The ball-piston gas of Bálint, Gilbert, Nándori, Szász and Tóth (2016/2017) is a deterministic billiard model of heat conduction built from hybrid cells — point-particle 'balls' moving in two-dimensional dispersing cells and one-dimensional 'pistons' penetrating neighbouring cells — designed so that, in a rare-interaction limit, a rigorous derivation of a mesoscopic stochastic energy process and thence of Fourier's law becomes realistically attainable. The source paper computed the transition kernel and master equation of the conjectured limiting Markov jump process and supported the Markov property numerically, but left four issues explicitly open: a rigorous proof of the micro-to-meso convergence (the minimal ball-piston pair is a three-dimensional semi-dispersing billiard with corner points), the hydrodynamic limit deriving Fourier's law with a fully justified heat conductivity including the dynamical-correlation contribution, a spectral-gap lower bound for the multi-cell generator, and rigorous control of extended chains and their collective transport coefficients.

Background

Fourier's law — the statement that heat flux through a material is proportional to the temperature gradient, with proportionality constant the heat conductivity — has never been derived with full mathematical rigor from the deterministic Newtonian dynamics of an interacting many-particle system; this is a long-standing challenge of mathematical statistical physics. A promising strategy, introduced by Gaspard and Gilbert in 2008 for a class of models now called GG-models, proceeds in two steps. A GG-model is a periodic network of cells bounded by dispersing walls (walls built from convex scatterers, which defocus bundles of trajectories), each confining a single point particle; neighbouring particles can collide through the interface between their cells, so energy — but not mass — is transported. Step one (micro-to-meso) takes the rare-interaction limit, in which a geometric parameter δ\delta controlling the interface openings is sent to zero: wall collisions within each cell then dominate, each cell relaxes to its microcanonical equilibrium (the uniform measure on the constant-energy surface) long before the next inter-cell collision, and the energy-exchange process is expected to converge, under the time rescaling that compensates the vanishing collision rate, to a Markov jump process (a continuous-time piecewise-constant random process) on the energy simplex (the set of admissible partitions of the conserved total energy among the cells) with an explicit generator. Step two (meso-to-macro) takes the hydrodynamic (diffusive) limit of this jump process on long chains to obtain the heat equation — Fourier's law — together with an expression for the heat conductivity.

The ball-piston gas is the member of the GG family designed to make step one mathematically realistic. Its cells are hybrid: a 'ball' is a point particle moving in a two-dimensional cell bounded by circular scatterers, while a 'piston' is a line segment constrained to move back and forth along a one-dimensional interval that partially penetrates the two neighbouring ball cells; energy moves between balls only through successive ball-piston collisions. The minimal system of one ball and one piston is isomorphic to a three-dimensional semi-dispersing billiard (a billiard in dimension three whose boundary is flat in some directions and convex in others) with corner points (non-smooth boundary points where two wall pieces meet at a non-zero angle), and the full gas is a one-dimensional chain of such units. The source paper derived, by geometric and measure-theoretic calculations on the collision surface, the conditional mean free times and the transition kernel of the conjectured limiting process — its master equation (eq. (32) there) gives the rates WW for energy transfers between a ball and a piston — and devised a numerical test of the Markov property based on the Kullback-Leibler divergence (a relative-entropy distance between probability distributions), with affirmative results as δ\delta decreases. The convergence itself, however, was established only heuristically. The authors identified the method of standard pairs — probability measures supported on one-dimensional unstable manifolds of billiard flows, the standard tool for controlling such highly singular initial data — as the intended route, noting that the corner points of the three-dimensional billiard pose the main obstacle.

On the macroscopic side, the conductivity of an energy-exchange process is given by a Green-Kubo formula (an exact expression of transport coefficients in terms of time integrals of equilibrium current correlations), which splits into a static contribution, expressible through one-step quantities of the kernel, and a dynamical-correlation contribution. Varadhan's variational formula characterizes this Green-Kubo conductivity as an infimum over local functions and is the main tool for its rigorous evaluation; Gaspard and Gilbert applied it to evaluate the dynamical contribution for the three-dimensional GG model of locally confined hard spheres (finding it small but nonzero), but stated that transposing the variational formula to the mixed ball-piston system poses technical problems that have yet to be overcome. Sasada proved for a class of stochastic energy-exchange models generalizing the GG processes that the conductivity equals the static contribution if and only if the model satisfies the gradient condition — the existence of a local function whose discrete gradient equals the microscopic current — and that the GG-derived models fail it, so their true conductivity includes a nonzero dynamical contribution. For the GG models of locally confined hard discs and hard spheres she also proved an N−2N^{-2} lower bound on the spectral gap of the NN-site generator (the gap between the two largest eigenvalues of the Markov generator, controlling relaxation; such bounds are key inputs to hydrodynamic-limit proofs), extending earlier mixing-rate results of Grigo, Khanin and Szász. Exponential decay of correlations for planar Sinai billiard flows with finite horizon (a billiard on a table with convex scatterers in which every free flight between collisions has uniformly bounded length) — the local-equilibration input needed in step one, valid also for standard-pair initial measures — was proven by Baladi, Demers and Liverani and extended to standard pairs by Bálint, Nándori, Szász and Tóth; both results, however, exclude tables with corner points.

