Catalysis: Homogeneous and Heterogeneous

A catalytic converter turns toxic engine exhaust into carbon dioxide, water, and nitrogen in milliseconds. The gas stream barely slows down as it passes through a honeycomb coated with platinum and rhodium. Nothing about the platinum is consumed, yet without it the same reactions would take years at room temperature. Catalysis is the single biggest lever chemists have over reaction rates. Roughly 90% of industrial chemical processes rely on a catalyst, and biology runs almost entirely on catalyzed reactions. Understanding how a catalyst works means understanding kinetics, mechanism, and thermodynamics all at once. This article builds the kinetic logic of catalysis from first principles. You will see how a catalyst opens an alternative mechanistic pathway. You will see why it cannot alter the equilibrium constant. Then you will meet the two great catalytic families, homogeneous and heterogeneous, along with the turnover concepts used to compare them.

Key takeaways

A catalyst provides an alternative reaction pathway with a lower activation energy, so it speeds up both the forward and reverse reactions equally.
Because the catalyst appears in the rate law but not the overall stoichiometric equation, it leaves $\Delta G^\circ$ and $K_{eq}$ completely unchanged.
Homogeneous catalysts share the same phase as reactants, giving high selectivity but hard separation; heterogeneous catalysts sit on a separate surface, giving easy separation but relying on adsorption kinetics.
Turnover number (TON) and turnover frequency (TOF) quantify catalyst productivity and efficiency, independent of concentration units.
Catalytic cycles are built from ordinary elementary steps, so the steady-state and rate-determining-step tools from mechanism analysis apply directly.
Key idea: a catalyst accelerates a reaction by opening a lower-energy mechanistic detour that regenerates itself at the end of every cycle, leaving thermodynamics untouched.

The core insight

A catalyst does not change how favorable a reaction is; it changes how the reaction gets there, replacing one high-barrier elementary step with a sequence of lower-barrier steps that consume and regenerate the catalyst in a closed cycle.

The physical picture

Every reaction has to pass through a transition state, a fleeting, high-energy arrangement of atoms sitting atop the energy barrier between reactants and products. The height of that barrier, $E_a$ , sets the rate through the Arrhenius factor $e^{-E_a/RT}$ . A catalyst cannot lower the energy of the reactant or product states — if it did, it would change $\Delta G^\circ$ for the reaction, which is thermodynamically forbidden for a species that is not consumed overall. Instead, a catalyst offers a detour: a multistep mechanism in which the catalyst binds a reactant, some bond reorganization happens in a more stable intermediate, and the catalyst is released unchanged at the end. Because the detour is built from several smaller steps, each individual step can have a much smaller barrier than the single uncatalyzed step it replaces — even though the net path from reactants to products still connects the same start and end points on the energy landscape. This is exactly analogous to the competing-pathway ideas from parallel reactions: the uncatalyzed and catalyzed routes run side by side, and whichever has the larger rate constant dominates the observed kinetics. Catalysts split into two operational classes based on phase. A homogeneous catalyst is dissolved in the same phase as the reactants — typically a transition-metal complex or an acid/base in solution. Every catalyst molecule is, in principle, equally accessible, so homogeneous catalysis tends toward high selectivity and well-defined mechanisms that can be studied by ordinary solution kinetics. A heterogeneous catalyst exists in a different phase, almost always a solid surface exposed to gas- or liquid-phase reactants. Reaction occurs only at surface sites, so the kinetics are governed by adsorption, surface diffusion, and desorption — a preview of surface catalysis and adsorption kinetics. Heterogeneous catalysts are prized industrially because filtering a solid out of a product stream is trivial compared with distilling a dissolved metal complex away from product.

Figure

Energy profile: uncatalyzed versus catalyzed pathway for A + B → C

Figure

Homogeneous vs heterogeneous catalytic cycle

The core relations

k_{obs} = k_{uncat} + k_{cat}[\text{Cat}]

where: the observed rate constant is a sum over parallel catalyzed and uncatalyzed pathways.

