Aromatic chemistry — EAS, Huckel

Intuition [Beginner]

Benzene is a ring of six carbons, each bonded to one hydrogen and two neighbouring carbons. Every carbon has one p-orbital perpendicular to the ring, and these six p-orbitals overlap to form a continuous ring of electron density above and below the plane. This is the aromatic pi system.

Benzene does not behave like an alkene. An alkene reacts readily with electrophiles via addition, breaking the pi bond. Benzene resists addition because breaking the aromatic pi system costs about 150 kJ/mol of extra stability. Instead, benzene undergoes electrophilic aromatic substitution (EAS): an electrophile attacks the ring, briefly disrupting aromaticity, but then a proton is expelled to restore the aromatic system. The net result is substitution (one H replaced by E), not addition.

The Huckel rule predicts which cyclic, planar, fully conjugated systems are aromatic: a molecule with $4n+2$ pi electrons is aromatic ($n=0,1,2,\dots$). Benzene has 6 pi electrons ($4\times 1+2$, $n=1$): aromatic. Cyclobutadiene has 4 pi electrons: destabilised (anti-aromatic). Cyclooctatetraene has 8 pi electrons and avoids anti-aromaticity by puckering into a non-planar tub shape.

When a substituent is already on the ring, it directs the next electrophile to specific positions. Electron-donating groups (amino, hydroxyl, methoxy, alkyl) activate the ring and direct ortho and para. Electron-withdrawing groups (nitro, carbonyl, cyano) deactivate the ring and direct meta.

Visual [Beginner]

Picture the benzene ring as a hexagon with carbons numbered 1 through 6. The six p-orbitals overlap in a continuous ring above and below the hexagon.

EAS mechanism. An electrophile E${}^{+}$ approaches the pi cloud. One of the six C-C pi bonds donates two electrons to form a new C-E bond. The carbon that received the electrophile changes from sp${}^2$ to sp${}^3$ (tetrahedral). The ring now has only four pi electrons spread over five carbons -- aromaticity is lost. This intermediate is the sigma complex (also called the Wheland intermediate or arenium ion).

To restore aromaticity, the hydrogen on the carbon that formed the new C-E bond leaves. The C-H bond breaks, and those two electrons re-enter the pi system. Aromaticity returns. Net result: H was replaced by E.

Ortho, meta, para positions. With a substituent X on C1: ortho is C2 and C6 (adjacent to X), meta is C3 and C5, para is C4 (opposite X).

Worked example [Beginner]

Nitration of toluene -- predict the major products.

Toluene is methylbenzene (${\text{CH}}_3{\text{-C}}_6{\text{H}}_5$). The methyl group sits on C1. Nitration uses nitric acid (${\text{HNO}}_3$) with sulfuric acid (${\text{H}}_2{\text{SO}}_4$) to generate the nitronium ion ${\text{NO}}_2^{+}$, the actual electrophile.

Step 1. The methyl group is electron-donating. Alkyl groups push electron density into the ring through hyperconjugation (the C-H bonds on the methyl group share electron density with the adjacent ring). This activates the ring toward electrophilic attack.

Step 2. The electrophile ${\text{NO}}_2^{+}$ attacks the ring. The methyl group directs to ortho (C2, C6) and para (C4). The reason: for ortho or para attack, one resonance structure of the sigma complex places positive charge on C1 -- the carbon bearing the methyl group. The methyl group stabilises this positive charge through hyperconjugation. For meta attack, the positive charge never lands on C1, so this extra stabilisation is unavailable.

Step 3. The major products are ortho-nitrotoluene (approximately 60%) and para-nitrotoluene (approximately 40%), with only trace meta-nitrotoluene (less than 5%). The ortho ratio is roughly 2:1 because there are two ortho positions but only one para position, though steric effects at ortho slightly reduce the expected ratio.

What this tells us: electron-donating groups direct ortho/para because the sigma complex for those positions has a resonance structure that the substituent can stabilise. The same logic explains why electron-withdrawing groups direct meta -- they destabilise the ortho/para sigma complex.

Check your understanding [Beginner]

Formal definition [Intermediate+]

Aromaticity is a property of cyclic, planar, fully conjugated pi systems that exhibit unusual thermodynamic stability, characteristic spectroscopic features (ring-current NMR shifts, UV absorption patterns), and distinct chemical behaviour (preference for substitution over addition). The quantitative criterion is Huckel's rule.

Huckel's rule. A planar, monocyclic, fully conjugated hydrocarbon is aromatic if and only if it contains $4n+2$ pi electrons, where $n=0,1,2,\dots$.

The rule derives from the Huckel molecular orbital treatment. For a cyclic polyene with $N$ atoms, the pi MOs have energies:

$${\epsilon }_k=\alpha +2\beta \cos \left(\frac{2\pi k}{N}\right),k=0,\pm 1,\pm 2,\dots$$

where $\alpha$ is the Coulomb integral (energy of an isolated p-orbital) and $\beta$ is the resonance integral (negative, representing the stabilisation from p-orbital overlap). For $N=6$ (benzene), the six MOs are at $\alpha +2\beta$, $\alpha +\beta$ (doubly degenerate), $\alpha -\beta$ (doubly degenerate), and $\alpha -2\beta$. The six pi electrons fill the three bonding MOs completely: $2+2+2=6=4(1)+2$.

For $N=4$ (cyclobutadiene), the four MOs are $\alpha +2\beta$, $\alpha$ (doubly degenerate), and $\alpha -2\beta$. The four pi electrons fill the bonding MO and place one electron each in the degenerate non-bonding MOs by Hund's rule. This open-shell configuration is destabilised relative to two isolated double bonds: anti-aromatic.

The $4n+2$ count ensures that all bonding MOs are completely filled and all antibonding MOs are empty -- a closed-shell configuration.

