The Molecular Orbitals of Butadiene

The four-orbital pattern derived here for butadiene is the literal starting point for understanding the Diels–Alder reaction: the reactive HOMO of the diene ($\psi_2$, with its large terminal-carbon coefficients) overlaps with the LUMO of a dienophile in exactly the geometry that the nodal pattern predicts, a frontier-orbital argument formalized by Woodward and Hoffmann. The same node-counting logic, extended to longer chains, underlies the electronic structure of carotenoid pigments (β-carotene, lycopene) whose extended conjugation pushes the HOMO–LUMO gap down into the visible region, producing the orange and red colors of carrots and tomatoes directly from orbital physics rather than any exotic chromophore chemistry. Retinal, the light-absorbing chromophore of rhodopsin in the vertebrate eye, uses precisely this diene/polyene orbital ladder — photon absorption promotes an electron from a bonding $\psi$ to an antibonding $\psi^*$, weakening a specific double bond just enough to trigger the cis–trans isomerization that initiates the visual signaling cascade. In materials science, conducting polymers such as polyacetylene are, in essence, "very long butadienes," and their band structure (in the solid-state-physics sense) is the $N\to\infty$ limit of exactly the cosine dispersion relation derived above for $N=4$. Even modern computational chemistry, which now routinely solves such systems with full density functional theory, still uses Hückel-level reasoning as the fast, intuitive first pass for predicting regiochemistry and reactivity before any expensive calculation is run.

Ask the tutor: How do the four butadiene π MOs change if we go to a cyclic system like cyclobutadiene instead of a linear chain?

Start with the $\sigma$-framework already in place: four $sp^2$-hybridized carbons, each bonded to its neighbors and to hydrogens, all lying in one plane (or very close to it, since conjugation strongly favors planarity). Each carbon carries one electron in a $2p_z$ orbital, oriented perpendicular to that plane — a dumbbell with a lobe above the plane and a lobe below it, related by a sign change. These four $2p_z$ orbitals, call them $\phi_1,\phi_2,\phi_3,\phi_4$ centered on C1–C4, are the basis set for the $\pi$ system. Everything downstream is linear algebra on this basis. The physical reason the atomic orbitals mix at all is orbital overlap: adjacent $p_z$ orbitals (C1–C2, C2–C3, C3–C4) overlap side-by-side, and quantum mechanics says that whenever two orbitals overlap, the true stationary states of the system are not the original atomic orbitals but linear combinations of them — one combination where the overlapping lobes have the same phase (constructive interference, bonding, lower energy) and one where they have opposite phase (destructive interference, antibonding, higher energy). With four atomic orbitals in a chain, that logic has to play out simultaneously at three separate bonds (1–2, 2–3, 3–4), and the interplay produces exactly four molecular orbitals, each a different global compromise between "constructive at every bond" and "destructive at every bond." A cleaner way to enumerate the possibilities, and the one that leads straight to the correct answer without guessing, is the particle-in-a-box / standing-wave analogy. The four $p_z$ orbitals are like four beads on a string, and the four $\pi$ MOs are the four allowed standing-wave patterns, indexed by how many times the wave changes sign between beads. Zero sign changes is the smoothest, lowest-energy wave; three sign changes (the maximum possible with four beads, since a sign change can only occur between adjacent atoms, and there are three such gaps) is the most oscillatory, highest-energy wave. This maps exactly onto the four MOs $\psi_1$ through $\psi_4$.

Notice the pattern that falls out once all four are drawn side by side: $\psi_1$ is all one color (bonding at every one of the three C–C links), $\psi_2$ has one sign flip in the middle (bonding at C1–C2 and C3–C4, antibonding at C2–C3), $\psi_3$ has two sign flips (antibonding at C1–C2 and C3–C4, bonding at C2–C3), and $\psi_4$ alternates every atom (antibonding everywhere). This is exactly the standing-wave sequence you'd draw for a vibrating string with increasing numbers of nodes, and that structural parallel is not a coincidence — both are eigenvalue problems for a linear chain with nearest-neighbor coupling.

The Hückel method reduces the full electronic Schrödinger equation to a tractable algebraic eigenvalue problem by keeping only $\pi$ electrons and only nearest-neighbor interactions. Two parameters carry all the physics:

where: $\alpha$ sets the zero-order reference energy (same for every carbon $2p_z$ by symmetry), and $\beta$ is the stabilization gained when two adjacent $p$ orbitals overlap constructively; $\beta<0$ means bonding combinations are lower in energy than $\alpha$ and antibonding combinations are higher. For a linear chain of $N$ conjugated $p$ orbitals with equal spacing, the Hückel eigenvalues have the exact closed form

where: $k$ is the MO index (ordered from lowest to highest energy since $\beta<0$ makes $\cos$ near $+1$ the most stabilizing), and $N$ is the number of atoms in the chain (here $N=4$). The coefficients of atomic orbital $r$ ($r=1,\dots,N$) in molecular orbital $k$ follow the companion formula

where: the $\sqrt{2/(N+1)}$ prefactor is the normalization constant guaranteeing $\sum_r c_{r,k}^2 = 1$, and the sine factor encodes the standing-wave shape — its sign changes are exactly the nodes seen in the figure above. Reading these two formulas is itself the payoff: $E_k$ tells you the energy ladder is not evenly spaced (it's a cosine, densest near the middle), and $c_{r,k}$ tells you where the electron density piles up in each MO and exactly which atoms carry the largest LCAO coefficients — information used immediately once you get to frontier-orbital reactivity.

