Order Parameter · aiwiki.page

An order parameter is a quantity used in statistical mechanics and condensed matter physics to distinguish phases and describe collective ordering. In the conventional description of a phase transition, it is often defined to vanish in a disordered phase and become nonzero in an ordered phase. It may be a scalar, vector, complex amplitude, or tensor, depending on the ordering involved. Its transformation under symmetry operations helps identify how one phase differs from another. (damtp.cam.ac.uk)

Definition and symmetry

An order parameter reduces a system’s many microscopic degrees of freedom to a small number of variables describing its macroscopic organization. It commonly takes the form of an expectation value of a microscopic observable or a spatially averaged quantity. A local order-parameter field additionally records how ordering varies from place to place. This description does not retain every microscopic detail; it selects the information relevant to a particular transition. (damtp.cam.ac.uk)

In spontaneous symmetry breaking, the governing equations respect a symmetry that an individual ordered equilibrium state does not. For example, reversing every spin leaves the zero-field Ising model unchanged, but exchanges its positively and negatively magnetized states. A nonzero order parameter identifies the selected state. Its magnitude describes the amount of ordering, while its sign or orientation distinguishes symmetry-related possibilities. Different transitions can require different order parameters, even within the same material. (damtp.cam.ac.uk)

Representative examples

For ferromagnetism, the order parameter is spontaneous magnetization. In an Ising system with $N$ spins $s_i=\pm1$ , a dimensionless magnetization is

m=\frac{1}{N}\sum_{i=1}^{N}\langle s_i\rangle .

Below the ordering temperature, the system can sustain $m\neq0$ without an applied field; above it, the zero-field equilibrium magnetization vanishes. Where spin orientations are not restricted to two values, magnetization is generally a vector rather than a scalar. (damtp.cam.ac.uk)

For a liquid–gas transition, the density difference between coexisting liquid and gas provides an order parameter. It decreases to zero at the critical point, where the two phases become indistinguishable. Unlike magnetization, this quantity does not describe the alignment of microscopic vectors. It measures a difference between macroscopic states. (damtp.cam.ac.uk)

In liquid crystals, orientational ordering need not produce a preferred molecular head-to-tail direction. Nematic order is therefore described by a second-rank tensor constructed from averages of molecular-axis products. This distinguishes an isotropic distribution from an aligned one without assigning physically different states to opposite directions along the same axis. (arxiv.org)

For superconductivity and superfluidity, an order parameter commonly has the complex form

\psi(\mathbf r)=|\psi(\mathbf r)|e^{i\theta(\mathbf r)}.

Both amplitude and phase matter. Spatial phase variations are central to superfluid flow, while phase winding around a vortex describes a topological defect. The order parameter is a collective field, not simply the wave function of an isolated particle. (nobelprize.org)

Landau theory

Landau theory treats the order parameter as a variable in an effective free-energy expansion. For a real scalar $\eta$ with reversal symmetry, a simple free-energy density is

f(\eta)=f_0+\frac{a}{2}\eta^2+\frac{b}{4}\eta^4-h\eta , \qquad b>0.

Here $h$ is a field conjugate to the order parameter; for magnetization it corresponds, with suitable conventions, to an applied magnetic field. Equilibrium values minimize $f$ . At $h=0$ , reversal symmetry excludes odd powers. (damtp.cam.ac.uk)

When $a>0$ , the minimum is $\eta=0$ . When $a<0$ , two minima appear:

\eta=\pm\sqrt{-a/b}.

If $a$ changes linearly with $T-T_c$ , this predicts a continuous onset proportional to $(T_c-T)^{1/2}$ . Other permitted terms or additional competing minima can instead produce a discontinuous transition. The expansion’s form is constrained by symmetry, whereas its coefficients depend on the system. (damtp.cam.ac.uk)

Ginzburg–Landau theory extends this construction to spatially varying fields. A term such as $\kappa|\nabla\eta|^2/2$ , involving the gradient, assigns a free-energy cost to spatial variation. This allows the description of interfaces, textures, and fluctuations rather than only uniform states. (damtp.cam.ac.uk)

Critical behavior and measurement

Near a continuous transition, the equilibrium order parameter often follows a power law,

|\eta|\propto(-t)^\beta,\qquad t=\frac{T-T_c}{T_c}<0,

where $\beta$ is a critical exponent. Fluctuations become correlated over increasingly large distances, characterized by the correlation length. The renormalization group explains why systems with different microscopic interactions can share critical behavior, and why simple mean-field predictions need not give the correct exponents. (pmc.ncbi.nlm.nih.gov)

Measurement requires attention to averaging. Oppositely ordered states can cancel in an average, so a vanishing mean alone does not establish the absence of order. Studies may therefore examine the order parameter’s distribution, its moments, and spatial correlations. Finite-size scaling relates these quantities to their behavior as the system approaches the thermodynamic limit. (www2.icp.uni-stuttgart.de)

Limits of the concept

Not every distinction between phases is captured by a conventional local symmetry-breaking order parameter. Some quantum phases, including those associated with the quantum Hall effect, require topological characterization beyond the Landau classification. Nonlocal quantities can serve as generalized indicators of order, but they should not be confused with the local expectation values used for ordinary magnetic or structural ordering. (damtp.cam.ac.uk)