Problem Statement

Complete, with full mathematical rigor, the limiting program for the rarely interacting ball-piston gas, as explicitly deferred by Bálint–Gilbert–Nándori–Szász–Tóth (2017). The single scientific objective is the rigorous derivation of the macroscopic heat-transport behavior of this deterministic billiard system from its Newtonian microscopic dynamics, and it decomposes into three linked stages fixed by the two-step strategy:

  1. Micro-to-meso limit. Prove that, for the multi-cell ball-piston chain, the continuous-time process of the particles' energies — jumping at the binary ball-piston collision events, with time rescaled to compensate the vanishing binary-collision rate — converges in distribution, as the rare-interaction parameter δ→0\delta\to0, to the Markov jump process on the energy simplex whose generator is the master equation with the pairwise rates WW computed in the source paper (its eqs. (30)–(32)). The proof must handle the semi-dispersing three-dimensional billiard with corner points to which the minimal ball-piston pair is isomorphic — for instance by standard-pair or Markov-approximation techniques — and must show that the limiting jump rates are independent of the non-equilibrium history of the energy partition.

  2. Spectral gap. Prove a lower bound, expected to be of order N−2N^{-2}, on the spectral gap of the Markov generator of the NN-cell ball-piston energy process — the analogue of Sasada's bound for GG-models, made harder by the hybrid two-dimensional/one-dimensional nature of the cells.

  3. Meso-to-macro limit. Derive the hydrodynamic limit of the multi-cell energy process — Fourier's law under diffusive scaling — and justify a precise expression for the heat conductivity of the ball-piston gas, including both the static contribution and the dynamical-correlation contribution, whose evaluation requires transposing the variational formula to the mixed ball-piston kernel.

Stage 1 provides the input of stages 2–3, and stages 2–3 are the two ingredients (relaxation control and transport-coefficient identification) of the hydrodynamic limit; the extension to chains and lattices of many ball-piston cells is the natural setting of stages 2–3 and belongs to the same objective. A complete answer establishes Fourier's law with a justified conductivity for the deterministic ball-piston gas; each stage is scientifically meaningful on its own, but partial progress does not discharge the stated objective.

