Heat capacity from energy levels

Result

Heat capacity from energy levels

T-086Home CU-303Threads energy · quantum

Statement

The Einstein and Debye models of solids.

Why it matters

molecular-partition-function and boltzmann-populations together give the statistical machinery to compute how molecules distribute themselves across available energy levels and how that distribution encodes bulk thermodynamic quantities; heat capacity from energy levels is the direct payoff of that machinery applied to solids, explaining a real and historically puzzling experimental observation: measured heat capacities of solids, at low temperature, are not the constant value classical physics predicted, but fall toward zero as \(T\to0\). Getting this right was one of the earliest triumphs of quantum statistical mechanics applied outside of spectroscopy.

Hypotheses

The Einstein model treats every atom in the solid as three independent quantum harmonic oscillators, all vibrating at the identical characteristic frequency \(\nu_E\).Real solids have a whole spectrum of vibrational (phonon) frequencies, not one, since neighbouring atoms are coupled to each other rather than vibrating independently; the single-frequency approximation is a deliberate simplification that captures the qualitative low-temperature falloff correctly but gets the precise low-temperature functional form wrong (Fails without/Common errors), a shortcoming fixed by the more elaborate Debye model. equipartition-theorem's classical result, \(\tfrac12kT\) per quadratic degree of freedom, applies only in the high-temperature limit where the thermal energy \(kT\) is large compared to the spacing between quantised vibrational energy levels.At low temperature, \(kT\) becomes comparable to or smaller than the vibrational quantum \(h\nu\), and the discreteness of the levels (rather than their being effectively continuous) becomes the dominant physical effect — exactly the regime equipartition cannot describe, and exactly the regime this result is built to handle correctly.

Proof

\varepsilon_v = \left(v+\tfrac12\right)h\nu_E, \qquad v=0,1,2,\ldots

Each of the \(3N\) independent oscillators (three per atom, \(N\) atoms) has the standard quantised harmonic-oscillator energy-level spacing \(h\nu_E\), a result carried over directly from the quantum treatment of molecular vibration. A

q_{\text{vib}} = \frac{1}{1-e^{-h\nu_E/kT}}, \qquad U-U(0) = 3N\cdot\frac{h\nu_E}{e^{h\nu_E/kT}-1}

Building the single-oscillator vibrational partition function from the level spacing of Step 1 (boltzmann-populations' machinery, summed as a geometric series) and using the standard thermodynamic relation between \(\ln q\) and internal energy gives the mean vibrational energy per oscillator; multiplying by \(3N\) independent oscillators (Hypotheses) gives the total internal energy of the solid above its zero-point value. B

C_V = \left(\frac{\partial U}{\partial T}\right)_V = 3Nk\left(\frac{\theta_E}{T}\right)^2\frac{e^{\theta_E/T}}{\left(e^{\theta_E/T}-1\right)^2}, \qquad \theta_E \equiv \frac{h\nu_E}{k}

Differentiating the internal energy of Step 2 with respect to temperature at constant volume gives the Einstein heat capacity, conventionally written using the Einstein temperature \(\theta_E\) (the characteristic vibrational frequency expressed as an equivalent temperature) as the single parameter fitted to a given solid's actual vibrational stiffness. B

T\gg\theta_E:\ C_V\to3Nk = 3R\ (\text{per mole}); \qquad T\ll\theta_E:\ C_V\to3Nk\left(\frac{\theta_E}{T}\right)^2e^{-\theta_E/T}\to0

Expanding Step 3's expression in the two limiting regimes recovers, at high temperature, exactly the classical equipartition result \(3R\) per mole (the empirical Dulong–Petit law, known long before any quantum explanation existed), while at low temperature the exponential term drives \(C_V\) to zero as \(T\to0\), qualitatively explaining the previously puzzling experimental falloff that classical equipartition could not account for at all. A

Result

C_V = 3R\left(\frac{\theta_E}{T}\right)^2\frac{e^{\theta_E/T}}{\left(e^{\theta_E/T}-1\right)^2}\ \text{(Einstein, per mole)}

Reading. Treating a solid's atoms as quantised oscillators, rather than classical ones, correctly predicts both the classical high-temperature heat capacity (Dulong–Petit's \(3R\)) and a low-temperature falloff toward zero, resolving a genuine failure of classical statistical mechanics.

