Calibration and commissioning results of the NUCLEUS experiment

Neutrinos, despite outnumbering every other particle in the entire universe, have eluded discovery for more than 25 years after being postulated and still pose a challenge both to theorists and experimentalists. There are diverse neutrino-matter interactions which span a broad range of energies, with a common denominator: the smallness of the cross-section. For MeV neutrinos, this cross-section is below $10^{-40}~\text{cm}^2$ and makes the observation of these particles in laboratory experiments extremely challenging, as years of dedicated neutrino oscillation experiments have demonstrated. One notable exception to this rule is the \cevns[.] This weak neutral current interaction stands out by having a large, by neutrino standards, Standard Model cross-section, which is more than 2 orders of magnitude higher than other neutrino processes. This is a game changer in the experimental landscape, as it allows the study of neutrinos using relatively small detectors (with masses ranging from 10~g to 1~kg), as opposed to ton or multitone experiments such as Borexino or Juno. On the other hand, the only \cevns observable is the energy of the induced nuclear recoil, which is at the 100~eV scale, meaning that this process is renown to be difficult to measure because detectors with an energy threshold as low as O(10~eV) are needed. Among the many physics motivations to study \cevns[,] a precision measurement of its cross-section would provide the means to unveil new physics beyond the Standard Model, like non-standard interactions or an unforeseen scaling of the weak mixing angle. Moreover, the detector miniaturization, allowed by the high cross-section, will give rise to new branches of neutrino applications for both civilian and military purposes. This thesis develops in the context of the NUCLEUS experiment, which aims to detect \cevns using the high anti-neutrino flux from the two 4.25~GW$_\text{th}$ reactors cores of the Chooz-B nuclear power plant in France. NUCLEUS will exploit an innovative detection system that consists of a 10~g array of cubic CaWO$_4$ and Al$_2$O$_3$ crystals as target detectors. The energy deposited in the crystals is read with superconductive \acl{TES}s deposited on the material surface. This technique allowed the NUCLEUS collaboration to develop detectors that reach the low energy thresholds required for a successful \cevns observation. At the start of this Ph.D. research, the NUCLEUS experiment was in its preliminary stage and only the basic setup for detector development was present. Over recent years, substantial improvements have been made to both the experimental setup and data analysis. A crucial issue facing cryogenic calorimeters, designed to measure energy depositions as low as O(10~eV), is the characterization of their response. To tackle this, NUCLEUS adopted a novel calibration technique based on the use of optical photons shining on the target calorimeter. The LANTERN project was developed in the context of this work to provide a highly scalable and cost-effective hardware setup to perform this calibration. LANTERN proved to be an elegant and simple solution that can be employed to achieve a full characterization of the detector response, both in terms of calibration and non-linearity evaluation, and will be deployed in the NUCLEUS setup in its final configuration. The NUCLEUS target detectors are expected to respond equally to both electron and nuclear recoils. To validate this, the NUCLEUS collaboration conducted a calibration campaign using the absorption of thermal neutrons from the nuclei present in the target detectors. The absorption produces an excited nuclear state that, by decaying to its ground state, generates a high energetic photon, that escapes detection, and a nuclear recoil of 100~eV, exactly in the region of interest of the NUCLEUS experiment. Detecting an interaction at 100~eV with a noise level of 10~eV (as for the NUCLEUS detectors) is difficult due to the low signal-to-noise ratio. Therefore, a suitable data analysis procedure was developed and used to observe the first direct detection of nuclear recoils with the NUCLEUS target detectors, achieving a $3\sigma$ significance. Since neutrinos interacting via \cevns produce nuclear recoils in the same 100~eV energy range, this data was the perfect test-bed to define the fundamental analysis aspects that need to be followed for a fruitful neutrino detection. Due to the importance of the developed data analysis, a thorough description is presented in this work along with the results of the nuclear recoil calibration. At the time of completing this thesis, the NUCLEUS collaboration commissioned its experimental setup and performed a long background characterization campaign. The interplay between the main subsystems, including LANTERN, was demonstrated and stable operation of the NUCLEUS detectors over an extended period of time (a few weeks) was achieved. The analysis of the data taken during this period is defined and presented in this work, with focus on the development of the first combined use of the active muon veto and target detectors as devised for the final configuration of NUCLEUS setup. At the end of this thesis, a discussion over the achieved background results is conducted. In summary, this thesis presents an overview of the \cevns interaction and the status of its searches with focus on the NUCLEUS experiment, which is the experimental context of this work. The central theme of this dissertation is the characterization and use of the cryogenic calorimeters employed for the detection of nuclear recoils at the few hundreds of electron volt scale. This topic is addressed from both the experimental and the analysis points of view, with the development of an optical calibration, as well as assessing the accuracy of these new procedures with the observation of calibrated nuclear recoils induced by neutron absorption. Finally, the results of the background characterization campaign and the setup commissioning are shown.