The formation and evolution of galaxies in an expanding universe

This PhD thesis is part of an ongoing effort in improving the theory of galaxy formation in a LCDM Universe. We include more realistic models of radiative cooling, star formation, and stellar feedback. A special attention has been given to the role of supernova explosions and stellar winds in the galaxy assembly. These processes happen at very small scales (parsecs), but they affect the inter-stellar medium (ISM) at Kpc-scales and regulate the formation of a whole galaxy. Previous attempts of mimicking these effects in simulations of galaxy formation use very simplified assumptions. We develop a much more realistic prescription for modeling the feedback, which minimizes any ad hoc sub-grid physics. We start with developing high resolution models of the ISM and formulate the conditions required for its realistic functionality: formation of a multi-phase medium with hot chimneys, super-bubbles, cold molecular phase, and very slow consumption of gas. We find that this can be achieved only by doing what the real Universe does: formation of dense (> 10 H atoms cm -3 ), cold ( T [approximate] 100 K) molecular phase, where star formation happens, and which young stars disrupt. Another important ingredient is the effect of runaway stars: massive binary stars ejected from molecular clouds when one of the companions becomes a supernova. These stars can move to 10-100 parsecs away from molecular clouds before exploding themselves as supernovae. This greatly facilitates the feedback. Once those effects are implemented into cosmological simulations, galaxy formation proceeds more realistically. For example, we do not have the overcooling problem. The angular momentum problem (resulting in a too massive bulge) is also reduced substantially: the rotation curves are nearly flat. The galaxy formation also becomes more violent. Just as often observed in absorption lines studies, there are substantial outflows from forming and active galaxies. At high redshifts we routinely find gas with few hundred km s -1 and occasionally 1000 - 2000 kms - 1 . The gas has high metallicity, which may exceed the solar metallicity. The temperature of the gas in the outflows and in chimneys can be very high: T = 10 7 - 10^8 K. The density profile of dark matter is still consistent with a cuspy profile. The simulations reproduce this picture only if the resolution is very high: better than 50 pc, which is 10 times better than the typical resolution in previous cosmological simulations. Our simulations of galaxy formation reach a resolution of 35 pc.

At the time in which most of the mass is assembled into a galaxy, a big fraction of the gas in the galactic disk has already been converted into stars. Therefore, we can assume that the remaining gas does not affect the evolution of the stellar distribution. In this approximation, all gasdynamical processes are neglected and we treat a galaxy as a pure collisionless system. Then we use N-body-only models to study the long-term evolution of an already formed stellar disk. During this evolution, the disk develops a bar at the center through disk instabilities. We find dynamical resonances between the bar and disk or halo material. These resonances can capture stars near certain resonant orbits. As a result, resonances prevent the evolution of the stars trapped around these orbits.