Validity domain of the rigorous derivation of the binary-ternary Boltzmann equation for a dense hard-sphere gas: long-time convergence, alternative scalings, and more general interaction laws

Ampatzoglou and Pavlović rigorously derived the binary-ternary Boltzmann equation — a kinetic equation whose collision term sums a classical binary operator and a ternary correction intended for denser gases — from a deterministic system of $N$ hard spheres of diameter $\epsilon_2$ that also undergo instantaneous ternary interactions inside zones of size $\epsilon_3$, in the scale-separated mixed scaling $N\epsilon_2^{d-1}\simeq N\epsilon_3^{d-1/2}\simeq 1$ with $\epsilon_2\ll\epsilon_3$. The convergence from the microscopic dynamics, proved in Maxwellian-weighted $L^\infty$-type spaces, holds only on a short time interval. This problem asks to extend the rigorous derivation beyond the three restrictions under which it is proved: (i) beyond the short-time regime, in the limit to global-in-time validity, or to a proven obstruction; (ii) to other density scalings and other relations between the binary and ternary interaction lengths, identifying the resulting limiting equations; (iii) to more general higher-order interaction mechanisms than hard-sphere and interaction-zone collision laws, such as genuine many-body potentials. All three strands are open for the deterministic particle model: the only long-time derivation from hard spheres, by Deng, Hani and Ma, covers the purely binary case, and the higher-order extension of Ampatzoglou, Pavlović and Warner is again short-time, restricted to a symmetric interaction-zone law, and exists so far only as a preprint.

Background

The Boltzmann equation describes the evolution of the one-particle probability density $f(t,x,v)$ of a rarefied gas, and its collision operator $Q_2(f,f)$ accounts only for binary encounters, which is adequate for dilute systems. The classical rigorous derivation goes back to Lanford: for $N$ hard spheres of diameter $\epsilon$ in $\mathbb{R}^d$ ($d\geq 2$) under the Boltzmann-Grad scaling $N\epsilon^{d-1}\simeq 1$, the hierarchy of equations for the particle marginals (the BBGKY hierarchy, named after Bogoliubov, Born, Green, Kirkwood and Yvon) converges, on a short time interval, to an infinite Boltzmann hierarchy whose factorized solutions solve the Boltzmann equation; convergence is measured in observables, i.e. integrated against test functions on well-separated configurations. Lanford's original argument was later made quantitative by Gallagher, Saint-Raymond and Texier, but even in the purely binary case the derivation remained restricted to short times until the recent work of Deng, Hani and Ma, who derived the Boltzmann equation from hard-sphere dynamics for arbitrarily long times, as long as the limiting equation is well-posed.

For denser gases, higher-order simultaneous encounters contribute materially to the dynamics (physics results on colloidal suspensions, cited in the source paper, show three-body interactions significantly affect the grand potential). This motivated a program of Ampatzoglou and Pavlović to derive kinetic equations with ternary collision terms. The ternary model uses an interaction-zone notion: three particles interact when a non-symmetric distance $d_3(x_i;x_j,x_k)=\sqrt{|x_i-x_j|^2+|x_i-x_k|^2}$ reaches $\sqrt{2}\,\epsilon_3$. Such ternary interactions are governed by the scaling $N\epsilon_3^{d-1/2}\simeq 1$, which is incompatible with the Boltzmann-Grad scaling if a single interaction length is used. The source paper of this problem resolves this conflict by endowing the same particles with both a hard-sphere diameter $\epsilon_2$ and a ternary zone $\epsilon_3$ with $\epsilon_2\ll\epsilon_3$, under the mixed scaling

$N\epsilon_2^{d-1}\simeq N\epsilon_3^{d-\frac{1}{2}}\simeq 1,$

and proves the first rigorous derivation of the binary-ternary Boltzmann equation $\partial_t f+v\cdot\nabla_x f=Q_2(f,f)+Q_3(f,f,f)$, where $Q_3$ is the ternary interaction-zone operator. The proof must control, under backward time evolution, arbitrary sequences of binary and ternary events, including recollisions (configurations where the interacting flow departs from free flow), grazing collisions (near-tangential encounters), and returns from a binary to a ternary interaction and vice versa; this requires new geometric and algebraic estimates. The main theorem establishes local well-posedness of the BBGKY and Boltzmann hierarchies in Maxwellian-weighted $L^\infty$ spaces (norms of the form $\sup |g|e^{\beta E}$ with $E$ the kinetic energy, with time-dependent weights) and convergence in observables on a time interval $[0,T]$ whose length $T=T(d,\beta_0,\mu_0)$ is dictated by that local theory.

