Exposing a dual-readout fibre calorimeter to electron beams, in preparation for HiDRa: High-resolution calorimeter for colliders

Contents lists available at ScienceDirect Nuclear Inst. and Methods in Physics Research, A journal homepage: www.elsevier.com/locate/nima Exposing a dual-readout fibre calorimeter to electron beams, in preparation for HiDRa: High-resolution calorimeter for 𝑒+𝑒−colliders Andrea Pareti 1 Università degli Studi di Pavia, Via Bassi 6, 27100, Pavia, Italy and INFN ARTICLE INFO Keywords: Calorimetry Silicon photomultiplier Detector R&D ABSTRACT Among the proposed accelerators for the post-LHC era are the Future Circular Collider (FCC) and the Circular Electron–Positron Collider (CEPC). Both projects would take advantage of the clean environment and low radiation damage of electron–positron collisions, providing the highest precision studies of electroweak and Higgs physics. The relatively high branching fraction of W, Z and Higgs bosons into hadronic jets makes the precise measurements of these objects an essential aspect on detector development for experiments at future collider facilities. One of the most promising calorimetry techniques for improved hadron energy reconstruction is the dual-readout method, which exploits signals from two different physics processes to correct for the 𝑓𝑒𝑚 of hadron showers, therefore boosting the standalone calorimeter performance. The usage of compact photosensors (e.g. Silicon PhotoMultipliers) for the readout enables a very fine segmentation, opening to particle-flow and advanced neural networks software reconstruction of events. In this contribution, the results of testing a small-scale dual-readout calorimeter prototype, characterised by a mixed PMT and SiPM readout solution, with an electron beam at the CERN SPS facility in the energy range [10, 120] GeV are presented. It follows the description of a larger prototype of the same type, currently under construction in the context of the INFN HiDRa project, that will be large enough to fully contain hadron showers. The design, construction technique and expected performance, as estimated through a Geant4 simulation parameterised on the previous prototype, are presented. 1. Introduction Excellent measurements of hadron jet energies are one of the open issues of currently operating collider experiments. These object are inherently complicated by the high multiplicity of low energy hadrons, each concurring with fluctuations on the electromagnetic fraction 𝑓𝑒𝑚 and on the invisible energy, which are the main limiting factors on the energy measurement for single hadrons. One of the major challenges on detector development for future 𝑒+𝑒−colliders, the FCC or CEPC, is the capability to recognise the Higgs-related processes also in completely hadronic final states (mainly 𝑒+𝑒−→𝑍𝐻 →4𝑗) from the 𝑒+𝑒−→ 𝑊+𝑊−∕𝑍𝑍 →4𝑗background. To achieve this, the W and Z bosons are required to be statistically separated, based on the invariant mass of the two pairs of hadron jets. This identification requirement poses a strict benchmark on calorimetric performances: an energy resolution of 4–5% for 50 GeV jets, or equivalently 𝜎∕𝐸= 30%∕√𝐸, are usually reported as target. The dual-readout method [1] aims to reach this resolution by measuring signals produced by active materials with different (ℎ∕𝑒)responses to the electromagnetic and non-electromagnetic components of E-mail address: [email protected]. 1On behalf of the ECFA-DRD6 Collaboration, dual-readout calorimeter working group. hadron showers, correcting for the 𝑓𝑒𝑚 on event-by-event basis. In most cases, Čerenkov light emission is used, alongside with a scintillating active material: Čerenkov light in hadron showers is mostly emitted by electrons and positrons, therefore providing a measurement of the electromagnetic component. The advantages of this method are that, once the correction for 𝑓𝑒𝑚 has been applied, the same energy scale for both electromagnetic and hadronic showers can be used, and the reconstructed energy distribution is gaussian-shaped. The IDEA detector, which was included in both the FCC and CEPC Conceptual Design Reports, uses a dual-readout calorimeter based on optical fibres. The fibres in such a calorimeter, parallel to the radial direction and pointing to the interaction point, are not longitudinally segmented, and play the role of both the active material and wave guide for signals extraction from inside the calorimeter. A few early results from the DD4hep simulation of this calorimeter can be found in [2]. A different detector layout, consisting of a crystal-based dual-readout calorimeter section for enhanced measurements of electromagnetic showers in front of the hadronic one, has also been proposed. Fast photosensors with single photon detection capabilities (SiPMs) would be used for the readout of https://doi.org/10.1016/j.nima.2024.169890 Received 29 June 2024; Received in revised form 5 August 2024; Accepted 15 September 2024 Nuclear Instruments and Methods in Physics Research A 1069 (2024) 169890 Available online 18 September 2024 0168-9002/© 2024 The Author. Published by Elsevier B.V. