Key Concepts of Phase Transitions

Why This Matters

Phase transitions reveal how matter fundamentally reorganizes itself at the microscopic level. They connect thermodynamics, statistical mechanics, symmetry principles, and scaling theory, making them central to condensed matter physics.

When you encounter a phase transition problem, you should immediately ask: Is this first-order or continuous? What's the order parameter? What symmetry is being broken? Understanding the "why" behind each concept will serve you far better on exams than rote memorization of formulas.

---

Types of Phase Transitions

The fundamental distinction lies in how thermodynamic quantities behave at the transition point. First-order transitions show discontinuities in first derivatives of free energy, while continuous transitions show discontinuities only in higher derivatives.

First-Order Phase Transitions

• Discontinuous jump in entropy and volume: the first derivatives of the Gibbs free energy ($S = -\partial G/\partial T$ and $V = \partial G/\partial P$) change abruptly at the transition temperature
• Latent heat required: energy must be added or removed to complete the transition, even at constant temperature
• Coexistence of phases: both phases exist simultaneously at the transition point, as seen in ice-water mixtures at $0°C$

Second-Order (Continuous) Phase Transitions

• No latent heat: the transition occurs continuously without energy being absorbed or released at a single temperature
• Diverging susceptibilities: second derivatives of free energy (like heat capacity $C_p$ or magnetic susceptibility $\chi$) show singularities at the critical point
• Ferromagnetic-paramagnetic transition: the classic example occurs at the Curie temperature, where spontaneous magnetization vanishes continuously

Critical Points

• Phase boundaries terminate here: the distinction between two phases disappears entirely (e.g., the liquid-gas critical point at $T_c = 647$ K for water)
• Scale invariance emerges: the system looks statistically identical at all length scales, with correlation length $\xi \to \infty$
• Fluctuations dominate: density or magnetization fluctuations occur at all sizes, causing phenomena like critical opalescence

---

Characterizing Order and Symmetry

Phase transitions fundamentally involve changes in the organization of a system. Order parameters quantify this organization, while symmetry breaking explains how distinct phases emerge from symmetric high-temperature states.

Order Parameters

The order parameter is the central quantity that distinguishes the ordered phase from the disordered one. It takes a nonzero value in the ordered phase and zero in the disordered phase.

• Magnetization $M$ for ferromagnets: measures net spin alignment; $M = 0$ above $T_c$, $M \neq 0$ below
• Density difference $\rho_l - \rho_g$ for liquid-gas: vanishes at the critical point where liquid and gas become indistinguishable
• Superfluid density (or condensate wavefunction) for superfluids: a complex-valued order parameter whose magnitude reflects the superfluid fraction

Symmetry Breaking

• High-temperature phase is symmetric: all directions or configurations are equally probable
• Low-temperature phase selects a direction: spins align along a particular axis even though the Hamiltonian has no preferred direction
• Spontaneous symmetry breaking: the ground state has lower symmetry than the governing equations, a concept central to both condensed matter and particle physics

---

Theoretical Frameworks

Several theoretical approaches allow us to predict and understand phase transition behavior. These range from phenomenological expansions to sophisticated mathematical machinery that explains why different systems behave identically near criticality.

Landau Theory

Landau's approach is to expand the free energy as a power series in the order parameter $\phi$, respecting the symmetries of the system:

$F = F_0 + a(T)\phi^2 + b\phi^4 + \cdots$

The coefficient $a(T)$ changes sign at $T_c$. Above $T_c$, $a > 0$ and the minimum sits at $\phi = 0$ (disordered). Below $T_c$, $a < 0$ and the minimum shifts to $\phi \neq 0$ (ordered). Typically one writes $a(T) = a_0(T - T_c)$ so the sign change is explicit.

Landau theory predicts mean-field critical exponents (e.g., $\beta = 1/2$), which are correct above the upper critical dimension ($d = 4$ for the Ising class) but fail in lower dimensions where fluctuations matter.

Mean-Field Theory

• Replaces fluctuating neighbors with an average field: each spin or particle feels the average effect of all others
• Exact in high dimensions or for long-range interactions: becomes increasingly accurate as coordination number increases
• Misses critical fluctuations: predicts incorrect exponents in $d < 4$ dimensions because it ignores correlations between fluctuations

Renormalization Group Theory

The renormalization group (RG) systematically accounts for fluctuations at all length scales. The basic idea proceeds in steps:

1. Coarse-grain: integrate out (average over) the shortest-wavelength degrees of freedom
2. Rescale: shrink the system back to its original size
3. Renormalize: adjust the coupling constants so the partition function is preserved

Repeating this procedure generates a flow in parameter space. Fixed points of this flow determine critical behavior: systems that flow to the same fixed point share the same critical exponents, regardless of microscopic details. This is the quantitative explanation of universality, and it earned Kenneth Wilson the 1982 Nobel Prize.

