Simulation for the development of cryogenic detectors for direct dark matter searches

More than 80% of the total matter budget of the universe is non-luminous matter, so-called dark matter. This missing mass can be explained by introducing a new type of particle which interacts via gravitation and, at most, interacts only weakly with ordinary matter. The possible mass range of this proposed dark matter particle spans several orders of magnitude. Recently theoretical models predicting a strong force among dark matter particles gained a lot of interest, since they have the potential to solve some outstanding astrophysical problems. The mass scale of this strongly interacting dark matter candidates is rather light and expected to be between several MeV to a few GeV. The CRESST experiment, currently operated at the Gran Sasso national laboratory in Italy, is a dedicated dark matter experiment looking for elastic scatters of potential dark matter candidates with ordinary matter. The energy deposited by the scattering process is transformed to lattice excitations of the detector crystal, so-called phonons, and further on read out by transition edge detectors. No signal has been observed with previous data taken by the CRESST experiment, however, the exclusion limit set for dark matter candidates below 2 GeV is the best for all direct detection experiments. In order to further improve the sensitivity towards smaller dark matter masses and weaker cross-sections, the detection threshold has to be further lowered and the intrinsic radio-purity decreased. A good understanding of the intrinsic radio-purity is crucial for the success of the experiment. The expected intrinsic radio-purity in the signal region can be estimated with a dedicated simulation software using the measured alpha-decay activity as an input. The simulation provides information about the size and the type of the possible background contributions and provides important input to the design of the next generation of dark matter experiments. In addition a new simulation framework will be developed to model the detection of the phonons produced by the scattering process.

The most important result was the development of the GEANT4-based simulation framework ImpCRESST and its application to a CRESST detector module. No simulation framework was available for CRESST before, and the new tool allows estimates of the radiogenic background from internal and external background sources for the first time. The estimates are generated for the best-studied detector module at that point, with an energy detection threshold of 604+-2 eV. Energy spectra from identified radioactive isotopes are simulated with GEANT4 and normalized to the progenitor alpha-decays assuming secular equilibrium. With this data-driven approach, the background content in the region of interest (ROI) can be evaluated using high-energy alpha spectral analysis signals. The simulation cannot fully reproduce the data in the ROI, and because of the unknown rise towards the threshold, the energy range was taken from 1 till 40 keV. In the ROI 68+-16% of the events can be explained by known radioactive decays. However, the rising background below the ROI can not be reproduced by the simulation. This rising background at low energies was later on also observed at CRESST with different detector geometries and more detailed investigations are a central part of the just recently started second funding period of the CRC. The method described above relies heavily on determining the alpha-activity measured in the MeV energy regime. With moving towards smaller detection thresholds, the energy measurement of the alpha-decays is transferred to the saturation region of the energy measurement, and the activity measurement has to be adapted. Different approaches, like a truncated fit or a likelihood-based method, are currently under investigation. The ImpCRESST framework was initially developed for the application at the CRESST experiment only. In the meantime, two new experiments based on the same cryogenic technology are approved and under construction. The group at HEPHY, with the expertise built up during the first CRC funding period, extended the existing research profile towards these new experiments and provided valuable input for the design and construction of the new experiment. The ImpCRESST simulation package is still under development, including a planned public release and an accompanying publication during the currently ongoing funding period. The establishment of a database of the screening results for materials used in the experiment is also of high importance for the future performance of the experiment. This includes the operation of the database and the execution of the screening measurement to determine the radiogenic activity.