Dust in hydrodynamic and semi-analytic galaxy evolution simulations

Dust grains are fundamental components of the galaxies of our Universe, affecting both the physical processes of galaxy evolution and their observability. Therefore, incorporating dust physics into galaxy evolution simulations is essential. This Thesis investigates various aspects of dust within galaxy evolution using a state-of-the-art dust model integrated into two commonly adopted tools: hydrodynamic simulations and semi-analytic models (SAMs). The dust model includes both stellar production of dust and the growth/destruction of grains in the gaseous phase, allowing to track the dust content and, in an approximate manner, the size and chemical composition of grains. This model, originally developed in previous hydrodynamic simulations, was integrated into a SAM as part of the work done in this Thesis. The dust model in cosmological hydrodynamic simulations is exploited to study the size and chemical properties of grains across different galaxies, alongside with some other basic predictions concerning the simulated galaxy population. The key result is that grains features depend on the evolutionary history of galaxies. Massive galaxies tend to have a population of silicate-rich and smaller grains compared to less massive galaxies. This difference is attributed to the larger relevance of grain growth in the interstellar medium (ISM) of massive galaxies, driven by the large amount of dense gas, compared to dust production by stars. The SAM equipped with the dust model is used to explore various topics, including the evolution of cosmic dust abundance across redshift. Observations suggest a decrease in galactic dust abundance from z~1 to z=0. Predictions from the model presented here are in keeping with this trend, suggesting that the decrease may not be due to intrinsic dust-related processes but rather the adopted galaxy evolution framework. Specifically, the growth of supermassive black holes (SMBHs) during disc instabilities (DIs) is critical for reproducing the observed drop within the context of this model. The SAM-based model is employed also to investigate the star formation rate (SFR) and dust content of local galaxies within different regions of the cosmic web. An analysis using SDSS and GAMA data shows that galaxies in less dense environments (such as voids and walls) tend to be more star-forming and dust-rich up to a certain mass, beyond which environmental differences reduce. The model is in agreement with these observations, attributing the reduced environmental impact in massive galaxies to the in situ growth of SMBHs driven by DIs, which makes galaxy quenching less sensitive to the environment where galaxies reside. Finally, the galaxy catalog produced by the SAM is post-processed with a Radiative Transfer (RT) code to predict observable galaxy properties taking into account dust absorption and emission. The analysis reported here focuses on the sub-mm emission of galaxies in the local Green Valley (GV), i.e. the transition region between the blue cloud and red sequence of the color-mass diagram. In keeping with GAMA and H-ATLAS observations, the model predicts significant sub-mm emission in GV galaxies due to their substantial dust content, which declines more slowly compared to the SFR, resulting in their optical green color. Conversely, rejuvenating systems in the GV exhibit low sub-mm emission due to insufficient time for significant dust growth after a quenched period.