The equipartition theorem

Result

The equipartition theorem

T-084Home CU-303Threads energy · quantum

Statement

An average of (1/2)kT per quadratic degree of freedom.

Why it matters

boltzmann-populations and molecular-partition-function together provide the fully quantum-statistical machinery for computing a system's average energy from its discrete energy levels, but that full apparatus is more than is needed whenever the relevant energy spacing is small compared to \(kT\), a regime encountered constantly for translational and rotational motion at ordinary temperatures. The equipartition theorem is the simpler, classical-limit shortcut available in exactly that regime: it assigns a fixed average energy of \(\tfrac12 kT\) to every quadratic degree of freedom a molecule possesses, without requiring the full partition-function machinery to be worked through explicitly at all.

It is also the theorem that finally explains, from first principles, the classical heat capacities (\(C_v=\tfrac32 R\) for a monatomic ideal gas, and correspondingly larger values for polyatomic gases) that earlier, purely empirical thermodynamic treatments simply took as measured facts, and it sets up heat-capacity-from-levels' Einstein and Debye models precisely by identifying where and why the classical equipartition prediction eventually breaks down.

Hypotheses

Each degree of freedom's energy can be written as a quadratic function of the corresponding coordinate or momentum (e.g. \(\tfrac12mv_x^2\) for translational motion along \(x\), or \(\tfrac12kx^2\) for a harmonic vibrational displacement).This quadratic form is essential to the theorem's derivation via classical statistical mechanics; a degree of freedom whose energy depends on its coordinate in some other, non-quadratic way would not receive the same universal \(\tfrac12 kT\) contribution, and the theorem simply does not apply to such cases. The system is treated entirely classically, with energy levels spaced closely enough (relative to \(kT\)) that they may be treated as an effectively continuous distribution rather than as discrete quantum states.This is precisely the condition that fails for vibrational modes at ordinary temperatures, whose energy-level spacing is typically large compared to \(kT\) at room temperature, and is the reason equipartition reliably predicts translational and rotational contributions to heat capacity but systematically fails for vibrational contributions except at unusually high temperature (Fails-without logic, developed further in heat-capacity-from-levels). Each degree of freedom's energy is assumed independent of every other degree of freedom (no coupling terms in the total energy expression).Without this independence, the total average energy could not simply be obtained by summing a fixed contribution from each degree of freedom separately; strongly coupled, anharmonic, or otherwise non-separable systems require a more careful treatment beyond the additive form the theorem otherwise provides.

Proof

\langle \varepsilon_i \rangle = \frac{\int \varepsilon_i\, e^{-\varepsilon_i/kT}\,d q_i}{\int e^{-\varepsilon_i/kT}\,d q_i}, \qquad \varepsilon_i = c_i q_i^2

Treating one quadratic degree of freedom \(q_i\) classically, its average energy in thermal equilibrium at temperature \(T\) is given by the Boltzmann-weighted average over all accessible values of \(q_i\), exactly the same classical statistical-mechanical averaging procedure boltzmann-populations applies to discrete states, here carried out as a continuous integral instead. B

\langle \varepsilon_i \rangle = \tfrac12 kT \quad \text{(independent of the specific constant } c_i \text{)}

Evaluating the Gaussian integrals in Step 1 (a standard result for any quadratic exponent) gives exactly \(\tfrac12 kT\) per quadratic degree of freedom, remarkably independent of the specific value of the coefficient \(c_i\) (e.g. independent of a particle's mass for translational motion, or a bond's specific force constant for vibrational motion) — the defining, universal content of the equipartition theorem. B

\langle E \rangle = \tfrac12 N_{\text{quad}}\,kT \quad \text{(per molecule, summing over all quadratic degrees of freedom)}

Since each quadratic degree of freedom contributes independently (Hypotheses) and identically \(\tfrac12kT\) (Step 2), the total average molecular energy is simply \(\tfrac12kT\) multiplied by the total count \(N_{\text{quad}}\) of quadratic degrees of freedom a molecule possesses: three translational (\(\tfrac12mv_x^2\), \(\tfrac12mv_y^2\), \(\tfrac12mv_z^2\)) for any molecule, plus rotational and vibrational contributions depending on molecular geometry and whether vibrations are thermally active (Hypotheses' continuum condition). A

C_v = \frac{\partial \langle E \rangle}{\partial T} = \tfrac12 N_{\text{quad}}\,R \quad \text{(per mole)}

