Critical Phenomena

Critical phenomena occur at the endpoint of a phase diagram where two phases become indistinguishable, such as the liquid-gas critical point or the Curie point of a ferromagnet. At this point, thermodynamic response functions like compressibility and susceptibility diverge, and the correlation length ξ grows without bound. The central concept is universality: systems with different microscopic interactions but the same spatial dimensionality and symmetry exhibit identical critical exponents. This was explained by the renormalization group theory developed by Kenneth Wilson in the 1970s, which showed that critical behavior depends only on a few relevant parameters. For example, the 3D Ising model and the liquid-gas transition belong to the same universality class, sharing exponents like β ≈ 0.326 for the order parameter.¹

Critical exponents describe how quantities diverge or vanish as the critical point is approached, e.g., the order parameter m ~ (T_c - T)^β, susceptibility χ ~ |T - T_c|^{-γ}, and correlation length ξ ~ |T - T_c|^{-ν}. These exponents are not independent; they obey scaling relations such as the Rushbrooke inequality α + 2β + γ ≥ 2 and the hyperscaling relation 2 - α = dν, which involve the spatial dimension d. The scaling hypothesis, formulated by Widom and others, posits that the free energy is a homogeneous function, leading to these relations. In systems below the upper critical dimension (d = 4 for many models), hyperscaling holds, but above it, mean-field exponents apply. The concept of scaling has been extended to dynamic critical phenomena, where relaxation times diverge with a dynamic exponent z.

Critical phenomena have been observed in diverse systems: fluids, magnets, superfluids, and even in the early universe. Precision experiments on the specific heat of helium near the lambda point confirmed the predicted logarithmic divergence and the value of the critical exponent α. In recent decades, critical phenomena have been studied in quantum phase transitions, where fluctuations are driven by quantum rather than thermal effects, leading to new universality classes. The concept of scale invariance has also found applications in complex systems, such as financial markets and biological networks, though these are not true equilibrium critical points. In cosmology, critical phenomena appear in the formation of black holes, where the mass of the black hole scales as a power law with the initial conditions, a discovery by Choptuik in 1993.²

Beyond the standard textbook examples, critical phenomena include subtle effects like the Fisher renormalization of exponents when a field is held constant instead of a density. The concept of 'weak universality' allows for non-universal amplitude ratios while exponents remain universal. In systems with long-range interactions, the upper critical dimension changes, altering the validity of mean-field theory. Critical phenomena also appear in non-equilibrium systems, such as absorbing phase transitions, which exhibit scaling but lack an equilibrium free energy. The study of critical phenomena has led to the development of the renormalization group, which is now a fundamental tool in many areas of physics, from particle physics to turbulence. Additionally, the concept of 'critical opalescence'—the strong scattering of light near the critical point—was a key early observation that motivated the theory.³

Critical exponent

A number describing how a physical quantity diverges or vanishes near a critical point, e.g., β, γ, ν.

Correlation length

The typical distance over which fluctuations are correlated; it diverges at the critical point.

Universality class

A set of systems sharing the same critical exponents due to common symmetry and dimensionality.

Renormalization group

A mathematical framework that explains scale invariance and universality in critical phenomena.

Order parameter

A quantity that is zero in one phase and non-zero in another, such as magnetization or density difference.