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). A. Pareti Fig. 1. Side view of the calorimeter prototype, and sketch of the beam instrumentation setup for the 2021 test beam, taken from [3]. each optical fibre, allowing for a fine lateral segmentation useful for particle-flow and neural network applications. The timing information provided by SiPMs, with an appropriate readout electronics, could be exploited to obtain information on photon production depth inside the optical fibres, hence partially recover the longitudinal segmentation. 2. Electromagnetic-shower-size prototype In order to demonstrate the feasibility of single fibre readout with SiPMs associated to a dual-readout calorimeter, a first small-scale prototype able to contain electron-induced showers was built, shown in Fig. 1. The overall dimensions of this prototype are 10 ×10 ×100 cm3. Brass capillary tubes with an outer diameter of 2 mm and an inner one of 1.1 mm are used as absorber. The active material consists of alternating rows of scintillating and clear optical fibres for collecting Čerenkov light. The fibres, 1 mm diameter, are inserted into the brass tubes. The remaining volume of the calorimeter, between the capillaries, is filled with the glue deposited during the prototype construction procedure and, eventually, air. A modular design was selected for the prototype assembly, with each module consisting of 20 rows of tubes and 16 tubes per row. The whole prototype consists of 9 modules, with 320 tubes each. Concerning the readout, a mixed PMT and SiPM solution is used. The central module, where the test particle beam is impinging, has a single-fibre, SiPM-based (Hamamatsu S14160-1315PS) readout for in-depth granularity studies. The remaining eight modules on the perimeter are equipped with two PMTs each, Hamamatsu R8900 for scintillating and R8900-100 with extended UV range for Čerenkov channels, respectively. This prototype was firstly exposed to a positron beam at the DESY laboratories (Germany) for a first equalisation and calibration setup. It was then taken twice on the H8 beamline at the CERN SPS North Area for a beam test with positrons at energies up to 120 GeV, in 2021 and then 2023. A few results on the two SPS test beams are reported in the following, while more details on the first H8 beamline results are presented in [3]. The beam test setup included two Čerenkov counters for positron selection out of muon and hadron components upstream in the beamline, followed by two Delay Wire Chambers (DWCs) to estimate the impact point position on the calorimeter front face. A pre-shower detector was also used for complementary particle identification purposes. Three scintillator plates located before the calorimeter were used for the triggering. A fourth scintillator plate was also placed after the calorimeter for muon identification. A few problems were encountered during this first beam test: due to access limitations in the experimental area, the pre-shower detector could only be placed very far from the calorimeter, a few metres upstream the beamline, causing non-optimal containment of electromagnetic showers inside the limited prototype volume. The calorimeter was also only tilted 1◦on the horizontal direction with respect to the beamline axis. The small tilting angle resulted in a Fig. 2. Lateral profile for positron-originated showers, measured through the highly granular core of the calorimeter prototype. It shows the fraction of energy deposited inside a radius 𝑟from the shower barycentre, estimated through the centre-of-gravity method. Source: Taken from [3]. dependence of the overall signal on the beam impact position, due to channelling effects of particles inside the optical fibres, and to the very small lateral size of electromagnetic showers, that deposit most of their energy in a few fibres. With an insufficient rotation of the calorimeter, if a particle entered a fibre, only very small energy deposits were left in other fibres of the same type in different rows, due to the alternating scintillating and Čerenkov layout of the prototype. Despite these issues, several measurements could be made with this small-scale calorimeter prototype. Fig. 2 shows the electromagnetic lateral shower profile measured with 20 GeV positrons inside the highgranularity module, reaching a millimetre precision. The ratio of energy deposited from a distance 𝑟from the shower barycentre, with respect to the total energy deposit, is plotted. The behaviour is well described by the Geant4-based simulation mimicking the beam setup conditions, including the different profiles for scintillation and Čerenkov signals. In fact, in early shower development stages Čerenkov light falls outside the numerical aperture