K_{eq} = \frac{k_f}{k_r} \quad \text{(unchanged by any catalyst)}

where: $k_f$ and $k_r$ are both accelerated by the same factor, so their ratio, and thus $\Delta G^\circ = -RT\ln K_{eq}$ , is untouched.

k = A\,e^{-E_a/RT}

where: $A$ is the pre-exponential factor and $E_a$ is the activation energy of the rate-determining step of whichever pathway dominates.

\text{TON} = \frac{\text{moles product formed}}{\text{moles catalyst}}, \qquad \text{TOF} = \frac{\text{TON}}{t} = \frac{1}{[\text{Cat}]}\frac{d[P]}{dt}

where: turnover number counts total productivity per active site over the catalyst's lifetime; turnover frequency is the rate per active site, usually reported per second or per hour.

r = \frac{k_{cat}[\text{Cat}]_0[S]}{K_M + [S]}

where: a Michaelis–Menten-type rate law, common to enzyme and many surface-bound catalytic cycles, foreshadowed here and derived in full for enzymes.

Worked examples

Derivation AWhy a catalyst cannot change Keq: a two-step cycle analysis

Goal. Show algebraically that a catalytic cycle, built from elementary steps that satisfy microscopic reversibility, reproduces the same equilibrium constant as the uncatalyzed reaction — only the path to get there differs.

Step 1. Write the uncatalyzed reaction and its equilibrium constant.

A \rightleftharpoons P, \qquad K_{eq} = \frac{k_1}{k_{-1}}

This is fixed by the standard Gibbs energies of A and P alone; it has nothing to do with mechanism.

Step 2. Write the catalyzed cycle as two elementary equilibria.

A + \text{Cat} \underset{k_{-a}}{\overset{k_a}{\rightleftharpoons}} \text{Cat}\,\cdot\,\text{A}, \qquad \text{Cat}\,\cdot\,\text{A} \underset{k_{-b}}{\overset{k_b}{\rightleftharpoons}} \text{Cat} + P

The catalyst binds A, the bound complex rearranges to bound P, and P is released, regenerating free catalyst.

Step 3. Multiply the two step equilibrium constants.

K_a = \frac{k_a}{k_{-a}}, \qquad K_b = \frac{k_b}{k_{-b}}

The overall process $A \rightleftharpoons P$ is the sum of these two steps, so its equilibrium constant is the product:

K_{overall} = K_a K_b = \frac{k_a k_b}{k_{-a} k_{-b}}

Step 4. Compare with the direct path. Because A and P are the same molecules with the same standard Gibbs energies regardless of which route connects them, thermodynamics requires

K_{overall} = K_{eq} = \frac{[P]_{eq}}{[A]_{eq}}

The catalyst can make $k_a, k_{-a}, k_b, k_{-b}$ all much larger than $k_1, k_{-1}$ — that is exactly what speeds up the reaction — but their ratio is pinned by the same $\Delta G^\circ$ .

Boxed result.

\boxed{K_{eq}^{catalyzed} = K_{eq}^{uncatalyzed}; \text{ only the rate to reach equilibrium changes, never the equilibrium itself}}

Worked examples

Example 1 — Foundational: comparing catalyzed and uncatalyzed rate constants. The decomposition of hydrogen peroxide, $2\text{H}_2\text{O}_2 \rightarrow 2\text{H}_2\text{O} + \text{O}_2$ , has an uncatalyzed activation energy of $E_a = 76$ kJ/mol. Adding aqueous $\text{I}^-$ as a homogeneous catalyst lowers this to $E_a = 57$ kJ/mol via an iodide-mediated two-step mechanism. Assume both pathways share the same pre-exponential factor, $A = 1.0\times10^{13}\ \text{s}^{-1}$ , and compare rate constants at $T = 298$ K. Step 1 — uncatalyzed rate constant.

k_{uncat} = A e^{-E_a/RT} = (1.0\times10^{13})\exp\left(\frac{-76000}{8.314 \times 298}\right)

The exponent is

\frac{-76000}{8.314\times298} = \frac{-76000}{2477.6} = -30.68

k_{uncat} = (1.0\times10^{13})(4.75\times10^{-14}\ \text{s}) = 0.475\ \text{s}^{-1}

Step 2 — catalyzed rate constant.

