The Hammett Equation

What the Hammett equation does

Put a substituent on a benzene ring — a nitro group, a methoxy, a chlorine — and every reaction that happens elsewhere on that ring changes speed. A nitro group can accelerate one reaction a thousandfold and slow another by the same factor. The Hammett equation makes that mess quantitative. It says the change is a simple product of two numbers:

    log(k / k₀)  =  σ · ρ           (for rates)
    log(K / K₀)  =  σ · ρ           (for equilibria)

  k   = rate constant for the substituted compound
  k₀  = rate constant for the parent (H, unsubstituted)
  σ   = substituent constant  (property of the group + its position)
  ρ   = reaction constant     (property of the reaction + conditions)

The genius is the separation of variables. σ depends only on what the substituent is and where it sits (meta or para) — it is the same number in every reaction. ρ depends only on the reaction — the mechanism, the solvent, the temperature. Measure a handful of rates, plot log(k/k₀) against the tabulated σ values, and you get a straight line whose slope is ρ. That single number is a fingerprint of the transition state.

σ: the substituent constant, born from benzoic acid

Hammett needed a ruler for "electron pull." He chose the acidity of substituted benzoic acids in water at 25 °C, because acid ionization is clean, measurable to three decimal places, and the carboxyl group is a fixed distance from the ring. He defined:

    σ  =  log(Kₐ / Kₐ₀)  =  pKₐ₀ − pKₐ

  where Kₐ  = acid dissociation constant of X-C₆H₄-COOH
        Kₐ₀ = acid dissociation constant of C₆H₅-COOH (X = H)

The logic of the sign follows directly. An electron-withdrawing group pulls density out of the ring, stabilizes the carboxylate anion, makes the acid stronger (higher Kₐ), and therefore gets a positive σ. An electron-donating group destabilizes the anion, weakens the acid, and gets a negative σ. Hydrogen is the origin: σ(H) = 0.

• Strong acceptors (positive σ): p-NO₂ (+0.78), m-NO₂ (+0.71), p-CN (+0.66), p-CF₃ (+0.54), m-Cl (+0.37), p-Cl (+0.23).
• Reference: H (0.00).
• Donors (negative σ): p-CH₃ (−0.17), p-OCH₃ (−0.27), p-NH₂ (−0.66), p-N(CH₃)₂ (−0.83).

Notice that chlorine is a donor by resonance but an acceptor by induction; at the para position the inductive win gives σ(p-Cl) = +0.23. Methoxy is the opposite story: from meta its inductive pull dominates (σ(m-OCH₃) = +0.12, slightly positive), but from para its lone-pair donation into the ring wins (σ(p-OCH₃) = −0.27, negative). The same group has different σ from different positions — that positional dependence is exactly the resonance/induction bookkeeping the equation packages for you.

ρ: the reaction constant, and what its sign reveals

Where σ is fixed per group, ρ is fixed per reaction. It is the slope of the plot, and it answers two questions at once:

1. The sign tells you the charge. A positive ρ means electron-withdrawing groups (positive σ) speed the reaction up — so negative charge is building at, or near, the reacting center in the rate-determining step, and withdrawing groups stabilize it. A negative ρ means donating groups accelerate it — so positive charge is building, and donors stabilize it.
2. The magnitude tells you how much. A large |ρ| means the reacting center feels the ring's electronics strongly — a lot of charge develops, close to the ring. A small |ρ| means the transition state barely notices; the charge is small, distant, or delocalized.

By construction, benzoic acid ionization has ρ = +1.00. Ester hydrolysis, nitration, solvolysis, and enzyme reactions are all measured on that same scale. A few landmark values:

• ρ ≈ +2.5: alkaline hydrolysis of ethyl benzoates — hydroxide attacks the carbonyl, negative charge builds in the tetrahedral intermediate, withdrawing groups help a lot.
• ρ ≈ −4.5 (vs σ⁺): SN1 solvolysis of cumyl chlorides — a benzylic carbocation forms in the rate step; donors stabilize the positive charge dramatically.
• ρ ≈ −12 (vs σ⁺): electrophilic aromatic bromination — an unusually large negative ρ because the ring itself is the nucleophile and huge positive charge builds in the arenium-ion transition state.
• ρ ≈ 0: free-radical or pericyclic steps, where the reacting center carries little net charge.

