AIDA-2020 Advanced European Infrastructures for Detectors at Accelerators. Presentation

Transcription

1 AIDA-22-SLIDE AIDA-22 Advanced European Infrastructures for Detectors at Accelerators Presentation Energy Resolution and Timing Performance Studies of a W-CeF3 Sampling Calorimeter prototype with a Wavelength-Shifting Fiber Readout Nessi-Tedaldi, Francesca (ETHZ) 24 May 217 The AIDA-22 Advanced European Infrastructures for Detectors at Accelerators project has received funding from the European Union s Horizon 22 Research and Innovation programme under Grant Agreement no This work is part of AIDA-22 Work Package 14: Infrastructure for advanced calorimeters. The electronic version of this AIDA-22 Publication is available via the AIDA-22 web site < or on the CERN Document Server at the following URL: < Copyright c CERN for the benefit of the AIDA-22 Consortium

2 Energy Resolution and Timing Performance Studies of a W-CeF3 Sampling Calorimeter prototype with a Wavelength-Shifting Fiber Readout Francesca Nessi-Tedaldi on behalf of the W-CeF3 R&D group N. Akchurin j, R. Becker a, B. Betev a, L. Bianchini a, L. Brianza b, V. Candelise e,h, F. Cavallari c, N. Chiodini b,i, I. Dafinei c, G. Della Ricca e,h, D. del Re c,g, M. Diemoz c, G. D Imperio c,g, G. Dissertori a, L. Djambazov a, M. Donegà a, M. Droege a, M. Fasoli b,i, J. Faulkner j, S. Gelli c,g, A. Ghezzi b, P. Govoni b, C. Haller a, U. Horisberger a, Th. Klijnsma a, W. Lustermann a, A. Marini a, A. Martelli b,f, D. Meister a, F. Micheli a, P. Meridiani c, V. Monti d, F. Nessi-Tedaldi a, M. Nuccetelli c, G. Organtini c,g, F. Pandolfi a, R. Paramatti c,g, N. Pastrone d, F. Pellegrino c, M. Peruzzi a, S. Pigazzini b,f, M. Quittnat a, S. Rahatlou c,g, C. Rovelli c, F. Santanastasio c,g, M. Schönenberger a, L. Soffi c,g, T. Tabarelli de Fatis b,f, P. Trapani d, F. Vazzoler e,h, A. Vedda b,i a) INFN: b) Sez. Milano-Bicocca, c) Sez. Roma, d) Sez. Torino, e) Sez. Trieste f) Università di Milano-Bicocca g) Università di Roma La Sapienza h) Università di Trieste i) Dip. Scienza dei Materiali, Università di Milano-Bicocca Int. Conf. on Technology and Instrumentation in Particle Physics 217 Beijing, May 24 th, 217

3 Francesca Nessi-Tedaldi 2 Original motivation: HL-LHC radiation environment in CMS HL-LHC (~225): a harsh environment for electromagnetic calorimetry (ECAL) γ dose rate up to 5 Gy/h, dose up to ~1 MGy (at η = 3) Hadron fluences up to ~4x1 14 cm -2 (average energy a few GeV) Neutron fluences up to ~5x1 15 cm -2 (average energy 1 MeV) Radiation-induced transparency losses in PbWO4 resulting in an energy resolution degradation Upgrade for HL-LHC: complete replacement of ECAL endcaps + partial ECAL barrel upgrade Hadron fluence [cm -2 ], 14 TeV pp, 1 fb -1 S/S 1 MC simulation ECAL Barrel ECAL endcap Fraction of ECAL response PbWO4 1 fb -1, 5E+33 cm -2 s -1 1 fb -1, 1E+34 cm -2 s -1 5 fb -1, 2E+34 cm -2 s -1 1 fb -1, 5E+34 cm -2 s -1 2 fb -1, 5E+34 cm -2 s -1 3 fb -1, 5E+34 cm -2 s -1 CMS Coll., CMS DP-213/ CMS Coll., CERN-LHCC η

