Massive Black Hole Pairs in Clumpy, Self-Gravitating Circumnuclear Disks: Stochastic Orbital Decay

Abstract

We study the dynamics of massive black hole pairs in clumpy gaseous circumnuclear disks. We track the orbital decay of the light, secondary black hole M •2 orbiting around the more massive primary at the center of the disk, using N -body/smoothed particle hydrodynamic simulations. We find that the gravitational interaction of M •2 with massive clumps M cl erratically perturbs the otherwise smooth orbital decay. In close encounters with massive clumps, gravitational slingshots can kick the secondary black hole out of the disk plane. The black hole moving on an inclined orbit then experiences the weaker dynamical friction of the stellar background, resulting in a longer orbital decay timescale. Interactions between clumps can also favor orbital decay when the black hole is captured by a massive clump which is segregating toward the center of the disk. The stochastic behavior of the black hole orbit emerges mainly when the ratio M •2 /M cl falls below unity, with decay timescales ranging from ∼ 1 to ∼ 50 Myr. This suggests that describing the cold clumpy phase of the inter-stellar medium in self-consistent simulations of galaxy mergers, albeit so far neglected, is important to predict the black hole dynamics in galaxy merger remnants.

The Pairing of Accreting Massive Black Holes in Multiphase Circumnuclear Disks: the Interplay Between Radiative Cooling, Star Formation, and Feedback Processes

“…The latter produces a highly inhomogeneous, multiphase background, affecting the orbital decay by making it stochastic, and often leading to a much longer decay timescale (Figure 1). This general outcome confirms the findings of Fiacconi et al (2013), who used a simple constant-cooling model and had no star formation, BH accretion, nor feedback processes implemented. The degree of clumpiness, however, depends on the specific physics implemented in the runs.…”

“…Indeed, our clumps are generally more compact and much denser than the typical GMC in the disk of the Milky Way. Despite the richer physics of the simulations presented here, the sizes and densities of the clumps are similar to those found in Fiacconi et al (2013), for which it was shown that a good match exists with clouds in the Galactic Center region (Oka et al 2001), which, albeit less extreme, represents an environment with high gas densities and stronger stellar radiation background more akin to what one expects in the nucleus of a merger remnant, which is what our CND model should capture. We repeated the same comparison here ( Figure A4).…”

“…In run NAC, especially after 3-5 Myr, there are more than 10 clumps with masses from commensurate to ten times more massive than the secondary BH. As discussed in Fiacconi et al (2013) and Roškar et al (2015), clumps with masses comparable to or larger than the mass of the secondary BH can have a dramatic effect on the latter, with direct hits or even repeated softer perturbations. At the same time, Figure 6 shows that most of the clumps have masses about the same or less than the secondary BH, namely close to or Gas density face-on projection and secondary BH's trajectory for run C (see Table 1), similar to Figure 2.…”

“…The effects of a clumpy environment on the dynamics of a BH pair in a range of BH mass ratios (1:1 to 1:5) in a gaseous CND are studied in Fiacconi et al (2013). The "lumpiness" of the disk is triggered and tuned by considering an artificial cooling term in the energy equation satisfied by the gas particles.…”

“…The BH masses are chosen to be in the range of those relevant for the future evolved Laser Interferometer Space Antenna (eLISA). Concerning other previous similar works, Fiacconi et al (2013) carried out a large set of CND simulations with massive BH pairs, but did not consider the effects of star formation and SNe, nor BH accretion or AGN feedback. The work of del Valle et al (2015), which investigated the dynamics of BH pairs in starforming CNDs, supports the picture that the orbits can be stochastically perturbed by high-density gas clumps, although they also find that the strength of such interactions is sensitive to the BHs softening length, decaying when the softening decreases.…”