Physisorption versus Chemisorption

Sprinkle activated charcoal into a gas jar of bromine vapor. The color fades within seconds. The gas molecules stick to the carbon surface. Weak, fast, reversible forces hold them there. This is physisorption. If instead you flow hydrogen gas over a hot nickel surface, the H₂ molecule splits apart. It forms genuine Ni–H chemical bonds. This is a slower and often irreversible process called chemisorption. Distinguishing these two modes of surface binding matters. Catalysis, gas storage, sensing, and even chromatography all depend on which one dominates. Physisorption controls surface area measurements and gas capture. Chemisorption is the essential first step of most industrial catalytic reactions. Examples range from ammonia synthesis to catalytic converters. This article contrasts the two processes through enthalpy, reversibility, specificity, and coverage limits. You will see the physical picture first. Then the thermodynamic and kinetic distinctions follow. Finally, two worked numerical examples appear. They show how to identify which mechanism operates from experimental data.

Key takeaways

Physisorption arises from van der Waals forces; chemisorption involves electron sharing or transfer forming true chemical bonds.
Enthalpies of physisorption are small, typically 5–40 kJ/mol; chemisorption enthalpies are large, typically 40–800 kJ/mol.
Physisorption is reversible and can form multiple layers; chemisorption is often irreversible and limited to one monolayer.
Physisorption is non-specific and occurs on any surface below the gas's critical temperature; chemisorption is specific to particular adsorbate–surface chemical pairs.
Chemisorption usually requires an activation energy and shows strong temperature dependence, unlike most physisorption.
Key idea: The nature of the adsorbate–surface bond, van der Waals versus covalent/ionic, dictates every measurable difference between physisorption and chemisorption.

The core insight

Physisorption and chemisorption sit at opposite ends of the same phenomenon, molecules accumulating at a surface, but they are governed by fundamentally different forces: weak, additive dispersion and dipole interactions for physisorption versus true orbital overlap and bond formation for chemisorption, and this single distinction in bonding character propagates into every measurable contrast in enthalpy, reversibility, coverage limit, and chemical specificity.

The physical picture

When a gas molecule approaches a solid surface, it experiences an attractive potential well from van der Waals forces (induced-dipole/induced-dipole, dipole/induced-dipole, and dispersion interactions) present between any two bodies at close range. If nothing else happens, the molecule simply settles into this shallow well, weakly bound but chemically unchanged. Because the well is shallow, only a small amount of thermal energy is needed to knock the molecule back off. Molecules can stack on top of each other because the second layer feels the same weak forces from the first layer that the first layer felt from the solid. Multilayer buildup, similar to condensation, is a hallmark of physisorption. Chemisorption is a different story. As the molecule approaches, its orbitals begin to overlap with surface atom orbitals. A new chemical bond forms, often after the incoming molecule dissociates (as H₂ does on many transition metals). This bond is comparable in strength to a bond formed in an ordinary chemical reaction. Because the bond requires specific orbital symmetry and electron count matching, only certain adsorbate–surface combinations work efficiently, and once every accessible surface atom has formed a bond, no more chemisorption can occur: the layer is complete at one molecule (or fragment) thick. The energy landscape often contains both wells in sequence. A gas molecule first falls into the shallow physisorption well (weak precursor state), then, if it has enough energy to cross an activation barrier, drops into the much deeper chemisorption well. This two-step picture, sometimes drawn as a Lennard-Jones-style potential energy diagram, explains why chemisorption frequently needs heating to proceed at a measurable rate even though it releases far more energy overall.

Figure

Potential energy curves for physisorption (shallow, non-dissociative) and chemisorption (deep, dissociative) as a molecule approaches a metal surface

The core relations

\Delta H_{\text{ads}}^{\text{phys}} \approx -5 \text{ to } -40\ \text{kJ mol}^{-1}

where: this range reflects van der Waals interaction strengths, comparable to enthalpies of condensation or vaporization.

\Delta H_{\text{ads}}^{\text{chem}} \approx -40 \text{ to } -800\ \text{kJ mol}^{-1}

where: this range reflects true chemical bond strengths, comparable to bond dissociation enthalpies.

