A visualization of the damage in Lead Tungstate calorimeter crystals after exposure to high-energy hadrons

Abstract

a b s t r a c tThe anticipated performance of calorimeter crystals in the environment expected after the planned High-Luminosity upgrade of the Large Hadron Collider (HL-LHC) at CERN has to be well understood, before informed decisions can be made on the need for detector upgrades. Throughout the years of running at the HL-LHC, the detectors will be exposed to considerable fluences of fast hadrons that have been shown to cause cumulative transparency losses in Lead Tungstate scintillating crystals. In this study, we present direct evidence of the main underlying damage mechanism. Results are shown from a test that yields a direct insight into the nature of the hadron-specific damage in Lead Tungstate calorimeter crystals exposed to 24 GeV/c protons.

Predicting hadron-specific damage from fast hadrons in crystals for calorimetry

“…The effective reduction of light output is macroscopically observed as a loss in light transmission. Indirect evidence for it was the Rayleigh scattering λ −4 behaviour [15] of the damage amplitudes observed in [3], the visualisation of scattering centres by shining laser light through irradiated PWO and LYSO crystals and the fact that the scattered light is polarised [10,16]. All of these effects were not observed in cerium fluoride [11], as expected, since the material consists of light elements that do not undergo fission [17].…”

“…While segments in latent tracks are scattering centres for light, in the process of track etching, which is performed for visualisation purposes, those segments get joined, as explained in section 2 and in [20]. This is evident in the images of etched tracks, as found in figure 2 (left) or in [16] e.g., where no pattern of tracks split into individual segments can be observed. To reach a further validation of the simulations, densities have thus to be compared at the track level.…”

“…From [16], where fission tracks from lead tungstate have been visualised, one obtains an observed track density crossing a surface per proton entering the crystal that amounts to φ = (1.76 ± 0.14) × 10 −6 tracks cm −2 /(p cm −2 ). In the FLUKA simulations for lead tungstate, the irradiation has been performed with a proton fluence of 1 × 10 6 protons on the crystal front face, i.e.…”

“…It might be useful to point out that the track length was inferred from densities, rather than from a direct determination of the simulated track length: this is justified because we want to compare simulation findings with the length of observed etched tracks [16], which tend to be longer than latent ones. It should also be noted that the only tracks directly accessible to simulations are latent ones, and that their length in lead tungstate is found to be on average 2.2 µm, as can be inferred from figures 7 and 8.…”

Response evolution of the CMS ECAL and R&D studies for electromagnetic calorimetry at the high-luminosity LHC

“…The major concern for HL-LHC running is that the energetic hadron fluences from p-p collisions are predicted to cause a specific, permanent, cumulative loss of transmission in Lead Tungstate [2], [3], [17]. Since a loss of transmission causes a reduction and a non-uniformity of light output that may worsen the energy resolution up to a level inadequate for HL-LHC running, the CMS ECAL group has decided to pursue this issue through in-beam studies of hadron-irradiated crystal matrices, the development of Monte Carlo (MC) simulations, to extrapolate the performance evolution through HL-LHC running, and through dedicated studies of crystal damage recovery through light injection and thermal treatment [7], [8].…”