Skip to main content
CBE Nexus · Global University Physics

Unit 12: Statistical Physics

Microscopic foundations of thermodynamics: probability, ensembles, partition functions, quantum statistics, fluctuations, phase transitions and computational methods.

All 36 Physics UnitsPrevious unitNext unit

Unit overview

Statistical physics builds a bridge between microscopic mechanics and macroscopic thermodynamics. It uses probability theory and state counting to explain equilibrium, entropy, heat capacity, radiation, quantum gases, magnetic ordering and transport. The chapter develops the microcanonical, canonical and grand canonical ensembles, derives the major distribution functions, and applies them to solids, electrons, photons, phase transitions and numerical simulations.

Level: Undergraduate university physics. Prerequisites: calculus, probability, classical mechanics, introductory quantum mechanics and thermodynamics. Conventions: kB is Boltzmann's constant, h is Planck's constant, ℏ=h/(2π), β=1/(kBT), and energies are measured in joules unless stated otherwise.

Learning outcomes

Study standard: Always state the ensemble, constraints, particle statistics, approximations and limiting regime before using a distribution or partition function.

1. Statistical physics: scope and purpose

Statistical physics explains macroscopic thermodynamic behaviour from the collective properties of microscopic constituents. A macroscopic sample contains an enormous number of particles, so tracking every trajectory is impractical and unnecessary. Instead, probability distributions, ensembles and averages connect microscopic laws to measurable pressure, temperature, magnetisation, heat capacity and entropy.

The central distinction is between a microstate and a macrostate. A microstate specifies the detailed microscopic configuration, such as particle positions and momenta or quantum occupation numbers. A macrostate specifies a small set of bulk constraints. Many different microstates can correspond to the same macrostate; the multiplicity of those microstates is fundamental to entropy.

Statistical physics is not merely a method of averaging. It provides a quantitative explanation of equilibrium, fluctuations, phase transitions, thermal radiation, electronic properties and the behaviour of matter at low temperatures. It also clarifies when classical approximations fail and quantum statistics become essential.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

2. Microstates, macrostates and multiplicity

For a system with discrete accessible states, a microstate is one complete allowed specification at the microscopic level. A macrostate may be described by energy, volume, particle number, magnetisation or other macroscopic constraints. Multiplicity Ω is the number of microstates compatible with a specified macrostate.

If all accessible microstates are equally likely in an isolated system at equilibrium, the macrostate with the greatest multiplicity is overwhelmingly probable for a large system. This explains why equilibrium is associated with sharply concentrated macroscopic observables, even though microscopic motion continues.

Multiplicity depends on what is held fixed and on whether particles are distinguishable, indistinguishable, localised or free to move. Counting must respect physical constraints and quantum state occupancy. Incorrect counting can produce wrong entropy factors, missing factorials and incorrect thermodynamic predictions.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

3. Probability, distributions and expectation values

A probability distribution assigns a probability to each possible microscopic outcome. Probabilities are non-negative and sum to unity. For discrete states i, the expectation value of an observable A is ⟨A⟩=ΣᵢpᵢAᵢ. For continuous variables, sums are replaced by integrals over a normalised probability density.

The variance σ²=⟨A²⟩−⟨A⟩² measures fluctuations around the mean. A small relative fluctuation means a macroscopic observable is sharply defined. For many independent contributions, relative fluctuations often decrease approximately as N⁻¹/², explaining the stability of bulk measurements.

An ensemble average is a probability-weighted prediction over many hypothetical replicas or repeated preparations. It should not be confused with a time average unless suitable ergodic assumptions apply. Experimental comparison also requires accounting for finite sample size, detector resolution and correlated observations.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

4. The microcanonical ensemble

The microcanonical ensemble describes an isolated system with fixed energy E, volume V and particle number N. Its accessible microstates lie within a narrow energy shell. Under the equal a priori probability postulate, each accessible microstate has probability 1/Ω.

The entropy is S=kB ln Ω. The logarithm makes entropy additive for independent systems because multiplicities multiply: ΩAB=ΩAΩB, so SAB=SA+SB. Boltzmann's constant converts the dimensionless logarithm into SI entropy units of joules per kelvin.

Temperature follows from the entropy derivative: 1/T=(∂S/∂E)V,N. Pressure follows from 1/T times the entropy derivative with respect to volume, and chemical potential is related to the particle-number derivative. These relations connect microscopic counting to thermodynamic state variables.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

5. Boltzmann entropy and the second law

Boltzmann's formula S=kB ln Ω assigns greater entropy to macrostates with greater compatible multiplicity. It gives a statistical interpretation of the thermodynamic entropy introduced in phenomenological thermodynamics. The logarithm is essential for extensivity when independent systems are combined.

The second law is statistical: an isolated system overwhelmingly evolves toward macrostates occupying vastly more microstates. Microscopic equations may be time-reversal invariant, but overwhelmingly probable macroscopic behaviour has a direction because low-multiplicity macrostates occupy an extraordinarily small fraction of accessible phase space.

Entropy can fluctuate downward for small systems over short observation intervals, but large sustained decreases are extraordinarily improbable. Statistical physics therefore reconciles microscopic reversibility with the practical irreversibility of mixing, diffusion and thermal equilibration.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

6. Phase space and Liouville's theorem

Classical phase space contains one coordinate for every independent position and momentum component. A single classical microstate is a point in this high-dimensional space. An ensemble is represented by a density ρ(q,p,t), whose integral over a region gives the probability of finding the system there.

Hamiltonian dynamics preserves phase-space volume. Liouville's theorem states that the phase-space density is constant along a Hamiltonian trajectory, even though the shape of an ensemble distribution may stretch and fold. This property underlies classical statistical mechanics and the invariant phase-space measure.

The phase-space volume associated with a classical state must be made dimensionless when used in state counting. For N particles, a conventional cell scale involving h is introduced, and indistinguishability is handled by a factor N! in the classical partition function.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

7. The canonical ensemble and Boltzmann factor

The canonical ensemble describes a system in thermal contact with a large reservoir at fixed temperature T. The system can exchange energy with the reservoir while its volume and particle number remain fixed. The probability of a microstate of energy Ei is proportional to exp(−βEi), where β=1/(kBT).

Normalisation introduces the canonical partition function Z=Σᵢexp(−βEi). The probability is pi=exp(−βEi)/Z. Low-energy states are more probable, while higher-energy states remain populated according to their Boltzmann suppression.

