Author: Dimitri Misiak

Publisher:

Published: 2021

Total Pages: 0

ISBN-13:

The Coherent Elastic Neutrino-Nucleus Scattering (CENNS) is a process predicted nearly 40 years ago. In August 2017, the COHERENT experiment reported the first keV-scale detection at the 6.7 sigma level of this process, which is a probe for the new low energy physics, opening a window on a myriad of new physics opportunities. The RICOCHET experiment aims at measuring with high accuracy the CENNS process in order to probe various exotic physics scenarios in the electroweak sector. Using cryogenic bolometers operated in a cryostat 8 meters away from the core of the ILL research nuclear reactor, the experiment will benefit from an intense neutrino flux, allowing the results of COHERENT to be reproduced in a single week. The objective of an accurate measurement will be achieved after one year of data collection, by 2024. The CRYOCUBE is a compact cubic array of cryogenic detectors with the following specifications: a very low energy threshold of O(10) eV on the thermal signal, an electromagnetic background rejection of at least 10^3 and a total target mass of 1 kg distributed among 27 germanium crystals of about 30 g each. The objective of this thesis is to propose an optimized detector design for the CRYOCUBE, inspired by the cryogenic germanium detectors equipped with charge and temperature readings of the direct dark matter search experiment EDELWEISS. This joint R&D program is based on event discrimination realized in germanium semiconductor crystals. The recoil energy of an incident particle is derived either from the increase of the crystal temperature measured by a GeNTD thermistor (heat channel) or from the excited electric charges collected by electrodes on its surface (ionization channel). This double energy measurement makes it possible to distinguish the nuclear recoils produced by the CENNS or the dark matter from the electronic radioactive background. As these recoils are of the order of O(100) eV, this thesis work is focused on the development of a new generation of cryogenic low threshold germanium detectors with particle identification. It explores how to improve the resolution in heat and ionization energy up to O(10) eV while maintaining a good rejection of background events. This study is based on the testing of prototype detectors in the IP2I cryostat, which are compared to theoretical predictions from electro-thermal and electrostatic modeling of the detectors. This manuscript begins with the definition of the CENNS process, its scientific importance and the objectives of the RICOCHET experiment. It then presents the cryogenic installation allowing the surface operation of the detectors at 20 mK in optimal conditions. An electro-thermal model of the bolometers, compared with experimental data, is developed and applied to the simulation of the noise associated with the electronics of the heat signal. The thesis then formalizes the generation of the ionization signals arising from excited charge carriers drifting in the germanium crystal under the influence of the applied electric field. The expected resolution from a future low-noise electronics is modeled based on two detector designs. They are optimized by their electrostatic simulation in a finite element calculation software. A comparison of the theoretical and experimental performance of ionization is performed on the basis of the RED80 and REDN1 prototype detectors. This work ends with the characterization of the radioactive background in the cryogenic laboratory with the analysis of the data from RED80, and in particular its neutron component, used to estimate the expected background at the ILL site for RICOCHET.