Pi Bonds and the Double Bond

The σ+π decomposition is not a historical curiosity confined to textbook ethylene — it is the operating logic behind nearly every reactive organic functional group and behind entire classes of modern technology. In pharmaceutical chemistry, the electrophilic addition chemistry of alkenes (epoxidation, dihydroxylation, hydroboration, halogenation) is always initiated by the exposed π HOMO reacting with an electrophile or metal center, exactly as worked through in Example 3 above; drug candidates bearing Michael acceptors (α,β-unsaturated carbonyls) exploit the same π-system exposure to react covalently with cysteine thiols in targeted covalent inhibitors (e.g., osimertinib, ibrutinib). In materials science, the π systems of conjugated polymers (polyacetylene, polythiophene, and the materials behind OLED displays and organic photovoltaics) rely on extended, delocalized side-on p-orbital overlap across many centers — a direct scale-up of the two-center π bond developed here, and the entry point to conjugation. In spectroscopy, the π→π electronic transition (the HOMO-to-LUMO promotion between the bonding and antibonding combinations derived above) is the basis of UV–visible absorption in every conjugated chromophore, from retinal in the visual cycle to the dyes in sunscreens and textiles. And in structural biology, the planarity enforced by π-bond formation — the very geometric constraint mechanistically derived in this article — underlies the rigidity of the peptide bond (itself π-delocalized between C, O, and N) and is the direct conceptual ancestor of <a class="wiki-link" href="/wiki/chem-orgo-restricted-rotation-and-molecular-rigidity" data-track="wiki_article.body.wiki_link">Restricted Rotation and Molecular Rigidity</a>, which governs everything from cis/trans* proline isomerization in protein folding to the design of rigid pharmacophores.

Ask the tutor: How does the σ+π picture change once we move from an isolated double bond to a conjugated diene, where p orbitals overlap across more than two centers?

Start from a single carbon atom promoted and hybridized to sp². Mixing the 2s orbital with two of the three 2p orbitals produces three sp² hybrids lying in a plane, each pointing 120° from its neighbors — these will make three σ bonds (to two H atoms and to the other carbon, in ethylene). The mixing consumed only two of the three p orbitals; one 2p orbital, conventionally called $2p_z$, is left completely untouched, oriented perpendicular to the sp² plane, with one lobe above the plane and one lobe below it, of opposite phase (sign of the wavefunction). When two such sp²-hybridized carbons are brought together, first their sp² hybrids point directly at each other along the internuclear axis and overlap head-on, forming a $\sigma_{CC}$ bond exactly as in ethane — cylindrically symmetric, and rotatable in principle. But now each carbon also carries its leftover, perpendicular $2p_z$ orbital. If the two CH₂ planes are coplanar, these two $2p_z$ orbitals are parallel to each other, both perpendicular to the molecular plane, and their lobes can overlap side-on: top lobe with top lobe, bottom lobe with bottom lobe. Same-phase (same-sign) overlap on both the top and bottom gives constructive interference — a bonding π molecular orbital, with a probability density that arches over and under the σ framework, and a node exactly in the plane containing all six atoms of ethylene, where the wavefunction amplitude is exactly zero by symmetry (equal and opposite contributions cancel). The antibonding combination, $\pi^*$, comes from combining the p orbitals out of phase: now there are two nodal surfaces — the original σ-framework nodal plane, plus a new perpendicular nodal plane bisecting the C–C bond, where the bonding lobes of one carbon meet the antibonding-phase lobes of the other and cancel. $\pi^*$ is empty in ground-state ethylene and is the LUMO — the orbital that accepts electron density from nucleophiles or reducing agents.

The second figure encodes the geometric heart of the matter: the strength of side-on overlap scales with $\cos\theta$, where $\theta$ is the dihedral twist between the two $2p_z$ axes. At $\theta = 0°$ (planar, as in ground-state ethylene) overlap is maximal; at $\theta = 90°$ the two p orbitals are orthogonal and overlap vanishes exactly — this is the geometric statement underlying the roughly 65 kcal/mol rotational barrier in ethylene, since twisting to 90° costs essentially the entire π-bond strength.