Electrophilic aromatic substitution (EAS). The general mechanism for a substituted benzene ring:

$$\text{Ar-H}+{\text{E}}^{+}\stackrel{k_1}{\to }{\text{Ar(H)(E)}}^{+}\stackrel{k_2}{\to }\text{Ar-E}+{\text{H}}^{+}$$

Step 1 ($k_1$): the electrophile attacks the ring, forming the sigma complex (Wheland intermediate). Aromaticity is disrupted. The positive charge is delocalised over the ring carbons ortho and para to the site of attack.

Step 2 ($k_2$): a base removes a proton from the tetrahedral carbon, restoring aromaticity. The electrons from the C-H bond re-enter the pi system.

The rate-determining step is usually Step 1 (sigma complex formation), though for strongly deactivated rings Step 2 can become rate-limiting.

Directing effects. Substituents on the ring influence both the rate of EAS (activation or deactivation) and the regiochemistry (ortho/para or meta direction).

Ortho/para directors (activating, mostly): ${\text{-NH}}_2$, $\text{-OH}$, ${\text{-OCH}}_3$, ${\text{-CH}}_3$, ${\text{-C}}_6{\text{H}}_5$, halogens. The sigma complex for ortho/para attack has a resonance structure that places positive charge on the carbon directly bonded to the substituent, allowing the substituent to stabilise that charge directly. This lowers the activation energy for ortho/para attack relative to meta.

Meta directors (deactivating): ${\text{-NO}}_2$, $\text{-CN}$, $\text{-COOH}$, ${\text{-SO}}_3\text{H}$, $\text{-CHO}$, ${\text{-CF}}_3$. The sigma complex for ortho/para attack places positive charge on the carbon directly bonded to the electron-withdrawing group -- destabilising. Meta attack never places positive charge adjacent to the substituent, avoiding this destabilisation.

Halogens are a special case: deactivating (slower than benzene) but ortho/para directing. The inductive effect (electron-withdrawing through sigma bonds) deactivates the ring, but the resonance effect (lone-pair donation into the ring) directs ortho/para. The inductive effect dominates the rate; the resonance effect dominates the regiochemistry.

Counterexamples to common slips

• "Aromatic means it has a benzene ring." Aromaticity is an electronic property, not a structural one. Cyclopentadienyl anion (6 pi electrons), pyridine (6 pi electrons), and the tropylium cation (6 pi electrons) are all aromatic without being benzene.

• "All activating groups direct ortho/para." True for standard EAS, but steric effects can override: a very bulky activating group (e.g., tert-butyl) favours para over ortho because the ortho positions are sterically hindered.

• "Huckel's rule applies to all rings." The $4n+2$ rule is strictly for monocyclic, planar, fully conjugated systems. Polycyclic aromatics (naphthalene, anthracene) are aromatic but not described by a simple perimeter-electron count -- they require separate MO analysis or Clar's sextet theory.

Key theorem with proof [Intermediate+]

Proposition (Ortho/para direction by electron-donating groups). Let X be an electron-donating substituent on benzene at C1. The sigma complex for electrophilic attack at an ortho or para position has a resonance structure with positive charge on C1 (the carbon bearing X), while the sigma complex for meta attack does not. The additional stabilisation from X lowers the activation energy for ortho/para attack relative to meta attack.

Proof. Write the sigma complex resonance structures for attack at each position. The sigma complex has positive charge on the three sp${}^2$ carbons that do not bear the incoming electrophile and are not double-bonded to each other.

Ortho attack at C2. When the electrophile E${}^{+}$ attacks C2, the C1=C2 pi bond donates its electrons to form the C2-E bond. The positive charge initially lands on C1. Through further delocalisation, the charge migrates to C4 and C6. The three resonance structures place positive charge at C1, C4, and C6. Since X is on C1, X directly stabilises the positive charge on C1 through resonance donation of lone pairs (for ${\text{-OCH}}_3$, ${\text{-NH}}_2$) or hyperconjugation (for ${\text{-CH}}_3$).

Para attack at C4. When E${}^{+}$ attacks C4, the charge delocalises to C1, C3, and C5. One resonance structure places positive charge on C1 -- directly on the carbon bearing X. X provides the same stabilisation.

Meta attack at C3. When E${}^{+}$ attacks C3, the positive charge delocalises to C2, C4, and C6. The carbon bearing the substituent X (C1) does not carry positive charge in any resonance structure for meta attack. X cannot stabilise the sigma complex through direct resonance or hyperconjugation.

Since the sigma complex is closer in energy to the transition state than to the reactants (Hammond postulate -- sigma complex formation is endothermic), the lower sigma-complex energy for ortho/para attack translates to a lower activation energy. The rate constant for ortho/para attack exceeds that for meta attack. $■$

Corollary. For electron-withdrawing substituents (meta directors), the resonance structure placing positive charge on C1 (ortho/para attack) is destabilised by the electron-withdrawing group. Meta attack avoids this destabilisation and is therefore favoured.

Bridge. The ortho/para direction principle builds toward the quantitative partial-rate-factor analysis in the Master tier, where it appears again as the Ingold substituent-effect classification. The foundational reason for regioselectivity is that sigma-complex resonance structures determine relative activation energies -- this is exactly the mechanistic pattern that identifies electrophilic aromatic substitution as the aromatic counterpart of electrophilic addition to alkenes 15.05.01, where a carbocation intermediate is also formed but trapped by a nucleophile rather than expelled as a proton. The bridge is between the qualitative resonance argument here and the Hammett-Brown ${\sigma }^{+}$ quantitative treatment below, which generalises the selectivity prediction to any substituent with a single empirical parameter.

Exercises [Intermediate+]

Advanced results [Master]

Theorem 1 (Huckel 4n+2 rule -- spectral characterisation). A planar, monocyclic, fully conjugated hydrocarbon with $N$ atoms is aromatic if and only if the number of pi electrons equals $4n+2$ for some non-negative integer $n$. The proof (see Full proof set) follows from the degenerate eigenvalue structure of the cyclic Huckel matrix: $4n+2$ electrons fill all bonding levels and leave all antibonding levels empty, producing a closed-shell ground state. Systems with $4n$ electrons leave a degenerate pair of non-bonding orbitals singly occupied -- the open-shell destabilisation that defines anti-aromaticity ^[Huckel1931].

