| Popis: |
As dark matter detection experiments continue to report null results, the need for larger and more sensitive detectors means even more stringent design requirements. New calibration techniques are required and better calibration methods become possible with increased detector size. Additionally, previously ignored detector features such as convection become important, especially as internal, dissolvable sources become more common. Furthermore, convection also offers the possibility for reduction of the 222Rn backrounds via an offline analysis where atoms of 214Pb are tagged and followed throughout the detector via a technique dubbed the “radon self-veto”. In this thesis, we present the characterization of a deuterium-deuterium plasma fusion neutron generator optimized to perform the nuclear recoil calibration of XENON1T. Part of this characterization is done with liquid organic scintillator detectors, which are sensitive to both electonic and nuclear recoil interactions. We develop a new algorithm for discriminating between these two signal types using Laplace transforms and show that it performs better than traditional algorithms. A multipurpose source of dissolvable 220Rn is presented and measurements made of long-lived contaminants from this source. Finally, we present the first measurement of convection in XENON1T and report the results of a simple convection-agnostic implementation of the radon self-veto. |