The molecular orbitals formed from two atomic $2p_z$ orbitals, $\phi_A$ and $\phi_B$, are linear combinations:

where:

• $\phi_A, \phi_B$ — the unhybridized $2p_z$ atomic orbitals on carbons A and B, both perpendicular to the sp² plane
• $c_1, c_2$ — normalization coefficients (equal in magnitude for a symmetric, homonuclear case like C=C)
• $\psi_\pi$ — bonding π MO, lower in energy, occupied by the π electron pair
• $\psi_{\pi^*}$ — antibonding π* MO, higher in energy, empty in the ground state

The overlap integral that determines bonding strength depends explicitly on orientation:

where:

• $S(\theta)$ — the p–p overlap integral at twist angle $\theta$
• $S_0$ — the maximum overlap, achieved at $\theta = 0$ (perfectly parallel p orbitals)
• $\theta$ — the dihedral angle between the two p-orbital axes (0° in planar ethylene)

The energy splitting between $\psi_\pi$ and $\psi_{\pi^*}$, in simple two-orbital (Hückel-like) perturbation theory, is:

where:

• $\alpha$ — the Coulomb integral, approximately the energy of an electron in an isolated $2p_z$ orbital
• $\beta$ — the resonance (exchange) integral, negative in sign, whose magnitude grows with overlap $S(\theta)$
• $E_\pi, E_{\pi^*}$ — energies of the bonding and antibonding π MOs

Reading this: because $\beta$ is negative, $E_\pi = \alpha + \beta$ lies below $\alpha$ (stabilized) and $E_{\pi^*} = \alpha - \beta$ lies above $\alpha$ (destabilized) by an equal amount $|\beta|$ — and since $|\beta|$ scales with $S(\theta) \propto \cos\theta$, twisting the alkene continuously weakens the bonding stabilization until, at $\theta = 90°$, $\beta \to 0$ and both MOs collapse to the nonbonding energy $\alpha$: the π bond is gone.

Example 1 — Foundational: counting σ and π bonds in ethylene explicitly. Ethylene, $\mathrm{C_2H_4}$, molecular formula confirmed by mass spectrometry (28.05 g/mol) and by NMR (a single vinyl ¹H environment, singlet-like pattern under first-order approximation, consistent with the D₂ₕ-symmetric planar structure). Each carbon is sp² hybridized: 3 hybrid orbitals + 1 unhybridized $2p_z$. Bond inventory:

• C1–H bond: C1 sp² + H 1s, head-on, σ. (×2 on C1)
• C2–H bond: C2 sp² + H 1s, head-on, σ. (×2 on C2)
• C1–C2 bond: C1 sp² + C2 sp², head-on, σ. (×1)
• C1–C2 second bond: C1 $2p_z$ + C2 $2p_z$, side-on, π. (×1)

Total σ bonds: 5. Total π bonds: 1. Total bonds: 6, matching the Lewis structure H₂C=CH₂ where each atom achieves an octet (carbon) or duet (hydrogen): carbon has 3 σ + 1 π = 4 bonds = correct valence; hydrogen has 1 σ = correct valence. Geometric check: H–C–H bond angle in ethylene is measured (by electron diffraction/microwave spectroscopy) at 117.4°, close to but distinctly less than the idealized sp² angle of 120°. This small deviation (2.6°) is explained by Bent's rule reasoning (heavier π-bonding character pulls slightly more p-character into the C=C-directed hybrid, leaving marginally more s-character, hence a marginally wider angle, in the two C–H hybrids) — a refinement on top of, not a contradiction of, the basic σ+π picture. The bond length C=C is 1.339 Å, shorter than the C–C single bond of 1.54 Å in ethane by 0.20 Å, directly attributable to the extra π bond pulling the nuclei closer.

Example 2 — Applied: predicting and rationalizing the loss of conjugation upon 90° twist in a strained bicyclic alkene. Consider a hypothetical (illustrative) bridged bicyclic alkene in which ring strain forces the C=C dihedral twist to $\theta \approx 30°$ away from planarity (this order of twist is seen in real strained systems such as certain adamantylideneadamantanes and bridgehead-adjacent alkenes). Predict the fractional retention of π-bond strength.

Step 1. Apply the overlap relation derived above: $S(\theta) = S_0 \cos\theta$.

