Open problems of the renormalization-group treatment of the critical dynamics of the binary-fluid and gas-liquid transitions (model H): higher-order ε expansion, controlled d = 3 amplitudes, and nonuniversal backgrounds

The 1976 renormalization-group (RG) analysis by Siggia, Halperin, and Hohenberg established the asymptotic critical dynamics of the gas-liquid and binary-fluid transitions, i.e. the dynamic universality class of model H, to second order in the $\epsilon = 4 - d$ expansion: an exact scaling law tying the exponents of the diverging transport coefficients to static exponents for all $d < 4$, the exactness of the Kawasaki-Stokes diffusivity relation with a universal amplitude $R$ that differs slightly from the mode-coupling value, a weak shear-viscosity divergence in $d = 3$, and perturbative arguments that additional hydrodynamic modes and free-energy asymmetry are irrelevant. The analysis itself identified what it could not control: order $\epsilon^3$ terms and order $\epsilon^2$ Rayleigh-lineshape corrections for $k\xi \gg 1$ were not computed; the extrapolation from $d = 4$ to $d = 3$ was uncontrolled, with conflicting amplitude estimates and an unexplained discrepancy with the self-consistent Kawasaki-Lo solutions; the applicability of simplified model H to real fluids was supported only perturbatively; and nonuniversal backgrounds and vertex corrections were not quantified. Later work recomputed and substantially revised the two-loop values (Adzhemyan et al., 1999), but no three-loop model H calculation exists; the $d = 3$ question has since been pursued with molecular dynamics, real-time functional RG, and stochastic fluid-dynamics simulations (2016-2026), achieving partial but not controlled agreement with experiment (for example, a functional-RG viscosity exponent of about 0.047-0.06 versus the microgravity xenon measurement 0.0690(6)). The problem remains open.

Background

Near the critical point of a gas-liquid transition, or of a binary-fluid mixture (the consolute point below which two liquid components separate), equilibrium properties are governed by the Ising universality class, but time-dependent properties define a separate dynamic universality class. Dynamic critical phenomena concern critical slowing down — the divergence of the relaxation time of fluctuations as $\tau \propto \xi^{z}$, which defines the dynamic exponent $z$, where $\xi$ is the correlation length — together with the singular behavior of transport coefficients: the order-parameter diffusivity $D$ (the interdiffusion constant of a mixture, or the thermal diffusivity of a pure fluid) vanishes at the critical point, while the shear viscosity diverges weakly. A light-scattering experiment measures the Rayleigh line, whose linewidth is the decay rate of order-parameter fluctuations with wave number $k$, a function of the scaling variable $k\xi$.

Halperin and Hohenberg organized critical dynamics into dynamic universality classes (models A through J). The class relevant for fluids, model H, consists of a conserved scalar order parameter $\psi$ (a concentration difference or the entropy density) coupled reversibly, through a coupling $g$, to the conserved transverse momentum density (the velocity field) of the fluid. It contains two kinetic coefficients: the order-parameter mobility $\lambda$, related to the diffusivity by $D = \lambda/\chi_\psi$ with $\chi_\psi$ the static susceptibility, and the shear viscosity.

Before 1976 the theory rested on uncontrolled mode-coupling and dynamic-scaling arguments (Kadanoff and Swift; Kawasaki), i.e. on treating the decay of each fluctuation mode as driven by nonlinear coupling to pairs of other modes. These produced the Kawasaki scaling function $\Omega_K(x)$ for the Rayleigh linewidth (with $x = k\xi$), treated the correlation length as an effective Stokes radius (the radius of a hard sphere whose Stokes drag reproduces the observed diffusivity), and led to the Kawasaki-Stokes relation for the diffusivity,

$D = R\,\frac{k_B T_c}{\bar{\eta}\,\xi^{d-2}},$

in spatial dimension $d$, where $k_B$ is Boltzmann's constant, $T_c$ the critical temperature, $\bar{\eta}$ the shear viscosity, and $R$ a dimensionless amplitude. In the simplest approximation, which neglects the weak critical divergence of the shear viscosity, the estimate is $R = (6\pi)^{-1} \approx 0.053$ in $d = 3$. Refinements computed vertex corrections and frequency-dependent (nonlocal, time-retarded) viscosities self-consistently (Kawasaki and Lo; Ohta and Kawasaki; Ohta). The controlled alternative was the epsilon expansion, an expansion in $\epsilon = 4 - d$ about the upper critical dimension $d = 4$ (the dimension above which long-wavelength fluctuations are weak and mean-field theory becomes exact), combined with the renormalization-group (RG) procedure of integrating out fluctuation modes and rescaling.

