Potential of CCDs for the study of sterile neutrino oscillations via Coherent Neutrino-Nucleus Elastic Scattering

We study the potential of a detector based on CCD sensors (CONNIE experiment) to study neutrino oscillations to sterile states using reactor neutrinos. We calculate the number of events expected in a 1 kg detector and determine the sensitivity to oscillations νe → νs in the Δm412 vs. sin2θes parameter space for various exposures. The sensitivity is compared with the regions excluded by the Daya Bay experiment under the assumption θ24 = θ34 = 0. This work was carried out independently of the CONNIE Collaboration using published information, and its results are not official.


Introduction
While the majority of neutrino oscillation measurements fit well in the 3 massive neutrino model, a handful of anomalous results [1] suggest the existence of at least one sterile neutrino with a mass splitting Δm 2 ∼ 1eV 2 . Models considering sterile neutrinos (3+N) have been considered but different observations impose severe constraints. The 3+1 scenario can be probed with an experiment sensitive to all 3 active flavors (ν e , ν μ , ν τ ) via a neutral current interaction. In the Coherent Elastic Neutrino-Nucleus Scattering (CENNS) [2], a neutrino, or antineutrino, scatters off a nucleus as a whole. When the 4-momentum transfer is small compared to the reciprocal of the nucleus size (Q 2 < 1/R 2 ), the interaction with the constituent nucleons coherently enhances the cross section. The differential cross section is: where E rec is the nuclear recoil energy, G F is the Fermi constant, M is the mass of the nucleus, Z, and N are the number of protons and neutrons in the nucleus respectively, θ W is the weak mixing angle, and f (q) is the nuclear form factor, which is a function only of the 4-momentum transfer q. Since sin 2 θ W ∼ 0.231, the term involving Z nearly cancels numerically and the differential cross section is approximately proportional only to N 2 . CENNS is a Standard Model (SM) process that has never been observed experimentally (see Note in [3]) primarily because the nuclear recoil energies E rec are very small ( 15 keV), and escape the capabilities of most detectors. Charge Coupled Devices (CCD) with very low thresholds (∼10 eV) and relatively large sensitive mass (∼5 g per CCD), have been successfully used in the DAMIC experiment [4] at SNOLAB to perform direct DM searches in the low mass region < 10 GeV/c 2 , demonstrating the unique capabilities of CCDs for the detection of low-energy recoils. This channel could be used as a probe to study oscillations between active and sterile states [5].

The CONNIE experiment
The Coherent Neutrino Nucleus Interaction Experiment (CONNIE) aims to detect for the first time the CENNS using an array of CCDs installed at 30 m from the Angra-II reactor at the Angra dos Reis nuclear power plant in Brazil. A prototype [6] was installed in 2014. The detector was upgraded to 80 g of mass (14 CCDs) in 2016.

Expected event rates
Assuming a 1 kg detector, an energy threshold of ∼ 28 eV, and a Lindhard quenching factor gives an expected rate of ∼16.1 events kg −1 day −1 [7]. It is possible to achieve a 5σ significance (figure 3, right) with an exposure of 42 kg day: a 50 g detector operating for 840 days or a 100 g detector operating for 420 days or a 1 kg detector for 42 days.

Analysis
We divided the energy spectrum ( figure 5, left) in 20 logarithmical uniform bins from 0.01 keV to 0.5 keV and defined the statistic: where D i are the observed events and P i are the predicted events. The observed and predicted events have a signal and background component:  We constructed a reticule of 50×50 points in the parameter space. To calculate the sensitivity of a 1kg CONNIE detector we made the assumption that: which makes χ 2 min = 0. The 90% C.L region was extracted from the excursion χ 2 = 4.6.

Results
The figure 6 shows the expected sensitivity for different exposures. In the case θ 24 = θ 34 = 0 this can be compared with the Daya Bay and Bugey limits [8].

Conclusions
The sensititivity of a CCD experiment like CONNIE to sterile neutrino oscillations in a 3+1 model was calculated. A 10 kg year exposure is competitive in the region Δm 2 > 1eV with the recent Daya Bay limits assuming θ 24 = θ 34 = 0. This result can be combined with ν μ andν μ like MINOS to constraint ν μ → ν e andν μ →ν e transitions. This will be subject of a future work.