Measurement of the double-beta decay half-life and search for the neutrinoless double-beta decay of ${}^{48}\mathrm{Ca}$ with the NEMO-3 detector

The NEMO-3 experiment at the Modane Underground Laboratory investigates the double-beta decay of ${}^{48}\mathrm{Ca}$. Using 5.25 yr of data recorded with a 6.99 g sample of ${}^{48}\mathrm{Ca}$, approximately 150 double-beta decay candidate events are selected with a signal-to-background ratio greater than 3. The half-life for the two-neutrino double-beta decay of ${}^{48}\mathrm{Ca}$ is measured to be $T_{1/2}^{2\nu }=[6.4_{-0.6}^{+0.7}{(\mathrm{stat})}_{-0.9}^{+1.2}(\mathrm{syst})]\times 10^{19}\mathrm{yr}$. A search for neutrinoless double-beta decay of ${}^{48}\mathrm{Ca}$ yields a null result, and a corresponding lower limit on the half-life is found to be $T_{1/2}^{0\nu }>2.0\times 10^{22}\mathrm{yr}$ at 90% confidence level, translating into an upper limit on the effective Majorana neutrino mass of $⟨m_{\beta \beta }⟩<6.0–26\mathrm{eV}$, with the range reflecting different nuclear matrix element calculations. Limits are also set on models involving Majoron emission and right-handed currents.

Figure 1

An image of the NEMO-3 source foils in the vicinity of the ${}^{48}\mathrm{Ca}$ source, showing the rate of single-electron events (arbitrary scale) from the nine ${\mathrm{CaF}}_2$ disks mounted beneath the ${}^{96}\mathrm{Zr}$ source and next to the ${}^{150}\mathrm{Nd}$ source foil. The region below the ${}^{48}\mathrm{Ca}$ source is populated with Mylar foils of the type used in the source construction. The more distant ${}^{82}\mathrm{Se}$ and ${}^{100}\mathrm{Mo}$ sources do not contribute to backgrounds in the analysis of ${}^{48}\mathrm{Ca}$. A copper tube forming part of the calibration system lies between the ${}^{48}\mathrm{Ca}/{}^{96}\mathrm{Zr}$ sources and the ${}^{82}\mathrm{Se}$ foil.

Figure 2

The energy spectrum of electrons in events containing a single electron compared to the fitted background contributions (top panel) and as a ratio to the total Monte Carlo prediction (bottom panel). The dominant contribution is from ${}^{90}\mathrm{Sr}/{}^{90}\mathrm{Y}$ with an end point of 2.3 MeV. At higher energies the background from the radon daughter ${}^{214}\mathrm{Bi}$ becomes significant. There are subleading contributions from other internal backgrounds, external backgrounds and contamination of events from neighboring source foils.

Figure 3

The distribution of the total energy measured in events with one electron and one photon (top panel) and as a ratio to the total Monte Carlo prediction (bottom panel). The higher end of the distribution is dominated by the contribution due to the $\beta$ decay of ${}^{208}\mathrm{Tl}$ to ${}^{208}\mathrm{Pb}$ which always occurs with the emission of one or more photons, releasing a total energy of 5 MeV.

Figure 4

The length of delayed tracks in $1e1\alpha$ events showing the contribution from radon progeny deposited on the wires of the tracking detector (red) and surface of the source foil (blue). A small contribution consistent with having an origin inside the Mylar surrounding the ${\mathrm{CaF}}_2$ disks, peaking at small track lengths, is also shown. The ratio of data events to the total Monte Carlo prediction is shown in the bottom panel.

Figure 5

The distribution of the summed electron energy in two-electron events. At low energies ${}^{90}\mathrm{Sr}/{}^{90}\mathrm{Y}$ background events dominate, while other internal, external, radon and non-${}^{48}\mathrm{Ca}$ backgrounds are relatively small. The $2\nu \beta \beta$ signal is clearly visible at higher energies. Four open histograms represent limits on non-standard model double-beta decay processes, with 90% confidence level upper limits on the corresponding event yields $N_{\mathrm{obs}}$ also given. The ratio of data events to the total Monte Carlo prediction is shown in the bottom panel.