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

where: $k$ is the rate of adsorption, $A$ is a pre-exponential (attempt) factor, $E_a$ is the activation energy (often near zero for physisorption, substantial for activated chemisorption), $R$ is the gas constant, $T$ is temperature.

q_{\text{st}} = -\left(\frac{\partial \ln p}{\partial (1/T)}\right)_{\theta} R

where: $q_{\text{st}}$ is the isosteric enthalpy of adsorption extracted from adsorption isotherms measured at several temperatures at fixed surface coverage $\theta$ ; this Clausius–Clapeyron-style relation is the standard experimental route to $\Delta H_{\text{ads}}$ .

Derivation AWhy chemisorption saturates at one monolayer while physisorption does not

Goal. Show, from the nature of the binding forces, why coverage limited to $\theta = 1$ (a full monolayer) is expected for chemisorption but not for physisorption.

Step 1. Define coverage. Let $\theta$ be the fraction of surface sites occupied, so $\theta = 1$ means every distinct adsorption site on the solid holds one adsorbed species.

Step 2. Consider the origin of the attractive force in each case. In chemisorption, binding requires a chemical bond between the adsorbate and a specific surface atom (or a small ensemble of them), which draws on that atom's available valence electrons. Once a surface atom has used its bonding capacity, it cannot form a second chemisorption bond to another incoming molecule; the surface atoms are the finite resource being consumed.

Step 3. Contrast with physisorption. The van der Waals attraction responsible for physisorption acts between any pair of nearby atoms or molecules, not just surface atoms and adsorbate. A molecule already adsorbed in the first layer still presents an outward-facing surface of polarizable electron density. A second gas molecule approaching that first layer experiences essentially the same category of dispersion attraction that the first molecule felt from the solid.

Step 4. Compare the energetics of stacking. Because the physisorption well depth for a second layer (adsorbate-on-adsorbate) is similar in order of magnitude to the first layer's well depth (adsorbate-on-solid), multilayer condensation becomes thermodynamically reasonable as the gas pressure approaches its saturation vapor pressure. There is no equivalent driving force for a second chemisorbed layer, because there is no unsatisfied surface valence left to bond to.

Step 5. State the resulting coverage law. For chemisorption, the maximum attainable coverage under the simplest (Langmuir) picture is

\theta_{\max} = 1

one adsorbate (or dissociated fragment) per accessible surface site, whereas physisorption coverage can exceed unity,

\theta > 1

as multiple layers accumulate, ultimately merging into bulk-like condensed adsorbate near the saturation pressure $p^{*}$ .

Boxed result.

\boxed{\theta_{\max}^{\text{chemisorption}} = 1 \quad \text{(monolayer, bond-limited)}, \qquad \theta^{\text{physisorption}} \text{ can exceed } 1 \text{ (multilayer, force-unlimited)}}

Worked examples

Example 1 — Identifying the adsorption type from enthalpy data (foundational). Nitrogen gas is adsorbed onto a silica surface at 77 K, and calorimetry gives an adsorption enthalpy of $\Delta H_{\text{ads}} = -18\ \text{kJ mol}^{-1}$ . A separate experiment adsorbs hydrogen gas onto a platinum surface at 300 K and the surface shows the same H atoms remain bound even after evacuating the chamber to high vacuum, with a measured $\Delta H_{\text{ads}} = -130\ \text{kJ mol}^{-1}$ . Step 1 — compare enthalpy magnitudes to reference ranges. Physisorption typical range: 5 to 40 kJ/mol in magnitude. Chemisorption typical range: 40 to 800 kJ/mol in magnitude. For N₂ on silica:

|\Delta H_{\text{ads}}| = 18\ \text{kJ mol}^{-1}

This falls squarely in the physisorption range, and is close to the enthalpy of vaporization of liquid N₂ (5.6 kJ/mol at its own boiling point, scaled up somewhat by the extra solid-surface attraction), confirming a van der Waals mechanism. For H₂ on platinum:

