[en] The global properties of galaxies show a strong evolution of the star formation rate and stellar mass density at the epoch of galaxy assembly, driven by several competing physical processes (merging, accretion, feedback, environment,...). The morphological properties of galaxies are also strongly evolving over the same timescales. I investigate how the evolution of the morphological properties is connected to the spectrophotometric properties of galaxies since z∼6. The spectroscopic data obtained within the VIMOS Ultra Deep Survey (VUDS), a new unique spectroscopic survey of ∼10000 galaxies between redshifts z ∼ 2 and z ∼ 6 conducted at the ESO-VLT, combined with the available Hubble Space Telescope imaging surveys such as COSMOS or CANDELS provide a great opportunity of probing galactic evolution across this cosmic epoch. This thesis is broadly divided in 4 topics tackling galaxy morphology evolution. A first topic is related to the evolution in galaxy sizes at high redshift. I have developed a new algorithm that measures galaxy sizes in a non-parametric way by computing the total area covered by the galaxy above a fixed luminosity isophote. I compare my new measurements with those standardly used in the literature and find that in terms of typical evolution they differ substantially. I find that galaxies have the same typical size across 2 < z < 4.5 contrary to a moderate evolution found for the same population in the same period when using a standard size measurement. I attribute this difference to the often irregular shape of galaxies at these epochs. Importantly, I find that the dispersion of sizes is very large, with, at the same redshift, compact galaxies that are basically unresolved with HST, and very large galaxies with total sizes over 5 kpc. I discuss the implications of these findings and conclude that the wide range of galaxy sizes must be the result of several different paths for galaxy assembly. The second topic provides context for the search of galactic clumps at high redshift and the importance it has for galaxy assembly studies. Based on the algorithm used to compute the total extent of galaxies, I developed a new code to detect clumps within a given isophote. I find that the fraction of clumpy galaxies is high at 2 < z < 6 (∼ 35 - 55%) and increases at higher redshifts. An interesting finding is that the predominant class of clumpy galaxies has two clumps, and these are often found to be the brightest and largest of all clumps. I argue that these results point to a scenario where we are witnessing merging of galaxies in its final stages, and that it complements the merger fraction derived from pair finding studies. The faint and small clumps are more akin to the clumps produced in the simulation of discs, where violent disc instabilities produce several clumps migrating to form a central proto-bulge. Then I move to a brief summary on our understanding of the evolution of the Hubble sequence across cosmic time and an extension of a visual classification scheme to a high-redshift sample of star-forming galaxies. I find a prevalence of irregular galaxies at all redshifts 2 < z < 6 that amount to around ∼ 40-50% of the star-forming sample. With respect to their global physical properties I find a diverse scenario where different morphological types span a wide range in stellar masses, star formation rates, dust content and ages. I argue that star-forming galaxies at these redshifts show diverse paths of galaxy formation and that no single morphological type corresponds univocally to a single path for formation. Finally I provide some context about the physical mechanisms that are thought to power the emission of Ly-α photons in high redshift galaxies to introduce a study on the typical conditions for Ly-α emission in high redshift galaxies. With the help of several members of the VUDS collaboration I have produced a catalog of visually classified Ly-α emitters based on the shape of the line on the 2D spectra. I identify a class of spatially extended line emitters (with respect to the UV continuum emission) that is the typical object corresponding to Ly-α emitters (LAEs) due to its large equivalent width. These objects are found to be smaller, less dusty, less massive and less star-forming than other classes of line emitters and also objects with no emission at all. I argue that this class of objects is partially the explanation for the differences found between LAEs and normal star-forming galaxies that are found in the literature. From the results that I have obtained, I conclude that the different morphological properties of star-forming galaxies at 2 < z < 6 show that there are several mechanisms that take part in galaxy formation. The great diversity in galaxy sizes, the high number of bright two clump systems as well as faint multi-clump systems indicating merging and disk instabilities processes, respectively, and the wide range in physical properties that are populated by irregular galaxies all point to a diverse history for galaxy formation. This suggests that forming galaxies is far more complex than a scenario involving cold gas accretion alone. I present observational evidence in favor of several different path for galaxy formation and assembly, for which the contribution of key physical processes like merging, accretion and disc instabilities, as well as stellar and AGN feedback, are very different. (author)

