Sub-GeV dark matter studies and universal bound states exploration with CRESST-III

Abstract

The mystery of Dark matter (DM) has eluded us for about a century and has been one of the biggest open questions in modern-day physics. Numerous efforts have gone into understanding its nature, origin, and properties. Despite a huge number of attempts, there has been no convincing evidence of its detection or production. Many different models have been proposed to date that aims at explaining the underlying nature of the elusive substance, although none have been completely confirmed experimentally or observationally.Λ-CDM has been a prevalent cosmological model in describing the large-scale structures in the universe and their cosmic evolution. Although Λ-CDM has been successful at larger scales, it faces severe issues on smaller scales when compared to the observations.Introducing self-interactions between dark matter (SIDM) particles claims to solve the small-scale issues in Λ-CDM simulations while also being consistent with the large-scale observations. One of such SIDM theories considers some (or all) of the dark matter particles in the universe today to be a bound state of two or more dark particles called as a Darkonium. If such bound states exist, then the form factor of such bound states and the possibility of breakup of such states would impart a different signature on the direct detection experiments, that can be smoking gun evidence for its existence.CRESST is one of the leading experiments for the direct detection of DM in the sub-GeV DM mass range, and it provides the most stringent exclusion limits in the said mass region. CRESST is able to achieve this sensitivity as it employs bolometers operating at cryogenic temperatures and is able to achieve nuclear recoil thresholds of O(10 eV).Sensitivity to go even lower in threshold is restricted by the presence of an unknownbackground at low energies called as Low Energy Excess (LEE).This thesis has two main parts where the first part aims to present the data analysis chainin CRESST and show the results obtained with the latest run (Run36 ) of CRESST-III.In this run, different detector materials, modules, and holding designs were operatedwith the aim of understanding the origin of the LEE. During the run, the detectorswere warmed up multiple times to high temperatures in order to study the effect of thiswarm-up on the rate of LEE. The results obtained with these warm-ups for differentdetector modules are discussed, and the possible origins of the LEE are narrowed downfurther.A new method of calibration at sub-keV energies was recently proposed where a radiativeneutron capture on 182W would be followed by a de-excitation with a single γ-emissionthat gives a nuclear recoil peak at 112.4 eV. This method would be useful for calibration atlow energies, and more accurate than conventionally used X-ray sources, which generallylie in O(keV) energy regime. With the data obtained by irradiating an AmBe neutronsource in Run36, the observation of this expected peak at 112 eV in the CaWO4 detectorswill also be shown.With the blind data obtained from February 2021 - August 2021 in the same run,exclusion limits were extracted for different detector modules, and huge improvements inlimits were obtained for the spin-dependent DM-nuclei interaction scenario using lithiumaluminate targets. The analysis of these modules, results obtained, and implications arealso discussed.The second main part of the thesis focuses on the direct detection results of the SIDMmodel described above using CRESST-III. The study focuses on self-interactions atnon-relativistic velocities necessary to explain small-scale structures. Assuming an energyregion where the self-interaction cross-section approaches the S-wave unitarity bound,low-energy scattering properties rely on the large scattering length γ, specifically the selfinteractingscattering cross-section σχ−χ. Estimating γ from CRESST data establishesexclusion limits on its value, converted to exclusion limits on σχ−χ/mχ, dependent onthe relative momentum between dark matter particles. Results reveal the first exclusionlimits on the self-interaction cross-section from the direct detection window under thegiven physics model for various dark matter masses and their interaction cross-sectionswith the detector nucleus. Enhanced sensitivity of the experiments in the future enablesa deeper study of self-interactions in the given physics case.