The Final Fate of Spherical Inhomogeneous Dust Collapse

TL;DR

This paper investigates how initial density and velocity profiles influence the final outcome of spherical dust collapse, revealing conditions under which the collapse results in black holes or naked singularities, and generalizing previous results.

Contribution

It provides a detailed analysis of initial conditions leading to black holes or naked singularities in dust collapse, extending earlier work by removing symmetry assumptions.

Findings

Collapse can end in a naked singularity depending on initial density derivatives.

Homogeneous dust collapse always results in a black hole.

Relaxing symmetry assumptions allows for smooth transitions between outcomes.

Abstract

We examine the role of the initial density and velocity distribution in the gravitational collapse of a spherical inhomogeneous dust cloud. Such a collapse is described by the Tolman-Bondi metric which has two free functions: the `mass-function' and the `energy function', which are determined by the initial density and velocity profile of the cloud. The collapse can end in a black-hole or a naked singularity, depending on the initial parameters characterizing these profiles. In the marginally bound case, we find that the collapse ends in a naked singularity if the leading non-vanishing derivative of the density at the center is either the first one or the second one. If the first two derivatives are zero, and the third derivative non-zero, the singularity could either be naked or covered, depending on a quantity determined by the third derivative and the central density. If the first three derivatives are zero, the collapse ends in a black hole. In particular, the classic result of Oppenheimer and Snyder, that homogeneous dust collapse leads to a black hole, is recovered as a special case. Analogous results are found when the cloud is not marginally bound, and also for the case of a cloud starting from rest. We also show how the strength of the naked singularity depends on the density and velocity distribution. Our analysis generalizes and simplifies the earlier work of Christodoulou and Newman [4,5] by dropping the assumption of evenness of density functions. It turns out that relaxing this assumption allows for a smooth transition from the naked singularity phase to the black-hole phase, and also allows for the occurrence of strong curvature naked singularities.