Search for electron-neutrino transitions to sterile states in the BEST experiment

The Baksan Experiment on Sterile Transitions (BEST) probes the gallium anomaly and its possible connections to oscillations between active and sterile neutrinos. Based on the Gallium-Germanium Neutrino Telescope (GGNT) technology of the SAGE experiment, BEST employs two zones of liquid Ga target to explore neutrino oscillations on the meter scale. Oscillations on this short scale could produce deficits in the ${}^{71}\mathrm{Ge}$ production rates within the two zones, as well as a possible rate difference between the zones. From July 5th to October 13th 2019, the two-zone target was exposed to a primarily monoenergetic, 3.4-MCi ${}^{51}\mathrm{Cr}$ neutrino source 10 times for a total of 20 independent ${}^{71}\mathrm{Ge}$ extractions from the two Ga targets. The ${}^{71}\mathrm{Ge}$ production rates from the neutrino source were measured from July 2019 to March 2020. At the end of these measurements, the counters were filled with ${}^{71}\mathrm{Ge}$ doped gas and calibrated during November 2020. In this paper, results from the BEST sterile neutrino oscillation experiment are presented in details. The ratio of the measured ${}^{71}\mathrm{Ge}$ production rates to the predicted rates for the inner and the outer target volumes are calculated from the known neutrino capture cross section. Comparable deficits in the measured ratios relative to predicted values are found for both zones, with the $4\sigma$ deviations from unity consistent with the previously reported gallium anomaly. If interpreted in the context of neutrino oscillations, the deficits give best-fit oscillation parameters of $\mathrm{\Delta }{m}^2=3.3_{-2.3}^{+\infty }{\mathrm{eV}}^2$ and ${\sin }^{2}2\theta =0.{42}_{-0.17}^{+0.15}$, consistent with ${\nu }_e\to {\nu }_s$ oscillations governed by a surprisingly large mixing angle.

Figure 1

The Ga target and extraction piping diagram also indicating the source-handling apparatus.

Figure 2

Flowchart showing the experimental steps of the BEST experiment. Time spent on each step and relevant efficiencies are also presented.

Figure 3

Flat-top pulses measured in proportional counters. (top) Saturated event induced by high-energy $\alpha$ particles. (bottom) Background candidate event that originates from high-voltage breakdown.

Figure 4

Typical pulse shapes measured in proportional counters. (top) ${}^{71}\mathrm{Ge}$ K peak candidate event with $T_N=7.44$ ns. (bottom) Background candidate event induced by either Compton scattering or high-energy $\beta$ particles in K-peak energy region. The slow background event with extended ionization has much slower fall time ($T_N=30.67$ ns) when the pulse begins at $\approx 200$ ns than the true ${}^{71}\mathrm{Ge}$ candidate event.

Figure 5

(upper panel) Energy vs rise-time histogram of all events of the inner target after the shield-open cut observed in all ten exposures during the first 30 days after extraction. The live time is 245 days, and 1364 events are shown. (lower panel) The same histogram for the 481 events that occurred during an equal live-time interval beginning at 40 days after extraction.

Figure 6

(upper panel) energy vs rise-time histogram of all events of the outer target after the shield-open cut observed in all ten exposures during the first 30 days after extraction. The live time is 249 days, and 1387 events are shown. (lower panel) The same histogram for the 504 events that occurred during an equal live-time interval beginning at 40 days after extraction.

Figure 7

Plots with energy spectra of selected events for whole dataset showing K and L peaks. (top) Energy spectrum of the inner target. (bottom) Energy spectrum of the outer target.

Figure 8

The geometrical dimensions of the Ga targets. All dimensions are in centimeters. There is an empty segment in the spherical vessel due to the thickened wall and the bottom of the re-entrant tube.

Figure 9

(top) the measured K $+$ L peak rates of the inner target volume. (middle-top) the production rates of the inner target volume normalized to the reference time (UTC $+$ 3 h). The combined results for events in the L and K peaks are shown separately and compared with the predicted rate. (middle-bottom) the measured K $+$ L peak rates of the outer target volume. (bottom) The production rates of the outer target volume normalized to the reference time (UTC $+$ 3 h). The combined results for events in the L and K peaks are shown separately and compared with the predicted rate. The blue (red) region represents the predicted (measured) production rate. The dotted lines enclose the $\pm 1\sigma$ uncertainty regions.

Figure 10

Allowed regions for two BEST results. The best-fit point is ${\sin }^{2}2\theta =0.{42}_{-0.17}^{+0.15}$, $\mathrm{\Delta }{m}^2=3.3_{-2.3}^{+\infty }{\mathrm{eV}}^2$ and is indicated by a point.

Figure 11

Ratios of measured and predicted ${}^{71}\mathrm{Ge}$ production rates in all Ga source experiments. The combined result is shown as a blue shaded band.

Figure 12

Allowed regions for two GALLEX, two SAGE and two BEST results. The best-fit point is ${\sin }^{2}2\theta =0.{34}_{-0.09}^{+0.14}$, $\mathrm{\Delta }{m}^2=1.{25}_{-0.25}^{+\infty }{\mathrm{eV}}^2$ and is indicated by a point.

Figure 13

Exclusion contours of all gallium anomaly experiments: two GALLEX, two SAGE, and two BEST results. The blue solid line and the blue tightly dotted line with the blue shadings show the $2\sigma$ and $3\sigma$ confidence level, respectively. The figure also presents the exclusion contours from Prospect [62], DANSS [63], Stéréo [64], KATRIN [65], the combined analysis of RENO and NEOS data [66], reactor antineutrino anomalies (RAA) [22], interpretations of the MicroBooNE result for the oscillation hypothesis with fixed mixing angle (${\sin }^{2}2\theta$) and profiled over the angle [30], and the model-independent 95% upper bound on ${\sin }^{2}2\theta$ from all solar neutrino experiments [67]. The $2\sigma$ allowed region of Neutrino-4 [68] is also presented and the gray shading represents the merged exclusion of the very short baseline (VSBL) null results.

Figure 14

(top) Allowed regions for two BEST results without the first extractions. The best-fit point is ${\sin }^{2}2\theta =0.35$, $\mathrm{\Delta }{m}^2=6.7{\mathrm{eV}}^2$ and is indicated by a point. (bottom) Allowed regions for two GALLEX, two SAGE, and two BEST results without the first extractions. The best-fit point is ${\sin }^{2}2\theta =0.31$, $\mathrm{\Delta }{m}^2=2.85{\mathrm{eV}}^2$ and is indicated by a point.