Energy Linearity and Resolution of the ATLAS Electromagnetic Barrel Calorimeter in an Electron Test-Beam

A module of the ATLAS electromagnetic barrel liquid argon calorimeter was exposed to the CERN electron test-beam at the H8 beam line upgraded for precision momentum measurement. The available energies of the electron beam ranged from 10 to 245 GeV. The electron beam impinged at one point corresponding to a pseudo-rapidity of eta=0.687 and an azimuthal angle of phi=0.28 in the ATLAS coordinate system. A detailed study of several effects biasing the electron energy measurement allowed an energy reconstruction procedure to be developed that ensures a good linearity and a good resolution. Use is made of detailed Monte Carlo simulations based on Geant which describe the longitudinal and transverse shower profiles as well as the energy distributions. For electron energies between 15 GeV and 180 GeV the deviation of the measured incident electron energy over the beam energy is within 0.1%. The systematic uncertainty of the measurement is about 0.1% at low energies and negligible at high energies. The energy resolution is found to be about 10% sqrt(E) for the sampling term and about 0.2% for the local constant term.A module of the ATLAS electromagnetic barrel liquid argon calorimeter was exposed to the CERN electron test-beam at the H8 beam line upgraded for precision momentum measurement. The available energies of the electron beam ranged from 10 to 245 GeV. The electron beam impinged at one point corresponding to a pseudo-rapidity of eta=0.687 and an azimuthal angle of phi=0.28 in the ATLAS coordinate system. A detailed study of several effects biasing the electron energy measurement allowed an energy reconstruction procedure to be developed that ensures a good linearity and a good resolution. Use is made of detailed Monte Carlo simulations based on Geant which describe the longitudinal and transverse shower profiles as well as the energy distributions. For electron energies between 15 GeV and 180 GeV the deviation of the measured incident electron energy over the beam energy is within 0.1%. The systematic uncertainty of the measurement is about 0.1% at low energies and negligible at high energies. The energy resolution is found to be about 10% sqrt(E) for the sampling term and about 0.2% for the local constant term