Search

It has been well-established that particular centrophilic orbital families in non-spherical galaxies can, in principle, drive a black hole binary to shrink its orbit through three-body scattering until the black holes are close enough to strongly emit gravitational waves. Most of these studies rely on orbital analysis of a static SMBH-embedded galaxy potential to support this view; it is not clear, however, how these orbits transform as the second SMBH enters the center, so our understanding of which orbits actually interact with a SMBH binary is not ironclad. Here, we analyze two flattened galaxy models, one with a single SMBH and one with a binary, to determine which orbits actually do interact with the SMBH binary and how they compare with the set predicted in single SMBH-embedded models. We find close correspondence between the centrophilic orbits predicted to interact with the binary and those that are actually scattered by the binary, in terms of energy and Lz distribution, where Lz is the z component of a stellar particle's angular momentum. Of minor note: because of the larger mass, the binary SMBH has a radius of influence about 4 times larger than in the single SMBH model, which allows the binary to draw from a larger reservoir of orbits to scatter. Of the prediction particles and scattered particles, nearly half have chaotic orbits, 40% have fx:fy=1:1 orbits, 10% have other resonant orbits.

Galaxy centers are residing places for Super Massive Black Holes (SMBHs). Galaxy mergers bring SMBHs close together to form gravitationally bound binary systems which, if able to coalesce in less than a Hubble time, would be one of the most promising sources of gravitational waves for the Laser Interferometer Space Antenna (LISA). In spherical galaxy models, SMBH binaries stall at a separation of approximately one parsec, leading to the "final parsec problem" (FPP). On the other hand, it has been shown that merger-induced triaxiality of the remnant in equal-mass mergers is capable of supporting a constant supply of stars on so-called centrophilic orbits that interact with the binary and thus avoid the FPP. In this paper, using a set of direct N-body simulations of mergers of initially spherically symmetric galaxies with different mass ratios, we show that the merger-induced triaxiality is able to drive unequal-mass SMBH binaries to coalescence. The binary hardening rates are high and depend only weakly on the mass ratios of SMBHs for a wide range of mass ratios q. The hardening rates are significantly higher for galaxies having steep cusps in comparison with those having shallow cups at centers. The evolution of the binary SMBH leads to relatively shallower inner slopes at the centers of the merger remnants. The stellar mass displaced by the SMBH binary on its way to coalescence is ~ 1-5 times the combined mass of binary SMBHs. The coalescence times for SMBH binary with mass ~ million solar masses are less than 1 Gyr and for those at the upper end of SMBH masses (~ billion solar masses) are 1-2 Gyr for less eccentric binaries whereas less than 1 Gyr for highly eccentric binaries. SMBH binaries are thus expected to be promising sources of gravitational waves at low and high redshifts., Comment: Accepted for publication in the Astrophysical Journal (ApJ). 14 pages, 8 figures

In spherical galaxies, binary supermassive black holes (SMBHs) have difficulty reaching sub-parsec separations due to depletion of stars on orbits that intersect the massive binary - the final-parsec problem. Galaxies that form via major mergers are substantially nonspherical, and it has been argued that the centrophilic orbits in triaxial galaxies might provide stars to the massive binary at a high enough rate to avoid stalling. Here we test that idea by carrying out fully self-consistent merger simulations of galaxies containing central SMBHs. We find hardening rates of the massive binaries that are indeed much higher than in spherical models, and essentially independent of the number of particles used in the simulations. Binary eccentricities remain high throughout the simulations. Our results constitute a fully stellar-dynamical solution to the final-parsec problem and imply a potentially high rate of events for low-frequency gravitational wave detectors like LISA., Comment: 9 pages, 11 figures. Accepted for publication in The Astrophysical Journal