\frac{-57000}{2477.6} = -23.01

k_{cat} = (1.0\times10^{13})e^{-23.01} = (1.0\times10^{13})(1.01\times10^{-10}) = 1.01\times10^{3}\ \text{s}^{-1}

Step 3 — ratio.

\frac{k_{cat}}{k_{uncat}} = \frac{1.01\times10^{3}}{0.475} \approx 2.1\times10^{3}

A drop of only 19 kJ/mol in activation energy — about the strength of a single hydrogen bond — speeds the reaction up roughly 2100-fold at room temperature. This is the leverage of the exponential in the Arrhenius law: modest changes in $E_a$ produce enormous changes in rate. Sanity check: $19\ \text{kJ/mol}/(RT) \approx 7.67$ , and $e^{7.67}\approx 2.1\times10^3$ , consistent with the ratio above.

Example 2 — Applied, exam-level: turnover frequency and catalyst loading in an industrial hydrogenation. A heterogeneous Pd/C catalyst hydrogenates an alkene in a batch reactor. The reactor holds 2.50 L of solution containing 0.400 mol of alkene substrate. The catalyst loading is 0.0150 mol of surface Pd sites (from chemisorption titration). After 12.0 minutes, gas uptake measurements show 0.360 mol of $\text{H}_2$ consumed, corresponding to 0.360 mol of alkene converted (1:1 stoichiometry). Step 1 — turnover number so far.

\text{TON} = \frac{\text{mol product}}{\text{mol catalyst}} = \frac{0.360\ \text{mol}}{0.0150\ \text{mol}} = 24.0

Each active Pd site has, on average, turned over 24 substrate molecules in 12 minutes. Step 2 — turnover frequency.

t = 12.0\ \text{min} = 720\ \text{s}

\text{TOF} = \frac{\text{TON}}{t} = \frac{24.0}{720\ \text{s}} = 3.33\times10^{-2}\ \text{s}^{-1}

Equivalently, TOF $= 120\ \text{h}^{-1}$ , meaning each site processes about 120 substrate molecules per hour — a typical value for a good heterogeneous hydrogenation catalyst (industrial benchmarks span roughly 10–10,000 h⁻¹ depending on the system). Step 3 — how much catalyst is needed to finish the job in 2.00 hours instead? Assuming the same TOF holds (pseudo-zero-order in this concentration regime), the remaining 0.040 mol of alkene must convert in $2.00\ \text{h} = 7200\ \text{s}$ :

n_{Cat} = \frac{n_{product}}{\text{TOF}\times t} = \frac{0.040\ \text{mol}}{(3.33\times10^{-2}\ \text{s}^{-1})(7200\ \text{s})} = \frac{0.040}{240} = 1.7\times10^{-4}\ \text{mol}

Only a small additional charge of catalyst is needed because the time window is generous; this illustrates how TOF lets an engineer trade catalyst loading against reaction time. Sanity check: $1.7\times10^{-4}$ mol is far less than the original 0.0150 mol charge, consistent with slowing down and using less catalyst over a longer batch time.