Siggia, Halperin, and Hohenberg (1976) applied this program to model H. They proved an exact scaling law tying the exponents of the diverging transport coefficients to static exponents for all $d < 4$ (a relation anticipated by mode-coupling arguments), showed that the Kawasaki-Stokes relation holds exactly but with a universal amplitude that differs slightly from its mode-coupling value, computed the transport exponents and the amplitude to second order in $\epsilon$ (predicting a weak divergence of the shear viscosity, $\eta \propto (T - T_c)^{-0.04}$ in $d = 3$, which explains why the Kawasaki approximation, which neglects it, agrees so well with experiment), and argued that additional hydrodynamic modes (sound, longitudinal currents) and free-energy asymmetry are irrelevant at the fixed point, so that simplified model H shares the asymptotic dynamics of real fluids. They also reviewed the corrections to Kawasaki theory and made suggestions for refining those calculations. The open questions listed here are precisely the limitations that the 1976 analysis identified in itself: higher-order $\epsilon$ terms were not computed; the extrapolation from the expansion about $d = 4$ to the physical dimension $d = 3$ was uncontrolled; the irrelevance arguments were perturbative only; and the nonuniversal background parts of the transport coefficients, which dominate outside the asymptotic regime actually probed by experiment, were not quantified.

Problem Statement

The common objective is a quantitatively reliable and error-controlled theory of the critical dynamics of the gas-liquid and binary-fluid transitions in the physical dimension $d = 3$, within (or explicitly beyond) the renormalization-group treatment of model H given by Siggia, Halperin, and Hohenberg. The following linked questions, identified as open by that analysis, remain to be answered.

(a) Higher-order epsilon expansion: compute the contributions beyond second order in $\epsilon = 4 - d$ that the 1976 work did not evaluate, in particular the order $\epsilon^3$ terms in the renormalization-group flow, the transport exponents, and the amplitudes (including the further effects of the dissipative self-coupling of the order parameter that enter at this order), and the order $\epsilon^2$ corrections to the Rayleigh-lineshape (linewidth) scaling function in the critical regime $k\xi \gg 1$.

(b) Controlled extrapolation to $d = 3$: resolve the ambiguity in passing from the $\epsilon$ expansion to three dimensions, where direct substitution $\epsilon = 1$ and hybrid recipes combining self-consistent approximations with perturbation theory give different numerical amplitudes (for example different values of the universal amplitude $R$ in the Kawasaki-Stokes relation $D = R\,k_B T_c/(\bar{\eta}\,\xi^{d-2})$, with the direct value 0.042 and the hybrid value anchored on the Kawasaki estimate $(6\pi)^{-1} \approx 0.053$), and explain the origin of the discrepancies with numerical self-consistent solutions such as those of Kawasaki and Lo. Deliver a prescription, with an error estimate, for the $d = 3$ values of the universal amplitudes and exponents.

(c) Applicability of model H to real fluids: establish whether the additional hydrodynamic couplings (sound and longitudinal modes) and the free-energy asymmetry of real fluids are dynamically irrelevant in $d = 3$ in the asymptotic critical regime, as the $\epsilon$ expansion near $d = 4$ suggests, so that the simplified model H shares the dynamics of all real fluids arbitrarily close to $T_c$.

(d) Nonuniversal backgrounds and controlled corrections: quantify the nonuniversal background contributions (background viscosity and background diffusivity) and the correct background-subtraction procedure in the experimentally accessible temperature range, accounting for frequency dependence, background terms, static Fisher-Burford corrections (an approximate form of the static correlation function that corrects the standard mean-field, Ornstein-Zernike form of the susceptibility away from the asymptotic regime), and vertex (proper-diagram) corrections, with controlled estimates of the omitted terms, so that the theoretical linewidths and transport coefficients can be compared with experiment beyond the asymptotic regime with the few-percent accuracy the 1976 analysis aimed at.

Items (a)-(d) form one program: they jointly deliver the controlled quantitative predictions for three-dimensional fluids that the 1976 analysis could only estimate, and each item supplies an ingredient (expansion coefficients, extrapolation with error control, model validation, background separation) that the others require for a complete, experimentally testable prediction.