Step 2. Compute $\cos(30°)$:

Step 3. Since bond stabilization tracks $|\beta| \propto S(\theta)$ in the simple linear approximation, the retained fraction of π-bond strength is:

Step 4. Using the ethylene-calibrated π-bond strength of ≈65 kcal/mol as $E_\pi(0)$:

So a 30° twist costs about $65 - 56.3 = 8.7$ kcal/mol of lost π stabilization — consistent in order of magnitude with the extra strain energy (several kcal/mol to over ten, depending on system) measured calorimetrically or computed for twisted-alkene natural products and reactive intermediates, and consistent with such alkenes showing anomalously high reactivity toward electrophiles (a weaker, more exposed, higher-energy HOMO π bond is easier to attack).

Step 5 — limiting case check. As $\theta \to 0°$ (planar limit), $\cos\theta \to 1$ and $E_\pi(\theta) \to E_\pi(0) = 65$ kcal/mol exactly — full π strength recovered, as required, confirming the formula degenerates correctly to the untwisted reference case.

Example 3 — Exam/competition level: mechanistic reasoning for regiochemistry using the π HOMO, applied to bromination of an unsymmetrical alkene. Consider 2-methylpropene (isobutylene), $(\mathrm{CH_3})_2\mathrm{C}=\mathrm{CH_2}$, reacting with $\mathrm{Br_2}$. Explain, using the π-bond picture developed above, why the bromonium ion opens to place the second bromide on the more substituted carbon (i.e., why Markovnikov-type selectivity is observed even though this is a mechanism without a pre-formed carbocation intermediate at the addition step).

Step 1 — Identify the reactive orbital. The alkene's HOMO is the π MO built from the two unhybridized $2p_z$ orbitals on the C=C carbons, as derived above — electron density bulging above and below the C=C axis, exposed and available as a nucleophile toward $\mathrm{Br_2}$'s σ* orbital (the low-lying antibonding orbital of the Br–Br bond, which acts as the electrophilic partner).

Step 2 — Curved-arrow mechanism, bromonium formation. The π electron pair (arrow originates in the middle of the C=C bond, at the π cloud) attacks one bromine atom of $\mathrm{Br_2}$; simultaneously, the Br–Br σ bond breaks heterolytically (arrow from the Br–Br bond to the departing bromide) to give a cyclic three-membered bromonium ion, with both carbons now bonded to the bridging Br⁺ via what remains of the former π system, and Br⁻ released into solution.

Step 3 — Analyze the bromonium ion's electronic asymmetry. Although drawn as a symmetric three-membered ring, the C–Br⁺ bonds are not equal in bond order: the more substituted carbon (bearing two methyl groups) can better stabilize partial positive charge through hyperconjugation from the two adjacent C–H σ bonds of the methyls, so the C–Br⁺ bond to that carbon is weaker/longer, and that carbon carries more of the residual positive character (the bromonium ion is unsymmetrically bridged, closer to an open carbocation on the more substituted side).

Step 4 — Curved-arrow mechanism, nucleophilic opening. Bromide ion (regenerated in Step 2) attacks the carbon bearing greater positive character — the more substituted carbon — in an $S_N2$-like backside attack that simultaneously breaks the C–Br⁺ bond on that same carbon (arrow from that C–Br bond to the bromine, which becomes neutral, ring-opened bromonium bromine).

Step 5 — Assign the product and check regiochemistry. The result is 1,2-dibromo-2-methylpropane, $\mathrm{BrCH_2C(CH_3)_2Br}$, with the new nucleophile (Br⁻) delivered to the more substituted carbon — the observed, experimentally confirmed regiochemical outcome for bromination of trisubstituted alkenes, fully rationalized without invoking a free (unbridged) carbocation, purely from (a) the π bond being the nucleophilic HOMO that initiates attack and (b) the electronic asymmetry the π system's original substitution pattern imposes on the resulting bridged cation.

Step 6 — Limiting case check. For a perfectly symmetric alkene (e.g. 2-butene, both alkene carbons equally substituted), the bromonium ion is symmetric by symmetry, both carbons carry equal partial positive character, and Br⁻ attacks with statistically equal probability at either carbon — degenerating the "selectivity" prediction correctly to "no regiochemical preference," which matches the observation that internal, symmetrically-substituted alkenes give a single achiral-relationship product regardless of which carbon bromide formally attacks.