5.2 Equipartition theorem

Equipartition theorem fundamentals

The equipartition theorem tells you how energy gets shared among the different ways a particle can move or vibrate. In a system at thermal equilibrium, each independent "way" of storing energy (called a degree of freedom) gets the same average amount of energy. This connection between microscopic motion and macroscopic quantities like temperature and heat capacity makes it one of the most useful results in classical statistical mechanics.

Definition and basic principles

The core statement: every quadratic term in the system's Hamiltonian contributes $\frac{1}{2}k_BT$ of average energy per particle.

A "quadratic term" means any term proportional to the square of a coordinate or momentum, like $\frac{1}{2}mv_x^2$ (kinetic energy in one direction) or $\frac{1}{2}kx^2$ (potential energy of a spring). The theorem applies to systems with many particles in thermal equilibrium and covers translational, rotational, and vibrational modes of motion.

Historical development

• James Clerk Maxwell laid the groundwork through his kinetic theory of gases in the 1860s, showing that molecular velocities follow a specific distribution
• Ludwig Boltzmann extended this into a broader statistical framework, connecting energy distribution to probability
• J. Willard Gibbs formalized the theorem using statistical ensembles, giving it the rigorous mathematical foundation used today
• In the early 20th century, failures of equipartition (particularly for heat capacities) became key evidence motivating quantum mechanics

Assumptions and limitations

The theorem rests on several assumptions, and knowing when they break down is just as important as knowing the theorem itself:

• Classical behavior: Particles are treated classically. This fails at low temperatures where quantum effects matter.
• Thermal equilibrium: The system must have had time to reach a steady-state energy distribution.
• Quadratic energy terms: The $\frac{1}{2}k_BT$ result holds specifically for terms quadratic in coordinates or momenta. Non-quadratic terms require a generalized treatment.
• Weak interactions: Strong inter-particle interactions can invalidate the theorem's predictions.
• Non-relativistic regime: At very high energies where particle speeds approach the speed of light, the classical Hamiltonian no longer applies.

Degrees of freedom

A degree of freedom is an independent way a system can store energy. Counting degrees of freedom correctly is the key step in applying the equipartition theorem, since the total average energy is just $\frac{1}{2}k_BT$ multiplied by the number of active quadratic degrees of freedom.

Translational degrees of freedom

These describe the motion of a particle's center of mass through space. In three dimensions, there are three translational degrees of freedom (motion along x, y, and z). Each contributes $\frac{1}{2}k_BT$ to the average kinetic energy.

For a monatomic ideal gas (like argon), translation is the only type of motion available. The average kinetic energy per atom is therefore $\frac{3}{2}k_BT$.

Rotational degrees of freedom

Molecules can also rotate around their center of mass, and each independent axis of rotation counts as a degree of freedom.

• Linear molecules (like $CO_2$ or $N_2$) have two rotational degrees of freedom. Rotation about the molecular axis doesn't count because the moment of inertia along that axis is negligibly small.
• Non-linear molecules (like $H_2O$) have three rotational degrees of freedom, one for each principal axis.

Each rotational degree of freedom contributes $\frac{1}{2}k_BT$ to the average energy.

Vibrational degrees of freedom

Atoms within a molecule can oscillate back and forth relative to each other. Each vibrational mode acts like a harmonic oscillator, which has both kinetic and potential energy terms. That means each vibrational mode contributes two quadratic terms, giving $k_BT$ per mode (not just $\frac{1}{2}k_BT$).

A molecule with $N$ atoms has $3N - 5$ vibrational modes if linear, or $3N - 6$ if non-linear. Vibrational modes typically require higher temperatures to become fully active, which is why heat capacities of polyatomic gases increase with temperature.

Energy distribution

Average energy per degree of freedom

The central quantitative result of equipartition:

• Each quadratic degree of freedom contributes $\frac{1}{2}k_BT$ on average
• Total average energy = (number of quadratic terms) $\times \frac{1}{2}k_BT$
• Monatomic ideal gas (3 translational DOF): average energy per molecule = $\frac{3}{2}k_BT$
• Diatomic gas at moderate temperature (3 translational + 2 rotational): average energy per molecule = $\frac{5}{2}k_BT$
• Each vibrational mode contributes $k_BT$ ($\frac{1}{2}k_BT$ kinetic + $\frac{1}{2}k_BT$ potential)

Equipartition and temperature

Temperature, from the equipartition perspective, is directly proportional to the average energy per quadratic degree of freedom:

$\langle E_{\text{per DOF}} \rangle = \frac{1}{2}k_BT$

This gives temperature a concrete molecular meaning: it measures how much kinetic energy is carried, on average, by each mode of motion. It also explains why temperature is an intensive property. Adding more particles increases total energy but doesn't change the average energy per degree of freedom.

