Relaxation Methods and Fast Reactions

Some reactions finish before you can mix the reagents. Proton transfers, enzyme conformational changes, and radical recombinations can occur in microseconds. A hand-driven mixing experiment simply cannot resolve them. Relaxation methods solve this by never mixing anything in real time at all. Instead, a system already at equilibrium gets kicked by a sudden change in temperature, pressure, or electric field. Chemists then watch it relax back, and that relaxation time encodes both rate constants at once. This article builds the theory of the temperature jump from first principles. You will see how a perturbed equilibrium decays exponentially. You will learn why the relaxation time depends on both forward and reverse rate constants together. Then you will see how stopped-flow complements temperature-jump for slower reactions, with two full numeric examples.

Key takeaways

Relaxation methods perturb a system already at equilibrium, rather than mixing reactants from scratch.
Temperature-jump (T-jump) uses a rapid heating pulse, typically completed in microseconds, to shift $K$ suddenly.
Stopped-flow rapidly mixes two solutions and monitors the approach to a new equilibrium or completion, resolving milliseconds.
Near equilibrium, any reversible reaction relaxes exponentially with a single relaxation time $\tau$ .
The reciprocal relaxation time is a simple sum of forward and reverse rate constants, weighted by concentration.
Key idea: Linearizing kinetics around equilibrium turns any reversible reaction into a first-order decay, so $1/\tau$ delivers $k_f$ and $k_r$ separately, even for reactions too fast to mix.

The core insight

Near equilibrium, a small displacement in concentration obeys a linear, first-order differential equation regardless of the underlying reaction order, so the system relaxes exponentially with a single time constant $\tau$ whose reciprocal is a specific algebraic combination of the forward and reverse rate constants and the equilibrium concentrations — measure $\tau$ at several equilibrium concentrations and both rate constants fall out.

The physical picture

Consider a simple reversible reaction,

A \underset{k_r}{\overset{k_f}{\rightleftharpoons}} B

sitting quietly at equilibrium. Concentrations are fixed at $[A]_{eq}$ and $[B]_{eq}$ , with $k_f[A]_{eq} = k_r[B]_{eq}$ . Nothing observable happens — forward and reverse fluxes cancel exactly. Now perturb it. In temperature-jump (T-jump), a capacitor discharges a few-kilovolt pulse through the sample, or a laser pulse dumps energy into the solvent, raising the temperature by 5–10 K in a few microseconds. Because $K = k_f/k_r$ depends on temperature through the van't Hoff relation, the equilibrium position itself shifts almost instantaneously — the jump in $T$ is far faster than any subsequent chemistry. The system finds itself with the old concentrations but the new equilibrium constant. It is no longer at equilibrium, and it relaxes toward the new one. That relaxation is what gets recorded, usually by absorbance or fluorescence through a fast photomultiplier, with the electronics and optical path together defining a "dead time" — the shortest interval after the jump during which data is trustworthy. Modern T-jump instruments have dead times of about 1 microsecond, enabling access to reactions with half-lives of microseconds to milliseconds, far beyond manual mixing. Stopped-flow](/wiki/Stopped-Flow) takes a different route to the same territory. Two syringes drive reactant solutions through a rapid mixer into an observation cell, then a stopping syringe halts the flow almost instantly, triggering the detector. Because mixing itself limits the time resolution, stopped-flow typically resolves reactions with half-lives down to about 1 millisecond — three orders of magnitude slower than T-jump, but perfectly matched to enzyme kinetics, ligand binding, and many organic mechanisms. The two techniques are complementary: T-jump for perturbations of an existing equilibrium, stopped-flow for driving a fresh mixture toward equilibrium or completion.

Figure

Temperature-jump concept: a sudden T rise shifts K, and the system relaxes exponentially back to a new equilibrium

The core relations

A \underset{k_r}{\overset{k_f}{\rightleftharpoons}} B, \qquad K = \frac{k_f}{k_r} = \frac{[B]_{eq}}{[A]_{eq}}

where: $K$ is the equilibrium constant at the (new) temperature, related to $k_f$ and $k_r$ by microscopic reversibility.

\frac{1}{\tau} = k_f + k_r

where: $\tau$ is the relaxation time for a unimolecular (or pseudo-first-order) reversible reaction; note it is a sum, not a difference.

