|
Abstract:
|
Formation mechanism of supermassive black holes (SMBHs) observed in the early
Universe is still not fully understood. The goal of this thesis is to nd under what
conditions black hole (BH) remnants of Population III stars can form SMBH with mass
109 M by redshift z = 7. We use Millennium and Millennium-II N-body cosmological
simulations to investigate BH growth on cosmological scales. In order to exploit both
high mass resolution in the Millennium-II simulation and large box size in the Millennium
simulation, we develop a method to combine these two simulations together. BHs can
grow through mergers with other BHs and through episodes of gas accretion triggered by
major mergers of dark matter haloes. As a constraint in our model, we use observed BH
mass function at redshift z = 6. We nd that BH seeds with masses 100 M could grow
to SMBHs in distant quasars if e ective Eddington ratio is xed at fEdd = 3:7 and each
accretion episode is limited to 50 Myr.
During a BH merger asymmetric emission of gravitational radiation can lead to BH
kick. Gravitational wave recoil can completely eject BH from it's host if the kick velocity
is larger than the escape velocity from the galaxy. Since gravitational wave recoil could
a ect SMBH growth through mergers, recoiling BHs are investigated in di erent models
of host galaxies.
BH trajectories are investigated in static and evolving dark matter halo potential described
by NFW and Einasto density distributions. We nd that evolution of dark matter
haloes clearly impact their capability to retain recoiling BHs since escape velocities are
lower for smaller haloes at high redshifts. If the Einasto pro le is considered, then a larger
number of complete BHs ejections is expected compared to NFW potential.
Further, we construct analytical and numerical host galaxy models whose components
are dark matter halo, bulge and disc. If baryonic component of a galaxy is included
escape velocity is higher compared to a purely dark matter halo potential. Major (1:1)
and minor (1:10) galaxy remnants are modeled separately. In numerical models BHs are
ejected from their host centre before galaxy merger is completed, so escape velocities are
generally lower in numerical models compared to analytical models where galaxy potential
is unperturbed.
Even though BHs could occasionally escape the most massive hosts, our model is not
considerably sensitive to the gravitational wave recoil except for mergers of equal mass
BHs in the least massive haloes at high redshifts where kick velocities of Vk . 100 km=s
could permanently eject BHs from their hosts. |