The Boltzmann factor is derived by considering the multiplicity of the reservoir after the system acquires energy Ei. Expanding the reservoir entropy around its equilibrium energy gives a term linear in Ei, producing exp(−Ei/kBT). This derivation identifies the assumptions of a large reservoir and weak coupling.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

8. Canonical partition function and thermodynamic properties

The partition function contains the equilibrium statistical information for a canonical system. Once Z(T,V,N) is known, the Helmholtz free energy is F=−kBT ln Z. The mean energy is U=−∂ln Z/∂β, with the derivative taken at fixed volume and particle number.

Entropy follows from S=kB(ln Z+βU). Pressure is obtained from p=kBT(∂ln Z/∂V)T,N. Heat capacity follows from the temperature derivative of mean energy and is related to energy fluctuations.

The partition function factorises for independent subsystems because their energies add and the sums over states separate. This makes large systems tractable. Interactions generally prevent simple factorisation and require approximations, numerical methods or more advanced theories.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

9. Energy fluctuations and heat capacity

In the canonical ensemble, the energy variance is ⟨(ΔE)²⟩=⟨E²⟩−⟨E⟩². Differentiating the partition function shows that this variance equals kBT²CV for a system with fixed volume and particle number.

This fluctuation relation gives a microscopic interpretation of heat capacity: a system with large energy fluctuations responds strongly to a temperature change. It also provides a route to calculate heat capacity from simulation data without repeatedly differentiating noisy energy measurements.

For macroscopic systems away from critical points, absolute fluctuations can be large while relative fluctuations are small. Near a critical point, correlation lengths and fluctuations can grow substantially, producing anomalous response functions and critical behaviour.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

10. The grand canonical ensemble

The grand canonical ensemble describes a system that exchanges both energy and particles with a reservoir. Temperature, volume and chemical potential are fixed, while energy and particle number fluctuate. The statistical weight of a state with energy Ei and particle number Ni is proportional to exp[−β(Ei−μNi)].

The grand partition function is Ξ=ΣNΣᵢexp[−β(Ei−μN)]. The grand potential is ΩG=−kBT ln Ξ. The mean particle number is obtained from a derivative with respect to chemical potential, while number fluctuations are linked to particle-number response.

This ensemble is useful for gases, adsorption, electron systems, photon gases and systems where particle number is not fixed. Photons in thermal equilibrium have zero chemical potential because the number of photons is not conserved in ordinary emission and absorption processes.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

11. Chemical potential and particle exchange

Chemical potential μ measures the change in an appropriate thermodynamic potential when one particle is added while the relevant external variables are held fixed. In the grand canonical description, it controls the energetic cost or benefit of particle exchange with a reservoir.

For two systems able to exchange particles, equilibrium requires equality of their chemical potentials. If chemical potentials differ, particle transfer tends to reduce the appropriate total thermodynamic potential. This is the statistical basis of diffusion and equilibrium composition.

The chemical potential is not generally equal to a single-particle energy. It includes the effects of temperature, interactions, density and quantum statistics. For ideal classical gases it can be expressed in terms of density and thermal wavelength.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

12. Classical Maxwell-Boltzmann statistics

Maxwell-Boltzmann statistics apply when quantum wave packets overlap negligibly and occupation of individual quantum states is small. Particles behave classically and quantum exchange effects can be neglected. The occupation probability is approximately proportional to exp[−β(ε−μ)].

The classical regime is often characterised by nλth³ much less than one, where n is number density and λth is the thermal de Broglie wavelength. Low density or high temperature favours this regime.

The classical ideal gas partition function includes a factor 1/N! for indistinguishable particles. This correction resolves the Gibbs paradox and ensures entropy is extensive. Classical statistics remains useful for dilute gases and many engineering applications.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

13. Thermal de Broglie wavelength and quantum degeneracy

The thermal de Broglie wavelength λth=h/√(2πmkBT) estimates the spatial scale of a particle's thermal wave character. It increases as temperature falls and as particle mass decreases.

When the mean interparticle spacing becomes comparable to λth, quantum wavefunctions overlap and exchange symmetry matters. The dimensionless degeneracy parameter nλth³ indicates whether classical statistics is adequate.

Electrons in ordinary metals are quantum-degenerate at room temperature because their density is high and their Fermi temperature is much greater than room temperature. Dilute atomic gases can be driven into quantum degeneracy by cooling and trapping.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

14. Identical particles and quantum state counting

Identical quantum particles cannot be distinguished by labels in the physical description. Their many-particle wavefunctions must be symmetric under exchange for bosons and antisymmetric for fermions. This exchange symmetry changes the allowed occupation patterns.

Bosons can share a single quantum state without restriction, whereas fermions obey the Pauli exclusion principle and each one-particle state can hold at most one fermion per spin state. These rules generate Bose-Einstein and Fermi-Dirac statistics.

The distinction is not a small correction in degenerate systems. It explains blackbody radiation, the stability of matter, electron degeneracy pressure, superfluidity, superconductivity and the electronic properties of solids.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

15. Bose-Einstein distribution

For bosons in thermal equilibrium, the mean occupation of a one-particle state of energy ε is n̄(ε)=1/[exp(β(ε−μ))−1]. The chemical potential cannot exceed the lowest single-particle energy in a stable ideal gas.

The denominator contains a minus one because multiple bosons may occupy the same state. At high temperature and low occupation, the distribution approaches the Maxwell-Boltzmann form.

Bose-Einstein statistics describe photons, phonons and integer-spin atoms. For an ideal uniform three-dimensional gas, cooling below a critical temperature produces macroscopic occupation of the ground state, known as Bose-Einstein condensation.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

16. Fermi-Dirac distribution and Pauli exclusion

For fermions, the mean occupation is n̄(ε)=1/[exp(β(ε−μ))+1]. The plus one follows from the restriction that a fermionic quantum state can be occupied only once per allowed spin state.

At zero temperature, states below the Fermi energy are occupied and states above it are empty for an ideal ground-state Fermi gas. At finite temperature, the sharp step is broadened over an energy scale of order kBT.

Fermi-Dirac statistics explain electron populations in metals and semiconductors, white-dwarf electron degeneracy pressure and neutron-star matter. The Fermi energy, Fermi momentum and density of states are central quantities in these applications.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

17. Deriving quantum distributions from grand partition functions

For a single bosonic state of energy ε, allowed occupation numbers are 0,1,2 and so on. Its grand partition function is a geometric series: Ξε=Σn exp[−βn(ε−μ)]=1/[1−exp(−β(ε−μ))]. Differentiating ln Ξε gives the Bose mean occupation.

