Landau-Ginzburg Functional and Critical Phenomena

Recommended resources

The freely available resources this lesson will direct you to.

Hello! Welcome to the final lesson in our module on Landau-Ginzburg theory.

So far, we have constructed the Landau-Ginzburg (LG) free energy functional, used it to derive mean-field behavior and the Ornstein-Zernike correlation function, and, most recently, rigorously established that mean-field theory fails for dimensions $d \leq 4$ due to the overwhelming effect of fluctuations.

Today, we will synthesize all these ideas to address our final learning outcome for this module: to explain the role of the Landau-Ginzburg functional as a universal effective field theory for critical phenomena. We will piece together what we've learned to understand why this remarkable theoretical structure is the cornerstone of the modern theory of phase transitions.

The Phenomenon of Universality

Before we discuss the theory, let's remind ourselves of the experimental reality that motivates it. Near a critical point, a startling phenomenon occurs: the detailed microscopic nature of a substance seems to be "washed away," and systems as different as a fluid and a magnet exhibit identical behavior.

To see this in action, first watch this classic demonstration of critical opalescence in a liquid-gas system.

2. Lec 1 (continued); The Landau-Ginzburg Approach Part 1

This segment from an MIT OpenCourseWare lecture vividly shows what happens as a fluid is cooled through its critical point. Notice how the system, which is transparent away from the critical point, becomes opaque (opalescent) precisely at the transition. This is direct visual evidence of huge density fluctuations occurring on the scale of the wavelength of light.

Watch the clip from 25:08 to 28:17. The key takeaway is that at the critical point, fluctuations become macroscopic, justifying the need for a theory that focuses on these collective, long-wavelength modes rather than individual atoms.

This phenomenon of diverging fluctuations, and the diverging correlation length we've discussed, is at the heart of universality. It suggests that the physics at criticality is governed by large-scale collective behavior, not microscopic details. For a more formal introduction to this idea, the following text provides an excellent overview.

[PDF] Statistical Field Theory - DAMTP

David Tong's lecture notes offer a wonderfully clear and insightful introduction to the concept of universality. He frames it in the context of two deep facts: 'Nature is organised by symmetry' and 'Nature is organised by scale'.

Please read the introduction (pages 2-4) of this chapter. It eloquently argues that universality arises because, at large scales, the system loses memory of its microscopic composition, and its behavior is dictated purely by symmetry.

With this experimental and conceptual motivation, we can now ask: how can we build a theory that has this universality baked in? The answer is the Landau-Ginzburg functional.

Constructing a Universal Theory

The power of the LG approach lies in the fact that it is not derived from the microscopic details of a specific material. Instead, it is postulated based on the most general principles that should govern any continuous phase transition: symmetry, dimensionality, and locality.

The following video by Prof. Mehran Kardar explains this construction masterfully.

3. The Landau-Ginzburg Approach Part 2

In this clip, Prof. Kardar outlines the logic for 'guessing' the form of the LG free energy functional. He emphasizes that by writing down all terms consistent with the system's symmetries, we arrive at a generic model that should describe any system in the same universality class.

Watch from 02:29 to 10:25. Pay close attention to these key points: The partition function is expressed as a functional integral over a coarse-grained order parameter field, m(x). The 'Hamiltonian' or free energy functional, \beta H, is written as an expansion in powers of m(x) and its gradients. The terms included in the expansion are only those that respect the symmetries of the problem (e.g., rotational symmetry means only even powers like m^2, m^4, and (( abla m)^2) are allowed). The coefficients (t, u, K, etc.) are phenomenological parameters that depend on temperature and the specific material, but the form of the functional is universal.

To summarize the logic:

Coarse-Graining and the Order Parameter Field: We accept that at criticality, the important degrees of freedom are long-wavelength fluctuations. We therefore average our microscopic variables (spins, densities) over small regions to define a smooth, continuous order parameter field, often denoted $ϕ (x)$ or $m (x)$ . This is the "field" in "effective field theory."
Symmetry Constraints: We write down the most general possible free energy functional, $F [ϕ (x)]$ , as a Taylor-like expansion in $ϕ$ and its gradients. We then discard any terms that violate the known symmetries of the system.
- Locality: Interactions are short-ranged, so the free energy density at a point $x$ should only depend on $ϕ$ and its derivatives at $x$ . This leads to an integral form: $F = \int d^{d} x f (ϕ (x), \nabla ϕ (x), ...)$ .
- Order Parameter Symmetry: For an Ising-like system, the physics is unchanged under $ϕ \to - ϕ$ . This means the free energy density $f$ can only contain even powers: $ϕ^{2}, ϕ^{4}, ...$ .
- Spatial Symmetry: In an isotropic system, there is no preferred direction, so gradient terms must also be rotationally invariant, like $(\nabla ϕ)^{2}$ .

