The 2019 note of Bodineau, Gallagher and Saint-Raymond asked whether Fourier's law of heat conduction can be derived from Newtonian hard-sphere dynamics. They proved the chain hard spheres to linearized Boltzmann equation to Stokes-Fourier system only for near-equilibrium fluctuations on a two-dimensional torus, and listed six gaps: thermostated boundary conditions at the particle level, nonlinear and large perturbations, the extension to dimension d at least 3, super-exponential collision trees with multiple recollisions, a missing renormalized modulated entropy for the nonlinear Boltzmann equation with inhomogeneous wall temperature, and physically unnatural iterated-logarithmic scale separations. The boundary-free gaps were closed in 2024-2025 by Deng, Hani and Ma, who derived the full nonlinear Boltzmann equation from hard spheres on R^d (d at least 2) and on the torus T^d (d = 2, 3) for arbitrarily long times without smallness, and derived the incompressible Navier-Stokes-Fourier and compressible Euler equations from Newton's laws on tori. The surviving open core is the physically central boundary-driven regime: a Boltzmann-Grad convergence theorem for hard spheres with thermostated (Maxwellian) walls at prescribed, possibly non-uniform temperature — open even at short times, since the only random-wall particle result (Le Bihan) is short-time with energy-preserving reflection, which yields an adiabatic rather than thermostat limit — together with the nonlinear diffusive limit of the Boltzmann equation with inhomogeneous wall temperature, which still lacks the uniform a priori bound (the renormalized modulated entropy) whose definition the source left open, and the question whether near-equilibrium smallness is merely technical.
Background
Fourier's law, , relates the heat flux through a material to the gradient of its temperature , with the thermal conductivity; it implies that the temperature obeys a diffusion (heat) equation. Although experimentally well tested, it is a phenomenological constitutive relation, and deriving it from the Newtonian mechanics of the constituent particles — in the spirit of Hilbert's sixth problem, and as phrased by Bonetto, Lebowitz and Rey-Bellet as a challenge to theorists — remains a central open problem of mathematical physics.
For a rarefied gas the reference microscopic model is a system of hard spheres of diameter interacting by elastic collisions. The Boltzmann equation describes the one-particle density in the low-density limit; the relevant scaling is the Boltzmann-Grad limit , with , where is the Knudsen number (the mean free path relative to the observation scale), so that corresponds to a fast-relaxation, diffusive regime. Lanford's 1975 theorem justifies the convergence to the Boltzmann equation, but only for short times: its proof propagates the BBGKY hierarchy (the coupled equations for the -particle marginals of the -particle distribution) and requires control of recollisions — collisions between already-correlated particles that the free-transport picture ignores — and of collision trees with super-exponentially many branchings. Diffuse (Maxwellian) reflection is the standard model of a wall held at fixed temperature: an incoming particle at is re-emitted with velocity drawn from the wall Maxwellian (the Gaussian equilibrium velocity distribution) at the prescribed temperature, so the wall acts as a heat reservoir; when the prescribed wall temperature is spatially non-uniform, the gas is genuinely out of equilibrium and the total energy of the particle system fluctuates.
At the kinetic level, Bodineau, Gallagher and Saint-Raymond (2019) proved that the linearized Boltzmann equation (with the linearization of the collision operator around the equilibrium Maxwellian ), equipped with diffuse-reflection boundary conditions at wall temperature with (the note assumes the non-equilibrium perturbation to be very small, so that the nonlinear convection terms are higher-order corrections; at the endpoint they survive and the formal limit is instead of Navier-Stokes-Fourier type), converges as to the Stokes-Fourier system — the incompressible Stokes equations for the velocity coupled with the heat equation and , which is Fourier's law. The crux is a uniform bound on the fluctuation , obtained by measuring it against a Maxwellian adapted to the wall temperature (a modulated fluctuation). For the full nonlinear Boltzmann equation this route breaks down: the only known global weak solutions are renormalized solutions (the DiPerna-Lions notion: weak solutions of the equation for the rescaled unknown , for which only an entropy inequality is available), and this entropy inequality carries a boundary contribution that has no definite sign when ; the note explicitly leaves open whether a renormalized modulated entropy restoring uniform bounds can be defined.
At the particle level, the same authors (2017) had proved the full chain — hard spheres to linearized Boltzmann equation to Stokes-Fourier system — for near-equilibrium fluctuations of order around the equilibrium Gibbs measure on the two-dimensional torus, on diffusive time scales. Their 2019 note lists the remaining gaps: the thermostated boundary conditions are not incorporated (no derivation of the Boltzmann equation from particles with such non-equilibrium boundaries was known); the argument is restricted to because the comparison between the hard-sphere Gibbs measure (the equilibrium measure including the exclusion constraint) and the tensor product , via a uniform lower bound on the partition function, fails for ; the nonlinear regime is out of reach; the super-exponential-tree and multiple-recollision estimates are only partly general; and the required scale separation is physically unnatural.
