New limits on spin-independent and spin-dependent couplings of low-mass WIMP dark matter with a germanium detector at a threshold of 220 eV

An energy threshold of $(220\pm 10)\mathrm{eV}$ was achieved at an efficiency of 50% with a four-channel ultralow-energy germanium detector, each with an active mass of 5 g. This provides a unique probe to weakly interacting massive particles (WIMP) dark matter with mass below 10 GeV. With a data acquisition live time of 0.338 kg-day at the Kuo-Sheng Laboratory, constraints on WIMPs in the galactic halo were derived. The limits improve over previous results on both spin-independent WIMP-nucleon and spin-dependent WIMP-neutron cross-sections for WIMP mass between 3 and 6 GeV. Sensitivities for full-scale experiments are projected. This detector technique makes the unexplored sub-keV energy window accessible for new neutrino and dark matter experiments.

Figure 1

The distributions of noise fluctuation of RT events and the maximum amplitudes of physics events in various energy bins. The discriminator threshold level is also shown.

Figure 2

(a) Measured energy spectrum of the ULEGe with ${}^{55}\mathrm{Fe}$ source together with x-ray peaks from various materials. The black histogram represents events selected by PSD cuts. Deviations from the expected flat spectra contribute to PSD efficiencies. (b) Scatter plots of the ${\mathrm{SA}}_6^P$ vs ${\mathrm{SA}}_{12}^{\mathrm{T}}$ signals, for both calibration and physics events. The PSD selection is shown.

Figure 3

Selection efficiencies of the PSD cut, as derived from the ${}^{55}\mathrm{Fe}$ calibration and in situ data with ACV tags. Also shown are the best-fit $1\sigma$ region and the trigger efficiency for physics events recorded by the DAQ system.

Figure 4

The survival fraction of events at $E=200–300\mathrm{eV}$ with ACV tags in PSD selection vs the relative timing between ACV signals and ULEGe triggers. Overlaid are the coincidence time intervals derived independently from hardware timing. The data point at negative time is due to events without ACV tags, and corresponds to $(1.7\pm 0.3)\times {10}^{-4}$.

Figure 5

The measured spectrum of ULEGe with 0.338 kg-day of data, after CRV, ACV, and PSD selections. Background spectra of the CRESST-I experiment [10] and the HPGe [12] are overlaid for comparison. The expected nuclear recoil spectra for two cases of ($m_{\chi }$, ${\sigma }_{\chi N}^{\mathrm{SI}}$) are superimposed onto the spectrum shown in linear scales in the inset.

Figure 6

Exclusion plots of (a) the spin-independent $\chi N$ and (b) the spin-dependent $\chi$-neutron cross-sections versus WIMP mass, displaying the KS-ULEGe limits and those defining the current boundaries [10, 27]. The DAMA-allowed regions [5] are superimposed. The striped regions are those favored by SUSY models [4]. Projected sensitivities of full-scale experiments are indicated as dotted lines. (c) Constraints at $m_{\chi }=5\mathrm{GeV}$ on the effective axial four-fermion $\chi$-proton ($a_p$) and $\chi$-neutron ($a_n$) spin-dependent couplings, in units of $2\sqrt{2}{\mathrm{G}}_{\mathrm{F}}$ following the formulation of Ref. [29]. The shaded area at the origin is the combined allowed region.