Differentiating the total molar average energy (\(N_A\) times the per-molecule result of Step 3, with \(N_Ak=R\)) with respect to temperature gives the molar heat capacity directly as \(\tfrac12 N_{\text{quad}}R\); for a monatomic ideal gas, with only three translational quadratic degrees of freedom and no rotational or vibrational contribution, this correctly reproduces the well-known classical result \(C_v=\tfrac32R\). A

Result

\langle \varepsilon \rangle = \tfrac12 kT \ \text{per quadratic degree of freedom}, \qquad C_v = \tfrac12 N_{\text{quad}}\,R

Reading. In the classical limit, every independent quadratic contribution to a molecule's energy — translational, rotational, or vibrational — carries, on average, the identical thermal energy \(\tfrac12kT\), regardless of the physical nature of that degree of freedom or the specific mass, moment of inertia, or force constant involved.

Scope. Reliable for translational and (at ordinary temperature) rotational degrees of freedom, whose energy-level spacing is small compared to \(kT\); vibrational degrees of freedom generally require the full quantum treatment (heat-capacity-from-levels) at ordinary temperatures, since their much larger level spacing violates the Hypotheses' classical-continuum condition except at unusually high \(T\).

Corollaries & converses

heat-capacity-from-levels's Einstein and Debye models are needed specifically to correct the equipartition prediction for vibrational modes, showing how heat capacity smoothly interpolates from near-zero (quantum, low-temperature regime, well below equipartition's classical prediction) to the full equipartition value (high-temperature, classical regime) as temperature rises.
molecular-partition-function's full quantum treatment reduces exactly to the equipartition result in the appropriate classical, closely-spaced-level limit, confirming equipartition is not an independent physical law but a limiting case of the more general partition-function framework.
statistical-entropy's \(S=k\ln W\) and the equipartition theorem are both classical-limit consequences of the same underlying statistical-mechanical machinery, applied respectively to counting microstates and to averaging energy over a continuous, quadratic energy landscape.

Fails without

Apply the theorem to a vibrational mode whose level spacing is large compared to \(kT\) (violating the classical-continuum hypothesis): the predicted heat-capacity contribution of a full \(kT\) per mode substantially overestimates the true, measured contribution at ordinary temperature, since most of that vibrational mode's population remains frozen in its ground state.
Drop the independence assumption and treat strongly coupled or anharmonic degrees of freedom as if additive: the simple sum-of-\(\tfrac12kT\)-contributions result (Step 3) no longer holds once cross-terms couple the different coordinates together, and a more careful, non-separable treatment is required instead.

Common errors

Applying equipartition's full classical prediction to vibrational degrees of freedom at room temperature without checking whether the vibrational energy-level spacing is actually small compared to \(kT\) (Hypotheses); for most molecular vibrations at room temperature it is not, and equipartition overestimates the true vibrational contribution to heat capacity substantially.
Assigning \(\tfrac12kT\) per atom rather than per independent quadratic degree of freedom; a polyatomic molecule's translational, rotational, and vibrational degrees of freedom must each be correctly counted and classified before the theorem can be applied (Step 3).
Forgetting that a vibrational degree of freedom, when thermally active, contributes a full \(kT\) (not \(\tfrac12kT\)), since it carries two separate quadratic terms — kinetic and potential energy — each independently contributing \(\tfrac12kT\) by Step 2.
Treating equipartition as an exact, universally valid law of nature rather than the classical, high-temperature (or closely-spaced-level) limit of the more fundamental quantum statistical treatment.

Discussion

The equipartition theorem was developed through the nineteenth century, with major contributions from James Clerk Maxwell and Ludwig Boltzmann, as part of the broader kinetic theory of gases; it successfully explained the observed heat capacities of monatomic and many diatomic gases, but its well-known failure to correctly predict vibrational heat capacity contributions (and, historically, its failure to predict blackbody radiation spectra correctly, the "ultraviolet catastrophe") was among the specific classical puzzles that motivated the development of quantum theory in the early twentieth century.

The theorem's classical failure for vibrational modes was resolved specifically by Albert Einstein's 1907 quantum treatment of solid heat capacities, later refined by Peter Debye; both models show heat capacity is temperature-dependent in a way pure classical equipartition cannot explain, converging to the equipartition prediction only once temperature is high enough that the relevant quantised energy-level spacing again becomes small compared to \(kT\), directly confirming equipartition as a genuine high-temperature limiting case rather than an independently wrong theory.