σ⁺ and σ⁻: when resonance reaches the reacting center

The ordinary σ was calibrated on a reaction where the substituent is not in direct resonance with the reacting center — the carboxylate of benzoic acid is insulated by the C–C bond. When your reacting center can conjugate directly through the ring to a para substituent, ordinary σ under-counts the effect. Hammett's successors built two extra scales:

• σ⁺ (H. C. Brown & Okamoto, 1958) — calibrated on cumyl chloride solvolysis, where a full positive charge builds one bond from the ring. Use it when positive charge develops at the reacting center. Para donors get much more negative: p-OCH₃ σ⁺ = −0.78 (vs σ = −0.27), p-N(CH₃)₂ σ⁺ = −1.70. A lone pair on the para donor pushes straight into the empty orbital.
• σ⁻ — used when negative charge develops that a para acceptor can delocalize onto by resonance, as in phenol ionization or nucleophilic aromatic substitution. Para acceptors get much more positive: p-NO₂ σ⁻ = +1.27 (vs σ = +0.78), p-CN σ⁻ = +1.00.

The practical test is empirical and beautiful: fit your data against σ, σ⁺, and σ⁻. Whichever gives the straightest line tells you which resonance channel is open. If cumyl-type solvolysis only lines up against σ⁺, you have direct evidence that a resonance-stabilized cation forms in the rate step.

Why "linear free-energy relationship" — the thermodynamic backbone

The Hammett equation is the archetypal linear free-energy relationship (LFER), and the name is not decoration. Rate constants relate to free energy of activation by the Eyring equation, log k ∝ −ΔG‡/2.303RT, and equilibrium constants relate to reaction free energy, log K = −ΔG°/2.303RT. So log(k/k₀) = σρ is really a statement that the change in activation free energy caused by a substituent is linearly proportional to the change in ionization free energy it causes in benzoic acid:

    −ΔΔG‡ (your reaction)   =   ρ · (−ΔΔG° of benzoic acid ionization)
    ───────────────────         ─────────────────────────────────────
       what you measure                 the σ ruler, scaled by ρ

That linearity is not guaranteed by any law — it is an empirical gift. It works when the substituent perturbs every member of a reaction series through the same physical mechanism (the same blend of induction and resonance), so a single scaling factor ρ captures the whole family. When two different mechanisms compete, the linearity breaks — and that break is the most informative thing a Hammett plot can do (see below).

Hammett vs related structure–reactivity treatments

Treatment	What it correlates	Best for	Key parameter
Hammett (σ, ρ)	Rate/equilibrium vs meta/para electronics	Aromatic side-chain reactions	ρ (charge & sensitivity)
Hammett σ⁺ / σ⁻	Same, with direct resonance to the center	Cationic (σ⁺) or anionic (σ⁻) TSs	ρ⁺, ρ⁻
Taft (σ*, Es)	Aliphatic + ortho, splits polar vs steric	Non-aromatic and ortho systems	ρ*, δ (steric)
Yukawa–Tsuno	Blends σ and σ⁺ with a resonance weight r	Partial resonance demand	ρ, r (0→1)
Brønsted (α, β)	Rate vs acidity/basicity of catalyst	General acid/base catalysis	α, β (proton transfer)
Grunwald–Winstein (m, Y)	Solvolysis rate vs solvent ionizing power	Solvent effects on SN1	m (charge development)

Worked example: which mechanism hydrolyzes an ester?

Suppose you measure the alkaline hydrolysis (saponification) of a series of ethyl benzoates, X-C₆H₄-CO₂Et, with hydroxide at 25 °C, and get these relative rates:

    X = p-NO₂   log(k/k₀) = +2.0    σ = +0.78
    X = p-Cl    log(k/k₀) = +0.6    σ = +0.23
    X = H       log(k/k₀) =  0.0    σ =  0.00
    X = p-CH₃   log(k/k₀) = −0.4    σ = −0.17
    X = p-OCH₃  log(k/k₀) = −0.7    σ = −0.27

Plot log(k/k₀) on the y-axis against σ on the x-axis. The five points fall on a straight line through the origin with slope

    ρ  =  Δ[log(k/k₀)] / Δσ  ≈  (+2.0 − (−0.7)) / (0.78 − (−0.27))  ≈  +2.6

Read it off: ρ is positive, so negative charge builds at the reacting carbon — hydroxide is adding to the carbonyl and the carbonyl carbon becomes an electron-rich tetrahedral alkoxide (the B_AC2 mechanism). ρ ≈ +2.6, larger than the benchmark +1, tells you the reacting center feels the ring strongly because the negative charge sits close to it. The plot correlates against ordinary σ (not σ⁻), confirming the developing charge is on oxygen of the tetrahedral center, insulated from direct ring resonance. One graph, and you have the mechanism, the charge, and its location.

