A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The Einstein model treats N atoms as 3N independent harmonic oscillators all with frequency ω_E. Heat capacity C_V = 3Nk (Θ_E/T)2 exp(−Θ_E/T) / [exp(−Θ_E/T)−1]2, where Θ_E = ℏω_E/k. It captures the high-temperature limit C_V = 3R but predicts C_V → 0 too steeply at low T, lacking the T3 behavior of the Debye model.
The puzzle that motivated Einstein in 1907 was the Dulong-Petit law: at room temperature, almost all elemental solids have a molar heat capacity of about 3R ≈ 25 J/(mol·K). Classical statistical mechanics explains this through the equipartition theorem — each atom has 3 kinetic and 3 potential degrees of freedom, each contributing ½kT to the energy, giving 3kT per atom or 3R per mole. But experiments showed that heat capacity drops toward zero as temperature falls. Diamond is particularly dramatic — at room temperature its heat capacity is well below 3R. Classical mechanics had no explanation for this.
Einstein's insight was to apply quantum mechanics to the lattice vibrations. Each atom sits in a potential well created by its neighbors and oscillates — a harmonic oscillator. A classical oscillator can have any energy continuously; a quantum oscillator can only have discrete energies εₙ = (n + ½)ℏω. From your knowledge of the partition function, you can sum the Boltzmann factors over these discrete levels to get the mean energy of one oscillator: ⟨ε⟩ = ℏω/[exp(ℏω/kT) − 1] + ½ℏω. The heat capacity is dU/dT for all 3N oscillators. The Einstein temperature Θ_E = ℏω_E/k sets the scale: when T >> Θ_E, thermal energy easily excites all modes and C_V → 3Nk = 3R (classical limit recovered). When T << Θ_E, the oscillators are "frozen" in their ground states — it costs too much thermal energy to excite the first quantum level, so C_V → 0 exponentially.
The model's success was striking: it explained, for the first time, why diamond has a low heat capacity at room temperature (its high bond stiffness gives a large ω_E and hence a large Θ_E ≈ 1320 K, so room temperature is in the "frozen" regime). But the prediction at very low temperatures is wrong. Experiments find C_V ∝ T³ as T → 0; Einstein's model predicts exponential decay C_V ∝ exp(−Θ_E/T), which falls too steeply. The fault is the assumption that all 3N oscillators vibrate at the same frequency ω_E. Real solids have a spectrum of vibrational frequencies — low-frequency, long-wavelength sound waves (acoustic modes) that remain thermally active at low T and produce the T³ behavior. This is what the Debye model corrects by using a realistic frequency distribution.
The Einstein model is therefore a historically decisive first step: it demonstrated that quantum discreteness was necessary to understand heat capacities, introduced the idea of phonons (quantized lattice vibrations), and recovered the classical Dulong-Petit law as a high-temperature limit — all from a single assumption that each atom is an independent quantum oscillator. Understanding where it fails (the low-T exponential rather than power-law behavior) is itself instructive, because it points directly toward the physics the Debye model must add: the coupling between atoms that gives rise to collective vibrational modes spanning a range of frequencies.