Scintillator ageing of the T2K near detectors from 2010 to 2021

2022 JINST 17 P10028 Published by IOP Publishing for Sissa Medialab Received:August 1, 2022 Accepted:September 4, 2022 Published:October 17, 2022 Scintillator ageing of the T2K near detectors from 2010 to 2021 The T2K Collaboration E-mail: [email protected] Abstract: The T2K experiment widely uses plastic scintillator as a target for neutrino interactions and an active medium for the measurement of charged particles produced in neutrino interactions at its near detector complex. Over 10 years of operation the measured light yield recorded by the scintillator based subsystems has been observed to degrade by 0.9–2.2% per year. Extrapolation of the degradation rate through to 2040 indicates the recorded light yield should remain above the lower threshold used by the current reconstruction algorithms for all subsystems. This will allow the near detectors to continue contributing to important physics measurements during the T2K-II and Hyper-Kamiokande eras. Additionally, work to disentangle the degradation of the plastic scintillator and wavelength shifting fibres shows that the reduction in light yield can be attributed to the ageing of the plastic scintillator. The long component of the attenuation length of the wavelength shifting fibres was observed to degrade by 1.3–5.4% per year, while the short component of the attenuation length did not show any conclusive degradation. Keywords: Gamma detectors (scintillators, CZT, HPGe, HgI etc); Neutrino detectors; Performance of High Energy Physics Detectors; Scintillators, scintillation and light emission processes (solid, gas and liquid scintillators) ArXiv ePrint:2207.12982 c 2022 CERN. Published by IOP Publishing Ltd on behalf of Sissa Medialab. Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. https://doi.org/10.1088/1748-0221/17/10/P10028 2022 JINST 17 P10028 Contents 1 Introduction 2 1.1 The T2K Experiment 2 1.2 Scintillator ageing 2 2 The T2K scintillator detectors 3 2.1 ND280 3 2.1.1 PØD 4 2.1.2 FGD 5 2.1.3 ECal 7 2.1.4 SMRD 7 2.2 INGRID 8 3 Light yield measurements 9 3.1 Data samples 10 3.2 ND280 10 3.2.1 PØD 11 3.2.2 FGD 12 3.2.3 ECal 12 3.2.4 SMRD 13 3.3 INGRID 13 3.4 Light yield stability uncertainty 14 3.4.1 ND280 14 3.4.2 INGRID 15 4 Annual light yield reduction 15 5 Projected future response 20 6 Separation of ECal scintillator and fibre degradation 22 6.1 Scintillator degradation 22 6.2 Fibre degradation 25 7 Conclusions 27 The T2K collaboration 31 –1– 2022 JINST 17 P10028 1 Introduction 1.1 The T2K Experiment T2K (Tokai-to-Kamioka) is a long-baseline neutrino oscillation experiment [ 1 ] located in Japan, measuring muon (anti-)neutrino disappearance and electron (anti-)neutrino appearance from a muon (anti-)neutrino beam produced by the J-PARC (Japan Proton Accelerator Research Complex) synchrotron [ 2 ]. The experiment consists of a far detector at a distance of 295 km from J-PARC, a near detector complex 280 m downstream of the proton beam target, and the beam facility itself. The far detector is Super-Kamiokande [ 3 ], a 50 kt water Cherenkov detector positioned 2 . 5 ◦ off the beam axis. The near detector complex contains the ND280 [ 4 ] and INGRID [ 5 ] detectors which started operation in 2010. In addition, WAGASCI-BabyMIND [ 6 , 7 ] was installed in 2019. INGRID is located directly on the beam axis, while the ND280 is situated at the same off-axis angle as Super-Kamiokande. ND280 measures the rate of neutrino interactions before oscillation has occurred. This provides information on the neutrino flux, cross section and neutrino type which is necessary to predict the interaction rate at the far detector. INGRID monitors the neutrino beam direction and profile as well as the neutrino interaction event rate with high statistics. Most subsystems of the ND280, along with the INGRID detector, use plastic scintillator bars as an active detector medium. Whilst traversing the detector, charged particles excite electrons within the scintillator material to higher orbitals. The de-excitation of the electrons produces the emission of scintillation light which is used to track the passage of these particles. The scintillation light is collected by 1 mm diameter Kuraray wavelength-shifting (WLS) fibres [ 8 ] for transmission to Hamamatsu Multi-Pixel Photon Counters (MPPC) [ 9 ], a type of Silicon Photon Multiplier (SiPM) located at one or both ends of the scintillator bars. T2K was the first experiment to employ MPPCs on a large scale, utilising ∼ 65 , 000 MPPCs across the near detectors. The observed MPPC failure rate has been very low, at around ∼ 0 . 5% of the total over the current lifetime of the experiment, and so their failure is not currently a concern for the future operation of the T2K near detectors. However, during 10 years of operation some degradation in the light yield produced by the scintillator bars has occurred. Similar degradation has also been observed in other experiments, such as MINOS [ 10 ] and MINER 𝜈 A [ 11 ]. Understanding this effect is important for the accurate calibration of the detectors, for monitoring their long-term efficiency and predicting the future performance. 1.2 Scintillator ageing The issue of plastic scintillator ageing is long known [ 12 ], and there are many studies aimed at measuring, characterizing and developing stabilisation methods for these widely used materials (see for example [ 13 – 19 ]). These studies often consider the impact of potentially controllable environmental factors such as temperature and humidity on the long-term performance of the materials, as well as ways to chemically stabilise them. The exact mechanism for scintillator ageing occurring within the T2K near detectors is unknown, but there are a number of potentially contributing factors:1 1 The low levels of radiation produced at the near detectors by the T2K neutrino beam, with only a few beam neutrino interactions per spill, means that radiation damage is expected to be negligible. –2– 2022 JINST 17 P10028 • Mechanical stressing of the scintillator causing the development of crazes or shears within the material [ 20 ]. These inhibit the uniform scattering of light within the scintillator, preventing transmission through total internal reflection. • Fogging of the scintillators due to water penetrating into the material and condensing [ 21 ]. This increases the opacity of the scintillator and is a significant problem where the materials are exposed to very high humidity conditions with large temperature variations. • Oxidation of the scintillator through photochemical processes that lead to the production of peroxides causing the yellowing of the material [ 22 ]. This reduces the light yield from the scintillator and has been observed in the accelerated ageing test performed on the scintillator bars used by the MINOS experiment [ 10 ], which are materially identical to the INGRID, FGD, ECal and PØD subsystems of T2K as described in section 2.1. Within this paper the relevant T2K near detector subsystems are described in section 2and the data samples and light yield measurement methods used are detailed in section 3. The rate of degradation of the T2K scintillator is presented in section 4, along with predictions for the future response of the detectors in section 5and an attempt to disentangle whether the ageing is dominated by the degradation of the scintillator or wavelength shifting (WLS) fibres in section 6. 2 The T2K scintillator detectors 2.1 ND280 The ND280 detector, figure 1, is composed of a set of subsystems enclosed within the refurbished UA1 magnet [23]. The subsystems are as follows: • A detector composed of scintillator, water and brass target planes designed to identify 𝜋0 s (PØD) [24]. • The tracker region, consisting of three time projection chambers (TPCs) [ 25 ] and two plastic scintillator fine-grained detectors (FGDs) [ 26 ], optimised to study charged current interactions of incoming neutrinos. The upstream FGD1 is entirely composed of scintillator planes, the downstream FGD2 consists of alternating modules of scintillator and water-filled volumes. • Plastic scintillator and lead sampling electromagnetic calorimeters (ECals) that surround the PØD and tracker region [27]. • Plastic scintillator side muon range detectors (SMRDs) [ 28 ] situated in the magnet flux return yokes. The coordinate system has 𝑧 along the neutrino beam direction, and 𝑥 and 𝑦 are horizontal and vertical, respectively. The material used in the PØD, FGD and ECal scintillator bars is polystyrene Dow Styron 663 W, doped with 1% PPO (2,5-diphenyloxazole) and 0.03% POPOP (1,4-bis(5-phenyloxazol-2-yl) benzene) and co-extruded with a surface layer of polystyrene loaded with 15% TiO 2 to allow diffuse reflection of scintillation light. The bars for the PØD and ECal were manufactured in the extrusion –3– 2022 JINST 17 P10028 beam y z x Figure 1. Exploded diagram of the ND280 off-axis near detector displaying the different detector subsystems. Table 1. Scintillator production dates for each ND280 detector. INGRID is included in the table for completeness. Detector Production Period PØD 2007–2008 ECal 2007–2009 FGD 2006 SMRD 2007–2008 INGRID 2007–2008 facility at Fermi National Accelerator Laboratory (FNAL) and their composition is identical to that of the scintillator bars used in the MINOS experiment [ 10 ]. For the FGD, scintillator bars of the same composition were produced by extrusion procedure at Celco Plastics Ltd, Surrey, British Columbia. The scintillator bars of the SMRD use polystyrene doped with 1.5% PTP (1,4-Diphenylbenzene) and 0.01% POPOP and were chemically etched to produce a reflective coating. These were manufactured by the Uniplast company in Vladimir, Russia. All the ND280 scintillator bars were produced between 2006 and 2009 as shown in table 1. All subsystems use Kuraray Y-11 blue to green WLS fibres for photon transmission to Hamamatsu (S10362-13-050C) MPPCs [ 29 ]. The specific WLS formulations used are Y-11(175) S-35 J-type (PØD), Y-11(200) S-35 J-type (FGD), Y-11(200) CS-35 J-type (ECal) and Y-11(150) S-70 S-type (SMRD). The particular geometry of each subsystem is described below. 