4 Strategy: a simple geometry Use an inorganic scintillator that is adequately radiation-tolerant Build a sampling calorimeter Extract the light by WLS fibers running along depolished chamfers F. N.-T. et al., CALOR 214, JoP Conf. Ser. 587 (215) 1239 chamfer minimising the machining and construction complexity, thus saving on costs minimising the light path, thus reduces radiation damage effects crystal Photodetector optimising the Molière radius, thus cell size, for pile-up mitigation Two setups: Single-channel prototype energy resolution, uniformity 5 x 3 channel matrix angular dependence Fibre Francesca Nessi-Tedaldi 3

5 CeF3 crystal scintillator Density [g/cm 3 ] 6.16 Refractive index 1.62 Peak luminescence [nm] 34 Decay time [ns] ~3 dly/dt [%/ o C].14 Ionising radiation: Can be made to recover Studied in the 9 for CMS 1) Studies performed on new crystals from Tokuyama, Japan 2) Hadron fluences: Can be made to recover Proven on 2 y old crystal 3) UV fast Transmittance [%] no build-up Tokuyama CeF 3 transmission and 6Co LY spectrum (inset) after γ-irradiation Wavelength [nm] ] -1 (34 nm) [m IND µ 4 2 Counts before and after overlaid! after 8kGy@42 Gy/h Induced absorption coefficient recovery after p-irradiation CeF 3 Recovered.4 hours Before 1.2hours 2.7hours 5.3hours 6.7hours 8.hours 5days Pulse Height [ch] p - irradiated 1) E. Auffray (CERN) et al., NIM A 383 (1996) ) F. N.-T. et al., SCINT 215, Berkeley (USA) Francesca Nessi-Tedaldi t [days] 3) G. Dissertori et al., NIM A 622 (21)

6 Francesca Nessi-Tedaldi 5 Single-channel prototype Dimensions chosen within CMS ECAL PbWO 4 crystals envelope (1 mm CeF mm W) x 15 layers = 19.5 cm = 25 X depolished chamfers, 3 mm wide, accomodate 1 mm fibers Effective R M = 23 mm, transverse dimensions 24 mm x 24 mm, SF = 38% For the tests in beam, surrounded by BGO crystals for shower containment Kuraray 3HF-SC (15) plastic fibers as WLS Each WLS fiber read out independently by a PMT Energy resolution and uniformity studies

7 Francesca Nessi-Tedaldi 6 Intrinsic energy resolution CERN SPS, October 214 H4 electron beam 2-15 GeV Trigger Scintillators smallest one 1 x 1 cm 2 Central events selection: 6 x 6 mm 2 of front face Energy resolution measured Wire Chambers and fiber Hodoscopes.5 mm impact point accuracy Single channel energy resolution dominated by lateral containment Good agreement between data and Monte Carlo The Monte Carlo extrapolation to a 5 x 5 channel matrix shows that an energy resolution stochastic term < 1% is achievable R. Becker et al., NIM A 84 (215) 79-83

8 Francesca Nessi-Tedaldi 7 W/CeF3 channel Response Uniformity Response vs impact point can be studied through precise electron tracking. Although Light Collection effects are not corrected for, we observe: Uniform response across central part of the channel Lateral non-uniformities dominated by shower non-containment Data/simulation in good agreement (within 5%) Agreement on non-central region could be improved by including the light collection in the simulation R. Becker et al., NIM A 84 (215) 79-83

9 WLS: Ce-doped quartz fibres Ce-doped photoluminescent quartz (Ce:SiO 2) is a good WLS candidate with CeF3: Ce:SiO 2 core (where light is produced) + cladding for light transport Ce:SiO 2 fibres developed for application to dosimetry 1) : Radiation hardness anticipated up to fluences >1 15 cm -2 Absorption spectrum matches CeF 3 emission Suitable as a WLS with CeF 3 Fast time response (3 ns), green emission Development of rad-hard Ce:SiO 2 in progress Normalized PLE λ max =329 nm λ em =465 nm CeF3 emission peak λ =464 nm max λ exc =3 nm Normalized PL (photons/nm) 1) A. Vedda, N. Chiodini, M. Fasoli et al., Appl. Phys. Lett., Vol. 85 (24) 6356 and priv. comm. (U. Milano Bicocca) Francesca Nessi-Tedaldi Wavelength (nm) 8