of optical fibres, due to its collimated direction with respect to incoming particles. Concerning the energy measurements, a good response in the whole [10–120] GeV energy range was observed. However, due to the limited shower containment that would have been caused by the far pre-shower detector, the energy resolution was only estimated for energies up to 30 GeV. At such low energies, sufficient hadron rejection could be done with the upstream Čerenkov counters. With this setup, a stochastic term of 17.5% was estimated [3]. Similar results were also obtained by the same Geant4 simulation, which also suggested that better results could be obtained by increasing the rotation angle of the calorimeter and therefore reducing the impact point modulation effect. A stochastic term of 14.5% was obtained just by increasing the rotation from 1◦to 2.5◦on the horizontal direction, and from 0◦to 2.5◦on the vertical one. With these new understandings, the same prototype was taken again on the same beamline during July 2023 for a more in-depth characterisation. A positron beam with improved purity was observed, and the pre-shower detector could be positioned closer to the prototype, about 15.5 mm from the calorimeter front face. A full angle scan on the horizontal direction up to ±4.5◦was performed, and data were taken with two different tilting angles (0, 2.5)◦on the vertical direction. Fig. 3 shows the signal modulation inside in the S channel (for 20 GeV positrons) dependency on impact point position, for the two considered vertical angles. As expected, increasing the rotation Nuclear Inst. and Methods in Physics Research, A 1069 (2024) 169890 2 A. Pareti Fig. 3. Preliminary results from the 2023 beam test: it shows the mean scintillating channel signal dependence over the impact point position, inside the high-granularity module with SiPM readout. A comparison using two different vertical angles is shown, using a 20 GeV positron beam. The red line shows the modulation effect, described in detail in [3]. The blue one shows that the modulation can be correctly addressed by increasing the prototype rotation angle. Fig. 4. HiDRa modules after the construction on the assembly table. A UV lamp was placed in front of the modules front face, to see the scintillating (light blue) and transparent (purple) fibres light collected at the rear face of the calorimeter. angle allows to sample the electromagnetic shower in more fibres, and remove the channelling problem, hence reducing the dependency effect. Early results from the analysis of the collected data show that a definitely improved energy resolution is in reach, despite some noisy PMT could result in higher constant term on the overall calorimeter energy resolution. Besides positrons, data with muons and positively charged pions were also taken, to estimate the response of the prototype to these particles. Data analysis is still ongoing at present. 3. HiDRa hadronic calorimeter prototype In order to assess the performance on hadrons of dual-readout calorimeters in association with SiPM readout, a prototype able to contain hadron-originated showers is under development. Similarly to the previous prototype, a modular design consisting of 80 independent modules based on steel capillary tubes, and alternating rows on scintillating and clear optical fibres for Čerenkov light measurement, is used. The same construction technique, with glue filling the volume between the tube rows, has been used. For this prototype, BCF-12 scintillating optical fibres from Saint-Gobain, guaranteeing longer attenuation lengths with respect to previous prototype, were selected. With a cross section of 65 ×65 cm2and a length of 2.5 metres (about 10.5 interaction lengths), the deposited energy for hadron showers inside the calorimeter volumes, estimated through Geant4 simulations in the [10–100] GeV energy range, is around 93%. Concerning the readout of this prototype, SiPMs will be used for the central 10 out Fig. 5. Geant4 simulation-based expected energy resolution for hadron showers originated by positively charged pions impinging on the HiDRa dual-readout prototype. of the 80 modules. Two different SiPM solutions, optimised for the different emission spectra and light yield of Čerenkov and scintillation emission, have been chosen. For scintillating fibres, a 10 μm pitch model, S16676-10(ES1) from Hamamatsu, was selected, for an improved dynamic range due to the large light yield. Clear optical fibres will conversely be coupled to a 15 μm pitch solution, S1667615(ES1) also from Hamamatsu, guaranteeing higher photon detection efficiency, to cope with the low amount of optical photons emitted through the Čerenkov process. The same Hamamatsu PMTs that were used for the previous prototype, R8900 for S channels and R8900-100 for C channels, will be then used for all other modules, for scintillating and clear fibres respectively, collecting light from 512 fibres