Cluster-expansion description of condensation for continuum fluids: does the Mayer/virial singularity coincide with gas-liquid coexistence, and can the cluster formalism reach the condensed phase?

Around Mayer's cluster theory of condensation, Max Born's 1937 Physica report posed a program of questions: explicit evaluation of irreducible cluster integrals for realistic potentials, extension of the cluster formalism to the condensed (liquid) regime where cluster integrals may depend on the total volume, a rigorous statistical-mechanical treatment of gas-liquid equilibrium, and a precise domain of validity for the saddle-point asymptotics. Numerical evaluation of cluster integrals (thread 1) is now routine for gas-phase coefficients via Mayer-sampling Monte Carlo (Lennard-Jones coefficients through sixteenth order), and convergence-validity (thread 4) was transformed into rigorous convergence theory (Penrose; Lebowitz-Penrose; Jansen-Kolesnikov), though the Groeneveld-Brydges conjecture on the virial domain remains open at best for non-negative potentials. The surviving open core (threads 2-3) is: for continuum fluids in three dimensions with genuinely short-range, stable, finite-range pair potentials of the Lennard-Jones class, (i) decide whether the leading singularity of the Mayer activity series (and of the virial series) coincides with the onset of condensation or is strictly separated from it; (ii) derive gas-liquid coexistence (non-analytic free energy, flat isotherm with Maxwell equal-area construction, saturated pressure) rigorously for such a potential, or establish a no-go result; and (iii) delimit or exclude any extension of the cluster-expansion formalism into the condensed regime with volume-dependent cluster integrals. All rigorous coexistence results to date hold only in the Kac (mean-field) limit, with unphysical multi-body interactions, with special liquid-vapor symmetry, or in dilute low-temperature scaling windows.

Background

The classical theory of the liquid-gas transition begins with the observation that the thermodynamics of a dilute gas admits a systematic expansion. In the grand-canonical ensemble, the pressure of a fluid of N interacting particles in volume V at temperature T is an analytic function of the fugacity $z=\exp(\mu/k_BT)$ (with $\mu$ the chemical potential), whose Taylor coefficients are the cluster integrals $b_l$: configuration-space integrals over connected clusters of l molecules, built from the Mayer bond function $f(r)=\exp(-v(r)/k_BT)-1$ where $v(r)$ is the pair potential. Re-expanding the pressure in powers of the density $\rho$ gives the virial expansion, whose coefficients (experimentally measurable as deviations from the ideal gas law) are expressed through the irreducible cluster integrals $\beta_\nu$ (cluster integrals whose bond diagrams cannot be split by removing one molecule); e.g. $B_2=-\beta_1/2$ and $B_3=-2\beta_2/3$. J. E. Mayer proposed (1937) that condensation be read off directly from this formalism: the equation of state is parametrized by the roots $z(v)$ of a generating relation, and condensation was identified with the coalescence of roots, i.e. with a singularity of the cluster series at the saturated fugacity. Max Born's Physica 1937 report, presented at the 1937 Amsterdam conference commemorating van der Waals, recast the combinatorial coefficient as a contour integral evaluated by the steepest-descent (saddle-point) method — the standard asymptotic technique of expanding an integral around the point where its integrand is extremal — examined Mayer's criterion, and recorded in a note added in proof that the large-N asymptotic formula ceases to be valid for specific volumes below a critical value $v_s$. The report's documented reception (the Born-Fuchs 1938 Proceedings paper) records the central contemporary doubt in the words: how can the gas molecules 'know' when they have to coagulate to form a liquid? Four program-level questions thereby attach to the source: (1) explicit numerical evaluation of $\beta_\nu$ for realistic intermolecular potentials; (2) extension of the combinatorial and asymptotic formulae to the condensed regime, where cluster integrals may depend explicitly on the total volume V; (3) a rigorous treatment of gas-liquid equilibrium, in particular whether singularity of the cluster generating functions corresponds to a true thermodynamic phase transition; (4) a precise delineation of the validity of the saddle-point asymptotics. The modern rigorous framework for these questions is equilibrium statistical mechanics in the thermodynamic limit ($N,V\to\infty$ at fixed $\rho$): a phase transition is a non-analyticity of the limiting free energy, and coexistence of gas and liquid means a flat $\mu$-isotherm (Maxwell's equal-area construction replacing the van der Waals loop) with a saturated pressure. A Kac potential is a pair potential whose range is taken to infinity while its strength is scaled to zero, so that the interaction becomes exactly mean-field; it is precisely the regime in which the Maxwell construction has been proven (Lebowitz-Penrose 1966), and it is not the regime of real liquids.