\frac{1}{\tau} = k_f\big([A]_{eq}+[B]_{eq}\big) + k_r

where: this is the general form for $A + B \rightleftharpoons C$ when species concentrations differ; the exact expression depends on stoichiometry (derived below for the bimolecular case).

\ln K = -\frac{\Delta H^\circ}{R}\left(\frac{1}{T}\right) + \frac{\Delta S^\circ}{R}

where: the van't Hoff equation gives the size of the equilibrium shift, $\Delta \ln K$ , produced by a temperature jump $\Delta T$ , via

\Delta \ln K \approx \frac{\Delta H^\circ}{RT^2}\,\Delta T .

Worked examples

Derivation ARelaxation time for a reversible first-order reaction

Goal. Show that a small displacement from equilibrium in $A \rightleftharpoons B$ decays exponentially, and find $\tau$ in terms of $k_f$ and $k_r$ .

Step 1. Write the full rate law. Let $x(t)$ be the instantaneous deviation of $[A]$ from its new equilibrium value $[A]_{eq}$ , so

[A](t) = [A]_{eq} + x(t), \qquad [B](t) = [B]_{eq} - x(t)

since total concentration $[A]+[B]$ is conserved. The net rate of loss of $A$ is

-\frac{d[A]}{dt} = k_f[A] - k_r[B].

Step 2. Substitute the displacement variable.

-\frac{dx}{dt} = k_f\big([A]_{eq}+x\big) - k_r\big([B]_{eq}-x\big)

-\frac{dx}{dt} = \big(k_f[A]_{eq} - k_r[B]_{eq}\big) + \big(k_f + k_r\big)x

Step 3. Use the equilibrium condition. At the new equilibrium, $k_f[A]_{eq} = k_r[B]_{eq}$ by definition, so the constant term vanishes exactly:

-\frac{dx}{dt} = (k_f+k_r)\,x

Step 4. Recognize first-order decay. This is a linear, homogeneous, first-order ODE regardless of the true molecularity of the forward and reverse steps — linearization near equilibrium always produces this form for small enough $x$ . Its solution is

x(t) = x(0)\, e^{-t/\tau}, \qquad \frac{1}{\tau} = k_f + k_r

Step 5. Interpret. The relaxation time depends on the sum of rate constants, never their difference or ratio alone. A single $\tau$ measurement cannot separate $k_f$ from $k_r$ , but combining it with the independently known equilibrium constant $K = k_f/k_r$ gives two equations in two unknowns.

\boxed{\dfrac{1}{\tau} = k_f + k_r, \qquad K = \dfrac{k_f}{k_r} \ \Rightarrow\ k_f = \dfrac{K}{1+K}\cdot\dfrac{1}{\tau}, \quad k_r = \dfrac{1}{1+K}\cdot\dfrac{1}{\tau}}

:::svg[Bimolecular relaxation: 1/τ plotted against (A]eq+[B]eq) gives slope k_f and intercept k_r]

Worked examples

Example 1 — Foundational: extracting $k_f$ and $k_r$ from a single T-jump trace. A weak acid HA undergoes a proton-transfer reaction monitored by a T-jump instrument, treated as

\text{HA} \underset{k_r}{\overset{k_f}{\rightleftharpoons}} \text{H}^+ + \text{A}^-.

Suppose (as an idealized unimolecular-style relaxation, valid at fixed pH buffering so pseudo-first-order conditions hold) the absorbance signal after the jump decays with a measured relaxation time

\tau = 8.0\ \mu\text{s} = 8.0\times10^{-6}\ \text{s}.

Independently, the equilibrium constant at the jumped temperature (298 K, from a separate pH titration) is

K = 1.8\times10^{-5}.

Step 1: total rate constant sum.