Example 1 (foundational): Verifying node count and symmetry by direct inspection of coefficients. Using the closed-form coefficient formula with $N=4$,

compute all sixteen coefficients for $k=1,2,3,4$ and $r=1,2,3,4$, and count sign changes along $r=1\to4$ for each $k$. For $k=2$ ($\psi_2$): $c_{1,2}=\sqrt{0.4}\sin(72^\circ)=0.632(0.951)=0.601$; $c_{2,2}=\sqrt{0.4}\sin(144^\circ)=0.632(0.588)=0.372$; $c_{3,2}=\sqrt{0.4}\sin(216^\circ)=0.632(-0.588)=-0.372$; $c_{4,2}=\sqrt{0.4}\sin(288^\circ)=0.632(-0.951)=-0.601$. So $\psi_2 = 0.601\phi_1+0.372\phi_2-0.372\phi_3-0.601\phi_4$: signs read $(+,+,-,-)$ — one sign change, between C2 and C3. That single node sits at the midpoint of the C2–C3 bond, which is exactly why $\psi_2$ is antibonding across that central bond even though it's bonding across C1–C2 and C3–C4 individually. This directly explains the experimentally observed partial double-bond character concentrated at the terminal bonds when $\psi_2$ (the HOMO) dominates reactivity — asymmetric, non-round numbers, but the qualitative sign pattern is exactly what the earlier hand-drawn figure showed, confirming the formula is self-consistent.

Example 2 (applied): Predicting the preferred conformation of butadiene from $\psi_1$. Butadiene exists predominantly as the s-trans conformer (planar, ~side groups pointing away), with a rotational barrier around the C2–C3 bond of roughly $25$–$30\ \text{kJ mol}^{-1}$ — far higher than a pure single bond's negligible rotation barrier, and far lower than a full C=C's ~250 kJ/mol barrier. Why? $\psi_1$, the doubly-occupied lowest MO, has coefficients (from Step 6 of the first derivation) $0.371, 0.601, 0.601, 0.371$ on C1–C4 — all the same sign. That means $\psi_1$ is bonding not only at C1–C2 and C3–C4 (the "formal" double bonds) but also across the central C2–C3 bond, since adjacent same-sign lobes overlap constructively there too. This is the direct orbital-level origin of partial $\pi$ character in the "single" C2–C3 bond: rotating around that bond breaks the C2–C3 $p$-orbital overlap, which costs real energy (destabilizing $\psi_1$ by losing some of that bonding interaction) even though C2–C3 is formally drawn as a single bond in the Lewis structure. Quantitatively, the bond order contribution from $\psi_1$ alone across C2–C3 is proportional to $2\times c_{2,1}c_{3,1} = 2(0.601)(0.601)=0.722$, a substantial fraction of a full bond ($=1$), which is why rotation is hindered but nowhere near as hindered as true double-bond rotation (bond order there would be closer to $1.9$–$2.0$ when both $\psi_1$ and the corresponding bonding character are summed for a true C=C).

Example 3 (exam/competition level): Predicting the UV–vis $\pi\to\pi^*$ transition energy trend using the HOMO–LUMO gap. Using the Hückel energies derived above, compute the HOMO–LUMO gap of butadiene ($\psi_2\to\psi_3$) and compare qualitatively to ethylene's gap, to rationalize why butadiene absorbs at longer wavelength (lower energy, $\lambda_{\max}\approx 217\ \text{nm}$) than ethylene ($\lambda_{\max}\approx 165\ \text{nm}$). For butadiene:

Since $\beta$ is negative, $-1.236\beta$ is a positive number: the gap magnitude is $1.236|\beta|$. For ethylene ($N=2$, $E=\alpha\pm\beta$):

Comparing the two coefficients: butadiene's gap ($1.236|\beta|$) is smaller than ethylene's gap ($2|\beta|$), by a ratio of

A smaller HOMO–LUMO gap means a lower-energy $\pi\to\pi^*$ transition, which by $E=hc/\lambda$ means a longer absorption wavelength — precisely the experimental trend ($217\,\text{nm}>165\,\text{nm}$, i.e., butadiene absorbs at lower energy/longer wavelength than ethylene). This single ratio, $0.618=1/\varphi$, is again the golden ratio falling directly out of the tridiagonal chain mathematics, and it is the qualitative reasoning that extends (with more nodes, smaller gaps) all the way to why long polyene chains and carotenoids absorb visible light while short ones absorb only in the UV — the conceptual seed of Frontier Molecular Orbitals: HOMO and LUMO.

Erich Hückel developed the method bearing his name in 1930–31 while working on the electronic structure of unsaturated and aromatic hydrocarbons, seeking a way to explain benzene's peculiar stability without brute-force numerical solution of the full many-electron Schrödinger equation. His approach — separate the $\sigma$ skeleton from the $\pi$ system, treat only nearest-neighbor $p$-orbital interactions, and reduce everything to a single small parameter $\beta$ — was radical in its simplicity for the time, and it succeeded precisely because conjugated hydrocarbons are a system where that approximation captures the essential physics (nodal structure, symmetry, relative ordering) even while sacrificing quantitative accuracy in absolute energies. Hückel had trained as a physicist under Peter Debye and brought a physicist's comfort with simplified models to a chemistry problem that other chemists were still trying to solve with purely empirical resonance-structure bookkeeping.