References

1. E. D. Siggia, B. I. Halperin, and P. C. Hohenberg, "Renormalization-group treatment of the critical dynamics of the binary-fluid and gas-liquid transitions," Physical Review B 13, 2110-2123 (1976). DOI: 10.1103/PhysRevB.13.2110, https://doi.org/10.1103/PhysRevB.13.2110.
2. P. C. Hohenberg and B. I. Halperin, "Theory of dynamic critical phenomena," Reviews of Modern Physics 49, 435-479 (1977). DOI: 10.1103/RevModPhys.49.435, https://doi.org/10.1103/RevModPhys.49.435.
3. Leo P. Kadanoff and Jack Swift, "Transport Coefficients near the Liquid-Gas Critical Point," Physical Review 166, 89-101 (1968). DOI: 10.1103/PhysRev.166.89, https://doi.org/10.1103/PhysRev.166.89.
4. Kyozi Kawasaki, "Dynamics of Critical Fluctuations. I," Progress of Theoretical Physics 39, 1133-1152 (1968). DOI: 10.1143/PTP.39.1133, https://doi.org/10.1143/PTP.39.1133.
5. Kyozi Kawasaki and Shih-Min Lo, "Nonlocal Shear Viscosity and Order-Parameter Dynamics near the Critical Point of Fluids," Physical Review Letters 29, 48-51 (1972). DOI: 10.1103/PhysRevLett.29.48, https://doi.org/10.1103/PhysRevLett.29.48.
6. Shih-Min Lo and Kyozi Kawasaki, "Vertex Correction Contribution to the Decay Rate of Concentration Fluctuations in Binary Liquid Critical Mixtures," Physical Review A 5, 421-424 (1972). DOI: 10.1103/PhysRevA.5.421, https://doi.org/10.1103/PhysRevA.5.421.
7. Shih-Min Lo and Kyozi Kawasaki, "Frequency-Dependence Correction to the Order-Parameter Decay Rates near the Critical Point of Fluids," Physical Review A 8, 2176-2177 (1973). DOI: 10.1103/PhysRevA.8.2176, https://doi.org/10.1103/PhysRevA.8.2176.
8. T. Ohta and K. Kawasaki, "Mode Coupling Theory of Dynamic Critical Phenomena for Classical Liquids. I: Dynamic Critical Exponents," Progress of Theoretical Physics 55, 1384-1395 (1976). DOI: 10.1143/PTP.55.1384, https://doi.org/10.1143/PTP.55.1384.
9. T. Ohta, "Mode Coupling Theory of Dynamic Critical Phenomena for Classical Liquids. II: Non-Markoffian Effects," Progress of Theoretical Physics 64, 536-543 (1980). DOI: 10.1143/PTP.64.536, https://doi.org/10.1143/PTP.64.536.
10. H. C. Burstyn, J. V. Sengers, J. K. Bhattacharjee, and R. A. Ferrell, "Dynamic scaling function for critical fluctuations in classical fluids," Physical Review A 28, 1567-1578 (1983). DOI: 10.1103/PhysRevA.28.1567, https://doi.org/10.1103/PhysRevA.28.1567.
11. L. Ts. Adzhemyan, A. N. Vasiliev, Yu. S. Kabrits, and M. V. Kompaniets, "H-model of critical dynamics: Two-loop calculations of RG functions and critical indices," Theoretical and Mathematical Physics 119, 454-470 (1999). DOI: 10.1007/BF02557344, https://doi.org/10.1007/BF02557344.
12. L. Ts. Adzhemyan and A. N. Vasiliev, "H-model of critical dynamics: Choice of dynamic variables, elimination of sound modes, and equations for sound waves in the neighborhood of T_c," Theoretical and Mathematical Physics 117, 1223-1239 (1998). DOI: 10.1007/BF02557163, https://doi.org/10.1007/BF02557163.
13. R. Folk and G. Moser, "Critical dynamics in mixtures," Physical Review E 58, 6246-6274 (1998). DOI: 10.1103/PhysRevE.58.6246, https://doi.org/10.1103/PhysRevE.58.6246.
14. R. F. Berg, M. R. Moldover, and G. A. Zimmerli, "Frequency-dependent viscosity of xenon near the critical point," Physical Review E 60, 4079-4098 (1999). DOI: 10.1103/PhysRevE.60.4079, https://doi.org/10.1103/PhysRevE.60.4079.
15. Sutapa Roy, S. Dietrich, and Felix Höfling, "Structure and dynamics of binary liquid mixtures near their continuous demixing transitions," The Journal of Chemical Physics 145, 134505 (2016). DOI: 10.1063/1.4963771, arXiv:1606.05595, https://arxiv.org/abs/1606.05595.
16. Yong-rui Chen, Yang-yang Tan, and Wei-jie Fu, "Critical dynamics of Model H within the real-time FRG approach," Physical Review D 111, 094025 (2025). DOI: 10.1103/PhysRevD.111.094025, arXiv:2406.00679, https://arxiv.org/abs/2406.00679.
17. Chandrodoy Chattopadhyay, Josh Ott, Thomas Schäfer, and Vladimir V. Skokov, "Simulations of Stochastic Fluid Dynamics near a Critical Point in the Phase Diagram," Physical Review Letters 133, 032301 (2024). DOI: 10.1103/PhysRevLett.133.032301, arXiv:2403.10608, https://arxiv.org/abs/2403.10608.
18. Johannes V. Roth, Yunxin Ye, Sören Schlichting, and Lorenz von Smekal, "Universal scaling of transport coefficients near the liquid-gas critical point," arXiv:2603.17874 (2026). DOI: 10.48550/arXiv.2603.17874, https://arxiv.org/abs/2603.17874.