Statistical mechanics Study Guide

📖 Core Concepts Statistical mechanics – uses probability to connect microscopic particle behavior to macroscopic thermodynamic properties. Ensemble – a huge collection of virtual copies of a system, each possibly in a different microstate; the ensemble’s probability distribution gives observable averages. Phase space (classical) – the set of all possible positions + momenta; a single system is a point, an ensemble is a distribution over points. Hilbert space (quantum) – the space of all pure quantum states; ensembles are described by a density operator (matrix). Mechanical vs. statistical equilibrium – mechanical equilibrium stops macroscopic motion; statistical equilibrium allows microscopic motion while the probability distribution stays unchanged. Fundamental postulate – for an isolated system, the equilibrium distribution depends only on conserved quantities (energy, particle number, …). Equal‑a‑priori probability – every accessible microstate of an isolated system with fixed energy is equally likely. --- 📌 Must Remember Three primary equilibrium ensembles Microcanonical: fixed \(E, N, V\). Canonical: fixed \(T\) (system in a heat bath). Grand canonical: fixed \(T\) and chemical potential \(\mu\) (energy + particles exchange). Thermodynamic limit – \(N\to\infty\), \(V\to\infty\) with constant density ⇒ all ensembles give the same macroscopic results. Ergodic hypothesis – a system eventually visits every accessible microstate, justifying equal probabilities. Liouville equation (classical) & von Neumann equation (quantum) govern the exact, reversible evolution of ensemble distributions. Linear response – small perturbations from equilibrium produce responses proportional to the perturbation. Fluctuation–Dissipation theorem – the same microscopic fluctuations that occur at equilibrium dictate how the system dissipates energy when perturbed. --- 🔄 Key Processes Building an equilibrium ensemble Identify constraints (e.g., fixed \(E\), fixed \(T\), fixed \(\mu\)). Choose the appropriate ensemble (micro‑, canonical, or grand‑canonical). Assign equal probability to all microstates consistent with the constraints (microcanonical) or weight them by the Boltzmann factor for canonical/grand‑canonical cases. Monte Carlo sampling (Metropolis–Hastings) Propose a new microstate. Compute the ratio of statistical weights (e.g., Boltzmann factors). Accept the move with probability \(\min(1, \text{ratio})\); otherwise keep the current state. Repeat to generate a representative set of states for averaging. Molecular dynamics for microcanonical averages Initialise positions and momenta consistent with the desired energy. Integrate Hamilton’s equations (classical) or Schrödinger equation (quantum) in time. Use time‑averaged observables as ensemble averages, assuming ergodicity. Linear response calculation Apply a weak external field \(F(t)\). Measure the induced observable \(A(t)\). Relate \(A(t)\) to the equilibrium correlation function \(\langle B(0)A(t)\rangle\) via the Kubo formula. --- 🔍 Key Comparisons Microcanonical vs. Canonical Constraints: \(E,N,V\) fixed vs. \(T\) fixed (energy can fluctuate). Probability: all allowed microstates equally likely vs. weighted by \(e^{-E/kBT}\). Classical vs. Quantum Ensembles Representation: probability density \(\rho(q,p)\) in phase space vs. density operator \(\hat\rho\) in Hilbert space. Evolution: Liouville equation vs. von Neumann equation. Mechanical equilibrium vs. Statistical equilibrium No net macroscopic forces vs. stationary probability distribution despite ongoing microscopic motion. --- ⚠️ Common Misunderstandings “Equal a priori probability means all microstates have the same energy.” – It means all accessible microstates (those that satisfy the energy constraint) are equally probable, not that they share the same energy. “Liouville’s equation predicts irreversibility.” – It is perfectly reversible; irreversibility emerges only after coarse‑graining or invoking stochastic approximations. “Canonical ensemble averages are always identical to microcanonical averages.” – They coincide only in the thermodynamic limit; finite or critical systems can show noticeable differences. --- 🧠 Mental Models / Intuition Ensemble as a “statistical microscope.” Imagine looking at many identical copies of a system side‑by‑side; the average over copies is what you measure experimentally. Boltzmann factor as “energy penalty.” Higher‑energy states are exponentially suppressed; the temperature sets how harsh the penalty is. Ergodicity as “shuffling the deck.” Over long times the system “shuffles” through all cards (microstates), so a single long run mimics many independent copies. --- 🚩 Exceptions & Edge Cases Small (finite) systems – ensemble equivalence can fail; microcanonical fluctuations are larger. Phase transitions – long‑range correlations break ergodicity, making the choice of ensemble critical. Long‑range interacting systems (e.g., gravitational) – may not reach the usual thermodynamic limit; ensemble predictions can diverge. --- 📍 When to Use Which Microcanonical – isolated lab‑scale system with strictly fixed energy (e.g., adiabatic container). Canonical – system in contact with a large heat bath (most laboratory situations). Grand canonical – open system exchanging particles (chemical reactions, adsorption, semiconductor reservoirs). Monte Carlo – when analytical partition functions are unavailable but you need equilibrium averages. Molecular dynamics – when you need time‑dependent trajectories or want to compute microcanonical averages directly. Linear response / Kubo – for transport coefficients (conductivity, viscosity) near equilibrium. --- 👀 Patterns to Recognize “Fixed X, variable Y” → choose ensemble where the fixed quantity appears as a constraint (energy → microcanonical, temperature → canonical, temperature + chemical potential → grand canonical). “Weight = exp(−β E)” → whenever a Boltzmann factor appears, you’re in a canonical‑type setting. “Correlation function → transport coefficient” – see Green–Kubo relations; a time integral of an equilibrium autocorrelation gives the macroscopic coefficient. --- 🗂️ Exam Traps Distractor: “All ensembles are always equivalent.” – Only true in the thermodynamic limit; finite or critical systems break the equivalence. Distractor: “Liouville’s equation explains entropy increase.” – It conserves Gibbs entropy; irreversibility requires additional assumptions (coarse‑graining, stochasticity). Distractor: “Microcanonical ensemble requires a heat bath.” – Incorrect; the microcanonical ensemble describes isolated systems with no bath. Distractor: “Maximum entropy principle applies only to quantum systems.” – It is a general principle; it selects the most unbiased ensemble consistent with known constraints for both classical and quantum cases. ---