The Gold Standard of Thermodynamics | Fisher Renormalisation, Constrained Ensembles and Apparent Critical Exponents

Find the mechanism before the formula

A constraint can change the temperature path by which a system reaches criticality. Fisher renormalisation explains why fixing a hidden density or composition can make measured critical exponents differ from their unconstrained ensemble values without creating a new universality class.

We will derive the exponent map, calculate constrained three-dimensional Ising values, transfer the result to an observable ratio, and separate genuine asymptotic renormalisation from finite-range apparent exponents.

This guide belongs to the Gold Standard of Thermodynamics, the Gold Standard of Education and the How X Works hub.

Choose a reading route

Start with the question closest to the decision you need to make. Every route returns to the same worked evidence and transfer checks.

Chapter contents

Open the group you need, then use the chapter links. Teaching sections remain open on the page.

1. A constraint changes the thermodynamic path

An unconstrained ensemble may let an extensive quantity x fluctuate while its conjugate field u is fixed. An experiment can instead hold x fixed. If x couples to the energy-like scaling field, the physical reduced temperature is related nonlinearly to the ideal thermal field near criticality.

Examples include fixed impurity concentration, fixed composition or fixed particle density rather than fixed chemical potential. The local interactions can be unchanged while the path through state space changes. The observed singularity is then expressed in a different control variable.

• Name the fixed extensive quantity.
• Name its conjugate intensive field.
• State the ideal ensemble used for comparison.
• Separate the true asymptotic limit from the accessible range.

Reading routes · Contents · Next chapter

2. The hidden variable couples through the heat-capacity singularity

Fisher considered a hidden variable that fluctuates in the ideal system but is constrained in the real one. The constraint equation contains an analytic part and a singular energy-like contribution. When the ideal specific-heat exponent α is positive, the singular term can dominate the mapping between physical and ideal reduced temperatures.

The leading relation behaves schematically as |t*|∝|t|^1/(1−α), where t is the constrained reduced temperature and t* the ideal thermal scaling field. Substituting this nonlinear map into ideal power laws divides several exponents by 1−α.

This is a thermodynamic coordinate effect with universal consequences, not an arbitrary rescaling chosen to improve a fit.

Reading routes · Previous chapter · Contents · Next chapter

3. The Fisher map changes some exponents and preserves others

For α>0 under the standard Fisher assumptions, the constrained exponents are α_X=−α/(1−α), β_X=β/(1−α), γ_X=γ/(1−α), and ν_X=ν/(1−α). The critical-isotherm exponent δ and anomalous dimension η remain unchanged.

The negative α_X means the constrained heat capacity has a finite cusp rather than the ideal divergence. Ratios such as γ/β are preserved because numerator and denominator acquire the same factor. The map is involutory when formulated consistently: applying the inverse constraint relation recovers the ideal exponents.

Reading routes · Previous chapter · Contents · Next chapter

4. Worked example: renormalise 3D Ising exponents

Use representative three-dimensional Ising values α=0.1096, β=0.32653, γ=1.2373 and ν=0.63012. The denominator is 1−α=0.8904.

The constrained values are α_X≈−0.1231, β_X≈0.3667, γ_X≈1.3896 and ν_X≈0.7077. δ≈4.7893 remains unchanged. These numbers are derived from one ideal exponent set; mixing estimates from incompatible analyses would spoil scaling checks.

The shift is systematic rather than uniform subtraction. Exponents of thermal distance are divided by 0.8904 because the constraint changes how fast the ideal scaling field approaches zero.

Reading routes · Previous chapter · Contents · Next chapter

5. Worked example: an order-parameter ratio

Let an ideal order parameter scale as m∝|t*|^β. Move the physical reduced temperature four times closer to criticality, so |t₂/t₁|=0.25. Without the constraint mapping, the ratio is 0.25^0.32653≈0.636.

With Fisher renormalisation, use β_X≈0.3667. The constrained ratio is 0.25^0.3667≈0.601. The order parameter falls farther over the same physical-temperature ratio because the constrained exponent is larger.

The amplitude cancels only if both points lie in the same asymptotic regime and use the same constraint. A crossover between ensembles or a drifting composition invalidates the simple ratio.

Reading routes · Previous chapter · Contents · Next chapter

6. Scaling relations survive in transformed form

Insert the Fisher values into Rushbrooke: α_X+2β_X+γ_X≈−0.1231+0.7334+1.3896≈1.9999. Widom also survives because δ is unchanged and γ_X/β_X=γ/β.

