Unit 1: Kinetic Theory of Gases & Gas Transport Phenomena

The classical kinetic molecular theory of ideal gases establishes a direct, rigorous bridge between microscopic particle kinematics and macroscopic thermodynamic state variables.

Fundamental Postulates

1. Point-Mass Particles: The gas consists of an ensemble of $N$ identical molecules of mass $m$, occupying a container of volume $V$. The volume of the individual molecules is negligibly small compared to $V$.
2. Elastic Collisions: Intermolecular forces are negligible except during instantaneous, perfectly elastic collisions where total kinetic energy and linear momentum are conserved.
3. Random Isotropic Motion: Particles move continuously in straight-line trajectories governed by Newton's equations of motion, distributed isotropically in velocity space.

Microscopic Derivation of Ideal Gas Pressure

Consider a cubical container of side length $L$ and volume $V = L^3$. Let a molecule possess Cartesian velocity components $(v_x, v_y, v_z)$. Upon colliding with an elastic planar wall perpendicular to the $x$-axis at $x = L$, the velocity component $v_x$ reverses to $-v_x$, while $v_y$ and $v_z$ remain unaltered.

The linear momentum transferred to the wall in a single collision event is:

The time elapsed between successive collisions of this molecule with the same wall is:

The average microscopic force exerted by this single particle on the wall is given by Newton's second law:

Summing over all $N$ molecules, the total force on the wall of area $A = L^2$ is:

The pressure $P$ exerted on the wall is:

Because particle velocity is isotropic in three-dimensional space:

Substituting into the pressure expression yields the fundamental kinetic equation of gas pressure:

Comparing with the macroscopic ideal gas equation of state $P V = n R T = N k_B T$, where $k_B = R / N_A$ is the Boltzmann constant:

Each translational degree of freedom contributes precisely $\frac{1}{2} k_B T$ to the average kinetic energy of a molecule, in full accord with the classical Equipartition Theorem.

Reference Table: Gas Transport Properties & Molecular Dimensions at 298.15 K and 1.00 atm

Below are benchmark experimental transport properties and collision dimensions derived from viscosity and thermal conductivity measurements using the Lennard-Jones (12-6) potential model:

Chapman-Enskog Transport Theory for Realistic Potentials

In real gases, molecules are not rigid hard spheres; intermolecular attractions ($r^{-6}$) and Pauli repulsions ($r^{-12}$) alter collision dynamics. The rigorous Chapman-Enskog solution of the Boltzmann transport equation yields:

where $\sigma$ is the collision diameter, $T^ = k_B T / \epsilon$ is the reduced temperature relative to the Lennard-Jones well depth $\epsilon$, and $\Omega^{(l,s)}(T^)$ are dimensionless collision integrals. At high temperatures ($T^ \gg 1$), $\Omega^ \to 1$ and hard-sphere scaling is recovered; at low temperatures, attractive well trapping increases $\Omega^$, enhancing the effective collision cross-section.

The equilibrium distribution of molecular velocities is governed by the Boltzmann factor $\exp(-\epsilon / k_B T)$. In three dimensions, translational kinetic energy is quadratic and uncoupled in each Cartesian component: $\epsilon = \frac{1}{2} m (v_x^2 + v_y^2 + v_z^2)$.

Cartesian 1D Velocity Probability Density

The probability $f(v_x) dv_x$ that a molecule has an $x$-velocity between $v_x$ and $v_x + dv_x$ is normalized:

This is a Gaussian centered at $\langle v_x \rangle = 0$, with variance $\sigma_x^2 = \langle v_x^2 \rangle = k_B T / m$.

3D Velocity Distribution

Because the Cartesian velocity components are statistically independent:

Transformation to Scalar Speed Distribution $f(v)$

To obtain the probability distribution of scalar speeds $v = (v_x^2 + v_y^2 + v_z^2)^{1/2}$, we transform from Cartesian coordinates to spherical polar coordinates in velocity space:

Integrating over all solid angles $\int_0^\pi \sin\theta d\theta \int_0^{2\pi} d\phi = 4\pi$:

In terms of molar mass $M = m N_A$ and gas constant $R = N_A k_B$:

The factor $v^2$ represents the growing volume of spherical shells in velocity space (density of states), while the exponential factor represents the Boltzmann thermal decay. The competition between these two terms generates an asymmetric distribution with a positive skew.

