Misura della produzione di non-prompt Lambda_c in collisioni protone-protone a sqrt(s)=13 TeV in ALICE

The Quark Gluon Plasma (QGP) is a state of strongly-interacting matter in which quarks and gluons are deconfined. This state of matter can be created in ultrarelativistic heavy-ions collisions, such as those studied at the CERN Large Hadron Collider (LHC). The QGP is very short living, so it cannot be detected directly in the experiments, but it is possible to study its properties from the final-state particles produced in the collisions. The ALICE detector was specifically designed to study heavy-ion collisions at the LHC. Heavy-flavour (HF) quarks (charm or beauty) are recognised as ideal probes of the QGP because, due to their large mass, they are produced in hard scattering processes on time scales shorter than the QGP formation and they experience the full evolution of the system. The study of HF is also important in proton-proton (pp) collisions, as their production is a sensitive observable to test perturbative Quantum Chromodynamics (pQCD) calculations. In this context, the production of HF hadrons is typically described as a convolution of the parton distribution functions of the colliding protons, the partonic cross section, and the Fragmentation Functions (FFs). The latter describe the hadronisation of the heavy quarks in the different hadron species, and, since this process is non-perturbative, it is usually parametrised from measurements in e+e- collisions. However, recent studies by the ALICE and LHCb collaborations show that the ratio between the production of HF baryons with respect to mesons is significantly higher in hadronic collisions compared to e+e- interactions, invalidating the assumption that the FFs are independent of the collision system. The goal of this thesis is to study the production of Lambda_c^+ (Lc) baryons in pp collisions at sqrt(s)=13 TeV, separating the prompt (originating from the hadronisation of a charm quark) and the non-prompt (from beauty-hadron decays) contributions. The measurement is carried out at midrapidity via the decay channel Lc -> p + K0s. To select the Lc candidates, two pion tracks are combined to create a K0s candidate, which, due to its long lifetime, is substantially displaced from the interaction point. The trajectory of the K0s is prolonged towards the interaction point and combined with a proton track, resulting in a candidate Lc. The full reconstruction of the decay-vertex topologies is done with a Kalman-filter based algorithm. To separate the candidates belonging to prompt, non-prompt and background, the different decay-vertex topologies were exploited. In fact, the mean proper decay length of Lc baryons is 60.7 um, while that of beauty hadrons is about 440 um. For the separation of the three classes, a Boosted Decision Trees (BDT) was used. The computation of the non-prompt fraction was done using several selection criteria based on the BDT output, along with a fit-like minimisation of the chi^2, which leads to the best estimation of the yield of Lc from beauty-hadron decays. The production cross section for the non-prompt Lc baryons was measured to be about (10 +/- 2) ub for pT>2GeV/c, in agreement with the pQCD calculations. An estimation of the systematic uncertainties is also included. The results obtained in this thesis provide further elements for the understanding of the charm and beauty quarks production and their fragmentation into hadrons, and give a useful benchmark for future studies in pp and lead-lead collisions during the Run 3.