OPTIMIZATION OF THE SINGLE MODULE OF DETECTION FOR THE CUORE-0 EXPERIMENT

Neutrino physics is nowadays one of the most discussed research fields in the study of the fundamental constituents of matter and their interactions. Its popularity has been increasing in the last years, supported by the wealth of experimental facts that have given model-independent proof of neutrino oscillations: the observation of oscillating neutrinos, first discovered by SuperKamiokande, then confirmed by SNO, has contributed significantly in pushing the efforts of the physics community towards a revision of the Standard Model. The fact that neutrinos oscillate between flavours can be understood only if a mixing mechanism, similarly to what happens in the hadronic field, is introduced, which leads directly to the request that these particles are admitted to be massive; and, for the Standard Model to incorporate non-zero neutrino masses, a certain number of its foundations should be relaxed. Among them, the redefinition of neutrinos as equal to their own anti-particles, which is a condition defined as “Majorana nature”, is particularly appealing to theorists. Experiments on flavour oscillations by neutrinos will shortly follow the path of precision measurements. This perspective, however, although needed, does not lead to any radical clarification about the topics that are still open today: a measurement on oscillating neutrinos does not quantify their mass eigenvalues but only the difference between their squares, nor it is capable of determining which nature, Majorana or Dirac, is valid for them. The values of squared mass differences obtained by these experiments are compatible with three possible scenarios: one where all masses are of order »1 eV and very similar to each other, and two that require one mass to be much smaller (inverse hierarchy) or larger (normal hierarchy) than the other ones. It is true that the absolute values of neutrino mass eigenvalues could be determined by cosmological measurements or by end-point study in single ¯-decay, but the first is model-dependent and the second one cannot identify the correct nature. The only instrument capable of answering these questions at once is the search for Neutrinoless Double Beta Decay (0o-DBD). The observation of this very rare nuclear decay would put a lower limit on the mass of the electron neutrino, would clarify which mass scenario holds, and would imply that neutrinos are Majorana particles. In fact, 0o-DBD is assumed to occur by exchange of a light Majorana neutrino and would be impossible in the Dirac case; however, it has been proved that witnessing a 0o-DBD would always mean that neutrinos are Majorana fermions, even if the dominant mechanism differs from the one mentioned above. Moreover, the conservation of lepton number would be violated. The decay, which is possible for nuclides where single-¯ is energetically forbidden, is very rare and has never been observed, apart from a discussed claim on 76Ge. Experiments on 0o-DBD can explore a range of values for a peculiar combination of neutrino mass eigenvalues called effective Majorana mass m¯¯: current researches have explored the region of m¯¯ that would hold in the case of similar masses, but must now move down to effective Majorana mass values of order 0.1 meV. The goal is very challenging. Due to the rarity of the decay (foreseen half-lives are higher than 1020 years) large quantities of nuclides must be observed for long times, and in an environment free of radioactive contaminants whose counting rates could mask the ones searched for. 0o-DBD, in principle, is marked very clearly in the spectrum obtained by summing over the energies of both the electrons emitted in the decay. One should look for a peak at the Q-value of the decay enlarged only by the finite energy resolution of the detector; many different techniques are suitable to approach the problem. One of the most promising detection strategies relies on the use of macro-bolometers, which are phonon-mediated detectors with mass of order 1 kg operating at extremely low temperatures (»10 mK). Bolometers are feasible and guarantee outstanding energy resolutions. The Cuoricino experiment, a tower of 62 bolometers, has searched for the 0o-DBD of 130Te from 2003 to 2008 at the Italian Gran Sasso National Laboratory, reaching one of the best sensitivities on m¯¯ of the old generation. Its descendant for the investigation of the inverted hierarchy region will be the CUORE experiment. CUORE will start taking data in 2013 and will be constituted by nineteen Cuoricinolike towers, increasing the total mass of a factor 20. The possibility for CUORE to enter the inverted hierarchy range depends on many factors. A mere increase in mass would not be sufficient, and time would also cease to be effective after a few years of datataking. Therefore, a thorough review of all the detector’s aspects is necessary, from the parameters related more directly to the sensitivity of the experiment, to detection performances, paying also attention to the modularity of the final array which becomes extremely relevant when such a large number of bolometric units is to be assembled and operated. The starting point is fixed, of course, by the achievements gained by Cuoricino in the technique’s application. An intermediate step will precede the beginning of CUORE: its first tower will be cooled-down and measured as a stand-alone experiment, with the beginning of data-taking foreseen in the first part of 2011. CUORE-0 will provide an imminent verification of the R&D work performed, but will also constitute a powerful experiment on its own, capable of improving the limit on m¯¯ fixed by Cuoricino. The precise focus of the Ph.D. work that I pursued in the last three years and will present in this thesis is, therefore, the optimization of the single module of detection in the view of the incoming CUORE-0, which will validate the innovations for their introduction in CUORE. The Ph.D. activity has been performed both at the Insubria University’s Cryogenics Laboratory in Como, with the development of small prototypes and preliminary tests, and at the Gran Sasso National Laboratory, where bolometers of the ultimate size have been operated in conditions similar to those of the final experiment. My work focused on optimizing some crucial aspects of the single module of detection in Cuoricino, by contributing in the following R&D activities: the fight against the background contribution in the range of interest due to surface contaminations of the materials surrounding the detector; the uniforming of protocols for the production and processing of energy absorbers (aiming at an increase of both bolometric performances and radio-purity); a striking innovation in the geometry of the semiconductor sensors that convert temperature pulses in voltage signals; a dedicated study aiming at optimizing and automating the coupling between the energy absorbers and the sensors, which is a delicate point in the determination of performance reproducibility. The tuning of all of these aspects, as will be seen in the course of this thesis, is necessary for CUORE to reach the goal of sensitivity to m¯¯ in the range of inverted hierarchy.