Stability of rotating scalar boson stars with nonlinear interactions

We study the stability of rotating scalar boson stars, comparing those made from a simple massive complex scalar (referred to as mini boson stars), to those with several different types of nonlinear interactions. To that end, we numerically evolve the nonlinear Einstein-Klein-Gordon equations in 3D, beginning with stationary boson star solutions. We show that the linear, non-axisymmetric instability found in mini boson stars with azimuthal number m = 1 persists across the entire parameter space for these stars, though the timescale diverges in the Newtonian limit. Therefore, any boson star with m = 1 that is sufficiently far into the nonrelativistic regime, where the leading order mass term dominates, will be unstable, independent of the nonlinear scalar self-interactions. However, we do find regions of m = 1 boson star parameter space where adding nonlinear interactions to the scalar potential quenches the nonaxisymmetric instability, both on the nonrelativistic, and the relativistic branches of solutions. We also consider select boson stars with in = 2, finding instability in all cases. For the cases exhibiting instability, we follow the nonlinear development, finding a range of dynamics including fragmentation into multiple unbound nonrotating stars, and formation of binary black holes. Finally, we comment on the relationship between stability and criteria based on the rotating boson star's frequency in relation to that of a spherical boson star or the existence of a corotation point. The boson stars that we find not to exhibit instability when evolved for many dynamical times include rapidly rotating cases where the compactness is comparable to that of a black hole or neutron star.

Automated summary

The study shows that rotating scalar boson stars with m = 1 are always unstable, while adding nonlinear interactions to m = 1 boson stars may weaken this non-axisymmetric instability. Additionally, boson stars with m = 2 are unstable in all cases.