Optimal convergence rate and time of validity for Lanford's theorem in the half-space with specular reflection

Dolmaire (2023) proved an analog of Lanford's theorem for the Boltzmann-Grad limit of a hard-sphere gas in the half-space with specular reflection at the wall, with an explicit algebraic convergence rate $O(\varepsilon^\gamma)$ for every $\gamma\in(0,13/128)$ , but only on a time interval $[0,T^0]$ strictly shorter than the co-existence time $T$ of the two hierarchies, and he states that the underlying geometric estimates are not optimal. The open problem is to make this validity result quantitatively optimal and time-complete: improve the convergence rate toward the expected optimal scale $O(\varepsilon|\ln\varepsilon|)$ , and remove the reduction from $T$ to $T^0$ , with extensions to other domains and boundary conditions as further stated directions.

Background

The Boltzmann equation describes the evolution of the one-particle distribution $f(t,x,v)$ of a rarefied gas under a molecular-chaos assumption; it can be derived heuristically from the deterministic Newtonian dynamics of $N$ identical hard spheres of diameter $\varepsilon$ in the Boltzmann-Grad scaling $N\varepsilon^{d-1}=1$ , in which the mean free path stays of order one as $N\to\infty$ . Lanford's theorem (1975) made this derivation rigorous: the first marginal of the $N$ -particle density — the law of a single particle's phase-space position, obtained by integrating the full density over the other particles — converges, on a short time interval, to the solution of the Boltzmann equation. The proof passes through the BBGKY hierarchy (an infinite coupled system of equations for the marginals, named after Bogoliubov, Born, Green, Kirkwood, and Yvon), compares it term-by-term with the limiting Boltzmann hierarchy, and controls the difference between the pseudo-trajectories that define each term of the series; the essential difficulty is that particles can recollide, so the microscopic dynamics does not exactly superimpose the free transport underlying the limit equation. Gallagher, Saint-Raymond and Texier (2013) gave the complete modern version of this program. The physical boundary case is harder: when the gas occupies the half-space $\{x_1>0\}$ and particles hitting the wall bounce by specular reflection (incoming and outgoing velocities are mirror images at the contact point), the billiard trajectories gain wall bounces that interact with the recollision analysis. Dolmaire (2023) defined the collision operator and the functional space in which the BBGKY hierarchy is solved strongly in this domain, adapted the recollision control of Gallagher–Saint-Raymond–Texier to the wall, and proved: in the Boltzmann-Grad limit, the first marginal converges locally uniformly to the solution of the Boltzmann equation with specular reflection, on a time interval $[0,T^0]$ , with rate $O(\varepsilon^\gamma)$ for every $\gamma\in(0,13/128)$ , assuming $\sqrt{f_0}$ Lipschitz in the position variable. Two of the theorem's quantitative features are explicitly identified in the paper (Remark 8) as artifacts of the proof: the time reduction $T^0<T$ , where $T$ is the time up to which both hierarchies co-exist by a fixed-point argument — caused by weight losses when taking fractional powers of the initial data — and the small algebraic exponent, caused by geometric estimates around wall bounces that are 'clearly not optimal'. In the boundaryless case the quantitative theory reaches near-linear rates of order $\varepsilon|\ln\varepsilon|$ , and Dolmaire conjectures that, since a wall should not regularize the dynamics, one should not expect better than $\varepsilon\ln\varepsilon$ in the half-space either; proving any rate of that order with a boundary present is out of reach of current estimates. Separately, the short-time restriction of Lanford-type theorems was famously overcome by Deng, Hani and Ma (2024) — but only in the whole space $\mathbb{R}^3$ without boundary, and by a method (long-time cumulant expansion with diagrammatic 'molecule' counting) that has not been extended to domains with walls; a February 2026 survey by Bodineau, Gallagher, Saint-Raymond and Simonella confirms that the short-time limitation stands for boundary settings and lists no improvement of the half-space rate. The only adjacent boundary result is Le Bihan's (2022) qualitative, non-quantitative convergence in a box with isotropic boundary conditions — a different domain and boundary law, with no rate.

Problem Statement

Establish a quantitatively optimal and time-complete version of Lanford's theorem for the Boltzmann-Grad limit in the half-space $\{x_1>0\}$ with specular reflection at the wall. Specifically, for $N$ hard spheres of diameter $\varepsilon$ in Boltzmann-Grad scaling with the initial first-particle density converging to $f_0$ (with $\sqrt{f_0}$ Lipschitz in position), prove that the first marginal converges to the solution of the Boltzmann equation with specular reflection, uniformly on compact subsets of the phase space that do not intersect the wall nor contain grazing trajectories and uniformly in time. The problem is to (1) improve the known algebraic convergence bound: prove convergence at any rate $O(\varepsilon^{\gamma'})$ with $\gamma'>13/128$ ; (2) determine whether the near-linear rate $O(\varepsilon|\ln\varepsilon|)$ , which Dolmaire conjectures is optimal in the presence of a wall, is attainable or unattainable with a boundary — attainability of the near-linear rate would imply an improved algebraic bound and hence (1) as well; and (3) extend the validity to the full time interval $[0,T]$ of hierarchy co-existence, removing the current reduction to a strictly shorter $[0,T^0]$ caused by weight losses in the fixed-point argument. These three goals share one objective — a quantitative validity theory for the dilute hard-sphere gas with a specular wall that matches, up to the conjectured logarithmic loss, the precision of the boundaryless theory. Outside these three goals, and attached to the same program, the author states a further direction: extend the derivation to more general geometries and boundary conditions, for example nonconvex obstacles or stochastic boundary laws in which the direction of the reflected velocity of a particle hitting the wall is chosen randomly, as in Le Bihan's model.

References

Théophile Dolmaire, "About Lanford's theorem in the half-space with specular reflection", Kinetic & Related Models 16(2):207–268, 2023. DOI: 10.3934/krm.2022028; arXiv:2102.05513. https://doi.org/10.3934/krm.2022028
Oscar E. Lanford, "Time evolution of large classical systems", in Dynamical Systems, Theory and Applications, Lecture Notes in Physics 38, pp. 1–111, Springer, 1975. DOI: 10.1007/3-540-07171-7_1. https://doi.org/10.1007/3-540-07171-7_1
Isabelle Gallagher, Laure Saint-Raymond, Benjamin Texier, "From Newton to Boltzmann: Hard Spheres and Short-range Potentials", EMS Textbooks in Mathematics, 2013. DOI: 10.4171/129; arXiv:1208.5753. https://doi.org/10.4171/129
Corentin Le Bihan, "Boltzmann-Grad limit of a hard sphere system in a box with isotropic boundary conditions", Discrete and Continuous Dynamical Systems 42(4):1903–1932, 2022. DOI: 10.3934/dcds.2021177; arXiv:2104.04354 (preprint titled "Boltzmann-Grad limit of a hard sphere system in a box with diffusive boundary conditions"). https://doi.org/10.3934/dcds.2021177
Yu Deng, Zaher Hani, Xiao Ma, "Long time derivation of the Boltzmann equation from hard sphere dynamics", 2024. arXiv:2408.07818. https://arxiv.org/abs/2408.07818
Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, Sergio Simonella, "Derivation of the Boltzmann equation from hard-sphere dynamics (after Y. Deng, Z. Hani, and X. Ma)", 2026. arXiv:2602.04407. https://arxiv.org/abs/2602.04407