Problem Statement

For a classical continuum fluid in three dimensions whose molecules interact through a stable, finite-range pair potential $v(r)$ of genuinely short range with an attractive tail (finite range means that $v$ has compact support, i.e. vanishes beyond some radius $b$, following the convention of Jansen 2012; stability means that the total interaction energy of any $N$ particles is bounded from below by $-BN$ for some constant $B\ge 0$ independent of $N$ and of the configuration; the attractive tail means that $v(r)<0$ for $r$ in some interval just inside the edge of the support, so that purely repulsive fluids are excluded; the class is motivated by the Lennard-Jones form $v(r)=A/r^{12}-B/r^{6}$, a repulsive core plus attractive tail, suitably truncated beyond a finite radius, since the untruncated $r^{-6}$ tail has infinite range; in particular not a Kac-type long-range potential, not supplemented by multi-body interactions, and not a special-symmetry model such as the Widom-Rowlinson model of two species with a repulsive hard core between unlike species), resolve the cluster-theoretic description of condensation left open by the 1937 program: (1) Prove or disprove that the leading singularity of the Mayer activity expansion (respectively, the boundary of the domain of convergence of the virial expansion) coincides with the onset of gas-liquid condensation. In particular, settle for which potentials the bounds $\beta^{-1}\log R^{May}(\beta)\le\mu^{sat}(\beta)$ and $\max(R^{vir}(\beta),\rho^{May}(\beta))\le\rho^{sat}(\beta)$ of Jansen 2012 are strict — where $\beta=(k_BT)^{-1}$ is the inverse temperature, $R^{May}(\beta)$ and $R^{vir}(\beta)$ are the radii of convergence of the activity and virial series, and $\mu^{sat}(\beta)$, $\rho^{sat}(\beta)$ are the chemical potential and density at the onset of condensation — and determine what the radii of convergence physically represent when they do not correspond to non-analyticities of the free energy — the open target being the class-level strictness classification, taken over the class members for which the onset of condensation exists (for a class member that does not condense, $\mu^{sat}$ and $\rho^{sat}$ are undefined and the strictness question does not arise). (2) Derive the gas-liquid coexistence region rigorously within this class (non-analytic free energy, flat $\mu$-isotherm with the Maxwell equal-area construction and a sharp saturated vapor pressure), or prove that such a derivation is impossible within the stated class. (3) Delimit or exclude an extension of the cluster-expansion formalism into the condensed regime, where the cluster integrals $b_l$ and irreducible integrals $\beta_\nu$ acquire explicit dependence on the total volume $V$, so that the equation of state is described from first principles across gaseous and condensed regimes. A rigorous result is required here: either a volume-dependent cluster expansion whose equation of state agrees with the Mayer series in the gas phase and continues across the transition to describe coexistence, or a rigorous impossibility statement with this scope; a purely formal or heuristic expansion does not resolve the thread. The stated short-range continuum class is the target throughout: the Kac (mean-field) limit, dilute low-temperature scaling windows, and lattice or special-symmetry models are outside its scope.