\frac{1}{\tau} = k_f + k_r = \frac{1}{8.0\times10^{-6}\ \text{s}} = 1.25\times10^{5}\ \text{s}^{-1}

Step 2: since $K = k_f/k_r \ll 1$ , the reverse (recombination) step dominates the sum, so $k_r \approx 1/\tau$ to a good first approximation. Solve exactly using the boxed relations:

k_r = \frac{1}{1+K}\cdot\frac{1}{\tau} = \frac{1}{1.0000180}\times(1.25\times10^{5}\ \text{s}^{-1}) \approx 1.25\times10^{5}\ \text{s}^{-1}

k_f = \frac{K}{1+K}\cdot\frac{1}{\tau} = (1.8\times10^{-5})(1.25\times10^{5}\ \text{s}^{-1}) \approx 2.25\ \text{s}^{-1}

Sanity check: $k_f/k_r = 2.25/(1.25\times10^5) = 1.8\times10^{-5} = K$ . ✓ The recombination rate constant, $1.25\times10^5\,\text{s}^{-1}$ , is far below the diffusion limit (~ $10^{10}\ \text{M}^{-1}\text{s}^{-1}$ for a bimolecular process, but here we are quoting a pseudo-first-order value under the given buffering), consistent with a modest proton-transfer barrier.

Example 2 — Applied: designing a T-jump versus stopped-flow choice, and using van't Hoff to size the jump. A biochemist studies a protein–ligand binding step,

\text{P} + \text{L} \underset{k_r}{\overset{k_f}{\rightleftharpoons}} \text{PL},

with $\Delta H^\circ = -40\ \text{kJ mol}^{-1}$ for binding. At $T = 298\ \text{K}$ , $K = 2.0\times10^{6}\ \text{M}^{-1}$ . Part (a): how much does a 6 K jump shift $K$ ? Using the linearized van't Hoff form,

\Delta \ln K \approx \frac{\Delta H^\circ}{RT^2}\,\Delta T

\Delta \ln K \approx \frac{(-40{,}000\ \text{J mol}^{-1})}{(8.314\ \text{J K}^{-1}\text{mol}^{-1})(298\ \text{K})^2}\times 6\ \text{K}

\Delta \ln K \approx \frac{-40{,}000}{738{,}600}\times 6 \approx (-0.05416)\times 6 = -0.325

So $K$ drops by a factor of $e^{-0.325} = 0.72$ , i.e., about 28% lower after the jump — comfortably within the linear-response regime needed for the relaxation-time formula to hold (small perturbation). Part (b): choosing the technique and interpreting $\tau$ . For a bimolecular association, the relevant relaxation relation (derived analogously to the unimolecular case, keeping first-order terms in the displacement) is

\frac{1}{\tau} = k_f\big([P]_{eq}+[L]_{eq}\big) + k_r.

Suppose the experiment is run at total concentrations giving $[P]_{eq}+[L]_{eq} = 1.0\times10^{-5}\ \text{M}$ , and the measured relaxation time is $\tau = 25\ \mu\text{s}$ . Then

\frac{1}{\tau} = \frac{1}{2.5\times10^{-5}\ \text{s}} = 4.0\times10^{4}\ \text{s}^{-1}

Repeating at a second concentration, $[P]_{eq}+[L]_{eq} = 5.0\times10^{-5}\ \text{M}$ , gives $\tau' = 6.7\ \mu\text{s}$ , so

\frac{1}{\tau'} = 1.49\times10^{5}\ \text{s}^{-1}

The slope of $1/\tau$ versus $([P]_{eq}+[L]_{eq})$ gives $k_f$ :

k_f = \frac{(1.49\times10^{5}-4.0\times10^{4})\ \text{s}^{-1}}{(5.0-1.0)\times10^{-5}\ \text{M}} = \frac{1.09\times10^{5}\ \text{s}^{-1}}{4.0\times10^{-5}\ \text{M}} \approx 2.7\times10^{9}\ \text{M}^{-1}\text{s}^{-1}

and the intercept gives $k_r$ :

k_r = 4.0\times10^{4}\ \text{s}^{-1} - (2.7\times10^{9}\ \text{M}^{-1}\text{s}^{-1})(1.0\times10^{-5}\ \text{M}) \approx 1.3\times10^{4}\ \text{s}^{-1}

Check: $K = k_f/k_r = 2.7\times10^9/1.3\times10^4 \approx 2.1\times10^5\ \text{M}^{-1}$ , reasonably close to the given order of magnitude given rounding — consistent with tight, near-diffusion-limited binding. Because $\tau$ here is tens of microseconds, only T-jump (dead time ~1 µs) resolves it; stopped-flow's ~1 ms dead time would completely miss this relaxation.