2.1.1 PØD The PØD detector consists of 40 scintillator modules called PØDules, as shown in figure 2. Each PØDule consists of two orthogonally oriented bar layers sandwiched between an inactive target and –4– 2022 JINST 17 P10028 sheets (4.5 mm thick). The water target is formed from two units, the upstream and central water target Super-PØDules. The upstream (central) water target Super-PØDule comprises 13 PØDules alternating with 13 (12) water bag layers (each of which is 28 mm thick), and 13 (12) brass sheets (1.28 mm thick), as shown in Fig. 6. The dimensions of the entire PØD active target are 2103 mm ! 2239 mm !2400 mm (width !height !length) and the mass of the detector with and without water is 15,800 kg and 12,900 kg respectively. The PØD is housed inside a detector basket structure that supports the central off-axis detectors inside the magnet. The remainder of this paper describes in detail the design, fabrication, and performance of the PØD. The production of the scintillator bars and their assembly into planks and PØDules will be presented followed by a description of how the individual PØDules were combined into the four Super-PØDules, and are read out using photosensors. The detector component performance, starting with scans of the PØDules using a radioactive source, dark noise measurements, and tests with the light injection system, is presented. The paper concludes with a description of the calibration and performance of the full detector. 2. Design and construction of the PØDule The PØDule is the basic structural element of the PØD active region, and is constructed of scintillator bars sandwiched between sheets of high-density polyethelene (HDPE, thickness 6.4 mm). The entire structure is surrounded by PVC frames that support the PØDule as well as providing mounts for the required services such as the MPPC light sensors, and the light injection system. The polystyrene triangular scintillating bars that make up the PØDules were fabricated by co-extruding polystyrene with a reflective layer of TiO 2 and a central hole for the WLS fiber. The light seal for the tracking plane is maintained by light manifolds that collect the WLS fibers into optical connectors. These manifolds also provide access to the fibers for the light injection system. Because of the large number of scintillating bars and the available space limitations, it was impractical to route the fibers outside the magnetic volume therefore the Hamamatsu MPPC photosensors, which are immune to the magnetic field, were attached directly to each WLS fiber just outside the PVC PØDule frame, as shown in Fig. 5. 2.1. Design of the PØDule The PØDule was designed to both provide the active tracking region and to serve as a structural element. This was achieved using a laminated structure of crossed scintillator bars between polystyrene skins. The final PØDule has been shown to have a rigidity similar to a solid mass of polystyrene of similar thickness. The edge of the central scintillator and skin structure of the PØDule is surrounded by a machined PVC frame. Each PØDule is instrumented on one side (both yand xlayers) with MPPCs and on the other a UV LED light injection system. The bottom PVC frame supports the weight of the PØDule within the ND280 detector basket. The frames also provide the fixed points needed to assemble the PØDule into the four Super-PØDules via two precision holes located in the four corners of each PØDule as well as a set of seven holes spaced along each side through which tensioning rods were passed. The PØDules, after installation into the finished PØD, are oriented such that the most upstream layer of scintillator has the bars oriented approximately along the vertical axis while the downstream layer has its bars oriented along the horizontal axis. This arrangement results in a local coordinate system defined such that the x,yand zaxes are approximately congruent with the global coordinate system where xis horizontal, yis vertical, and zpoints downstream toward Super-Kamiokande. The external dimensions of the PØDule are 2212 mm (x) by 2348 mm (y)by 38.75 mm (z). Fig. 5. A close-up view of the edge of a PØDule showing how the WLS fibers exit the scintillator bars and couple to the MPPCs. The optical connectors will be described on more detail in Section 2.2.5. Fig. 6. Expanded view of water target PØDule, brass radiator and water bladder containment frame. Upstream ECal Upstream Water Target Central WaterTarget Central ECal Legend Lead Light-tight Cover Brass Water Scintillator Wavelength-shifting Fiber Fig. 4. A schematic of the four PØD Super-PØDules as installed in the detector. Beam direction: left to right. S. Assylbekov et al. / Nuclear Instruments and Methods in Physics Research A 686 (2012) 48–6350 Figure 2. A schematic of the four PØD Super-PØDules as installed in the detector. The neutrino beam enters from the left hand side of the figure. radiator material. The PØDules are perpendicular to the beam direction and are assembled into four constituent units called Super-PØDules, these are defined in the following way: 1. Super-PØDule 0: PØDules 0–6, the Upstream ECal. 2. Super-PØDule 1: PØDules 7–19, the Upstream Water Target. 3. Super-PØDule 2: PØDules 20–32, the Central Water Target. 4. Super-PØDule 3: PØDules 33–39, the Central ECal. The scintillator bars in the PØD are triangular in cross section with a height of 17 mm and a width of 33 mm. Each bar has a single 2.6 mm diameter coaxial hole through which a WLS fibre is inserted, as shown in figure 3(a). The horizontal bars are 2133 mm long and the vertical bars are 2272 mm long. The fibres are not secured within the bar, leaving an air gap between the bar and WLS fibre. The WLS fibres are mirrored with a vacuum deposition of aluminium on one end and are optically coupled to an MPPC on the other. A ferrule is glued to the end of the fibre which couples to a housing holding the MPPC, see figure 3(b). A 3 mm thick polyethylene disk behind the MPPC provides pressure between the fibre and MPPC epoxy window. This design of fibre to MPPC coupling is also used by the ECal (see section 2.1.3). 2.1.2 FGD The FGD scintillator bars are perpendicular to the beam in either the horizontal (X) or vertical (Y) direction, and are arranged into “XY” modules. Each module consists of a layer of 192 bars in the –5– 2022 JINST 17 P10028 (a) End view of a PØD scintillator bar. sheets (4.5 mm thick). The water target is formed from two units, the upstream and central water target Super-PØDules. The upstream (central) water target Super-PØDule comprises 13 PØDules alternating with 13 (12) water bag layers (each of which is 28 mm thick), and 13 (12) brass sheets (1.28 mm thick), as shown in Fig. 6. The dimensions of the entire PØD active target are 2103 mm ! 2239 mm !2400 mm (width !height !length) and the mass of the detector with and without water is 15,800 kg and 12,900 kg respectively. The PØD is housed inside a detector basket structure that supports the central off-axis detectors inside the magnet. The remainder of this paper describes in detail the design, fabrication, and performance of the PØD. The production of the scintillator bars and their assembly into planks and PØDules will be presented followed by a description of how the individual PØDules were combined into the four Super-PØDules, and are read out using photosensors. The detector component performance, starting with scans of the PØDules using a radioactive source, dark noise measurements, and tests with the light injection system, is presented. The paper concludes with a description of the calibration and performance of the full detector. 2. Design and construction of the PØDule The PØDule is the basic structural element of the PØD active region, and is constructed of scintillator bars sandwiched between sheets of high-density polyethelene (HDPE, thickness 6.4 mm). The entire structure is surrounded by PVC frames that support the PØDule as well as providing mounts for the required services such as the MPPC light sensors, and the light injection system. The polystyrene triangular scintillating bars that make up the PØDules were fabricated by co-extruding polystyrene with a reflective layer of TiO 2 and a central hole for the WLS fiber. The light seal for the tracking plane is maintained by light manifolds that collect the WLS fibers into optical connectors. These manifolds also provide access to the fibers for the light injection system. Because of the large number of scintillating bars and the available space limitations, it was impractical to route the fibers outside the magnetic volume therefore the Hamamatsu MPPC photosensors, which are immune to the magnetic field, were attached directly to each WLS fiber just outside the PVC PØDule frame, as shown in Fig. 5. 