Impact of neutrino decays on the Cosmic Neutrino Background anisotropies

TL;DR

This work tests invisible, non-relativistic neutrino decays ν_H → ν_l + φ as a non-standard interaction by implementing late-time decays in a linear Einstein–Boltzmann solver and computing the CνB angular power spectra for several lifetimes and decay channels. Using a modified CLASS++ code that handles a massive daughter and dark radiation, the authors show that such decays imprint order-unity to tens-of-percent changes in the CνB spectra for ℓ ≤ 17, while leaving CMB signatures far smaller, thus enabling potential detection by future PTOLEMY measurements of the first few multipoles. The results indicate that the CνB anisotropies can distinguish decay scenarios (solar vs atmospheric mass gaps, single vs two-channel) and motivate polarized-tritium experiments to probe parameter regions inaccessible to the CMB. The paper also provides a public code implementation and outlines future work to connect CνB anisotropies to capture-rate maps, non-linear effects, and broader decaying-warm-dark-matter models.

Abstract

The anisotropies of the Cosmic Neutrino Background (C$ν$B) offer an ideal tool to test non-standard neutrino interactions, since they directly trace the perturbations in the neutrino distribution function. Here, we study how invisible neutrino decays impact the C$ν$B anisotropies, in a framework where neutrinos decay non-relativistically to dark radiation and lighter neutrinos in a manner consistent with the measured mass splittings. For this purpose, we perform the first implementation of such a late-time neutrino decay scenario within a linear Einstein-Boltzmann solver, and compute the C$ν$B angular power spectra from the Boltzmann hierarchy solutions for a range of lifetimes and decay channels. We find that neutrino decays leave very strong signatures on the C$ν$B angular spectra, about two orders of magnitude larger than on the CMB angular spectra, particularly for lifetimes comparable to the age of the Universe. We show that a future polarized tritium target run of the PTOLEMY experiment, with sufficient counting statistics to measure just the first $\sim 15$ multipoles of the neutrino sky map, could test neutrino decay models that remain undetectable with CMB data.

Figures (6)

• Figure 1: Final phase-space distributions of $\nu_l$ in the neutrino decay scenarios A2 (left panel) and A3 (right panel) for a range of neutrino lifetimes $\tau_\nu$ and fixed daughter mass $m_{\nu l}$. These are compared to a standard Fermi-Dirac distribution. The comoving momentum $q$ is given in units of $T_{\nu 0}$.
• Figure 2: Similar as \ref{['fig:PSD_A1_A2']}, but for scenarios B1 (left panel) and B2 (right panel).
• Figure 3: Residuals of the CMB (left panel) and C$\nu$B (right panel) temperature power spectra in the neutrino decay scenario A2, for several neutrino lifetimes $\tau_\nu$ and a fixed daughter mass $m_{\nu l} = 0.03 \ \rm{eV}$. All residuals are taken with respect to the stable limit ($\tau_\nu \rightarrow \infty$). Various contributions to the 1$\sigma$ fractional uncertainty are shown: Planck 2018 measurement errors for the CMB, and cosmic variance and PTOLEMY counting statistics (assuming a total of $N=10^5$ capture events) for the C$\nu$B. The C$\nu$B curves are smoothed with a Savitzky-Golay filter.
• Figure 4: Similar as \ref{['fig:residuals_A2']}, but for the neutrino decay scenario A3 and a daughter mass $m_{\nu l} = 0.05 \ \mathrm{eV}$.
• Figure 5: Similar as \ref{['fig:residuals_A2']}, but for the neutrino decay scenario B1 and a daughter mass $m_{\nu l} = 0.03 \ \mathrm{eV}$.