each. With the purpose of reducing the complexity of the readout system for the high-granularity core of the HiDRa prototype, consisting of 10240 SiPMs, an analogical sum of eight adjacent SiPM signals of the same type will be implemented directly on the front-end boards. The CAEN FERS system, which was already used for the smaller electromagnetic prototype, will also be exploited for the readout of the SiPM modules: two FERS boards, able to read up to 64 channels, will be required for each module. A great effort is put on making the electronics and readout systems as compact as possible, therefore avoiding the need to fan out the fibres from the back of capillaries. The mechanical integration of SiPMs is in advanced state with dummy components, while real components design is almost finalised. The construction of the prototype started in November 2023 at the INFN laboratories in Pavia, Italy, in collaboration with other INFN groups. At the time of writing, 33 modules have been assembled, some of which are shown in Fig. 4. A first test beam with 36 out of the 80 modules of the full prototype will be carried out at the end of August 2024, only using a PMT-based readout, for a first characterisation and validation with hadron beams of the prototype’s Geant4 simulation. A few simulation-based results are reported in Figs. 5 and 6. The first one shows the expected energy resolution for single particle hadron showers, taken from a gaussian fit of the measured energy reconstructed through the dual-readout method. A linear sum of the stochastic and constant components of the relative energy resolution was used in the fit, as a better agreement was found with respect to a quadratic sum. Good results are obtained considering the shower particles leakage outside the detector volume mostly from the sides of the detector. Fig. 6 shows the expected spatial resolution for electromagnetic showers, obtained as the difference between the Geant4 simulation truth information and reconstructed shower barycentre. An excellent spatial resolution at a level of the millimetre is found in the [10, 100] GeV Nuclear Inst. and Methods in Physics Research, A 1069 (2024) 169890 3 A. Pareti Fig. 6. Geant4 simulation-based spatial resolution for positron-induced electromagnetic showers. Very similar results are also obtained for photon-induced ones. The combined resolution has been obtained by weighting for the S and C channels resolutions, maximising the component with the minimum standard deviation. energy range, thanks to the very high granularity offered by silicon photodetectors. The grouping of 8 fibres in one readout channel, that would be done through the front-end boards, has been reproduced by summing the signals on each independent SiPM. A 5 cm smearing on the longitudinal position has been added to the truth information provided by Geant4 on the optical photon production point. Such an information, that in a real case scenario would be measured through timing, is used to correct the reconstructed coordinates for the 2.5◦ calorimeter angle rotation on both horizontal and vertical directions that is used in the simulation. 4. Conclusions Experiments at future colliders will greatly benefit from precise measurements of hadron jets. The dual-readout calorimetry technique offers several advantages for the improvement of hadron energy reconstruction, while the SiPM readout of fibres allows further advanced software reconstruction techniques that are under investigation. Two dual-readout calorimeter prototypes of this kind have been presented: the first one, able to contain only electromagnetic showers, has been constructed and tested with particle beams, showing promising results. The second one, able to contain hadronic showers, is under construction and will be used to reach deep understanding of dual-readout, highly granular calorimeter design, in sight of the calorimeter construction for the IDEA experiment. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Andrea Pareti reports financial support was provided by AidaInnova. Andrea Pareti reports financial support was provided by EuroLabs. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under GA n◦ 101004761. References [1] S. Lee, M. Livan, R. Wigmans, Dual-readout calorimetry, Rev. Modern Phys. 90 (2) (2018) http://dx.doi.org/10.1103/revmodphys.90.025002. [2] A. Loeschcke-Centeno, Simulation of a capillary tube, fibre dual-readout calorimeter in DD4hep, in: Same proceedings, 2024. [3] N. Ampilogov, S. Cometti, J. Agarwala, V. Chmill, R. Ferrari, G. Gaudio, P. Giacomelli, A. Giaz, A. Karadzhinova-Ferrer, A. Loeschcke-Centeno, A. Negri, L. Pezzotti, G. Polesello, E. Proserpio, A. Ribon, R. Santoro, I. Vivarelli, Exposing a fibre-based dual-readout calorimeter to a positron beam, J. Instrum. 18 (09) (2023) P09021, http://dx.doi.org/10.1088/1748-0221/18/09/P09021. Nuclear Inst. and Methods in Physics Research, A 1069 (2024) 169890 4