Background in the CMS muon detectors: simulation and measurements with pp collision data
PPrepared for submission to JINST iWoRiD 2019, 21 th International Workshop on Radiation Imaging Detectors, 7-12 Jul2019, Kolymbary, Chania, Crete (Greece)
Background in the CMS muon detectors: simulation andmeasurements with pp collision data
Silvia Costantini a on behalf of CMS Collaboration a CERN, CH-1211 Geneva 23, Switzerland
E-mail: [email protected]
Abstract: The CMS muon system at the LHC is built of different detector technologies. Themeasurement of the background hit rates in the different muon detectors during the LHC Run-2is of prime importance for an assessment of the longevity of the chambers and their on-boardelectronics, and therefore for the expected performance of the system at HL-LHC. Moreover,an accurate modelling of the backgrounds using simulations is also critical, as an estimation ofthe expected radiation background for the Phase-2 upgrade can only rely on Montecarlo-basedpredictions in the regions where new detectors are being installed. The present understanding ofthe backgrounds measured with data collected during the LHC Run-2, as well as at CERN high-intensity gamma irradiation facility, GIF++, is discussed, together with the work made to improvethe accuracy of the background modelling in Fluka and GEANT4 simulations.Keywords: Muon spectrometers, Performance of High Energy Physics Detectors, Radiation cal-culations a r X i v : . [ phy s i c s . i n s - d e t ] F e b ontents The High Luminosity LHC (HL-LHC) upgrade [1] is expected to increase the sensitivity to newphysics by increasing the center-of-mass energy for proton-proton (pp) collisions to 14 TeV and theinstantaneous luminosity to 5 · cm − s − (7 . · cm − s − in the ultimate scenario), withan average number of additional pp interactions (pileup) per bunch-crossing expected to reach 140(200 in the ultimate scenario). The integrated luminosity will increase in the coming two decadesfrom the design value of 300 fb − of the present LHC “Phase-1”, to 3000 fb − (or to 4000 fb − ,expected ultimate value) during the so-called LHC “Phase-2”, as shown in figure 1.The detailed description of the CMS detector, together with a definition of the coordinatesystem used, can be found in Ref. [2]. Muons with pseudorapidity in the range | η | < . The estimation of the dose and rate expected in the muon system from radiation backgroundis essential to choose the appropriate detector technologies for the upgrade, and to design thedetectors and electronics that will be able to cope with the HL-LHC conditions. The new detectortechnologies (the GEMs and the so-called “improved RPCs”, referred to as iRPCs in the following)that are to be installed as part of the muon system upgrade, are designed to maintain excellentperformance throughout the whole HL-LHC operation. Both the new and the existing detectors– 1 – igure 1 . Development of the instantaneous and integrated luminosities vs. time for the design HL-LHC.The present “Phase-1” data taking period will end in 2023, followed by a shutdown for the HL-LHC upgrade.The high luminosity data taking period, “Phase-2”, is expected to last from 2026 till 2038.
Figure 2 . Schematic view, in the R-z plane, of one quadrant of the CMS detector, with the axis parallel tothe beam (z) running horizontally and the radius (R) increasing upward. The interaction region is at the lowerleft corner. The position of the present RPC chambers is shown in blue. The RPCs are both in the barrel(“RB” chambers) and in the endcaps (“RE”) of CMS. The DT chambers are labeled “MB”(“muon barrel”)and the CSC chambers are labeled “ME” (“muon endcap”). The steel disks are displayed as dark gray areas.Also shown in orange and red are the chambers of the proposed upgrade scenario of the CMS Muon system,including Gas Electron Multiplier detectors (labeled “ME0” and “GE”) and iRPCs (“RE3/1” and “RE4/1”). (DTs, CSCs, RPCs) are being tested (Section 4) to demonstrate, or confirm, their radiation hardnessat the level required to handle the HL-LHC background conditions.Proton-proton primary interactions and particle transport through the CMS detector are simu-lated using FLUKA [5, 6], which is also used to estimate the expected fluences, doses and fluxesfrom neutrons, photons and charged particles. Hit rates are in turn estimated by normalizing thefluxes by the detector sensitivities determined with GEANT4 [7–9]. Very accurate descriptions ofthe CMS detector and the CMS cavern are implemented in the FLUKA and GEANT4 simulations,well reproducing the Run1 and Run2 measurements as shown in the next Section. As an example,figures 3 and 4 show typical detector sensitivities in the GEM chambers as a function of the particleenergies, together with the resulting expected fluxes in the detector.– 2 –he DTs, CSCs and RPCs are replacing for Phase-2 their front-end electronics in the chambersexposed to the largest background with new electronics tested to cope with the expected neutronfluences and total ionization doses [1].
