3D Tracking with the CYGNO/INITIUM experiment

The nature of dark matter (DM) remains one of the most fundamental open questions in modern physics. Among the main DM candidates are Weakly Interacting Massive Particles (WIMPs), which are expected to interact with ordinary matter through scattering. Detecting such interactions requires highly sensitive and low-background detectors. Directional detection offers a promising way to overcome the limits of conventional WIMP search techniques – especially the neutrino floor – by identifying the direction of WIMP-induced nuclear recoils (NR) and thereby inferring the incoming WIMP direction. The motion of Earth around the Sun and within the Galaxy creates an apparent WIMP wind, which results in an anisotropic distribution of recoil tracks in detectors on Earth, an effect that no other background can mimic. This thesis contributes to this effort through the development and application of 3D reconstruction techniques for the CYGNO/INITIUM experiment. CYGNO employs a He:CF4 gas mixture at atmospheric pressure and a triple-GEM amplification system, producing scintillation light that is read by CMOS sensors and photomultiplier tubes (PMTs). This work presents the development of a 3D reconstruction strategy by combining CMOS (XY) and PMT (Z) data. A dedicated framework was developed to extract information from PMT waveforms, synchronize it with CMOS data, and reconstruct the full 3D topology and direction of ionizing tracks. This reconstruction was applied to LIME data, CYGNO’s largest prototype, operating underground at LNGS. The 3D analysis was initially focused on alpha particles, whose ionization profiles closely mimic WIMP-induced NRs. By combining CMOS and PMT data, alpha tracks were reconstructed with accurate direction and head-tail determination. This method was used to characterize the alpha background in LIME, including spatial localization within different sectors of the detector. The results confirmed the presence of Rn-222 diffused in the sensitive gas volume and the likelihood that its decay daughters are produced with a positive charge. The spatial and angular distributions of alphas also revealed contamination from U-238 and Th-232 in the detector materials, namely field cage rings and GEMs, showcasing the robustness of the developed analysis. These findings helped explain discrepancies between Monte Carlo simulations and measured data and led to design improvements in future CYGNO detectors, namely CYGNO-04. Finally, the thesis reports the first steps toward implementing Negative Ion Drift (NID) operation with MANGO prototype. A novel analysis strategy is presented for the literature first-ever reported NID PMT waveforms. These results highlight the potential of the CYGNO/INITIUM approach for directional, low-background dark matter detection and support its development into future large-scale detectors.