10 Francesca Nessi-Tedaldi 71 µm 9 Tests with Ce:SiO2 fibres CERN SPS H4 beam, June 215 Polymicro Kuraray A bundle of SiO 2:Ce fibres in each of 3 corners: 1 bundle from U. Milano-Bicocca 1) 71 µm 2 bundles from Polymicro/Texas Tech 2) One Kuraray 3HF plastic fibre as reference Fibers (3-2-1 configuration) micro Fibers (3-2-1 confi µm 3 3 μm 571 μm WLS efficiency is lower wrt Kuraray fibres: Fibers (3-2-1 configuration) micro Fibers (3-2-1 confi µm 3 Bicocca Polymicro Fibres exposed to 365 nm light over ~ 15 cm bundle of Ce-doped quartz fibres: factor ~1 less light than a plastic fibre smaller diameter fibres: bundle of 6 SiO2:Ce fibres in each corner Bicocca Polymicro 1 Polymicro 2 1) A. Vedda, N. Chiodini, M. Fasoli et al., Appl. Phys. Lett., Vol. 85 (24) 6356 and priv. comm. 2) Jordan Damgov, N. Akchurin et al., SCINT215, Berkeley (USA)

11 Energy resolution with Ce:SiO2 WLS fibres Central events selection: 3 x 3 mm 2 of front face Slightly different energy resolution for the different bundles of Ce:SiO2 fibres Worse resolution compared to plastic fibres, consistent with ratio of photoluminescent light yields Events / GeV Electron Beam µ = 183 σ = 6 σ/µ = 3.4 ±.1 % Events / GeV Electron Beam 45 µ = 63 σ = 39 σ/µ = 6.1 ±.2 % single channel single Kuraray fiber single channel single Ce:SiO2 bundle Amplitude [ADC Count] 1 GeV Amplitude [ADC Count] 1 GeV F. Micheli et al., IEEE NSS 215, San Diego (USA) Francesca Nessi-Tedaldi 1

12 Energy resolution vs. energy Resolution with Ce:SiO2 WLS fibres Dominated by the photoluminescent light yield, a factor ~1 lower than for Kuraray plastic fibers Higher light-yield fibres would improve the energy resolution Energy Resolution [%] Data fibre 2 S = 25.94% single channel single Kuraray fiber Electron Beam ±.98 S = (26. ± 1.)% C = 2.12% (2.1 ± ±.1)%.11 Energy Resolution [%] Data fibre 1 S = 54.58% Electron Beam ± 1.97 S = (54.6 ± 2.)% C = 2.45% (2.5 ± ±.3)%.31 single channel single Ce:SiO2 bundle Beam Energy [GeV] F. Micheli (ETH) et al., IEEE NSS 215, San Diego (USA) Francesca Nessi-Tedaldi Beam Energy [GeV] 11

13 Signal shape characteristics PMT GHz (4 ns window) full waveform acquired SiO 2:Ce fibres exhibit: WLS emission time constant typical of Cerium (folded in twice!) a fast, Cherenkov component with a rise time of a few ns (dominated by PMT response time) applications in timing measurements? Perform dedicated timing studies Signal Amplitude [mv] GeV Electron Beam W-CeF3 Kuraray WLS PRELIMINARY WLS signal Time [ns] Signal Amplitude [mv] Cherenkov PRELIMINARY 1 GeV Electron Beam W-CeF3 Ce:SiO2 WLS WLS signal Time [ns] F. Micheli (ETH) et al., IEEE NSS 215, San Diego (USA) Francesca Nessi-Tedaldi 12

14 Francesca Nessi-Tedaldi 13 Timing studies with Ce:SiO2 CERN SPS H4 beam, October 215 One blind bundle of Ce:SiO 2 fibers: black paper inserted between it and the W-CeF3 stack No WLS signal, collect just Cherenkov and direct scintillation signal from the fiber Fiber bundle read out with a Hamamatsu SiPM Reference time from MicroChannelPlate (MCP) device 1) in front of channel, which has time resolution 2-3 ps, negligible MCP 1) L. Brianza et al., Nucl. Instr. Meth. A797 (215)