2.1. Design of the PØDule The PØDule was designed to both provide the active tracking region and to serve as a structural element. This was achieved using a laminated structure of crossed scintillator bars between polystyrene skins. The final PØDule has been shown to have a rigidity similar to a solid mass of polystyrene of similar thickness. The edge of the central scintillator and skin structure of the PØDule is surrounded by a machined PVC frame. Each PØDule is instrumented on one side (both yand xlayers) with MPPCs and on the other a UV LED light injection system. The bottom PVC frame supports the weight of the PØDule within the ND280 detector basket. The frames also provide the fixed points needed to assemble the PØDule into the four Super-PØDules via two precision holes located in the four corners of each PØDule as well as a set of seven holes spaced along each side through which tensioning rods were passed. The PØDules, after installation into the finished PØD, are oriented such that the most upstream layer of scintillator has the bars oriented approximately along the vertical axis while the downstream layer has its bars oriented along the horizontal axis. This arrangement results in a local coordinate system defined such that the x,yand zaxes are approximately congruent with the global coordinate system where xis horizontal, yis vertical, and zpoints downstream toward Super-Kamiokande. The external dimensions of the PØDule are 2212 mm (x) by 2348 mm (y)by 38.75 mm (z). Fig. 5. A close-up view of the edge of a PØDule showing how the WLS fibers exit the scintillator bars and couple to the MPPCs. The optical connectors will be described on more detail in Section 2.2.5. Fig. 6. Expanded view of water target PØDule, brass radiator and water bladder containment frame. Upstream ECal Upstream Water Target Central WaterTarget Central ECal Legend Lead Light-tight Cover Brass Water Scintillator Wavelength-shifting Fiber Fig. 4. A schematic of the four PØD Super-PØDules as installed in the detector. Beam direction: left to right. S. Assylbekov et al. / Nuclear Instruments and Methods in Physics Research A 686 (2012) 48–6350 (b) A close-up view of the edge of a PØDule showing how the WLS fibers exit the scintillator bars and couple to the MPPCs. Figure 3. PØD bar image (a) and MPPC connection schematic (b). was placed into a holding jig so that a photo of the polished end could be taken. Software to find the TiO2boundaries and hole boundaries was used to measure the critical dimensions. Fig. 3 shows a photo of the bar profile with the measured dimensions for a typical bar. Figure 3: Photo taken with a CCD camera of a typical scintillator bar produced at Celco Plastics. A MATLAB R  edge-finding routine is used to find the edges of the TiO2 to measure the TiO2thickness on all 4 sides, and the edge of the center hole to measure the hole diameter. •After every shift the bars produced during the shift were taken back to TRIUMF and 4- 8 bars out of every 100 were scanned on the bar scanner (see section 2.5) to make sure the light yield was acceptable. 2.4. Bar width results This section summarizes the results of the dial gauge measurements on the bars as they came o↵ the production run. There were four dial gauges in total: two to measure the width of the bar and two to measure the height. The resolution of the dial gauges was 0.01 mm. The distributions are shown in Fig. 4. The RMS of the deviations ranges from 0.013 mm for the height to 0.019 mm for the width. Figure 4: Measured distributions of the mean widths and heights of the bars. Details of the measurements are given in the text. 2.5. Bar scanner measurements In order to look for dead spots (areas of the bar which scintillate less than the rest of the bar) and to compare light yields among the bars as they came o↵the production line, the bars were scanned by moving a 106Ru beta source along the length of a bar. The light output from a WLS fiber threaded through the hole in the bar was measured with an unbiased photodiode whose current was read out by a Keithley picoammeter. When looking for dead spots, measurements were made every 2 cm; when comparing light yields between bars they were made every 50 cm. A special table and computer controlled movable source holder were constructed so that these measurements could be carried out reproducibly. The same fiber was used throughout. The results of a typical scan are shown in Fig. 5 together with an exponential fit to the data, which yields a normalized light yield, I0. During production, about 1 in every 4 bars was scanned and the distribution of light yields was found to be Gaussian with a width of 4.5 %. 6 (a) Photo taken with a CCD camera of a typical FGD scintillator bar. (b) Partial view of the end of a FGD scintillator layer with alternating fibres connected to MPPCs. Figure 4. An image of a FGD scintillator bar (a) and a FGD scintillator layer (b). horizontal direction glued to 192 bars in the vertical direction. The scintillator bars have a square cross section with a side width of 9.6 mm. The length of the bars is 1864 mm and each has a 1.8 mm diameter hole through its centre containing the WLS fibre, see figure 4(a). One end of the fibre is mirrored with a vacuum deposition of aluminium to improve light collection efficiency, the other end is connected to an MPPC. The upstream FGD1 contains fifteen such modules while the downstream FGD2 contains seven modules interspersed with inactive water target layers. Each FGD module has dimensions of 1864 × 1864 × 20 . 2mm 3 (not including electronics). There is an air gap between the scintillator and the fibre. The fibre extends only a few centimetres from one end of the bar to reach an MPPC as shown in figure 4(b). The fibre is connected to the MPPC with a custom two part connector, one part glued to the fibre and the other holding the MPPC, latched together by mechanical force. Bicron BC600 glue was chosen to fix the coupler to the fiber. Within each layer, alternate bars are read out from alternating ends. –6– 2022 JINST 17 P10028 (a) End view of an ECal scintillator bar and WLS fibre. (b) Exploded diagram of WLS fibre to MPPC connector. Figure 5. ECal scintillator bar with WLS fibre (a) and diagram of the WLS fibre to MPPC connector assembly (b). 2.1.3 ECal The ECal scintillator bars have a cross section of 40 × 10 mm 2 with a 2 mm diameter hole down the centre through which a WLS fibre passes, see figure 5(a). There is an air gap between the scintillator and the fibre. The ECal is comprised of thirteen modules each of which uses one or two different lengths of scintillator bar in their construction. The Downstream (DS) ECal module has 1700 bars of length 2000 mm oriented in alternating vertical and horizontal layers perpendicular to the beam direction. Across the six Barrel ECal modules there are 3990 bars of length 3840 mm (Z bars) lying parallel to the beam direction. The four top and bottom barrel modules contain 6144 bars of length 1520 mm (X bars), and the two side barrel modules contain 3072 bars of length 2280 mm (Y bars). The Barrel X and Y bars are oriented perpendicular to the beam direction and have fibres which are mirrored on one end with a vacuum deposition of aluminium, while the other end is connected to an MPPC as shown in figure 5(b). The Barrel Z and Downstream ECal bars have fibres which are connected to MPPCs on both ends. As will be described in section 3, the analysis methods used within this paper require the 3D reconstruction of muon tracks. As a result the six PØD ECal modules (not described) are not used in the studies presented as all the scintillator bars run parallel to the beam direction making the 3D reconstruction of particle tracks, needed for ageing studies, challenging. It should be noted that the Downstream ECal was installed into the ND280 in early-2010, while the Barrel ECal modules were installed in late-2010. 2.1.4 SMRD There are two types of SMRD scintillator bars with different sizes, horizontal (7 × 167 × 875 mm 3 ) and vertical (7 × 175 × 875 mm 3 ), grouped in modules of 4 and 5 respectively, see figure 6(a). There are 404 modules in total with 768 (1240) horizontal (vertical) bars. The modules are placed in layers in the air gaps of the magnetic flux return yokes. The magnet yokes are numbered 1-8 going downstream along the beam direction. All yokes host three horizontal layers and yokes 1 through 5 also host three vertical layers. The most downstream yokes host more vertical layers: yoke 6 hosts four, and yokes 7 and 8 host six. Figure 6(a) shows the placement of the first layer of modules in a –7– 2022 JINST 17 P10028 (a) Schematic view of a layer of SMRD modules. izontal modules and are 2.2m long for all vertical modules. 