Advanced Reconstruction Techniques for CUORE: Searching Beyond the Standard Model with Cryogenic Calorimeters

Abstract

Located within the Laboratori Nazionali del Gran Sasso (LNGS), the Cryogenic Underground Observatory for Rare Events (CUORE) is an experiment primarily searching for neutrinoless double beta decay in ¹³⁰Te. It is the largest operating sub-kelvin cryogenic detector array, instrumenting 988 TeO₂ detector channels at temperatures below 20 mK. CUORE uses the cryogenic calorimeter technique, resolving the thermal signatures from nuclear/particle interactions within crystal absorbers for precise determination of deposited energy. This work establishes methods and analysis techniques to treat CUORE as a segmented detector in aggregate, with a focus towards identifying and reconstructing track-like signatures induced by high-energy through-going particles traversing the detector array. Implementations of such high-multiplicity techniques are used to validate that CUORE can resolve the remaining underground flux of muons within LNGS. This result demonstrates CUORE’s unprecedented size and acceptance as compared to previous cryogenic calorimeter arrays, and has applications towards future searches for neutrinoless double beta decay for which muon-induced backgrounds are non-negligible. Additionally, these methods open up new avenues for CUORE to search for exotic beyond-the-Standard Model particles and interactions, such as particles with fractional electric charge. If realized in nature, fractionally charged particles (FCPs) could be present within the underground flux of cosmic radiation and would leave faint track-like signatures across the detector. We report on a search for FCPs using the first tonne-year of CUORE’s exposure, finding no excess of FCP track candidates over background, and setting leading limits at 90% C.L. on the possible underground flux of FCPs with charges between 1/24 − 1/5 that of an electron. Lastly, we introduce differentiable programming methods for the end-to-end training of neural ordinary differential equations to model thermal pulse dynamics within CUORE calorimeter channels. These methods and results improve understanding of detector response, enable improved in situ background characterization, and open novel opportunities for CUORE and future tonne-scale cryogenic calorimeters to search for physics beyond the Standard Model.