Large-Mass Ultra-Low Noise Germanium Detectors: Performance and Applications in Neutrino and Astroparticle Physics

TL;DR

This work addresses measuring coherent neutrino-nucleus scattering from reactor antineutrinos, requiring large-mass targets and sub-keV energy thresholds. The authors present a large-mass ultra-low-noise p-type modified-electrode HPGe diode (mass ~0.475 kg) with threshold ~330 eV and capacitance ~1 pF, achieving FWHM ~140 eV. They demonstrate pulse-shape discrimination enabling multi-site event identification and show quenching factors in agreement with Lindhard theory, with background control via shielding, anti-Compton veto, and PSD; the approach is scalable to ~10 kg. Beyond CNS, the detector offers sensitivity to light WIMPs and neutrino magnetic moments down to ~2e-11 mu_B, with potential utility for neutrinoless double-beta decay studies.

Abstract

A new type of radiation detector, a p-type modified electrode germanium diode, is presented. The prototype displays, for the first time, a combination of features (mass, energy threshold and background expectation) required for a measurement of coherent neutrino-nucleus scattering in a nuclear reactor experiment. The device hybridizes the mass and energy resolution of a conventional HPGe coaxial gamma spectrometer with the low electronic noise and threshold of a small x-ray semiconductor detector, also displaying an intrinsic ability to distinguish multiple from single-site particle interactions. The present performance of the prototype and possible further improvements are discussed, as well as other applications for this new type of device in neutrino and astroparticle physics (double-beta decay, neutrino magnetic moment and WIMP searches).

Figures (12)

• Figure 1: Energy resolution (FWHM) and effective gain shift observed using low-energy gamma emissions from a collimated $^{241}$Am source at different positions along the longitudinal HPGe crystal axis. The tiny axial dependence of the second (0.15% maximum variation) demonstrates an optimal charge collection even in the presence of a modified electrode configuration.
• Figure 2: A comparison of the energy threshold ($\sim$ 330 eV, 5 sigma from noise centroid) in the modified electrode HPGe with that of a conventional coaxial detector of the same mass ($\sim 475$ g), typically in the few keV region (the particular one used for the figure being relatively low in noise). No instabilities in the threshold have been observed in five months of continuous detector operation. Energies are electron-equivalent, i.e., ionization.
• Figure 3: Effect of detector threshold and energy resolution in the differential rate of recoils expected from reactor antineutrinos (see text).
• Figure 4: Exposure of the prototype to a monochromatic 24 keV reactor neutron beam custom-built to mimic reactor antineutrino recoils other. A Titanium postfilter allows to switch off the dominant 24 keV beam component and with it the neutrino-like recoils, leaving the scarce backgrounds intact other. To illustrate its effect, the vertical arrows mark the energy at which the endpoint of these soft recoils is predicted, based on a full MCNP-Polimi simulation polimi of the experiment and the expected 20% quenching factor. Inset: Signals time-coincident with the thermal peak of a large $^{6}$LiI[Eu] scintillator mounted on a goniometric table allow to select discrete Ge recoil energies, for which quenching factors can then be obtained other. An excellent agreement with Lindhard theory expectations has been observed over the range of recoil energies relevant to the upcoming reactor neutrino experiment (see text).
• Figure 5: Expected antineutrino signal in the planned demonstration experiment, clearly visible above the background goal. The background is scaled down from data acquired with a partial shielding in place, i.e., the spectral shape depicted is representative of actual observations.