3.3. Photosensors Multi Pixel Photon Counters (MPPC) [17] developed for T2K by Hamamatsu were chosen as the common photosensor for all ND280 scintillator based detectors. Key features of these devices are their insensitivity to magnetic fields and their compact size which make them well suited for applications near or in the magnetic field and within the limited space available inside the UA1 magnet. The total number of MPPCs used for the SMRD amounts to 4016. The custom made version of the MPPCs for T2K consist of an array of 667 independent 50×50 µm2avalanche photodiodes (pixels) operating in Geiger mode. The MPPC sensitive area of 1.3×1.3mm2is well suited to accept light from a 1 mm diameter Y11 fiber. The MPPC signal is a sum of pixel avalanches and this multi-pixel sensor operates as an analog photodetector with a dynamic range that is limited by the finite number of pixels. Typical light signals in SMRD counters are below 50 photoelectrons and therefore dynamic range issue are not aconcern. Eachpixelcanberepresentedasamicrocapacitor which quickly discharges during Geiger breakdown initiated by a photoelectron until the voltage difference across it has decreased below the breakdown voltage. The overvoltage, which is defined as the difference between the supplied bias voltage and the breakdown voltage, is the main parameter that affects the performance of MPPCs and the stability of its operation. MPPCs have an excellent single photoelectron resolution up to mean charges corresponding to about 30 photoelectrons that allow to perform an accurate calibration of each counter. At a temperature of T=25◦Cand an overvoltage of 1.6 V MPPCs are characterized by a typical gain of 7.5×105,aphoto detection efficiency of about 25% for green light as emitted by a Y11 fiber. The average dark rate amounts to 700 kHz with maximum values approaching up to 1 MHz, the estimated combined crosstalk and afterpulse probability is 20-25% and the recovery time of a single pixel is 30 ps. The MPPCs of the SMRD were operated in the T2K neutrino beam starting in 2009 and after more than 1.5 years of operation only one sensor (0.025%) is suspected to have failed. All MPPCs were tested extensively as function of bias voltage and temperature and in particular the gain and dark rate had to satisfy stringent criteria in order for a MPPC to be included in the SMRD. 3.4. Module Assembly and Installation At multiple stages of the detector production and assembly the performance of the scintillation counters with embedded WLS fiber were tested in response to cosmic rays. First, the scintillators were tested immediately after the extrusion process by measuring the light yield with a photomultiplier tube and in response to throughgoing muons. Secondly, after the endcaps were attached to the scintillation counters and the WLS fiber had been glued into the grooves with BC600 optical glue the counters were retested using MPPCs and double ended readout in responsee to central penetrating muons. Out of 2008 counters 20 were found to have a large asymmetry (more than 50%) in light yield between the two ends. The asymmetry was attributed to a damages of the fiber cladding encurred during the gluing process. These 20 counters were repaired by gluing a new fiber into a refurbished groove after milling out the previously glued fiber. All counters which passed the quality test, were wrapped by one layer of 0.1 mm thick Tyvek paper which leads to a further increase in light yield of 15%. Ferrule Ferrule Endcap Endcap Y11 WLS fiber (Kuraray) Scintillator (Tyvek) Reflective layer Stainless steel container Figure 5: SMRD counter sliced view. Successively the counters were wrapped in a lightproof stainless steel container (Fig. 5). The container was attached to scintillator and endcaps with DP-490 black epoxy glue and a double sided 20×0.15 mm Tesa adhesive tape. Additionally all joints between the container surface and the endcap were covered by ablackTesaadhesivetapeof25×0.065 mm. Each fiber end inside the endcap ferrule was cut by a cylindrical mill and polished to provide good optical contact with the MPPC. After assembly, the dark noise of each SMRD counter was measured with MPPCs and an oscilloscope to ensure the absence of light leaks. In total, 2130 counters (800 – 167 mm wide and 1330 –175mmwidecounters)wereassembledandtestedtobeof good quality. After shipment to Japan 230 counters were found to have developed a sub-millimeter sized air gap between the end of the fiber and the face of the ferule, resulting in a small loss in light yield. Hence the endcaps of all counters were additionally fastened by 2 stainless steel screws each in orderto minimize the risk of future counter degradation. After refurbishment of the problematic counters all counters were re-tested and demonstrated to show excellent performance. Single counters are assembled into bigger units named modules to facilitate installation and to stabilize the position of the counters in the magnet slits. In order to match the different dimensions of vertical and horizontal magnet slits two types of SMRD modules were built. Modules intended for vertical slits consist of five counters (each 175mm wide) while the horizontal ones consist of four counters (each 167 mm wide). Extruded Aluminum H-profiles are used to inter-connect counters into modules. The boundary edges of the first and last counter in each module were protected with aluminium U-channels as shown in fig. 6. The counters and the extrusions are tightly wrapped with capton tape in three locations. In order to stabilize a module inside a magnet slit tape springs made of phosphorbronze are mounted on both sides of the modules as indicated in Fig. 7. Two springs are mounted on each side of the vertical modules while three per side are attached to the H-profiles of horizontal modules. Any lateral and longitudinal forces from the springs act on the H-profiles and not on the 5 (b) SMRD counter sliced view. Figure 6. Schematic view of SMRD module positions (a) and image of an SMRD counter design (b). yoke segment. For better collection of the scintillator light and to improve the positional accuracy in the SMRD, S-shaped (curvature of ∅= 58 mm) WLS fibres run down the length or the bars as shown in figure 6(b). The fibres are bent and glued into grooves within the scintillator bars using BC600 Bicron glue. It is worth noting that any degradation of this glue with time could have an impact on the light yield measured by the SMRD. The design results in nearly uniform response across the surface while reducing the number of channels to read out. The signal is read out from both ends of the bar via MPPCs. Each fibre exits through a ferrule which is part of a custom made endcap, glued and screwed to the scintillator, to which a connector with the MPPC is attached. A foam spring ensures a reliable coupling between the photosensor and the fibre. 2.2 INGRID The INGRID detector consists of 16 identical iron and plastic scintillator detector modules. Each module is constructed of 11 tracking plastic scintillator planes interleaved with 9 passive iron plates, as shown in figure 7(the final pair of scintillator planes lacks an interleaved iron plate). Each scintillator plane lies perpendicular to the beam direction and consists of 24 horizontally (X) orientated bars glued to a further 24 vertically (Y) orientated bars. Each bar is 1203 mm long and has a cross section of 50 ×10 mm2. In common with the PØD and ECal scintillator bars, the INGRID bars were produced at Fermilab in 2007–2008. As such they have the same material composition of Dow Styron 663 W polystyrene, doped with 1% PPO and 0.03% POPOP, and co-extruded with a TiO 2 rich material to allow diffuse reflection of scintillation light. Unglued Kuraray Y11(200) M-type WLS fibres collect the light from the bars and are coupled on one end to Hamamatsu (S10362-13-050C) MPPCs as shown in figure 8. The uninstrumented ends of the bars and fibres are painted with a reflective coating of ELJEN ® EJ-510. The design of the fibre-MPPC coupling for INGRID is the same as used in the FGDs. –8– 2022 JINST 17 P10028 Table 3. Absolute range and modal light yield stability uncertainties in PEU for each subsystem. Also shown are the range and modal uncertainties as a percentage of the recorded MPV in each time bin. Subsystem Uncertainty Range Modal Uncertainty Absolute Value (PEU) % of MPV Absolute Value (PEU) % of MPV PØD 0.02–0.57 0.11–2.57 ∼0.20 ∼0.90 ECal (Single-ended) 0.07–2.19 0.28–8.24 ∼0.15 ∼0.80 ECal (Double-ended) 0.05–1.22 0.33–7.35 ∼0.10 ∼0.90 FGD 0.10–0.29 0.51–1.33 ∼0.15 ∼0.70 SMRD 0.79–1.33 1.4–2.3 1.33 2.3 INGRID 0.07–0.74 0.30–3.28 ∼0.30 ∼1.50 all SMRD data points. The range of uncertainties, and modal uncertainty, across all data periods are shown in table 3. Most uncertainty values lie close to modal value, with a few exceptions which push out the maximum range to higher values. 