How Binaries Accrete: Hydrodynamic Simulations with Passive Tracer Particles

Abstract

Linear analysis of gas flows around orbiting binaries suggests that a centrifugal barrier ought to clear a low-density cavity around the binary and inhibit mass transfer onto it. Modern hydrodynamics simulations have confirmed the low-density cavity, but show that any mass flowing from large scales into the circumbinary disk is eventually transferred onto the binary components. Even though many numerical studies confirm this picture, it is still not understood precisely how gas parcels overcome the centrifugal barrier and ultimately accrete. We present a detailed analysis of the binary accretion process, using an accurate prescription for evolving grid-based hydrodynamics with Lagrangian tracer particles that track the trajectories of individual gas parcels. We find that binary accretion can be described in four phases: (1) gas is viscously transported through the circumbinary disk up to the centrifugal barrier at the cavity wall, (2) the cavity wall is tidally distorted into accretion streams consisting of near-ballistic gas parcels on eccentric orbits, (3) the portion of each stream moving inwards of an accretion horizon radius r ¯ ≃ a —the radius beyond which no material is returned to the cavity wall—becomes bound to a minidisk orbiting an individual binary component, and (4) the minidisk gas accretes onto the binary component through the combined effect of viscous and tidal stresses.

Contracting and Expanding Binary Black Holes in 3D Low-mass AGN Disks: The Importance of Separation

“…Figure 5 shows the evolution of the accretion rate onto each BH as a function of time for one simulation with a b = R H /4. As can be seen in Figure 5, an equal-mass and circular BBH accretes symmetrically with m m 1 2   » , as expected from previous studies (e.g., Muñoz et al 2019;Tiede et al 2022).…”

On the Role of Dynamical Cooling in the Dynamics of Circumbinary Disks

“…As is demonstrated in Appendix, given the parameters we have (carefully) chosen, our results are largely unaffected by this treatment, and a quasisteady state for accretion and angular momentum transfer can be achieved without resolving the minidisks. On the other hand, we show in Appendix that this approach is likely inappropriate to study thinner disks (H/r < 0.1), where the eccentricity of the cavity becomes significantly larger, and an excised central domain would erroneously prevent materials on eccentric orbits around the cavity from returning to the disk (Tiede et al 2022).…”

“…We note that while many recent simulations of CBDs fully enclose the binary in the simulation domain, the properties of the CBDs, particularly the angular momentum flux through the disk, are insensitive to this choice as long as there is no significant mass ejection from the central binary. Our choice of parameters is not subject to such a phenomenon, as justified in Tiede et al (2022), as well as by comparing simulation results of Miranda et al (2017) and Muñoz et al (2019) for the same parameters.…”

“…For such an eccentric disk, the "diode" inner boundary condition is no longer valid. Materials pass close to the binary will slingshot to a larger radius (Tiede et al 2022;e., we expect both inflow and outflow at r in = a B ), while the "diode" inner boundary forbids such materials to return to the simulation domain. As can be…”

“…The former is computationally more efficient and better conserves angular momentum for the outer CBDs, which we adopt in this work. We also note that this approach is valid when materials entering the co-orbital region of the central binary do not get ejected across the chosen inner boundary, as has been found to be the case for the parameters that we choose (Muñoz et al 2019;Tiede et al 2022). In a companion paper (Wang et al 2022), we will expand the parameter space and make our computational domain enclose the central binary.…”