The boundary-free gaps were removed in 2024-2025 by Deng, Hani and Ma, who derived the full nonlinear Boltzmann equation from hard spheres on and on the torus () for arbitrarily long times and without smallness, and, combining this with known hydrodynamic limits, derived the incompressible Navier-Stokes-Fourier equations (the viscous analogue of the Stokes-Fourier system) and the compressible Euler equations from Newton's laws on tori (near equilibrium for the former). What survives of the source question is precisely the physically central case: the boundary-driven, non-equilibrium regime in which heat flows between reservoirs at prescribed temperatures.
Problem Statement
Let () be a smooth bounded domain. Consider the system of hard spheres of diameter in , with elastic inter-particle collisions and a thermostated boundary: a particle reaching is re-emitted with velocity drawn from the wall Maxwellian (the Gaussian distribution of outgoing velocities) at the prescribed temperature with (diffuse, or Gaussian, reflection; the restriction is the source's assumption that the non-equilibrium perturbation is very small, under which the nonlinear convection terms are higher-order corrections and the diffusive limit is the Stokes-Fourier system, whereas at convection persists and the limit is of Navier-Stokes-Fourier type), so that the boundary acts as a heat reservoir, the total energy of the system fluctuates, and no invariant equilibrium measure with product structure is available. Take initial -particle data that are chaotic (asymptotically factorizing as tensor products of a one-particle distribution as ) and close to the canonical equilibrium Gibbs measure at uniform temperature , (the source's : the hard-sphere exclusion set with partition function and Maxwellian velocity density at temperature ; under the thermostated dynamics it has no invariant counterpart with this product structure), with fluctuation amplitude of order — the linearized, near-equilibrium regime of the source's torus theorem, which the source authors expected to extend to thermostated systems — in the Boltzmann-Grad scaling . In the sequential limit below, the initial fluctuation is absorbed in the Boltzmann-Grad step at fixed , and, since , the wall drive enters the intermediate kinetic description as a small forcing of the linearized Boltzmann equation around the global Maxwellian, as in the source's kinetic theorem.
Prove that, in the combined low-density and diffusive limit (first , at fixed , then ; or in a single scaling if possible), the rescaled one-particle fluctuation of the gas converges, in the sense with respect to the equilibrium Maxwellian (the source's convention), to an infinitesimal Maxwellian whose macroscopic profiles solve the Stokes-Fourier system
which is Fourier's law for the hard-sphere gas driven by reservoirs at prescribed temperatures.
This single objective comprises two linked enabling steps, reflecting the two open items discussed in the source note: (i) at the particle level, a Boltzmann-Grad convergence theorem for the thermostated hard-sphere process — currently unavailable even at short times — and then its extension to the long, diffusive time scale; and (ii) at the kinetic level, a uniform a priori bound for the fluctuations of the nonlinear Boltzmann equation with inhomogeneous wall temperature, i.e. the renormalized modulated entropy whose definition the source leaves open. The broader, source-faithful target is the same derivation beyond the near-equilibrium regime, together with the question whether the smallness assumptions are purely technical or whether genuinely new phenomena obstruct diffusion-type limits under boundary drive far from equilibrium; a rigorous obstruction result of that kind would be a valid answer of the refutation type. The deliverable required here is the near-equilibrium convergence stated above: of the two steps just listed, only the particle-level Boltzmann-Grad convergence theorem (i) is needed for it, while the nonlinear a priori bound (ii) and the beyond-near-equilibrium derivation are optional extensions of the source's programme, and any proof of that convergence is acceptable. Results on boundary-free domains, on deterministic (specular) or energy-preserving walls, or at Euler rather than diffusive scale constitute recognizable partial progress but do not close the problem.