Boltzmann distribution connection

The equipartition theorem isn't a separate postulate; it's a consequence of the Boltzmann distribution. In the canonical ensemble, the probability of finding a system in a state with energy $E$ is proportional to $e^{-E/k_BT}$. When you compute the thermal average of a quadratic energy term $\frac{1}{2}\alpha q^2$ using this distribution, the Gaussian integral yields exactly $\frac{1}{2}k_BT$. This derivation is worth working through, since it shows precisely why the result only holds for quadratic terms.

Applications in classical systems

Ideal gas model

Equipartition directly predicts the thermodynamic properties of ideal gases:

• Internal energy of a monatomic ideal gas: $U = \frac{3}{2}nRT$ (since each molecule has 3 translational DOF)
• The ideal gas law $PV = nRT$ can be derived by connecting the average translational kinetic energy to pressure via momentum transfer to container walls
• For diatomic gases at room temperature: $U = \frac{5}{2}nRT$

Specific heat capacity

Heat capacity predictions follow directly from the internal energy:

• Monatomic ideal gas: $C_V = \frac{3}{2}R \approx 12.5 \text{ J/(mol·K)}$
• Diatomic ideal gas (room temperature, rotations active): $C_V = \frac{5}{2}R \approx 20.8 \text{ J/(mol·K)}$
• Solids at high temperature: each atom has 3 vibrational modes, each contributing $k_BT$, giving $C_V = 3R \approx 25 \text{ J/(mol·K)}$. This is the Dulong-Petit law, and it works well for most solids above their Debye temperature.

Deviations from these predictions at low temperatures were historically important because they pointed toward quantum mechanics.

Brownian motion

The random, jittery motion of small particles suspended in a fluid (like pollen grains in water) is a direct consequence of thermal energy. Equipartition predicts that each translational degree of freedom of a Brownian particle carries $\frac{1}{2}k_BT$ of kinetic energy, regardless of the particle's size. This means:

$\frac{1}{2}m\langle v_x^2 \rangle = \frac{1}{2}k_BT$

Larger particles move more slowly, but they carry the same average kinetic energy per degree of freedom as the surrounding fluid molecules. Einstein's analysis of Brownian motion, grounded in equipartition, provided some of the first direct evidence for the existence of atoms.

Quantum mechanical considerations

Quantum vs. classical equipartition

Classical equipartition assumes energy can take any continuous value. In quantum mechanics, energy levels are discrete, and this discreteness matters when the spacing between levels is comparable to or larger than $k_BT$.

At high temperatures ($k_BT \gg$ energy level spacing), quantum systems recover the classical equipartition result. This is the correspondence principle in action. At low temperatures, degrees of freedom with large energy spacings effectively "freeze out" and stop contributing to the heat capacity.

Low temperature deviations

The most dramatic failures of classical equipartition occur at low temperatures:

• Solids: Heat capacity drops well below the Dulong-Petit value of $3R$ as temperature approaches zero, eventually going to zero as required by the third law of thermodynamics. The Einstein and Debye models account for this using quantized vibrational modes.
• Diatomic gases: Rotational contributions to heat capacity disappear below a characteristic temperature (around 80 K for $H_2$). Vibrational contributions freeze out at even higher temperatures.
• Exotic phenomena like Bose-Einstein condensation and superconductivity represent complete departures from classical energy distribution.

Quantum harmonic oscillator

The quantum harmonic oscillator is the standard model for understanding how equipartition breaks down. Its energy levels are:

$E_n = \left(n + \frac{1}{2}\right)\hbar\omega, \quad n = 0, 1, 2, \ldots$

The average energy at temperature $T$ is:

$\langle E \rangle = \frac{\hbar\omega}{2} + \frac{\hbar\omega}{e^{\hbar\omega/k_BT} - 1}$

• When $k_BT \gg \hbar\omega$: the average energy approaches $k_BT$, recovering the classical equipartition result.
• When $k_BT \ll \hbar\omega$: the average energy approaches the zero-point energy $\frac{1}{2}\hbar\omega$, and the mode contributes negligibly to heat capacity.

The ratio $\hbar\omega / k_BT$ determines whether a given mode behaves classically or quantum mechanically.

Equipartition in statistical ensembles

The equipartition theorem can be derived within different ensemble frameworks. The choice of ensemble depends on what constraints are imposed on the system.

Microcanonical ensemble

This ensemble describes an isolated system with fixed energy $E$, volume $V$, and particle number $N$. Equipartition follows from the fundamental postulate that all accessible microstates at a given energy are equally probable. By computing averages over the constant-energy surface in phase space, you recover the $\frac{1}{2}k_BT$ result.

Canonical ensemble

The canonical ensemble describes a system in thermal contact with a heat bath at temperature $T$. Energy fluctuates, but temperature is fixed. This is the most common setting for deriving equipartition, since the Boltzmann factor $e^{-H/k_BT}$ makes the Gaussian integrals over quadratic terms straightforward. The partition function approach gives the cleanest derivation.

Grand canonical ensemble

Here both energy and particle number fluctuate, with the system exchanging particles and energy with a reservoir. The chemical potential $\mu$ and temperature $T$ are fixed. Equipartition still applies to the average energy per degree of freedom, and this ensemble is particularly useful for studying open systems and systems near phase transitions.

