of orbital mixing would have no physical content.

• Molecular orbital theory 14.05.01 pending. MO theory is the complementary bonding model. Where VB theory assigns electron pairs to individual bonds constructed from hybrid orbitals, MO theory distributes electrons over molecular-wide orbitals constructed from all atomic orbitals simultaneously. Both describe the same quantum-mechanical reality; the difference is in the basis set used to approximate the wave function. For localised bonding in saturated organic molecules, the VB/hybridization picture is more intuitive; for delocalised $\pi$ systems and spectroscopy, MO theory is more natural.

• Organic stereochemistry 15.01.01. The three-dimensional consequences of hybridization are the foundation of organic stereochemistry. The tetrahedral geometry of $s{p}^3$ carbon gives rise to chirality (four different substituents produce non-superimposable mirror images). The planar geometry of $s{p}^2$ carbon produces E/Z (cis/trans) isomerism: the $\pi$ bond locks the substituents in a plane, preventing rotation. The linear geometry of $sp$ carbon constrains the substituent arrangement to a line. Coulson's theorem predicts the exact bond angles in each case.

• Spectroscopy 14.12.01. Infrared and NMR spectroscopy provide experimental confirmation of hybridization predictions. $sp$ C-H stretches appear at $\sim 3300{\text{cm}}^{-1}$, $s{p}^2$ at $\sim 3100{\text{cm}}^{-1}$, $s{p}^3$ at $\sim 2900{\text{cm}}^{-1}$, reflecting the increasing bond strength (and force constant) with greater $s$-character. In NMR, the C-H coupling constant ${}^1{J}_{CH}$ increases with $s$-character ($s{p}^3$: $\sim 125$ Hz, $s{p}^2$: $\sim 160$ Hz, $sp$: $\sim 250$ Hz) because the Fermi contact interaction is proportional to the electron density at the nucleus, which increases with $s$-character.

Historical & philosophical context [Master]

The valence bond theory was founded by Walter Heitler and Fritz London in 1927 with their quantum-mechanical treatment of the hydrogen molecule ^{[Heitler & London 1927]}. Their calculation demonstrated that the exchange integral (arising from the antisymmetry of the wave function) produces an attractive interaction between two hydrogen atoms, forming a bond. This was the first quantum-mechanical explanation of chemical bonding. The Heitler-London paper appeared in Z. Physik 44, 455-472, and established the conceptual framework that Pauling would later expand.

Linus Pauling extended and popularised the theory in a series of papers (1931-1935) and in The Nature of the Chemical Bond (1939, 2nd ed. 1940, 3rd ed. 1960) ^{[Pauling 1960]}, which introduced the hybridization concept and the resonance description of molecules. Pauling's book is one of the most influential chemistry texts ever written; it unified structural chemistry, crystallography, and the new quantum mechanics into a coherent framework. Pauling received the Nobel Prize in Chemistry in 1954 for this work. John Slater independently developed equivalent ideas (1931) ^{[Slater 1931]}, and the Slater-Pauling theory became the standard bonding model.

The competition between VB theory (Pauling, Slater) and MO theory (Mulliken, Hund) defined the theoretical chemistry of the mid-20th century. By the 1960s MO theory had gained the upper hand in computational chemistry because it was better suited to computer implementation: the Hartree-Fock self-consistent field method is naturally formulated in the MO framework. VB theory experienced a revival in the 1980s through the work of Gerriten, van Lenthe, and especially Sason Shaik and Philippe Hiberty, who showed that VB theory provides qualitative insight that MO theory sometimes obscures -- particularly for understanding bond dissociation, reaction barriers, and charge-shift bonding ^{[Shaik & Hiberty 2008]}.

Coulson's 1937 paper on the directional character of covalent bonds ^{[Coulson 1937]} provided the first quantitative link between hybrid orbital composition and bond angles. Walsh's 1953 series on molecular shapes ^{[Walsh 1953]} offered the MO-theory counterpart: geometry prediction from electron count. Bent's 1961 formulation of his rule ^{[Bent 1961]} completed the qualitative framework for non-ideal hybridization. Together, these three contributions form the modern quantitative basis for hybridization theory.

The concept of hybridization has been criticised as a bookkeeping device rather than a physical mechanism. The criticism is correct in the sense that hybridization is a basis change, not a dynamical process. But as a heuristic for predicting and rationalising molecular geometry, hybridization remains one of the most powerful tools in chemistry. Pauling himself was explicit about this: hybridization is a way of thinking about the wave function that makes its symmetry properties visible, not a claim about the time-evolution of orbitals.

Bibliography [Master]

• Heitler, W. & London, F., "Wechselwirkung neutraler Atome und homopolare Bindung nach der Quantenmechanik", Z. Physik 44 (1927), 455-472. The founding paper of valence bond theory.
• Pauling, L., "The Nature of the Chemical Bond. I-IV", J. Am. Chem. Soc. 53 (1931), 1367-1400 and subsequent papers. Introduction of hybridization.
• Pauling, L., The Nature of the Chemical Bond, 3e (Cornell University Press, 1960). The canonical monograph.
• Slater, J. C., "Directed Valence in Polyatomic Molecules", Phys. Rev. 37 (1931), 481-489. Independent development of directed valence theory.
• Coulson, C. A., "The Directional Character of Covalent Bonds", J. Chem. Soc. (1937), 1570. Coulson's theorem on s-character and bond angles.
• Walsh, A. D., "The Electronic Orbitals, Shapes, and Ground State Configurations of Polyatomic Molecules", J. Chem. Soc. (1953), 2260-2331. Walsh diagrams for geometry prediction.
• Bent, H. A., "Structural Chemistry of Donor-Acceptor Interactions", Chem. Rev. 61 (1961), 275-311. Bent's rule.
• Shaik, S. & Hiberty, P. C., A Chemist's Guide to Valence Bond Theory (Wiley, 2008). Modern VB theory revival, charge-shift bonding.
• Clayden, J., Greeves, N., Warren, S. & Wothers, P., Organic Chemistry, 2e (Oxford, 2012), Ch. 4. Hybridization in organic chemistry context.
• Tro, N. J., Chemistry: A Molecular Approach, 6e (Pearson, 2023), Ch. 10. Introductory treatment.

---

Deepened in Cycle 4 (2026-05-21). All hooks_out targets are proposed. Status remains draft pending Tyler's review and external chemistry reviewer per CHEMISTRY_PLAN §6.