Hyperscaling becomes 2−α_X=dν_X when it held for the ideal system. With d=3, both sides are about 2.123. These checks show that Fisher renormalisation reorganises a consistent exponent set rather than changing individual exponents independently.

A reported “constrained exponent” that breaks the transformed scaling relations may reflect backgrounds, crossover, finite size or an incorrect ensemble identification.

Reading routes · Previous chapter · Contents · Next chapter

7. Transfer check: susceptibility grows more strongly

For susceptibility χ∝|t|^−γ, reducing |t| by a factor of four gives an ideal increase 4^1.2373≈5.56. Under the constraint, γ_X≈1.3896 gives 4^1.3896≈6.87.

The same temperature step therefore looks more singular in susceptibility while the constrained heat capacity is less singular, becoming a cusp. That combination is a distinctive consequence of the exponent map, not a contradiction.

The transfer uses the sign in the observable’s power law. Order parameters vanish with positive β; response functions grow with negative powers of γ or ν.

Reading routes · Previous chapter · Contents · Next chapter

8. Finite-range fits can mimic a new exponent

Experiments rarely reach arbitrarily small |t|. Analytic backgrounds, corrections to scaling and a crossover between unconstrained and constrained regimes can bend a log–log plot. Fitting one straight line across that bend produces an effective exponent that depends on the chosen window.

An apparent exponent between β and β_X does not by itself confirm Fisher renormalisation. Test whether the constraint is actually maintained, whether the conjugate field drifts, and whether several observables transform consistently.

Publish local slopes or moving-window fits. A genuine asymptotic mechanism should approach the predicted set, not merely offer one convenient slope over a short interval.

Reading routes · Previous chapter · Contents · Next chapter

9. The sign of α decides the leading outcome

When the ideal α is positive, the singular energy term can dominate the constraint equation and produce the leading Fisher exponent map. When α<0, the analytic term generally dominates, so the leading exponents remain unrenormalised, although constraint-induced corrections and slow crossover can still be important.

The marginal case α=0 requires special care because logarithmic corrections can replace a simple power-law argument. Kenna, Hsu and von Ferber extended Fisher renormalisation to logarithmic-correction exponents.

Do not divide by 1−α automatically for every constrained system. First verify the coupling, sign and dominance assumptions used in the derivation.

Reading routes · Previous chapter · Contents · Next chapter

10. Limits and misconceptions

• A fixed control variable matters only if it couples to the relevant scaling field.
• Fisher renormalisation is asymptotic, not a licence to refit any discrepancy.
• The ideal and constrained reduced temperatures are different coordinates.
• Positive, negative and zero α require different leading analyses.
• Corrections, finite size and imperfect constraint can create effective exponents.
• One transformed exponent is weaker evidence than a consistent transformed set.

Universality class and measurement ensemble are related but distinct ideas. A constraint can alter exponents expressed in the laboratory control variable while preserving the underlying fixed-point structure and scaling relations.

Reading routes · Previous chapter · Contents · Next chapter

11. Frequently asked questions

Does the constraint create a new universality class?

Not in the basic Fisher mechanism. It changes the thermodynamic path and exponent representation while retaining linked universal structure.

Why does α become negative?

The constraint suppresses the fluctuation responsible for the ideal heat-capacity divergence. The transformed singularity is a finite cusp with α_X<0.

Can δ stay unchanged while β and γ change?

Yes. δ describes the critical isotherm, while β and γ describe approach along the thermal field. The nonlinear thermal-coordinate map changes the latter pair together and preserves Widom scaling.

Reading routes · Previous chapter · Contents · Next chapter

12. A reproducible constrained-scaling checklist

The gold standard is to identify the ensemble before fitting, transform a complete exponent set, and show that several observables approach the same constrained theory. A constraint is a physical boundary condition, not an after-the-fact explanation.

Reading routes · Previous chapter · Contents · Continue reading

Continue through the Gold Standard ecosystem

Return to the Gold Standard of Thermodynamics, the Gold Standard of Education parent, or the How X Works hub.

Read next in this subject

• Critical Exponents, Scaling Laws and the Limits of Mean-Field Equations of State
• Spinodal Curvature, Metastable Branches and Nucleation Limits
• Helmholtz-Energy Equations of State, Residual Properties and Consistency Checks
• How Science Works | Why Different Materials Can Behave Alike Near a Critical Point

Authoritative reading

• Fisher, Renormalization of Critical Exponents by Hidden Variables
• Kenna, Hsu and von Ferber, Fisher renormalization for logarithmic corrections
• Izmailian and Kenna, Critical phenomena for systems under constraint
• Campostrini and colleagues, three-dimensional Ising critical exponents