A BaF2 calorimeter for Mu2e-II

Transcription

1 A BaF2 calorimeter for Mu2e-II I. Sarra, on behalf of LNF group Università degli studi Guglielmo Marconi Laboratori Nazionali di Frascati NEWS General Meeting March 218

2 Proposal (1) q This technological development is driven by the aim of improving the state-of-the-art calorimetry in search for the conversion of a muon into an electron. ü The conversion process is forbidden in the Standard Model of Particle Physics and its observation will be a clear evidence for new physics. ü Two international experiments are already in construction for this search, one in Japan (COMET at JParc) and one in USA (Mu2e at Fermilab), with the goal of improving the previously reached sensitivity by four orders of magnitude. 1 I. NEWS General Meeting March 218

3 Proposal (2) q Our proposal finds a natural framework and timeline in the upgrade of the Mu2e experiment, Mu2e-II, that will increase of a factor of 1 the sensitivity allowing to observe signals with a branching fraction as small as 6x1-18. Figure 1: Λ versus κ sensitivity plots for the CLFV muon channels. Mu2e is a potential discovery experiment that is relevant in all possible scenarios. 2 I. NEWS General Meeting March 218

4 Proposal (3) q To achieve this goal, a very intense, pure negative muon beam has to be stopped on a thin target, at the high rate of 3 GHz, inside an evacuated detector region and in presence of 1 T axial magnetic field. ü A very fast calorimeter, with high timing resolution and extremely high rate capability, can be achieved by optically connecting the extremely fast barium fluoride (BaF 2 ) crystals to our proposal of a new generation UV extended Silicon Photomultipliers. ü To do this, we will take full advantage of the fast 22 nm scintillation component (t =.9 ns) strongly reducing the larger slow component at 3 nm (t = 65 ns) while preserving high gain, low noise and radiation hardness. 3 I. NEWS General Meeting March 218

5 State of the art q The existence of the very fast scintillation component in BaF 2 makes this crystal an attractive, if not the best, candidate for high-rate applications to the upgrade phase of Mu2e calorimeter. ü A lot of research and development on UV extended, solar blind, Avalanche Photodiodes (SB-APD) has been carried out from a consortium of Caltech/JPL/RMD in USA. ü The SB-APD works in the proportional regime (gain between 1-1), it is a really promising technology and will work properly in magnetic field but presents three disadvantages that we can overcome with our proposal: (i) the gain is too low thus requiring a high amplification in the front-end stage; (ii) the signal has a rise-time of 15 ns but a quenching time of about 3 ns; (iii) the related noise is too high to operate at room temperature. 4 I. NEWS General Meeting March 218

6 R&D Approach q The R&D approach is to transform a standard large-area (6 6 mm 2 ) SensL J-series SiPM into a UV device using a thin layer of nanoparticles. ü The possibility of wavelength shifting using nanoparticles was first tested by ANL in collaboration with the University of Illinois at Urbana- Champaign investigating the properties of Si nanoparticles. test of Si nanoparticles 5 nm steps from 25 nm -> 4 nm Figure 2: (Left) Response of SiPM through uncoated plastic film (squares), and through plastic film with Si nanoparticle coating (dots). (Right) Ratio between the response of the coated/uncoated configurations. 5 I. NEWS General Meeting March 218

7 SensL s J-Series with fast output The decay time, τ SPE, of a SiPM is determined by the quench resistor values and the total capacitance of the microcells, is typically in the range of tens of nanoseconds: τ "#$ = C ( R + + R -./1 N : q SensL has developed a SiPM with a third electrode coupled to individual diodes through low capacitance. As a result, the response signal has a very short pulse width. The capacitance of the third electrode toward the other SiPM electrodes is ~1% of C SiPM 6 I. NEWS General Meeting March 218

8 Nanoparticle Type LaYO q The constraint for the Mu2e-II experiment is to discriminate between the 22 nm (fast emission) and the above 3 nm (slow emission) components of the BaF 2 emission spectra. Ø LaYO nanoparticle is a good candidate for BaF2 Readout Note that the 22 nm emission of BaF2 is contained in the absorption peak of the LaYO nanoparticles Very little absorption for wavelengths >25 nm 7 I. NEWS General Meeting March 218

9 Working progress The major foreseen risk is to not reach an appropriate level of filtering for the BaF 2 slow component. Nonetheless, should this case occur, the lesson learned with the first prototypes will suggest us what to improve in order to produce a better filter in short time. The backup plan is to apply another level of filtering to the our device. A lower technical risk is related to the difference between the emission and the absorption times of the WLS process. Indeed the procedure must preserve the fast emission time of UV light from crystals (.9 ns for the barium fluoride). Our goal is to keep this time difference below few ns. 8 I. NEWS General Meeting March 218

10 First test with BaF2 and SiPM at LNF To evaluate the time resolution of the complex BaF2 crystal + SiPM, we have used an UV-sensitive MPPC from Meg. 9 I. NEWS General Meeting March 218

