Propagation Estimates for the Boson Star Equation

Sébastien Breteaux, Jérémy Faupin, Viviana Grasselli · 2025

Read Propagation Estimates for the Boson Star Equation by Sébastien Breteaux, Jérémy Faupin, Viviana Grasselli (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-20718.txt.

Introduction

Sébastien Breteaux, Jérémy Faupin, and Viviana Grasselli study the boson star equation, a model for stars held together by quantum particles and their mutual attraction. They ask how fast a disturbance in such a system can spread. The interaction between particle pairs is left general, covering many possible force laws.

They prove the solution cannot outrun the speed of light, apart from a remainder that shrinks exponentially and cannot be improved. For short-range forces with smooth, small starting states, they add sharper results. These bound how the solution travels through phase space over long times, including a minimum speed tied to the momentum of the final scattering state.

Catalog record

CREATORSébastien Breteaux, Jérémy Faupin, Viviana Grasselli
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE167 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-20718

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Propagation Estimates for the Boson Star Equation Sébastien Breteaux, Jérémy Faupin, Viviana Grasselli Abstract We consider the boson star equation with a general two-body interaction potential w w and initial data ψ 0 \psi_{0} in a Sobolev space. Under general assumptions on w w , namely that w w decomposes as a sum of a finite, signed measure and an essentially bounded function, we prove that the (local in time) solution cannot propagate faster than the speed of light, up to a sharp exponentially small remainder term. If w w is short-range and ψ 0 \psi_{0} is regular and small enough, we prove in addition asymptotic phase-space propagation estimates and minimal velocity estimates for the (global in time) solution, depending on the momentum of the scattering state associated to ψ 0 \psi_{0} . 1 Introduction In this paper we study the pseudo-relativistic (or semi-relativistic) Hartree equation \displaystyle\begin{cases}i\partial_{t}\psi=(\langle\nabla\rangle+w*|\psi|^{2})\psi,\quad x\in\mathbb{R}^{d},\,\quad t\geq 0,\\ \psi|_{t=0}=\psi_{0},\end{cases} (1.1) in any dimension d\geq 3 . Here and throughout the paper, we use the notation \langle\nabla\rangle for the Fourier multiplier \langle\nabla\rangle:=\sqrt{1-\Delta}=\mathcal{F}^{-1}\sqrt{1+|\xi|^{2}}\mathcal{F}, where \mathcal{F} is the Fourier transform normalized such that \mathcal{F} is unitary on L^{2} . Eq. ( 1.1 ) describes the effective dynamics, in the mean-field limit, of an N -body quantum system of pseudo-relativistic bosons of mass m=1 , with two-body gravitational interaction given by the convolution potential w ; see [ 18 ] . It is thus used as a model for a pseudo-relativistic boson star. Here we work in units such that Planck’s constant divided by 2\pi and the velocity of light are equal to 1 . We also assume that the mass of the bosons is 1 for simplicity, but we could consider the kinetic energy \sqrt{m^{2}-\Delta} (or \sqrt{m^{2}-\Delta}-m ), for any m>0 , without changing our results. A physically relevant choice of the interaction potential w is the attractive Newtonian potential, w(x)=-\kappa|x|^{-1} for some \kappa>0 , but we will consider in this paper general potentials in suitable spaces, imposing different conditions depending on the results. The assumption w\geq 0 corresponds to repulsive, or defocusing interactions, while w\leq 0 corresponds to attractive, or focusing ones. We use the shorthand \psi_{t} for a solution (t,x)\mapsto\psi(t,x) to ( 1.1 ). The boson star equation ( 1.1 ) exhibits (formally at this stage) three important conserved quantities: the mass, M(\psi_{t}):=\int_{\mathbb{R}^{d}}|\psi_{t}|^{2}; (1.2) the energy, defined by E(\psi_{t}):=\frac{1}{2}\int_{\mathbb{R}^{d}}\big|\langle\nabla\rangle^{\frac{1}{2}}\psi_{t}\big|^{2}+\frac{1}{4}\int_{\mathbb{R}^{d}}(w*|\psi_{t}|^{2})|\psi_{t}|^{2} (1.3) and the momentum, P(\psi_{t}):=-\frac{i}{2}\int_{\mathbb{R}^{d}}\bar{\psi}_{t}\nabla\psi_{t}. (1.4) We consider in this paper a translation invariant boson star equation, but several of our results might be extended to a boson star placed in an external potential V(x) , provided that suitable regularity and decay assumptions on V are imposed. In this case, of course, the momentum is not a conserved quantity anymore. For the attractive Newtonian potential w(x)=-\kappa|x|^{-1} with \kappa>0 , the convergence, in the mean-field limit, of the ground state energy of the pseudo-relativistic N -body Schrödinger Hamiltonian to the ground