For a fermionic state, occupation is restricted to zero or one. The single-state grand partition function is Ξε=1+exp[−β(ε−μ)]. Differentiation gives the Fermi mean occupation with the plus sign in the denominator.

These derivations show that the difference between Bose and Fermi distributions arises from allowed occupation numbers, not from arbitrary curve fitting. The classical distribution is recovered when exp[β(ε−μ)] is very large.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

18. Density of states in three dimensions

The density of states g(ε) counts the number of one-particle states per energy interval. For free particles in a three-dimensional box, states correspond to allowed wavevectors in reciprocal space. Counting wavevector-space cells gives a density proportional to the square root of energy.

For non-relativistic free particles, ε=ℏ²k²/(2m), so the three-dimensional density of states per unit volume is proportional to ε¹/². Spin degeneracy must be included consistently.

The density of states converts sums over energy levels into integrals. It is essential for calculating internal energy, particle number, heat capacity, electronic transport and photon or phonon populations.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

19. Classical ideal gas partition function

For a monatomic ideal gas, the single-particle translational partition function is q=V/λth³. For N indistinguishable non-interacting particles, ZN=qN/N!. The factorial prevents overcounting permutations of identical particles.

Using Stirling's approximation ln N!≈N ln N−N for large N gives the Helmholtz free energy and entropy. Differentiation recovers the ideal gas equation of state pV=NkBT and the Sackur-Tetrode entropy formula.

The classical partition function assumes dilute conditions, negligible interactions and a continuum approximation to translational energy levels. It fails when quantum degeneracy becomes important or when intermolecular interactions strongly affect the equation of state.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

20. Equipartition theorem and its limitations

In classical thermal equilibrium, each independent quadratic term in the energy contributes an average kBT/2. A monatomic ideal gas has three translational quadratic degrees of freedom, giving mean energy 3kBT/2 per particle.

Rotational and vibrational degrees of freedom contribute when they are thermally accessible. A classical diatomic molecule at moderate temperature has three translational and two rotational quadratic degrees of freedom; vibrational modes may be frozen at ordinary temperatures.

Equipartition fails for quantum degrees of freedom when level spacing is large compared with kBT. This explains why classical predictions of heat capacities can fail at low temperatures and why quantum statistical mechanics is needed.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

21. Quantum harmonic oscillator

The quantum harmonic oscillator has energy levels En=ℏω(n+1/2), where n is a non-negative integer. The zero-point term ℏω/2 remains even in the ground state and has no classical analogue.

Its partition function is Z=exp(−βℏω/2)/[1−exp(−βℏω)]. Differentiation gives mean energy (ℏω/2)coth(βℏω/2). At high temperature the mean energy approaches kBT; at low temperature it approaches the zero-point energy.

The oscillator model describes molecular vibrations, lattice phonons and quantised field modes. The temperature dependence of its heat capacity demonstrates quantum freezing of degrees of freedom.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

22. Blackbody radiation and photon gas

A blackbody absorbs and emits radiation in thermal equilibrium. Quantising electromagnetic modes and applying Bose-Einstein statistics with zero photon chemical potential produces Planck's spectral distribution.

Integrating Planck's law over frequency gives the Stefan-Boltzmann energy density proportional to T⁴. The spectral maximum shifts inversely with temperature, described by Wien's displacement law.

The photon gas has pressure u/3, where u is energy density, and its entropy and energy follow from the temperature dependence of radiation modes. These results are important in astrophysics, thermal radiation and early-universe cosmology.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

23. Phonons and lattice heat capacity

A crystal lattice supports collective vibrational normal modes. Quantisation of these modes produces phonons, which are bosonic quasiparticles with energy ℏω. Phonons transport heat and influence thermal expansion, sound propagation and electrical resistance.

The Einstein model treats atoms as independent oscillators with one characteristic frequency. It captures low-temperature suppression qualitatively but does not reproduce the full low-frequency mode spectrum of a real solid.

The Debye model approximates the phonon spectrum by a continuum up to a cutoff frequency chosen to give the correct number of modes. At low temperature it predicts CV proportional to T³; at high temperature it approaches the Dulong-Petit limit.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

24. Fermi gas and Fermi energy

A free Fermi gas fills the lowest available momentum states consistent with the Pauli principle. At zero temperature, all states are occupied up to the Fermi wavevector kF. For a three-dimensional spin-½ gas, kF=(3π²n)¹/³.

The Fermi energy EF=ℏ²kF²/(2m) sets a characteristic energy scale and TF=EF/kB is the Fermi temperature. If T is much smaller than TF, the gas is strongly degenerate even when the temperature seems high on an everyday scale.

The Fermi energy determines electron populations in metals and contributes to electronic pressure. It also sets the scale for low-temperature heat capacity and helps explain why only electrons near the Fermi surface participate strongly in many thermal and transport processes.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

25. Low-temperature heat capacity of solids

At sufficiently low temperature, the Debye phonon contribution to heat capacity varies as T³. In a metal, conduction electrons add a term approximately proportional to T. Thus CV≈γT+βT³ over an appropriate low-temperature interval.

A plot of CV/T against T² can separate the electronic coefficient γ from the phonon coefficient β. This method provides information about the electronic density of states and lattice vibrations.

Real materials may show additional contributions from magnetic ordering, nuclear spins, superconducting transitions or impurities. Experimental interpretation therefore requires checking the temperature range and material phase.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

26. Paramagnetism and spin statistics

A paramagnet contains magnetic moments that tend to align with an applied field but are opposed by thermal disorder. For independent spin moments in a field, the single-particle partition function sums the Boltzmann weights of the allowed magnetic orientations.

The magnetisation follows from the derivative of free energy with respect to magnetic field. At weak field and high temperature, susceptibility often follows Curie's law, χ proportional to 1/T.

At stronger fields or lower temperatures, magnetisation approaches saturation. Interactions among moments can produce cooperative magnetic ordering, which lies beyond the independent-spin model and may require mean-field or microscopic lattice approaches.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

27. Fluctuations, correlations and response

Fluctuations describe deviations from equilibrium averages. Correlations quantify how changes in one degree of freedom are statistically related to another. In equilibrium systems, response functions are connected to fluctuations through fluctuation-response relations.

For example, energy fluctuations determine heat capacity in the canonical ensemble, while particle-number fluctuations determine compressibility-related response in the grand canonical ensemble. Magnetic susceptibility is similarly linked to magnetisation fluctuations.

