Oscillating about coplanarity in the 4 body problem (Q2322815)

This paper analyzes the gravitational Newtonian 4-body problem for the spatial case in which, typically, the four point masses lie at the vertices of a tetrahedron. As the masses move about, at isolated instants this tetrahedron might degenerate so as to be contained in a single plane, such instant are named coplanar instants. A solution is called {\em bounded} if the distances between the bodies remain bounded. The main theorem of this paper states the following: For the 4-body problem in 3-space, any bounded zero angular momentum solution defined on an infinite time interval suffers infinitely many coplanar instants, that is, at infinite number of times all 4 bodies lie in the same plane. This result is a generalization of an earlier result due to the author in [Arch. Ration. Mech. Anal. 164, No. 4, 311--340 (2002; Zbl 1024.70005)], for collinear instants (``syzygies'') in the zero angular momentum planar 3-body problem. Furthermore, it is extended for the \(d + 1\) body problem in \(d\)-dimensional Euclidean space. Some key ingredients behind of the theorems and their proofs are given right. The concept of the {\em degeneration locus} within configuration space is defined as the set of configurations for which the \(d + 1\) masses all lie on a single affine hyperplane. In the case \(d = 3\), it corresponds to the set of coplanar configurations. The Newton's equations and the kinetic energy metric down are pushed to {\em shape space}, it means the quotient space of the \(N\)-body configurations space by the isometry group of \(\mathbb{R}^d\). The author chose to work on the oriented shape space by using the group \(SE(d)\) of orientation-preserving isometries to identify configurations. In fact, a crucial role in the proof is playing for the non-negativity of the curvature of oriented shape space. The concept ``downstairs'' is used for when the author works on the quotient, while ``upstairs'' means he is working on the original configuration space. The proof begins by making some computations in order to exploit the relations between Newton's equations at zero angular momentum upstairs and downstairs. The process of pushing theequations downstairs is referred to as {\em reduction}. This reduction procedure is a metric reduction, putting kinetic energy to the fore, as opposed to the oft-used symplectic reduction. Then, the translation-reduced configuration space is identifying with real \(d\times d\) matrices, and also the degeneration locus with the set of matrices having determinant zero, and the mass metric with the Frobenius (standard Euclidean) norm. In the final section, the author conclude by posing a number of open questions.