Linear Free-Energy Relationships

The extrathermodynamic idea behind the line

LFERs are called extrathermodynamic because no law of thermodynamics requires them. Free energy is a state function, but rate constants depend on the transition state, which lies outside equilibrium thermodynamics. The straight line is an empirical gift: it appears when a substituent perturbs the transition state and the product (or a model equilibrium) by proportional amounts.

Formally, for a reference equilibrium we define a substituent constant from ΔΔG°, and for the reaction of interest we measure ΔΔG‡. If

ΔΔG‡ = ρ · ΔΔG°_ref

then dividing by −2.303RT gives the familiar log(k/k₀) = ρσ. The reaction constant ρ is the slope; the substituent constant σ is a transferable number tabulated once and reused across thousands of reaction series. That transferability is the practical payoff: you measure σ from one clean reference reaction and predict rates everywhere else.

The Hammett equation as the prototype

Hammett chose the ionization of meta- and para-substituted benzoic acids in water at 25 °C as his reference. He defined σ = log(K_X/K_H), so an electron-withdrawing group that stabilizes the carboxylate anion gives a positive σ (e.g. p-NO₂ ≈ +0.78, p-Cl ≈ +0.23) and an electron-donating group gives a negative σ (p-OCH₃ ≈ −0.27, p-NH₂ ≈ −0.66). By definition σ = 0 for H and ρ = 1 for the reference reaction.

The sign and size of ρ for a new reaction is the mechanistic readout:

• ρ > 0 — negative charge builds up at the transition state (electron-withdrawing groups accelerate). Base-promoted ester hydrolysis of ethyl benzoates has ρ ≈ +2.5.
• ρ < 0 — positive charge builds up (electron-donating groups accelerate), typical of SN1 solvolysis and electrophilic aromatic substitution.
• ρ ≈ 0 — little charge change, or a radical/pericyclic pathway insensitive to polar substituents.

Because plain σ underestimates direct resonance with a charged center, Brown introduced σ⁺ (from cumyl chloride solvolysis) for reactions building cationic character and σ⁻ for anionic centers conjugated to the substituent, e.g. p-NO₂ reaching σ⁻ ≈ +1.27 in phenol ionization.

Reading breaks and curvature in the plot

A perfectly straight Hammett plot means every substituent reacts by the same mechanism and rate-determining step. Deviations are diagnostic, not noise:

• A break (two intersecting lines) signals a change in rate-determining step or mechanism as substituents change. Classic example: nucleophilic aromatic substitution where addition is rate-limiting for some substituents and elimination for others.
• Concave-upward curvature can indicate a shift between competing pathways (e.g. SN1 vs SN2) whose relative rates cross over with substituent.
• A single point off the line often flags direct resonance — the tell that you should switch to σ⁺ or σ⁻.

Because ρ absorbs the temperature and solvent factor (ρ ∝ 1/T), comparing ρ across conditions also probes how ‘tight’ the transition state is. A large |ρ| means a lot of charge and a strongly perturbed TS; a small |ρ| means an early, reactant-like TS in the Hammond sense.

Beyond aromatics: Taft, Brønsted, Swain–Scott, Grunwald–Winstein

Hammett's benzene scaffold holds substituent geometry fixed, but aliphatic systems mix electronic and steric effects. Taft (1952) separated them by comparing acid-catalyzed and base-catalyzed ester hydrolysis: the acidic pathway is nearly insensitive to polar effects, so its rate isolates steric strain (the parameter E_s), while the difference between the two isolates the polar effect (σ*). This gives a two-parameter LFER, log(k/k₀) = ρ*σ* + δE_s.

Other members of the family use entirely different reference axes:

• Brønsted catalysis law (1924): log k_cat = α log K_a + C for general acid catalysis (β for general base). The exponent α (0–1) reports the degree of proton transfer at the transition state — a direct measure of position along the reaction coordinate.
• Swain–Scott (1953): log(k/k₀) = s n, ranking nucleophiles by n (CN⁻, I⁻, thiolate high; water = 0) with substrate sensitivity s.
• Grunwald–Winstein (1948): log(k/k₀) = m Y, correlating solvolysis rate with solvent ionizing power Y (tert-butyl chloride reference). An m near 1 signals a fully ionized, SN1-like TS.

Marcus theory can be viewed as the LFER's curved big brother: it predicts a quadratic relation between ΔG‡ and ΔG° that reduces to a Brønsted-type line (slope 0.5) near thermoneutral reactions.

Why chemists still lean on LFERs

LFERs remain one of the cheapest, most information-dense mechanistic tools in physical organic chemistry:

• Diagnosing mechanism — the sign of ρ distinguishes cationic from anionic transition states without isotopes, spectroscopy, or computation.
• Predicting rates — once ρ is known for a series, a tabulated σ predicts the rate of an untested substrate, useful in process chemistry and reaction screening.
• Drug design and QSAR — Hansch's quantitative structure–activity relationships extended σ (and Hammett's lipophilicity partner π) to correlate biological activity with substituent electronics and hydrophobicity, seeding modern medicinal-chemistry QSAR.
• Catalysis and enzymology — Brønsted α/β values quantify transition-state proton transfer in general acid–base catalysis, from small molecules to enzyme active sites.

Their limit is honesty about assumptions: an LFER only holds while one mechanism and rate-determining step dominate. When that fails, the broken line itself becomes the discovery.

The main LFER families and what each parameter measures

Relationship	Parameter	What it probes	Reference reaction
Hammett	σ (meta/para)	Electronic effect of aryl substituents	Benzoic acid pKₐ in water
Hammett–Brown	σ⁺ / σ⁻	Direct resonance with cationic / anionic TS	Cumyl chloride solvolysis / phenol pKₐ
Taft	σ* and Eₛ	Separated polar and steric aliphatic effects	Ester hydrolysis (acid vs base)
Brønsted	pKₐ of catalyst	Proton transfer in general acid/base catalysis	Acid/base dissociation
Swain–Scott	n (nucleophilicity)	Nucleophile strength in SN2	CH₃Br in water (n = 0)