Cargando…
Mesure in situ de l’uniformité du calorimètre électromagnétique et recherche des premiers événements di-photons dans ATLAS.
The search for a light Higgs boson in the H --> gamma gamma channel with the ATLAS experiment demands excellent performance of the electromagnetic calorimeter, as well as a good understanding of the backgrounds. This work is organized in two main points. One point is about the operation and perfo...
Autor principal: | |
---|---|
Lenguaje: | fre |
Publicado: |
Laboratoire d'Annecy-le-Vieux de Physique des Particules
2010
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/1307098 |
Sumario: | The search for a light Higgs boson in the H --> gamma gamma channel with the ATLAS experiment demands excellent performance of the electromagnetic calorimeter, as well as a good understanding of the backgrounds. This work is organized in two main points. One point is about the operation and performance of the electromagnetic calorimeter. First, developments are brought to the Liquid Argon calorimeters online reconstruction code, and its algorithms are validated, which includes the development of an online monitoring. Then, two independent measurements of the electromagnetic calorimeter uniformity of response are proposed : one with the signal deposited by cosmic muons on an effective coverage of about 20 %, and one with pi0 --> gamma gamma events in the first collisions at sqrt{s} = 7 TeV with an integrated luminosity of L = 414.8 ub-1. The measurements with cosmic muons shows that the non-uniformities in the eta direction are found to not exceed 1.0 % in the middle layer and 1.7 % in the first layer, at the 95 % Confidence Level. With the pi0 --> gamma gamma events, these results are improved and lead to a dispersion of the response between data and simulation of less than 0.7 % in the three cryostats (barrel and endcaps), which is the level expected at the start-up of the experiment. The second main point of this thesis is the extraction of a non-resonant di-photon signal, for which two methods are presented. The first method is a two-step purity measurement using an extrapolation of the background from background-enriched control regions, into a signal region. The second method uses a 4 x 4 efficiency matrix fully determined from simulation and is applied on an event-by-event basis. These methods are applied to collision data corresponding to L = 2.82 pb-1. A direct di-photon signal yield of 108.6 +- 19.5 (stat.) +- 34.5 (syst.) is extracted, with a purity of (65.1 +- 8.6 (stat.) +- 8.7 (syst.)) %. |
---|