Theorem 2 (Frost-Musulin circle). The MO energies of a monocyclic polyene with $N$ atoms are given by the vertical projections of the vertices of a regular $N$-gon inscribed in a circle of radius $2|\beta |$ centred at $\alpha$, with one vertex at the bottom. This geometric construction, due to Frost and Musulin in 1953, is a direct consequence of the circulant-matrix structure of the Huckel Hamiltonian ^{[FrostMusulin1953]}.

Theorem 3 (Huckel resonance energy of benzene). The delocalisation energy of benzene computed at the Huckel level is $2|\beta |$, corresponding to approximately 150 kJ/mol using the empirically calibrated value $\beta \approx -75$ kJ/mol. This is in reasonable agreement with the thermochemical resonance energy of 150-160 kJ/mol from heats of hydrogenation.

Theorem 4 (Ingold substituent classification). Substituents on aromatic rings classify into four categories by their effect on EAS: activating/ortho-para directing (amino, hydroxyl, alkoxy, alkyl), weakly activating/ortho-para directing (phenyl), deactivating/meta directing (nitro, cyano, carbonyl, sulfonyl), and deactivating/ortho-para directing (halogens -- the unique case where inductive withdrawal dominates reactivity but resonance donation dominates regiochemistry). This classification, due to Ingold's school in the 1930s-40s, organises the qualitative directing-effect predictions.

Theorem 5 (Hammett-Brown ${\sigma }^{+}$ correlation). The logarithm of the partial rate factor for EAS at the para position correlates linearly with the substituent constant ${\sigma }^{+}$: $\log (f_{\text{para}})={\rho }^{+}{\sigma }^{+}$. The ${\sigma }^{+}$ scale, developed by Brown and Okamoto in 1958, differs from the Hammett $\sigma$ scale by incorporating resonance electron donation in the transition state. The ${\sigma }^{+}$ values for strong resonance donors (e.g., ${\text{-OCH}}_3$: ${\sigma }^{+}=-0.78$) are much more negative than their $\sigma$ values ($-0.27$), reflecting the additional resonance stabilisation in the EAS transition state absent in benzoic-acid dissociation.

Theorem 6 (Reactivity-Selectivity Principle). More reactive electrophiles exhibit lower positional selectivity in EAS. The partial rate factor $f_{\text{para}}$ for toluene nitration is approximately 60; for toluene chlorination it is approximately 600. The less reactive chlorine is more selective, amplifying the electronic differences between ring positions. This inverse relationship reflects the Hammond-postulate geometry of the transition state.

Theorem 7 (Clar's sextet rule). For polycyclic aromatic hydrocarbons (PAHs), the most important resonance structure is the one with the maximum number of disjoint aromatic sextets (benzene-like $\pi$-electron sets of six). This rule, due to Clar in 1958, predicts that naphthalene has one Clar sextet, while phenanthrene has two disjoint sextets and is more stable than its isomer anthracene.

Synthesis. The Huckel framework is the foundational reason that aromatic stability admits a unified theoretical treatment across organic, inorganic, and biological systems. The central insight is that the $4n+2$ rule is a spectral property of the adjacency matrix of a conjugated ring; putting these together with perturbation theory yields the Hammett-Brown quantitative structure-reactivity relationships. This is exactly the structure that identifies MO energy levels in aromatic heterocycles with the Frost-circle pattern for carbocyclic systems: the pattern generalises to polycyclic aromatics via Clar's sextet theory and to spherical systems via the Hirsch $2(N+1{)}^2$ rule for fullerenes. The bridge is between Huckel's 1931 MO treatment and modern DFT-calculated aromaticity indices (NICS, HOMA), and the pattern recurs in the EAS sigma complex, which appears again in 15.05.01 as the mechanistic analogue of the carbocation intermediate in electrophilic addition. The same MO filling argument that stabilises benzene builds toward 14.05.02 pending where homonuclear diatomic MO theory applies the same Aufbau logic to a different geometry.

Huckel MO theory and Frost circles [Master]

Huckel molecular orbital theory provides a quantitative model for the pi-electron system of conjugated hydrocarbons by reducing the full electronic-structure problem to a one-electron Hamiltonian on the p-orbital basis. The Huckel Hamiltonian matrix for a molecule with $N$ conjugated atoms is an $N\times N$ matrix with diagonal elements $\alpha$ (the Coulomb integral, representing the energy of an isolated p-orbital) and off-diagonal elements $\beta$ (the resonance integral, representing the energy of overlap between bonded p-orbitals). The resonance integral $\beta$ is negative, indicating stabilisation.

For a monocyclic polyene (annulene) with $N$ atoms, the Huckel matrix is a circulant matrix -- each row is a cyclic shift of the previous one. The eigenvalues of circulant matrices have a known closed form:

$${\epsilon }_k=\alpha +2\beta \cos \left(\frac{2\pi k}{N}\right),k=0,\pm 1,\dots$$

For even $N$, the levels are: a single non-degenerate level at $\alpha +2\beta$ (the most bonding), pairs of doubly degenerate levels, and a single non-degenerate level at $\alpha -2\beta$ (the most antibonding). All MO energies lie in the range $[\alpha +2\beta ,\alpha -2\beta ]$ -- no orbital is more than $2|\beta |$ from the atomic reference $\alpha$.

The Frost circle (Frost and Musulin 1953) gives a geometric construction for these energies ^{[FrostMusulin1953]}. Inscribe a regular $N$-gon in a circle of radius $2|\beta |$ centred at $\alpha$, with one vertex at the bottom. The vertical positions of the vertices give the MO energy levels. For benzene ($N=6$): the hexagon has one vertex at the bottom ($\alpha +2\beta$, bonding), two vertices at the sides at mid-height ($\alpha +\beta$, bonding, doubly degenerate), two vertices at the sides above centre ($\alpha -\beta$, antibonding, doubly degenerate), and one vertex at the top ($\alpha -2\beta$, antibonding).