|\Delta H_{\text{ads}}| = 130\ \text{kJ mol}^{-1}

This is far above the physisorption ceiling and consistent with a true Pt–H bond enthalpy (comparable to typical metal–hydrogen bond strengths of 100–160 kJ/mol). Step 2 — use the reversibility criterion as a cross-check. N₂ desorbs readily on gentle warming or evacuation at 77 K — reversible, consistent with physisorption. H₂ remains bound to Pt even under vacuum at 300 K, requiring much higher temperatures (or a chemical reducing/oxidizing step) to remove it — irreversible under mild conditions, consistent with chemisorption. Sanity check. Both the enthalpy magnitude and the reversibility test agree independently: N₂/silica is physisorption, H₂/Pt is chemisorption. This cross-validation is exactly the diagnostic logic used in real surface-science laboratories (temperature-programmed desorption plus calorimetry).

Example 2 — Estimating monolayer capacity and distinguishing mechanism from a pressure/temperature series (applied, exam-level). A gas is adsorbed onto a metal powder of total surface area $A = 12.0\ \text{m}^2$ , and the isosteric heat of adsorption, extracted from isotherms at $\theta = 0.5$ using the relation

q_{\text{st}} = -R\left(\frac{\partial \ln p}{\partial (1/T)}\right)_\theta

is measured from two isotherms at $T_1 = 250\ \text{K}$ and $T_2 = 350\ \text{K}$ . At fixed $\theta = 0.5$ , the equilibrium pressures are $p_1 = 2.00\ \text{kPa}$ (at 250 K) and $p_2 = 55.0\ \text{kPa}$ (at 350 K). Step 1 — apply the Clausius–Clapeyron-type isosteric equation in finite-difference form.

\ln\left(\frac{p_2}{p_1}\right) = -\frac{q_{\text{st}}}{R}\left(\frac{1}{T_2} - \frac{1}{T_1}\right)

Compute the pressure ratio:

\ln\left(\frac{55.0}{2.00}\right) = \ln(27.5) = 3.314

Compute the temperature term:

\frac{1}{T_2} - \frac{1}{T_1} = \frac{1}{350} - \frac{1}{250} = 2.857\times10^{-3} - 4.000\times10^{-3} = -1.143\times10^{-3}\ \text{K}^{-1}

Step 2 — solve for $q_{\text{st}}$ .

q_{\text{st}} = -R \cdot \frac{\ln(p_2/p_1)}{(1/T_2 - 1/T_1)} = -(8.314\ \text{J mol}^{-1}\text{K}^{-1})\cdot\frac{3.314}{-1.143\times10^{-3}\ \text{K}^{-1}}

q_{\text{st}} = -(8.314)\times(-2900\ \text{K}) = 2.41\times10^{4}\ \text{J mol}^{-1} = 24.1\ \text{kJ mol}^{-1}

Step 3 — interpret the mechanism. A magnitude of 24.1 kJ/mol lies within the physisorption range (5–40 kJ/mol) and well below typical chemisorption values, so this adsorbate is physisorbed on the metal powder — consistent with the pressure rising steeply with temperature, since physisorbed gas is easily driven off as thermal energy increases (an entropically favored desorption, matching the small binding enthalpy). Step 4 — as a follow-on calculation, estimate the physisorbed monolayer capacity if the adsorbate cross-sectional area is $\sigma = 0.162\ \text{nm}^2$ (the standard value for N₂).

N_m = \frac{A}{\sigma} = \frac{12.0\ \text{m}^2}{0.162\times10^{-18}\ \text{m}^2} = 7.41\times10^{19}\ \text{molecules}

Converting to moles:

n_m = \frac{7.41\times10^{19}}{6.022\times10^{23}\ \text{mol}^{-1}} = 1.23\times10^{-4}\ \text{mol}

Sanity check. A powder of 12 m² surface area holding roughly 0.12 mmol of monolayer gas is a physically reasonable BET-type result for a modest-surface-area catalyst support, and the low enthalpy confirms the physisorption assignment used in the monolayer capacity calculation (this is, in fact, the logic behind BET surface-area measurements using physisorbed N₂ at 77 K).

