In this thesis, I present my work on the characterization of quasar outflows which I carried out using the statistics and the ionization properties of Narrow Absorption Lines (NALs). The study is based on a new sample of intermediate resolution spectra of 100 high-redshift quasars (z_em = 3.5 − 4.5), obtained with X-shooter at the European Southern Observatory Very Large Telescope, in the context of the XQ-100 Legacy Survey (Lopez et al., 2016). The combination of high signal-to-noise ratio (S/N), wide wavelength coverage and moderate spectral resolution of this survey have allowed me to look for empirical signatures to distinguish between two classes of absorbers: intrinsic (produced in gas that is physically associated with the quasar) and intervening, without taking into account any a priori definition or velocity cut-off. Previous studies have shown that NALs tend to cluster near the quasar emission redshift, at z_abs = z_em. I detect a significant excess of absorbers over what is expected from randomly distributed intervening structures. This excess does not show a dependence on the quasar bolometric luminosity and it is not due to the redshift evolution of NALs. Most interestingly, it extends far beyond the standard 5000 km/s cut-off traditionally defined for associated absorption lines. I take advantage of the large spectral coverage of the XQ-100 spectra to study the relative numbers of NALs in different transitions, indicative of the ionization structure of the absorbers and their locations relative to the continuum source. Among the ions examined in this work, NV is the ion that best traces the effects of the quasar ionization field, offering an excellent statistical tool to identify intrinsic systems and derive the fraction of quasar driving outflows. I also test the robustness of the use of NV as additional criterium to select intrinsic NALs, performing a stack analysis of the Lya forest of the XQ-100 sample, to search for NV signal at large velocity offsets. Lastly, I compare the properties of the material along the quasar line of sight, derived from my sample, with results based on close quasar pairs investigating the transverse direction. I find a deficiency of cool gas (traced by CII) along the line of sight connected to the quasar host galaxy, in contrast with what is observed in the transverse direction in agreement with the predictions of the AGN unification models.
