Formation of disc galaxies from cosmological simulations: galactic outflows and chemical evolution

In this Thesis, I study the formation of late-type galaxies and the role that feedback from stars and supermassive black holes (SMBHs) plays in galaxy evolution across cosmic time. By carrying out cosmological hydrodynamical simulations, I investigate how different processes, such as the cosmological gas accretion from the large scale environment, star formation and chemical enrichment, stellar and AGN (Active Galactic Nucleus) feedback, affect the early stages of forming galaxies and contribute to determine their present-day properties. Driven by the challenging task of simulating late-type galaxies with a limited bulge and a dominant disc in a cosmological context, I study the impact of galactic outflow modelling on the formation and evolution of a disc galaxy. I find that galactic outflows regulate the timing of gas accretion and determine the star formation history of the forming galaxy. Also, I quantify the strong interplay between the adopted hydrodynamical scheme and the sub-resolution model describing star formation and stellar feedback. Throughout this Thesis, I devote particular emphasis to connect chemical evolution and gas dynamics, in order to interpret observations of metal abundance in the interstellar medium (ISM) and circumgalactic medium (CGM). I investigate the metal content of gas and stars, and explore how the variation of the essential elements contributing to define the model of chemical evolution determine final metal abundance trends. The investigation that I present is guided by observations and focusses on the results at redshift z = 0, with particular emphasis on the role played by the high-mass end of the IMF. Moreover, I introduce a novel methodology to generate synthetic stars from star particles. The technique that I developed takes properties of star particles from simulations as input and allows to obtain a catalogue of mock stars, provided with photometric properties, whose characteristics are drawn from the input features. The ultimate goal of this method is to translate the populations of star particles of a simulation into stellar populations, thereby enabling a direct and accurate comparison with observations. This technique will be of paramount importance with ongoing survey data releases (e.g. GAIA and surveys of resolved stellar populations). Also, I investigate the role of AGN feedback in regulating the formation and evolution of late-type galaxies. I introduce a new model aimed at investigating the interaction between the central SMBH and the various gas phases which coexist in the ISM of the host galaxy