Search for the Higgs boson at ATLAS/LHC, in associated production with a Z boson

A mechanism of spontaneous symmetry breaking was used to explain the mass of elementary particles and predicted the existence of the Higgs boson. The Higgs boson was discovered in 2012 by the ATLAS and CMS experiments at the LHC with a mass of about 125 GeV. It now becomes necessary to study this new boson in order to validate the Standard Model of elementary particles. The Standard Model Higgs boson with a mass of 125 GeV decays most of the times to a pair of b-quarks. However, this decay is very difficult to study in a proton-proton collider like the LHC, due to the production of a huge background of b-jets (and also non-b-jets). In the LHC, the only production process with some chance to be used in this study is the associated production with a vector boson, which can decay leptonically allowing the identification of the event. One can use three possibilities: a Z boson decaying to neutrinos (0-lepton channel), a W boson decaying to an electron or muon and a neutrino (1-lepton channel) or a Z boson decaying to a pair of electrons or muons (2-lepton channel). The latter channel is the subject of this thesis. It is an interesting channel because the two charged leptons reduce the multijet background more than the other channels and because this channel reconstructs all the particles in the event. The analysis of the 2-lepton channel, known as ZH analysis, considers only events with high quality reconstructed particles. The two charged leptons are required to reconstruct the Z boson and the b-quarks, which are reconstructed as jets, are required to be identified as b-jets (b-tagging) for the reconstruction of the Higgs boson. A statistical procedure uses the events selected as input to a likelihood fit to obtain normalisations of the simulated backgrounds from data and simulation comparisons and to extract the signal strength parameter, $\mu$, that multiplies the cross section times the branching ratio from the prediction for the Higgs boson from the Standard Model to adjust the ZH signal simulation process to the data. Both muons and electrons, used in the ZH analysis, are much better reconstructed than jets. So, most of the effort in this thesis was dedicated to the improvement of the jet reconstruction, in particular to the energy scale and uncertainty of the jets. I studied two methods to help reduce the systematic uncertainty of the jet energy scale, the leading experimental uncertainty of several physics analysis, with a important contribution to the ZH search, since the Higgs decays to b-quarks that hadronize into jets. First, I created a new trigger with the objective of selecting high-momenta isolated charged hadrons to be used in the determination of the calorimeter response to single hadrons, known as the single hadron response. In addition to extending the momentum range from 10 to 30~GeV, it did so in a high-pileup environment (pileup - number of interactions per proton bunch crossing). This contributed to the validation of the test beam determination of the jet energy scale uncertainty. For jets with $p_\mathrm{T}$ above 1.8~TeV, for which the current in-situ techniques are not able to set the jet energy scale uncertainty, the single hadron response is used to obtain this uncertainty. I have tested a new jet calibration in the ZH analysis: the global sequential calibration (GSC). This method was first developed to reduce the jet energy response dependence on the jet flavour, for gluon and light quark jets. My contribution was the validation and detailed performance studies for b-jets in general, and in particular for b-jets coming from the Higgs boson decay. This new calibration improved the jet energy response by about 2-3% and the jet energy resolution by about 20 %. Systematic uncertainties were evaluated for this calibration. I derived specific b-jet systematic uncertainties for this calibration as well as the uncertainties associated to differences between the ATLAS fast and full simulations. The Standard Model Higgs boson decaying in $b\bar b$ is expected to produce a very small excess of events in the invariant mass distribution of the two b-quarks. We need to have the best possible resolution on this variable and since GSC was found to reduce the b-jet jet energy response resolution, it was applied to the $m_{b \bar b}$ distribution. GSC improves the invariant mass resolution up to 18 %. I performed the ZH analysis using a cut-based method and a statistical procedure to obtain the value of the signal strength parameter. A value of $\mu = -0.69 \pm 1.29$ was obtained. This value is negative, but there is an enormous uncertainty, dominated by the statistical uncertainty, and the result is still compatible with the Standard Model prediction for the Higgs boson.