The six pi electrons of benzene fill the three bonding MOs: two in the lowest ($\alpha +2\beta$) and four in the degenerate pair ($\alpha +\beta$). The total pi-electron energy is:

$$E_{\pi }=2(\alpha +2\beta )+4(\alpha +\beta )=6\alpha +8\beta$$

The reference energy for three isolated double bonds (the Kekule structure) is $3\times 2\alpha +3\times 2\beta =6\alpha +6\beta$. The difference is the delocalisation energy or resonance energy: $E_{\text{res}}=2\beta \approx -150kJ/mol$ (reported as 150 kJ/mol of stabilisation). This Huckel estimate agrees with the experimental resonance energy of 150-160 kJ/mol derived from heats of hydrogenation -- a remarkably good match for a theory that neglects electron-electron repulsion entirely.

The $\alpha$ and $\beta$ parameters are calibrated empirically. The Coulomb integral $\alpha$ serves as the energy zero. The resonance integral $\beta$ is fitted to thermochemical data. A value of $\beta \approx -75$ kJ/mol reproduces the benzene resonance energy. The parameter is an effective quantity that absorbs much of the electron-correlation error.

Charged aromatic systems. The cyclopentadienyl anion (${\text{C}}_5{\text{H}}_5^{-}$) has $N=5$ and 6 pi electrons. The Frost circle for a pentagon gives levels at $\alpha +2\beta$, $\alpha +0.618\beta$ (degenerate), and $\alpha -1.618\beta$ (degenerate). Six electrons fill the bonding MO and one degenerate pair: $2+2+2=6=4(1)+2$, closed shell, aromatic. The stability of this anion is the basis for metallocene chemistry (ferrocene, $\text{Fe}({\text{C}}_5{\text{H}}_5{)}_2$).

The tropylium cation (${\text{C}}_7{\text{H}}_7^{+}$) has $N=7$ and 6 pi electrons. The Frost heptagon gives levels at $\alpha +2\beta$, $\alpha +1.247\beta$ (degenerate), and $\alpha -0.445\beta$ (degenerate). Six electrons fill the three lowest levels: $6=4(1)+2$, aromatic. The tropylium cation is a stable, isolable species confirmed by mass spectrometry of tropylium salts.

Heterocyclic systems. Pyridine has $N=6$ atoms (five C and one N) and 6 pi electrons. Nitrogen contributes one electron to the pi system (its lone pair is in an sp${}^2$ orbital in the ring plane, not a p-orbital) and has a more negative Coulomb integral (${\alpha }_{\text{N}}=\alpha +h\beta$ with $h\approx 0.5$). This perturbation lifts the degeneracy slightly but preserves the closed-shell configuration.

Pyrrole has $N=5$ atoms (four C and one N) and 6 pi electrons. The nitrogen contributes two electrons to the pi system (its lone pair is in a p-orbital perpendicular to the ring). The Frost pentagon applies, and 6 pi electrons give a closed-shell aromatic system.

Comparison with DFT. Density functional theory calculations on benzene give orbital energies qualitatively similar to Huckel theory: a lowest bonding orbital, a degenerate pair, another degenerate pair, and a highest antibonding orbital. The HOMO-LUMO gap from DFT is approximately 5 eV (480 kJ/mol), while the Huckel gap is $2|\beta |\approx 150$ kJ/mol -- a factor of 3 underestimate. The qualitative orbital ordering and symmetry labels are correct in Huckel theory, but quantitative energies require higher-level methods. The value of Huckel theory is its analytical tractability: the entire electronic structure follows from a single parameter $\beta$ and the topology of the conjugated network.

Quantitative directing effects: partial rate factors and the Hammett-Brown treatment [Master]

The selectivity of EAS is quantified by partial rate factors $f_X^p$, defined as the rate of substitution at one specific position of a monosubstituted benzene relative to one position of benzene itself, normalised for the number of equivalent positions:

$$f_{\text{ortho}}=\frac{k_{\text{ortho}}/2}{k_{\text{benz}}/6},f_{\text{meta}}=\frac{k_{\text{meta}}/2}{k_{\text{benz}}/6},f_{\text{para}}=\frac{k_{\text{para}}}{k_{\text{benz}}/6}.$$

A partial rate factor of 1 means that position reacts at the same per-position rate as benzene. Values greater than 1 indicate activation; values less than 1 indicate deactivation.

The following table gives representative partial rate factors for nitration at 25 ${}^{\circ }$C:

Substituent	$f_{\text{ortho}}$	$f_{\text{meta}}$	$f_{\text{para}}$	Classification
${\text{-NHCOCH}}_3$	~800	~3	~800	activating, o/p
${\text{-OCH}}_3$	~500	~3	~2500	activating, o/p
${\text{-CH}}_3$	~40	~3	~60	activating, o/p
$\text{-F}$	~0.03	~0.001	~0.1	deactivating, o/p
$\text{-Cl}$	~0.03	~0.001	~0.14	deactivating, o/p
$\text{-Br}$	~0.03	~0.001	~0.11	deactivating, o/p
${\text{-COOC}}_2{\text{H}}_5$	~$10^{-4}$	~$10^{-3}$	~$10^{-4}$	deactivating, meta
${\text{-NO}}_2$	~$10^{-8}$	~$10^{-3}$	~$10^{-8}$	deactivating, meta

Several patterns emerge. For activating ortho/para directors, $f_{\text{ortho}}$ and $f_{\text{para}}$ are orders of magnitude larger than $f_{\text{meta}}$. For deactivating meta directors, $f_{\text{meta}}$ is orders of magnitude larger than $f_{\text{ortho}}$ and $f_{\text{para}}$. Halogens show intermediate behaviour: all three $f$ values are less than 1 (deactivation), but $f_{\text{ortho}}$ and $f_{\text{para}}$ are 10-100 times larger than $f_{\text{meta}}$ (ortho/para direction).

