Probing unitarity violation of lepton flavor mixing matrix with reactor antineutrinos at JUNO and TAO

TL;DR

The paper develops a self-consistent framework to probe unitarity violation in the lepton flavor mixing matrix using reactor antineutrinos at JUNO and TAO. By modeling the non-unitary mixing as $N = T V$ with $T = {f 1}-\zeta$ (lower-triangular), it derives oscillation probabilities for disappearance and appearance and formulates IBD and elastic $\overline{\nu}_\alpha-e$ scattering cross sections that depend on $N N^\dagger$. The authors show that reactor fluxes are unchanged by non-unitarity, while detection rates carry explicit dependence on the non-unitarity parameters, enabling constraints from TAO and JUNO through a chi-square analysis. They find that TAO can set strong near-term bounds on $|\zeta_{\mu e}|$ and $\zeta_{\mu\mu}$, while JUNO can constrain diagonal differences like $|\zeta_{ee}-\zeta_{\tau\tau}|$ and, in complementary scenarios, $|\zeta_{\mu e}|$, illustrating the value of reactor experiments for independent, near-future tests of non-unitarity. The work emphasizes a self-consistent treatment and the potential to integrate these results with global electroweak and flavor data in full neutrino-mass models.

Abstract

Motivated by the precise measurements of neutrino oscillation parameters at Jiangmen Underground Neutrino Observatory (JUNO), we investigate the possibility of probing the unitarity violation of lepton flavor mixing matrix solely with reactor antineutrinos. First, we stress that it is necessary to reconsider the production and detection of neutrinos in a self-consistent way, apart from neutrino propagation, when analyzing experimental sensitivities to unitarity violation. Then, concentrating on JUNO and its satellite experiment Taishan Antineutrino Observatory (TAO), we demonstrate how the event rates of inverse beta decays (i.e., $\overlineν^{}_e + p \to e^+ + n$) for observing $\overlineν^{}_e \to \overlineν^{}_e$ oscillations, and those of elastic antineutrino-electron scattering (i.e., $\overlineν^{}_α+ e^- \to \overlineν^{}_α+ e^-$ with $α= e, μ, τ$) for $\overlineν^{}_e \to \overlineν^{}_μ$ and $\overlineν^{}_e \to \overlineν^{}_τ$ oscillations, depend on the parameters characterizing unitarity violation. Our investigation will be useful for JUNO and TAO to place independent constraints with more data in the near future.

Figures (1)

• Figure 1: The sensitivity of TAO to the non-unitarity parameters $\left|\zeta_{\mu e}^{}\right|$ and $\zeta_{\mu \mu}^{}$ in the case of normal neutrino mass ordering. The black, blue and red solid curves represent the $1\sigma$, $2\sigma$ and $3\sigma$ contours, respectively. The yellow shaded region indicates the constraints $\left|\zeta_{\mu e}^{}\right| < 1.8 \times 10^{-2}$ and $\zeta_{\mu \mu}^{} < 1.2 \times 10^{-2}$ from previous neutrino oscillation experiments at the 90% confidence level Blennow:2025qgd. As a comparison, the area enclosed by the blue dashed lines denotes the region of $\left|\zeta_{\mu e}^{}\right| < 2.4 \times 10^{-5}$ and $\zeta_{\mu \mu}^{} < 1.1 \times 10^{-5}$, which are the $2\sigma$ upper limits obtained from the global-fit analysis of current flavor and electroweak precision observables Blennow:2023mqx.