Common misconception: that equipartition simply "doesn't work" for real molecules and should be discarded in favour of the full quantum treatment entirely. In fact it remains an excellent, routinely used approximation for translational motion at essentially any accessible temperature and for rotational motion at all but the very lowest temperatures; its failure is specifically and narrowly localised to vibrational (and electronic) degrees of freedom under ordinary conditions, not a general breakdown of the whole framework.

Worked examples

\text{Monatomic ideal gas: 3 translational quadratic degrees of freedom, no rotation or vibration}

A monatomic atom has only translational kinetic energy (\(\tfrac12mv_x^2+\tfrac12mv_y^2+\tfrac12mv_z^2\), three quadratic terms) and no rotational or vibrational degrees of freedom at all (a point particle has no internal structure to rotate or vibrate); applying Step 3 with \(N_{\text{quad}}=3\) gives \(\langle E\rangle = \tfrac32kT\) per atom. A

C_v = \tfrac12(3)R = \tfrac32 R \approx 12.5\,\text{J/(mol·K)}

Applying Step 4 with \(N_{\text{quad}}=3\) and \(R=8.314\,\text{J/(mol·K)}\) reproduces the well-established classical, experimentally confirmed heat capacity of a monatomic ideal gas (e.g. helium, argon, and other noble gases at ordinary temperature), a direct, quantitative confirmation of the theorem's validity in this specific, translation-only case. A

C_v(\text{monatomic ideal gas}) = \tfrac32R \approx 12.5\,\text{J/(mol·K)}, \ \text{matches experiment}

Reading. Counting only the three translational quadratic degrees of freedom available to a structureless atom and applying the equipartition result directly reproduces the classical, experimentally well-established monatomic ideal-gas heat capacity exactly.

Scope. The identical counting-and-applying procedure, extended to include rotational (and, where thermally active, vibrational) degrees of freedom, predicts heat capacities for diatomic and polyatomic gases as well.

Problems

A linear diatomic molecule has 3 translational and 2 rotational quadratic degrees of freedom (rotation about the two axes perpendicular to the bond; rotation about the bond axis itself is not a genuine rotational degree of freedom for a linear molecule). Assuming vibration is not thermally active at the temperature of interest, predict \(C_v\).
Solution
Total quadratic degrees of freedom: \(N_{\text{quad}}=3+2=5\). By Step 4, \(C_v=\tfrac12(5)R=\tfrac52R\approx20.8\,\text{J/(mol·K)}\), the standard classical prediction for a diatomic gas well below the temperature at which its vibrational mode becomes thermally active.
Explain why including the diatomic molecule's vibrational mode (2 additional quadratic degrees of freedom: kinetic plus potential energy of the bond stretch) in the calculation of the previous problem would predict \(C_v=\tfrac72R\), and why this prediction is generally not observed at room temperature.
Solution
A thermally active harmonic vibration contributes two quadratic terms (kinetic energy \(\tfrac12\mu\dot{x}^2\) and potential energy \(\tfrac12kx^2\) of the oscillating bond), each independently worth \(\tfrac12kT\) by Step 2, for a full \(kT\) contribution; adding this to the \(5\) already-counted translational and rotational degrees of freedom gives \(N_{\text{quad}}=7\), predicting \(C_v=\tfrac72R\) by Step 4. This prediction generally fails at room temperature because most molecular vibrational energy-level spacings are large compared to \(kT\) at \(298\,\text{K}\), violating the Hypotheses' classical-continuum condition; the vibrational mode is therefore only partially or negligibly thermally active, and the true \(C_v\) typically falls well below the full equipartition prediction of \(\tfrac72R\) at ordinary temperature, converging toward it only as temperature is raised substantially.
Using only the Hypotheses' classical-continuum condition, explain qualitatively why raising a diatomic gas's temperature substantially would be expected to bring its measured \(C_v\) closer to the full \(\tfrac72R\) prediction that includes vibration.
Solution
The classical-continuum condition (Hypotheses) requires the relevant energy-level spacing to be small compared to \(kT\) for equipartition to apply. Raising temperature increases \(kT\) directly, so even though the vibrational energy-level spacing itself does not change, it becomes progressively smaller relative to the now-larger \(kT\); at sufficiently high temperature, \(kT\) can become comparable to or exceed the vibrational spacing, restoring the classical-continuum condition for the vibrational mode as well and bringing the observed \(C_v\) closer to the full equipartition prediction that treats vibration as thermally active.