First Constraints on Light Sterile Neutrino Oscillations from Combined Appearance and Disappearance Searches with the MicroBooNE Detector

We present a search for eV-scale sterile neutrino oscillations in the MicroBooNE liquid argon detector, simultaneously considering all possible appearance and disappearance effects within the $3+1$ active-to-sterile neutrino oscillation framework. We analyze the neutrino candidate events for the recent measurements of charged-current ${\nu }_e$ and ${\nu }_{\mu }$ interactions in the MicroBooNE detector, using data corresponding to an exposure of $6.37\times {10}^{20}$ protons on target from the Fermilab booster neutrino beam. We observe no evidence of light sterile neutrino oscillations and derive exclusion contours at the 95% confidence level in the plane of the mass-squared splitting $\mathrm{\Delta }{m}_{41}^2$ and the sterile neutrino mixing angles ${\theta }_{\mu e}$ and ${\theta }_{ee}$, excluding part of the parameter space allowed by experimental anomalies. Cancellation of ${\nu }_e$ appearance and ${\nu }_e$ disappearance effects due to the full $3+1$ treatment of the analysis leads to a degeneracy when determining the oscillation parameters, which is discussed in this Letter and will be addressed by future analyses.

Figure 1

Reconstructed neutrino energy of (a) fully contained ${\nu }_e$ $CC$ and (b) partially contained ${\nu }_e$ $CC$ events. The data points are shown with statistical error bars. The MC predictions of the $3\nu$ hypothesis for ${\nu }_e$ CC events (green) and different types of backgrounds are shown in the stack of histograms. The category “Others” corresponds to the background events originating from either beam neutrino interactions outside the fiducial volume or cosmic-ray muons. The dashed red histogram represents the MC prediction of the $4\nu$ best-fit with $\mathrm{\Delta }{m}_{41}^2=1.295{\mathrm{eV}}^2$, ${\sin }^2{\theta }_{14}=0.936$ (${\sin }^{2}2{\theta }_{ee}=0.240$), and ${\sin }^2{\theta }_{24}=0$ (${\sin }^{2}2{\theta }_{\mu e(\mu \mu )}=0$). The MC predictions and shaded error bands correspond to the central values and systematic uncertainties for each energy bin with constraints (Sec. VI A in Ref. [41]) from the ${\nu }_{\mu }$ CC and ${\pi }^0$ channels as used in the joint fit to the seven channels.

Figure 2

MicroBooNE ${\mathrm{C}\mathrm{L}}_{\mathrm{s}}$ exclusion contours at the 95% C.L. in the plane of $\mathrm{\Delta }{\mathrm{m}}_{41}^2$ and (a) ${\sin }^{2}2{\theta }_{\mu \mathrm{e}}$ or (b) ${\sin }^{2}2{\theta }_{\mathrm{ee}}$. The red solid (dashed) curve represents the MicroBooNE 95% ${\mathrm{C}\mathrm{L}}_{\mathrm{s}}$ data exclusion (Asimov sensitivity) limits after profiling over the mixing angle ${\sin }^2{\theta }_{24}$. The blue long-dashed curve represents the MicroBooNE 95% ${\mathrm{C}\mathrm{L}}_{\mathrm{s}}$ Asimov sensitivity in the scenario of (a) ${\nu }_e$ appearance only or (b) ${\nu }_e$ disappearance only as opposed to the full $3+1$ oscillation result. In (a), the LSND 90% and 99% C.L. allowed regions [25] using the ${\nu }_e$ appearance-only approximation are shown as the light blue and gray shaded areas, respectively. In (b), the cyan shaded area represents the $2\sigma$ allowed region of the gallium anomaly from the experimental results of GALLEX, SAGE, and BEST [20]. The $2\sigma$ allowed region of the Neutrino-4 experiment [24] is also shown in (b).