First measurement of the charge asymmetry in boosted top quark pair production in the atlas experiment using the grid-based tier-3 facility at IFIC-Valencia

Resumen

The Large Hadron Collider (LHC) at CERN is the largest collider ever built. It is assembled in the same 27 km tunnel where Large Electron-Positron Collider (LEP) used to be, $\sim$ 100 m underground, near Geneva (Switzerland). The main interaction points have four experiments installed: ATLAS, CMS, ALICE and LHCb. The LHC is designed to collide two proton beams, that circulate in opposite directions, each with an energy of up to 7 TeV, reaching a centre-of-mass energy up to 14 TeV. The ATLAS experiment is a general-purpose detector designed to exploit the potential of the LHC. With a total weight of 7000 tonnes, ATLAS is the heaviest experiment at the LHC. It has 42 m of total length and 11 m of radius. It is hosted under the ground of the Interaction Point 1 of the LHC. Its main components are four: the Inner Detector (ID), the Calorimeters, the Magnet system and the Muon Spectrometer. The first LHC running period is emphasized in this thesis because the analyses presented in this thesis are based on Run-I data. During this period, IFIC was involved in the Spanish Federated Tier-2 (ES-ATLAS-T2) and consequently in the Iberian ATLAS Cloud, providing computing resources to the ATLAS Collaboration. The Iberian ATLAS Cloud has responded very efficiently during the Run-I. The changes in the Computing Model allowed to improve the performance of the cloud in terms of connectivity, storage, replication, transfer, etc. All the sites of the cloud provided the resources of CPU, disk and tape needed to fulfil the ATLAS pledge. The required Distributed Analysis tools were provided to the users in order to use/store data and produce experimental results, i.e., the observation of a new particle in the search for the Standard Model Higgs boson. The LHC underwent a period of maintenance and planned upgrades, the First Long Shutdown, to prepare the collider for a higher energy and luminosity, and all the implications that this entails. During this period, installations connected with the experiments were improved, accelerators were upgraded, electronics and computing evolved, etc. This period served to optimize the experiences learned during the Run-I and improve them during the Run-II. Several of the LS1 upgrades were tested for a long period of time in an exercise called Data Challenge (DC14), which claimed to get ATLAS ready for Run-II physics. During DC14 the Integrated Simulation Framework (ISF) was commissioned in the context of physics analyses, large-scale jobs of the updated reconstruction algorithms and the new distributed computing tools were run, and finally, the Run-II Analysis Model was tested, resulting in a gain of experience with the Run-II analysis framework. The top quark is the most massive of the fundamental particles known so far. It was discovered in 1995 at the Tevatron, a $\sqrt{s}$ = 1.96 TeV $p\bar{p}$ collider built in Fermilab, by the CDF and D0 Collaborations. It is the only quark in the SM that decays before it hadronizes, the only fermion whose mass is close to the electroweak scale, and the only quark with large Yukawa coupling to the Higgs boson. Its properties are of great importance in many of the current BSM theories. Apart from direct searches for new resonances in $t\bar{t}$ production, experiments look for signatures of new physics by performing precise measurements of top quark production and decay. The Tevatron experiments reported a first measurement of the forward-backward asymmetry ($A_{FB}$) in top quark pair production that had considerable tension with respect to the SM prediction. At the LHC a related charge asymmetry ($A_C$) has been measured. Popular BSM models address some of the current open questions, such as the baryon asymmetry of the universe, the nature of dark matter or a mechanism to naturally stabilize the Higgs boson mass at its observed value of approximately 125 GeV. For some of these solutions to work, the new physics scale has to be around the TeV, within reach of collider experiments. The LHC is the first machine to directly probe these energies. In the most