Physical Chemistry
Maxwell-Boltzmann Distribution
Speed distribution of gas molecules — f(v) = 4π(m/2πkT)^(3/2) v² exp(−mv²/2kT)
The Maxwell-Boltzmann distribution f(v) = 4π(m/2πkT)^(3/2) v² exp(−mv²/2kT) gives the probability density of molecular speeds in a classical gas at thermal equilibrium. Derived by James Clerk Maxwell in 1860 and put on rigorous statistical-mechanics footing by Ludwig Boltzmann between 1868 and 1872, it predicts three characteristic speeds in the fixed ratio 1 : 1.128 : 1.225 for v_mp : v_avg : v_rms. For nitrogen at 300 K, v_rms ≈ 515 m/s; for helium at the same temperature, v_rms ≈ 1370 m/s — fast enough that helium continually leaks from Earth's gravity well. The mean free path at standard conditions is roughly 70 nm.
- Equationf(v) = 4π(m/2πkT)^(3/2) v² e^(−mv²/2kT)
- v_rms(3kT/m)^(1/2)
- N₂ at 300 Kv_rms ≈ 515 m/s
- He at 300 Kv_rms ≈ 1370 m/s
- Mean free path STP≈ 70 nm
- DerivedMaxwell 1860, Boltzmann 1872
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why the Maxwell-Boltzmann distribution matters
- Foundation of kinetic theory of gases. Pressure, viscosity, thermal conductivity, and diffusion coefficients all derive from integrals over f(v). Pressure on a wall is P = (1/3) n m ⟨v²⟩, which directly recovers PV = nRT once ⟨v²⟩ = 3kT/m is plugged in.
- Underwrites the Arrhenius equation. Reaction rates depend on the fraction of molecules with energy above an activation barrier E_a. That fraction is ∫_{E_a}^∞ f(E) dE ∝ exp(−E_a/RT) — directly giving k = A exp(−E_a/RT). This is why a 10 K rise typically doubles a reaction rate at room temperature.
- Explains atmospheric escape. Earth's escape velocity is 11.2 km/s. The Maxwell-Boltzmann tail predicts a tiny but non-zero fraction of hydrogen and helium with speed above this — over geological time hydrogen is essentially gone and helium continually leaks. Heavier molecules like N₂ have a v_rms of only 515 m/s and the tail above 11.2 km/s is exp(−471) — utterly negligible.
- Time-of-flight molecular beams. Cosine-law effusion through a small hole transmits molecules with the local Maxwell-Boltzmann distribution. Knudsen cells, Stern-Gerlach experiments, and modern ultra-high vacuum surface science all rely on this prediction.
- Doppler broadening of spectral lines. Atoms moving toward an observer emit blue-shifted photons; those moving away red-shift. The line profile is the Gaussian convolution of the natural line shape with the velocity distribution along the line of sight, giving width Δν/ν = (8kT ln 2 / mc²)^(1/2). Spectral-line widths in stellar atmospheres remotely measure temperature this way.
- Explains heat capacity equipartition. Each translational degree of freedom contributes ½kT to ⟨E⟩; for monatomic gases the total kinetic energy is exactly (3/2)kT per molecule, giving C_V = (3/2)R ≈ 12.5 J/(mol·K) for noble gases — confirmed to high precision in helium, neon, and argon.
- Sets diffusion and thermal-conduction scales. Diffusion coefficient D ∝ v_avg λ / 3 and thermal conductivity κ ∝ (1/3) n c_v v_avg λ both follow from the Maxwell-Boltzmann mean speed and mean free path. For air at 300 K and 1 atm: D(N₂ in air) ≈ 0.2 cm²/s, κ ≈ 0.026 W/(m·K).
Common misconceptions
- v_avg equals the speed at the peak of the curve. No — the peak is at v_mp, which is smaller. The distribution is asymmetric with a long high-speed tail, so v_avg > v_mp and v_rms > v_avg. The fixed ratios are 1 : 1.128 : 1.225.
- Most molecules travel near v_rms. Most molecules are clustered around v_mp, not v_rms. v_rms is dominated by the upper tail because squaring weights large speeds disproportionately.
- The distribution depends on the molecule's chemical identity. It depends only on mass and temperature. At a given T, all molecules of the same mass have the identical f(v). N₂ and CO have nearly the same molecular mass (28 g/mol) so their distributions are indistinguishable.
- Maxwell-Boltzmann is symmetric like a Gaussian. The velocity-component distribution is a Gaussian, but the speed distribution has the v² Jacobian — so it is asymmetric, going to zero at v = 0 and decaying exponentially at high v.
- It applies in any quantum regime. No. When λ_dB ≳ n^(−1/3), Bose-Einstein or Fermi-Dirac statistics dominate. Conduction electrons in metals at room temperature follow Fermi-Dirac, not Maxwell-Boltzmann — a common student trap.
- The temperature in the formula is the speed's average. Temperature is a property of the equilibrium ensemble, not a single molecule. A handful of molecules sampled from the gas does not have a temperature; the distribution they were drawn from does.
