Di-photon cross section measurement and sensitivity study for Higgs search in the H → γγ channel with the ATLAS detector

This work is done with the ATLAS collaboration. Three independent methods are proposed to measure the photon trigger efficiency. They are first evaluated using Monte Carlo simulation and then applied on 2010 data. The two photon-based methods show consistent results, with efficiency discrepancy at a few % level. For the method based on electron sample, the statistics is too low to draw conclusion. A detailed QCD di-photon cross-section measurement is performed on a data sample corresponding to a luminosity of (37.2 ± 1.2) pb-1, in which a 2D fit method is introduced to extract the signal yields. The differential distributions of the observables Mγγ, pTγγ and Δφγγ are derived and compared with the predictions from the DIPHOX and RESBOS generators. A good agreement is found for the Mγγ distribution, whereas discrepancies are observed in the pTγγ and Δφγγ distributions. In the study of the H → γγ channel based on a simulated sample, a deterioration of 4% in the exclusion limit is observed after incorporating the systematic uncertainty arising from the resolution of the di-photon invariant mass. In the real data analysis, the 2D fit method is also applied to decompose the backgrounds in the di-photon candidate invariant mass range [100, 150] GeV. The sensitivity of the Higgs boson search in the two photon final state is then reappraised. 3.2-4.2 times the standard model predicted cross-section in the 110-140 GeV mass range is expected to be excluded with 1 fb-1 of data at √(s)=7 TeV. (author)