Near critical points, long-range correlations can develop and fluctuations become large. Finite-size effects, slow relaxation and critical slowing down complicate both simulations and laboratory measurements.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

28. Phase transitions and order parameters

A phase transition is a qualitative change in macroscopic behaviour as control parameters such as temperature, pressure or field vary. An order parameter distinguishes phases, such as magnetisation in a ferromagnet or density difference between coexisting liquid and gas phases.

First-order transitions involve latent heat and discontinuities in first derivatives of free energy. Continuous transitions show continuous first derivatives but singular or divergent response functions in the thermodynamic limit.

Statistical models explain transitions through collective correlations and symmetry. Finite systems show rounded transitions rather than exact thermodynamic singularities, so numerical and experimental analyses must account for system size.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

29. Ising model and cooperative ordering

The Ising model represents spins on a lattice with two possible orientations. A common Hamiltonian is H=−JΣ⟨ij⟩sᵢsⱼ−BΣᵢsᵢ, where J describes nearest-neighbour coupling and B represents an applied field in suitable units.

For J>0, neighbouring spins favour alignment. At high temperature thermal disorder dominates; below a critical temperature in dimensions that support ordering, spontaneous magnetisation can occur even when the applied field is zero.

The model is a foundation for understanding ferromagnetism, critical phenomena, universality and Monte Carlo methods. Despite its simplicity, it exhibits collective behaviour that cannot be understood by considering isolated spins alone.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

30. Monte Carlo methods in statistical physics

Monte Carlo simulation estimates ensemble averages by sampling configurations according to a probability distribution. In the Metropolis algorithm, a proposed change with energy difference ΔE is accepted with probability min[1,exp(−βΔE)].

A Markov chain is constructed so that its stationary distribution is the desired equilibrium distribution. Detailed balance is a common sufficient condition, although ergodicity and adequate mixing must also be considered.

Reliable simulations require equilibration, independent or decorrelated samples, uncertainty estimates, finite-size analysis and checks for critical slowing down. A long run does not guarantee a good sample if the chain remains trapped in a metastable region.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

31. Molecular dynamics and statistical ensembles

Molecular dynamics integrates classical equations of motion for many interacting particles. Trajectories provide time-dependent positions, velocities and microscopic observables. Suitable averaging can estimate equilibrium and transport properties.

A microcanonical molecular-dynamics simulation conserves total energy, while thermostats can sample canonical-like distributions. Barostats can allow volume fluctuations and sample pressure-controlled conditions, but algorithms must be chosen and validated carefully.

Time-step convergence, force accuracy, boundary conditions, equilibration and finite-size effects are essential. Molecular dynamics is powerful for liquids, materials and biomolecules but remains limited by force-field quality and accessible time and length scales.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

32. Virial expansion and non-ideal gases

The virial equation expresses deviations from ideal-gas behaviour as a density expansion, for example p/(kBT)=n+B₂(T)n²+B₃(T)n³+…. The coefficients encode intermolecular interactions.

The second virial coefficient describes leading pair-interaction corrections in a dilute gas. Its temperature dependence can be related to the intermolecular potential through an integral over separation and the Boltzmann factor.

Virial expansions are useful at low and moderate densities but may converge poorly near condensation or criticality. Equation-of-state selection should be based on the phase region and required accuracy.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

33. Information theory and entropy

Shannon information entropy H=−Σipi ln pi quantifies uncertainty in a probability distribution. Multiplying by kB gives an entropy with thermodynamic units when the probabilities represent physical microstates under appropriate assumptions.

The maximum entropy principle selects the least biased distribution consistent with specified constraints. Maximising entropy with normalisation and mean-energy constraints yields the canonical Boltzmann distribution.

Information entropy and thermodynamic entropy are closely related but should not be conflated without specifying the physical interpretation. The former is dimensionless and depends on a probability model; the latter is a measurable state function.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

34. Ergodicity and equilibrium averages

A system is ergodic in a relevant region if a sufficiently long trajectory explores the accessible phase space in a way that time averages agree with ensemble averages. This is a powerful assumption in equilibrium statistical mechanics.

Real systems may exhibit slow relaxation, conserved quantities, broken ergodicity or metastable trapping. Glasses and complex materials can fail to reach equilibrium on experimental timescales, even though an equilibrium ensemble can be formally defined.

When using time averages, compare independent initial conditions, examine autocorrelation times and verify that observables have stabilised. Ensemble equivalence is most reliable for large systems away from pathological constraints and phase coexistence.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

35. Ensemble equivalence and thermodynamic limit

Microcanonical, canonical and grand canonical ensembles can yield the same bulk thermodynamic predictions in the thermodynamic limit for many short-range systems. The limit takes particle number, volume and energy large while holding densities fixed.

Finite systems can show differences among ensembles because energy and particle-number fluctuations are not negligible. Long-range interactions, non-additive systems and phase coexistence can also complicate equivalence.

Choosing an ensemble is therefore a modelling decision based on physical constraints. An isolated sample suggests microcanonical conditions; a sample in a heat bath suggests canonical conditions; particle exchange suggests grand canonical conditions.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

36. Thermodynamic limit, extensivity and finite size

The thermodynamic limit sends N and V to infinity at fixed density. Extensive quantities such as energy and entropy scale with system size, while intensive quantities such as temperature and pressure remain finite.

Surface contributions become negligible relative to bulk contributions for large systems, but they matter in nanoparticles, thin films, droplets and finite simulation boxes. Critical phenomena are especially sensitive to finite-size scaling.

Finite-size scaling studies how observables change with system size near criticality. It allows estimates of critical exponents and transition temperatures from simulations that cannot reach the infinite-system limit.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

37. Statistical physics of transport and diffusion

Equilibrium statistical mechanics describes static distributions, while transport theory studies how systems relax and carry heat, momentum and particles. Diffusion, viscosity and thermal conductivity depend on microscopic motion and correlations.

The Einstein relation connects mean-square displacement to the diffusion coefficient: in d dimensions, ⟨r²(t)⟩≈2dDt at sufficiently long times for normal diffusion. Deviations can signal confinement, anomalous transport or insufficient equilibration.

Green-Kubo relations express transport coefficients as time integrals of equilibrium current autocorrelation functions. These relations provide a bridge between microscopic dynamics and macroscopic transport laws.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

38. Critical phenomena and universality

Near a continuous phase transition, correlation lengths can become very large and thermodynamic response functions may follow power laws. Critical exponents characterise these singular behaviours.

Different microscopic systems can share the same critical exponents when they have the same dimensionality, symmetry and range of interactions. This is universality and explains why simple models can describe diverse materials.