3.4.2 INGRID INGRID takes a similar approach to the ND280 subsystems, aggregating the cosmic ray data over 3 day periods and extracting the standard deviation in the MOM extracted from those reduced periods. This provides uncertainties of 0.3-3.3% in each time bin. 4 Annual light yield reduction The distribution of the ND280 MIP MPV or INGRID MOM was extracted for each subsystem during each T2K Run and then fitted with a linear function in order to calculate the overall drop in light yield and annual decrements, see figure 13. The data are aggregated by T2K Run for the PØD, FGD and SMRD, with the time error being the standard deviation in time stamp of all MIP hits during each T2K Run. For the ECal and INGRID, the higher statistic allows for the data to be instead aggregated on a per month basis, or per J-PARC Main Ring Run, respectively. The fit is only applied to the data from January 2012 (December 2010) onwards for the ND280 (INGRID) subsystems as the current calibration procedures and cosmic ray triggering prescale were not finalised until that time. These differences in detector condition cannot be corrected for subsequently, but the early data are retained for completeness. As described in the previous section, for the PØD and ECal subsystems (see sections 3.2.1 and 3.2.3 respectively) the study has been performed for all sub-modules or bar types separately, see figures 13(a)–13(d) and 13(e)–13(h). The reduction in light yield extracted by the described method measures the loss in performance of the whole readout system; the scintillator, WLS fibre, MPPC and their couplings. However it is assumed that the bulk of the light yield reduction can be ascribed to the degradation of the plastic scintillator as this is a well known phenomenon (as described in section 1.2), and there has been no obvious degradation in MPPC performance (for example significant drift in overvoltage settings with time), and the stability of the WLS fibre will be addressed in section 6. Without knowledge of the ageing mechanism(s) degrading the ND280 and INGRID subsystems it is difficult to know what form the time dependence on the ageing rate should be expected to – 15 – 2022 JINST 17 P10028 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 Light Yield (PEU) Dule 0)∅Data (Super-P Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.4−0.2−0.0 0.2 0.4 0.6 0.8 1.0 1.2 Residual (a) PØD, Super-PØDule 0. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 Light Yield (PEU) Dule 1)∅Data (Super-P Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.8−0.6−0.4−0.2−0.0−0.2 0.4 0.6 0.8 1.0 1.2 Residual (b) PØD, Super-PØDule 1. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 18 19 20 21 22 Light Yield (PEU) Dule 2)∅Data (Super-P Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.5− 0.0 0.5 1.0 1.5 Residual (c) PØD, Super-PØDule 2. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 18 19 20 21 22 23 Light Yield (PEU) Dule 3)∅Data (Super-P Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.5− 0.0 0.5 1.0 1.5 Residual (d) PØD, Super-PØDule 3. Figure 13. Light yield change in each subsystem (upper pad) and residual when the linear fit result is subtracted from each data point (lower pad), for T2K Runs 3–9 (ND280) and Runs 2–11 (INGRID). The x-error bars (time) show the standard deviation in the hit times for each data aggregation period, and the y-error bars (light yield) show the quadratic sum of the light yield stability uncertainty and the uncertainty on the ND280 Landau-Gaussian MIP MPV or INGRID MOM. Hollow data points are excluded from the fits as the current calibration procedures and cosmic ray triggering prescale were not finalised at that time. The exponential fits used in section 5are not shown as the linear and exponential fit lines for all detectors overlap almost completely on the displayed timescales. – 16 – 2022 JINST 17 P10028 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 23 24 25 26 27 28 29 30 31 Light Yield (PEU) Data (ECal Barrel X) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 1− 0 1 2 3 4 Residual (e) ECal Barrel X. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 21 22 23 24 25 26 27 28 29 Light Yield (PEU) Data (ECal Barrel Y) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 1− 0 1 2 3 4 Residual (f) ECal Barrel Y. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 13 14 15 16 17 Light Yield (PEU) Data (ECal Barrel Z) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.5− 0.0 0.5 1.0 1.5 Residual (g) ECal Barrel Z. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 13 14 15 16 17 18 Light Yield (PEU) Data (Downstream ECal) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.5−0.0 0.5 1.0 1.5 2.0 Residual (h) Downstream ECal. Figure 13. Light yield change in each subsystem (upper pad) and residual when the linear fit result is subtracted from each data point (lower pad), for T2K Runs 3–9 (ND280) and Runs 2–11 (INGRID). The x-error bars (time) show the standard deviation in the hit times for each data aggregation period, and the y-error bars (light yield) show the quadratic sum of the light yield stability uncertainty and the uncertainty on the ND280 Landau-Gaussian MIP MPV or INGRID MOM. Hollow data points are excluded from the fits as the current calibration procedures and cosmic ray triggering prescale were not finalised at that time. The exponential fits used in section 5are not shown as the linear and exponential fit lines for all detectors overlap almost completely on the displayed timescales. – 17 – 2022 JINST 17 P10028 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 20.5 21.0 21.5 22.0 22.5 Light Yield (PEU) Data (FGD) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 0.3−0.2−0.1−0.0−0.1 0.2 0.3 0.4 Residual (i) FGD. 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 55 56 57 58 59 60 61 62 63 Light Yield (PEU) Data (SMRD) Linear Fit 0 1 2 3 4 5 6 7 8 9 Year (from 2010) 2−1−0 1 2 3 Residual (j) SMRD. 0 2 4 6 8 10 12 Year (from 2010) 20 21 22 23 24 25 Light Yield (PEU) Data (INGRID) Linear Fit 0 2 4 6 8 10 12 Year (from 2010) 1.5−1.0−0.5−0.0 0.5 1.0 1.5 2.0 Residual (k) INGRID. Figure 13. Light yield change in each subsystem (upper pad) and residual when the linear fit result is subtracted from each data point (lower pad), for T2K Runs 3–9 (ND280) and Runs 2–11 (INGRID). The x-error bars (time) show the standard deviation in the hit times for each data aggregation period, and the y-error bars (light yield) show the quadratic sum of the light yield stability uncertainty and the uncertainty on the ND280 Landau-Gaussian MIP MPV or INGRID MOM. Hollow data points are excluded from the fits as the current calibration procedures and cosmic ray triggering prescale were not finalised at that time. The exponential fits used in section 5are not shown as the linear and exponential fit lines for all detectors overlap almost completely on the displayed timescales. – 18 – 2022 JINST 17 P10028 Table 4. Linear fit parameters to PØD, FGD, SMRD, ECal and INGRID data from figure 13 and the annual percentage reduction, relative to the 2012 fit values. Single-ended readout bars are mirrored on one end. Subsystem Readout Type A (PEU) B (PEU/yr) 𝜒2/NDF Annual Reduction (%) Super-PØDule 0 Single-ended 19.97 ±0.15 0.35 ±0.03 4.35/5=0.87 1.82 ±0.16 Super-PØDule 1 Single-ended 21.17 ±0.16 0.36 ±0.04 10.39/5=2.08 1.76 ±0.20 Super-PØDule 2 Single-ended 21.15 ±0.17 0.36 ±0.03 9.14/5=1.83 1.76 ±0.15 Super-PØDule 3 Single-ended 21.33 ±0.16 0.37 ±0.03 5.46/5=1.09 1.80 ±0.15 FGD Single-ended 22.68 ±0.19 0.27 ±0.04 0.74/5=0.15 1.22 ±0.18 SMRD Double-ended 60.86 ±1.48 0.54 ±0.26 2.62/5=0.52 0.90 ±0.44 ECal Barrel X Single-ended 27.27 ±0.06 0.52 ±0.01 33.09/37 =0.89 1.98 ±0.04 ECal Barrel Y Single-ended 25.21 ±0.08 0.49 ±0.01 31.88/37 =0.86 2.02 ±0.04 ECal Barrel Z Double-ended 16.01 ±0.04 0.33 ±0.01 36.60/37 =0.99 2.15 ±0.07 Downstream ECal Double-ended 15.48 ±0.06 0.28 ±0.01 11.57/37 =0.31 1.87 ±0.07 INGRID Single-ended 24.50 ±0.11 0.42 ±0.02 89.32/33 =2.71 1.78 ±0.08 take. A priori it might be expected that an exponential function would be suitable, and fits of this form are used for projecting the future response of the most important subsystems in section 5. However, given the observed data distributions and timescale studied a simple linear fit is found to be appropriate, and are applied in the form: 𝑓(𝑡)=𝐴−𝐵𝑡 , (4.1) where 𝐴 is the fitted light yield in PEU at year 0 (2010), 𝐵 is the gradient of the fit in PEU per year, and 𝑡is the year since 2010. The fit parameters are shown in table 4. The degradation of the scintillator appears to be reasonably consistent across all subsystems. All show a reduction in light yield within the range ∼ 0 . 3 – 0 . 5PEU per year, equivalent to an annual light yield reduction of 0.9–2.2% relative to their 2012 fit values. The 1% difference separating the highest and lowest degradation rates between the materially identical FGD (1.2%) and ECal (Barrel Z 2.2%), is not surprising given the differences in production dates for the scintillator bars, and the varying environmental conditions they have experienced during their production, transportation, and positioning within the ND280. All of the aforementioned factors will have contributed to differences in the temperature, humidity and UV exposure of the bars across their lifetimes, and so impacted upon their respective ageing profiles. The higher statistics of the ECal allows for a finer assessment of its ageing ratee, see figures 13(e)– 13(h). An initial rapid ageing is observed within the first two years of operation, followed by a near linear reduction beyond 2012, however it is unclear if this is a real effect or just an artefact of the changes in calibration procedure and cosmic ray triggering prescale. The higher ECal light yield obtained by the Barrel X and Y bars is due to the combination of direct and reflected light signals for these single-ended (mirrored) readout channels, compared to direct transmission only for the double-ended readout of the Barrel Z and Downstream bars. Results from the MINOS experiment, which uses materially identical bars to the FGD, ECal, PØD and INGRID, showed ageing rates of ∼ 2% per year [ 10 ] over 3 and 4.5 year periods measured with their near and far detectors respectively, in good agreement with the higher rates we obtain from the ECal and PØD. – 19 – 2022 JINST 17 P10028 The MINER 𝜈 A experiment found a substantially higher rate of ageing for their scintillator bars, equivalent to a ∼ 7 . 