SCT Hybrid Testing and the Production of Direct Photons in the ATLAS experiment at the LHC

Reported in this thesis are the results of production tests of barrel hybrids, photon identification and an analysis of Monte Carlo direct photons. The testing of barrel hybrids assembled at Birmingham is now complete. Hybrids were mounted with chips, bonded and tested to meet the ATLAS acceptance criteria. They have had sensors subsequently attached, been placed on the semiconductor tracker barrels and are preparing to start their operational life. Photon identification has been studied over the Et range 20-450 GeV. Calorimeter identification has been optimised to an efficiency of ~ 90% for single photons, giving a rejection factor against QCD jets increasing with Et from 2600 at 20 GeV to 12700 at > 300 GeV. The addition of an isolation cut inceases this rejection by a factor 2-4 (20-300 GeV) although the high energy region suffers from a lack of statistics. The feasibility of performing a direct photon cross section measurement has been shown. Significant numbers of events are expected over a large range of Et, with the signal to background increasing from unity at 20 GeV to around 30 at 300 GeV. Methods for quantifying the remaining background have been presented and show it should be possible to recover accurately the direct photon signal. Direct photons have been shown to be sensitive to the various parameterisations of the gluon parton density function (PDF). Variations of 4% - 10% in photon eta are visible across the pt and eta ranges accessible by ATLAS. This corresponds to a sensitivity at low pt, to the low-x behaviour of the gluon and at high pt to the high-x behaviour of the gluon. As a result, the measurement of photon eta can be used to help constrain the gluon PDF.