Limiting cases & common traps

Situation	What actually happens	Common error
Catalyst amount doubled	Rate roughly doubles (first order in catalyst); $K_{eq}$ unchanged	Assuming yield or equilibrium position improves
Very high substrate concentration (surface or enzyme-like cycle)	Rate saturates at $k_{cat}[\text{Cat}]_0$ , becomes zero order in substrate	Expecting rate to keep rising linearly forever
Catalyst poisoning	Strongly-binding impurity occupies active sites irreversibly, TOF collapses	Confusing poisoning with simple dilution of catalyst
Comparing TOF across different catalysts	Only valid if "active sites" are counted consistently (e.g., by chemisorption)	Comparing TOF using total metal loading instead of accessible surface sites
Homogeneous vs heterogeneous choice	Homogeneous: high selectivity, hard separation. Heterogeneous: robust, easy separation, harder mechanistic control	Assuming one type is universally "better"

The chemist behind it

Paul Sabatier's systematic studies of metal-catalyzed hydrogenation, especially his discovery that finely divided nickel catalyzes the hydrogenation of unsaturated hydrocarbons, established heterogeneous catalysis as a controllable, mechanistic science rather than an empirical curiosity. His work directly enabled the industrial hydrogenation of vegetable oils and laid groundwork for the surface-binding concepts — the "Sabatier principle," that an optimal catalyst binds reactants neither too strongly nor too weakly — that still guide catalyst design today.

Origins · 1 minPaul Sabatier and how this was discovered

Paul Sabatier

1854–1941 · French chemist who shared the 1912 Nobel Prize in Chemistry for pioneering metal-catalyzed hydrogenation

Ask the tutor: How does the Michaelis–Menten equation emerge as the steady-state limit of a two-step enzyme-substrate binding cycle?

Does a catalyst change the reaction order?

Sometimes, yes. The uncatalyzed and catalyzed pathways are genuinely different mechanisms, so they can have different rate laws entirely. A gas-phase reaction that is second order overall might become pseudo-first-order in substrate once a metal surface or enzyme active site saturates it, because the rate-determining step has shifted to a different elementary process. The underlying principle from mechanism analysis still applies: the observed rate law reflects whichever step is rate-determining in the operating pathway, catalyzed or not.

Can a catalyst make an unfavorable reaction happen?

No. A catalyst cannot shift $K_{eq}$ , so it cannot make a thermodynamically unfavorable reaction ( $\Delta G^\circ > 0$ ) proceed to useful yield at equilibrium. What it can do is dramatically speed up the approach to whatever equilibrium already exists, or, in a flow reactor, allow continuous removal of product so the reaction is pulled forward by Le Chatelier's principle — but that trick works with or without a catalyst.

Why do heterogeneous catalysts often use expensive metals like platinum in tiny amounts?

Only surface atoms participate in catalysis; atoms buried inside a bulk metal particle are wasted. Catalyst manufacturers disperse platinum or palladium as nanoparticles on a high-surface-area support like alumina or carbon, maximizing the fraction of atoms exposed at the surface. This is why turnover frequency, calculated per accessible surface site rather than per gram of metal, is the fair way to compare catalyst efficiency.

How does this connect to enzyme kinetics?

Enzymes are essentially homogeneous (or membrane-bound) catalysts obeying the same core rule: bind substrate, lower the barrier to a transition state, release product, regenerate the free active site. The saturation rate law introduced above, with $K_M$ playing the role of a binding constant, is derived rigorously as the Michaelis–Menten equation, where $k_{cat}$ is precisely a turnover frequency at saturating substrate.

Why does an uncatalyzed side reaction sometimes still compete even with a good catalyst present?

Because the observed rate constant is a sum over parallel pathways, exactly as in competing reactions. If the uncatalyzed pathway has a non-negligible rate constant and the catalyst concentration is low, both channels contribute measurably to product formation, sometimes giving different products or stereochemistry. Selectivity requires the catalyzed channel's effective rate, $k_{cat}[\text{Cat}]$ , to dominate $k_{uncat}$ by orders of magnitude.

Catalysis: Homogeneous and Heterogeneous

The physical picture

The core relations

Worked examples

Worked examples

Limiting cases & common traps

The chemist behind it

Before this

Where this leads