5% annual reduction in light yield over a 2 year study period [ 11 ]. It is unclear why MINER 𝜈 A measured such a high rate of degradation as their scintillator composition is again identical to that used by MINOS and most T2K subsystems. It might be possible that MINER 𝜈 A has sampled an initial rapid ageing of their scintillator, as perhaps indicated in the earliest ECal data points as discussed above, and also anecdotally observed by MINOS [ 10 ]; and that further study of later data would show a reduced ageing rate in line with those measured by T2K and MINOS. For completeness, if a linear fit is applied only to the currently excluded Downstream ECal data recorded during the 2010–2011 period, an annual light loss rate of 1 . 33 ± 0 . 29 PEU per year on an initial light yield of 17 . 2 ± 0 . 3is obtained. This is equivalent to annual reduction in light yield of 7.7±1.7% which is in excellent agreement with the MINER𝜈A result. 5 Projected future response The PØD subsystem of the ND280 is being decommissioned in 2022 to allow for the upgrade of the ND280 detector [ 31 ]. The remaining ECal, FGD, SMRD and INGRID subsystems will be retained in their current form, and so an understanding of their future response will become increasingly important as the T2K near detectors continue operating into the T2K-II [ 32 ] and the Hyper-Kamiokande [33] eras. As such the future response of the ECal, FGD and INGRID subsystems has been projected through until 2040. The SMRD is excluded from this study as its initial light yield is substantially higher than for the other subsystems and its rate of degradation is lower. As such the likelihood of the light yield from this subsystem dropping below any reconstruction threshold is not considered to be an issue for the time period considered. Although a linear fit to the data in section 4results in a reasonable agreement, an exponential fit is better physically motivated. Figure 14 shows the projected future response from the earlier linear fits, and from the application of an exponential fit to the ECal and FGD data from 2012, and the INGRID data from 2010, onwards. The exponential fit is of the form: 𝑓(𝑡)=𝐴exp −𝑡 𝜏,(5.1) where 𝐴 is the fitted light yield in PEU at year 0 (2010), 𝜏 is the time constant of the exponent in years, and 𝑡is the year since 2010. The fit parameters are shown in table 5. The time constant 𝜏 is consistent, ∼ 44 – 49 years, for all ECal bar types, along with the light yield constant 𝐴 for the pairs of single-ended (mirrored), ∼ 26 PEU, and double-ended, ∼ 16 PEU, readout bars. The resultant 𝜒2/NDF for the exponential fits are marginally reduced by ∼ 0 . 1 – 0 . 2 compared to the corresponding linear fits shown in table 4. The INGRID time constant of 52 . 9 ± 2 . 4years is marginally longer than those recorded by the ECal, and the resultant 𝜒2/NDF for the exponential fit saw a slight reduction of 0.22. The FGD records a significantly longer time constant of 80 . 3 ± 11 . 1years and a negligible increase of 0.01 in its 𝜒2/NDF for the exponential fit compared to the linear fit. As with the linear ageing results (see table 4), some variation in the degradation rates between the different subsystems is expected due to the varying age and environmental exposure profiles of the scintillator bars, although why the FGD should be such an outlier is unclear. – 20 – 2022 JINST 17 P10028 0 5 10 15 20 25 30 Year (from 2010) 10 12 14 16 18 20 22 24 26 28 30 Light Yield (PEU) Data (ECal Barrel X) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (a) ECal Barrel X. 0 5 10 15 20 25 30 Year (from 2010) 10 12 14 16 18 20 22 24 26 28 Light Yield (PEU) Data (ECal Barrel Y) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (b) ECal Barrel Y. 0 5 10 15 20 25 30 Year (from 2010) 6 8 10 12 14 16 18 Light Yield (PEU) Data (ECal Barrel Z) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (c) ECal Barrel Z. 0 5 10 15 20 25 30 Year (from 2010) 6 8 10 12 14 16 18 Light Yield (PEU) Data (Downstream ECal) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (d) Downstream ECal. 0 5 10 15 20 25 30 Year (from 2010) 12 14 16 18 20 22 Light Yield (PEU) Data (FGD) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (e) FGD. 0 5 10 15 20 25 30 Year (from 2010) 12 14 16 18 20 22 24 Light Yield (PEU) Data (INGRID) Linear Fit 68% Confidence Interval Linear Fit 95% Confidence Interval Exponential Fit 68% Confidence Interval Exponential Fit 95% Confidence Interval (f) INGRID. Figure 14. Projected light yield for each ECal bar type, FGD and INGRID, showing the 68% and 95% confidence intervals extracted from both the linear and exponential fits to the data. Hollow data points are excluded from the data fits. Table 5. Exponential fit parameters to ECal, FGD and INGRID data from figure 14. ECal Bar Type Readout Type A (PEU) 𝜏(yr) 𝜒2/NDF Barrel X Single-ended (mirrored) 27.39 ±0.07 47.7±1.1 27.82/37 =0.75 Barrel Y Single-ended (mirrored) 25.34 ±0.08 45.7±1.3 28.27/37 =0.76 Barrel Z Double-ended 16.10 ±0.04 44.1±0.9 29.64/37 =0.80 Downstream Double-ended 15.55 ±0.07 49.2±2.6 10.28/37 =0.28 FGD Single-ended (mirrored) 22.72 ±0.20 80.3±11.1 0.68/5=0.14 INGRID Single-ended (mirrored) 24.61 ±0.11 52.9±2.4 82.27/33 =2.49 – 21 – 2022 JINST 17 P10028 The anticipated ECal response drops by ∼ 50% or ∼ 60% over thirty years for all bar types from extrapolations of the exponential and linear fits, respectively. This remains above the minimum charge threshold of 5.5 PEU required for use in the current ECal offline reconstruction algorithms. The value of this threshold has been chosen to avoid discrepancies between data and the current MC simulation at low charges. It should be possible to lower the current charge threshold through more detailed simulation of the detector response, for example including bar non-uniformity and improved MPPC dark-noise rate, and through enhancing the reconstruction algorithms. Without improvement there is a risk that information will be lost for particle interactions which deposit energy at values below the MIP MPV, potentially limiting the physics reach of analyses which utilise the ECals. The FGD and INGRID subsystems expect their MIP MPV or MOM response to reduce by ∼ 30% and ∼ 40%, respectively, over thirty years under the hypothesis of an exponential decline. For both detectors this increases by a further ∼ 5 – 10% for a linear decline. For both scenarios this remains far above the hit thresholds of 5.0 and 2.5 PEU, respectively, currently used by the offline reconstruction algorithms for these detectors. If the true rate of ageing were to be higher, such as the ∼ 50 – 60% light yield reduction currently projected by the ECal, this would still not be a concern for these subsystems. 6 Separation of ECal scintillator and fibre degradation The results shown in section 4combines the ageing of the scintillator bars with that of the WLS fibres, 2 therefore a second approach was applied to separate the two effects within the ECal data. Without applying the attenuation correction, the MIP MPV response is extracted at different distances from the sensor for each bar type during each T2K Run, see figure 15. The best fit to the data was achieved by applying a double-exponential fit, which accounts for the short and long components of the fibre attenuation, of the form: 𝑓(𝑥)=𝑆exp −𝑥 𝜆𝑆+𝐿exp −𝑥 𝜆𝐿,(6.1) where 𝑆 and 𝐿 are the fitted light yield in PEU at 0 cm from the MPPC for the short and long components of the exponential function, respectively; 𝜆𝑆 and 𝜆𝐿 are the associated short and long attenuation lengths; and 𝑥is the distance from the MPPC in cm. 6.1 Scintillator degradation The parameters of the double-exponential fits can be used to calculate the predicted total MIP light yield at a distance of 0 cm from the MPPC, 𝑓( 0 )=𝑆+𝐿 . This should remove the dependence on the propagation of the light down the WLS fibre and the decrease in evaluated light yield will only be dependent on the ageing of the scintillator. The results for this evaluation are shown in figure 16 with both a linear and exponential fit applied to the data from 2012. The linear fit is of the form: 𝑓(𝑡)=𝐴−𝐵𝑡 , (6.2) 2 Any degradation of the coupling between the fibre and the MPPC, either through loss of transparency of the epoxy or gradual displacement of the fibre also contributes to the results. – 22 – 2022 JINST 17 P10028 0 20 40 60 80 100 120 140 Distance (cm) 22 24 26 28 30 32 34 36 38 40 Light Yield (PEU) Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 (a) ECal Barrel X. 0 50 100 150 200 Distance (cm) 20 25 30 35 40 Light Yield (PEU) Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 (b) ECal Barrel Y. 0 50 100 150 200 250 300 350 Distance (cm) 5 10 15 20 25 30 Light Yield (PEU) Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 (c) ECal Barrel Z. 0 50 100 150 200 Distance (cm) 10 15 20 25 30 Light Yield (PEU) Run 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 Run 1 Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 (d) Downstream ECal. Figure 15. ECal light yield as a function of distance to the MPPC for each T2K Run. The errors on the data points are only the uncertainty on the Landau-Gaussian fit MPV at each distance point, no light yield stability uncertainty is applied. Results of the fits to the hollow data points are excluded from the subsequent data fits. 0123456789 Year (from 2010) 22 24 26 28 30 32 34 36 38 40 42 Light Yield (PEU) Barrel X Downstream ECal Barrel Y Barrel Z Linear Fit Exponential Fit Barrel X Downstream ECal Barrel Y Barrel Z Linear Fit Exponential Fit Figure 16. Light yield as evaluated at 0 cm from the MPPC for each ECal bar type. Hollow data points are excluded from the data fits. where 𝐴 is the fitted total MIP light yield ( 𝑆+𝐿 ) in PEU at year 0 (2010), 𝐵 is the gradient of the fit in PEU per year, and 𝑡is the time in years since 2010. The fit parameters are shown in table 6. For the single-ended (mirrored) readout bars the reduction in light yield from the scintillator ageing is ∼ 0 . 