MEASUREMENT OF TOP QUARK PAIR PRODUCTION CROSS SECTION AND CHARGE ASYMMETRY AT THE LHC WITH THE ATLAS EXPERIMENT

In March 2010 the Large Hadron Collider (LHC) at CERN started its operation at a centre of mass energy of 7 TeV. Since then, the ATLAS experiment has been collecting a large number of proton-proton collision events, resulting in an integrated luminosity of about 5.2 fb-1up to now. My PhD research has been focused on top quark physics, which is one of the milestones of the ATLAS experiment physics program. The production of top quarks is the dominant high-pT process in p-p collisions at multi-TeV energies, after QCD jets, W and Z bosons. Furthermore, top quark physics is a rich subject, in fact top quark events are used for detector commissioning and to provide a consistency test of the Standard Model. Finally the top quark sector is considered a good channel for new physics discovery. In some Beyond the Standard Model theories top quark pairs can be produced by the exchange of undiscovered heavy particles. There are also several models that predict the existence of heavy coloured resonances decaying into top quarks in the TeV energy range. In particular I have performed consistency tests of Standard Model theory using two top quark physics measurements. With first data I have focused on the measurement of top quark pair cross section and then I have moved to more stringent tests using the measurement of top quark charge asymmetry. The first part of my PhD activity has been dedicated to the top quark pair production cross section determination with a counting method. The aim of this analysis has been to provide the first measurement of top quark pair production cross section in p-p collisions at √s = 7 TeV in order to compare it with theoretical Standard Model predictions. This result has been published in Autumn 2010, as the best world measurement at that time in √s = 7 TeV proton-proton collisions. This analysis has been performed in the so called “semileptonic channel”. This decay channel, with one W boson decaying leptonically and the other one decaying hadronically, is characterized by the presence of one energetic electron or muon (events with a tau-lepton have not been considered, since they need a dedicated analysis), one neutrino and at least four energetic jets. Two of these jets come from a b-quark and they can be identified using b-tagging techniques. The neutrino doesn't interact in the detector, but its energy can be measured in the transverse plane as missing transverse energy. The most important background processes are QCD mul-tijet events and W+jets events, in which W boson is produced in association with hadronic jets. As a first step, top quark pair candidate events have been selected: an optimization of selection cuts has been done in order to select a signal sample as pure as possible. Another crucial point has been the evaluation of the background contamination. Since LHC collisions energy is ~4 times higher than the one of previous existing colliders, Monte Carlo predictions are characterized by large uncertainties. For what concerns the two main backgrounds, data driven techniques have been therefore designed in order to obtain an estimate directly from data, as independent as possible from Monte Carlo predictions. The systematic uncertainty coming from the selection and the background estimate has been evaluated. The cross section measured with 35 pb-1 in the electron and muon channels combined is (154 +49 -45) pb for a selection which does not make use of b-tagging information and (156 +36 -30) pb after requiring at least one b-tagged jet. Within their uncertainty, the two results are well compatible between each other and with Standard Model prediction. With the increase of available statistics and a better knowledge of the detector performance, the measurement of top quark pair cross section with a counting method became less competitive with respect to ft techniques. In addition the collection of higher statistics permitted to obtain competitive results on the measurement of other top quark properties, which were suffering for higher statistical uncertainty. In the second part of my PhD activity I have therefore performed studies for top quark charge asymmetry measurement. This analysis has been performed in the same channel of cross section measurement. This choice has permitted to take advantage of all the previous studies on signal selection, on background estimates and on systematic uncertainty evaluation. Top quark charge asymmetry can only occur in asymmetric initial states in top quark pair production, so the main contribution comes from q-q production mechanism. It consists in the fact that the top quark is preferably emitted in the direction of the incoming quark and not in the one of the incoming antiquark. This feature originates a difference in top and antitop quark rapidity distributions. The asymmetry foreseen at the LHC according to the Standard Model is small, as will be shown in the following. Some Beyond the Standard Model theories predict, at the opposite, a sizable asymmetry. As a consequence, this measurement can provide a window on new physics. Furthermore the CDF Collaboration measurement, performed at the Tevatron collider at Fermilab, has shown a deviation larger than 3 sigma from the Standard Model prediction in the large top-antitop invariant mass region. This analysis has been performed with 0.70 fb-1. With respect to previous cross section analysis, additional investigations have been done on background contamination, since this measurement is sensible not only to the normalisation, but also to the shape of background processes. Furthermore dedicated studies have been done in order to design algorithms to reconstruct top and antitop quarks from their decay products. Finally some work has been done to identify which observables are more sensitive to new physics and their dependence with respect to top quark pair kinematic variables, since different Beyond the Standard Model theories predict different dependencies and different relations between variables. For the first ATLAS measurement, the observable considered is: AC(|y|) = (N(Delta|y| > 0) - N(Delta|y| < 0))/( N(Delta|y| > 0) + N(Delta|y| < 0)) where Delta|y| is the difference between the absolute values of top and antitop rapidities (|ytop| - |yantitop|), N(Delta|y| > 0) is the number of selected events in which Delta|y| is positive, while N(Delta|y| < 0) is the number of selected events in which Delta|y| is negative. The measured asymmetry is AC = -0.024 ± 0.016 (stat.) ± 0.023 (syst.), combining electron and muon channel. Within present uncertainty the result is in agreement with the Standard Model prediction (from the mc@nlo Monte Carlo generator) of AC = 0:006. Work is ongoing in order to reduce the systematic uncertainty. With higher integrated luminosity, differential asymmetries will also be considered increasing the sensitivity to new physics. The first chapter of this thesis is dedicated to a review of top quark physics. After a brief introduction on the Standard Model theory, important measurements performed both at the Tevatron and at the LHC are described. Then the prospects for top quark physics at the LHC are discussed. The second chapter describes the LHC and the ATLAS detector. In particular the reconstruction algorithms for the objects involved in top quark pair events are presented. In the third chapter top quark pair cross section measurement with first 35 pb-1 is documented, while the last chapter is dedicated to top quark charge asymmetry analysis performed with 0.70 fb-1.