This dissertation focuses on the measurement of solar neutrinos using liquid scintillator detectors. Within the Sun’s core, stellar nucleosynthesis occurs through two primary reaction sequences: the proton-proton (pp) chain and the subdominant Carbon-Nitrogen-Oxygen (CNO) cycle. Both sequences convert Hydrogen into Helium, releasing various solar neutrinos in the process. The detection of these neutrinos provides a unique method to probe the mechanisms driving stellar energy production and offers an opportunity to address longstanding astrophysical questions, such as the unresolved solar metallicity puzzle. In this dissertation, I present my analyses of data from the Borexino experiment, located at the Laboratori Nazionali del Gran Sasso (LNGS), and simulations from the JUNO experiment, currently under construction in China. Two complementary methods are employed to detect solar neutrino signals. The first approach utilizes a spectral analysis technique that relies on the energy and position information of events. This technique primarily leverages the scintillation light produced during the interaction of solar neutrinos and background contaminants with the liquid scintillator. The second method, the Correlated and Integrated Directionality (CID) technique, was introduced in Borexino. It exploits the directional information encoded in the fast, sub-dominant Cherenkov emission produced within the detector. The combination of these two methodologies represents the most powerful approach to isolate solar neutrino signals from background-induced events. In the final stages of its data-taking period (January 2017 - October 2021), Borexino prioritized improving the accuracy of the first measurement of the neutrinos produced in the CNO cycle (referred to as CNO neutrinos) performed in 2020. My contributions to this endeavour are discussed in this thesis. I participated in the optimization of the simulated detector response and the re-tuning of Monte Carlo simulations. Additionally, I developed a novel method based on cosmogenic neutrons for monitoring the detector response throughout the detector volume over time. Finally, I conducted the spectral analysis, performed by constraining the contributions from pep neutrinos and 210Bi contamination from independent estimations, and evaluated all potential sources of systematic uncertainty. The outcomes indicate a significant increase in the measurement precision and rejection of the null hypothesis of no CNO neutrino flux. The resulting CNO neutrino interaction rate serves as an input to test the metallicity of the Sun and draw inferences about solar physics. This measurement has been published by the Borexino collaboration in Physical Review Letters. The CID method is used to determine for the first time the rate of CNO neutrinos using the complete Borexino dataset (May 2007 - October 2021). This measurement is unique as no assumptions on backgrounds present in the detector are required. Given the independence of information utilized by the CID and spectral methodologies, these techniques are combined to achieve the highest level of accuracy. I detail my contributions in implementing a novel two-dimensional spectral analysis that leverages the results from the CID analysis as independent constraints. This approach leads to the final and most precise CNO neutrino measurement achieved to date. These results are presented in a paper published by the Borexino collaboration in Physical Review D, where I am one of the four main authors. Leveraging on the expertise gained within the Borexino experiment, I have implemented analogous strategies to investigate the potential of the JUNO detector for solar neutrino measurements. In the second part of this thesis, I present my contributions to developing a spectral analysis framework used to determine JUNO’s sensitivity to solar neutrino. This analysis resulted in a collaboration paper, in which I am listed as one of the main authors, published in the Journal of Cosmology and Astroparticle Physics. Furthermore, I actively participated in developing the framework to apply the CID method within the JUNO software and combining the two methodologies. The work presented in this thesis serves as proof of principle for these approaches.
Luca Pelicci (Fri,) studied this question.