---

Universal Behavior and Scaling

One of the most remarkable discoveries in phase transition physics is that vastly different systems exhibit identical critical behavior. This universality arises from the diverging correlation length, which washes out microscopic details.

Universality Classes

Systems are grouped not by their microscopic Hamiltonian or material composition, but by just two features:

• Spatial dimensionality $d$
• Symmetry of the order parameter (its number of components and the symmetry group)

The Ising universality class (scalar order parameter, $Z_2$ symmetry) includes uniaxial ferromagnets, binary alloys, and the liquid-gas critical point. The XY class (two-component order parameter, $O(2)$ symmetry) covers superfluid helium and planar magnets. The Heisenberg class (three-component, $O(3)$) describes isotropic ferromagnets.

Critical Exponents

Near $T_c$, thermodynamic quantities follow power laws in the reduced temperature $t = (T - T_c)/T_c$:

• Correlation length: $\xi \sim |t|^{-\nu}$
• Order parameter: $M \sim |t|^{\beta}$ (for $T < T_c$)
• Susceptibility: $\chi \sim |t|^{-\gamma}$
• Heat capacity: $C \sim |t|^{-\alpha}$

Only two exponents are independent. Scaling relations connect them all:

• $\alpha + 2\beta + \gamma = 2$ (Rushbrooke)
• $\gamma = \nu(2 - \eta)$ (Fisher)
• $\nu d = 2 - \alpha$ (Josephson/hyperscaling)

These are universal within a class and depend only on $d$ and order parameter symmetry.

Scaling Laws

• Data collapse: plotting $M|t|^{-\beta}$ vs. $H|t|^{-\beta\delta}$ yields the same curve for all systems in a universality class
• Finite-size scaling: describes how critical behavior is modified in systems of finite size $L$, with $\xi$ effectively capped at $L$
• Widom scaling hypothesis: the singular part of the free energy is a generalized homogeneous function of its arguments, from which all exponent relations follow

---

Models and Representations

Simplified models capture the essential physics of phase transitions while remaining mathematically tractable. They serve as testing grounds for theoretical ideas and provide exact results in certain limits.

Ising Model

The Ising model places spins $s_i = \pm 1$ on a lattice with the Hamiltonian:

$H = -J\sum_{\langle ij \rangle} s_i s_j - h\sum_i s_i$

Here $J$ is the nearest-neighbor coupling ($J > 0$ for ferromagnetic) and $h$ is an external field. Despite its simplicity, this is the minimal model that exhibits a nontrivial phase transition. Onsager's exact solution of the 2D case (1944) proved that mean-field theory fails in low dimensions. No exact 3D solution exists; results come from numerical methods and RG.

Phase Diagrams

• Map stability regions in parameter space: typically temperature vs. pressure, or temperature vs. magnetic field
• Phase boundaries are coexistence lines: first-order transitions occur along lines, critical points are isolated points where those lines terminate
• Triple points: where three phases coexist simultaneously (e.g., solid-liquid-gas at a unique $(T, P)$)
• Multicritical points: where multiple critical lines or phase boundaries meet (e.g., tricritical points, bicritical points)

---

Dynamics and Kinetics

Phase transitions don't occur instantaneously. Nucleation, growth, and metastability govern how transitions actually proceed in time.

Nucleation and Growth

Nucleation is typically the rate-limiting step in a first-order transition. A new-phase droplet must reach a critical radius before it can grow spontaneously:

$r^* = \frac{2\gamma}{\Delta g}$

where $\gamma$ is the surface tension (energy cost of the interface) and $\Delta g$ is the bulk free energy gain per unit volume. Nuclei smaller than $r^*$ shrink back; larger ones grow.

• Homogeneous nucleation: occurs in a perfectly uniform system; requires large supercooling or supersaturation
• Heterogeneous nucleation: impurities, surfaces, or defects lower the free energy barrier $\Delta G^*$, making this the dominant mechanism in practice

Metastable States

• Local but not global free energy minimum: superheated liquid or supercooled vapor can persist until a nucleation event occurs
• Spinodal decomposition: when the system is driven past the spinodal line into the thermodynamically unstable region (where $\partial^2 F / \partial \phi^2 < 0$), phase separation occurs spontaneously without any nucleation barrier
• Practical importance: glasses, amorphous solids, and many technological materials exist in metastable states

---

Quick Reference Table

---

Self-Check Questions

1. What distinguishes a first-order phase transition from a second-order transition in terms of free energy derivatives and latent heat?

2. The 3D Ising model and the liquid-gas critical point belong to the same universality class. What shared features place them in this class, and what microscopic details are irrelevant?

3. Compare and contrast Landau theory and renormalization group approaches: when does each give accurate predictions for critical exponents?

4. A supercooled liquid exists below its freezing point without solidifying. Explain this metastable state in terms of nucleation theory and identify what would trigger the transition.

5. If you're given critical exponents $\beta$ and $\gamma$ for a system, how would you use scaling relations to determine other exponents? Write out at least one such relation.