Return to search

Analysis and modelling for CRESST II

The dark matter search CRESST-II completed its most recent run, Run 32, in 2011. Compared to previous runs, the quantity of data taken in this run increased , significantly. In this work, Oxrop, analysis software in use within the CRESST collaboration, is upgraded to analyse this new data. At the same time, Oxrop's internal structure is improved so that it can now handle data from detectors across different experiments consistently. This upgrade was performed with a view to developing Oxrop's candidacy for use with EURECA, a future dark matter experiment. Oxrop is then used to model CRESST-II data .. First, light detector response to scintillation light produced in y interactions in CRESST -Il' s target crystals is examined. A factor influencing detector efficiency is the time constant of scintillation light production, and this light detector examination is performed with a view to extracting the scintillation time constants of the target crystals. A simple model of light detector response of one exponential rise and two exponential decay times is initially considered. It is shown that this simple model does not closely match the light detector response to y interactions in the crystal scintillator. Empirical extensions to this expected model are then made, allowing for additional decay times. These extensions allow the light detector response to crystal scintillator interactions to be well modelled, and allow estimates of the millikelvin y scintillation time of Ca W04 and Zn W04. This model is then also applied to X-ray interactions directly in the light detectors. It is seen that, even with these model extensions, interactions directly in the light detector still show significant tension with the applied model. This implies that direct calibration of light detectors with X-rays is not possible without a further understanding of light detector response, or that future direct calibrations should be done with optical photons. Position dependent effects in Run 32 calibration data are then studied. A phenomenon that has previously been considered as unrelated to position dependence, the anti-correlation effect between phonon and light detector signals, is shown to exhibit a position dependent effect in at least one lightjphonon detector pair under study. Additionally, the collection efficiency of the light detector is shown to be related to the mean interaction position. Collection efficiency is found to reduce when mean interaction position is close to the cylindrical surfaces of CRESST's Ca W04 target crystals. The magnitude of the difference in light collection: efficiency between surface and bulk interactions is also seen to be correlated with high energy light detector resolution. The WIMP-nucleon cross section limits resulting from the CRESST-II commissioning run (2007) are also reanalysed in this work. The original analysis of the commissioning run accounted only for tungsten recoils in the Ca W04 crystals used in CRESST - II. Here, interactions from calcium and oxygen nuclei are also accounted for. The resulting WIMP-nucleon cross section limits were improved at light WIMP masses -0(10 GeV j c2). These limits show a mild tension with a recent dark matter analysis of Run 32, particularly for WIMP masses below 10 GeV j c2. Possible causes of this tension are discussed

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:568066
Date January 2011
CreatorsBrown, Andrew
ContributorsKraus, Hans
PublisherUniversity of Oxford
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation

Page generated in 0.0019 seconds