The Hammond Postulate

What the Hammond postulate does

A transition state is the single highest point along a reaction's path — the fleeting arrangement of atoms at the top of the energy barrier. It lasts about the time of one bond vibration (≈10⁻¹³ s) and can never be isolated, crystallized, or spectroscopically caught. So how do chemists reason about its structure? The Hammond postulate is the answer: it lets you replace the invisible transition state with a species you can draw.

The statement, from George Hammond's 1955 paper, is this:

If two states, as for example a transition state and an unstable intermediate, occur consecutively during a reaction process and have nearly the same energy content, their interconversion will involve only a small reorganization of the molecular structure.

In plainer terms: along a smooth energy landscape, things that are close in energy are close in geometry. So the transition state resembles whichever stable species — reactant or product — sits nearest to it on the energy diagram.

• Exothermic step (ΔH < 0). The barrier is closer to the reactants in energy. The transition state comes early, and looks like the reactants: bonds are barely broken or formed.
• Endothermic step (ΔH > 0). The barrier is closer to the products in energy. The transition state comes late, and looks like the products: bonds are almost fully broken and formed.
• Thermoneutral step (ΔH ≈ 0). The transition state is symmetric, roughly midway between reactant and product in both energy and structure.

The immediate payoff: anything that stabilizes a late transition state's resemblance-partner also stabilizes the transition state itself, and so lowers the barrier and speeds the reaction. That single sentence connects thermodynamic stability (of an intermediate or product) to kinetic rate — the bridge that makes the postulate so useful.

Reading it off the reaction-coordinate diagram

Draw energy on the vertical axis against reaction progress on the horizontal. The transition state is the maximum. Hammond's postulate is a statement about where along the horizontal axis that maximum falls.

  EXOTHERMIC step               ENDOTHERMIC step
  (fast, early TS)              (slow, late TS)

   E |   ‡  (early)              E |          ‡  (late)
     |  / \                        |         / \
     | /   \                       |        /   \____ products
  R__|/     \____                  |       /
     |       products           R__|______/
     +------------→ rxn            +------------→ rxn
     TS looks like R               TS looks like P

Notice the geometry of the curve does the work. In the exothermic case the peak is shifted left, sitting over the reactant valley — hence "early" and reactant-like. In the endothermic case the peak is shifted right, perched over the product valley — "late" and product-like. This is the Leffler-Hammond picture, and the fraction of the way the transition state sits toward the product is often written as the parameter α (0 = reactant-like, 1 = product-like).

The physical-organic logic, step by step

The postulate earns its keep on multistep mechanisms that pass through a high-energy intermediate. Take a generic two-step reaction going through an unstable intermediate I:

    R  ──(step 1, slow, endothermic)──→  I  ──(step 2, fast)──→  P
                    ‡₁ is LATE                       ‡₂ is EARLY
                    (looks like I)                   (looks like I)

1. Identify the rate-determining step. The tall barrier — usually the endothermic formation of the reactive intermediate I.
2. Locate its transition state. Because step 1 is uphill, its transition state (‡₁) is late: it resembles the intermediate I far more than the starting material R.
3. Transfer stability from I to ‡₁. Whatever makes I more stable — an extra alkyl group donating hyperconjugation into a carbocation, an aryl group offering resonance to a radical, a polar solvent solvating a charge — is already present at the transition state, because the transition state looks like I. So the barrier for the fast-forming, more-stable intermediate is lower.
4. Predict the faster pathway. The reaction that produces the more stable intermediate is the faster reaction. Thermodynamic stability of a species you never isolate dictates the kinetic outcome.

This is why you can reason about carbocation "stability" (a thermodynamic property) to predict SN1 and E1 rates (a kinetic property). The two are linked only because the postulate makes the transition state a stand-in for the cation.

Worked example: SN1 carbocation selectivity

Solvolysis of an alkyl halide in the SN1 regime is rate-limited by ionization to the carbocation — a strongly endothermic step (breaking C-X, separating charge in solution). By Hammond, its transition state is late and closely resembles the carbocation.

    R-X  ──slow, endothermic──→  R⁺  +  X⁻   (rate-determining, LATE TS)
    R⁺   ──fast──→  R-Nu                       (trapping, EARLY TS)

Now compare a tertiary versus a secondary substrate. The tertiary carbocation is stabilized by three hyperconjugating C-H/C-C bonds and by inductive donation; it sits far lower in energy than the secondary cation (in the gas phase the difference is on the order of 60-80 kJ/mol). Because the ionization transition state looks like the cation, that stabilization is felt at the barrier top:

• Tertiary (e.g. tert-butyl chloride). Low-energy cation → low, late transition state → fast solvolysis.
• Secondary (e.g. isopropyl chloride). Higher-energy cation → higher transition state → slow.
• Primary/methyl. Cation is prohibitively unstable → SN1 essentially doesn't happen; the mechanism switches to SN2.

Measured 2-adamantyl vs. tert-butyl and the classic tert-butyl : isopropyl : ethyl solvolysis series span many orders of magnitude in rate — a rate range that maps directly onto the stability range of the cations, exactly as the late-transition-state picture requires. The same logic explains why Markovnikov addition of HX to an alkene protonates to give the more substituted (more stable) carbocation: the protonation step is endothermic, its transition state resembles the cation, and the more stable cation forms faster.

