Convergence and non-convergence to Bose-Einstein condensation
Shuzhe Cai, Xuguang Lu
TL;DR
This work analyzes how interaction potentials in the space-homogeneous Bose–Einstein Boltzmann equation influence convergence to Bose–Einstein condensation (BEC) at low temperature. By classifying collision kernels into two regimes—one with a lower bound and another with an upper bound on the scattering cross-section—and employing measure-valued isotropic solutions, the authors establish strong convergence to BE equilibrium for the lower-bound regime under suitable initial data, while proving non-convergence in the upper-bound regime when condensation is initially absent. A semi-strong framework links entropy production to a semi-norm distance from BE equilibrium, yielding explicit decay rates that depend on the potential parameter $\eta$. The results highlight the critical effect of the microscopic interaction potential on the macroscopic condensation dynamics, providing both constructive initial data examples and rigorous non-convergence scenarios.
Abstract
The paper is a continuation of our previous work on the strong convergence to equilibrium for the spatially homogeneous Boltzmann equation for Bose-Einstein particles for isotropic solutions at low temperature. Here we study the influence of the particle interaction potentials on the convergence to Bose-Einstein condensation (BEC). Consider two cases of certain potentials that are such that the corresponding scattering cross sections are bounded and 1) have a lower bound ${\rm const.}\min\{1, |{\bf v-v}_*|^{2η}\}$ with ${\rm const}.>0, 0\le η<1$, and 2) have an upper bound ${\rm const.}\min\{1, |{\bf v-v}_*|^{2η}\}$ with $η\ge 1$. For the first case, the long time convergence to BEC i.e. $\lim\limits_{t\to\infty}F_t(\{0\})=F_{\rm be}(\{0\})$ is proved for a class of initial data having very low temperature and thus it holds the strong convergence to equilibrium. For the second case we show that if initially $F_0(\{0\})=0$, then $ F_t(\{0\})=0$ for all $t\ge 0$ and thus there is no convergence to BEC hence no strong convergence to equilibrium.
