Electron Recoil rejection by decay time measurement in large liquid Xenon detectors

Abstract

Very large Liquid Xenon (LXe) Time Projection Chambers (TPC) are employed to search for Dark Matter (DM). The DM particles are supposed to interact with the whole nucleus, compared to background of γ-rays, which interact with the electrons. Therefore, DM signals are caused by Nuclear Recoil (NR) instead of the Electron Recoils (ER). In ER and NR events differ in pulse shape since the ratios of light from direct scintillation and recombination are different. To discriminate against residual ER events would be possible if one can distinguish the differences in decay times. This method can be successfully applied in Liquid Argon TPCs. In LXe, however, it is generally assumed that these differences are too small to be distinguished at low energies. The easiest algorithm of Pulse Shape Discrimination (PSD) distinguishes the event type based on the number of photons emitted much later than the longest decay time. At low energies too much of the timing information is lost, and this method does not perform well. However, the timing of all photons does contain sufficient information. If we use sufficiently fast PMTs, have a large enough bandwidth in the Front End electronics, and avoid reflections then we should reach a background rejection better than 10−2 even at 2 keVee. In our Decay Time Measurement (DTM) method the decay curves are compared with a model on an event by event basis. Statistically this is independent from the charge over light (`S2/S1') cut normally applied in Dual Phase detectors. Applying both rejection mechanisms a LXeTPC can become `quasi background free'.