paper

Exploring the String Axiverse with Precision Black Hole Physics

arXiv:1004.3558 · doi:10.1103/PhysRevD.83.044026

Abstract

It has recently been suggested that the presence of a plenitude of light axions, an Axiverse, is evidence for the extra dimensions of string theory. We discuss the observational consequences of these axions on astrophysical black holes through the Penrose superradiance process. When an axion Compton wavelength is comparable to the size of a black hole, the axion binds to the black hole "nucleus" forming a gravitational atom in the sky. The occupation number of superradiant atomic levels, fed by the energy and angular momentum of the black hole, grows exponentially. The black hole spins down and an axion Bose-Einstein condensate cloud forms around it. When the attractive axion self-interactions become stronger than the gravitational binding energy, the axion cloud collapses, a phenomenon known in condensed matter physics as "Bosenova". The existence of axions is first diagnosed by gaps in the mass vs spin plot of astrophysical black holes. For young black holes the allowed values of spin are quantized, giving rise to "Regge trajectories" inside the gap region. The axion cloud can also be observed directly either through precision mapping of the near horizon geometry or through gravitational waves coming from the Bosenova explosion, as well as axion transitions and annihilations in the gravitational atom. Our estimates suggest that these signals are detectable in upcoming experiments, such as Advanced LIGO, AGIS, and LISA. Current black hole spin measurements imply an upper bound on the QCD axion decay constant of 2 x 10^17 GeV, while Advanced LIGO can detect signals from a QCD axion cloud with a decay constant as low as the GUT scale. We finally discuss the possibility of observing the gamma-rays associated with the Bosenova explosion and, perhaps, the radio waves from axion-to-photon conversion for the QCD axion.

54 pages, 15 figures; v2: PRD version, small correction in eq. 48 and 53 as well as fig. 10, conclusions unchanged

References in corpus (14)

Cited by in corpus (468)