Experiments and simulations test scaling relations, collapse data across system sizes and distinguish genuine critical behaviour from crossover effects, finite-size rounding and measurement artefacts.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

39. Applications in astrophysics and cosmology

Statistical physics describes stellar interiors, degenerate matter, radiation fields and the early universe. Fermi pressure helps support compact stars, while photon and neutrino distributions influence thermal histories and energy transport.

The cosmic microwave background is close to a blackbody spectrum, providing a precise record of an early hot radiation-dominated universe. Statistical distributions also describe particle populations in astrophysical plasmas.

Astrophysical applications often require relativistic particles, strong gravity, non-equilibrium radiation and nuclear reactions. The correct statistical ensemble and equation of state depend on the physical timescales and conservation laws.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

40. Experimental methods and statistical data analysis

Statistical physics experiments may measure heat capacity, magnetic susceptibility, electrical resistance, thermal radiation, particle distributions or phase-transition signatures. Each observable must be connected to a model through a calibrated measurement chain.

A useful experimental plan identifies independent variables, controlled conditions, repeat measurements, instrument calibration, environmental effects and a predicted scaling law. Plots should include units, uncertainty bars and residual analysis where appropriate.

Data analysis should compare alternative models, estimate parameter uncertainty and report limitations. Correlated noise, drift, hysteresis and finite sample effects can be more important than simple counting statistics.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

41. Research practice, reproducibility and model validation

A statistical-physics model is defined by its degrees of freedom, Hamiltonian or energy function, constraints, probability measure and approximations. A reproducible report states these choices explicitly, along with numerical methods and parameter values.

Validation compares model predictions with limiting cases, known analytical results, independent simulations or experimental data. Verification checks whether the equations and algorithms have been implemented correctly.

Research quality requires transparent data handling, versioned code, uncertainty estimates, sensitivity analysis and clear distinction between numerical evidence and physical interpretation. A visually smooth simulation is not proof of a correct model.

Mathematical method. Begin with the allowed microscopic states and the constraints imposed by the physical preparation. Define the probability assigned to each state, normalise it, and then calculate the expectation value of the observable of interest. Keep track of whether a sum is over distinguishable particles, indistinguishable particles, one-particle states or many-particle configurations. These are different counting problems and must not be interchanged.

Physical interpretation and checks. Test the expression in the high-temperature, low-density, weak-field or large-system limit when those limits are meaningful. Confirm that probabilities are normalised, mean energies lie within allowed bounds, and dimensions are consistent. A formula derived for a classical dilute gas should not be applied automatically to a degenerate electron system or a low-temperature quantum fluid.

Application and evidence. In an experiment or simulation, identify the observable corresponding to the theoretical quantity, the ensemble most closely approximated by the apparatus, and the relevant time and length scales. State how the data are averaged, how correlated observations are treated, and how finite-size or finite-time effects influence the result. Theoretical predictions become useful only when these operational details are addressed.

Common examination issue. Students often quote a distribution without explaining its assumptions, confuse the chemical potential with a particle energy, or omit state degeneracy from a partition sum. Define every symbol, state whether spin degeneracy is included, and distinguish a single-particle partition function from a many-particle partition function. This makes the derivation reproducible and prevents hidden factors from entering numerical answers.

Key mathematical derivations

The following derivations connect the probability postulates to measurable thermodynamic quantities. Work through each algebraic step and identify the physical assumptions, rather than memorising the final expression.

Derivation: canonical probability from a heat reservoir

Consider a small system S in weak thermal contact with a much larger reservoir R. The combined isolated system has fixed total energy Etotal. If the small system occupies a microstate i with energy Ei, the reservoir has energy Etotal−Ei. The probability of i is proportional to the number of reservoir states available at that energy: pi∝ΩR(Etotal−Ei). Using Boltzmann entropy, lnΩR=SR/kB, and expanding about Etotal gives SR(Etotal−Ei)≈SR(Etotal)−Ei(∂SR/∂E)Etotal. Since (∂S/∂E)=1/T, the result is pi∝exp[−Ei/(kBT)]. Normalising over all system states gives pi=exp(−βEi)/Z, with Z=Σi exp(−βEi). The derivation assumes a reservoir much larger than the system, weak coupling and a well-defined reservoir temperature.

Derivation: mean energy and heat capacity from the partition function

Start with Z=Σi exp(−βEi). Differentiating gives ∂Z/∂β=−ΣiEi exp(−βEi). Dividing by Z yields ∂lnZ/∂β=−⟨E⟩, so U=−∂lnZ/∂β. A second derivative gives ∂²lnZ/∂β²=⟨E²⟩−⟨E⟩². Because β=1/(kBT), differentiating U with respect to T gives CV=(⟨E²⟩−⟨E⟩²)/(kBT²). Thus CV is non-negative for a canonical system with ordinary stable conditions and is directly connected to the variance of energy. This result is both a theoretical identity and a practical simulation diagnostic.

Derivation: Bose-Einstein occupation

For one bosonic mode with energy ε, any occupation n=0,1,2,… is allowed. In the grand canonical ensemble the weight of n particles is exp[−βn(ε−μ)]. Let x=exp[−β(ε−μ)]. The single-mode grand partition function is Ξε=1+x+x²+…=1/(1−x), for x<1. The mean occupation is n̄=x∂lnΞε/∂x=x/(1−x). Substituting x gives n̄=1/[exp(β(ε−μ))−1]. The result depends on unrestricted bosonic occupancy and requires μ below the lowest allowed energy for the ideal gas.

Derivation: Fermi-Dirac occupation

For a fermionic one-particle state, Pauli exclusion allows n=0 or n=1 only. The single-state grand partition function is Ξε=1+exp[−β(ε−μ)]. The mean occupation is obtained by differentiating lnΞε with respect to βμ, giving n̄=exp[−β(ε−μ)]/[1+exp(−β(ε−μ))]. Multiplying numerator and denominator by exp[β(ε−μ)] gives n̄=1/[exp(β(ε−μ))+1]. The plus sign follows directly from the two allowed occupancy values.

Derivation: density of states for free particles

In a cubic box of side L with periodic boundary conditions, allowed wavevectors are spaced by 2π/L. Each state occupies volume (2π/L)³ in k-space. The number of states with wavevector magnitude below k is the volume of a sphere, 4πk³/3, divided by the cell volume, multiplied by any spin degeneracy. Differentiating with respect to k and using ε=ℏ²k²/(2m) converts the count into g(ε). In three dimensions g(ε) per unit volume is proportional to ε¹/². The exact prefactor depends on whether spin degeneracy is included.