[en] Nowadays, efforts are being devoted to the study of alternative cosmological scenarios, in which, modifications of the General Relativity theory have been proposed to explain the late cosmic acceleration without assuming the existence of the dark energy component. In this scenario, we investigate the R${}^2$-corrected Appleby-Battye model, or R${}^2$-AB model, which consists of an f(R) model with only one extra free parameter b, besides the cosmological parameters of the flat-ΛCDM model: H${}_0$ and Ω${}_{m,0}$. Regarding this model, it was already shown that a positive value for b is required for the model to be consistent with Solar System tests, moreover, the condition for the existence of a de Sitter state requires b ≥ 1.6. To impose observational constraints on the R${}^2$-AB model we consider three datasets: 31 H(z) measurements from Cosmic Chronometers (CC), 20 [fσ${}_8$](z) measurements from Redshift-Space Distortion (RSD), and the most recent type Ia Supernovae (SNe Ia) sample from Pantheon+. Next, we perform two diferent analyses: we have cosidered only SNe Ia data and the combined likelihood SNe + CC + RSD. The first one has provided b=2.28${}_{-<!--\; ???\; -->0.55}^{+6.52}$, while the second one b=2.18${}_{-<!--\; ???\; -->0.55}^{+5.41}$. In the first case it was necessary to set the absolute magnitude M${}_B$ = -19.253 from SH0ES collaboration, while in the second we did a marginalization over the Hubble constant H${}_0$ in the normalized growth function. We have also observed that the H${}_0$-M${}_B$ degenerecency was broken by adding CC data to the SNe data. Additionally, we perform illustrative analyses that compare this f(R) model with the flat-ΛCDM model, considering several values of the parameter b, for diverse cosmological functions like the Hubble function H(z), the equation of state w${}_{eff}$(z), the parametrized growth rate of cosmic structures [fσ${}_8$](z), and σ${}_8$(z). From our results, we conclude that the R${}^2$-AB model fits well current observational data, although the model parameter b was not unambiguously constrained in the analyses.

No abstract available

No abstract available

No abstract available

No abstract available

No abstract available

No abstract available

[en] For this work, a hydroalcoholic extract was prepared from the aerial parts of the Malvaviscus arboreus (Malvaceae) species, popularly known as hibisco-colibri in Brazil; and then, investigated by infrared and NMR spectroscopy. A class of special metabolites containing a porphyrin skeleton and pyrrole units was determined by typical absorption spectra, mainly ¹H and ¹³C NMR, thus indicating the capability of showing anti-corrosion effect. Therefore, the M. arboreus extract was examined as a corrosion inhibitor for AISI 1020 carbon steel in H₂SO₄ 1 mol L^-1, using a mass loss (gravimetric) technique. Results showed good corrosion inhibition efficiency, at both time and temperature variation. The maximum efficiency observed at 500 ppm was 97.5%, at 3 h, 24 h and 48 h. The surface morphology was analyzed by scanning electron microscopy (SEM), through which a higher homogeneity was observed in the presence of the M. arboreus extract. The hydroalcoholic extract of hibisco-colibri can be considered an efficient green agent for preventing corrosion. (author)

[en] The diseases that affect the thalamus are heterogeneous in their etiologies, including infectious, inflammatory, vascular, toxic metabolic, and neoplastic causes. It is often difficult to make the clinical differentiation between different entities. Within this context, computed tomography and magnetic resonance imaging have come to be of fundamental importance for defining the etiology and planning the treatment. In this pictorial essay, we will illustrate the main causes of diseases affecting the thalamus, discussing the possible differential diagnoses, as well as the most relevant imaging aspects. (author)