15 Francesca Nessi-Tedaldi 14 Ce:SiO2 signal amplitude map Data taken with beam centred on blind Ce:SiO 2 fiber bundle, 1 x 1 cm 2 trigger Ce:SiO 2 fibers pulse amplitude map using impact point coordinates from beam hodoscope fiber region Y [mm] X [mm] Amplitude of pulses used to identify event category: Amplitude [ADC] Polymicro fiber (with cladding) Nural/Polymicro Fibers (3-2-1 configuration) 3 µm 571 µm core active cladding) (with fiber configuration) (3-2-1 Fibers Milano 3 µm 55 µm active core fiber (with cladding) configuration) (4-2 Fibers Nural/Polymicro µm 3 µm 55 core active Bicocca Kuraray Polymicro Fiber event: beam in fiber region, while signal from Kuraray fiber < threshold

16 Timing resolution fiber events Amplitude > 15 ADC PRELIMINARY The timing resolution depends on amplitude The beam energy is irrelevant Merge time resolution data for all energies and estimate the resolution for events on fiber and events on channel For amplitude > 1 ADC counts, timing resolution σt ~ 1 ps [ps] σ t 6 5 fiber events 2 GeV 15 GeV 1 GeV 5 GeV 2 GeV C = 88 ± 6 ps PRELIMINARY PRELIMINARY Amplitude [ADC] Francesca Nessi-Tedaldi 15

17 Francesca Nessi-Tedaldi 16 W-CeF3 prototype matrix 5 x 3 channel matrix built, for ultimate energy and angular resolution studies 12x(6 mm CeF mm W) 25X (= 144 mm) High granularity: effective RM = 17 mm (for pile-up rejection), transverse dimensions 17 mm x 17 mm, SF = 22% 3 mm-wide, depolished chamfers as before, to favour scintillation light escape towards WLS, dimensioned to accommodate fibres WLS Kuraray 3HF-SC fibres for readout 4 fibres signals onto one photodetector but for one inner channel, where they are read out independently APD readout, Hamamatsu S , 5 x 5 mm2, as for CMS ECAL barrel

18 Francesca Nessi-Tedaldi 17 First results CERN SPS H4 beam, June 216 Energy resolution studied for central events ( 4 x 4 mm 2 ) σ/e = (2. ±.1)% single matrix channel 1 GeV electron beam PRELIMINARY Single channel resolution 2% at 1 GeV Result scales as expected with the sampling fraction wrt single-channel prototype Single-channel stochastic term compatible with 2%/ E Considerable electronic noise would require new readout

19 71 µm 217 Tests with new Ce:SiO2 fibres CERN SPS H4 beam, June μm Polymicro Kuraray A bundle of SiO 2:Ce fibres in each of 3 corners: 1 bundle from U. Milano-Bicocca 1) 3 μm 2 bundles of type IV fibers from Polymicro/ Texas Tech 2) One Kuraray 3HF plastic fibre as reference Fibers (3-2-1 configuration) micro Fibers (3-2-1 confi µm 3 Bicocca Polymicro Fibres exposed to 365 nm light Polymicro type IV Francesca Nessi-Tedaldi Polymicro type IV fibers: different Ce-distribution (on a ring) higher WLS efficiency expected through optimised light transport 1) A. Vedda, N. Chiodini, M. Fasoli et al., Appl. Phys. Lett., Vol. 85 (24) 6356 and priv. comm. 2) N. Akchurin, this conference Visible light through fibers 18