11 Performance W. Ootani et al., Development of Deep-UV Sensitive MPPC for LXe Scintillation Detector, NDIP214 Gain >5 15 (for single segment chip) 2 Dark count rate ~1 Hz/mm Measured performances for single cell are for ΔV all numbers Excellent PDE (>15%) is achieved. Correlated noise probability Largest (~1%) in estimation of expe After-pulsing: Voltage = Vbr+2.5, G = 8 x 15<1% PDE =uncertainties 2% #488-chip1 #488-chip2 #488-chip3 #488-chip4 #489-chip1 #489-chip2 #489-chip3 #489-chip p 2 I. NEWS General Meeting 218 y r a n i m i l re Over voltage [V] Probability of crasstalk+afterpuse [%] Gain 13 Photon detection efficiency [%] photons impinging MPPC (geometrical acceptance Optical crosstalk ~35% reflection on surrounding materials) PDE Gain Correlated noise 6 3 #488-chip1 #488-chip2 #488-chip3 #488-chip4 #489-chip1 #489-chip2 #489-chip3 #489-chip4 5 4 #488-chip1 #488-chip2 #488-chip3 #488-chip4 #489-chip1 #489-chip2 #489-chip3 #489-chip4 3 ryina am n i im i prelprel March Over Over voltage [V] volta

12 Measured performance for series 11 I. NEWS General Meeting March 218

13 8 8 Trasmittance Test of CsI/BaF2 + Meg MPPC with CR CsI Italiano Opto Materials CsI Russo ISMA CsI 55 SICCAS 6Riferim 65 7 Wavelength (nm) 2 Then we have setup a cosmic rays test station and made the comparison Timing Resolution between the performance of the CsI crystals (used in Mu2e) and the BaF2 Cosmic Rays (CRs) crystals Exploiting (proposed for Mu2e-II) Wavelength (nm) Timing Resolution Crystals between two scintillation counters Experimental setup MPPC readout Exploiting Cosmic Rays (CRs) Different wrapping materials Coupling both withtwo andscintillation without optical Crystals between counters grease MPPC readout Different wrapping materials Coupling both with and without optical Distance from PMT [cm] grease!!!!!!! 1 Analysis thecnique Fit function -> pol4 Fit range: (.1 85)% of the max amplitude Constant fraction method ± Distance from PMT [cm] q 1.52 ± ±.42 Constant fraction method used q 1.52 ±.4174 ps. Entries / 1 ps Entries 6 Constant ± 3.6 Constant fraction method used 12 1 Mean.9891 ± Sigma.3285 ± I. NEWS General Meeting March 218

14 Signals shape - CsI Entries Mean Mean y RMS RMS y Underflow 4 Overflow I. NEWS General Meeting March 218

15 Waveform CsI Amplitude [mv] Amplitude [mv] p 4.99e+4 ± p1 273 ±.24 p ±.2439 p ± 2.871e 5 p ± 3.254e Finger Crystal 2ns p 138 ± p ±.1713 p ±.211 p3.11 ± 2.35e 5 p e 5 ± 2.336e Amplitude [mv] Amplitude [mv] Crystal 1ns p ± p ±.128 p ±.1567 p ± 1.76e 5 p e 6 ± 1.77e Crystal 4ns p 5.512e+4 ± 12.9 p ±.1918 p2.346 ±.2395 p ± 2.661e 5 p4.11 ± 2.525e Amplitude [mv] I. NEWS General Meeting March 218

16 Signals shape - BaF Entries Mean Mean y 81.3 RMS RMS y Underflow Overflow I. NEWS General Meeting March 218

17 Waveform BaF 2 Amplitude [mv] Amplitude [mv] p 5.465e+4 ± 15.9 p ±.1958 p ±.2327 p ± 2.676e 5 p ± 2.969e p 4232 ± p1 p ± ± 2.75e 5 p2 p4.728 ± ± 2.853e 7 Finger Crystal 2ns Amplitude [mv] 35 p 1.111e+4 ± p1 p ± ± 2.88e 5 p2 p ± ± 2.22e Amplitude [mv] p 1.27e+4 ± p1 p ± ± 2.913e 5 p2 p ± ± 2.811e Crystal 1ns Crystal 4ns Amplitude [mv] I. NEWS General Meeting March 218

18 CsI Time Resolution 1 Gsps/s Fingers Time resoltion CsI + MPPC time resolution N. Entries / 2 ps Entries 831 Constant ± 5.7 Mean.6727 ±.93 Sigma.2498 ±.7 N. Entries / 1 ps Entries 831 Constant ± 3.98 Mean.1381 ± Sigma.5549 ± I. NEWS General Meeting March 218

19 BaF2 Time Resolution 1 Gsps/s ü The first results are really encouraging Fingers Time resoltion BaF2 + MPPC time N. Entries / 2 ps Entries 831 Constant ± 5.7 Mean.6727 ±.93 Sigma.2498 ±.7 N. Entries / 1 ps resolution Entries 1328 Constant ± 8.7 Mean.1699 ±.826 Sigma.2967 ± I. NEWS General Meeting March 218

20 Conclusions The framework and timeline of the project find a natural application in the calorimeter upgrade for the phase-ii of the Mu2e experiment (Mu2e-II) at Fermilab. The requirement for this innovative silicon photosensor is that of being able to readout the fast component (22 nm) of the light emitted by the BaF2 crystals while achieving practical blindness to the slow component (> 3 nm). Using these sensors and the BaF2 crystals, we aim to build a radiation hard calorimeter with good energy resolution and extremely high performance in timing resolution, rate capability and pileup discrimination power for 1 MeV electrons. All of the above has to be achieved in the presence of a strong magnetic field (1 T), in a radiation hard environment and with 1 GHz muon beam impinging on a thin target. 19 I. NEWS General Meeting March 218