Scope. The Einstein model's single-frequency approximation (Hypotheses) predicts the low-temperature falloff exponentially rather than as the experimentally observed \(T^3\) power law; the Debye model, using a realistic spectrum of vibrational frequencies up to a cutoff, corrects exactly this low-temperature discrepancy while retaining the same high-temperature Dulong–Petit limit.

Corollaries & converses

equipartition-theorem's \(3R\) high-temperature prediction (Step 4) is recovered here as a limiting case, not a universally valid law — this result explicitly identifies exactly the temperature regime (\(T\gg\theta_E\)) in which the classical equipartition value can be trusted, and the regime in which it cannot.
statistical-entropy's \(S=k\ln W\) framework is the same statistical machinery used implicitly throughout Step 2's partition-function construction; heat capacity and entropy are simply different thermodynamic derivatives of the identical underlying \(q(T)\).
Converse: fitting a measured heat-capacity curve to the Einstein (or, more accurately, Debye) functional form is a standard experimental method for extracting a solid's characteristic vibrational (Einstein or Debye) temperature directly from calorimetric data.

Fails without

Drop the single-frequency assumption and apply the Einstein formula at very low temperature expecting quantitative accuracy: real solids' measured heat capacity follows a \(T^3\) power law at low \(T\), while Step 3's Einstein expression instead falls off exponentially; the qualitative conclusion (\(C_V\to0\)) survives, but the quantitative low-\(T\) prediction does not match experiment (Result, Scope).
Apply the classical equipartition value \(C_V=3R\) (Step 4's high-\(T\) limit) at a temperature well below the solid's characteristic \(\theta_E\): this substantially overpredicts the true heat capacity, since the quantised level spacing (Step 1), not the continuous classical picture, controls the physics in that regime.

Common errors

Assuming \(C_V=3R\) holds at all temperatures rather than only in the high-temperature limit \(T\gg\theta_E\) (Step 4); at low temperature this substantially overpredicts the true, measured heat capacity.
Expecting the Einstein model's low-temperature falloff to match the correct experimental \(T^3\) power law exactly; the Einstein model's single-frequency simplification instead gives an exponential falloff, quantitatively wrong at very low \(T\) even though qualitatively correct in predicting \(C_V\to0\).
Confusing the Einstein temperature \(\theta_E\) with the actual physical vibrational frequency \(\nu_E\) directly; \(\theta_E=h\nu_E/k\) is a temperature-equivalent recasting of that frequency, convenient for comparing directly against the thermodynamic temperature \(T\).

Discussion

Albert Einstein published this model in 1907, one of the earliest applications of quantum theory outside of blackbody radiation and the photoelectric effect, explicitly aiming to explain the experimentally observed decline in solids' heat capacity at low temperature that classical physics (via equipartition) could not account for at all. Peter Debye refined the model in 1912, replacing Einstein's single vibrational frequency with a realistic continuous spectrum of frequencies up to a maximum cutoff, correctly recovering the experimentally observed low-temperature \(T^3\) law while preserving Einstein's high-temperature Dulong–Petit limit.

The empirical Dulong–Petit law itself (that molar heat capacities of many solid elements cluster near \(3R\approx25\,\text{J/mol/K}\) at room temperature) had been known since 1819, decades before any theoretical justification existed; both the Einstein and Debye models are, in this sense, further instances of the historically recurring empirical-regularity-before-full-theoretical-explanation pattern seen elsewhere in this network (Hess's law, the ideal gas law).