Worked example: chlorine vs bromine radical selectivity

Free-radical halogenation is the textbook demonstration of the postulate, because two nearly identical reactions land on opposite sides of it. The selectivity-determining step is hydrogen abstraction by the halogen atom:

    X•  +  R-H  ──→  X-H  +  R•

  Cl•  +  R-H:   ΔH ≈ -21 kJ/mol (secondary C-H)   EXOTHERMIC → early TS
  Br•  +  R-H:   ΔH ≈ +45 kJ/mol (secondary C-H)   ENDOTHERMIC → late TS

Because chlorine abstraction is exothermic, its transition state is early: the C-H bond is barely stretched, the incipient carbon radical is barely formed, and the barrier is nearly blind to how stable the eventual radical would be. Chlorine therefore abstracts 3°, 2° and 1° hydrogens with only mild preference — the relative per-hydrogen reactivities at 25 °C are roughly 5.0 : 3.8 : 1.0.

Bromine abstraction is endothermic, so its transition state is late: the C-H bond is nearly broken, the carbon radical is nearly fully developed, and the barrier feels the full difference in radical stability. Bromine is therefore extremely selective — relative reactivities of roughly 1600 : 82 : 1 for 3° : 2° : 1°. This is why radical bromination is a clean synthetic tool for tertiary and benzylic C-H bonds while radical chlorination gives statistical mixtures.

Early vs late transition state

	Early (reactant-like) transition state	Late (product-like) transition state
Step energetics	Exothermic (ΔH < 0)	Endothermic (ΔH > 0)
Position on reaction coordinate	Close to reactants (small α)	Close to products (large α)
Bond making/breaking	Barely begun	Nearly complete
Geometry resembles	The reactants	The products / intermediate
Sensitivity to product stability	Low — poor selectivity	High — strong selectivity
Typical rate	Fast (low barrier)	Slow (high barrier)
Bell-Evans-Polanyi α	≈ 0-0.3	≈ 0.7-1.0
Canonical example	Cl• hydrogen abstraction	Br• hydrogen abstraction; SN1 ionization

The quantitative cousins: Bell-Evans-Polanyi and LFERs

Hammond's postulate is qualitative, but it has a quantitative shadow. The Bell-Evans-Polanyi principle states that within a family of related reactions the activation energy varies linearly with the reaction enthalpy:

    Ea  =  E₀  +  α · ΔH        (0 ≤ α ≤ 1)

The slope α is precisely the "lateness" of the transition state. When α is near 0 the transition state is reactant-like (early) and the barrier barely responds to ΔH; when α is near 1 the transition state is product-like (late) and the barrier tracks ΔH almost one-for-one. So a more exothermic reaction (more negative ΔH) has a lower barrier — a faster reaction — and the effect is strongest for late transition states.

The experimental expressions of this same α are the linear free-energy relationships: the Brønsted catalysis law (slope β) and the Hammett equation (slope ρ). A Brønsted α near 1 signals a late, product-like proton-transfer transition state; a Hammett ρ tells you how much charge has built up on the ring at the transition state. All of these are ways of measuring what Hammond described structurally.

Limitations, breakdowns, and anti-Hammond effects

• It applies only to a single elementary step. Never invoke Hammond across a whole multistep mechanism — apply it step by step, and only to the rate-determining one for kinetics.
• One-dimensional assumption. The postulate treats the reaction coordinate as a single variable. When two things change independently — bond formation and proton transfer in an E2, for instance — a one-dimensional picture fails and you need a two-dimensional More O'Ferrall-Jencks diagram. There, a perturbation can move the transition state perpendicular to the reaction coordinate, producing an "anti-Hammond" shift that seems to violate the simple rule.
• Perpendicular (anti-Hammond) effects. Stabilizing a corner of the More O'Ferrall-Jencks diagram pulls the transition state toward that corner along the perpendicular axis and away along the parallel axis. The net structural change can be the opposite of what the naive Hammond argument predicts.
• Nearly thermoneutral steps. When ΔH ≈ 0, "closest in energy" is ambiguous and the transition state is genuinely intermediate; the postulate offers little predictive traction.
• It's a postulate, not a theorem. It has no proof — it rests on the empirical smoothness of potential energy surfaces. It usually works because real surfaces are smooth, but it is a heuristic, not a law.
• Don't confuse it with the Curtin-Hammett principle. Curtin-Hammett governs product ratios from rapidly interconverting conformers; Hammond governs the structure of a single transition state. They are separate ideas that often appear on the same exam.

Historical note: George Hammond, 1955

George Simms Hammond published "A Correlation of Reaction Rates" in the Journal of the American Chemical Society in 1955 (J. Am. Chem. Soc. 1955, 77, 334) while at Iowa State College. He was building on ideas already present in the work of John E. Leffler, and the relationship is sometimes called the Leffler-Hammond postulate in recognition. Leffler's 1953 paper had proposed that the transition state's character interpolates between reactant and product in proportion to the reaction's energetics; Hammond sharpened this into the structural statement chemists now use.

Hammond went on to a distinguished career in photochemistry (the Hammond mechanism of triplet sensitization bears his name) and became a leading figure in American chemistry, but the postulate from this single early paper remains his most widely taught contribution — a fixture of every sophomore organic course and every graduate physical-organic sequence.