2008 JINST 3 S08004
3.5.1 CMS ECAL In Situ Performance
The energies and directions of prompt electrons and photons from pp collisions in CMS are identified and reconstructed starting from the energies deposited in each crystal in the ECAL. These deposits are clustered by algorithms discussed in Section 4.4, and
67
1 General Overview CMS–ECAL TDR
8
Fig. 1.3: Different contributions to the energy resolution of the PbWO4 calorimeter.
Angular and mass resolution
The two-photon mass resolution depends on the energy resolution and the error on the measured angle between the two photons. If the vertex position is known, the angular error is negligible. However, a contribution of about 1.5 GeV to the di-photon mass resolution (at a mass of around 100 GeV) is expected from the uncertainty in the position of the interaction vertex, if the only information available is the r.m.s spread of about 5.3 cm of the interaction vertices. At low luminosity, where the number of superimposed events is small, the longitudinal position of the Higgs production vertex can be localized using high-pT tracks originating from the Higgs event.
Studies indicate that even at high luminosity the correct vertex can be located for a large fraction of events using charged tracks. However, this result depends on the precise knowledge of the minimum-bias pileup at LHC energies. We thus retain the possibility of inserting a barrel preshower device consisting of a lead/silicon layer. The information from the preshower, when combined with that of the crystal calorimeter, could provide the measurement of the photon direction at high luminosity, with an accuracy of about 45 mrad/√E.
1.4.4 Radiation environment
At a luminosity of 1034 cm–2 s–1 about 109 inelastic proton–proton interactions per second will generate a hostile radiation environment.
The simulations of the radiation environment use minimum-bias events obtained from the DPMJET-II event generator. The uncertainty in the estimate of the neutron fluence is about a factor of 2 due to approximations in the geometrical descriptions of the subdetectors, and somewhat smaller for the dose in and around the ECAL. All estimates are presented for an integrated luminosity of 5 × 105 pb–1 assumed to be appropriate for the first ten years of LHC operation.
0.1 1 10
1 10 100 1000
σ/E[%]
Intrinsic All
Noise Photo
E[GeV]
resolution, measured by fitting a Gaussian function to the reconstructed energy distributions, has been parameterized as a function of energy:
σ E
2
=
S
√E 2
+
N
E 2
+C2, (1.2)
whereSis the stochastic term,N the noise andCthe constant term. The values of these parameters are listed in the figure.
E (GeV)
0 50 100 150 200 250
(E)/E (%)σ
0 0.2 0.4 0.6 0.8 1 1.2
1.4 3x3
S=3.63 +/− 0.1%
N=124 MeV 3x3 Hodo Cuts S= 2.83 +/− 0.3%
N=124 MeV C= 0.26 +/− 0.04%
C=0.26 +/− 0.01%
Figure 1.7: ECAL supermodule energy resolution,σE/E, as a function of electron energy as measured from a beam test. The upper series of points correspond to events taken with a 20×20 mm2trigger and reconstructed using a containment correction described in Sec- tion4.3.2.2. The lower series of points correspond to events selected to fall within a 4×4 mm2 region. The energy was measured in an array of 3×3 crystals with electrons impacting the central crystal.
1.5.4 Hadron calorimeter
The design of the hadron calorimeter (HCAL) [3] is strongly influenced by the choice of mag- net parameters since most of the CMS calorimetry is located inside the magnet coil (Fig.CP 1) and surrounds the ECAL system. An important requirement of HCAL is to minimize the non-Gaussian tails in the energy resolution and to provide good containment and her- meticity for theETmissmeasurement. Hence, the HCAL design maximizes material inside the magnet coil in terms of interaction lengths. This is complemented by an additional layer of scintillators, referred to as the hadron outer (HO) detector, lining the outside of the coil.
Brass has been chosen as absorber material as it has a reasonably short interaction length, is easy to machine and is non-magnetic. Maximizing the amount of absorber before the mag- net requires keeping to a minimum the amount of space devoted to the active medium. The tile/fibre technology makes for an ideal choice. It consists of plastic scintillator tiles read out with embedded wavelength-shifting (WLS) fibres. The WLS fibres are spliced to high-
Figure 3.21: CMS ECAL energy resolution as a function of energy. Left: Differ- ent contributions according to the design [79]: the noise term (magenta line labeled
“Noise”), the stochastic term (red line labeled “Photo”), the sum in quadrature of the stochastic and constant terms (brown line labeled “Intrinsic”), and the total res- olution (blue line labeled “All”). Right: Measurement for a 3×3 array of crystals with electrons from a test beam incident in an area of 20×20 mm2 (solid line) and 4×4 mm2 (dashed line) in the center of the front face of an arbitrary crystal. [66,83].
associated with hypotheses as to the nature of the particles that produced them, as discussed in Sections 4.5 and 4.6. The energy deposited in each crystal, together with other properties of the clusters are then used to estimate the energy of the electrons and photons originating at the interaction point. As discussed in more detail in the remainder of this section, the reconstructed electron and photon energy resolution is significantly worse than the resolution measured in test beams.
This is due to several factors, the most important of which is the material of the tracker, its cables and services in front of the ECAL, and the additional material of the preshower in front of the ECAL endcap. Other factors contributing to the resolution in situ include the energy scale determination using Z → e+e− and µµγ events, the crystal-by-crystal calibration accuracy, the electronic noise, and the accuracy of the GEANT4 simulation [84].
As discussed in detail in Section 4.4, the in-situ reconstruction of electrons takes into account electron showering and photon conversions in the tracker material, that lead to a spread of the electromagnetic showers along the φ-direction due to the
bending of electrons and positrons in the magnetic field. As a result of the spreading of energy over a larger angular range, there is a greater likelihood that part of the shower goes into the inter-crystal, inter-module and inter-supermodule cracks, and so more of the energy is lost on average and the fluctuations in these losses contribute to the resolution. In addition, since the energy is spread among a greater number of crystals, the overall contribution to the resolution from electronic noise is increased for showering electrons relative to non-showering electrons.
In addition to the effects of translating the detector energies to the particle energies and the associated changes in clustering, there are further effects that worsen the energy resolution. In the ECAL barrel (|η| < 1.5), the per-crystal electronics noise has risen due to the rise in the dark current in the avalanche photodiodes (APDs) caused by irradiation (see Figures 3.22 and 3.23), and there is an additional small effect due to the radiation-induced loss of transparency of the crystals (see Figure 3.19) that amplifies the noise relative to the signal for a given particle energy.
date (month/year)
07/11 01/12 07/12
A) µ Dark current (
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
2.2
-1 )Integrated Luminosity (fb
0 10 20 30 40 50 60 70 80 90 CMS Preliminary 2011-2012 100
ECAL Barrel
Average dark current per channel (2 APDs)
η=0
=0.45 η η=0.8
=1.15 η
=1.45 η
Integrated Luminosity
Figure 3.22: CMS ECAL APD dark current evolution in time [82].