A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The molecular partition function q is the sum of Boltzmann factors over all molecular energy levels. For an ideal gas, the total partition function factorizes into independent contributions: q = q_trans · q_rot · q_vib · q_elec, because translational, rotational, vibrational, and electronic degrees of freedom are (approximately) independent. Each contribution has a characteristic form: q_trans ∝ V(2πmkT/h²)3/2; q_rot depends on the rotational constants; q_vib = ∏[1−exp(−hν_i/kT)]−1 for harmonic oscillators; q_elec is usually just the ground-state degeneracy unless excited states are thermally accessible. Thermodynamic properties are then obtained as derivatives of ln q.
Evaluate each partition function contribution for a simple diatomic like N₂ at 298 K and 1000 K. Observe how q_trans is enormous (many translational states accessible), q_rot is moderate, and q_vib is close to 1 (vibrational states barely excited at room temperature).
Statistical mechanics connects microscopic quantum energy levels to macroscopic thermodynamic properties through a single central object: the partition function. For a single molecule, the molecular partition function q = Σᵢ exp(−εᵢ/kT) is a weighted count of all accessible quantum states — each state's weight is its Boltzmann factor, which is large for low-energy states and small for high-energy states. If you know q as a function of temperature, you can calculate any thermodynamic property by differentiation: internal energy from ∂(ln q)/∂(1/kT), entropy from T-derivatives of ln q, and so on.
For an ideal gas molecule, the total energy is approximately the sum of independent contributions: translational kinetic energy, rotational energy, vibrational energy, and electronic energy. Because these modes are (approximately) independent, the partition function factorizes: q = q_trans · q_rot · q_vib · q_elec. This is an enormous simplification — instead of summing over every combined quantum state of a molecule with hundreds of modes, you can compute each factor separately and multiply.
Each factor has a characteristic magnitude at room temperature, determined by how the energy level spacing compares to the thermal energy kT ≈ 2.5 kJ/mol at 298 K. Translational energy levels in a macroscopic container are incredibly closely spaced — the spacing is proportional to 1/L², where L is the container size — so kT exceeds the spacing by a factor of roughly 10³⁰, meaning q_trans is enormous. Rotational level spacings are larger (set by molecular moments of inertia), so q_rot is moderate — perhaps 10–100 for a small diatomic. Vibrational level spacings hν are often comparable to or larger than kT, so exp(−hν/kT) ≈ 0 for the first excited vibrational state, and q_vib ≈ [1 − exp(−hν/kT)]⁻¹ ≈ 1. The practical consequence: most molecules at room temperature are in their vibrational ground state, and vibrational modes contribute negligibly to the heat capacity — they are "frozen out."
The distinction between the single-molecule partition function q and the N-molecule partition function Q = qN/N! is subtle but critical. The N! corrects for indistinguishability: quantum mechanics treats identical particles as fundamentally indistinguishable, so swapping two N₂ molecules does not produce a new microstate. Without the N! correction, the calculated entropy is too large — a problem known as the Gibbs paradox, where mixing two samples of the same ideal gas would spuriously increase entropy. The N! also connects to the chemical potential and ensures that the ideal gas entropy obeys all thermodynamic requirements.
Once you have q and its temperature derivative, every thermodynamic property follows analytically. This is the power of the partition function approach: complex macroscopic quantities reduce to calculus on a sum of exponentials, grounded in the quantum energy levels you can calculate or look up in spectroscopic databases.