The Ingold classification. Ingold's school (1930s-1940s) systematised the qualitative directing effects into the framework still taught today. The classification rests on whether the substituent donates or withdraws electron density by resonance and by induction. Resonance donation activates and directs ortho/para. Inductive withdrawal deactivates. The net effect depends on which mechanism dominates.

For halogens: the resonance donation (lone pairs on Cl, Br, I in p-orbitals that overlap with the ring) and inductive withdrawal (electronegativity difference) are comparable in magnitude. The inductive effect lowers electron density everywhere on the ring, slowing the reaction. But the resonance effect selectively stabilises the ortho/para sigma complexes, directing regiochemistry.

The Hammett-Brown ${\sigma }^{+}$ scale. The standard Hammett equation correlates reaction rates with substituent constants $\sigma$: $\log (k_X/k_H)=\rho \sigma$. For EAS, the correlation using $\sigma$ fails for strong resonance donors because the transition state has substantial positive charge delocalised onto the substituent -- a resonance interaction not captured by the standard $\sigma$ constants (which are based on benzoic acid dissociation, a reaction without such charge delocalisation).

Brown and Okamoto (1958) introduced the ${\sigma }^{+}$ scale, determined from the rates of solvolysis of substituted cumyl chlorides -- a reaction with a transition state resembling the EAS sigma complex. The ${\sigma }^{+}$ values for resonance donors are much more negative than their $\sigma$ values: ${\sigma }^{+}({\text{-OCH}}_3)=-0.78$ vs $\sigma ({\text{-OCH}}_3)=-0.27$. The difference ${\sigma }^{+}-\sigma$ measures the resonance contribution to transition-state stabilisation. The correlation $\log (f_{\text{para}})={\rho }^{+}{\sigma }^{+}$ holds well across most substituents.

Yukawa-Tsuno equation. For substituents with substantial resonance demand, the linear free-energy relationship requires an explicit resonance term:

$$\log \frac{k_X}{k_H}=\rho \left[\sigma +r\left({\sigma }^{+}-\sigma \right)\right]$$

where $r$ is the resonance-demand parameter. For electrophiles with highly developed positive charge in the transition state (e.g., nitration), $r\to 1$ and the equation reduces to the Brown ${\sigma }^{+}$ correlation. For electrophiles with less charge development, $r<1$. The Yukawa-Tsuno equation interpolates between the Hammett and Brown correlations.

Selectivity and reactivity. The Reactivity-Selectivity Principle states that more reactive electrophiles are less selective. For toluene nitration, $f_{\text{para}}\approx 60$; for toluene chlorination, $f_{\text{para}}\approx 600$; for toluene bromination, $f_{\text{para}}\approx 2500$. The less reactive bromine is the most selective. The principle follows from the Hammond postulate: a very reactive electrophile has a very early transition state (reactant-like), so the energy difference between ortho, meta, and para attack is small. A less reactive electrophile has a later transition state (product-like, closer to the sigma complex), amplifying the energy differences between positions.

Beyond Huckel: antiaromaticity, Mobius aromaticism, and Clar's sextet theory [Master]

Huckel's $4n+2$ rule addresses monocyclic, planar, fully conjugated systems. The broader landscape of aromaticity includes anti-aromatic destabilisation, twisted-ring topology, triplet-state aromaticity, spherical shells, and polycyclic systems -- each requiring an extension or modification of the basic rule.

Antiaromaticity. A planar, monocyclic, fully conjugated system with $4n$ pi electrons is anti-aromatic: it is less stable than an acyclic polyene with the same number of pi electrons. The canonical example is cyclobutadiene (${\text{C}}_4{\text{H}}_4$, 4 pi electrons). The Huckel MOs are $\alpha +2\beta$, $\alpha$ (doubly degenerate), and $\alpha -2\beta$. Four pi electrons fill the bonding MO (2 electrons) and place one electron each in the degenerate non-bonding pair -- an open-shell diradical. The open-shell configuration incurs exchange repulsion and Jahn-Teller distortion: cyclobutadiene is not square but rectangular, with alternating short and long bonds.

Cyclobutadiene is so reactive that it dimerises at temperatures above 35 K. It has been isolated in an argon matrix at 8 K. The anti-aromatic destabilisation makes it more reactive than 1,3-butadiene (an acyclic diene), which is the defining feature of anti-aromaticity -- it is not merely non-aromatic but actively destabilised.

Pentalene (${\text{C}}_8{\text{H}}_6$, a fused bicyclic system with 8 pi electrons) is also anti-aromatic and has never been isolated as a stable compound, though its dianion (10 pi electrons, $4n+2$ with $n=2$) is stable.

Mobius aromaticism. In 1964 Heilbronner showed that a cyclic pi system with a Mobius twist (an odd number of phase inversions around the ring) reverses the Huckel rule: $4n$ pi electrons become aromatic and $4n+2$ become anti-aromatic ^{[Heilbronner1964]}. The eigenvalues of a Mobius cyclic system are:

$${\epsilon }_k=\alpha +2\beta \cos \left(\frac{(2k+1)\pi }{N}\right)$$

For a Mobius system with 8 pi electrons ($N=8$), the filling pattern produces a closed-shell configuration -- aromatic. Mobius aromaticity was a theoretical prediction for decades. In 2003, Herges and co-workers synthesised a molecule with Mobius topology (a twisted [16]annulene derivative) and confirmed its aromatic character by NICS calculations and X-ray crystallography. Mobius aromaticism is now recognised as a general principle in transition-state aromaticity: pericyclic reactions proceed through aromatic transition states, and whether that state is Huckel or Mobius aromatic determines the stereochemical outcome (Woodward-Hoffmann rules).