Derivation: Fermi energy of a three-dimensional gas

At zero temperature, fill all states up to kF. For spin degeneracy two, particle density is n=2×[volume of k-space sphere]/[(2π)³]=kF³/(3π²). Therefore kF=(3π²n)¹/³. The highest occupied energy is EF=ℏ²kF²/(2m). This shows EF increases with density as n²/³ and provides the Fermi temperature TF=EF/kB. The result assumes a uniform, non-interacting, non-relativistic spin-½ gas.

Derivation: classical monatomic ideal-gas partition function

The one-particle translational partition function is q=V/h³∫exp[−βp²/(2m)]d³p. Evaluating the three Gaussian momentum integrals gives q=V(2πmkBT)³/²/h³=V/λth³. For N identical non-interacting particles, ZN=qN/N!. The N! removes overcounting of permutations. From F=−kBT lnZN and Stirling's approximation, thermodynamic derivatives recover pV=NkBT and the classical ideal-gas entropy.

Derivation: equipartition theorem

For a canonical distribution, the probability density is proportional to exp(−βH). Consider one quadratic contribution Hx=ax². The average is ⟨ax²⟩=∫ax²e−βax²dx/∫e−βax²dx. Differentiating the Gaussian integral with respect to βa gives ⟨ax²⟩=1/(2β)=kBT/2. Each independent quadratic degree contributes this amount. The theorem assumes classical continuous degrees of freedom and fails when quantum energy levels are not closely spaced compared with kBT.

Derivation: Planck distribution

Electromagnetic radiation in a cavity can be decomposed into independent normal modes. Each mode is a quantum harmonic oscillator with energies nℏω when the zero-point term is omitted from thermal occupation. Photons are bosons and their number is not conserved, so μ=0. The mean occupation is n̄=1/[exp(ℏω/kBT)−1]. Multiplying by photon energy ℏω and the electromagnetic mode density gives the spectral energy density. The result tends to the Rayleigh-Jeans law at low frequency and decays exponentially at high frequency.

Derivation: Debye low-temperature heat capacity

The Debye model approximates acoustic phonon modes with a linear dispersion relation ω=vk up to a cutoff ωD. The density of modes is proportional to ω². Internal energy is an integral of ℏω times the Bose occupation over frequency. At low temperature, the upper cutoff can be extended to infinity after changing variable x=ℏω/kBT. The integral scales as T⁴, so differentiating with respect to temperature gives CV proportional to T³. This is the Debye law for a three-dimensional insulating crystal at sufficiently low temperature.

Derivation: Einstein relation for diffusion

For a random walk with independent zero-mean displacement increments, the mean-square displacement is the sum of the variances of the increments because cross terms average to zero. At long times, the number of increments is proportional to time, so ⟨r²⟩ grows linearly with t. Isotropic diffusion in d dimensions obeys ∂P/∂t=D∇²P; multiplying by r² and integrating over space gives d⟨r²⟩/dt=2dD. Integrating from zero displacement gives ⟨r²⟩=2dDt.

Derivation: entropy and multiplicity for independent systems

If systems A and B are independent, each microstate of A can be paired with each microstate of B. Thus ΩAB=ΩAΩB. Applying Boltzmann's definition gives SAB=kB ln(ΩAΩB)=kB lnΩA+kB lnΩB=SA+SB. The logarithmic form is therefore required for additive entropy under independent composition.

Worked numerical examples

Use SI units and retain constants to a sensible number of significant figures. These examples illustrate how microscopic statistical expressions produce measurable macroscopic predictions.

Worked example 1: Boltzmann multiplicity

A macrostate has Ω=10²⁰ accessible microstates. Its Boltzmann entropy is S=kB ln Ω=(1.380649×10⁻²³)ln(10²⁰)=6.36×10⁻²² J K⁻¹.

Worked example 2: Canonical two-level system

A particle has energies 0 and ε. Z=1+e⁻ᵝᵋ. The excited-state probability is e⁻ᵝᵋ/(1+e⁻ᵝᵋ)=1/(eᵝᵋ+1).

Worked example 3: Mean energy from partition function

For Z=1+e⁻ᵝᵋ, U=−∂lnZ/∂β=ε/(eᵝᵋ+1). The result tends to zero at low T and ε/2 at high T.

Worked example 4: Entropy of a two-state system

For probabilities p and 1−p, S=−kB[p ln p+(1−p)ln(1−p)]. Entropy is largest at p=1/2.

Worked example 5: Thermal wavelength

For an atom of mass 4.0×10⁻²⁶ kg at 300 K, λth=h/√(2πmkBT)≈1.30×10⁻¹¹ m. Compare this with mean particle spacing to test classical validity.

Worked example 6: Fermi energy estimate

For electron density n=8.5×10²⁸ m⁻³, kF=(3π²n)¹/³≈1.36×10¹⁰ m⁻¹. EF=ℏ²kF²/(2me)≈7.1 eV.

Worked example 7: Photon energy and occupation

At frequency 5.0×10¹⁴ Hz and T=3000 K, x=hν/kBT≈8.0. The mean photon occupation 1/(eˣ−1)≈3.4×10⁻⁴.

Worked example 8: Equipartition energy

A classical monatomic gas has three translational quadratic degrees of freedom. Mean translational energy per particle is 3kBT/2; for one mole it is 3RT/2.

Worked example 9: Harmonic oscillator partition function

For levels ℏω(n+1/2), Z=e⁻ᵝℏω/²/(1−e⁻ᵝℏω). The mean energy is ℏω/2+ℏω/(eᵝℏω−1).

Worked example 10: Canonical energy fluctuations

If CV=12 J K⁻¹ at T=300 K, variance is kBT²CV=1.49×10⁻¹⁷ J². The standard deviation is 3.86×10⁻⁹ J.

Worked example 11: Grand-canonical mean occupancy

For a fermionic state with ε−μ=0, mean occupancy is 1/[e⁰+1]=1/2 at any finite temperature.

Worked example 12: Debye low-temperature scaling

If a Debye solid has CV=βT³ with β=2.0×10⁻³ J mol⁻¹ K⁻⁴, at 5 K CV=0.25 J mol⁻¹ K⁻¹.

Worked example 13: Einstein solid high-temperature limit

For 3N independent oscillators, the high-temperature heat capacity tends to 3NkB, or 3R per mole, the Dulong-Petit value.