Figure 3 . GEM sensitivities in the GE1/1 chambers as a function of the particle energies for neutrons,photons, electrons and positrons. The sensitivities are simulated with GEANT4.
Figure 4 . Expected hit rates, at the HL-LHC design luminosity, in the GE1/1 chambers. Rates values aredetermined normalizing fluxes estimated with FLUKA via average sensitivities obtained with GEANT4 andshown for instance in figure 3.
The data collected during the LHC Run1 and Run2 have offered the opportunity to extensively studythe radiation background inside the CMS cavern and in the Muon system. A linear dependenceas a function of the LHC instantaneous luminosity has been observed by all subdetectors overseveral orders of magnitude of the instantaneous luminosity, as shown as an example in figure 5.The highest rates are observed in the external endcap chambers, more exposed to the neutrons– 3 –ermeating the CMS cavern. Assuming that the linear dependence on the luminosity, observedfrom ≈ cm − s − to ≈ cm − s − , will hold at 5 · cm − s − , these measurements canbe extrapolated to estimate the expected background rates at the HL-LHC in the existing detectors.Figure 6 shows the RPC measurements in the barrel and endcap compared to the expected hit ratesin the detectors. Good agreement is observed in the central (barrel) region of the detector, wherethe differences are at the level of 10% to 30%. In the endcap regions, the simulation is able toreproduce the shape of the measurements, with a quantitative agreement within a factor of two. Asimilar agreement, within a factor two, is observed by the CSCs. The RPC average sensitivitiesdetermined with GEANT4 are: neutrons ( . ± . ) %; photons: ( . ± . ) %; e+/e-: ( ± ) %. Figure 5 . Rates measured in the endcaps of the RPCs as a function of the instantaneous luminosity, at acenter-of-mass energy of 13 TeV, for values of the instantaneous luminosities up to ≈ . · cm − s − . The expected higher background conditions at the HL-LHC are realistically simulated at the CERNGamma Irradiation Facility (GIF++) [10], where beams of high energy charged particles, mainly100 GeV muons, are provided with photons from an intense (14 TBq) Cs source. The dosedistribution inside the GIF++ bunker is shown in figure 7 [10]. The energy of the emitted photons,662 keV, is within the typical range (0.1 to 10 MeV) of neutron-induced photons emitted at the LHC.Accelerated irradiation tests are performed, allowing to study the performance and stability, and toasses the longevity of detectors expected to cope for several decades with the HL-LHC conditions,i.e. with doses and rates of the order of five times the LHC ones (as mentioned in Section 1). Effectsof prolonged exposition to radiation may result in loss of gas gain; rise in spurious signal rates (darkrate); increase in leakage currents (dark current); development of self-sustained discharges set offat high radiation rates (Malter effect). Existing and new detectors are therefore being certifiedfor the doses and rates expected at the HL-LHC. Maximum hit rates and charges expected to beaccumulated in the chamber gas volume, relevant for the detector longevity, are summarized intable 1 [1].Chambers of all the muon subdetectors are being irradiated at the GIF++. We refer to [11] fora detailed discussion of the longevity studies performed by the DTs. The CSCs have collected a– 4 – igure 6 . Comparison between measurements and simulation in the barrel and endcap stations of theRPC detector. The stations RB1-RE1 (top left), RB2-RE2 (top right), RB3-RE3 (bottom left) and RB4-RE4(bottom right) are shown, respectively. The simulation is able to reproduce the shape of the data, with aquantitative agreement at the level of 10% to 30% in the barrel, and within a factor 2 in the endcap.