This procedure universally leads to the $ϕ^{4}$ LG functional for any system with a single-component order parameter and up-down ( $Z_{2}$ ) symmetry:

F [ϕ (x)] = \int d^{d} x [\frac{r}{2} ϕ^{2} + \frac{u}{4} ϕ^{4} + \frac{K}{2} (\nabla ϕ)^{2} - h ϕ]

This same functional form describes uniaxial magnets, the liquid-gas critical point, and binary fluid mixtures. The differences between these systems are hidden in the non-universal, material-dependent coefficients $r, u, K$ .

Test your understanding!

What would be the leading terms in a Landau-Ginzburg functional for an isotropic three-dimensional ferromagnet (like iron) approaching its Curie point? The order parameter is a 3-component vector field $m (x)$ .

Show answer

The free energy must be invariant under O(3) rotations of the order parameter $m$ . Therefore, all terms must be scalars formed from $m$ . The leading terms are:

F [m (x)] = \int d^{d} x [\frac{r}{2} (m \cdot m) + \frac{u}{4} (m \cdot m)^{2} + \frac{K}{2} i = 1 \sum 3 (\nabla m_{i})^{2} - h \cdot m]

Note that $m \cdot m = m^{2}$ is a scalar. The term $(m \cdot m)^{2} = m^{4}$ is also a scalar. A term like $(m_{1}^{2} + m_{2}^{4} + m_{3}^{6})$ would violate the O(3) symmetry. This is the LG functional for the n=3 Heisenberg universality class.

From Microscopic to Effective: A Derivation

We've established that the LG functional is postulated on general grounds. However, for certain microscopic models, it can be formally derived. This is a crucial step in seeing how it truly functions as an effective theory that emerges from a more fundamental description.

Given your background, you'll appreciate the following derivation for the Ising model, which uses the Hubbard-Stratonovich transformation—a standard technique in many-body and quantum field theory.

[PDF] Chapter 2 Ginzburg-Landau Phenomenology

This section provides a first-principles derivation of the Ginzburg-Landau Hamiltonian for the Ising model. It's a mathematically elegant process that makes the concept of an effective field theory concrete.

Please study Section 2.4, "Derivation of GL Hamiltonian for the Ising model". Follow the four steps: Hubbard-Stratonovich Decoupling: A continuous field \Psi_i is introduced to decouple the spin-spin interaction term e^{K \sigma_i \sigma_j}. Integrating out Microscopic Variables: The original Ising spins \{\sigma_i\} are summed over, leaving a partition function that depends only on the new field \{\Psi_i\}. Re-exponentiating: The result is put back into the form of an effective Hamiltonian for the \Psi field. Expansion: The resulting Hamiltonian is expanded for small \Psi and long wavelengths (small momentum q), recovering the familiar \phi^4 form of the LG functional.

This derivation is powerful. It explicitly shows how the microscopic interaction strength $K$ and lattice structure get absorbed into the phenomenological LG parameters $t$ and $K$ , while the general $ϕ^{4}$ structure emerges from the mathematics of the transformation and expansion.

Synthesis: The Complete Picture

We can now assemble the full definition: The Landau-Ginzburg functional is a universal effective field theory.

Effective Theory: It is not a fundamental description of matter. It emerges after we have coarse-grained (or "integrated out") the microscopic degrees of freedom. Its parameters ( $r, u, K$ ) are not fundamental constants but phenomenological coefficients that encode all the complicated, non-universal microscopic physics.
Field Theory: Its fundamental variable is a continuous field $ϕ (x)$ , the order parameter. This is essential because near a critical point, the correlation length diverges, and the physics is dominated by collective, long-wavelength fluctuations that cannot be described by focusing on individual particles or spins.
Universal: Its mathematical structure is determined only by the symmetries of the order parameter and the dimensionality of space. As long as different physical systems share these same general properties, they will be described by the same LG theory near their critical point. This is the theoretical origin of the experimental phenomenon of universality.

The LG functional is the common language used to discuss critical phenomena. It provides the starting point for the most powerful tool we have for understanding universality and calculating correct critical exponents: the Renormalization Group (RG).

Conclusion

Key Takeaways:

The Landau-Ginzburg functional is an effective field theory that describes the long-wavelength physics of a system near a critical point.
It is universal because its structure is dictated by general principles of symmetry and dimensionality, not by the specific microscopic details of a material.
The process of coarse-graining bridges the microscopic world and the effective field theory, absorbing complex details into a few phenomenological parameters.
Systems with different microscopic origins but the same order parameter symmetry and spatial dimension (e.g., Ising magnets and liquid-gas systems) belong to the same universality class and are described by the same LG functional.

Preview of the Next Lesson:

We have now reached the limits of what Landau-Ginzburg theory can do on its own. It provides the correct framework, but as we've seen, a simple mean-field analysis of it only works for $d > 4$ . To get the correct physics in our world ( $d = 3$ ), we need to analyze the full LG functional, including all fluctuations, in a more sophisticated way.

Our next module, "Scaling and Universality," will build the conceptual foundation for this. We will start by introducing the scaling hypothesis, a brilliant phenomenological idea that postulates specific relationships between critical exponents. This will formalize the concept of universality classes and prepare us for the introduction of the Renormalization Group, the machinery that ultimately proves the scaling hypothesis and explains the deep connection between the LG functional and the universal nature of critical points.