Boosting background suppression in the NEXT experiment through Richardson-Lucy deconvolution

Abstract

Next-generation neutrinoless double beta decay experiments aim for halflife sensitivities of ∼ 10 27 years, requiring suppressing backgrounds to a level of < 1 count/tonne/year. To reach those levels any extra background rejection handle, in addition to excellent energy resolution and the use of extremely radiopure materials, is of utter importance. The NEXT experiment exploits differences in the spatial ionization patterns of double beta decay and single-electron events to discriminate signal from background. While the former display two Bragg peak dense ionization regions at the opposite ends of the track, the latter typically have only one such feature. Thus, comparing the energies at the track extremes provides an additional rejection tool. The unique combination of the topology-based background discrimination and excellent energy resolution (1% FWHM at the Q-value of the decay) is the distinguishing feature of the NEXT experiment. Previous studies demonstrated a topological background rejection factor of ∼ 5 when reconstructing electron-positron pairs in the 208 Tl 1.6 MeV double escape peak (with Compton events as background), recorded in the NEXT-White demonstrator at the Laboratorio Subterráneo de Canfranc, while retaining ∼ 72% of the signal. These results were recently improved through the use of a deep convolutional neural network to yield a background rejection factor of ∼ 10 with 65% signal efficiency. Here, we present a new reconstruction method, based on the Richardson-Lucy deconvolution algorithm, which allows reversing the blurring induced by electron diffusion and electroluminescence light production in the NEXT time projection chamber. The new method yields highly refined 3D images of reconstructed events, and, as a result, significantly improves the topological background discrimination. When applied to real-data 1.6 MeV electron-positron pairs (at this stage without using deep neural networks), it leads to a background rejection factor of 27 at 57% signal efficiency.C Examples of successful RL-based classification vs. failure of the classical analysis 31

Measurement of the ${}^{136}$Xe two-neutrino double beta decay half-life via direct background subtraction in NEXT

“…E b,min is defined as a function of the energy of the event, optimized by means of MC studies. The ratio of the signal efficiency over the square root of the background acceptance ranges from ∼2.3 to ∼3.1 for events between 1 and 3 MeV energy, consistently with [13]. According to this selection, the left and right panels of Fig.…”

“…For the former topology, only calibration events inside the double-escape peak at 1.6 MeV produced by 2.6 MeV 208 Tl gamma-rays are used. As discussed in [13,16], events inside (outside) this peak are mostly populated by pair-production (Compton scattering) events. The overall ββ selection efficiency for double-electron (single-electron) calibration tracks of E evt > 1 MeV energy is measured to be 24.7±0.5% (2.24±0.06%) in Run-V, and 27.5±0.6% (2.34±0.07%) in Run-VI.…”

“…A second reconstruction stage is performed in order to reverse the blurring induced by the electron diffusion and EL light production. A Richardson-Lucy deconvolution is applied to the 3D hits relying on a point spread function derived from 83m Kr events [13]. The deconvolved 3D hits are then grouped into volume elements of (5 mm) 3 , which are used to build tracks following the connectivity criteria established by a breadth-first search algorithm [14].…”

“…The deconvolved 3D hits are then grouped into volume elements of (5 mm) 3 , which are used to build tracks following the connectivity criteria established by a breadth-first search algorithm [14]. The energies of the end-points of each track, hereafter 'blobs', are defined by integrating the energy of the hits contained within spheres of 18 mm radius centered in the identified extremes [13]. Figure 1 shows examples of two observed tracks of 1.7 MeV.…”

Development of very-thick transparent GEMs with wavelength-shifting capability for noble element TPCs

“…The visible photon PSF is 2-3 times wider than for VUV, yet recent results show that optical PSFs even wider than the one in Fig. 3(bottom-right) are compatible with track reconstruction with 1-2 mm accuracy [42].…”