References
- Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, A microscopic view of the Fourier law, Comptes Rendus Physique (2019), doi:10.1016/j.crhy.2019.08.002; arXiv:1912.03947, https://arxiv.org/abs/1912.03947
- Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, From Hard Sphere Dynamics to the Stokes-Fourier Equations: An Analysis of the Boltzmann-Grad Limit, Annals of PDE 3 (2017), doi:10.1007/s40818-016-0018-0, https://doi.org/10.1007/s40818-016-0018-0
- Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, The Brownian motion as the limit of a deterministic system of hard-spheres, Inventiones mathematicae 203 (2016), doi:10.1007/s00222-015-0593-9, https://doi.org/10.1007/s00222-015-0593-9
- Yu Deng, Zaher Hani, Xiao Ma, Long time derivation of Boltzmann equation from hard sphere dynamics, arXiv:2408.07818 (2024, v3 2025), https://arxiv.org/abs/2408.07818
- Yu Deng, Zaher Hani, Xiao Ma, Hilbert's sixth problem: derivation of fluid equations via Boltzmann's kinetic theory, arXiv:2503.01800 (2025), https://arxiv.org/abs/2503.01800
- Corentin Le Bihan, Boltzmann-Grad limit of a hard sphere system in a box with isotropic boundary conditions, Discrete and Continuous Dynamical Systems (2022), doi:10.3934/dcds.2021177, https://doi.org/10.3934/dcds.2021177; preprint arXiv:2104.04354, posted under the title 'Boltzmann-Grad limit of a hard sphere system in a box with diffusive boundary conditions', https://arxiv.org/abs/2104.04354
- Corentin Le Bihan, Long Time Validity of the Linearized Boltzmann Equation for Hard Spheres: A Proof Without Billiard Theory, Communications in Mathematical Physics (2025), doi:10.1007/s00205-025-02105-z; preprint arXiv:2212.04392, https://arxiv.org/abs/2212.04392
- Corentin Le Bihan, Long-Time Validity of the Linearized Boltzmann Uncut-Off and the Linearized Landau Equations From the Newton Law, Communications on Pure and Applied Mathematics (2026), doi:10.1002/cpa.70035, https://doi.org/10.1002/cpa.70035
- Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, Sergio Simonella, Statistical dynamics of a hard sphere gas: fluctuating Boltzmann equation and large deviations, Annals of Mathematics 198 (2023), doi:10.4007/annals.2023.198.3.3, https://doi.org/10.4007/annals.2023.198.3.3
- Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, Sergio Simonella, Long-time derivation at equilibrium of the fluctuating Boltzmann equation, Annals of Probability (2024), doi:10.1214/23-aop1656; preprint arXiv:2201.04514, https://arxiv.org/abs/2201.04514
- Théophile Dolmaire, About Lanford's theorem in the half-space with specular reflection, arXiv:2102.05513 (2021), https://arxiv.org/abs/2102.05513
- Oscar E. Lanford, Time evolution of large classical systems, Lecture Notes in Physics 38, Springer (1975), doi:10.1007/3-540-07171-7_1, https://doi.org/10.1007/3-540-07171-7_1
- Nader Masmoudi, Laure Saint-Raymond, From the Boltzmann equation to the Stokes-Fourier system in a bounded domain, Communications on Pure and Applied Mathematics 56 (2003), doi:10.1002/cpa.10095, https://doi.org/10.1002/cpa.10095
- Ning Jiang, Nader Masmoudi, Boundary Layers and Incompressible Navier-Stokes-Fourier Limit of the Boltzmann Equation in Bounded Domain I, Communications on Pure and Applied Mathematics (2016), doi:10.1002/cpa.21631; preprint arXiv:1510.02977, https://arxiv.org/abs/1510.02977
- Ning Jiang, Xu Zhang, The Boltzmann equation with incoming boundary condition: global solutions and Navier-Stokes limit, arXiv:1701.04144 (2017), https://arxiv.org/abs/1701.04144
- R. Esposito, Y. Guo, C. Kim, R. Marra, Non-Isothermal Boundary in the Boltzmann Theory and Fourier Law, Communications in Mathematical Physics 323 (2013), doi:10.1007/s00220-013-1766-2, https://doi.org/10.1007/s00220-013-1766-2
- Hongxu Chen, Chanwoo Kim, Gradient Decay in the Boltzmann Theory of Non-isothermal Boundary, Communications in Mathematical Physics (2024), doi:10.1007/s00205-024-01956-2, https://doi.org/10.1007/s00205-024-01956-2
- Hongxu Chen, Renjun Duan, Diffusive limit of the Boltzmann equation around Rayleigh profile in the half space, arXiv:2512.04403 (2025), to appear in Bull. Inst. Math. Acad. Sin., https://arxiv.org/abs/2512.04403
- R. Medina, The Boltzmann equation with non-isothermal Maxwell boundary conditions, arXiv:2604.13572 (2026), https://arxiv.org/abs/2604.13572
- Yunbai Cao, Juhi Jang, Chanwoo Kim, Passage from the Boltzmann equation with diffuse boundary to the incompressible Euler equation with heat convection, Journal of Differential Equations (2023), doi:10.1016/j.jde.2023.04.028; preprint arXiv:2104.02169, https://arxiv.org/abs/2104.02169
- Juhi Jang, Chanwoo Kim, Incompressible Euler Limit from Boltzmann Equation with Diffuse Boundary Condition for Analytic Data, Annals of PDE (2021), doi:10.1007/s40818-021-00108-z; preprint arXiv:2005.12192, https://arxiv.org/abs/2005.12192
- F. Bonetto, J. L. Lebowitz, L. Rey-Bellet, Fourier's Law: A Challenge to Theorists, in Mathematical Physics 2000, Imperial College Press (2000), doi:10.1142/9781848160224_0008, https://doi.org/10.1142/9781848160224_0008