From the Maxwell-Boltzmann scalar speed distribution, three characteristic molecular speeds quantify different aspects of kinetic motion.

1. Most Probable Speed ($v_{\text{mp}}$)

The most probable speed corresponds to the local maximum of the probability density function $f(v)$. Setting $\frac{df(v)}{dv} = 0$:

Dividing by $v \exp(-m v^2 / 2 k_B T) \neq 0$:

2. Mean (Average) Speed ($\bar{v}$ or $\langle v \rangle$)

The mean speed is the first moment of the distribution:

Using standard Gaussian definite integrals $\int_0^\infty x^3 e^{-a x^2} dx = \frac{1}{2 a^2}$, with $a = \frac{m}{2 k_B T}$:

3. Root-Mean-Square Speed ($v_{\text{rms}}$)

The root-mean-square speed is the square root of the second moment:

Using $\int_0^\infty x^4 e^{-a x^2} dx = \frac{3}{8} \sqrt{\frac{\pi}{a^5}}$:

Numerical Ordering & Comparison

At any temperature $T$, $v_{\text{mp}} < \bar{v} < v_{\text{rms}}$ because the long exponential tail at high speeds pulls the higher-order moments towards higher velocities.

University Honors Research Monograph: The Stern-Gerlach Experiment & Spatial Velocity Selection

In 1922, Otto Stern and Walther Gerlach utilized thermal effusion of silver atoms from an electrically heated oven into a high-vacuum chamber to test spatial quantization:

• Effusion Collimation: Silver atoms effused through a narrow pinhole ($d \approx 0.1\text{ mm}$) at $T = 1300\text{ K}$, passing through two micro-machined slits to form a ribbon-like molecular beam with transverse angular divergence $< 0.1^\circ$.
• Inhomogeneous Magnetic Field: The effusing beam traversed a $3.5\text{ cm}$ magnetic gap between a knife-edge pole piece and a grooved opposing pole, creating an intense transverse field gradient:

• Deflection Force: The net magnetic force acting on an effusing silver atom of magnetic moment $\vec{\mu}$ was:

• Velocity Dispersion Elimination: Because classical Maxwell-Boltzmann effusion produces a broad velocity distribution ($f(v) \propto v^3 \exp(-M v^2 / 2 R T)$), classical mechanics predicted a continuous smeared distribution on the glass plate detector. Instead, Stern and Gerlach observed the beam splitting cleanly into two discrete, symmetrically displaced parabolic deposits ($z = \pm 0.1\text{ mm}$), providing the first direct experimental proof of electron spin quantization ($m_s = \pm 1/2$) and spatial orientation quantization in atomic physics.

Molecules in a gas do not traverse infinite distances unobstructed; they undergo frequent binary collisions that randomize trajectories.

Hard-Sphere Collision Cross-Section

Consider spherical molecules of diameter $d$. Two identical molecules collide when the distance between their centers is $r \le d$. The collision cross-section $\sigma$ is the circular target area presented by one molecule to another:

Relative Speed of Colliding Pairs

When two identical molecules collide with velocities $\vec{v}_1$ and $\vec{v}_2$, their relative velocity is $\vec{v}_{\text{rel}} = \vec{v}_1 - \vec{v}_2$. Taking the ensemble average over isotropic distributions:

Collision Frequency of a Single Molecule ($z$)

In time $\Delta t$, a target molecule sweeps out a collision cylinder of volume $V_{\text{cyl}} = \sigma \bar{v}_{\text{rel}} \Delta t$. With number density $\mathcal{N} = N/V = P/(k_B T)$, the number of collisions per unit time experienced by one molecule is:

Total Collision Density ($Z_{AA}$)

The total number of binary collisions per unit volume per unit time in a pure gas is:

The factor $\frac{1}{2}$ prevents double-counting the collision pair.

Mean Free Path ($\lambda$)

The mean free path $\lambda$ is the average distance traversed by a molecule between successive collisions:

Key scaling relationships:

• $\lambda \propto T$ at constant pressure $P$.
• $\lambda \propto \frac{1}{P}$ at constant temperature $T$.
• $\lambda$ is independent of temperature $T$ at constant volume/density.

The rate at which gas molecules strike a surface governs effusion, adsorption, chemical vapor deposition, and heterogeneous reaction rates.