The payoff: what a curved or broken plot means

A Hammett plot that is not a single straight line is the most valuable result of all, because the equation only stays linear while a single mechanism operates. Three classic shapes:

• Two straight branches, V-shaped (concave up). Electron-rich rings react by one pathway (say, rate-limiting nucleophilic addition, negative ρ), and at some crossover substituent electron-poor rings switch to a different pathway (say, rate-limiting departure of a leaving group, positive ρ). The minimum marks the mechanistic switch. Nucleophilic aromatic substitution and some E1cb eliminations show this.
• Concave-down curve. Often a change in the rate-determining step within a two-step mechanism: as substituents shift the relative barriers, control passes from one step to the other, and the slope glides from one ρ to another.
• Better fit to σ⁺ than σ, or vice versa. Not a break, but a diagnosis of how much resonance reaches the center. A large positive ρ against σ⁻ is a signature of anionic transition states; a large negative ρ against σ⁺ is the signature of a benzylic cation.

The Yukawa–Tsuno equation (1959) formalizes the middle ground: log(k/k₀) = ρ[σ + r(σ⁺ − σ)], where the resonance parameter r runs from 0 (pure σ behavior) to 1 (pure σ⁺ behavior), quantifying partial resonance demand at the transition state.

Where the equation earns its keep

• Mechanism assignment. The everyday use in physical organic chemistry: measure five to eight rates, plot, read the sign and slope, and constrain the transition state before you ever draw an arrow-pushing mechanism.
• Drug design (QSAR). Corwin Hansch extended σ into quantitative structure–activity relationships in the 1960s, combining Hammett σ with a hydrophobicity parameter π (from octanol/water partition) to correlate biological potency with structure. Modern medicinal chemistry's log P and electronic descriptors trace directly back to Hammett's σ.
• Catalyst tuning. Ligand electronics on a metal catalyst are optimized by building a Hammett series of para-substituted aryl ligands (e.g. triarylphosphines) and correlating turnover or enantioselectivity against σ — a fast way to know whether the catalyst wants more electron density or less.
• Enzyme and materials chemistry. ρ values from substituted substrates probe charge development in enzyme active sites; substituent tuning of conjugated organic materials uses σ to dial HOMO/LUMO levels.

Limitations and where it fails

• Ortho substituents are out. The whole framework is meta/para only; ortho groups add steric and field effects that no single σ can carry. Use the Taft treatment for those.
• Saturated and non-benzenoid systems. σ is defined for benzene-ring communication. Aliphatic chains, saturated rings, and heteroaromatics need their own parameter sets (σ*, or ring-specific σ tables for pyridine, thiophene, etc.).
• Mixed or changing mechanisms. If the mechanism shifts across the series, you get a broken plot — informative, but the single-ρ equation no longer describes the whole set.
• Wrong σ scale. Forcing σ where σ⁺ or σ⁻ is physically appropriate produces scatter that looks like noise but is really a resonance mismatch. Always test all three scales.
• Solvent and temperature dependence. ρ is not a constant of nature — it varies with solvent (more so for charge-separating reactions) and with temperature (ρ generally shrinks toward zero as T rises, since RT in the denominator grows). Always report the conditions.

Historical note: Hammett, Columbia, 1937

Louis Plack Hammett (1894–1987), a physical chemist at Columbia University, published "The Effect of Structure upon the Reactions of Organic Compounds. Benzene Derivatives" in the Journal of the American Chemical Society in 1937. He was drawing together scattered observations — including J. F. J. Dippy's careful benzoic-acid pKₐ measurements — into a single quantitative law and tabulating the first σ constants. Three years later, in his 1940 textbook Physical Organic Chemistry, he coined the name of an entire subdiscipline and placed the equation at its center. The follow-on scales came from his intellectual descendants: Herbert C. Brown and Yukio Okamoto defined σ⁺ in 1958, Robert Taft split polar and steric effects for aliphatic systems in the early 1950s, Yukawa and Tsuno blended the scales in 1959, and Corwin Hansch carried σ into pharmacology in the 1960s. Nearly ninety years on, the Hammett plot is still the first thing a physical organic chemist reaches for when a mechanism is in doubt.