ARAPUCA, light trapping device for the DUNE experiment

TL;DR

This work develops ARAPUCA and X-ARAPUCA light-trapping devices for the DUNE Photon Detection System, demonstrating efficient capture of liquid argon scintillation light at 127 nm and guiding material choices to reach the experiment's light-yield goals. It reports full prototype characterizations, including a new wavelength shifter (FB118) that yields PDEs up to $2.9\%$ in X-ARAPUCA configurations, and shows a roughly 50% PDE enhancement when using the FB118 WLS over EJ-286. The study also validates the concept in ProtoDUNE-SP, where xenon doping is explored as a robust strategy to recover light yield under nitrogen quenching, with evidence of substantial recovery (up to about 95%) under certain conditions. Together, these results support X-ARAPUCA as a promising option for the DUNE far detector PDS and provide practical guidance for materials, validation tests, and potential Xe-doping implementations to mitigate impurities in large LArTPCs.

Abstract

The Deep Underground Neutrino Experiment (DUNE) will be the first mega-science program on the US soil and will shade light on some of the open questions in neutrino physics. The experiment foresees the realization of an intense neutrino beam at Fermilab (Chicago - USA), of a near detector to monitor the beam and a far detector installed in the Sanford Underground Research Facility (SURF), 1300 km far away. Four 10 kt Liquid Argon Time Projection Chambers (LArTPC) will compose the 40 kt far detector modules to perform the precise measurements required for DUNE. The experimental technique uses charge and light signal from ionizing radiations in liquid argon to fully reconstruct neutrino interactions with excellent spatial resolution, calorimetric measurements and particle identification. The liquid argon scintillation light is produced around 127 nm within a few nanoseconds from the radiation passage. Its detection may improve the LArTPC calorimetric besides giving the time stamp of non-beam events, which is vital in fiducializing nucleon-decay events. The DUNE Photon Detection System (PDS) is responsible to efficiently collect this light in order to fiducialize the active volume of the detector with $>99\%$ efficiency. The PDS will be composed by 1,500 X-ARAPUCA modules, a light trapping device that was the main subject of this thesis. In this work, the research and development of the ARAPUCA devices will be presented. The full characterization of the prototypes were performed in dedicated liquid argon tests, in Brazil and Italy, where comparable efficiencies of (2.2 $\pm$ 0.4)% and (1.9 $\pm$ 0.1)% were found. A new wavelength shifter developed in Italy in collaboration with the Universita degli Studi di Milano-Bicocca will be presented. It allowed to increase the light detection efficiency of about 50% and resulted in an overall efficiency of (2.9 $\pm$ 0.1)% of the X-ARAPUCA.

Figures (104)

• Figure 1: (Left) Allowed region within 90% C.L. for $\sin[2](2\theta_{13})$ and $\delta_{CP}$ parameters for 7, 10 and 15 years of beam exposure. The global fit to neutrino oscillation data (NuFIT 4.0) 90% C.L. region is represented by the yellow area and the black star is the "true value" assumed to be the central value of the global fit. (Right) Allowed region within 90% C.L. for $\sin[2](\theta_{23})$ and $\delta_{CP}$ parameters DUNE_vol2.
• Figure 2: Solar chain for electron neutrino production vissani.
• Figure 3: Predicted solar neutrino energy spectra. The solid lines represent neutrinos produced by the $pp$ and $pep$ chains and the dashed blue lines represent neutrinos from the CNO cycle. The theoretical fractional uncertainties are labeled and are shown for each source solar_nu_flux_2005.
• Figure 4: (Left) Predicted atmospheric neutrinos flux in absence of oscillation, average over the zenith angle was taken. (Right) Cartoon of the up and down symmetry on the neutrino flux vissani. Without considering oscillations, the flux of atmospheric neutrinos would be symmetric following the flux in the left panel.
• Figure 5: Zenith angle distribution of $\mu$-like (bottom panels) and $e$-like (top panels) events for sub-GeV and multi-GeV events in the Super Kamiokande. Upward-going and downward-going particles have $\cos\Theta<0$ and $\cos\Theta>0$, respectively. The dashed regions are Monte Carlo expectation for no oscillations and the bold line is the best-fit expectation for $\nu_\mu \rightarrow \nu_\tau$ oscillations SuperK_plots.