Limiting cases & common traps

Situation	What happens	Common trap
$K \gg 1$ (forward dominant)	$1/\tau \approx k_f(\text{concn})$ , reverse term negligible	Forgetting the small $k_r$ still matters for the intercept in a concentration series
$K \ll 1$ (reverse dominant)	$1/\tau \approx k_r$	Assuming $\tau$ alone gives $k_f$ directly — it never does without $K$
Very fast reaction, $\tau <$ instrument dead time	Relaxation invisible; only endpoints seen	Concluding "reaction too slow to matter" from a flat trace
Multiple coupled equilibria	Two or more relaxation times $\tau_1, \tau_2$ appear	Fitting a single exponential to a genuinely biexponential trace
Small $\Delta T$ or $\Delta H^\circ \approx 0$	Negligible equilibrium shift, weak/no signal	Blaming the detector when the real problem is a small $\Delta H^\circ$

The chemist behind it

Relaxation kinetics owes its existence to Manfred Eigen, who in the 1950s at the Max Planck Institute for Biophysical Chemistry realized that instead of trying to mix faster, one could perturb an equilibrium already established and watch it snap back. He developed temperature-jump, pressure-jump, and electric-field-jump methods, extending measurable rate constants into the microsecond and even nanosecond range — reactions like proton transfer in water, previously thought immeasurably fast, suddenly became textbook numbers. His work opened quantitative kinetics of enzyme catalysis, nucleic acid base-pairing, and diffusion-limited reactions to direct rate measurement.

Origins · 1 minManfred Eigen and how this was discovered

Manfred Eigen

1927–2019 · Shared the 1967 Nobel Prize in Chemistry for developing relaxation methods to study reactions too fast for conventional mixing

Ask the tutor: How would the relaxation analysis change for a reaction network with three coupled equilibria showing two distinct relaxation times?

How is a temperature jump actually produced experimentally

Most classic instruments discharge a charged capacitor (a few kilovolts) through a conducting sample cell, resistively heating the solution by 5–10 K within 1–10 microseconds. Modern versions use pulsed infrared lasers tuned to a water absorption band, heating the solvent almost instantaneously and giving sub-microsecond jumps. Either way, the heating step must be much faster than the chemical relaxation being studied, otherwise the "jump" itself smears out the kinetics you are trying to resolve.

Why does relaxation always look like a single exponential even for reactions that aren't first order

Because relaxation kinetics deliberately looks only at small displacements from equilibrium. Any smooth rate law, expanded in a Taylor series around the equilibrium concentrations, reduces to a linear term in the displacement $x$ once you keep only first order in $x$ . Linear differential equations of this form always integrate to single exponentials. Far from equilibrium the full nonlinear rate law would apply, but relaxation methods are designed to stay in the small-perturbation regime.

Can relaxation methods handle reactions with more than two species or coupled equilibria

Yes, but the mathematics generalizes to a set of coupled linear equations, and diagonalizing that system produces multiple relaxation times rather than one, each a specific combination of all the rate constants in the network. A three-species coupled equilibrium typically shows two resolvable relaxation times, $\tau_1$ and $\tau_2$ , and disentangling all rate constants requires measuring both across several concentration conditions. This is standard practice in complex enzyme mechanisms with multiple intermediates.

How does relaxation kinetics relate to the Arrhenius equation and activation energy

Once $k_f$ and $k_r$ are extracted at several temperatures using the T-jump method, each obeys the Arrhenius equation independently. Plotting $\ln k_f$ and $\ln k_r$ against $1/T$ gives two activation energies, and their difference must equal $\Delta H^\circ$ for the reaction, providing an internal consistency check. This connects fast-reaction kinetics directly back to the thermodynamic and Arrhenius framework covered for ordinary reversible reactions.

What is the fastest reaction rate constant ever measured by relaxation methods

Proton transfer between $\text{H}_3\text{O}^+$ and $\text{OH}^-$ in water is essentially diffusion-limited, with $k_r$ (recombination) around $1.4\times10^{11}\ \text{M}^{-1}\text{s}^{-1}$ , among the fastest bimolecular rate constants ever measured in solution. Eigen's original T-jump studies on water autoionization in the late 1950s were what first demonstrated the technique's power, resolving relaxation times on the order of tens of microseconds for a process previously considered immeasurable by any mixing-based method.

Relaxation Methods and Fast Reactions

The physical picture

The core relations

Worked examples

Worked examples

Limiting cases & common traps

The chemist behind it

Before this