Mathematical Statistical Mechanics — Mathematics Knowledge System

Overview

Mathematical statistical mechanics provides the rigorous foundations for deriving thermodynamic properties of macroscopic systems from the microscopic laws of classical or quantum mechanics. The subject uses measure theory (Gibbs measures, DLR formalism), ergodic theory (justifying the equality of time averages and ensemble averages), functional analysis (transfer matrices, thermodynamic limit), and probability theory (large deviations, phase transitions). David Ruelle's 1969 monograph established the mathematical framework that is now standard.

Intuition

A gas of 10²³ particles cannot be tracked individually. Statistical mechanics instead describes the probability distribution over all possible microstates. The canonical ensemble assigns each microstate with energy E a probability proportional to e^{−E/kT} (Boltzmann factor). Thermodynamic quantities (pressure, entropy, free energy) are then ensemble averages. Ergodic theory justifies using ensemble averages in place of time averages for real systems.

Formal Definition

Definition

The canonical ensemble is a probability measure on the phase space of a system; thermodynamic quantities are expectations under this measure.

\mu_{\beta}(d\omega) = \frac{e^{-\beta H(\omega)}}{Z(\beta)}\,\lambda(d\omega),\quad Z(\beta) = \int e^{-\beta H(\omega)}\,\lambda(d\omega)

Canonical (Gibbs) measure at inverse temperature beta = 1/(kT)

F(\beta) = -\frac{1}{\beta}\ln Z(\beta)

Helmholtz free energy

S = -k_B \sum_i p_i \ln p_i = -k_B \int \ln\left(\frac{d\mu}{d\lambda}\right) d\mu

Boltzmann-Gibbs entropy

f(\beta) = -\frac{1}{\beta}\lim_{N\to\infty}\frac{1}{N}\ln Z_N(\beta)

Thermodynamic limit of free energy per particle

\langle A \rangle_{\mu} = \int A(\omega)\,\mu_{\beta}(d\omega)

Ensemble average of observable A

Notation

Notation	Meaning
$\beta = 1/(k_B T)$	Inverse temperature
$Z(\beta)$	Partition function
$F = -k_B T \ln Z$	Helmholtz free energy
$\mu_\beta$	Gibbs (canonical) measure
$H$	Hamiltonian (energy function on phase space)

Theorems

Theorem 1: Existence of the Thermodynamic Limit

\lim_{N \to \infty} \frac{1}{N} \ln Z_N(\beta) \text{ exists for systems with stable, tempered interactions}

Theorem 2: Ergodic Theorem (Birkhoff)

\lim_{T \to \infty} \frac{1}{T}\int_0^T A(\phi^t \omega)\,dt = \langle A \rangle_\mu \quad \mu\text{-almost surely}

Theorem 3: Lee-Yang Theorem

\text{The zeros of } Z_N(z) \text{ as a polynomial in the fugacity } z \text{ lie on the unit circle } |z|=1

Theorem 4: DLR (Gibbs) Equilibrium Condition

\mu \text{ is a Gibbs measure iff } \mu(d\omega_\Lambda | \omega_{\Lambda^c}) = \frac{e^{-\beta H_\Lambda(\omega)}}{Z_\Lambda(\omega_{\Lambda^c})}\lambda_\Lambda(d\omega_\Lambda)

Worked Examples

1
Partition function: sum over two microstates.
$Z = e^{-\beta \cdot 0} + e^{-\beta\varepsilon} = 1 + e^{-\beta\varepsilon}$
2
Average energy: E = −∂/∂β ln Z.
$\langle E \rangle = -\frac{\partial}{\partial\beta}\ln Z = -\frac{-\varepsilon e^{-\beta\varepsilon}}{1 + e^{-\beta\varepsilon}} = \frac{\varepsilon}{e^{\beta\varepsilon} + 1}$
3
At high T (β → 0): ⟨E⟩ → ε/2 (equipartition). At low T (β → ∞): ⟨E⟩ → 0 (ground state).
$\langle E \rangle \to \varepsilon/2 \text{ as } T\to\infty$

✓ Answer

⟨E⟩ = ε/(e^{βε} + 1) — the Fermi-Dirac distribution for a two-level fermionic system.

Practice Problems

Hardproof writing

Show that the Helmholtz free energy F = −kT ln Z is related to internal energy U and entropy S by F = U − TS.

Common Mistakes

Common Mistake

The thermodynamic limit is merely taking N → infinity without care

The thermodynamic limit requires simultaneously taking N → ∞ and V → ∞ with N/V = ρ fixed (fixed density). Without stability and tempering conditions on the interaction, the free energy per particle need not converge.

Common Mistake

All systems are ergodic

Ergodicity is a special property — not guaranteed in general. Integrable systems and systems with KAM tori are non-ergodic. Verifying ergodicity rigorously is very difficult and remains open for many physical systems.

Quiz

The canonical ensemble describes a system in thermal contact with a heat bath. Which quantity is fixed?

Energy ETemperature T (or equivalently, inverse temperature β)Entropy SChemical potential μ

The ergodic theorem justifies equating which two quantities?

Kinetic and potential energyTime averages and ensemble (phase-space) averagesEntropy and free energyPartition function and Hamiltonian

Historical Background

Statistical mechanics was developed by Maxwell, Boltzmann, and Gibbs in the 19th century as a bridge between microscopic mechanics and thermodynamics. Gibbs's 1902 'Elementary Principles of Statistical Mechanics' introduced the canonical and grand canonical ensembles. The mathematical rigour came much later: Ruelle and Lanford in the 1960s established the thermodynamic limit rigorously; Dobrushin, Lanford, and Ruelle (DLR) formulated Gibbs measures for infinite-volume systems in 1969–1972; and the Lee–Yang theorem (1952) provided the first rigorous result on phase transitions.

1872
Boltzmann derives the H-theorem and Boltzmann equation for gas dynamics
Ludwig Boltzmann
1902
Gibbs publishes 'Elementary Principles in Statistical Mechanics'
J. Willard Gibbs
1952
Lee and Yang prove theorems on the location of zeros of the partition function (phase transitions)
Tsung-Dao Lee, Chen-Ning Yang
1969
Ruelle publishes 'Statistical Mechanics: Rigorous Results'
David Ruelle
1969
Dobrushin, Lanford, Ruelle formulate the DLR equations for Gibbs measures
Roland Dobrushin, Oscar Lanford, David Ruelle

Summary

The canonical ensemble assigns probability mu_beta proportional to e^{−βH} to each microstate; Z is the partition function normalising this.
Free energy F = −kT ln Z; all thermodynamic quantities follow from F by differentiation.
The thermodynamic limit f = lim_{N→∞} (−1/(βN)) ln Z_N exists for stable, tempered interactions (Ruelle).
Ergodic theory justifies ensemble averages: time average = ensemble average for ergodic systems (Birkhoff).
Phase transitions occur where the free energy is non-analytic; the DLR formalism and Lee-Yang zeros characterise them rigorously.

References

BookRuelle, D. — Statistical Mechanics: Rigorous Results (1969), W.A. Benjamin; reprinted World Scientific 1999
BookGeorgii, H.-O. — Gibbs Measures and Phase Transitions, 2nd ed. (2011), de Gruyter
BookGallavotti, G. — Statistical Mechanics: A Short Treatise (1999), Springer
WebsiteWikipedia — Statistical mechanics
WebsiteMathWorld — Statistical Mechanics