logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-11182024-235337


Tipo di tesi
Tesi di laurea magistrale
Autore
BIANCO, LORENZO
URN
etd-11182024-235337
Titolo
A new calorimetry for future cLFV searches based on a large LYSO crystal coupled to MPPCs
Dipartimento
FISICA
Corso di studi
FISICA
Relatori
relatore Prof.ssa Papa, Angela
Parole chiave
  • calorimetria
  • calorimetry
  • detector
  • rivelatore
  • simulation
  • simulazione
Data inizio appello
09/12/2024
Consultabilità
Tesi non consultabile
Riassunto
This thesis aims to design, construct, and commission an electromagnetic calorimeter to detect photons with energies around 50 MeV. This calorimeter functions as a prototype for future technologies in the field of charged lepton flavor violation (cLFV) and as an auxiliary detector for calibrating the liquid xenon calorimeter (XEC) of the MEG-II experiment at the Paul Scherrer Institute (PSI).

The prototype consists of a large cylindrical scintillator crystal (h = 10 cm, Ø 8.5 cm) made of LYSO, with 115 x 2 Multi-Pixel Photon Counters (MPPCs) Hamamatsu S13360-6025PE attached to its front and back faces for signal readout. LYSO was chosen for its properties, which make it ideal for high energy and timing resolutions, while the high “granular” readout using dual photo-sensor panels offers enhanced spatial resolution compared to traditional crystal + PMT systems. This design is particularly suited to cLFV applications, where pile-up from high-intensity beams must be addressed, as well as the position and timing evaluation of the impinging radiation to complement the energy measurement, with a single detector.

The project focuses on two major aspects related to the detector itself: the photosensor characterization (hardware) and the extraction of the expected detector performances based on a detailed MC simulation, which includes the radiation-matter interaction up to the waveforms, and the kinematical variables reconstruction algorithms (software).

On the hardware side, an in-depth characterization of the selected MPPC photosensor model was conducted at PSI. By acquiring dark noise spectra across different temperatures, the gain as a function of bias voltage was determined, allowing extraction of the breakdown voltage and measurement of the dark rate as a function of overvoltage for various working points. A mass characterization of all 300 available photosensors was also conducted, coupling them with small plastic scintillator tiles (7 x 7 x 0.5 mm3) and exposing them to a Sr-90 source, ensuring proper functionality.

On the software side, a significant portion of this thesis involved developing a stand-alone simulation chain, with the ultimate goal of integrating these modules into the experiment’s software. Specifically, a full Monte Carlo simulation of the detector was implemented (using Geant4), along with a digitizer (using ROOT libraries) to model the response of photosensors and readout electronics, reproducing output waveforms. These waveforms are constructed by summing single-photoelectron pulses, with templates that are user-selectable based on MPPC characterization data. This enables simulations tailored to any desired working point. Analysis algorithms were also developed to reconstruct key kinematic variables (energy, time, and position) from the waveforms, allowing identical methods to be applied to both experimental and simulation data.

Finally, the best estimators and the expected resolutions were evaluated through a full simulation of the XEC calibration, where a charge exchange reaction produces a 55 MeV gamma ray to be detected in the LYSO detector. This energy lies precisely in the $\mu^+ \to e^+ \gamma$ signal region. From this analysis, it was demonstrated that, without imposing acceptance cuts (which would improve resolution at the cost of reduced efficiency), the expected energy resolution is $\sigma_E / E \simeq 3.7\%$.

Furthermore, it was shown that with an efficient Monte Carlo reproduction, it is possible to achieve timing and position resolutions of $\sigma_t \simeq 60 \, \text{ps}$ and $\sigma_{x,y} \simeq 3.5 \, \text{mm}$, respectively.
File