75 PEU per year, and for the double-ended readout bars it is ∼ 0 . 50 PEU per year. – 23 – 2022 JINST 17 P10028 Table 6. Linear fit parameters to ECal light yield at 0 cm from the MPPC from figure 16, and annual percentage reduction in light yield, relative to 2012 fit value. Reference results, in parentheses, from the linear fit in table 4are included for comparison ECal Bar Type Readout Type A (PEU) B (PEU/yr) 𝜒2/NDF Annual Reduction (Ref.) (%) Barrel X Single-ended 38.21 ±0.50 0.76 ±0.09 0.45/5=0.09 2.07 ±0.25 (1.98 ±0.04) Barrel Y Single-ended 36.94 ±0.47 0.73 ±0.08 1.40/5=0.28 2.06 ±0.23 (2.02 ±0.05) Barrel Z Double-ended 27.65 ±0.18 0.50 ±0.03 3.66/5=0.73 1.88 ±0.11 (2.15 ±0.07) Downstream Double-ended 27.87 ±0.32 0.49 ±0.05 2.69/5=0.54 1.82 ±0.18 (1.87 ±0.07) Table 7. Exponential fit parameters to ECal light yield at 0 cm from the MPPC from figure 16. Reference time constants, in parentheses, from the exponential fit in table 5are included for comparison. ECal Bar Type Readout Type A (PEU) 𝜏(Ref.) (yr) 𝜒2/NDF Barrel X Single-ended 38.4±0.5 45.2±5.1(47.7±1.1)0.37/5=0.07 Barrel Y Single-ended 37.2±0.5 45.2±4.8(45.7±1.3)1.11/5=0.22 Barrel Z Double-ended 27.8±0.02 49.5±3.0(44.1±0.9)3.22/5=0.64 Downstream Double-ended 28.0±0.3 51.6±5.3(49.2±2.6)2.82/5=0.56 This is a reduction of ∼ 2 . 1% for the single-ended (mirrored) bars, and ∼ 1 . 9% per year for the double-ended bars. The exponential fit is of the form: 𝑓(𝑡)=𝐴exp −𝑡 𝜏,(6.3) where 𝐴 is the fitted total MIP light yield ( 𝑆+𝐿 ) in PEU at year 0 (2010), 𝜏 is the time constant of the exponent in years, and 𝑡is the year since 2010. The fit parameters are shown in table 7. For the linear fits the annual reduction in light yield is consistent within ∼ 1 𝜎 of the reference degradation shown in table 4, and similarly the time constant for the exponential fits is consistent within ∼ 1 𝜎 of the reference values shown in table 5. This suggests the ageing is dominated by the degradation of the scintillator rather than the WLS fibres. The exception to this is the Barrel Z results which lie ∼ 2 𝜎 from the reference values and imply a slower rate of degradation than those shown in the earlier results of section 4and 5. This is likely due to some loss in MIP hit efficiency at the furthest distances from the MPPCs as the scintillator degrades. This would truncate the rising edge of the MIP light yield distribution, see for example at a distance of 360 cm in figure 17, shifting the extracted MIP MPV to a slightly higher value than might be otherwise expected. The result would be an underestimate in the degradation rate extracted with this technique for the Barrel Z bars, leading to the discrepancy when making comparisons with the reference values. Fortunately any loss in hit efficiency at the furthest distances from the MPPCs will have negligible impact on the overall hit reconstruction efficiency as the MPPC on the opposing end of the bars will continue to efficiently reconstruct these hits, as only one MPPC is required to reconstruct a hit on the double-ended readout bars. This is confirmed by separate studies monitoring hit efficiency in the ECal modules which observed a negligible reduction ( ∼ 0 . 1%) in the single-hit efficiency (requiring a hit in the singleended readout bars, or at least one hit on either end of the double-ended readout bars) across all bar – 24 – 2022 JINST 17 P10028 The T2K collaboration K. Abe,60 N. Akhlaq,49 R. Akutsu,66 A. Ali,71,66 C. Alt,10 C. Andreopoulos,58,37 M. Antonova,19 S. Aoki,32 T. Arihara,63 Y. Asada,73 Y. Ashida,33 E.T. Atkin,21 S. Ban,33 M. Barbi,50 G.J. Barker,70 G. Barr,46 D. Barrow,46 M. Batkiewicz-Kwasniak,14 F. Bench,37 V. Berardi,22 L. Berns,69 S. Bhadra,74 A. Blanchet,57 A. Blondel,57,12 S. Bolognesi,5T. Bonus,72 S. Bordoni,12 S.B. Boyd,70 A. Bravar,12 C. Bronner,60 S. Bron,12 A. Bubak,56 M. Buizza Avanzini,36 N.F. Calabria,23 S. Cao,20 A.J. Carter,52 S.L. Cartwright,55 M.G. Catanesi,22 A. Cervera,19 J. Chakrani,36 D. Cherdack,16 G. Christodoulou,11 M. Cicerchia, 24,𝑎 J. Coleman, 37 G. Collazuol, 24 L. Cook, 46,29 A. Cudd, 6 Yu.I. Davydov, 39 A. De Roeck, 11 G. De Rosa,23 T. Dealtry,34 C.C. Delogu,24 C. Densham,58 A. Dergacheva,26 F. Di Lodovico,31 S. Dolan,11 D. Douqa,12 T.A. Doyle,34 O. Drapier,36 K.E. Duffy,46 J. Dumarchez,57 P. Dunne,21 K. Dygnarowicz,68 A. Eguchi,59 S. Emery-Schrenk,5A. Ershova,5S. Fedotov,26 P. Fernandez,37 A.J. Finch,34 G.A. Fiorentini Aguirre74 G. Fiorillo,23 M. Friend,15,𝑏 Y. Fujii,15,𝑏 Y. Fukuda,41 K. Fusshoeller, 10 C. Giganti, 57 V. Glagolev, 39 M. Gonin, 28 E.A.G. Goodman, 13 A. Gorin, 26 M. Grassi, 24 M. Guigue,57 D.R. Hadley,70 J.T. Haigh,70 P. Hamacher-Baumann,53 D.A. Harris,74 M. Hartz,66,29 T. Hasegawa,15,𝑏 S. Hassani,5N.C. Hastings,15 A. Hatzikoutelis,34,𝑐 Y. Hayato,60,29 A. Hiramoto,33 M. Hogan,7J. Holeczek,56 A. Holin58 T.J. Holvey,46 N.T. Hong Van,20,27 T. Honjo,45 F. Iacob,24 A.K. Ichikawa,69 M. Ikeda,60 T. Ishida,15,𝑏 M. Ishitsuka,64 H.T. Israel,55 S.J Ives,21 K. Iwamoto,59 A. Izmaylov,26 N. Izumi,64 M. Jakkapu,15 B. Jamieson,71 S.J. Jenkins,55 C. Jesús-Valls,17 J.J. Jiang,43 P. Jonsson, 21 C.K. Jung, 43,§ P.B. Jurj, 21 M. Kabirnezhad, 74 A.C. Kaboth, 52,58 T. Kajita, 61,𝑑 H. Kakuno, 63 J. Kameda,60 S.P. Kasetti,38 Y. Kataoka,60 Y. Katayama,73 T. Katori,31 M. Kawaue,33 E. Kearns,3,29,𝑑 M. Khabibullin,26 A. Khotjantsev,26 T. Kikawa,33 H. Kikutani,59 S. King,31 J. Kisiel,56 A. Knight,70 T. Kobata,45 T. Kobayashi,15,𝑏 L. Koch,18 G. Kogan,21 A. Konaka,66 L.L. Kormos,34 Y. Koshio,44,𝑑 A. Kostin,26 K. Kowalik,42 Y. Kudenko,26,𝑒 S. Kuribayashi,33 R. Kurjata,68 T. Kutter,38 M. Kuze,62 M. La Commara,23 L. Labarga,1K. Lachner70 J. Lagoda,42 S.M. Lakshmi,42 M. Lamers James,34,58 I. Lamont,34 M. Lamoureux,24 D. Last,47 N. Latham,70 M. Laveder,24 M. Lawe,34,∗Y. Lee,33 C. Lin,21 T. Lindner,66 S.-K. Lin,38 R.P. Litchfield,13 S.L. Liu,43 A. Longhin,24 K.R. Long,21,58 L. Ludovici,25 X. Lu,70 T. Lux,17 L.N. Machado,23 L. Magaletti,22 K. Mahn,40 M. Malek,55 M. Mandal,42 S. Manly,51 A.D. Marino,6L. Marti-Magro,73 D.G.R. Martin,21 M. Martini,57, 𝑓 J.F. Martin,65 T. Maruyama,15,𝑏 T. Matsubara,15 V. Matveev,26 C. Mauger,47 K. Mavrokoridis,37 E. Mazzucato,5N. McCauley,37 J. McElwee,55 K.S. McFarland,51 C. McGrew,43 A. Mefodiev,26 G.D. Megias,54,61 L. Mellet,57 C. Metelko, 37 M. Mezzetto, 24 A. Minamino, 73 O. Mineev, 26 S. Mine, 4 M. Miura, 60,𝑑 L. Molina Bueno, 19 S. Moriyama, 60,𝑑 Th.A. Mueller, 36 D. Munford, 16 L. Munteanu, 5 K. Nagai 73 Y. Nagai, 9 T. Nakadaira, 15,𝑏 K. Nakagiri,59 M. Nakahata,60,29 Y. Nakajima,59 A. Nakamura,44 H. Nakamura,64 K. Nakamura,29,15,𝑏 Y. Nakano,60 S. Nakayama,60,29 T. Nakaya,33,29 K. Nakayoshi,15,𝑏 C.E.R. Naseby,21 T.V. Ngoc,20,𝑔 V.Q. Nguyen,57 K. Niewczas,72 Y. Nishimura,30 K. Nishizaki,45 F. Nova,58 P. Novella,19 J.C. Nugent,13 H.M. O’Keeffe,34 L. O’Sullivan,55 T. Odagawa,33 T. Ogawa,15 R. Okada,44 K. Okumura,61,29 T. Okusawa,45 R.A. Owen,49 Y. Oyama,15,𝑏 V. Palladino,23 V. Paolone,48 M. Pari,24 J. Parlone,37 S. Parsa,12 J. Pasternak,21 M. Pavin,66 D. Payne,37 G.C. Penn,37 J.D. Perkin,55 D. Pershey,8 L. Pickering, 52 C. Pidcott, 55 G. Pintaudi, 73 C. Pistillo, 2 B. Popov, 57,ℎ K. Porwit, 56 M. Posiadala-Zezula, 67 Y.S. Prabhu,42 B. Quilain,36 T. Radermacher,53 E. Radicioni,22 B. Radics,10 P.N. Ratoff,34 M. Reh,6 C. Riccio,43 E. Rondio,42 S. Roth,53 A. Rubbia,10 A.C. Ruggeri,23 C.A. Ruggles,13 A. Rychter,68 K. Sakashita,15,𝑏 F. Sánchez,12 G. Santucci,74 C.M. Schloesser,10 K. Scholberg,8,𝑑 M. Scott,21 Y. Seiya,45,𝑖 T. Sekiguchi,15,𝑏 H. Sekiya,60,29,𝑑 D. Sgalaberna,10 A. Shaikhiev,26 A. Shaykina,26 M. Shiozawa,60,29 W. Shorrock,21 A. Shvartsman,26 K. Skwarczynski,42 M. Smy,4J.T. Sobczyk,72 H. Sobel, 4,29 F.J.P. Soler, 13 Y. Sonoda, 60 R. Spina, 22 H. Su, 48 I.A. Suslov, 39 S. Suvorov, 26,57 A. Suzuki, 32 S.Y. Suzuki,15,𝑏 Y. Suzuki,29 A.A. Sztuc,21 M. Tada,15,𝑏 S. Takayasu,45 A. Takeda,60 Y. Takeuchi,32,29 H.K. Tanaka,60,𝑑 Y. Tanihara,73 M. Tani,33 V.V. Tereshchenko,39 N. Teshima,45 N. Thamm,53 – 31 – 2022 JINST 17 P10028 L.F. Thompson,55 W. Toki,7C. Touramanis,37 T. Towstego,65 K.M. Tsui,37 T. Tsukamoto,15,𝑏 M. Tzanov,38 Y. Uchida,21 A. Vacheret,21 M. Vagins,29,4Z. Vallari,43, 𝑗 D. Vargas,17 G. Vasseur,5 C. Vilela,11 W.G.S. Vinning,70 T. Vladisavljevic,58 T. Wachala,14 A.V. Waldron,46,𝑘 J.G. Walsh,34 Y. Wang,43 L. Wan,3D. Wark,58,46 M.O. Wascko,21 A. Weber,18 R. Wendell,33,𝑑 M.J. Wilking,43 C. Wilkinson,35 J.R. Wilson,31 K. Wood,35 C. Wret,51 J. Xia,61 Y.