Study and mitigation of spurious electron emission from cathodic wires in noble liquid time projection chambers

Abstract

Noble liquid radiation detectors have long been afflicted by spurious electron emission from their cathodic electrodes. This phenomenon must be understood and mitigated in the next generation of liquid xenon (LXe) experiments searching for WIMP dark matter or neutrinoless double beta decay, and in the large liquid argon (LAr) detectors for the longbaseline neutrino programmes. We present a systematic study of this spurious emission involving a series of slow voltage-ramping tests on fine metal wires immersed in a two-phase xenon time projection chamber with single electron sensitivity. Emission currents as low as 10 −18 A can thus be detected by electron counting, a vast improvement over previous dedicated measurements. Emission episodes were recorded at surface fields as low as ∼10 kV/cm in some wires and observed to have complex emission patterns, with average rates of 10-200 counts per second (c/s) and outbreaks as high as ∼10 6 c/s. A fainter, less variable type of emission was also present in all untreated samples. There is evidence of a partial conditioning effect, with subsequent tests yielding on average fewer emitters occurring at different fields for the same wire. We find no evidence for an intrinsic threshold particular to the metal-LXe interface which might have limited previous experiments up to fields of at least 160 kV/cm. The general phenomenology is not consistent with enhanced field emission from microscopic filaments, but it appears instead to be related to the quality of the wire surface in terms of corrosion and the nature of its oxide layer. This study concludes that some surface treatments, in particular nitric acid cleaning applied to stainless steel wires, can bring about at least order-of-magnitude improvements in overall electron emission rates, and this should help the next generation of detectors achieve the required electrostatic performance.

Investigation of background electron emission in the LUX detector

“…For example, the ultraviolet xenon scintillation and EL photons carry an energy of ∼7 eV, which is above the work function of many metals and other species; therefore, the photoelectric effect can liberate electrons from the electrode surfaces and certain impurities dissolved in the liquid xenon [9,11,39]. In addition, electrons may be emitted from the cathodic electrode wire surfaces, where the electric field can reach very high values if physical or chemical defects are present [24,40]. Electron emission from electrodes can lead to high voltage instabilities or even breakdowns and has prevented several TPC experiments from operating at the designed field configurations [29,41].…”

“…Similar transient electron emission from hot spots have also been reported in Refs. [24,40], and are attributed to grid emission. A total of three statistically significant rate spikes are identified in this 10-day dataset, and additional ones with smaller amplitudes are observed but cannot be definitively differentiated from statistical fluctuations of the background rate.…”

“…Several authors have studied electron emission in xenon TPCs, and developed viable hypotheses that explain certain background populations [9,11,17,18,[22][23][24][25]. In this work, we strive to obtain a coherent picture of these background effects through a systematic investigation of all observed electron pathologies in the Large Underground Xenon (LUX) dark matter experiment [26].…”

Extending light WIMP searches to single scintillation photons in LUX

“…Additionally, the Hamamatsu R11410 PMTs used in LZ are found to have higher DPE probabilities which would also aid in a more efficient analysis. Finally, there are prospects for lower coincidence backgrounds from spurious electron and photon emission from its grids as this has been the subject of significant research in recent times [19].…”

“…PMT dark counts consist almost entirely of 1 phe signals and hence did not pose a significant background. This method retains z-position (depth) information and ER-NR discrimination for these events, and hence can provide a bridge between standard analyses with both S1 and S2 signals and the more challenging S2-only analyses [17,18], that can suffer from higher background levels (e.g., from radioactivity or spurious electron emission from electrode grids [19]). An example event that fell below the twofold requirement but in which the S2 was preceded by a single detected photon of a large pulse area (indicative of DPE emission) is shown in Fig.…”