CMS DT Longevity Studies

The CMS drift tubes (DT) muon detector, built for standing up the LHC expected integrated and instantaneous luminosities, will be used also in the High Luminosity LHC (HL-LHC) at a 5 times larger instantaneous luminosity and, consequently, much higher levels of radiation, reaching ~10 times the LHC integrated luminosity. Initial irradiation tests of a spare DT chamber at the CERN gamma irradiation facility (GIF++), at large (~O(100)) acceleration factor, showed aging effects resulting in a degradation of the DT cell performance [1]; however, full CMS simulations have shown almost no impact in the muon reconstruction efficiency over the full barrel acceptance and for the full integrated luminosity. A second spare DT chamber was moved inside the GIF++ bunker in October 2017. The chamber is being irradiated at lower acceleration factors, and only 2 out of the 12 layers of the chamber are switched at working voltage when the radioactive source is active, being the other layers in standby. In this way the other layers remain not aged and are used as reference and as a precise and unbiased telescope of muon tracks for the efficiency computation of the aged layers of the chamber, when set at working voltage for measurements. An integrated dose equivalent to two times the expected integrated luminosity of the HL-LHC run has been absorbed by this second spare DT chamber and the final impact on the muon reconstruction efficiency is under study. Direct inspection of some extracted aged anode wires presented a resistive deposition of materials. Investigation on the outgassing of cell materials and of the gas components used at the GIF++ are under way. Strategies to mitigate the aging effects are also being developed. From the long irradiation measurements of the MB2 spare DT chamber, the effects of radiation in the performance of the CMS DTs expected during the HL-LHC will be presented.

References: [1] CMS-TDR-016, The Phase-2 Upgrade of the CMS Muon Detectors (https://cds.cern.ch/record/2283189)

Hit efficiency definition

The efficiency to detect a single hit in a cell of a layer was defined and measured as the ratio between the number of detected and expected hits. The position of expected hits was determined using as probes sets of well reconstructed track segments with associated hits in at least 4 layers in SL3 and at least 1 layer in SL1 but the layer under test. The intersection of this track segment with the layer under study determined the position, therefore the cell, where a hit was expected: the cell was considered efficient if a hit was found within it.

Ilustration of a DT chamber and a reconstructed track segments with 4 associated hits in SL3 extrapolated to the SL1:

Trigger used for the data taking

During the test beam, high momentum muons (~50 GeV/c) of the SPS H4 beam line were recorded using GIF++ external scintillators as trigger; the beam spot usually illuminated few cells in both projections.

When no beam, cosmic muons were recorded using the DT auto trigger requiring a track in both projections on just SL2 and SL3 of the DT chamber; with such a trigger no bias was introduced on the SL1 layers under study. The recorded cosmic muon tracks illuminated almost uniformly the chamber.

Dose definition and extrapolation

At GIF++, the dose rate is measured with a REMUS dosimeter situated inside the GIF++ bunker. Several dose rate measurements were done with a portable dosimeter in different positions of the MB2 chamber to extrapolate the REMUS measurement to the dose rate on the surface of the MB2 chamber.

The dose rate and integrated dose are converted to expected instantaneous and integrated luminosities for the MB1 chambers on the external wheels (YB±2) expected at the HL-LHC from extrapolation of Run2 measurements. Conversion factors were calculated comparing currents in the MB2 chamber at GIF++ for a given dose rate in not aged layers with currents of DT chambers at P5 for a given instantaneous luminosity.

Since the gain at the high dose rate used to irradiate at GIF++ is lower than at HL-LHC, the dose rate to instantaneous luminosity conversion was calculated at HL-LHC background rate: 10³⁴ cm^-2s^-1 = 0.0109 mGy/h, while the integrated dose to integrated luminosity was calculated at GIF++ aging rate (~10xHL-LHC): 1 fb^-1 = 0.42 mGy.