On the side of the limit equation (as opposed to its derivation from particles), the binary-ternary Boltzmann equation is by now fairly well understood: global well-posedness near vacuum in weighted $L^\infty$ spaces was proved by Ampatzoglou, Gamba, Pavlović and Tasković, global mild solutions in an integrable space by Wei and Yu, and generation and propagation of moments for the space-homogeneous case again by Ampatzoglou, Gamba, Pavlović and Tasković. These results do not address the convergence from the microscopic dynamics, which is the subject of this problem. On the hierarchy side, Ampatzoglou, Miller, Pavlović and Tasković proved global-in-time existence and uniqueness for the classical binary Boltzmann hierarchy using polynomially weighted $L^\infty$ bounds and the Klainerman-Machedon board game combinatorics (a combinatorial method that reorganizes the terms of a hierarchy expansion into overlapping pairings that can be estimated collectively), but not for the mixed binary-ternary hierarchy. For the derivation itself, the long-time techniques of Deng, Hani and Ma (a cumulant ansatz keeping memory of the full collision history, plus a cutting algorithm for associated diagrams) exist only for binary collision trees.

Problem Statement

The problem is to determine the validity domain of the rigorous microscopic derivation of the binary-ternary Boltzmann equation established by Ampatzoglou and Pavlović, by removing its three proved restrictions. The underlying particle system consists of $N$ particles in $\mathbb{R}^d$, $d\geq 2$, evolving deterministically, in which two particles undergo an elastic hard-sphere collision when their centers are at distance $\epsilon_2$, and a particle interacts with two others when the non-symmetric distance $d_3(x_i;x_j,x_k)=\sqrt{|x_i-x_j|^2+|x_i-x_k|^2}$ equals $\sqrt{2}\,\epsilon_3$, both collision laws preserving momentum and energy; the proved derivation assumes the scale-separated mixed scaling $N\epsilon_2^{d-1}\simeq N\epsilon_3^{d-1/2}\simeq 1$ with $\epsilon_2\ll\epsilon_3$, and yields convergence of the BBGKY hierarchy to the Boltzmann hierarchy — hence, for chaotic initial data (initial data whose $k$-particle marginals factorize as $f^{\otimes k}$ in the limit $N\to\infty$), of the one-particle density to the solution of the binary-ternary Boltzmann equation — only on a short time interval $[0,T]$ determined by a local well-posedness theory in Maxwellian-weighted $L^\infty$-type spaces. The three linked targets, which together answer the single question of how far beyond these restrictions the kinetic description remains valid, are:

1. Time. Establish the convergence from the microscopic dynamics beyond the short-time regime — a genuine extension of the proved time horizon suffices, whether or not global-in-time validity of the binary-ternary Boltzmann equation is reached (that is, validity for as long as the limiting equation is well-posed, paralleling the binary-case result of Deng, Hani and Ma) — or alternatively prove a rigorous obstruction showing that the derivation genuinely fails beyond some finite time horizon.

2. Scaling. Determine the behavior of the particle system and the resulting kinetic description under other density scalings and other relations between the binary and ternary interaction lengths — for example comparable or non-separated $\epsilon_2$ and $\epsilon_3$, or denser regimes: identify which limiting equations, if any, arise in such regimes, and extend the rigorous derivation to them or prove non-validity. Regimes in which the ternary contribution vanishes in the limit, so that the problem reduces to the already-derived purely binary case, do not by themselves resolve this target; the target concerns regimes beyond the proved scale-separated mixed scaling in which the ternary interaction genuinely persists or the scale separation fails.

3. Interaction laws. Extend the rigorous derivation to more general higher-order interaction mechanisms than the hard-sphere and interaction-zone collision laws of the source paper — for instance genuine three-body (many-body) potentials, different geometric interaction conditions, or other collision laws — including the construction of the corresponding particle dynamics, hierarchies, and collision operators. It suffices for this target to carry out the full derivation for one concretely specified class of such interaction laws — for example a family of smooth short-range three-body potentials; the target does not require covering every more general mechanism.