Derivation of Collision Frequency with Walls ($Z_w$)

Consider an element of wall area $A$ in the $xy$-plane at $z = 0$. Molecules with positive $z$-velocity $v_z > 0$ located within distance $v_z \Delta t$ will strike area $A$ during time interval $\Delta t$.

The number of molecules with velocity between $v_z$ and $v_z + dv_z$ striking area $A$ in $\Delta t$ is:

Integrating over all positive velocities $v_z \in [0, \infty)$:

Evaluating the integral:

Thus:

Using $\mathcal{N} = P / (k_B T)$:

Knudsen Effusion & Graham's Law

If a pinhole orifice of area $A_0$ has dimensions much smaller than the mean free path ($d_{\text{hole}} \ll \lambda$, Knudsen number $\text{Kn} = \lambda / d_{\text{hole}} \gg 1$), molecules escape into vacuum without undergoing collisions in the aperture. This is effusion.

The molar rate of effusion is:

The mass rate of effusion is:

For two different gases at identical temperature and pressure:

This provides the rigorous kinetic derivation of Graham's Law of Effusion.

Transport phenomena describe the macroscopic non-equilibrium flux of physical quantities—momentum, thermal energy, and mass—driven by gradients in macroscopic fields.

Phenomenological Definition of Viscosity

Newton's law of viscosity states that when a shear velocity gradient $\frac{du_x}{dz}$ exists in a fluid, a shear stress $\tau_{xz}$ (momentum flux per unit area) opposes the shear:

where $\eta$ is the dynamic viscosity coefficient (SI units: $\text{Pa}\cdot\text{s}$ or $\text{kg}/(\text{m}\cdot\text{s})$).

Kinetic Theory Derivation of Viscosity

Consider a gas with a macroscopic velocity profile $u_x(z)$ moving in the $x$-direction, where $u_x$ increases with $z$. Molecules crossing a reference plane at $z = z_0$ last collided on average at distance $\lambda$ above or below the plane.

1. Downward flux of molecules from $z_0 + \frac{2}{3}\lambda$:

Molecules carrying momentum $p_x^+ = m u_x\left(z_0 + \frac{2}{3}\lambda\right) \approx m \left[ u_x(z_0) + \frac{2}{3}\lambda \frac{du_x}{dz} \right]$.

1. Upward flux of molecules from $z_0 - \frac{2}{3}\lambda$:

Molecules carrying momentum $p_x^- = m u_x\left(z_0 - \frac{2}{3}\lambda\right) \approx m \left[ u_x(z_0) - \frac{2}{3}\lambda \frac{du_x}{dz} \right]$.

The net momentum flux $J_{p_x}$ transported downward across the plane per unit area per second is:

Comparing with Newton's law:

where $\rho = \mathcal{N} m$ is the mass density.

Substituting $\lambda = \frac{1}{\sqrt{2} \pi d^2 \mathcal{N}}$ and $\bar{v} = \sqrt{\frac{8 k_B T}{\pi m}}$:

Surprising Physical Consequences

1. Pressure Independence: Because $\rho \propto P$ and $\lambda \propto 1/P$, their product $\rho \lambda$ is independent of pressure. Dilute gas viscosity is independent of gas pressure (Maxwell's celebrated prediction, verified by experiment).
2. Positive Temperature Dependence: $\eta \propto \sqrt{T}$. As temperature increases, gases become more viscous, in sharp contrast to liquids.

Thermal conductivity represents the transport of kinetic energy down a temperature gradient $\frac{dT}{dz}$.

Phenomenological Law (Fourier's Law)

The heat flux vector $J_q$ (energy per unit area per unit time) is proportional to the negative thermal gradient:

where $\kappa$ is the thermal conductivity coefficient (SI units: $\text{W}/(\text{m}\cdot\text{K})$).

