First Direct Search for Inelastic Boosted Dark Matter with COSINE-100

A search for inelastic boosted dark matter (IBDM) using the COSINE-100 detector with 59.5 days of data is presented. This relativistic dark matter is theorized to interact with the target material through inelastic scattering with electrons, creating a heavier state that subsequently produces standard model particles, such as an electron-positron pair. In this study, we search for this electron-positron pair in coincidence with the initially scattered electron as a signature for an IBDM interaction. No excess over the predicted background event rate is observed. Therefore, we present limits on IBDM interactions under various hypotheses, one of which allows us to explore an area of the dark photon parameter space that has not yet been covered by other experiments. This is the first experimental search for IBDM using a terrestrial detector.

Figure 1

(a) Production of relativistic BDM ${\chi }_1$ in the Galactic center by the annihilation of heavier dark matter ${\chi }_0$. (b) Illustration of multiple-site hits from an inelastic interaction of BDM for the case of two interactions occurring in two different NaI(Tl) and liquid scintillator (LS) detectors. (c) Illustration of bremsstrahlung radiation-induced hits on two NaI(Tl) or LS detectors.

Figure 2

Schematic of the COSINE-100 detector. The NaI(Tl) (106 kg) detectors are immersed in the 2200 L LAB-LS that are surrounded by layers of shields.

Figure 3

The spectrum of the summed energy from all crystals (filled circle), after the application of all selection criteria of the IBDM candidate events but with no requirement on the amount of energy observed in the crystals, is compared with the expected background (solid histogram). Contributions to the background from muon and $\beta /\gamma$ events caused by radionuclide contaminations are indicated. In the region of interest (energy greater than 4 MeV), muons are the only contribution.

Figure 4

Comparison of the event rate as a function of the crystal multiplicity in data with an energy greater than 3 MeV per crystal (open circle) and the expected background from muons (solid line). The expected signal shapes for two different model parameters (dotted line and dashed line) are also shown. For an easy comparison of the shapes, the dotted line and dashed line spectra are multiplied by a factor of 7 and 3, respectively.

Figure 5

Measured 90% C.L. upper limits from 59.5 days of COSINE-100 data in the $L_{\mathrm{lab}}^{\max }$-$\sigma$ plane are presented for three different benchmark models. These results are compared with the experimental sensitivities of XENON1T with 34.2 days data [46] and DEAP-3600 with 4.4 days data [47] calculated in Ref. [22].

Figure 6

Measured 90% C.L. upper limits (lines) from 59.5 days of COSINE-100 data in terms of the $m_X$ and $\epsilon$ parameters assuming the mediator to be a dark photon are presented for three different benchmark models. These results are compared with the currently excluded parameter space (shaded region) from direct dark photon search experiments E141 [48], NA48 [49], NA64 [50], and BABAR [51] and bounds from the electron’s anomalous magnetic moment $(g-2{)}_e$ [52].