The Restricted Three-Body Problem as a Perturbed Duffing Equation

TL;DR

This paper reformulates the restricted three-body problem as a perturbed Duffing equation, proving the existence of homoclinic tangles for large negative Jacobi constants across various mass ratios, using a simplified and self-contained approach.

Contribution

It introduces a new coordinate transformation to express the problem as a perturbed Duffing equation and applies a streamlined Poincaré-Melnikov method without McGehee's analysis.

Findings

Homoclinic tangles exist for large negative Jacobi constants.

Invariant manifolds depend analytically on mass ratio and Jacobi constant.

The approach simplifies previous proofs and is self-contained.

Abstract

This paper investigates the restricted circular planar three-body problem. We prove that for every negative Jacobi constant of sufficiently large magnitude, the surface of unperturbed parabolic solutions breaks to induce homoclinic tangle for all but at most finitely many mass ratios of primaries. This result is not covered by \cite{G} as the required large magnitude of the Jacobi constant is uniform across the mass ratios. Our approach consists of three main ingredients. First, by introducing new coordinate transformations, we reformulate the restricted three-body problem as a perturbed Duffing equation. Second, we adopt the method recently introduced in \cite{CW} to derive integral equations for the primary stable and unstable solutions. This enables us to effectively capture the order of the singularities involved and to further establish the existence and analytic dependence of the invariant manifolds on the mass ratio of the primaries and the Jacobi constant. Third, in evaluating the Poincar\'e-Melnikov integral, we take advantage of the explicit homoclinic solution of the unperturbed Duffing equation. Compared to existing works, our proof is significantly simpler and more direct. Moreover, the paper is self-contained: we do not rely on McGehee's analysis in \cite{Mc} to justify the applicability of the Poincar\'e-Melnikov method.