Baird's rule for triplet states. In 1972 Baird showed that the aromaticity rules reverse for the lowest triplet state ($S=1$) of cyclic pi systems ^[Baird1972]: $4n$ pi electrons become aromatic and $4n+2$ become anti-aromatic. This reversal follows from the different orbital occupancy in the triplet: the degenerate pair that is half-filled in the singlet ground state of a $4n$-electron system gains exchange stabilisation in the triplet, making the $4n$ system more stable than expected. Baird's rule explains the photochemistry of aromatic compounds: triplet-state benzene is anti-aromatic in the triplet and undergoes ring-opening and addition reactions that ground-state benzene resists.

Spherical aromaticity. Fullerenes are three-dimensional pi systems for which the planar $4n+2$ rule does not apply. Hirsch, Chen, and co-workers (2000) proposed the $2(N+1{)}^2$ rule: a spherical shell with $2(N+1{)}^2$ pi electrons is aromatic, by analogy with the noble-gas electron configurations. For $N=0$: 2 electrons; $N=1$: 8 electrons; $N=2$: 18 electrons; $N=3$: 32 electrons. ${\text{C}}_{60}$ has 60 pi electrons, which does not match $2(N+1{)}^2$ exactly, explaining why ${\text{C}}_{60}$ is not a particularly strong aromatic system -- its stability derives from its geometry (isolated pentagon rule) rather than from a single dominant aromatic shell.

Clar's sextet theory. For polycyclic aromatic hydrocarbons (PAHs), the simple $4n+2$ perimeter count fails because the pi system is not monocyclic. Clar (1958) proposed that the most important resonance structures are those with the maximum number of disjoint aromatic sextets -- benzene-like rings with six pi electrons that behave as local aromatic units. The remaining carbons carry either double bonds or no pi electrons at all.

For naphthalene (${\text{C}}_{10}{\text{H}}_8$): only one Clar sextet can be drawn at a time. The most important resonance structure has the sextet on one ring, with a fixed double bond and a single bond on the other. This predicts that both rings of naphthalene have less than full benzene-like aromatic character, consistent with naphthalene's resonance energy (250 kJ/mol) being less than twice that of benzene (300 kJ/mol).

For phenanthrene vs anthracene (both ${\text{C}}_{14}{\text{H}}_{10}$ isomers): phenanthrene can be drawn with two disjoint Clar sextets (the outer rings), while anthracene can have at most one sextet and a delocalised middle ring. Clar's theory correctly predicts that phenanthrene is more stable and more reactive at the 9,10-positions (the least aromatic part of the molecule).

Aromaticity criteria. Modern computational chemistry uses several quantitative aromaticity indices. The Nucleus-Independent Chemical Shift (NICS), introduced by Schleyer in 1996, measures the magnetic shielding at the ring centre (or 1 A above it). Aromatic rings have negative NICS values (diamagnetic ring current), anti-aromatic rings have positive values (paramagnetic ring current). NICS(0) for benzene is approximately $-10$ ppm; for cyclobutadiene it is approximately $+30$ ppm. The Harmonic Oscillator Model of Aromaticity (HOMA) measures bond-length equalisation: aromatic compounds have nearly equal bond lengths, while non-aromatic and anti-aromatic compounds show bond-length alternation. HOMA = 1 for perfect aromaticity, 0 for non-aromatic systems, and less than 0 for anti-aromatic systems.

Named EAS reactions and side-chain chemistry [Master]

The general EAS mechanism (sigma complex formation, proton loss) operates across a range of named reactions that differ in the electrophile generated and the conditions required. Each reaction has specific synthetic applications and limitations.

Friedel-Crafts alkylation. An alkyl halide ($\text{R-Cl}$ or $\text{R-Br}$) reacts with benzene in the presence of a Lewis acid catalyst (${\text{AlCl}}_3$, ${\text{FeBr}}_3$, ${\text{BF}}_3$) to install an alkyl group. The Lewis acid abstracts the halide, generating a carbocation (or a Lewis acid-base complex that delivers the carbocation character to the ring).

Limitation 1: carbocation rearrangements. Primary carbocations rearrange before attacking the ring. Propyl chloride (${\text{CH}}_3{\text{CH}}_2{\text{CH}}_2\text{Cl}$) generates a primary carbocation that undergoes a 1,2-hydride shift to give the secondary isopropyl carbocation. The product is isopropylbenzene (cumene), not propylbenzene. To install a linear alkyl chain, use Friedel-Crafts acylation followed by reduction instead.

Limitation 2: polyalkylation. The alkyl group installed by Friedel-Crafts alkylation is an electron-donating, ortho/para-directing substituent. After the first alkylation, the product is more reactive than the starting material. The reaction tends toward polyalkylation unless the benzene is used in large excess.

Limitation 3: scope restriction. Friedel-Crafts alkylation fails on strongly deactivated rings (nitrobenzene, benzaldehyde, benzoic acid) because the ring is too electron-poor to attack the carbocation. It also fails with aryl and vinyl halides (sp${}^2$ carbons do not form stable carbocations under these conditions).

Friedel-Crafts acylation. An acyl chloride ($\text{RCOCl}$) reacts with benzene in the presence of ${\text{AlCl}}_3$ to install an acyl group, giving a ketone ($\text{Ar-COR}$). The electrophile is the acylium ion ${\text{RCO}}^{+}$, generated by Lewis acid abstraction of chloride.

The acylium ion is resonance-stabilised: ${\text{R-C}}^{+}\text{=O}\leftrightarrow {\text{R-C=O}}^{+}$. This resonance stabilisation prevents carbocation rearrangements -- the acyl group does not migrate. Additionally, the ketone product is a meta-directing, deactivating group, so polyacylation does not occur. These two advantages (no rearrangement, no polyacylation) make Friedel-Crafts acylation synthetically more reliable than alkylation.

The ketone product can be reduced to the corresponding alkyl chain by Clemmensen reduction (Zn(Hg)/HCl) or Wolff-Kishner reduction (hydrazine, KOH, heat). This two-step sequence installs an unrearranged alkyl chain on the aromatic ring.

