Open Access

Alba Soto Ontoso

• Particle physics is living it’s golden age: petabytes of high precision data are being recorded at experimental facilities such as the Relativistic Heavy Ion Collider (RHIC) and the Large Hadron Collider (LHC). Despite the significant theoretical progress achieved in the last years the complete understanding of the internal structure of protons, not computable with perturbative QCD, remains as one of the most challenging unsolved problems in the physics of elementary interactions. Besides its fundamental interest, pinning down the relevant degrees of freedom and their properties, such as their spatial distribution, has profound implications in several phenomenological aspects of high-energy collisions. Currently one of the subjects undergoing intense study is the possibility that droplets of quark-gluon plasma (QGP) are being created not only in heavy ion collisions but in more dilute systems such as high-multiplicity proton-proton interactions. This is a data driven debate as it is rooted in the similar patterns observed across the different collision systems at the LHC (p+p,p+Pb and Pb+Pb) in the flow harmonic analyses: one of the golden probes of QGP formation specially sensitive to the initial collision geometry. Another domain in which the proton structure plays a central role is the description of multi-parton interactions, the mechanism that dominates the underlying event at LHC energies, in Monte Carlo event generators. All in all a precise characterization of the hadronic structure is a crucial ingredient of the physics program of the LHC. A full characterization of a hadron would require momentum, spatial and spin information, the so-called Wigner distribution. So far obtaining this information experimentally has not been achieved. From a theoretical point of view several complications arise such as non-universality and breaking of factorization theorems. Then, in general, the description of the hadron structure relies on phenomenological tools that require theoretical modeling constrained by experimental data. The main goal of this thesis to characterize the transverse structure of the proton. For that purpose, a wide variety of phenomenological problems that are sensitive to the proton structure have been addressed. First, elastic scattering data on proton-proton interactions constitutes a powerful probe of the geometry of the collision. A dedicated analysis of this observable focusing on the extraction of the inelasticity density from it at √s=62.5 GeV and √s=7 TeV is presented. In the TeV regime, a unexpected phenomenon, dubbed the hollowness effect, arises: the inelasticity density, a measurement of how effective is the collision producing secondary particles, reaches its maximum at non-zero impact parameter. We provide the first dynamical explanation of the hollowness effect by constructing the elastic scattering amplitude in impact parameter representation according to the Glauber model. For that purpose we relied on a composite description of the proton. More concretely, the relevant degrees of freedom that participate in the scattering process were considered to be gluonic hot spots. The probability distribution for the transverse positions of hot spots inside the proton includes repulsive short-range correlations between all pairs of hot spots controlled by an effective repulsive core rc that effectively enlarges the mean transverse separation distance between them. The main results extracted from this work are as follows. To begin with, we found that our model was not able to describe a growing behavior of the inelasticity density at zero impact parameter in the absence of non-trivial spatial correlations. However, even in the presence of correlations, the emergence of the hollowness effect couldn’t be described when the number of hot spots was smaller than 3. Both features set solid constraints in the proton structure within our model. Finally, we pinpoint the transverse diffusion of the hot spots with increasing collision energy to be the dynamical mechanism underlying the onset of the hollowness effect. A convenient playground to test further implications of this novel geometric description of the proton are the initial state properties of high energy proton-proton interactions in the context of QGP physics. The parametrization of the geometry of the collision is mandatory in any theoretical model attempting to describe the striking experimental results that suggest collective behavior in proton-proton interactions at the LHC, such as the non-zero value of the flow harmonic coefficients (vn). A quantitative way to characterize the initial geometry anisotropy of the overlap region is to compute the spatial eccentricity moments (εn) that fluctuate on an event by event basis. For that purpose we develop a Monte Carlo Glauber event generator. A systematic investigation of the effect of non-trivial spatial correlations in the spatial eccentricity moments from ISR to LHC energies within our Monte Carlo Glauber approach is presented. We found that both the eccentricity (ε2) and the triangularity (ε3) are affected by the inclusion of short-range repulsive correlations. In particular, the correlated scenario yielded larger values of ε2(3) in ultra-central collisions while reducing them in minimum bias. Moreover, we explore not only the eccentricities mean but their fluctuations in terms of symmetric cumulants. The experimental measurement by the CMS Collaboration at √s=13 TeV indicates a anti-correlation of v2 and v3 around the same number of tracks in the three collision systems available at the LHC. We lay out, for the first time in the literature, a particular mechanism that permits an anti-correlation of ε2 and ε3 in the highest centrality bins as dictated by data. When modeling the proton as composed by 3 gluonic hot spots, the most common assumption in the literature, we find that the inclusion of spatial correlations is indispensable to reproduce the negative sign. Further, we perform a systematic investigation of the parameter space of the model i.e. radius of the hot spot, radius of the proton, repulsive core and number of hot spots in each proton. Our results suggest that the interplay of the different scales is decisive and confirm the discriminating power of this observable on initial state models. Together with their drastic impact on the description of the hollowness effect and the absolute values of the eccentricities, the symmetric cumulant study adds evidence to the fact that the inclusion of spatial correlations between the sub nucleonic degrees of freedom of the proton modifies the initial state properties of p+p interactions at LHC energies.