Convolution estimates for the Boltzmann gain operator with hard spheres
Ioakeim Ampatzoglou, Tristan Léger
TL;DR
This work establishes moment-preserving convolution estimates for the Boltzmann gain operator in hard potentials, including hard spheres, by developing a purely kinetic approach based on Bobylev variables, angular averaging, and a cancellation mechanism for energy-absorbing collisions. The authors prove a sharp bilinear bound for $Q^+$ in polynomially weighted spaces: for $0<\gamma\le1$, $1<p<2<q<\infty$, $2<r\le\infty$, and suitable weights $k\ge l>1$, $\|\langle v\rangle^{k+1-\gamma} Q^+(f,g)\|_{L^r} \lesssim \|\langle v\rangle^l f\|_{L^p}\|\langle v\rangle^k g\|_{L^q} + \|\langle v\rangle^l f\|_{L^q}\|\langle v\rangle^k g\|_{L^p}$ for $f,g$ in a natural weighted space $\bm{\mathcal{X}_{p,q}^k}$. The paper also extends Maxwell-molecule convolution estimates to arbitrary $r$ and provides a detailed, self-contained kinetic proof of these results without reliance on Fourier-analytic machinery. The Cancellation Lemma plays a central role by showing that rare but energy-absorbing collisions do not destroy the average smoothing, enabling moment preservation in the hard-potential regime. These results advance the understanding of regularization and moment behavior in the Boltzmann equation with hard potentials and have potential implications for long-time dynamics and hydrodynamic limits.
Abstract
We prove new moment-preserving polynomially weighted convolution estimates for the gain operator of the Boltzmann equation with hard potentials, including the critical case of hard-spheres. Our approach relies crucially on a novel cancellation mechanism dealing with the pathological case of energy-absorbing collisions (that is, collisions that accumulate energy to only one of the outgoing particles). This difficulty is specific to hard potentials, and is not present for Maxwell molecules. Our method quantifies the heuristic that, while energy-absorbing collisions occur with non-trivial probability, they are statistically rare, and therefore do not affect the overall averaging behavior of the gain operator. At the technical level, our proof relies solely on tools from kinetic theory, such as geometric identities and angular averaging.
