Open fronts in the rigorous Newton-to-Boltzmann derivation: short-range and long-range potentials, physical boundaries, and non-perturbative data

The monograph of Gallagher, Saint-Raymond and Texier (2012/2014) gave a complete, quantitative proof of Lanford's theorem: in the Boltzmann-Grad limit, Newtonian hard-sphere systems in the whole space -- and, under a technical monotonicity assumption on the potential, particles interacting through smooth compactly supported repulsive potentials -- converge to the solution of the Boltzmann equation, but only on a time interval that is a fraction of the mean free time, for analytic-type data. Its concluding remarks formulate the surviving open problems: longer validity times, long-range and grazing-dominated interactions (leading to singular collision kernels and the Landau equation), bounded domains with physical boundary conditions, a microscopic counterpart of DiPerna-Lions renormalization for arbitrary data, and -- as later made explicit by Deng-Hani-Ma -- optimal convergence rates. Deng, Hani and Ma (2024-2025) have since resolved the hard-sphere long-time question, conditionally on the Boltzmann equation having a regular solution. This record concerns the surviving core: extending the rigorous derivation to short-range potentials on arbitrary times, singular kernels including the nonlinear Landau equation, physical boundaries, and non-perturbative data.

Background

A central problem of mathematical physics, part of Hilbert's sixth problem (1900) on the axiomatization of physics, is the rigorous derivation of the Boltzmann equation of kinetic theory from Newton's laws for a system of $N$ identical particles. The Boltzmann equation governs the one-particle distribution function $f(t,x,v)$ of a rarefied gas through $(\partial_t + v\cdot\nabla_x)f = Q(f,f)$, where the quadratic collision operator $Q$ encodes binary elastic collisions with a cross-section $b$ determined by the intermolecular interaction. In the Boltzmann-Grad scaling one sends the particle number $N\to\infty$ and the interaction length $\varepsilon\to 0$ with $N\varepsilon^{d-1}=O(1)$, so that each particle suffers $O(1)$ collisions per unit time; the mean free time (the average time between successive collisions of a particle) is the natural time unit. The closure assumption behind the equation is molecular chaos (the Stosszahlansatz): particles about to collide are statistically independent. Newtonian dynamics creates correlations -- in particular recollisions, i.e. pairs of particles that meet more than once, whose histories are then no longer independent -- and the derivation must show these are negligible in the limit. The standard route is the BBGKY hierarchy, the chain of equations for the $s$-particle correlation functions $f_s$, whose formal limit is the Boltzmann hierarchy, solved by tensorized products of the Boltzmann solution. Lanford (1975) proved this term-by-term convergence for hard spheres, but only on a time interval that is a small fraction of the mean free time, and for data with analytic-type bounds. Gallagher, Saint-Raymond and Texier (source monograph, 2012/2014) gave a complete and quantitative proof in the whole space $\mathbb{R}^d$, for hard spheres and -- under a technical assumption, condition (8.3.1) of the monograph, namely $\rho\Phi''(\rho)+2\Phi'(\rho)\ge 0$ for $0<\rho<1$, with $\rho$ the interparticle distance in units of the interaction range, which guarantees that the two-body scattering deflection angle is a strictly monotone function of the impact parameter so that the cross-section is well defined -- also for smooth, compactly supported, repulsive potentials. Their concluding remarks formulate the open problems that define this record. First, the time of validity: the convergence holds exactly as long as uniform a priori estimates in the analytic-type function spaces hold, and the obstacle is the control of spatial concentrations and high-order correlations beyond such spaces. Second, the potential class: for long-range potentials the cross-section becomes singular, the limiting equation being the Boltzmann equation with non-cutoff (angularly singular) kernel, for which Alexandre and Villani (2002) built the PDE-level theory of renormalized solutions; when grazing (small-deflection) collisions dominate, the formal limit is instead the Landau equation, a nonlinear Fokker-Planck equation in velocity describing the cumulative effect of many weak deflections, which is the kinetic equation of weakly coupled plasmas; its derivation from particle dynamics was, in the monograph's words, essentially open. Third, the geometry: the analysis is restricted to the whole space, where free transport is along straight lines. Fourth, the data: the only global existence theory of the Boltzmann equation for initial data of arbitrary size is the DiPerna-Lions theory of renormalized solutions (1989), obtained by testing the equation against $\beta(f)$ for smooth $\beta$; the monograph states that using it would require understanding a counterpart of renormalization at the level of the microscopic dynamics. Fifth, optimal convergence rates, a question made explicit in the subsequent work of Deng-Hani-Ma. The landscape changed with Deng, Hani and Ma (2024-2025), who proved the hard-sphere derivation for arbitrarily long times: uniformly on any interval on which the Boltzmann equation has a bounded solution with Maxwellian (Gaussian) velocity decay, the correlation functions factorize up to $\varepsilon^{\theta}$ in $L^1$, in dimensions $d\ge 2$, in the whole space (for the grand-canonical ensemble, in which the number of particles is Poisson-distributed at fixed density) and, in a companion paper, on the two- and three-dimensional tori, which they combine with hydrodynamic-limit theory to claim the resolution of Hilbert's sixth problem for a rarefied hard-sphere gas. Their result remains conditional on the Boltzmann Cauchy problem: for generic large data, global solutions are known only in the renormalized (DiPerna-Lions) class, and their own list of open directions includes the extension to short-range potentials, which they believe true but which requires new ingredients to handle simultaneous interactions of more than two particles through the potential.