Derivation from equipartition and isotropy
Maxwell's 1860 derivation rests on two physical assumptions: the gas is isotropic (no preferred direction in velocity space) and the velocity components v_x, v_y, v_z are statistically independent. Independence plus isotropy forces the velocity distribution to factor as f(v_x) f(v_y) f(v_z) and depend only on v_x² + v_y² + v_z² — a functional equation whose only solution is the product of three Gaussians, f(v_i) ∝ exp(−α v_i²). Equipartition fixes α = m/(2kT). Boltzmann's later work in 1872 derived the same result more rigorously from the H-theorem: any deviation from Maxwell-Boltzmann form decreases the H-functional under classical collisions, so equilibrium uniquely produces this distribution.
To get the speed distribution, transform from velocity vector to magnitude. The volume element in velocity space at speed v is 4π v² dv (a spherical shell). Multiplying gives f(v) = 4π v² (m/2πkT)^(3/2) exp(−mv²/2kT). Three characteristic speeds drop out: setting df/dv = 0 yields v_mp = (2kT/m)^(1/2); integrating ∫v f(v) dv gives v_avg = (8kT/πm)^(1/2); and ∫v² f(v) dv gives v_rms = (3kT/m)^(1/2). Numerically v_mp : v_avg : v_rms = 1 : (4/π)^(1/2) : (3/2)^(1/2) ≈ 1 : 1.128 : 1.225.
The kinetic energy distribution follows from f(E) dE = f(v) dv with E = ½mv². Substituting gives f(E) = 2(E/π)^(1/2) (1/kT)^(3/2) exp(−E/kT), the gamma distribution with shape parameter 3/2. Mean energy is ⟨E⟩ = (3/2)kT — recovering equipartition exactly. The fraction of molecules with energy above a threshold E_a in the high-energy tail is well-approximated by exp(−E_a/kT) for E_a ≫ kT, which is the Boltzmann factor that powers the Arrhenius rate law.
v_mp vs v_avg vs v_rms across gases at 300 K
| Gas | Molar mass (g/mol) | v_mp (m/s) | v_avg (m/s) | v_rms (m/s) | Notes |
|---|---|---|---|---|---|
| Hydrogen (H₂) | 2.016 | 1574 | 1776 | 1928 | Fastest molecular gas; escapes Earth's atmosphere |
| Helium (He) | 4.003 | 1116 | 1259 | 1367 | Second-fastest; Sun-derived He still leaks |
| Methane (CH₄) | 16.04 | 557 | 628 | 682 | Important atmospheric tracer |
| Water (H₂O) | 18.02 | 525 | 593 | 644 | Vapor at 300 K above the liquid |
| Nitrogen (N₂) | 28.01 | 422 | 476 | 517 | Air's main component |
| Oxygen (O₂) | 32.00 | 395 | 446 | 483 | 21% of dry air |
| Argon (Ar) | 39.95 | 353 | 399 | 432 | 1% of dry air; monatomic |
| Carbon dioxide (CO₂) | 44.01 | 337 | 380 | 413 | Triatomic; high heat capacity |
| Xenon (Xe) | 131.3 | 195 | 220 | 239 | Heaviest stable noble gas |
Maxwell-Boltzmann vs Bose-Einstein vs Fermi-Dirac vs Maxwell-Jüttner
| Distribution | Year / author | Particle type | Functional form (occupation) | Regime |
|---|---|---|---|---|
| Maxwell-Boltzmann | 1860 / 1872 | Classical, distinguishable | exp(−E/kT) | Dilute gases, λ_dB ≪ spacing |
| Bose-Einstein | 1924 | Bosons (integer spin) | 1/(exp((E−μ)/kT) − 1) | Photons, ⁴He, BECs |
| Fermi-Dirac | 1926 | Fermions (half-integer spin) | 1/(exp((E−μ)/kT) + 1) | Electrons in metals, white dwarfs |
| Maxwell-Jüttner | 1911 / Jüttner | Relativistic classical | γ²β² exp(−γmc²/kT)/K₂(mc²/kT) | kT comparable to or larger than mc² |
| Boltzmann (energy form) | 1872 | Classical, energy levels | g_i exp(−E_i/kT)/Z | Discrete-level gases |
| Power-law / κ-distribution | Astrophysical fits | Out-of-equilibrium plasmas | (1 + E/(κkT))^(−κ−1) | Solar wind, magnetospheres |
Applications
- Reaction kinetics — Arrhenius and collision theory. The fraction of molecules with translational kinetic energy above E_a is the Maxwell-Boltzmann tail; this fraction times the collision frequency gives the bimolecular rate constant. Combustion engineers, atmospheric chemists, and pharmacologists all use this directly.
- Atmospheric escape and planetary science. The Jeans escape rate for hydrogen from Earth and Mars is computed by integrating the Maxwell-Boltzmann tail above the local escape velocity. This explains why Mars lost its hydrogen-bearing water and why Earth's atmosphere is nitrogen-rich rather than hydrogen-rich.
- Spectroscopy and Doppler thermometry. Plasma temperatures in tokamaks, stellar atmospheric temperatures, and interstellar cloud temperatures are routinely measured from line widths using the Maxwell-Boltzmann velocity distribution along the line of sight.