An answer may address any one of the three targets completely; each has a recognizable completion boundary in the Lanford-derivation tradition (a stated particle model, a stated scaling, a convergence theorem on a stated time window).

References

1. I. Ampatzoglou and N. Pavlović, 'Rigorous derivation of a binary-ternary Boltzmann equation for a dense gas of hard spheres', arXiv:2007.00446 (2020), https://arxiv.org/abs/2007.00446; published as 'Rigorous derivation of a binary-ternary Boltzmann equation for a non ideal gas of hard spheres', Forum of Mathematics, Sigma 13 (2025) e52, DOI 10.1017/fms.2025.11, https://doi.org/10.1017/fms.2025.11
2. I. Ampatzoglou and N. Pavlović, 'Rigorous derivation of a ternary Boltzmann equation for a classical system of particles', Communications in Mathematical Physics 387 (2021) 793-863, DOI 10.1007/s00220-021-04202-y, https://doi.org/10.1007/s00220-021-04202-y; arXiv:1903.04279, https://arxiv.org/abs/1903.04279
3. I. Ampatzoglou, I. M. Gamba, N. Pavlović and M. Tasković, 'Global well-posedness of a binary–ternary Boltzmann equation', Annales de l'Institut Henri Poincaré C, Analyse non linéaire 39 (2022) 327-369, DOI 10.4171/aihpc/9, https://doi.org/10.4171/aihpc/9; arXiv:1910.14476, https://arxiv.org/abs/1910.14476
4. I. Ampatzoglou, I. M. Gamba, N. Pavlović and M. Tasković, 'Moment estimates and well-posedness of the binary-ternary Boltzmann equation', Pure and Applied Analysis 8 (2026) 189-245, DOI 10.2140/paa.2026.8.189, https://doi.org/10.2140/paa.2026.8.189; arXiv:2210.09600, https://arxiv.org/abs/2210.09600
5. J. Wei and Q. Yu, 'On the global existence of mild solutions to binary-ternary Boltzmann equation in an integrable space', Discrete and Continuous Dynamical Systems - B 29 (2024) 2903-2927, DOI 10.3934/dcdsb.2023206, https://doi.org/10.3934/dcdsb.2023206
6. I. Ampatzoglou, J. K. Miller, N. Pavlović and M. 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, https://doi.org/10.1016/j.jfa.2025.111079; arXiv:2402.00765, https://arxiv.org/abs/2402.00765
7. Y. Deng, Z. Hani and X. Ma, 'Long time derivation of the Boltzmann equation from hard sphere dynamics', arXiv:2408.07818 (2024, preprint; latest version v3, 2025, no journal version listed), https://arxiv.org/abs/2408.07818
8. I. Ampatzoglou, N. Pavlović and W. Warner, 'Derivation of the Higher Order Boltzmann Equation for Hard Spheres', arXiv:2409.05669 (2024, preprint), https://arxiv.org/abs/2409.05669
9. E. Cárdenas, N. Pavlović and W. Warner, 'Derivation of a Boltzmann equation with higher-order collisions from a generalized Kac model', SIAM Journal on Mathematical Analysis 56 (2024) 5409-5444, DOI 10.1137/23m1606150, https://doi.org/10.1137/23m1606150; arXiv:2211.02758, https://arxiv.org/abs/2211.02758
10. I. Ampatzoglou, J. K. Miller and N. Pavlović, 'A Rigorous Derivation of a Boltzmann System for a Mixture of Hard-Sphere Gases', SIAM Journal on Mathematical Analysis 54 (2022) 2320-2372, DOI 10.1137/21m1424779, https://doi.org/10.1137/21m1424779
11. M. Wilkinson, 'Maximal Codimension Collisions and Invariant Measures for Hard Spheres on a Line', Journal of Statistical Physics 191 (2024) 93, DOI 10.1007/s10955-024-03310-y, https://doi.org/10.1007/s10955-024-03310-y
12. O. E. Lanford, 'Time evolution of large classical systems', in Dynamical Systems, Theory and Applications, 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
13. I. Gallagher, L. Saint-Raymond and B. Texier, 'From Newton to Boltzmann: Hard Spheres and Short-range Potentials', Zurich Lectures in Advanced Mathematics 18, European Mathematical Society (2014), DOI 10.4171/129, https://doi.org/10.4171/129