List of Plots

Figure	Caption
	[1] Integrated dose [mGy] as a function of time for the irradiation period of the CMS DT spare chamber ‘MB2’ at GIF++. The layers 1 and 4 from superlayer 1 were irradiated with HV on (3550 V). The rest of the chamber was irradiated with HV in standby (1900 V) and used as reference. The vertical grey lines correspond to the 2017 and 2018 winter breaks when the chamber is kept off and the vertical orange lines mark the two irradiation and data-taking periods (eras). During summer 2018, 8 wires were extracted for later inspection and replaced with new non-aged wires. The axis on the right shows the equivalent expected luminosity for MB1 chambers in the external wheels for the HL-LHC.
	[2] Hit efficiency for cosmic muons as a function of integrated luminosity for the aged layers (SL1L1 and SL1L4) at 3550 V. Each of the points in this plot corresponds to the data-taking during HV scans (with the source off) that were collected about every week. The DT auto trigger was used for cosmic muon tracks identifying a track in both projections on SL2 and SL3 of the DT chamber, avoiding any bias on the layers irradiated with HV on. The integrated luminosity used in these studies corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC. No atmospheric condition corrections have been applied.
	[3] Hit efficiency for cosmic muons as a function of HV for the aged layers (SL1L1 and SL1L4) when the source is off and the front-end threshold 30 mV. The DT auto trigger was used for cosmic muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber, avoiding any bias on the layers irradiated with HV on. There are three series of datasets each corresponding to a different integrated luminosity: beginning, middle and end of the irradiation period of MB2. The integrated luminosity used in these studies corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC.
	[4] Hit efficiency for cosmic muons as a function of HV for the aged layers (SL1L1 and SL1L4) when the source is off. It shows the comparison of the hit efficiency for the front-end threshold values of 20 mV and 30 mV at the end of the irradiation period. The DT auto trigger was used for cosmic muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber, avoiding any bias on the layers irradiated with HV on. The integrated luminosity, 5737 fb-1, corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC.
	[5] Hit efficiency for cosmic muons as a function of instantaneous dose for the aged layers (SL1L1 and SL1L4) at 3550 V and for SL1L3 (reference) at 3600 V at a front-end threshold of 20 mV. The DT auto trigger was used for cosmic muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber, avoiding any bias on the layers irradiated with HV on. There are three series of datasets each corresponding to a different integrated luminosity. The integrated luminosity used in these studies corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC.
	[6] Hit efficiency for test beam muons as a function of instantaneous dose for the aged layers (SL1L1 and SL1L4) at 3550 V and for SL1L3 (non-aged) at 3600 V at a front-end threshold of 20 mV. The external trigger scintillators were used for muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber in coincidence with the scintillators, avoiding any bias on the layers irradiated with HV on. The test beam was performed after an irradiation integrated dose equivalent to an HL-LHC integrated luminosity of 3600 fb-1.
	[7] Hit efficiency for test beam muons as a function of instantaneous dose for the aged layers (SL1L1 and SL1L4) at 3550 V at a front-end threshold of 20 mV. The external trigger scintillators were used for muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber in coincidence with the scintillators, avoiding any bias on the layers irradiated with HV on. The test beam was performed after an irradiation integrated dose equivalent to an HL-LHC integrated luminosity of 3600 fb-1. An exponential is used to fit the hit efficiency of SL1L1 and SL1L4 layers resulting in: * for SL1L1 * for SL1L4.
	[8] Normalized Current for aged (SL1L1 and SL1L4) and non-aged (SL1L3) layers at 3550 V. The Normalized Current is defined as the ratio between the instantaneous current and the instantaneous dose. Values are scaled with respect to the initial value measured in SL1L3. Currents are corrected for the measured pressure in the bunker. Each of the points in this plot corresponds to the mean current of reading along a day. Gaps are caused by the loss of monitoring. The integrated luminosity used in these studies corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC.
	[9] Hit efficiency for cosmic muons as a function of integrated luminosity for the aged layer SL1L1 taken with the source on at a background rate slightly higher than the expected background rate at the HL-LHC (corresponding to a instantaneous luminosity of 5.8x1034 cm-2s-1) at 3550 V and at a front-end threshold of 20 mV. The data are fitted using an exponential model, . The DT auto trigger was used for cosmic muon tracks, identifying a track in both projections on SL2 and SL3 of the DT chamber, avoiding any bias on the layers irradiated with HV on. The integrated luminosity corresponds to the integrated luminosity expected for the MB1 chambers in the external wheels of the CMS detector (YB+/-2) during the HL-LHC.
	[10] Expected hit efficiencies at the end of the HL-LHC for all the DT chambers of the CMS muon system, MB4 chambers (upper) and for MB1, MB2 and MB3 (lower). These efficiencies have been estimated considering a safety factor of 2 for the expected HL-LHC background rate (10x1034 cm-2s-1) and a safety factor of 2 for the expected integrated luminosity (6000 fb-1) to obtain the expected hit efficiency for the MB1 chambers in wheels ±2 and extrapolating to the rest of the CMS muon system using the expected integrated charge at the end of HL-LHC.
	[11] Muon reconstruction efficiency at CMS as a function of absolute η for non aged DT chambers and aged DT chambers with the efficiencies shown in Fig. 10. The efficiencies have been calculated using simulated sample of dimuon events.

-- BarbaraAlvarezGonzalez - 2019-02-25