state energy of the static, pseudo-relativistic Hartree equation corresponding to ( 1.1 ), was established by Lieb and Yau in their seminal paper [ 47 ] . The fermionic case is also considered in [ 47 ] , in relation with the Chandrasekhar theory of stellar collapse. As mentioned before, the convergence in the mean-field limit of the time-dependent pseudo-relativistic Schrödinger equation to the pseudo-relativistic Hartree equation ( 1.1 ) was proven in [ 18 ] . See also [ 40 , 41 ] for explicit rates of convergence in the mean-field limit. In [ 45 ] , local and global well-posedness are studied in dimension d=3 for ( 1.1 ) with a general external potential V and a Yukawa-type interaction potential, w(x)=\kappa|x|^{-1}e^{-\mu|x|} with \kappa\in\mathbb{R} , \mu\geq 0 , and for initial data \psi_{0} in the Sobolev space H^{s} , with s\geq\frac{1}{2} (for attractive interaction potentials, a smallness condition must be imposed, either on \psi_{0} or on w ). The local well-posedness is extended in [ 35 ] to low-regularity initial states, namely \psi_{0}\in H^{s} with s>\frac{1}{4} . Here it should be noticed that the energy space corresponds to the Sobolev regularity H^{1/2} . In [ 12 , 13 , 11 ] , global existence – and small initial data scattering – are proven for (sums of) short-range interaction potentials of the form w_{i}(x)=\kappa_{i}|x|^{-\alpha_{i}} , for some suitable values of \alpha_{i} ; see also [ 36 ] for related results in the case of Yukawa-type potentials w(x)=\kappa|x|^{-1}e^{-\mu|x|} with \mu>0 . We will compare some of these results to ours below, after the statement of our main contributions. For long-range potentials w(x)=\kappa|x|^{-1} , “modified scattering states” must be introduced; see [ 53 ] for global existence and small initial data scattering in this case. In relation with gravitational collapse, finite time blow-up for ( 1.1 ) is proven in [ 27 ] for attractive Newtonian potentials w(x)=-\kappa|x|^{-1} with \kappa>0 and the mass M(\psi_{0})=\|\psi_{0}\|_{L^{2}}^{2} of the initial state larger than some critical value. The blow-up phenomenon is analysed in the mean-field limit in [ 51 ] . For the existence of solitary waves and stability results around them, under suitable assumptions, we refer to [ 25 , 26 ] . We will not study finite time blow-up nor the existence of solitary waves here. Our main concern is the speed of propagation of boson stars whose dynamics is given by ( 1.1 ). We generally consider a large class of interaction potentials, w\in\mathcal{M}+L^{\infty}, (1.5) where \mathcal{M} stands for the Banach space of finite, signed Radon measures on \mathbb{R}^{d} . In particular, we can take w as a Dirac delta measure and hence consider a pure-power cubic non-linearity. Depending on the results, we will restrict the class of admissible potentials. We will begin with establishing local and global existence of solutions to ( 1.1 ). Our proof follows the usual strategy of applying a fixed point argument to solve Duhamel’s equation associated to ( 1.1 ) in a suitable function space. For the global existence, we will distinguish two distinct regimes: the long-range regime, w\in L^{d/2+1,\infty}+L^{\infty} , and the short-range regime, w\in\mathcal{M}+L^{q} , 1\leq q<\frac{2d}{3} , with possibly a smallness condition involving the initial state \psi_{0} . Compared to the extensively studied non-linear Schrödinger or Hartree equations (see e.g. [ 9 , 10 , 61 , 29 ] and references therein), a difficulty, as in previous works, comes from the loss of derivatives in the dispersive estimates associated to the semi-relativistic kinetic energy, which, for the L^{1}\to L^{\infty} estimate, may be written as \big\|e^{-it\langle\nabla\rangle}f\big\|_{L^{\infty}}\lesssim|t|^{-\frac{d}{2}}\|\langle\nabla\rangle^{\frac{d}{2}+1}f\|_{L^{1}}, (1.6) see Appendix B . Given a solution \psi_{t} to ( 1.1 ), we will aim at estimating the speed of propagation of the boson star, in the sense of proving time-decay estimates for the probability that the velocity \frac{x}{t} of the state \psi_{t} belongs to a certain domain, possibly depending on the position and momentum of the initial state \psi_{0} . We will establish both a maximal and a minimal velocity estimate, expressed under different forms. Our maximal velocity bound is of the form \|\mathbf{1}_{Y}\psi_{t}\|_{L^{2}}\leq e^{t-\mathrm{dist}(X,Y)}\|\mathbf{1}_{X}\psi_{0}\|_{L^{2}}, (1.7) for any convex subsets X,Y\subset\mathbb{R}^{d} , where \mathrm{dist}(X,Y) stands for the distance from X to Y . It holds for any local in time solution \psi_{t} to ( 1.1 ), for any interaction potential w\in\mathcal{M}+L^{\infty} , provided that the initial state

Read the complete text