Problem Statement

Prove, or disprove by explicit construction, the convergence of the deterministic Newtonian N-particle dynamics to the limiting kinetic equation in each of the regimes still excluded by the rigorous theory, at the standard of rigor of the Lanford/Gallagher-Saint-Raymond-Texier and Deng-Hani-Ma theorems -- i.e. quantitative convergence, in the appropriate low-density scaling (Boltzmann-Grad $N\varepsilon^{d-1}=O(1)$ for short-range interactions, and the corresponding grazing/weak-coupling scalings for long-range interactions, in the sense of the references cited for this axis), of the $s$-particle correlation functions to tensorized powers of the solution of the limiting equation, in $L^1$ or an equivalent topology, with explicit control of the error. The four components are: (P1) Smooth short-range potentials, long times: extend the Deng-Hani-Ma long-time validity theorem from hard spheres to Hamiltonian systems $\dot x_i=v_i$, $\dot v_i=-\frac{1}{\varepsilon}\sum_{j\ne i}\nabla\Phi((x_i-x_j)/\varepsilon)$ (the scaled dynamics of equation (1) of the source monograph and equation (1.27) of Deng-Hani-Ma) with $\Phi$ smooth, compactly supported and repulsive, whose two-body deflection angle is not a piecewise constant function of the impact parameter (this degenerate sub-case is excluded: as the source monograph's concluding remarks record, the Boltzmann equation is not expected there to be a good approximation of the dynamics), not assumed to satisfy the source monograph's condition (8.3.1), $\rho\Phi''(\rho)+2\Phi'(\rho)\ge 0$, proving validity on any time interval on which the corresponding Boltzmann solution exists and remains bounded in the same sense as Theorem 1 of Deng-Hani-Ma, i.e. with a Maxwellian-weighted $L^\infty$ bound. (P2) Singular kernels: establish the low-density limit for long-range potentials and grazing-dominated regimes -- the derivation of the inhomogeneous nonlinear Boltzmann equation with non-cutoff (angularly singular) cross-section from Newtonian dynamics, and of the nonlinear Landau equation from deterministic Newtonian particle systems in the appropriate grazing/weak-coupling scaling (the many-weak-deflections regime described in the background) -- beyond the existing time-zero consistency results, linearized/fluctuation-level derivations, and Kac-type randomized models. (P3) Physical boundaries: extend the derivation to gases confined to bounded domains with physical boundary conditions -- specular reflection (mirror-like rebound: the tangential velocity component is preserved and the normal component reversed) beyond the half-space case, and diffusive or accommodation boundaries (particles absorbed and re-emitted with a Maxwellian velocity distribution at the wall temperature) beyond the existing short-time box results -- including long time intervals. (P4) Non-perturbative data: remove the hypothesis that the Boltzmann equation possesses a regular solution on the validity interval, by connecting the microscopic derivation with the DiPerna-Lions theory of renormalized solutions -- the conceptual step being a microscopic counterpart of renormalization -- so as to obtain validity for physically admissible initial data of arbitrary size, globally in time. A complete resolution establishes all four components; the components are independent, so an affirmative resolution of any single component is a valid partial resolution of the question, while a complete resolution requires all four. Two of the components bundle two independently resolvable sub-claims each: (P2) splits into (P2a) the derivation of the inhomogeneous nonlinear Boltzmann equation with non-cutoff cross-section and (P2b) the derivation of the nonlinear Landau equation, and (P3) splits into (P3a) specular reflection beyond the half-space and (P3b) diffusive or accommodation boundaries beyond the existing short-time box results; a component counts as resolved only when all of its sub-claims are settled, and an affirmative resolution of a single sub-claim is itself a valid partial resolution of the question. A refutation of any component exhibits a regime in which the limiting kinetic equation provably fails to approximate the particle dynamics. A refutation of a single sub-claim exhibits such a regime for the corresponding limiting equation and settles that sub-claim, the other sub-claim of its component remaining open. Quantitative refinements -- optimal convergence rates and the optimal dependence of the admissible validity time on the scaling parameter, which Deng-Hani-Ma list among the open questions of their theory (Remark 1.5) -- are desirable but not required to resolve the question, while the four components above constitute the acceptance target.