Worked example 14: Diffusion from mean-square displacement

In two dimensions, a particle has ⟨r²⟩=4Dt. If the mean-square displacement is 0.008 m² after 20 s, D=1.0×10⁻⁴ m²/s.

Worked example 15: Curie-law comparison

If χ=C/T, doubling absolute temperature halves susceptibility, provided the material remains in the Curie-law regime and interactions are negligible.

Worked example 16: Entropy from equiprobable states

For 100 equally likely microstates, S=kB ln100=6.36×10⁻²³ J K⁻¹.

Worked example 17: Bose occupation

For β(ε−μ)=2, n̄=1/(e²−1)=0.156 for a bosonic mode.

Worked example 18: Fermi occupation

For β(ε−μ)=2, n̄=1/(e²+1)=0.119 for a fermionic state.

Worked example 19: Ideal-gas partition function

For one particle in volume V, q=V/λth³. Increasing V at fixed T increases the number of accessible translational states in direct proportion to V.

Worked example 20: Fermi temperature

If EF=5.0 eV, TF=EF/kB≈5.80×10⁴ K. At room temperature T/TF≈0.0052, so the electron gas is strongly degenerate.

University examination questions and model answers

These original questions cover conceptual reasoning, derivations, calculations and applications. They are independent practice materials, not official examination papers.

Question 1. Define microstate and macrostate.

Model answer: A microstate specifies a complete microscopic configuration. A macrostate specifies bulk constraints such as energy, volume, particle number or magnetisation. Many microstates can correspond to one macrostate.

Question 2. State Boltzmann's entropy formula and explain it.

Model answer: S=kB ln Ω, where Ω is the number of accessible microstates compatible with the macrostate. The logarithm makes entropy additive for independent systems.

Question 3. What is an ensemble?

Model answer: An ensemble is a statistical collection of hypothetical systems prepared under the same macroscopic constraints but occupying different microstates with probabilities determined by the ensemble.

Question 4. Describe the microcanonical ensemble.

Model answer: It models an isolated system with fixed energy, volume and particle number. Accessible states in a narrow energy shell are assigned equal probability.

Question 5. Derive the canonical Boltzmann factor.

Model answer: A system state of energy E leaves the reservoir with energy ER−E. Expanding reservoir entropy gives SR(E−E)≈SR(ER)−E/T; reservoir multiplicity is therefore proportional to exp(−E/kBT).

Question 6. Define the canonical partition function.

Model answer: Z=Σi exp(−βEi), where β=1/(kBT). It normalises canonical probabilities and generates thermodynamic quantities.

Question 7. Show how to obtain mean energy from Z.

Model answer: U=ΣipiEi=−∂lnZ/∂β, at fixed volume and particle number. Differentiation weights states by their energies.

Question 8. Explain the relation between energy fluctuations and heat capacity.

Model answer: In the canonical ensemble, variance of energy equals kBT²CV. Large heat capacity corresponds to larger equilibrium energy fluctuations.

Question 9. What is the grand canonical ensemble?

Model answer: It describes a system that exchanges energy and particles with reservoirs. T, V and μ are fixed; energy and particle number fluctuate.

Question 10. Define chemical potential.

Model answer: Chemical potential is the change in an appropriate thermodynamic potential when particle number changes at fixed natural variables. It controls particle exchange and diffusion.

Question 11. When is Maxwell-Boltzmann statistics valid?

Model answer: When the gas is dilute or hot enough that quantum wavefunctions overlap negligibly and occupation of each quantum state is small; nλth³≪1.

Question 12. Explain thermal de Broglie wavelength.

Model answer: λth=h/√(2πmkBT) estimates the quantum spatial extent of a particle at temperature T. Quantum degeneracy matters when this approaches the mean particle spacing.

Question 13. State the difference between bosons and fermions.

Model answer: Bosons have symmetric many-particle wavefunctions and can share a state. Fermions have antisymmetric wavefunctions and obey Pauli exclusion.

Question 14. Write the Bose-Einstein distribution and explain the minus sign.

Model answer: n̄=1/[exp(β(ε−μ))−1]. The minus sign results from summing all non-negative occupation numbers for a bosonic state.

Question 15. Write the Fermi-Dirac distribution and explain the plus sign.

Model answer: n̄=1/[exp(β(ε−μ))+1]. It follows because a fermionic state can be either empty or occupied once.

Question 16. How does the classical limit emerge from quantum statistics?

Model answer: When exp[β(ε−μ)]≫1, the ±1 in the denominator is negligible, giving n̄≈exp[−β(ε−μ)], the Maxwell-Boltzmann form.

Question 17. Explain density of states.

Model answer: It gives the number of allowed one-particle states per energy interval. It converts sums over levels into energy integrals for particle number and energy.

Question 18. State and explain the equipartition theorem.

Model answer: Each independent quadratic term in classical energy contributes average kBT/2. Quantum level spacing can freeze degrees of freedom and invalidate the classical result.

Question 19. Why does a quantum oscillator have zero-point energy?

Model answer: Its ground-state energy is ℏω/2, required by quantum uncertainty and the oscillator eigenvalue spectrum; it remains even at zero temperature.

Question 20. Explain Planck's radiation law statistically.

Model answer: Electromagnetic modes are bosonic oscillators with zero photon chemical potential. Applying Bose-Einstein occupation yields Planck's spectrum.

Question 21. Compare Einstein and Debye models.

Model answer: Einstein assigns one frequency to all lattice oscillators. Debye uses a continuum of vibrational modes and correctly predicts low-temperature CV∝T³.

Question 22. Define Fermi energy.

Model answer: At zero temperature it is the highest occupied one-particle energy. It sets the characteristic energy and temperature scales of a degenerate Fermi gas.

Question 23. Why do electrons in metals show degeneracy at room temperature?

Model answer: Their Fermi temperature is typically far above room temperature because electron density is high; only states near the Fermi surface are thermally affected.

Question 24. What is Bose-Einstein condensation?

Model answer: At sufficiently low temperature, a macroscopic fraction of bosons occupies the lowest quantum state, producing a coherent quantum phase.

Question 25. Explain the Ising model.

Model answer: It is a lattice model of two-state spins with neighbour coupling and an external field. It illustrates collective magnetic ordering and critical phenomena.

Question 26. What is an order parameter?

Model answer: A macroscopic quantity distinguishing phases, such as spontaneous magnetisation. It may vanish in a symmetric phase and become non-zero in an ordered phase.

Question 27. Explain detailed balance in Monte Carlo simulation.