Vilsmeier-Haack formylation. This reaction installs a formyl group ($\text{-CHO}$) on activated aromatic rings using DMF and ${\text{POCl}}_3$. The electrophile is an iminium ion: $({\text{CH}}_3{)}_2{\text{N}}^{+}\text{=CHCl}$. Attack by the aromatic ring, followed by hydrolysis, gives the aldehyde. The reaction works on activated rings (phenols, anilines, heterocycles like pyrrole and indole) but not on benzene itself or deactivated rings.

Gattermann-Koch formylation. This reaction installs a formyl group using CO and HCl in the presence of ${\text{AlCl}}_3$/CuCl. The electrophile is the formyl cation ${\text{HCO}}^{+}$. The reaction works on benzene (where Vilsmeier-Haack fails), making the two reactions complementary. Gattermann-Koch requires high-pressure CO equipment, limiting its practical use.

Benzylic halogenation (NBS). The benzylic position is especially reactive toward radical halogenation. The benzylic C-H bond is weakened by resonance stabilisation of the resulting benzylic radical -- the unpaired electron delocalises into the aromatic ring. N-Bromosuccinimide (NBS) maintains a low, steady-state concentration of ${\text{Br}}_2$, preventing electrophilic bromination of the ring itself. The mechanism is radical: NBS generates ${\text{Br}}_2$, which cleaves to ${\text{Br}}^{\bullet }$. The bromine radical abstracts a benzylic hydrogen, generating the benzylic radical, which reacts with ${\text{Br}}_2$ to give the benzylic bromide and regenerate ${\text{Br}}^{\bullet }$.

Side-chain oxidation. Alkylbenzenes with at least one benzylic hydrogen are oxidised to benzoic acids by strong oxidising agents (alkaline ${\text{KMnO}}_4$ or ${\text{Na}}_2{\text{Cr}}_2{\text{O}}_7$/H${}_2$SO${}_4$). Toluene gives benzoic acid. Ethylbenzene also gives benzoic acid -- both carbons of the ethyl group are removed. The reaction proceeds through sequential oxidation: alkyl $\to$ alcohol $\to$ aldehyde/ketone $\to$ carboxylic acid. tert-Butylbenzene has no benzylic hydrogens and is resistant to oxidation.

Full proof set [Master]

Proposition 1 (4n+2 closed-shell rule from eigenvalue spectrum). Let $N$ be a positive even integer and let the Huckel eigenvalues of an $N$-atom monocyclic polyene be ${\epsilon }_k=\alpha +2\beta \cos (2\pi k/N)$ for $k=0,\pm 1,\dots ,\pm (N/2-1),N/2$. If the number of pi electrons is $4n+2$ for some non-negative integer $n$, then all bonding MOs are completely filled and all antibonding MOs are empty (closed-shell configuration). If the number of pi electrons is $4n$ with $n\ge 1$, the highest occupied MOs are a degenerate pair with one electron each (open-shell configuration).

Proof. The MO energies come in a single non-degenerate level at $\alpha +2\beta$ (bonding), a single non-degenerate level at $\alpha -2\beta$ (antibonding), and $(N-2)/2$ pairs of doubly degenerate levels between these extremes. Each degenerate pair accommodates 4 electrons (2 per orbital). The bonding orbitals are those with ${\epsilon }_k<\alpha$ (i.e., $\cos (2\pi k/N)>0$). The number of bonding MOs (counting degeneracy) is:

• 1 non-degenerate bonding MO (at $\alpha +2\beta$): capacity 2 electrons
• $(N-2)/4$ degenerate bonding pairs (each below $\alpha$): capacity $4\times (N-2)/4=N-2$ electrons

Total bonding capacity = $2+(N-2)=N$ electrons. When $N=4n+2$, exactly $4n+2$ electrons fill $n+1$ bonding MOs (counting the non-degenerate bottom plus $n$ degenerate pairs), using all bonding capacity and leaving antibonding orbitals empty.

When the electron count is $4n$ (for $n\ge 1$), the highest bonding or non-bonding degenerate pair receives only 2 electrons instead of 4, leaving two singly-occupied orbitals (open shell). The energy of this configuration exceeds that of the closed-shell reference because of exchange repulsion and the absence of a bonding contribution from the singly-occupied orbitals. $\square$

Proposition 2 (Ortho/para direction -- quantitative activation-energy argument). Let X be a substituent at C1 of benzene with a resonance or hyperconjugative ability to donate electron density. The activation energy for EAS at the para position, $\Delta G_{\text{para}}^{‡}$, is lower than the activation energy for meta attack, $\Delta G_{\text{meta}}^{‡}$, by an amount proportional to the resonance stabilisation that X provides when positive charge appears on C1.

Proof. The activation energy for EAS at a given position is governed by the energy of the sigma complex (Wheland intermediate) via the Hammond postulate. For para attack at C4, the resonance structures of the sigma complex place positive charge on C1, C3, and C5. When positive charge appears on C1, X donates electron density through resonance (for ${\text{-OCH}}_3$, ${\text{-NH}}_2$) or hyperconjugation (for ${\text{-CH}}_3$), stabilising this structure by an amount $\Delta E_{\text{res}}$.

For meta attack at C3, the resonance structures place positive charge on C2, C4, and C6. None of these carbons is C1. X cannot provide direct resonance stabilisation. The sigma complex for meta attack lacks the $\Delta E_{\text{res}}$ stabilisation.