energetic collisions heavy particles (such as top quarks, W or Z bosons) are produced with a transverse momentum that greatly exceeds their rest mass. The Lorentz boost of these heavy particles changes the topology of their decay products. The classical method to identify and to reconstruct their hadronic decays is not adequate for these "boosted objects". In order to manage this boosted topology, new techniques have been developed that reconstruct the boosted object as a single jet and distinguish it from the background through a substructure analysis. In order to select and to reconstruct top quark pairs at the LHC, two methods for selecting lepton+jets events are discussed: the classical, resolved approach, developed at the Tevatron for top quark production at rest, and a new technique designed specifically to deal with boosted top quark production. The latter has superior acceptance and resolution for top quark pairs with a mass above 1 TeV. A SM template is constructed using a combination of MC and data-driven techniques. Data has been used to estimate two important backgrounds in the boosted analysis: W+jets and multi-jet production. Finally, a detailed comparison of data and the MC template is presented for the boosted analysis. Apart from an understood discrepancy in the yield, a reasonable agreement is found. A search for production of new heavy particles decaying to $t\bar{t}$ in the lepton+jets decay channel was carried out with the ATLAS experiment at the LHC. The search uses data corresponding to an integrated luminosity of 20.3 fb$^{-1}$ of proton-proton collisions at a centre-of-mass energy of 8 TeV. No excess of events beyond the SM predictions is observed in the $t\bar{t}$ invariant mass spectra. Upper limits on the $\sigma \times BR$ are set for a broad (15.3% width) Randall-Sundrum Kaluza-Klein gluon. Based on these results, the existence of a broad Kaluza-Klein gluon with mass 0.4 < $m_{g_{KK}}$ < 2.2 TeV is excluded at 95% CL, while masses below 2.3 TeV are expected to be excluded. These results probe new physics at higher mass than previous ATLAS searches for the same signature, and the results are applicable to a broader variety of heavy resonances. The charge asymmetry in $t\bar{t}$ production in 8 TeV pp collisions is measured for events with highly boosted top quark pairs. The dataset corresponds to an integrated luminosity of 20.3 fb$^{-1}$ collected by the ATLAS detector in 2012. The observed $|\Delta |y||$ distribution, the difference in absolute rapidities of top and anti-top quarks, is corrected to bring the measurement to the parton-level phase space: $m_{t\bar{t}}$ > 0.75 TeV, $| \Delta |y||$ < 2. A matrix unfolding procedure, which uses the FBU algorithm, corrects for the effect of kinematic migrations in mass and rapidity due to reconstruction. An acceptance correction is applied to remove the selection bias. The $A_C$ in the high mass, central fiducial region is found to be $A_C$ = 4.2% $\pm$ 3.2 (tot.) for 0.75 TeV < $m_{t\bar{t}}$ < 4 TeV and $|\Delta |y||$ < 2. The uncertainty is dominated by the signal modelling uncertainty followed by the statistical uncertainty of the 8 TeV data set. Uncertainties on the detector response play a minor role. Results of a differential $A_C$ measurement in three $m_{t\bar{t}}$ intervals are: $m_{t\bar{t}}$ interval: [0.75-0.9] TeV | [0.9-1.3] TeV | [1.3-4.0] TeV measurement: 2.2 $\pm$ 7.3 % | 8.6 $\pm$ 4.4 % | -2.9% $\pm$ 15.0 % SM prediction: 1.42 $\pm$ 0.04 % | 1.75 $\pm$ 0.05 % | 2.55 $\pm$ 0.18 % Powheg+Pythia: 0.5 $\pm$ 0.2% | 0.8 $\pm$ 0.3% | 0.1 $\pm$ 0.7% MC@NLO+Herwig: 1.1 $\pm$ 0.2% | 0.9 $\pm$ 0.2% | 0.5 $\pm$ 0.6% The bin from 0.9 to 1.3 TeV, where the $A_C$ has been measured with a precision of 4.3%, extends the $t\bar{t}$ mass reach of previous analyses into the TeV regime. The SM prediction is also given, as well the expectations in two MC generators. The largest deviation is observed in the intermediate mass bin, where the difference between the measured and predicted $A_C$ corresponds to 1.7 standard deviations. This measurement agrees with the SM and other measurements at somewhat lower mass. They provide a stringent constraint of the parameter space of several extensions of the SM.