Evolution of a black hole cluster in full general relativity

We evolve for the first time in full general relativity a small, collisional N-body black hole cluster of arbitrary total mass M. The bound cluster is initially compact (radius R/M~10), stable, and consists of 25 equal-mass, nonspinning black holes. The dynamical interactions of compact objects in N-body clusters is of great interest for the formation of black holes in the upper mass gap as well as intermediate and supermassive black holes. These are potential sources of gravitational waves that may be detected by both current and future observatories. Unlike previous N-body Newtonian and post-Newtonian simulations, no "subgrid physics" is required to handle collisions and mergers. We can therefore confirm in full general relativity several predictions from these simulations and analytic estimates: the runaway growth of a large black hole via repeated mergers; spindown of the central black hole with increasing captures; the ejection of a black hole with a large asymptotic velocity due to a several-body interaction; and a regime where mergers occur primarily via direct collisions on highly eccentric orbits instead of quasicircular inspirals. We extract the gravitational wave signal and find it has several distinct features associated with the compact cluster regime. Our results suggest the signal is sufficiently loud that next generation observatories would likely be able to detect similar events across most of the observable universe. This work is a preliminary proof-of-principle study that we hope will open up a new arena for numerical relativity and the study of N-body compact systems.