The energy difference between the two sigma complexes is approximately $\Delta E_{\text{res}}$. Since sigma complex formation is the rate-determining step and the Hammond postulate places the transition state close in energy to the sigma complex:

$$\Delta G_{\text{meta}}^{‡}-\Delta G_{\text{para}}^{‡}\approx \Delta E_{\text{res}}>0$$

The rate ratio follows from the Eyring equation: $k_{\text{para}}/k_{\text{meta}}\approx \exp (\Delta E_{\text{res}}/RT)$. For toluene nitration, $\Delta E_{\text{res}}\approx 10$ kJ/mol gives $k_{\text{para}}/k_{\text{meta}}\approx 60$, consistent with the observed partial rate factor. $\square$

Connections [Master]

• Electrophilic addition to alkenes 15.05.01. EAS is the aromatic counterpart of electrophilic addition. Both begin with electrophilic attack on a pi system forming a carbocation-like intermediate, but in EAS the intermediate restores aromaticity by proton loss while in alkene addition it is trapped by a nucleophile. The sigma complex in EAS is mechanistically analogous to the carbocation intermediates in 15.05.01. This unit's resonance analysis of the sigma complex builds toward the same charge-stabilisation logic used in 15.05.01 for predicting Markovnikov regioselectivity.

• Stereochemistry of aromatic systems 15.01.01. Substituted biphenyls with large ortho substituents can exhibit axial chirality (atropisomerism) when rotation about the biaryl bond is restricted. The design of axially chiral ligands for asymmetric catalysis uses EAS to install the ortho substituents that create the steric barrier. The planarity requirement for aromaticity connects to the conformational analysis in 15.01.01.

• MO theory for homonuclear diatomics 14.05.02 pending. The Huckel MO framework is the pi-system analogue of the diatomic MO theory in 14.05.02 pending: both construct molecular orbitals from atomic basis functions, both fill levels by the Aufbau principle, and both predict that closed-shell configurations are more stable than open-shell ones. The Frost circle for cyclic systems plays the same pedagogical role as the correlation diagram for homonuclear diatomics.

• Carbonyl chemistry 15.07.01. Friedel-Crafts acylation installs a carbonyl group on the ring, connecting aromatic chemistry to carbonyl reactivity. The acylium ion electrophile is a resonance-stabilised carbonyl species, and the aryl ketone products undergo the reactions covered in 15.07.01.

• Biomolecules: aromatic amino acids 17.01.01. The aromatic amino acids phenylalanine, tyrosine, and tryptophan are aromatic because of Huckel stability. Their UV absorption (used in protein quantification at 280 nm), their participation in pi-stacking interactions that stabilise protein and nucleic acid structures, and the tyrosine hydroxyl group's ortho/para-directing character in enzymatic modifications all follow from the aromatic chemistry covered here.

Historical & philosophical context [Master]

Friedrich Kekule proposed the cyclic structure of benzene in 1865 ^[Kekule1865]. His proposal -- a hexagonal ring of six carbons with alternating single and double bonds -- explained the lack of isomers in monosubstituted benzenes but could not explain why benzene does not undergo addition reactions like an alkene, nor why all C-C bond lengths in benzene are equal. The "Kekule structure" implies that 1,2-disubstituted benzene should have two isomers (adjacent single vs adjacent double bond), but only one is observed.

Erich Huckel's 1931 quantum-mechanical treatment provided the mathematical framework ^[Huckel1931]. By reducing the full Schrodinger equation for the pi system to a one-electron Hamiltonian on the p-orbital basis, Huckel derived the $4n+2$ rule from the eigenvalue structure of the cyclic adjacency matrix. Frost and Musulin provided the visual mnemonic in 1953 ^{[FrostMusulin1953]}. Linus Pauling's resonance theory (1930s) described benzene as a hybrid of two Kekule structures, with the true structure being an average -- a description that is qualitatively correct but lacks the predictive power of the MO treatment.

The question of whether a chemical bond in benzene is "really" single or double has no answer at the quantum-mechanical level: all C-C bonds in benzene are equivalent with bond order 1.5, and the two "Kekule structures" are representations of the same delocalised wavefunction, not two structures in rapid equilibrium.

Bibliography [Master]

@article{Kekule1865,
  author = {Kekul\'e, F. A.},
  title = {Sur la constitution des substances aromatiques},
  journal = {Bull. Soc. Chim. Fr.},
  volume = {3},
  year = {1865},
  pages = {98--110},
}

@article{Huckel1931,
  author = {H\"uckel, E.},
  title = {Quantentheoretische Beitr\"age zum Benzolproblem},
  journal = {Z. Phys.},
  volume = {70},
  year = {1931},
  pages = {204--286},
}

@article{FrostMusulin1953,
  author = {Frost, A. A. and Musulin, B.},
  title = {Mnemonic device for molecular-orbital energies},
  journal = {J. Chem. Phys.},
  volume = {21},
  year = {1953},
  pages = {572--573},
}

@article{Heilbronner1964,
  author = {Heilbronner, E.},
  title = {H\"uckel molecular orbitals of M\"obius-type conformations of annulenes},
  journal = {Tetrahedron Lett.},
  volume = {5},
  year = {1964},
  pages = {1923--1928},
}

@article{Baird1972,
  author = {Baird, N. C.},
  title = {Quantum organic photochemistry. {II}. {R}esonance and aromaticity in the lowest ${}^3\pi\pi^*$ state of cyclic hydrocarbons},
  journal = {J. Am. Chem. Soc.},
  volume = {94},
  year = {1972},
  pages = {4941--4948},
}

@book{Clayden2012,
  author = {Clayden, J. and Greeves, N. and Warren, S.},
  title = {Organic Chemistry},
  edition = {2nd},
  publisher = {Oxford UP},
  year = {2012},
}

@book{March2013,
  author = {Smith, M. B.},
  title = {March's Advanced Organic Chemistry},
  edition = {7th},
  publisher = {Wiley},
  year = {2013},
}

@book{CareySundberg2007,
  author = {Carey, F. A. and Sundberg, R. J.},
  title = {Advanced Organic Chemistry, Part A: Structure and Mechanisms},
  edition = {5th},
  publisher = {Springer},
  year = {2007},
}

@book{AnslynDougherty2006,
  author = {Anslyn, E. V. and Dougherty, D. A.},
  title = {Modern Physical Organic Chemistry},
  publisher = {University Science Books},
  year = {2006},
}