- Vacuum technology. Mean free path calculations from MB distribution determine pump throughput, sensor placement, and contamination rates in semiconductor fabs and particle accelerators. Below 10⁻⁴ Pa, λ exceeds typical chamber dimensions and molecular flow regime takes over.
- Mass spectrometry and isotope separation. Time-of-flight mass spectrometers exploit the (m)^(−1/2) scaling of v_rms — lighter ions arrive at the detector first. Gaseous diffusion enrichment of uranium during the Manhattan Project relied on the (1/m)^(1/2) speed difference between ²³⁵UF₆ and ²³⁸UF₆.
Frequently asked questions
How are v_mp, v_mean, and v_rms different?
All three are characteristic speeds of the same distribution but they answer different questions. The most probable speed v_mp = (2kT/m)^(1/2) is the peak of f(v) — the speed at which the largest fraction of molecules are found per unit speed. The mean speed v_avg = (8kT/πm)^(1/2) is the arithmetic average — the integral of v · f(v) — and is the relevant quantity for collision frequencies. The root-mean-square speed v_rms = (3kT/m)^(1/2) comes from the kinetic energy per molecule (½mv²) = (3/2)kT and is the speed used in pressure derivations. The ratio is fixed: v_mp : v_avg : v_rms = 1 : 1.128 : 1.225, regardless of gas or temperature. For N₂ at 300 K, these are 422, 476, and 515 m/s respectively.
What is the mean free path and how does it follow from the distribution?
The mean free path λ is the average distance a molecule travels between collisions. From kinetic theory, λ = kT / (√2 π d² P), where d is the molecular diameter. For air at standard temperature and pressure (273 K, 1 atm) and an effective d ≈ 0.37 nm, λ ≈ 70 nm — about 200 molecular diameters. The √2 factor arises because the relative speed between two Maxwell-Boltzmann molecules is √2 times the mean speed of one. The mean free path scales inversely with pressure, so in a vacuum chamber at 10⁻⁶ Pa, λ exceeds 50 metres — molecules cross the chamber without colliding, which is why this is called the molecular-flow regime.
Why does the curve have the v² Jacobian factor?
The underlying probability density of velocity components is a 3D Gaussian, f_3D(v_x, v_y, v_z) ∝ exp(−m(v_x² + v_y² + v_z²)/2kT). Converting from velocity vector to speed magnitude integrates over all directions for a given speed v — the surface area of a sphere of radius v in velocity space is 4π v². So the speed distribution picks up the factor 4π v², which makes f(v) = 0 at v = 0 (no molecules with zero speed even though zero is the most probable single velocity) and shifts the peak to v_mp > 0. The competition between the rising v² and the decaying Gaussian is what gives the distribution its characteristic asymmetric peak with a long high-speed tail.
When does Maxwell-Boltzmann break down?
When quantum statistics become important. At high density and low temperature, fermions like electrons follow Fermi-Dirac statistics with a sharp Fermi surface, and bosons follow Bose-Einstein with possible condensation. The crossover happens when the thermal de Broglie wavelength λ_dB = h/(2π m kT)^(1/2) becomes comparable to the inter-particle spacing n^(−1/3). For air at 300 K and atmospheric density, λ_dB ≈ 18 pm versus a spacing of 3.4 nm — Maxwell-Boltzmann is excellent. For conduction electrons in copper at 300 K, λ_dB ≈ 4 nm and the spacing is about 0.23 nm — Fermi-Dirac dominates. Maxwell-Boltzmann is also inadequate at relativistic speeds (T > mc²/k), where the Maxwell-Jüttner distribution replaces it.
How was the distribution verified experimentally?
The first quantitative confirmation came from Otto Stern's molecular beam experiments at Hamburg in the 1920s, using a slotted rotating drum to time-of-flight molecules from an oven. Later, Estermann, Simpson, and Stern in 1947 measured the speed distribution of cesium atoms with sub-1% accuracy. Modern verification is routine: time-of-flight mass spectrometry, laser-induced fluorescence, and Doppler-broadening of spectral lines all reproduce the predicted f(v) and its temperature scaling. Doppler broadening of the 656 nm hydrogen line at 6000 K (solar photosphere) shows exactly the Gaussian profile expected from Maxwell-Boltzmann velocities along the line of sight, with width Δλ/λ = (8kT ln 2/(mc²))^(1/2) ≈ 4 × 10⁻⁵.
Why does activation energy matter so much for reaction rates?
The Arrhenius rate constant k = A exp(−E_a/RT) is proportional to the fraction of molecules with kinetic energy above E_a — exactly the high-energy tail of the Maxwell-Boltzmann distribution. Because the tail decays exponentially, small temperature increases cause large rate increases. For E_a = 50 kJ/mol, going from 300 K to 310 K raises the fraction of molecules above E_a by a factor of about 1.94 — close to the empirical rule of thumb that reaction rates double per 10 K. For larger activation energies the multiplier is even bigger. This explains why food spoilage roughly doubles per 10 K, why enzymes evolved at body temperature, and why combustion reactions ignite at sharp threshold temperatures.