A prototype neutron-detector array for future deep-underground s-process studies

TL;DR

This work tests a prototype SHADES-style neutron detector consisting of a liquid EJ-309 scintillator surrounded by 3He counters to study neutron detection for deep-underground s-process experiments. Using neutrons from the 7Li(p,n)7Be reaction at FRANZ, the studyCharacterizes scintillator quenching, neutron/gamma discrimination, and the neutron-coincidence behavior with the counters. A Gaussian-Mixture Variational Auto-Encoder (GMVAE) enhances low-energy neutron tagging, achieving discrimination down to $E_n \approx 163$ keV, while coincidence timing reveals a robust neutron-signature around $3{-}7\,\mu s$ with substantial background suppression. Geant4 simulations corroborate the measured timing and support scaling toward the full SHADES array, which aims to measure the $^{22}$Ne($\alpha$,n)$^{25}$Mg cross-section at astrophysical energies in an underground setting with reduced background.

Abstract

We report a novel neutron-detection approach employing an EJ-309 liquid scintillator surrounded by six 3He proportional counters. Tests were performed at the FRANZ facility of the Goethe-University Frankfurt using the 7Li(p,n0)7Be reaction, producing neutrons across energies 50-720 keV. The scintillator's neutron energy quenching is determined, and its neutron/gamma-ray discrimination performance is evaluated. The lowest detectable neutron energy is 163 keV. EJ-309 - 3He counter neutron coincidences are compared with those from simulations. This array forms the prototype of a larger design, called scintillator-3He array for deep-underground experiments on the S-process, currently undergoing construction and testing for an upcoming deep-underground study of the 22Ne(alpha,n)25Mg reaction cross-section at the freshly-commissioned Bellotti Ion Beam facility of the INFN Laboratori Nazionali del Gran Sasso. This upcoming project is expected to achieve exceptionally low sensitivity for measuring the cross section at energies of interest for the astrophysical 'weak' and 'main' slow neutron-capture processes.

Figures (13)

• Figure 1: Experimental setup
• Figure 2: Neutron spectra calculated from PINO for the proton energies used in this measurement. (Refer to online plots for color).
• Figure 3: Energy spectra measured by the EJ-309 for fixed $\gamma$-ray sources a) $^{137}$Cs and b) $^{60}$Co, zoomed into the Compton edges. Solid red lines represent the total fit, with quoted $\chi^{2}$/ndf. Dotted green lines represent the background modeled by a 2$^{\mathrm{nd}}$ order polynomial (omitting the peak region of interest). Dashed blue lines represent only the Gaussian component of the total fit. c) DAQ energy to quenched energy ($E_{\mathrm{Q}}$) calibration. (Refer to online plots for color).
• Figure 4: a) Energy resolution of the EJ-309 scintillator as a function of quenched energy. b) Quenched energy to neutron energy calibration (points). Three different fits are shown: a rational function (short-dashed), a quadratic (long-dashed) and an exponential function (solid). Literature examples are also plotted from Takeda Takeda_2011 (purple dashed) and Enqvist Enqvist_2013 (orange short-dashed), for the same crystal size as this study. (Refer to online plots for color).
• Figure 5: Traditional PSD vs quenched energy. Proton beam energy = 2449 keV. Inset: Sample waveforms for neutron (PSD = 0.21) and $\gamma$-ray (PSD = 0.09). The long and short integral ranges are also shown. (Refer to online plots for color).