Table 1 . Expected hit rate and accumulated charge for the CMS muon detectors after the Phase-2 upgrades,at the end of HL-LHC running, assuming an instantaneous luminosity of 5 · cm − s − . Only the worst-case values in the most exposed chambers of each subdetector are given. The numbers shown do not includesafety factors. The column RPC refers to the RPC chambers already present during Phase-1, while the newchambers to be installed are referred to as iRPCs. DT CSC RPC iRPC GE1/1 GE2/1 ME0 | η | range 0-1.2 0.9-2.4 0-1.9 1.8-2.4 1.6-2.15 1.6-2.4 2.0-2.8Hit rate (Hz/cm ) 50 4500 200 700 1500 700 48000Charge per wire (mC/cm) 20 110Charge per area (mC/cm ) 280 330 6 3 280total integrated charge of about 330 mC/cm, corresponding to three times the integrated charge perwire expected at the HL-LHC (figure 8) with no evidence of gas gain loss. The average gas gain ofthe most exposed ME2/1 chambers is normalized to the one of the least exposed ME1/3 chambers,used as a reference. The RPCs (figure 9) see no noticeable effects of detector degradation for valuesthe integrated charge up to 600 mC/cm corresponding to roughly 70% of the ones expected atthe HL-LHC. Longevity tests are also performed on large size prototype of Phase-2 RPC detectors,showing stability of the main detector parameters. The GEMs have observed no aging effect with– 5 –otal accumulated charge of 125 mC/cm , corresponding to 10 years of GE1/1 (GE2/1) operationat the HL-LHC, and about 45% of the total ME0 operation. Irradiation tests for ME0 have beenperformed also with an X-ray source, allowing to collect an integrated charge of 875 mC/cm ,corresponding to 10 years of operation (figure 10). Safety factors of at least 3 are included for allsubdetectors. Figure 7 . Dose distribution inside the GIF++ bunker. The labels U (upstream) and D (downstream) referto the two independent irradiation regions of the GIF++ bunker, classified as “upstream” or “downstream”relative to the incoming charged particle beam. The chambers being irradiated are located near the positionslabelled as D4 (DTs), U3 (RPCs), D2 (CSCs), and D1 (GEMs).
Figure 8 . Relative gas gain in the CSC ME2/1 chambers, the most exposed to radiation background,normalized to the least exposed ME1/3 chambers, taken as a reference in the comparison.
Extensive measurements have been performed during the LHC Run1 and Run2 data taking periods,leading to an accurate understanding of the radiation background in the CMS Muon system.The simulation, with FLUKA being mainly used for flux and dose estimation, and GEANT4 forsensitivity studies, well describes the data in the central part of the detector, with differences in the– 6 – igure 9 . RPC hit efficiency at the working point as a function of the γ background rate per unit area, forvalues of the integrated charge up to 600 mC/cm , for both irradiated (IRR) and non-irradiated reference(REF) chambers. The measurements are shown at different values (0, 18 and 31%) of the total integratedcharge expected at the HL-LHC. Figure 10 . Normalized gas gain in the GEMs as a function of the accumulated charge. – 7 –ange of 10% to 30%. The agreement between data and simulation is within a factor two in theendcap. The level of agreement with present measurements, in the regions covered by the existingdetectors, allows to estimate the expected radiation background where new detectors are going tobe installed at the HL-LHC, and it is essential for detector design ad future operation. Longevitytests are being performed at the CERN GIF++ facility. They show no evidence of aging effects atthe total integrated charge collected so far.
References [1] CMS Collaboration,
The Phase-2 Upgrade of the CMS Muon Detectors , CMS-TDR-016 (2017).[2] CMS Collaboration,
The CMS experiment at the CERN LHC , JINST (2008) S08004.[3] F. Sauli, GEM: A new concept for electron amplification in gas detectors , Nucl.Instr. Meth. A (1997) 531.[4] CMS Collaboration,
Performance of the CMS muon detector and muon reconstruction withproton-proton collisions at √ s =
13 TeV , JINST (2008) P06015.[5] A. Ferrari et al., FLUKA: a multi-particle transport code , CERN-2005-10 (2005).[6] G. Battistoni et al.,
Overview of the FLUKA code , Ann. Nucl. Energy (2015), 10.[7] GEANT4 Collaboration, Geant4 - a simulation toolkit , Nucl.Instr. Meth. A (2003) 250.[8] Geant4 Collaboration,
Geant4 developments and applications , IEEE Trans. Nucl. Scien. (2006)270.[9] Geant4 Collaboration, Recent developments in Geant4 , Nucl.Instr. Meth. A (2016) 186.[10] D. Pfeiffer et al.,
The radiation field in the Gamma Irradiation Facility (GIF++) at CERN , Nucl.Instr.Meth. A (2017) 91.[11] G. Abbiendi et al.„
Study of the effects of radiation on the CMS Drift Tubes Muon Detector for theHL-LHC , JINST (2019) C12010.(2019) C12010.