eduKateSG · How Science Works
Choose your reading route
Inside an unstable part of a phase diagram, an apparently uniform mixture can lower its free energy when small composition fluctuations grow. Separation begins throughout the material rather than waiting for a rare critical nucleus.
Full chapter index · Check your understanding · Materials Science · Science master hub
A mechanism-focused enrichment guide for curious students, parents and teachers. Worked numbers are explicitly hypothetical; research links identify supporting scientific studies. This is not an examination-syllabus claim or a laboratory operating procedure.
Inside this guide
Chapters 1–3 · Understand the mechanism
1. Begin with a uniform state that is unstable
Chapters 4–5 · Work through the model
Chapters 6–8 · Evaluate the evidence
6. Know what real materials add
CHAPTER 1 OF 8 · Understand the mechanism
1. Begin with a uniform state that is unstable
A freshly quenched mixture can look uniform while being thermodynamically unstable. Microscopic composition fluctuations are always present. In the spinodal region, some of those fluctuations grow instead of being smoothed away.
This differs from nucleation-and-growth in a metastable region, where forming a new phase generally requires overcoming a free-energy barrier associated with a sufficiently large nucleus.
Spinodal decomposition describes the early amplification and later evolution of composition patterns. It does not imply that every two-phase material followed this route.
Let f(c) be homogeneous free-energy density as a function of composition c. The curvature f″(c) tells whether a small local composition change is penalised or favoured in the simplest local picture.
When f″ is positive, small fluctuations tend to raise local free energy. When f″ is negative, the homogeneous state is locally unstable: slightly richer and poorer regions can lower it.
A gradient-energy term still penalises abrupt spatial changes. This prevents arbitrarily fine fluctuations from dominating and gives interfaces a finite energetic cost.
CHAPTER 3 OF 8 · Understand the mechanism
3. Follow conserved composition with Cahn–Hilliard theory
A common model defines chemical potential μ = f′(c) − κ∇²c and conserved evolution ∂c/∂t = M∇²μ. M is a mobility and κ weights composition gradients.
The equation conserves the total amount of each component under closed, no-flux conditions. Material rearranges; the model does not create one component from nothing.
Simulations of early spinodal decomposition test how nonlinear terms alter the evolving pattern. The linear theory is most useful while fluctuations remain small.
CHAPTER 4 OF 8 · Work through the model
4. Calculate which dimensionless modes grow
For a simplified dimensionless linear model, take growth rate s(q) = q²(1 − q²), where q is disturbance wavenumber. Modes with 0 < q < 1 grow; q > 1 modes decay.
At q = 0.5, s = 0.1875. At q = 1/√2 ≈ 0.707, s = 0.25, the maximum of this teaching model. At q = 1.2, s ≈ −0.634.
If amplitude obeys A(t)=A₀est, the fastest early mode grows most rapidly. These dimensionless numbers are illustrative; converting them to a real wavelength and time requires material-specific scaling.
| Dimensionless q | Growth rate s(q) | Early behaviour |
|---|---|---|
| 0.5 | 0.1875 | Grows |
| 0.707 | 0.250 | Fastest in this model |
| 1.2 | −0.634 | Decays |
CHAPTER 5 OF 8 · Work through the model
5. Distinguish early wavelength from final domain size
Linear stability predicts early amplification. As composition contrasts become large, nonlinear evolution matters and domains commonly coarsen: characteristic regions grow while interfacial area decreases.
The initially fastest-growing wavelength is therefore not a permanent ruler stamped into the material. Processing may arrest coarsening, while elasticity, flow or anisotropy can redirect morphology.
Research on spinodal decomposition and coarsening fronts illustrates the multi-stage nature of pattern formation. One early length scale does not fully specify the late structure.
Elastic strain can couple to composition when phases prefer different lattice spacings. Surfaces or particles may favour one component and bias where domains form.
Phase-field research with embedded particles studies the interplay of phase separation and wetting. The resulting morphology depends on particle preference and mixture composition.
Temperature history matters because it changes free energy and mobility. Two samples ending at the same temperature can differ if one spent more time coarsening on the way.
Microscopy reveals real-space domains but covers a limited field and resolution. Scattering can reveal a characteristic reciprocal-space peak across a larger volume. Each measurement compresses the structure differently.
A shifting scattering peak can support coarsening, but converting peak position into a single domain size assumes a suitable morphology and analysis convention. Report the full pattern, uncertainty and image processing.
Compare multiple times after a controlled quench. A final two-colour-looking image shows separation, not by itself whether the early route was spinodal rather than nucleated.
Does spinodal separation need a critical nucleus? In the ideal unstable region, small fluctuations can grow without crossing the classical nucleation barrier.
Do all wavelengths grow? No. Gradient energy suppresses sufficiently short wavelengths in the standard model.
Does the fastest early wavelength equal the final domain size? No. Nonlinear evolution and coarsening change the pattern. Continue with Materials Science to connect processing, microstructure and properties.