https://arxiv.org/pdf/1606.08871v1.pdf Xiv:1606.08871v1 [astro-ph.GA] 28 Jun 2016 The origin of most massive black holes at high-z: BLUETIDES and the next quasar frontier: ABSTRACT : The growth of the most massive black holes in the early universe, consistent with the detection of highly luminous quasars at z > 6 implies sustained, critical accretion of material to grow and power them. Given a black hole seed scenario, it is still uncertain which conditions in the early Universe allow the fastest black hole growth. Large scale hydrodynamical cosmological simulations of structure formation allow us to explore the conditions conducive to the growth of the earliest supermassive black holes. We use the cosmological hydrodynamic simulation BlueTides, which incorporates a variety of baryon physics in a (400h −1Mpc)3 volume with 0.7 trillion particles to follow the earliest phases of black hole critical growth. At z = 8 the most massive black holes (a handful) approach masses of 108 M with the most massive (with MBH = 4×108 M) being found in an extremely compact spheroid-dominated host galaxy. Examining the large-scale environment of hosts, we find that the initial tidal field is more important than overdensity in setting the conditions for early BH growth. In regions of low tidal fields gas accretes ’cold’ onto the black hole and falls along thin, radial filaments straight into the center forming the most compact galaxies and most massive black holes at earliest times. Regions of high tidal fields instead induce larger coherent angular momenta and influence the formation of the first population of massive compact disks. The extreme early growth depends on the early interplay of high gas densities and the tidal field that shapes the mode of accretion. Mergers play a minor role in the formation of the first generation, rare massive BHs. Conclusion: We use the hydrodynamical cosmological simulation BlueTidesto study the origin of the most massive black holes at early times. Here we have focussed our analysis on the massive galaxies and black holes at z > 8 (for earlier discussion of z = 6 SDSS quasars, see e.g. Di Matteo et al. 2012; Feng et al. 2014). In Feng et al. (2015) we discussed how the most massive galaxies at these epochs are likely disk dominated. Here we have shown that while the most massive disks can still host massive black holes the most extreme early black hole growth (reaching a MBH a few 108 M at z = 8) occurs in spheroid-dominated extremely compact galaxies. The sites of extremely rapid black hole growth hosted by spheroid-dominated galaxies are the result of large-scale filamentary accretion of cold gas from which the halo and eventually black holes can accrete new material radially (at low angular momentum). The large scale, thin filamentary structures surrounding these halo is a direct result of the relatively low tidal field strength due to the surrounding large-scale density field. In contrast, for halos located in strong tidal fields, the surrounding filamentary structure is larger than the halo allowing material to be accreted from different directions producing more gradual accretion (deceleration along t1) parallel to the large scale filament - and acceleration (along t3) - perpendicular to the filaments, allowing material to accrete angular momentum more coherently. We find that star formation in the spheroidal hosts of massive BHs prompts very rapid transformation of gas into stars and with BH feedback heating and somewhat quenching the inner regions which result in an older stellar populations. Disk dominated hosts tend to have even higher star formation rates (less quenched by AGN feedback) and which arise after a gradual ramp up in time as gas accretion is somewhat delayed compared to the spheroids. We plan to explore the observational signatures of these populations of stars in the most massive disks versus the most massive black hole hosts in future work. Our results suggest a new scenario for the origin of the most massive early black holes which does not rely on major gas rich mergers. This scenario offers clues to the origin of the most massive, rare high-z black holes. These extreme black holes can be grown at the highest rates in high density regions in the early Universe but not necessarily the most extreme ones: what is important is that they grow most rapidly in regions with relatively low tidal fields due to their environments. The fact that the brightest quasars at these early times host the most massive black holes, but do not necessarily have to live in the densest regions of the universe, imply that they are unlikely to be the precursors of the massive black holes in cluster of galaxies today. More likely their descendants will not be in privileged sites but in rather isolated galaxies (e.g. Thomas et al. 2016) Our work and the finding that black hole growth is linked to tidal field theory offers some explanation as to the origin of the most massive BHs in most compact galaxies. It is interesting that Barber et al. (2016) using the EAGLE simulation at z = 0 also found a strong link between compactness of the host and the most (positive) outliers in the MBH − M∗ relation. These authors also find that the most extreme outliers grow rapidly in MBH at early times to lie well above the present day MBH − M∗ relation and are consistent with undisturbed morphologies implied by late time observations. In the context of our findings it is likely that indeed the most extreme massive black hole outliers are linked to regions with lower tidal fields in the initial conditions. While it is currently unfeasible (with any current HPC) to run BlueTidesto z = 0, we also plan to carry out a Dark-Matter Only BlueTideswhich will allow us to study the descendants of the the earliest supermassive BHs. The fastest black hole growth is then a direct consequence of the initial conditions of the density field. Although the limited time evolution and somewhat limited numerical resolution of the large volumes of BlueTidesprecludes more detailed studies and investigations we plan to use these clues to re-simulate some of these systems at higher resolution (Feng et al. 2014, as we did with MB simulation, see e.g.) with the goal of studying in more detail the dependence and constrain seed black hole scenarios. We note that (Danovich et al. 2015, and references therein) have studied in detail the angular-momentum buildup in high − z massive galaxies using high-resolution zoom cosmological simulations and have explicitly studied the link with early disc formation and tidal field theory. In our work we illustrate that black hole formation and angular momentum buildup can be understood within the same context.