References

1. Max Born, 'The statistical mechanics of condensing systems', Physica 4, 1034-1044 (1937). DOI 10.1016/S0031-8914(37)80201-3, https://doi.org/10.1016/S0031-8914(37)80201-3
2. Joseph E. Mayer, 'The Statistical Mechanics of Condensing Systems. I', The Journal of Chemical Physics 5, 67-73 (1937). DOI 10.1063/1.1749933, https://doi.org/10.1063/1.1749933
3. Joseph E. Mayer and Philip G. Ackermann, 'The Statistical Mechanics of Condensing Systems. II', The Journal of Chemical Physics 5, 74-83 (1937). DOI 10.1063/1.1749934, https://doi.org/10.1063/1.1749934
4. Max Born and Klaus Fuchs, 'The statistical mechanics of condensing systems', Proceedings of the Royal Society of London. Series A 166, 391-414 (1938). DOI 10.1098/rspa.1938.0100, https://doi.org/10.1098/rspa.1938.0100
5. C. N. Yang and T. D. Lee, 'Statistical Theory of Equations of State and Phase Transitions. I. Theory of Condensation', Physical Review 87, 404-409 (1952). DOI 10.1103/PhysRev.87.404, https://doi.org/10.1103/PhysRev.87.404
6. T. D. Lee and C. N. Yang, 'Statistical Theory of Equations of State and Phase Transitions. II. Lattice Gas and Ising Model', Physical Review 87, 410-419 (1952). DOI 10.1103/PhysRev.87.410, https://doi.org/10.1103/PhysRev.87.410
7. Oliver Penrose, 'Convergence of Fugacity Expansions for Fluids and Lattice Gases', Journal of Mathematical Physics 4, 1312-1320 (1963). DOI 10.1063/1.1703906, https://doi.org/10.1063/1.1703906
8. J. L. Lebowitz and O. Penrose, 'Convergence of Virial Expansions', Journal of Mathematical Physics 5, 841-847 (1964). DOI 10.1063/1.1704186, https://doi.org/10.1063/1.1704186
9. J. L. Lebowitz and O. Penrose, 'Rigorous Treatment of the Van Der Waals-Maxwell Theory of the Liquid-Vapor Transition', Journal of Mathematical Physics 7, 98-113 (1966). DOI 10.1063/1.1704821, https://doi.org/10.1063/1.1704821
10. David Ruelle, 'Existence of a Phase Transition in a Continuous Classical System', Physical Review Letters 27, 1040-1041 (1971). DOI 10.1103/PhysRevLett.27.1040, https://doi.org/10.1103/PhysRevLett.27.1040
11. J. L. Lebowitz, A. E. Mazel and E. Presutti, 'Rigorous Proof of a Liquid-Vapor Phase Transition in a Continuum Particle System', Physical Review Letters 80, 4701-4704 (1998). DOI 10.1103/PhysRevLett.80.4701, https://doi.org/10.1103/PhysRevLett.80.4701 (also arXiv:cond-mat/9809144)
12. Andrea Collevecchio, Wolfgang König, Peter Mörters and Nadia Sidorova, 'Phase Transitions For Dilute Particle Systems with Lennard-Jones Potential', Communications in Mathematical Physics 299, 603-630 (2010). DOI 10.1007/s00220-010-1097-5, https://doi.org/10.1007/s00220-010-1097-5
13. Jayant K. Singh and David A. Kofke, 'Mayer Sampling: Calculation of Cluster Integrals using Free-Energy Perturbation Methods', Physical Review Letters 92, 220601 (2004). DOI 10.1103/PhysRevLett.92.220601, https://doi.org/10.1103/PhysRevLett.92.220601
14. Andrew J. Schultz and David A. Kofke, 'Sixth, seventh and eighth virial coefficients of the Lennard-Jones model', Molecular Physics 107, 2309-2318 (2009). DOI 10.1080/00268970903267053, https://doi.org/10.1080/00268970903267053
15. Chao Feng, Andrew J. Schultz, Vipin Chaudhary and David A. Kofke, 'Eighth to sixteenth virial coefficients of the Lennard-Jones model', The Journal of Chemical Physics 143, 044504 (2015). DOI 10.1063/1.4927339, https://doi.org/10.1063/1.4927339
16. Sabine Jansen, 'Mayer and Virial Series at Low Temperature', Journal of Statistical Physics 147, 678-706 (2012). DOI 10.1007/s10955-012-0490-1, https://doi.org/10.1007/s10955-012-0490-1 (also arXiv:1109.6568)
17. Sabine Jansen, 'Revisiting Groeneveld's approach to the virial expansion', Journal of Mathematical Physics 62, 023302 (2021). DOI 10.1063/5.0030148, https://doi.org/10.1063/5.0030148 (also arXiv:2009.09211)
18. Sabine Jansen and Leonid Kolesnikov, 'Cluster expansions: Necessary and sufficient convergence conditions', Journal of Statistical Physics 189, 33 (2022). DOI 10.1007/s10955-022-02992-6, https://doi.org/10.1007/s10955-022-02992-6 (also arXiv:2112.13134)
19. Qidong He, Ian Jauslin, Joel Lebowitz and Ron Peled, 'Liquid-vapor transition in a model of a continuum particle system with finite-range modified Kac pair potential' (2025). arXiv:2510.24825, https://arxiv.org/abs/2510.24825
20. Kiyoharu Kawana, 'On Virial Expansion in Hard Sphere Model' (2025). arXiv:2501.15766, https://arxiv.org/abs/2501.15766