Experimental validations

Dulong-Petit law

In 1819, Dulong and Petit observed that the molar heat capacity of many solid elements is approximately $3R \approx 25 \text{ J/(mol·K)}$. Equipartition explains this: each atom in a solid vibrates in three dimensions, giving 6 quadratic terms (3 kinetic + 3 potential), and $6 \times \frac{1}{2}k_BT = 3k_BT$ per atom.

Deviations at low temperatures (diamond, for instance, has an anomalously low heat capacity at room temperature due to its high Debye temperature) were among the first clues that classical physics was incomplete.

Specific heat measurements

• Gas-phase measurements confirm $C_V = \frac{3}{2}R$ for monatomic gases and $C_V = \frac{5}{2}R$ for diatomic gases at room temperature
• Tracking heat capacity as a function of temperature reveals the stepwise activation of rotational and then vibrational modes
• Low-temperature calorimetry shows the $T^3$ dependence predicted by the Debye model, confirming quantum corrections to equipartition

Molecular spectroscopy

Rotational and vibrational spectra of molecules provide direct evidence for quantized energy levels. At high temperatures, the density of occupied states becomes large enough that the system behaves quasi-classically, consistent with equipartition. At low temperatures, only the lowest energy levels are populated, and classical predictions fail. Spectroscopic data combined with statistical mechanical calculations have validated the equipartition framework across many molecular systems.

Limitations and breakdowns

High temperature limits

At extremely high temperatures, particles can reach relativistic speeds, and the kinetic energy is no longer simply $\frac{1}{2}mv^2$. The classical Hamiltonian breaks down, and relativistic statistical mechanics or quantum field theory becomes necessary. In astrophysical plasmas and particle physics contexts, simple equipartition no longer applies.

Low temperature failures

This is the most physically important limitation. As $T \to 0$:

• Heat capacities vanish (consistent with the third law of thermodynamics)
• Degrees of freedom freeze out one by one as $k_BT$ drops below their characteristic energy spacing
• Collective quantum phenomena (superfluidity, superconductivity, Bose-Einstein condensation) emerge, which have no classical analog

Non-equilibrium systems

Equipartition requires thermal equilibrium. Systems driven out of equilibrium, such as turbulent flows, actively reacting chemical mixtures, or systems under external forcing, can have highly non-uniform energy distributions. Non-equilibrium statistical mechanics is needed to handle these cases, and the energy per degree of freedom can differ dramatically from $\frac{1}{2}k_BT$.

Extensions and generalizations

Generalized equipartition theorem

The standard equipartition result can be generalized. For any phase-space variable $x_i$ and Hamiltonian $H$:

$\left\langle x_i \frac{\partial H}{\partial x_j} \right\rangle = \delta_{ij} k_BT$

This reduces to the familiar $\frac{1}{2}k_BT$ per quadratic term as a special case. For a general power-law term $\alpha |q|^n$ in the Hamiltonian, the average energy contribution is $\frac{k_BT}{n}$, not $\frac{1}{2}k_BT$. This allows treatment of anharmonic potentials and other non-quadratic systems.

Non-quadratic Hamiltonians

Real systems often have energy terms that aren't purely quadratic. Anharmonic corrections to molecular vibrations, relativistic kinetic energy, and systems with constraints all fall into this category. The generalized equipartition theorem handles these, though the calculations become more involved. The key point is that the average energy per term depends on the functional form of that term, not just on temperature.

Equipartition in complex systems

Equipartition ideas have been extended beyond traditional physics into areas like biophysics and soft matter. For example, the fluctuations of a polymer chain or a biological membrane can be analyzed using equipartition applied to the normal modes of the system. Each mode's mean-square amplitude is determined by $k_BT$ and the mode's stiffness, providing a way to extract mechanical properties from thermal fluctuation measurements.

Computational methods

Molecular dynamics simulations

Molecular dynamics (MD) simulations integrate Newton's equations of motion for a system of interacting particles. You can directly measure the average kinetic energy per degree of freedom and verify that it equals $\frac{1}{2}k_BT$ once the system equilibrates. MD is particularly useful for testing equipartition in systems too complex for analytical treatment, such as proteins in solution or liquids near phase transitions.

Monte Carlo methods

Monte Carlo simulations use random sampling to explore the configuration space of a system weighted by the Boltzmann factor. They're efficient for computing thermodynamic averages in high-dimensional systems. While they don't track dynamics, they can verify equipartition predictions for equilibrium properties and are especially powerful near phase transitions where analytical methods struggle.

Equipartition in numerical analysis

Equipartition concepts also appear in computational contexts. For instance, in MD simulations, thermostats (like the Nosé-Hoover thermostat) are designed specifically to enforce equipartition across all degrees of freedom. Checking that each degree of freedom carries $\frac{1}{2}k_BT$ of energy is a standard diagnostic for whether a simulation has properly equilibrated.