No Arabic abstract
We study the flow structure in 3D magnetohydrodynamic (MHD) simulations of accretion onto Sagittarius A* via the magnetized winds of the orbiting Wolf-Rayet stars. These simulations cover over 3 orders of magnitude in radius to reach $approx$ 300 gravitational radii, with only one poorly constrained parameter (the magnetic field in the stellar winds). Even for winds with relatively weak magnetic fields (e.g., plasma $beta$ $sim$ $10^6$), flux freezing/compression in the inflowing gas amplifies the field to $beta$ $sim$ few well before it reaches the event horizon. Overall, the dynamics, accretion rate, and spherically averaged flow profiles (e.g., density, velocity) in our MHD simulations are remarkably similar to analogous hydrodynamic simulations. We attribute this to the broad distribution of angular momentum provided by the stellar winds, which sources accretion even absent much angular momentum transport. We find that the magneto-rotational instability is not important because of i) strong magnetic fields that are amplified by flux freezing/compression, and ii) the rapid inflow/outflow times of the gas and inefficient radiative cooling preclude circularization. The primary effect of magnetic fields is that they drive a polar outflow that is absent in hydrodynamics. The dynamical state of the accretion flow found in our simulations is unlike the rotationally supported tori used as initial conditions in horizon scale simulations, which could have implications for models being used to interpret Event Horizon Telescope and GRAVITY observations of Sgr A*.
We present Athena++ grid-based, hydrodynamic simulations of accretion onto Sagittarius A* via the stellar winds of the $sim 30$ Wolf-Rayet stars within the central parsec of the galactic center. These simulations span $sim$ 4 orders of magnitude in radius, reaching all the way down to 300 gravitational radii of the black hole, $sim 32$ times further in than in previous work. We reproduce reasonably well the diffuse thermal X-ray emission observed by Chandra in the central parsec. The resulting accretion flow at small radii is a superposition of two components: 1) a moderately unbound, sub-Keplerian, thick, pressure-supported disc that is at most (but not all) times aligned with the clockwise stellar disc, and 2) a bound, low-angular momentum inflow that proceeds primarily along the southern pole of the disc. We interpret this structure as a natural consequence of a few of the innermost stellar winds dominating accretion, which produces a flow with a broad distribution of angular momentum. Including the star S2 in the simulation has a negligible effect on the flow structure. Extrapolating our results from simulations with different inner radii, we find an accretion rate of $sim$ a few $times 10^{-8} M_odot$/yr at the horizon scale, consistent with constraints based on modeling the observed emission of Sgr A*. The flow structure found here can be used as more realistic initial conditions for horizon scale simulations of Sgr A*.
We present 3D general relativistic magnetohydrodynamic (GRMHD) simulations of the accretion flow surrounding Sagittarius A* that are initialized using larger-scale MHD simulations of the $sim$ 30 Wolf--Rayet (WR) stellar winds in the Galactic center. The properties of the resulting accretion flow on horizon scales are set not by ad hoc initial conditions but by the observationally constrained properties of the WR winds with limited free parameters. For this initial study we assume a non-spinning black hole. Our simulations naturally produce a $sim 10^{-8} M_odot$ yr$^{-1}$ accretion rate, consistent with previous phenomenological estimates. We find that a magnetically arrested flow is formed by the continuous accretion of coherent magnetic field being fed from large radii. Near the event horizon, the magnetic field is so strong that it tilts the gas with respect to the initial angular momentum and concentrates the originally quasi-spherical flow to a narrow disk-like structure. We also present 230 GHz images calculated from our simulations where the inclination angle and physical accretion rate are not free parameters but are determined by the properties of the WR stellar winds. The image morphology is highly time variable. Linear polarization on horizon scales is coherent with weak internal Faraday rotation.
The magnetic fields of accretion disks play an important role in studying their evolution. We may assume that its generation is connected to the dynamo mechanism, which is similar with that in the galactic disks. Here, we propose a model of the magnetic field of the accretion disk that uses the same approaches that have been used for galaxies. It is necessary to obtain the field, which is expected to be less than the equipartition value, and without destroying the disk. To do so, it is necessary to formulate the basic properties of the ionized medium and to estimate the parameters governing the dynamo. We used the no-z approximation that has been developed for thin disks. We also take different boundary conditions that can change the value of the field significantly. We show that the magnetic field strictly depends on the boundary conditions. Taking zero conditions and the fixed magnetic field condition on the inner boundary, which are connected to the physical properties of the accretion disk, we can avoid solutions that are greater than the equipartition field.
Millimeter wave Very Long Baseline Interferometry (mm-VLBI) provides access to the emission region surrounding Sagittarius A*, the supermassive black hole at the center of the Milky Way, on sub-horizon scales. Recently, a closure phase of 0+-40 degrees was reported on a triangle of Earth-sized baselines (SMT-CARMA-JCMT) representing a new constraint upon the structure and orientation of the emission region, independent from those provided by the previously measured 1.3mm-VLBI visibility amplitudes alone. Here, we compare this to the closure phases associated with a class of physically motivated, radiatively inefficient accretion flow models, and present predictions for future mm-VLBI experiments with the developing Event Horizon Telescope (EHT). We find that the accretion flow models are capable of producing a wide variety of closure phases on the SMT-CARMA-JCMT triangle, and thus not all models are consistent with the recent observations. However, those models that reproduce the 1.3mm-VLBI visibility amplitudes overwhelmingly have SMT-CARMA-JCMT closure phases between +-30 degrees, and are therefore broadly consistent with all current mm-VLBI observations. Improving station sensitivity by factors of a few, achievable by increases in bandwidth and phasing together multiple antennas at individual sites, should result in physically relevant additional constraints upon the model parameters and eliminate the current 180 degree ambiguity on the source orientation. When additional stations are included, closure phases of order 45--90 degrees are typical. In all cases the EHT will be able to measure these with sufficient precision to produce dramatic improvements in the constraints upon the spin of Sgr A*.
Supermassive stars (SMS; ~ 10^5 M_sun) formed from metal-free gas in the early Universe attract attention as progenitors of supermassive black holes observed at high redshifts. To form SMSs by accretion, central protostars must accrete at as high rates as ~ 0.1-1 M_sun/yr. Such protostars have very extended structures with bloated envelopes, like super-giant stars, and are called super-giant protostars (SGPSs). Under the assumption of hydrostatic equilibrium, SGPSs have density inverted layers, where the luminosity becomes locally super-Eddington, near the surface. If the envelope matter is allowed to flow out, however, a stellar wind could be launched and hinder the accretion growth of SGPSs before reaching the supermassive regime. We examine whether radiation-driven winds are launched from SGPSs by constructing steady and spherically symmetric wind solutions. We find that the wind velocity does not reach the escape velocity in any case considered. This is because once the temperature falls below ~ 10^4 K, the opacity plummet drastically owing to the recombination of hydrogen and the acceleration ceases suddenly. This indicates that, in realistic non-steady cases, even if outflows are launched from the surface of SGPSs, they would fall back again. Such a wind does not result in net mass loss and does not prevent the growth of SGPSs. In conclusion, SGPSs will grow to SMSs and eventually collapse to massive BHs of ~ 10^5 M_sun, as long as the rapid accretion is maintained.