References

1. Isabelle Gallagher, Laure Saint-Raymond, Benjamin Texier, 'From Newton to Boltzmann: Hard Spheres and Short-range Potentials', Zurich Lectures in Advanced Mathematics, EMS Press, 2014. DOI: 10.4171/129; ISBN 978-3-03719-129-3; arXiv:1208.5753. https://doi.org/10.4171/129
2. O. E. Lanford, 'Time evolution of large classical systems', in Dynamical Systems, Theory and Applications (Battelle Seattle 1974 Rencontres), Lecture Notes in Physics 38, Springer, Berlin, 1975, pp. 1-111. DOI: 10.1007/3-540-07171-7_1. https://doi.org/10.1007/3-540-07171-7_1
3. R. J. DiPerna, P.-L. Lions, 'On the Cauchy problem for Boltzmann equations: global existence and weak stability', Annals of Mathematics (2) 130 (1989), no. 2, 321-366. DOI: 10.2307/1971423. https://doi.org/10.2307/1971423
4. Reinhard Illner, Marvin Shinbrot, 'The Boltzmann equation: global existence for a rare gas in an infinite vacuum', Communications in Mathematical Physics 95 (1984), 217-226. DOI: 10.1007/BF01468142. https://doi.org/10.1007/BF01468142
5. R. Alexandre, C. Villani, 'On the Boltzmann equation for long-range interactions', Communications on Pure and Applied Mathematics 55 (2002), no. 1, 30-70. DOI: 10.1002/cpa.10012. https://doi.org/10.1002/cpa.10012
6. Mario Pulvirenti, Chiara Saffirio, Sergio Simonella, 'On the validity of the Boltzmann equation for short range potentials', Reviews in Mathematical Physics 26 (2014), no. 2, 1450001. DOI: 10.1142/S0129055X14500019; arXiv:1301.2514. https://doi.org/10.1142/S0129055X14500019
7. A. V. Bobylev, M. Pulvirenti, C. Saffirio, 'From particle systems to the Landau equation: a consistency result', Communications in Mathematical Physics 319 (2013), 683-702. DOI: 10.1007/s00220-012-1633-6. https://doi.org/10.1007/s00220-012-1633-6
8. Yu Deng, Zaher Hani, Xiao Ma, 'Long time derivation of the Boltzmann equation from hard sphere dynamics', arXiv:2408.07818 (v1 August 2024, v3 July 2025). https://arxiv.org/abs/2408.07818
9. 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
10. 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)', Bourbaki Seminar, January 2026, arXiv:2602.04407. https://arxiv.org/abs/2602.04407
11. Shan Gao, 'Comment on "Hilbert's Sixth Problem: Derivation of Fluid Equations via Boltzmann's Kinetic Theory" by Deng, Hani, and Ma', arXiv:2504.06297 (2025). https://arxiv.org/abs/2504.06297
12. Théophile Dolmaire, 'About Lanford's theorem in the half-space with specular reflection', Kinetic and Related Models 16 (2023), 207-268. DOI: 10.3934/krm.2022028; arXiv:2102.05513. https://doi.org/10.3934/krm.2022028
13. Corentin Le Bihan, 'Boltzmann-Grad limit of a hard sphere system in a box with diffusive boundary conditions', arXiv:2104.04354 (2021). https://arxiv.org/abs/2104.04354
14. Corentin Le Bihan, 'Long time validity of the linearized Boltzmann equation for hard spheres: a proof without billiard theory', arXiv:2212.04392 (2022). https://arxiv.org/abs/2212.04392
15. Corentin Le Bihan, 'Long time validity of the linearized Boltzmann uncut-off and the linearized Landau equations from the Newton Law', arXiv:2408.03597 (v1 August 2024, v2 January 2026). https://arxiv.org/abs/2408.03597
16. Xuanrui Feng, Zhenfu Wang, 'Kac's Program for the Landau Equation', Archive for Rational Mechanics and Analysis 250 (2026), no. 4. DOI: 10.1007/s00205-026-02230-3; arXiv:2506.14309. https://doi.org/10.1007/s00205-026-02230-3
17. Shuchen Guo, 'From Kac particles to the Landau equation with hard potentials: BBGKY hierarchy method', arXiv:2508.10697 (2025). https://arxiv.org/abs/2508.10697
18. Ioakeim Ampatzoglou, Joseph K. Miller, Nataša Pavlović, Maja Tasković, 'On the global in time existence and uniqueness of solutions to the Boltzmann hierarchy', Journal of Functional Analysis 289 (2025), 111079. DOI: 10.1016/j.jfa.2025.111079; arXiv:2402.00765. https://doi.org/10.1016/j.jfa.2025.111079
19. Thierry Bodineau, Isabelle Gallagher, Laure Saint-Raymond, Sergio Simonella, 'Long-time derivation at equilibrium of the fluctuating Boltzmann equation', Annals of Probability 52 (2024). DOI: 10.1214/23-AOP1656; arXiv:2201.04514. https://doi.org/10.1214/23-AOP1656
20. François Golse, 'Validité de la théorie cinétique des gaz: au-delà de l'équation de Boltzmann [d'après T. Bodineau, I. Gallagher, L. Saint-Raymond, S. Simonella]', Séminaire Bourbaki, November 19, 2022; arXiv:2308.11919 (2023). https://arxiv.org/abs/2308.11919
21. Mario Pulvirenti, Sergio Simonella, 'The Boltzmann-Grad limit of a hard sphere system: analysis of the correlation error', Inventiones mathematicae 207 (2017), no. 3, 1135-1237. DOI: 10.1007/s00222-016-0682-4; arXiv:1405.4676. https://doi.org/10.1007/s00222-016-0682-4