Limiting cases & common traps

Feature	Physisorption	Chemisorption
Bonding force	van der Waals (dispersion, dipole)	Covalent/ionic, orbital overlap
Typical $\Delta H_{\text{ads}}$	5–40 kJ/mol	40–800 kJ/mol
Reversibility	Highly reversible	Often irreversible without chemical intervention
Specificity	Non-specific, any gas-solid pair below $T_c$	Highly specific, particular adsorbate-surface pairs
Layers	Multilayer, $\theta$ can exceed 1	Monolayer only, $\theta_{\max} = 1$
Activation energy	Usually negligible	Often significant (activated adsorption)
Temperature dependence of rate	Rate falls as $T$ rises (less condensation)	Rate can rise with $T$ up to a point (Arrhenius-like)
Effect on adsorbate structure	Molecule unchanged	Molecule often dissociates or is structurally altered

The chemist behind it

Irving Langmuir, working at General Electric in the 1910s, was the first to systematically distinguish adsorption that saturated at a single monolayer from adsorption that did not, framing the physical picture in terms of surface sites with finite, chemically specific binding capacity. His 1918 isotherm treatment, built explicitly on the idea of a fixed number of surface sites each holding at most one adsorbed species, is the direct conceptual ancestor of the modern physisorption/chemisorption distinction and earned him the 1932 Nobel Prize in Chemistry for work on surface chemistry.

Origins · 1 minIrving Langmuir and how this was discovered

Irving Langmuir

1881–1957 · American chemist and physicist who founded modern surface science and derived the monolayer adsorption isotherm at General Electric

Ask the tutor: How does the Langmuir isotherm mathematically build on the monolayer limit of chemisorption described here?

How can I quickly tell physisorption from chemisorption in a lab experiment?

Measure the adsorption enthalpy (via calorimetry or an isosteric plot of $\ln p$ against $1/T$ at fixed coverage) and check reversibility. If $|\Delta H_{\text{ads}}|$ is under about 40 kJ/mol and the gas desorbs completely on mild evacuation or gentle warming, it is physisorption. If the enthalpy exceeds roughly 40 kJ/mol and the adsorbate resists removal under vacuum, or requires a chemical reaction to release, it is chemisorption. Temperature-programmed desorption profiles and the shape of the isotherm (monolayer plateau versus multilayer rise) provide additional confirmation.

Can a single adsorbate-surface system show both physisorption and chemisorption?

Yes. Many real systems follow a two-step path: the molecule first physisorbs into a shallow precursor well, then crosses an activation barrier into a much deeper chemisorbed state, as shown in the potential energy diagram above. Hydrogen on many transition metals behaves this way at low temperature, trapping in the physisorbed well first before slowly converting to dissociative chemisorption as the surface is warmed or given enough time.

Why does chemisorption often require heating to proceed at a useful rate, even though it releases more energy than physisorption?

Because chemisorption frequently has to cross an activation energy barrier corresponding to bond-breaking or bond-reorganization at the surface, similar to any ordinary chemical reaction. Even though the final chemisorbed state is much lower in energy (more exothermic) than the physisorbed state, the rate of getting there depends on the Arrhenius factor $\exp(-E_a/RT)$ . Raising temperature increases the fraction of molecules with enough energy to cross the barrier, so the process is called "activated adsorption."

Does chemisorption enthalpy set an upper coverage limit the same way for every surface?

The idealized limit of $\theta_{\max} = 1$ assumes one adsorption site per surface atom and one adsorbate per site, the basic Langmuir picture. Real surfaces can deviate: some molecules bridge two or three metal atoms (lowering effective site count), while some small atoms occupy interstitial sites (allowing slightly different capacities). The qualitative point still holds: chemisorption capacity is fixed by the finite number of chemically reactive surface atoms, unlike physisorption's open-ended multilayer capacity.

How does this topic connect to catalysis and the Langmuir isotherm?

Heterogeneous catalysis almost always begins with chemisorption of at least one reactant, since forming a real chemical bond to the catalyst surface is what activates the molecule for subsequent reaction steps. The mathematical description of how coverage $\theta$ depends on pressure, the Langmuir adsorption isotherm, was built directly from the monolayer, site-limited picture of chemisorption described here, and it underlies rate laws for catalytic reactions at surfaces.