A GRAIN of SAND for DUNE: Development of simulations and reconstruction algorithms for the liquid Argon target of the SAND detector in DUNE

Neutrinos are the most abundant of known matter particles in the Universe and the study of their properties has produced many surprises, including the evidence for physics beyond the Standard Model of particle physics. In particular, the phenomenon of neutrino oscillations, transitions in flight between the different types of neutrinos, provides compelling evidence that neutrinos have small, but non-zero, masses and that their mass states are mixtures of their flavor states. The field of neutrino physics has advanced rapidly and there is currently a good understanding of most experimental results in the context of neutrino oscillations. Presently, the focus of current and future experiments has therefore shifted to making precise measurements of the oscillation parameters and trying to understand the nature of neutrino masses. The Deep Underground Neutrino Experiment (DUNE) is a future neutrino oscillation experiment that is designed to achieve the sensitivity required to finally solve long-standing open questions in neutrino physics, such as finding the ordering of neutrino mass states and measuring the possible matter-antimatter asymmetry in this sector of the Standard Model. The latter would be a very important step in our understanding of the Universe: current observations show in fact a large asymmetry, with matter strongly dominating over antimatter from the small to the large structures of the cosmos, despite our cosmological models predicting that the Big Bang should have created equal amounts of matter and antimatter. If an intrinsic asymmetry exists within the laws of physics, it would explain why the balance was broken in favor of matter, rather than antimatter. A small asymmetry has already been found in the quark sector of the Standard Model, but it is still not sufficient to explain the observations. A precise measurement of the same effect in the neutrino sector is therefore crucial. DUNE will be employing a high-intensity neutrino beam produced at the Fermi National Accelerator Laboratory (Fermilab) in Illinois (US) and directed towards an underground detector at the Sanford Underground Research Facility (SURF) in South Dakota (US). Neutrino oscillations are measured by comparing recorded neutrino events between two detectors at the near and far sites. The Far Detector will consist of a modular 68 kt liquid argon time projection chamber, by far the largest liquid argon neutrino detector ever built. Neutrinos cannot be detected directly, but charged articles resulting from their interactions in liquid argon produce two signals, vacuum ultraviolet (VUV) light from the excitation of Ar atoms and electric charge from their ionization. Both these signals are collected and used to achieve a complete reconstruction of the neutrino interaction. The Near Detector will instead be a suite of three detectors, as it needs to address some of the most important challenges of the experiment. Apart from characterizing the unoscillated neutrino beam, the precision needed to achieve DUNE goals requires to carefully constrain the systematic uncertainties of the measurement. The Near Detector is therefore expected to fulfill this task by providing corrections for extrapolating the beam to the far site, measuring ν-Ar cross-sections and tuning the interaction models. SAND (System for on-Axis Neutrino Detection) will be one of the three components of the Near Detector complex, acting as the primary beam monitor of the experiment. It will be a magnetized detector, equipped with a high resolution tracker to measure the momentum of particles by their curvature and a high performance electromagnetic calorimeter. Moreover, it will also host several different targets to study neutrino interactions on different nuclei. In particular, since the presence of Ar targets in the Near Detector is essential for comparison with the Far Detector, SAND will contain a small (1 t) liquid argon cryostat inside its magnetic volume called GRAIN (Granular Argon for Interaction of Neutrinos). GRAIN is not foreseen simply as a passive target, but it will also be instrumented to actively contribute to the reconstruction of neutrino interactions occurring inside. In principle, liquid argon provides information via scintillation light and ionization charge. However, given the high rate environment and the shape of the cryostat, collecting ionization charge as it is done in the Far Detector will not be feasible. On the other hand, equipping GRAIN with photodetectors only will allow to measure the total energy released in the medium and give a time reference to the event, but the spatial reconstruction ability would still be limited. The proposed improvement to recover the spatial reconstruction is to place imaging devices on the inner walls of GRAIN to take pictures of neutrino interactions. Such devices are an innovative concept for liquid argon detectors, as scintillation light has never been exploited to this extent. Since the VUV wavelength and the cryogenic environment rule out commercially available cameras, a custom-made solution is being pursued. The active photosensor will be a 2D array of Silicon PhotoMultipliers (SiPMs), each acting as an independent pixel, while two technologies are currently being investigated for the imaging element of the camera: coded aperture masks and lenses. The purpose of this thesis has been the study of the lens-based option and its performance for the reconstruction of ν-Ar interactions in SAND. An optimized device was designed to address the challenges derived from operating in liquid argon and validated using Monte Carlo simulations, leading to the definition of a preliminary geometrical configuration of these cameras in GRAIN. An optical simulation of GRAIN was then developed to simulate its response to neutrino interactions within the SAND framework. This software package includes a full description of the scintillation mechanism and light propagation, returning as output the set of images recorded by each camera. A preliminary reconstruction algorithm was also developed to analyze these 2D images and combine them together to obtain a 3D reconstruction of the event. The overall reconstruction performance was then studied by integrating together information coming from all SAND subsystems. In addition to the simulation efforts, the first camera prototype was built and tested in a warm environment, showing agreement with the expectations, while cold tests are in preparation for the near future. Despite more development being needed, this innovative optical readout for GRAIN is expected to provide a significant contribution to the selection of ν-Ar samples and the fulfillment of the Near Detector goals.