Results on light dark matter particles with a low-threshold CRESST-II detector

TL;DR

Using data from the CRESST-II Phase 2 module Lise with a 0.3 keV nuclear-recoil threshold, this work performs a blind analysis to search for light dark matter scattering off CaWO4 nuclei. It quantifies the trigger efficiency, energy calibration, and signal-survival probabilities, and derives a data-driven background model to set 90% CL upper limits on the DM-nucleon cross-section via Yellin's optimum interval method. The analysis demonstrates that lowering the energy threshold is crucial for sensitivity to sub-GeV DM, extending the reach down to ~0.5 GeV/c^2 and establishing a new benchmark for CaWO4-based detectors. It also outlines CRESST-III upgrades (smaller absorbers, improved light collection, and in-house crystals) expected to push thresholds toward the 100 eV scale.

Abstract

The CRESST-II experiment uses cryogenic detectors to search for nuclear recoil events induced by the elastic scattering of dark matter particles in CaWO$_4$ crystals. Given the low energy threshold of our detectors in combination with light target nuclei, low mass dark matter particles can be probed with high sensitivity. In this letter we present the results from data of a single detector module corresponding to 52 kg live days. A blind analysis is carried out. With an energy threshold for nuclear recoils of 307 eV we substantially enhance the sensitivity for light dark matter. Thereby, we extend the reach of direct dark matter experiments to the sub-region and demonstrate that the energy threshold is the key parameter in the search for low mass dark matter particles.

Figures (8)

• Figure 1: Fraction of heater pulses triggering, injected with discrete energies $E_{\text{inj}}$. The error bars indicate the statistical (binomial) uncertainty at the respective energy. The solid red curve is a fit with the sum of a scaled error function and a constant pile-up probability $p_{\text{pile-up}}$ (blue dashed line). The latter is independent of the injected energy since it is caused by random coincidences of the injected heater pulses with any other trigger. The fit yields an energy threshold of $E_{\text{th}}=\unit[(307.3\pm 3.6)]{eV}$ and a width of [$\sigma=(82.0\pm 4.2)$]eV.
• Figure 2: The solid lines represent the signal survival probability after successive application of the cuts, as discussed in the text. The simulated pulses correspond to nuclear recoil events at discrete energies starting from the threshold of [0.3]keV (data points).
• Figure 3: Spectrum of all events (blue) up to [9]keV, truncated at a bin content of 100. The red line corresponds to a data-driven background model including a linearly decreasing background (dashed black line) and contributions from $^{179}$Ta (M1) ([2.7]keV), $^{55}$Fe ([6.0]keV & [6.6]keV), and Cu fluorescence ([8.1]keV). The drop towards lower energies results from the declining signal survival probability (see figure \ref{['fig:Efficiencies']}). This plot illustrates that the detector Lise exhibits an almost flat background level down to the threshold energy.
• Figure 4: Trigger efficiency (black, left y-axis) as a function of the complete measurement time. Each data point is determined as the fraction of [0.4]keV heater pulses triggering in the respective time bin. For the same time bins also the fitted peak positions and the fit errors for Mn K$_{\alpha}$ (blue) and K$_{\beta}$ (magenta) from the $^{55}$Fe-source are drawn (scale on the right y-axis). Only small deviations from the fitted total mean values (dashed lines) are observed. The constant spread of the corresponding data (grey points) illustrates a stable line width.
• Figure 5: Data taken with the detector module Lise depicted in the light yield - energy plane. The solid lines mark the [90]% upper and lower boundaries of the e$^-/\gamma$-band (blue), the band for recoils off oxygen (red) and off tungsten (green). The dashed blue line corresponds to the lower [5]$\sigma$ boundary of the e$^-/\gamma$-band, events outside are marked with a blue circle. The upper boundary of the acceptance region (yellow area) is set to the middle of the oxygen band (dashed dotted red), the lower one to the [99.5]% lower boundary of the tungsten band. Events therein are additionally highlighted in red.