Kinetic Theory Derivation of $\kappa$

By analogy with momentum transport, molecules crossing a reference plane at $z_0$ carry the average thermal energy characteristic of their last collision at $z_0 \pm \frac{2}{3}\lambda$:

Noting that $\frac{d\epsilon}{dT} = c_v = \frac{C_{V, m}}{N_A}$ is the heat capacity per molecule at constant volume:

Comparing with Fourier's law yields:

Substituting $\eta = \frac{1}{3} \rho \bar{v} \lambda$:

Eucken Correction for Polyatomic Gases

For a monatomic gas with only translational degrees of freedom, $C_{V, m} = \frac{3}{2} R$. Rigorous Chapman-Enskog kinetic theory yields a factor of $2.5$ for translational motion:

For polyatomic gases carrying rotational and vibrational energy, the Eucken formula partitions transport into translational and internal modes:

Like viscosity, dilute gas thermal conductivity $\kappa$ is independent of pressure and scales as $\sqrt{T}$.

Effusion phenomena and free-jet gas expansions provide the experimental foundation for modern gas-phase reaction dynamics, molecular beam spectroscopy, and high-temperature vaporization thermodynamics.

1. Knudsen Effusion Mass Spectrometry (KEMS)

Knudsen Effusion Mass Spectrometry (developed by Inghram, Chupka, and Drowart) is the primary metrological method for measuring vapor pressures of refractory materials and determining dissociation energies of high-temperature gaseous species ($D_0^\circ$).

A sample is placed inside a sealed Knudsen cell (made of chemically inert tungsten, molybdenum, or alumina) maintained at temperature $T$ inside an ultra-high vacuum chamber ($P_{\text{chamber}} < 10^{-7}\text{ Torr}$). Vapor molecules escape through an ideal circular orifice of area $A_0$ having knife-edge walls ($L_w \ll d_{\text{hole}}$). The Knudsen number satisfies:

ensuring that molecular collisions within the orifice are negligible and thermodynamic vapor-liquid/solid equilibrium inside the cell is undisturbed.

The molecular effusion beam is ionized by an electron impact source ($e^- + M \longrightarrow M^{+\bullet} + 2 e^-$) or tunable synchrotron VUV radiation, and ion intensities $I_i^+$ are measured by a quadrupole or time-of-flight mass spectrometer. The partial vapor pressure $P_i$ inside the cell is related to ion intensity by the fundamental KEMS formula:

where $k_{\text{cal}}$ is an instrumental calibration constant (determined from silver or gold vapor standards), $\sigma_i$ is the electron impact ionization cross-section, and $\gamma_i$ is detector multiplier gain.

By recording $P_i(T)$ over temperature, the standard enthalpy of vaporization $\Delta_{\text{vap}} H^\circ$ is extracted from the Clausius-Clapeyron relation:

2. Supersonic Free-Jet Aerodynamic Expansion

In contrast to effusive beams where particles exit independently without collisions, a supersonic free jet is created by expanding a high-pressure carrier gas ($P_0 = 1 - 50\text{ bar}$, typically He or Ar) through a tiny pinhole nozzle ($d = 50 - 200\;\mu\text{m}$) into high vacuum.

Within the first few nozzle diameters downstream of the orifice, molecules undergo tens of thousands of binary collisions. In this continuum hydrodynamic expansion zone:

• Random thermal kinetic energy is converted into directed, uniform forward velocity $u$.
• The local Mach number exceeds unity ($M = u / a > 1$).
• The velocity distribution narrows dramatically, collapsing the translational temperature $T_{\text{trans}}$ to cryogenic values ($T_{\text{trans}} < 1 - 5\text{ K}$).

Cooling Dynamics & Non-Equilibrium Freezing

Because collision rates scale with density ($Z \propto n$), the collision frequency drops precipitously as the gas expands, causing different molecular degrees of freedom to decouple at distinct downstream locations ("freezing"):

1. Translational Cooling: Energy transfer between translational modes occurs in $1 - 2$ collisions; temperatures drop to $0.5 - 2\text{ K}$.
2. Rotational Cooling: Rotational-translational ($R-T$) relaxation requires $\sim 10 - 20$ collisions; rotational temperatures reach $2 - 10\text{ K}$, simplifying complex rovibrational spectra into isolated ground-state lines.
3. Vibrational Freezing: Vibrational-translational ($V-T$) relaxation requires $10^4 - 10^6$ collisions. Because density drops before sufficient collisions occur, vibrational cooling freezes early at $T_{\text{vib}} \approx 50 - 150\text{ K}$.

A conical skimmer extracts the central core of the supersonic jet, forming a pristine, collision-free molecular beam with narrow velocity dispersion ($\Delta v / v < 1\%$).