The metal enrichment of the intergalactic medium

The Big Bang Nucleosyhtesis produced a cosmic gas composed by Hydrogen and Helium, with virtually no trace of heavier elements (metals). Observations show that the di use and pervasive cosmic gas surrounding galaxies { the intergalactic medium (IGM) { is ubiquitously enriched with metals, that can only be produced by stars inside galaxies. In turn, the formation and evolution of galaxies is determined by the complex interplay between the interstellar medium and IGM, as mediated by the circumgalactic medium (CGM). Understanding how metal travelled to the IGM, when and from which galaxies are among the current challenge for cosmological models. The pre-enrichment scenario is a theoretical framework that auto-consistently accounts for the evolution of galaxy formation and the cosmic enrichment history. In this picture, massive (M? > 1011M ) galaxies are able to retain their metals, while low mass (M? < 108M ) galaxies are prone to material ejection because of their shallower potential well. At high redshift (z > 7), metal pollution is due to low mass galaxies ejecting metals via supernovae. Additionally, the same low mass galaxies driving the enrichment can play an important role in the rst stages of cosmic reionization, since they are the ideal hosts for the rst metal-free Population III stars, which may in turn be the responsible for an early reionization. Understanding cosmic metal enrichment is therefore fundamental to explain both the formation and the evolution of galaxies and the reionization history. Aim of this work is to study the evolution of metal enrichment on galactic and intergalactic scales. Our analysis is based on cosmological simulations. Because of the huge dynamical range of the underlying physical phenomena, a true auto-consistent simulation is impossible, and a viable modelization can be achieved via subgrid models. We devise simulations by accounting for the trade o between implementing highly re ned physical models and considering a cosmic volume large enough to allow a fair comparison with observations. Among our key results, we show that galaxies develop a mass-metallicity (M?-Z) relation by z = 6. Our M?-Z relation agrees with z = 4 observations. The presence of cold enriched gas in the IGM/CGM implies that these metals must have been injected at epochs early enough to allow an e cient cooling as expected in a pre-enrichment scenario. Due to the physical conditions of the di use phases, C IV absorption line experiments can probe only ' 2% of the total carbon present in the IGM/CGM. By analyzing each galactic environment, we nd that the CGM density pro les are selfsimilar, once scaled with the virial radius of the parent dark matter halo. We then construct an analytical model for the H I absorption, testing it against synthetic spectra. When compared with available data, our CGM model nicely predicts the observed pro le in z < 2 galaxies, and supports the idea that the CGM pro le does not evolve with redshift. We produce [C II] emission mock maps that can be directly compared with ALMA observations. At z ' 6 we nd that the [C II] galaxy ux is correlated with MUV, and such relation is in very good agreement with recent observations. In our mock maps we nd that C II ions in the CGM/IGM can be seen as ' 0:1 Jy/beam [C II] emission/absorptions features via CMB resonant scattering. Such signals are very challenging to be detected with current facilities. Finally, we analyse the case of CR7 { the brightest z = 6:6 Ly emitter (LAE) known to date { that might represent the rst case of Pop III detection. If CR7 is powered by Pop III, we predict that in the COSMOS/UDS/SA22 elds, 14 out of the 30 LAEs with L > 1043:3erg s􀀀1 should also host Pop III stars producing an observable LHeII > 1042:7erg s􀀀1. As an alternate explanation, we explore the possibility that CR7 is instead powered by accretion onto a Direct Collapse Black Hole. Our model predicts L , LHeII, and X-ray luminosities that are in agreement with the observations. We propose that further X-ray observations can distinguish between the two above scenarios.