20 Francesca Nessi-Tedaldi 19 Conclusions An innovative sampling calorimeter geometry has been built and exposed to particle beams up to 15 GeV Materials used (Cerium Fluoride, Ce-doped quartz) potentially suitable for HL-LHC running in response time, radiation hardness, signal amplitudes, granularity For cell dimensions as in present CMS ECAL (24 x 24 mm 2, RM=23 mm), and sampling fraction 38%, 5x5 energy resolution: ~1%/ E For high-granularity cell dimensions (17 x 17 mm 2, RM=17 mm) and sampling fraction 22%, analysis is in progress Timing resolution <1 ps (preliminary) Further results using new generation Ce:SiO2 fibers expected on energy resolution

21 Francesca Nessi-Tedaldi 2 Backup

22 [AD Frascati W/CeF3 results 2 CeF Cosmic runs Front Scintillator [ADC Channel / 1] (a) Correlation between ADC channel value for signals in the front scintillator and the sum of ADC channel values for the four WLS fiber signals Position [mm] (b) For a beam centered on the W-CeF 3 tower and triggering the front scintillator, beam profiles along the x- and y-axis, as determined in the beam hodoscope Energy resolution Figure 3: For 491 MeV electrons, signal distributions in beam counters. Event selection Resolution Entries / (2 ADC Channels) Francesca Nessi-Tedaldi Cosmic Data W-CeF 3 Single Tower Single photoelectron fit Q = 27.3 ±.2 1 = 8.1 ±.1 σ 1 ADC Channel ADC Channel / 1 Beam Energy [MeV] (a) For cosmic muons: pedestal-subtracted (b) For 491 MeV electrons: correlation between ADCthe spectrum sum offor ADC thechannel four WLS values Figure fiberfrom signals 5: Relative from the energy WCeFresolution 3 tower: inas black, a function of the electron beam energy for single ADC spectrum for one fiber signal Figure of4: the Summed W-CeF 3 tower. Q 1 is the fitted allposition events, in red two whenfibers single and electron the sumevents of ADC are selected, channel events. and values Filled in blue circles suchrepresent events arethe central. experimental data, compared with the expectation from of the single photoelectron peak, s 1 is its from the other two fibers, one tion entry (empty per circles). The energy resolution expected from simulation for a 5 5 matrix o width. event. towers is also shown (black squares), fitted with a function including only the stochasti 161 or more electrons can reach the setup simultaneously. energy The resolution. requirement of Eq. 3.2 allows to select Figure single 2: electrons. W-CeF 3 tower The further signals. 162 requirement of just one cluster in the hodoscope restricts the selection 163 to single electrons impinging centrally in the W-CeF 3 tower. This criterion eliminates electrons 164 hitting the tower near the edges, where the energy containment is smaller, thus minimizing the tail placed in front of the tower and aligned with its axis. For each, x- and y- direction, it consists of of the distribution at low ADC channels. 491 MeV Electron Beam 25 layers 1) of R. 8 Becker, scintillating F. fibers, N.-T. 1 et mm al., in 215 diameter, JINST staggered, 1 P72 whose signals have been added, 2 by 2, in a multi-anode Hamamatsu PMT of type R59-L16, and input to an ADC. 2.5 Readout and data acquisition system Entries / (4 ADC Channels) MeV Electron Beam Single Tower W-CeF 3 All events - Single e selection 2 Central 8x8 mm Energy Resolution [%] Resolution [%] Data (1x1) MC (1x1) MC (5x5) 9.7 % / Electron Beam E [GeV] W-CeF 3 Single Tower

23 Francesca Nessi-Tedaldi 22 Low-energy performance Frascati Beam Test Facility Bunched electron beam MeV Single channel prototype: Channel surrounded by BGO crystals Kuraray 3HF-SC (15) fibers Hamamatsu R145 PMT Front Scintillator single e- selection Hodoscope 8 x 8 mm 2 central impact Fibers read out individually Single channel energy resolution dominated by lateral containment Good agreement between data and Monte Carlo No sensitivity to constant term Monte Carlo extrapolation to 5 x 5 channel matrix shows that resolution better than 1%/ E is achievable R. Becker et al., 215 JINST 1 P72 Energy Resolution [%] Electron Beam Data (1x1) MC (1x1) MC (5x5) 9.7 % / E [GeV] W-CeF 3 Single Tower Beam Energy [MeV]