Model answer: For each pair of configurations, equilibrium probability flow from one to the other balances the reverse flow. This helps ensure the desired stationary distribution.

Question 28. Why is equilibration required before collecting simulation averages?

Model answer: Initial configurations may not represent equilibrium. The system must relax before samples are collected; autocorrelation and metastability must also be considered.

Question 29. What does ergodicity mean?

Model answer: A long trajectory explores the relevant accessible phase space sufficiently that time averages agree with ensemble averages under appropriate conditions.

Question 30. Explain ensemble equivalence.

Model answer: For many large short-range systems, different ensembles give the same bulk thermodynamics in the thermodynamic limit. Finite systems and long-range interactions may differ.

Question 31. What is the thermodynamic limit?

Model answer: N and V tend to infinity at fixed density. Bulk quantities become extensive or intensive as appropriate, and surface effects are often negligible.

Question 32. State the Einstein relation for diffusion.

Model answer: In d dimensions, normal diffusion gives ⟨r²(t)⟩≈2dDt at long times. D is the diffusion coefficient.

Question 33. What are critical exponents?

Model answer: They describe power-law behaviour of quantities near continuous phase transitions, such as heat capacity, susceptibility and correlation length.

Question 34. Explain universality.

Model answer: Different systems can share critical exponents when they have common dimensionality, symmetry and interaction range, despite microscopic differences.

Question 35. What is the Gibbs paradox and how is it resolved?

Model answer: Classical counting of identical particles as distinguishable gives non-extensive entropy. Dividing the partition function by N! corrects overcounting.

Question 36. Distinguish information entropy from thermodynamic entropy.

Model answer: Shannon entropy quantifies uncertainty in a probability distribution. Thermodynamic entropy is a physical state function; they are related when probabilities describe physical microstates.

Question 37. Why do fluctuations grow near critical points?

Model answer: Correlations extend over larger distances and collective modes become important, increasing response functions and producing critical behaviour.

Question 38. Describe how statistical physics explains the second law.

Model answer: High-entropy macrostates correspond to vastly more microstates than low-entropy ones, making evolution toward equilibrium overwhelmingly probable for macroscopic systems.

Question 39. What are Green-Kubo relations used for?

Model answer: They calculate transport coefficients from time integrals of equilibrium current autocorrelation functions, linking microscopic dynamics to macroscopic transport.

Question 40. List checks required for a reliable statistical-physics simulation.

Model answer: Check model assumptions, boundary conditions, equilibration, time-step or sampling convergence, autocorrelation, independent runs, finite-size effects and uncertainty estimates.

Laboratory and computational investigations

1. Calorimetry and heat capacity

Measure a sample mass, temperature response and heat supplied using a calibrated heater or calorimeter. Record the time series and allow thermal equilibrium at each point. Determine heat capacity from the slope of energy against temperature after accounting for the calorimeter constant and heat loss. Repeat at several temperatures to identify departures from classical equipartition. Report sensor calibration, thermal lag, uncertainty in mass and electrical measurements, and the model used to correct environmental exchange.

2. Low-temperature heat capacity

Use a cryostat and calibrated thermometer to measure sample heat capacity over a controlled temperature range. For a metal, plot CV/T against T² in the regime where electronic and Debye phonon terms dominate. The intercept estimates the electronic coefficient and the slope estimates the phonon coefficient. Check addenda from the sample holder, thermal contact resistance, thermometer self-heating and reproducibility across warming and cooling runs.

3. Magnetic susceptibility

Place a paramagnetic sample in a controlled magnetic field and measure magnetisation or an equivalent magnetic response while varying temperature. In the weak-field Curie regime, plot inverse susceptibility against temperature. Departures from linearity may indicate interactions, saturation, background diamagnetism or a phase transition. Calibrate field and sensor response, measure sample geometry and subtract holder contributions.

4. Blackbody radiation

Use a cavity radiator or calibrated thermal source and a spectral detector to measure emitted intensity at known temperatures. Compare the measured spectrum with Planck's law and test the displacement of the spectral maximum with temperature. Account for detector spectral response, emissivity, window transmission, background radiation and alignment. Do not treat a non-ideal surface as a perfect blackbody without an emissivity correction.

5. Brownian motion and diffusion

Track suspended tracer particles with a microscope and video or image-analysis system. Convert pixel displacements to physical distances using a calibration scale. Compute mean-square displacement versus time lag and fit the long-time slope to obtain D. Check for drift, convection, particle interactions, tracking errors and finite field-of-view effects. Compare with the Stokes-Einstein prediction when spherical-particle and continuum-fluid assumptions are justified.

6. Monte Carlo study of the Ising model

Implement a two-dimensional spin lattice with periodic boundaries and nearest-neighbour coupling. Initialise spins, perform random single-spin proposals and apply the Metropolis acceptance rule. Discard equilibration sweeps, then record energy and magnetisation over many decorrelated samples. Repeat for several temperatures and lattice sizes. Examine hysteresis, autocorrelation, finite-size rounding and statistical uncertainty before estimating a transition region.

Statistical physics diagrams

OccupationEnergy relative to μBose-EinsteinMaxwell-BoltzmannFermi-Dirac (schematic)
Figure 12.1. Schematic occupation trends. Exact curves depend on temperature, chemical potential, energy scale and degeneracy; the Fermi distribution approaches a step at zero temperature.
EnergyState indexEFOccupied states below EF at T=0
Figure 12.2. Schematic filling of fermion states up to the Fermi energy at zero temperature.
g(ω)ωωDg(ω) ∝ ω²
Figure 12.3. Debye density of vibrational modes in three dimensions, with a cutoff chosen to give 3N modes.
↑↑↓↑↑↓↓↑↓↓↑↑Nearest-neighbour coupling favours collective alignment for J > 0.
Figure 12.4. Two-dimensional Ising lattice: each site carries a two-state spin interacting with neighbouring sites.

Formula summary

Before applying any formula, identify the ensemble and the correct state-counting convention. Distinguish energy per particle from total energy, number density from total particle number, and a one-particle partition function from a many-particle partition function.

Revision checklist

Further study and academic integrity

For deeper study, consult established university texts on statistical mechanics and thermal physics, original research papers for specialist topics, and your institution's current reading list. Verify constants, sign conventions and notation against the source used in your course. These notes are independent educational materials and do not replace an institution's approved syllabus or official assessment materials.

CBE Nexus · Undergraduate Physics · Unit 12: Statistical Physics · Print-ready learning resource.

Official Education References

Use these official organisations for curriculum guidance, examinations and further learning information.