Low-mass dark matter search using ionization signals in XENON100

We perform a low-mass dark matter search using an exposure of $30\mathrm{kg}\times \mathrm{yr}$ with the XENON100 detector. By dropping the requirement of a scintillation signal and using only the ionization signal to determine the interaction energy, we lowered the energy threshold for detection to 0.7 keV for nuclear recoils. No dark matter detection can be claimed because a complete background model cannot be constructed without a primary scintillation signal. Instead, we compute an upper limit on the WIMP-nucleon scattering cross section under the assumption that every event passing our selection criteria could be a signal event. Using an energy interval from 0.7 keV to 9.1 keV, we derive a limit on the spin-independent WIMP-nucleon cross section that excludes WIMPs with a mass of $6\mathrm{GeV}/c^2$ above $1.4\times {10}^{-41}{\mathrm{cm}}^2$ at 90% confidence level.

Figure 1

Rate of events ($\mathrm{S}2>80\mathrm{PE}$) as a function of the time difference from the previous recorded event. A cut is set at 10 ms to remove a population of events of small S2 signals (e.g., photo-ionization) that appears within a few ms from the previous trigger.

Figure 2

S2 asymmetry parameter for ${}^{241}\mathrm{AmBe}$ calibration data in the liquid and a population of events produced in the xenon gas phase. We select interactions in the gas by requiring an S1 signal, small drift time, and a large S2 width using ${}^{60}\mathrm{Co}$ and DM search data. An S2 asymmetry cut set at 0.17 is used to reject the gas event population in the dark matter data.

Figure 3

The analysis acceptance (red triangles) and the trigger efficiency (blue circles). The purple dashed line is the analysis threshold (80 PE).

Figure 4

Energy distribution of the events remaining in the data set after all data selection cuts. As an example, the expected spectrum for a WIMP of $6\mathrm{GeV}/c^2$ and a spin-independent WIMP-nucleon scattering cross section of $1.5\times {10}^{-41}{\mathrm{cm}}^2$ is also shown. The corresponding nuclear recoil energy scale is indicated on the top axis. The charge yield model assumed here has a cutoff at 0.7 keV, which truncates the WIMP spectrum. The optimum interval (thick red line) is found in the S2 range [98, 119] PE and contains 1173 events.

Figure 5

Charge yield ($Q_y$) as a function of energy for nuclear recoils (keV). This analysis employs the conservative nuclear recoil charge yield model of Bezrukov et al. (electric field independent) [15], given by the green line. It agrees with the measurement of XENON100 ($\mathrm{E}=0.53\mathrm{kV}/\mathrm{cm}$) [14] (red triangles), while the NEST model ($\mathrm{E}=0.73\mathrm{kV}/\mathrm{cm}$) [16] (dashed black) and the recent measurement of LUX ($\mathrm{E}=0.18\mathrm{kV}/\mathrm{cm}$) [17] (blue points) predict higher yields. To account for the mild discrepancies at low energies, we use the model from Bezrukov et al. and conservatively assume $Q_y=0$ below 0.7 keV.

Figure 6

S2 spectrum of ${}^{241}\mathrm{Am}\mathrm{Be}$ calibration data compared to simulations using the $Q_y$ from Bezrukov et al. [15] with no energy cutoff.

Figure 7

WIMP exclusion limit on the spin-independent WIMP-nucleon scattering cross section at 90% confidence level. Limits from the LUX [17], XENON100 [10], SuperCDMS [21], CDMSlite [22], XENON10 [8], CRESST-II [23] and PICO-2L [24] experiments are shown. The claims from DAMA/LIBRA experimental data [25] and CDMS-II (Si detectors) [7] are also shown. The limit from this analysis is shown with the thick blue line, and it improves the XENON100 result [10] (dashed blue line) for WIMP masses below $\sim 7.4\mathrm{GeV}/c^2$.