References

  1. Péter Bálint, Thomas Gilbert, Péter Nándori, Domokos Szász, Imre Péter Tóth, 'On the Limiting Markov Process of Energy Exchanges in a Rarely Interacting Ball-Piston Gas', Journal of Statistical Physics 166(3–4), 903–925 (2017). DOI: 10.1007/s10955-016-1598-5; arXiv:1510.06408. https://arxiv.org/abs/1510.06408
  2. Péter Bálint, Péter Nándori, Domokos Szász, Imre Péter Tóth, 'Equidistribution for Standard Pairs in Planar Dispersing Billiard Flows', Annales Henri Poincaré 19, 979–1042 (2018). DOI: 10.1007/s00023-018-0648-8; arXiv:1707.03068. https://arxiv.org/abs/1707.03068
  3. Péter Bálint, Thomas Gilbert, Domokos Szász, Imre Péter Tóth, 'What mathematical billiards teach us about statistical physics?' (2020). arXiv:2009.06284. https://arxiv.org/abs/2009.06284
  4. Mark F. Demers, Carlangelo Liverani, 'Recent Progress in the Application of Transfer Operators to Dispersing Billiards' (2026). arXiv:2606.10155. https://arxiv.org/abs/2606.10155
  5. Pierre Gaspard, Thomas Gilbert, 'Dynamical contribution to the heat conductivity in stochastic energy exchanges of locally confined gases', Journal of Statistical Mechanics: Theory and Experiment (2017) 043210. DOI: 10.1088/1742-5468/aa60ce; arXiv:1611.05809. https://arxiv.org/abs/1611.05809
  6. Makiko Sasada, 'Thermal conductivity for stochastic energy exchange models' (2016). arXiv:1611.08866. https://arxiv.org/abs/1611.08866
  7. Makiko Sasada, 'Spectral gap for stochastic energy exchange model with nonuniformly positive rate function', The Annals of Probability 43(4), 1663–1711 (2015). DOI: 10.1214/14-AOP916; arXiv:1305.4066. https://doi.org/10.1214/14-AOP916
  8. Makiko Sasada, 'On the Green–Kubo formula and the gradient condition on currents', The Annals of Applied Probability 28(5) (2018). DOI: 10.1214/17-AAP1369. https://doi.org/10.1214/17-AAP1369
  9. A. Grigo, K. Khanin, D. Szász, 'Mixing rates of particle systems with energy exchange', Nonlinearity 25(8), 2349–2376 (2012). DOI: 10.1088/0951-7715/25/8/2349. https://doi.org/10.1088/0951-7715/25/8/2349
  10. P. Gaspard, T. Gilbert, 'Heat Conduction and Fourier's Law by Consecutive Local Mixing and Thermalization', Physical Review Letters 101, 020601 (2008). DOI: 10.1103/PhysRevLett.101.020601. https://doi.org/10.1103/PhysRevLett.101.020601
  11. Pierre Gaspard, Thomas Gilbert, 'On the derivation of Fourier's law in stochastic energy exchange systems', Journal of Statistical Mechanics: Theory and Experiment (2008) P11021. DOI: 10.1088/1742-5468/2008/11/P11021. https://doi.org/10.1088/1742-5468/2008/11/P11021
  12. Péter Nándori, 'Local Equilibrium in Inhomogeneous Stochastic Models of Heat Transport', Journal of Statistical Physics 164, 410–437 (2016). DOI: 10.1007/s10955-016-1551-7; arXiv:1603.04960. https://arxiv.org/abs/1603.04960
  13. Dmitry Dolgopyat, Péter Nándori, 'The first encounter of two billiard particles of small radius' (2016). arXiv:1603.07590. https://arxiv.org/abs/1603.07590
  14. Wael Bahsoun, Fanni M. Sélley, 'Map Lattices Coupled by Collisions: Hitting Time Statistics and Collisions Per Lattice Unit', Annales Henri Poincaré 23, 2919–2947 (2022). DOI: 10.1007/s00023-022-01164-2; arXiv:2104.10233. https://arxiv.org/abs/2104.10233
  15. Wael Bahsoun, Maxence Phalempin, 'Rare Events Statistics for Zd\mathbb{Z}^d Map Lattices Coupled by Collision', Communications in Mathematical Physics 407, article 57 (2026). DOI: 10.1007/s00220-026-05557-w; arXiv:2412.12803. https://arxiv.org/abs/2412.12803
  16. D. Szász, 'Markov approximations and statistical properties of billiards', in: The Abel Prize 2013–2017, Springer, 299–319 (2019). DOI: 10.1007/978-3-319-99028-6_13; arXiv:1702.01261. https://arxiv.org/abs/1702.01261
  17. Viviane Baladi, Mark F. Demers, Carlangelo Liverani, 'Exponential decay of correlations for finite horizon Sinai billiard flows', Inventiones mathematicae 211, 39–177 (2018). DOI: 10.1007/s00222-017-0745-1. https://doi.org/10.1007/s00222-017-0745-1
  18. Jacopo De Simoi, Imre Péter Tóth, 'An Expansion Estimate for Dispersing Planar Billiards with Corner Points', Annales Henri Poincaré 15, 1223–1243 (2014). DOI: 10.1007/s00023-013-0272-6. https://doi.org/10.1007/s00023-013-0272-6
  19. Lorenzo Caprini, Luca Cerino, Alessandro Sarracino, Angelo Vulpiani, 'Fourier's Law in a Generalized Piston Model', Entropy 19(7), 350 (2017). DOI: 10.3390/e19070350; arXiv:1707.04023. https://doi.org/10.3390/e19070350
  20. Yao Li, Lingchen Bu, 'From billiards to thermodynamic laws: Stochastic energy exchange model', Chaos 28(9), 093105 (2018). DOI: 10.1063/1.5040296; arXiv:1805.02133. https://arxiv.org/abs/1805.02133