Common misconception: that quantum effects on heat capacity only matter for exotic or extremely cold systems. In fact, many solids' Einstein or Debye temperatures lie well above room temperature (diamond's is unusually high, well above \(1000\,\text{K}\)), meaning quantum suppression of heat capacity below the classical \(3R\) value is observable, and chemically relevant, at temperatures far from absolute zero for such materials.

Worked examples

\text{A solid with } \theta_E = 300\,\text{K}, \text{ evaluated at } T=300\,\text{K} \ (T=\theta_E)

Substituting \(\theta_E/T=1\) into Step 3: \(C_V = 3R(1)^2\dfrac{e^1}{(e^1-1)^2} = 3R\times\dfrac{2.718}{(1.718)^2} = 3R\times\dfrac{2.718}{2.952} = 3R\times0.921 \approx 2.76R\), noticeably below the classical \(3R\) limit even at \(T=\theta_E\) exactly. A

\text{Same solid at } T=3\theta_E=900\,\text{K}

Here \(\theta_E/T=1/3\); substituting gives \(C_V=3R(1/3)^2\dfrac{e^{1/3}}{(e^{1/3}-1)^2} = 3R\times0.111\times\dfrac{1.396}{(0.396)^2}=3R\times0.111\times8.90\approx2.96R\), much closer to the classical \(3R\) limit, confirming Step 4's high-temperature convergence. A

C_V(T=\theta_E)\approx0.92\times3R; \qquad C_V(T=3\theta_E)\approx0.99\times3R

Reading. Even at temperatures comparable to or somewhat above a solid's characteristic vibrational temperature, the heat capacity is measurably below the classical Dulong–Petit value, and only asymptotically approaches it as \(T\) rises further above \(\theta_E\).

Scope. Any solid's Einstein temperature can be substituted into the same formula to predict its heat capacity at any given temperature, within the model's stated low-temperature limitations.

Problems

Show that Step 3's Einstein heat-capacity expression reduces to \(C_V\to3R\) in the limit \(T\gg\theta_E\) (i.e. \(\theta_E/T\to0\)), using the small-\(x\) expansion \(e^x-1\approx x\).
Solution
Let \(x=\theta_E/T\to0\). Then \(e^x\to1\) and \(e^x-1\to x\), so \(C_V=3Rx^2\dfrac{e^x}{(e^x-1)^2}\to3Rx^2\dfrac{1}{x^2}=3R\), recovering exactly the classical Dulong–Petit limit, confirming Step 4's high-temperature behaviour directly from the general formula.
Explain qualitatively why diamond, with an unusually high Einstein/Debye temperature (well above room temperature), has a noticeably lower molar heat capacity at room temperature than a typical metal with a much lower characteristic temperature.
Solution
Diamond's very stiff, strongly bonded lattice gives an unusually high vibrational frequency, and hence a high \(\theta_E\) (or Debye temperature). At room temperature, \(T/\theta_E\) is small for diamond (deep in the "low-temperature" regime relative to its own characteristic temperature), so by Step 4 its heat capacity sits well below the classical \(3R\) limit. A typical metal with a much lower \(\theta_E\) has \(T\gg\theta_E\) already at room temperature, placing it much closer to the classical, fully equipartitioned \(3R\) value.
Using the Einstein model, at what value of \(\theta_E/T\) does the model predict \(C_V\) has fallen to exactly half of the classical Dulong–Petit value, \(1.5R\)? Describe (without full numerical solution) how you would find this value.
Solution
This requires solving \(3R\,x^2\dfrac{e^x}{(e^x-1)^2}=1.5R\), i.e. \(x^2\dfrac{e^x}{(e^x-1)^2}=0.5\), for \(x=\theta_E/T\); since this transcendental equation has no simple closed-form solution, it is solved numerically or graphically (e.g. by tabulating the left-hand side over a range of \(x\) and finding where it crosses \(0.5\)), giving a specific numerical value of \(x\) somewhat greater than \(1\) (i.e. \(T\) somewhat below \(\theta_E\)), consistent with Worked Example 1's finding that \(C_V\) is already below \(3R\) at \(T=\theta_E\) itself.