Searches for double beta decay of ${}^{134}\mathrm{Xe}$ with EXO-200

Searches for double beta decay of ${}^{134}\mathrm{Xe}$ were performed with EXO-200, a single-phase liquid xenon detector designed to search for neutrinoless double beta decay of ${}^{136}\mathrm{Xe}$. Using an exposure of $29.6\mathrm{kg}·\mathrm{yr}$, the lower limits of ${\mathrm{T}}_{1/2}^{2\nu \beta \beta }>8.7\times {10}^{20}\mathrm{yr}$ and ${\mathrm{T}}_{1/2}^{0\nu \beta \beta }>1.1\times {10}^{23}\mathrm{yr}$ at 90% confidence level were derived, with corresponding half-life sensitivities of $1.2\times {10}^{21}\mathrm{yr}$ and $1.9\times {10}^{23}\mathrm{yr}$. These limits exceed those in the literature for ${}^{134}\mathrm{Xe}$, improving by factors of nearly $10^5$ and 2 for the two antineutrino and neutrinoless modes, respectively.

Figure 1

Energy spectral agreement between data and MC simulation in SS (top) and MS (bottom) events. The comparison is presented between 460 keV and 1600 keV for two calibration sources, ${}^{228}\mathrm{Th}$ and ${}^{226}\mathrm{Ra}$, deployed near the cathode of the detector for calibration runs. The background-subtracted ${}^{136}\mathrm{Xe}\text{−}2\nu \beta \beta$ spectrum is compared to MC only in SS events. A worsening of the agreement in SS $\gamma$-like spectra is observed for energies $\lesssim 600\mathrm{keV}$. This effect was considered in the searches as discussed in Sec. 4.

Figure 2

Profile-likelihood ratio, $\mathrm{\Delta }\mathrm{NLL}$, for ${}^{134}\mathrm{Xe}$ $2\nu \beta \beta$ and $0\nu \beta \beta$. The dashed lines represent the $1\sigma$ and 90% C.L.’s, assuming the validity of Wilks’s theorem [22, 23]. The latter intersects the profile curves at 5900 and 730 counts for $2\nu \beta \beta$ and $0\nu \beta \beta$, respectively.

Figure 3

The top plot shows the SS energy projection for the results of the ${}^{134}\mathrm{Xe}\text{−}2\nu \beta \beta$ analysis, using the fitted central values. While the $0\nu \beta \beta$ search is performed in a separate fit with a higher energy threshold, it is overlaid here for comparison since the backgrounds at these energies are the same. The last bin is the summed contents of all bins between 1700 keV and 9800 keV. The middle plot shows the same results, where the ${}^{136}\mathrm{Xe}\text{−}2\nu \beta \beta$ component is subtracted from data and best-fit values. The bottom plot presents the residuals between data and best fit normalized to the Poisson error.

Figure 4

Contour lines of the profile-likelihood ratio, $\mathrm{\Delta }\mathrm{NLL}$, scanned for the ${}^{85}\mathrm{Kr}$ $\beta$ and ${}^{134}\mathrm{Xe}$ $2\nu \beta \beta$ components using two NLL functions. The blue solid lines were obtained using the same function as the ${}^{134}\mathrm{Xe}$ $2\nu \beta \beta$ search, which accounts for all the systematic errors. The green dashed lines were evaluated without consideration of the normalization errors. The table shows $\mathrm{\Delta }\mathrm{NLL}$ values of various C.L.’s for 1 and 2 d.o.f.