-h. Xu,34 K. Yamamoto,45,𝑖 C. Yanagisawa,43,𝑙 G. Yang,43 T. Yano,60 K. Yasutome,33 N. Yershov,26 U. Yevarouskaya,57 M. Yokoyama,59,𝑑 Y. Yoshimoto,59 M. Yu,74 R. Zaki,74 A. Zalewska,14 J. Zalipska,42 K. Zaremba,68 G. Zarnecki,42 X. Zhao,10 T. Zhu,21 M. Ziembicki,68 E.D. Zimmerman,6M. Zito57 and S. Zsoldos31 1Department of Theoretical Physics, University Autonoma Madrid, ES-28049 Madrid, Spain 2 Albert Einstein Center for Fundamental Physics, Laboratory for High Energy Physics (LHEP), University of Bern, Sidlerstrasse 5, Bern, Switzerland 3Department of Physics, Boston University, 590 Commonwealth Avenue, Boston, Massachusetts, U.S.A. 4Department of Physics and Astronomy, University of California, Irvine, 4129 Frederick Reines Hall, Irvine, California, U.S.A. 5IRFU, CEA, Université Paris-Saclay, F-91191 Gif-sur-Yvette, France 6Department of Physics, Univeristy of Colorado at Boulder, 2000 Colorado Avenue, Boulder, Colorado, U.S.A. 7Department of Physics, Colorado State Univeristy, 400 Isotope Drive, Fort Collins, Colorado, U.S.A. 8Department of Physics, Duke University, 120 Science Drive, Durham, North Carolina, U.S.A. 9Department of Atomic Physics, Eötvös Loránd University, Pázmańy Péter sétány 1/A, Budapest, Hungary 10 Institute for Particle Physics and Astrophysics, ETH Zurich, Otto-Stern-Weg 5, Zurich, Switzerland 11 CERN European Organisation for Nuclear Research, CH-1211, Genéve 23, Switzerland 12 University of Geneva, section de Physique, DPNC, 24 rue Ernest-Ansermet, CH-1211 Genéve 4, Switzerland 13 School of Physics and Astronomy, University of Glasgow, Glasgow G12 8QQ, United Kingdom 14 H. Niewodniczanski Institute of Nuclear Physics PAN, ul. Radzikowskiego 152, Cracow, Poland 15 KEK, High Energy Accelerator Research Organization, 1-1 Oho, Tsukuba, Japan 16 Department of Physics, University of Houston, 3507 Cullen Boulevard, Houston, Texas, U.S.A. 17 Institut de Física d’Altes Energies (IFAE) — The Barcelona Institute of Science and Technology, Campus UAB, Bellaterra (Barcelona), Spain 18 Institut für Physik, Johannes Gutenberg-Universität Mainz, Staudingerweg 7, 55128 Mainz, Germany 19 IFIC (CSIC & University of Valencia), Catedrático José Beltrán, Valencia, Spain 20 Institute For Interdisciplinary Research in Science and Education (IFIRSE), ICISE, 7 Science Avenue, Gheng Rang Ward, Quy Nhon, Vietnam 21 Department of Physics, Imperial College London, South Kensington Campus, London, United Kingdom 22 Dipartimento Interuniversitario di Fisica, INFN Sezione di Bari and Università e Politecnico di Bari, via Edoardo Orabona 4, Bari, Italy 23 Dipartimento di Fisica, INFN Sezione di Napoli and Università di Napoli, Complesso Universitario di Monte Sant’Angelo, via Cinthia 6, Napoli, Italy 24 Dipartimento di Fisica, INFN Sezione di Padova and Università di Padova, via Francesco Marzolo 8, Padova, Italy 25 INFN Sezione di Roma and Università di Roma “La Sapienza”, Piazzale Aldo Moro 2, Roma, Italy 26 Institute for Nuclear Research of the Russian Academy of Sciences, prospekt 60-letiya Oktyabrya 7a, Moscow, Russia 27 International Centre of Physics, Institute of Physics (IOP), Vietnam Academy of Science and Technology (VAST), 10 Dao Tan, Ba Dinh, Hanoi, Vietnam 28 International Laboratory for Astrophysics, Neutrino and Cosmology Experiments, University of Tokyo, 5-1-5 Kashiwa-no-Ha, Kashiwa, Japan 29 Kavli Institute for the Physics and Mathematics of the Universe (WPI), The University of Tokyo Institutes for Advanced Study, University of Tokyo, 5-1-5 Kashiwa-no-Ha, Kashiwa, Japan 30 Department of Physics, Keio University, 3-14-1 Hiyoshi, Kanagawa, Japan 31 Department of Physics, King’s College London, Strand, London WC2R 2LS, United Kingdom 32 Department of Physics, Kobe University, 1-1 Rokkodai, Nada-ku, Kobe, Japan 33 Department of Physics, Kyoto University, Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto, Japan – 32 – 2022 JINST 17 P10028 34 Physics Department, Lancaster University, Lancaster LA1 4YB, United Kingdom 35 Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, U.S.A. 36 Laboratoire Leprince-Ringuet, IN2P3-CNRS, Ecole Polytechnique, Rue de Fresnel, Palaiseau, France 37 Department of Physics, University of Liverpool, Oxford Street, Liverpool L69 7ZE, United Kingdom 38 Department of Physics and Astronomy, Louisiana State University, 202 Nicholson Hall, Baton Rouge, Louisiana, U.S.A. 39 Joint Institute for Nuclear Research, Dubna, Moscow Region, Russia 40 Department of Physics and Astronomy, Michigan State University, 567 Wilson Road, East Lansing, Michigan, U.S.A. 41 Department of Physics, Miyagi University of Education, 149 Aramaki-aza-Aoba, Aoba-ku, Sendai, Japan 42 National Centre for Nuclear Research, ul. Andrzeja Sołtana 7, 05-400 Otwock, Warsaw, Poland 43 Department of Physics and Astronomy, State University of New York at Stony Brook, Nicolls Road, Stony Brook, New York, U.S.A. 44 Department of Physics, Okayama University, 3-1-1 Tsushima-naka , Kita-ku, Okayama, Japan 45 Department of Physics, Osaka City University, 1-1 Machikaneyama, Toyonaka, Osaka, Japan 46 Department of Physics, Oxford University, Keble Road, Oxford OX1 3RH, United Kingdom 47 Department of Physics and Astronomy, University of Pennsylvania, 209 South 33rd Street, Philadelphia, PA 19104, U.S.A. 48 Department of Physics and Astronomy, University of Pittsburgh, 3941 O’Hara Street, Pittsburgh, Pennsylvania, U.S.A. 49 Department of Physics, Queen Mary, University of London, Mile End Road, London E1 4NS, United Kingdom 50 Department of Physics, University of Regina, 3737 Wascana Parkway, Regina, Saskatchewan, Canada 51 Department of Physics and Astronomy, University of Rochester, 206 Bausch and Lomb Hall, Rochester, New York, U.S.A. 52 Department of Physics, Royal Holloway, University of London, Egham Hill, Egham, Surrey TW20 0EX, United Kingdom 53 III. Physikalisches Institut, RWTH Aachen University, Otto-Blumenthal-Straße, Aachen, Germany 54 Departamento de Física Atómica, Molecular y Nuclear, Universidad de Sevilla, 41080 Sevilla, Spain 55 Department of Physics and Astronomy, University of Sheffield, Hounsfield Road, Sheffield S3 7RH, United Kingdom 56 Institute of Physics, University of Silesia, ul. 75 Pułku Piechoty 1, Katowice, Poland 57 Laboratoire de Physique Nucléaire et de Hautes Energies (LPNHE), CNRS/IN2P3, Sorbonne Université, Université Paris Diderot, 4 place Jussieu, Paris, France 58 STFC, Rutherford Appleton Laboratory, Harwell Oxford, and Daresbury Laboratory, Warrington, United Kingdom 59 Department of Physics, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan 60 Institute for Cosmic Ray Research, Kamiokaka Observatory, University of Tokyo, Higashi-Mozumi 456, Kamioka, Japan 61 Institute for Cosmic Ray Research, Research Center for Cosmic Neutrinos, 5-1-5 Kashiwa-no-Ha, Kashiwa, Japan 62 Department of Physics, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo, Japan 63 Department of Physics, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji-shi, Tokyo, Japan 64 Department of Physics, Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba, Japan 65 Department of Physics, University of Toronto, 60 Saint George Street, Toronto, Ontario, Canada 66 TRIUMF, 4004 Wesbrook Mall, Vancouver, British Columbia, Canada 67 Faculty of Physics, University of Warsaw, ul. Pasteura 5, Warsaw, Poland 68 Institute of Radioelectronics and Multimedia Technology, Warsaw University of Technology, ul. Nowowiejska 15/19, Warsaw, Poland 69 Department of Physics, Faculty of Science, University of Tohoku, 6-3 Aramaki-aza-Aoba, Aoba-ku, Sendai, Japan 70 Department of Physics, University of Warwick, Coventry CV4 7AL, United Kingdom 71 Department of Physics, University of Winnipeg, 515 Portage Avenue, Winnipeg, Manitoba, Canada 72 Faculty of Physics and Astronomy, Wroclaw University, pl. Maxa Borna 9, Wroclaw, Poland – 33 – 2022 JINST 17 P10028 73 Department of Physics, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama, Japan 74 Department of Physics and Astronomy, York University, 4700 Keele Street, Toronto, Ontario, Canada ∗Corresponding author 𝑎Also at INFN-Laboratori Nazionali di Legnaro 𝑏Also at J-PARC, Tokai, Japan 𝑐Now at San José State University, San José, California, U.S.A. 𝑑Affiliated member at Kavli IPMU (WPI), the University of Tokyo, Japan 𝑒Also at Moscow Institute of Physics and Technology (MIPT), Moscow region, Russia and National Research Nuclear University "MEPhI", Moscow, Russia 𝑓Also at IPSA-DRII, France 𝑔Also at the Graduate University of Science and Technology, Vietnam Academy of Science and Technology ℎAlso at JINR, Dubna, Russia 𝑖Also at Nambu Yoichiro Institute of Theoretical and Experimental Physics (NITEP) 𝑗Now at California Institute of Technology, Pasadena, California, U.S.A. 𝑘Now at Imperial College London, London, United Kingdom 𝑙Also at BMCC/CUNY, Science Department, New York, New York, U.S.A. – 34 –