Cosmological and astrophysical bounds on a heavy sterile neutrino and the KARMEN anomaly

TL;DR

The paper investigates whether a $33.9\,\mathrm{MeV}$ sterile neutrino as a solution to the KARMEN anomaly can survive cosmological and astrophysical constraints. It develops a detailed but approximate kinetic framework to track the heavy neutrino’s production, decay, and impact on light-neutrino spectra during Big Bang Nucleosynthesis, translating these effects into bounds on the lifetime $\tau_{\nu_s}$ that depend on the effective number of neutrino species $\Delta N$. In parallel, it analyzes SN 1987A constraints, showing that both small and large mixing scenarios demand $\tau_{\nu_s}\gtrsim6\times10^4$ s to avoid excessive energy loss or transport, effectively ruling out the KARMEN interpretation of a 33.9 MeV sterile neutrino. Consequently, the combined cosmological and SN bounds leave no viable lifetime window for this particle under standard assumptions, unless new physics (e.g., anomalous nucleon interactions) decouples SN bounds while preserving BBN sensitivity. The results underscore the strong tension between the KARMEN anomaly and existing cosmological/astrophysical data.

Abstract

Constraints on the lifetime of the heavy sterile neutrino, that was proposed as a possible interpretation of the KARMEN anomaly, are derived from primordial nucleosynthesis and SN 1987A. Together with the recent experimental bounds on the nu_s lifetime, SN 1987A completely excludes this interpretation. Nucleosynthesis arguments permit a narrow window for the lifetime in the interval 0.1-0.2 sec. If nu_s possesses an anomalous interaction with nucleons, the SN bounds may not apply, while the nucleosynthesis ones would remain valid.

Figures (7)

• Figure 1: The normalized number density, $x^3\, n_{\nu_s}$, as a function of $x$ for the lifetimes 0.1 sec (solid), 0.2 sec (dashed) and 0.3 sec (dotted).
• Figure 2: The normalized energy density, $x^4\, \rho_{\nu_s}$, as a function of $x$ for the lifetimes 0.1 sec (solid), 0.2 sec (dashed) and 0.3 sec (dotted).
• Figure 3: A snap-shot of the distribution function, $f_{\nu_s}(y)$, at the time $x=1$ for the lifetimes 0.1 sec (solid), 0.2 sec (dashed) and 0.3 sec (dotted).
• Figure 4: The energy density of all the active neutrinos divided by the energy density in the electromagnetic plasma, $\sum \rho_{\nu}/\rho_{EM}$, as a function of $x$ for lifetimes 0.1 sec (solid), 0.2 sec (dashed) and 0.3 sec (dotted).
• Figure 5: A snap-shot of the spectrum of $\nu_e$, namely $y^2 f_{\nu_e}$ and the distortion $y^2 \delta f_{\nu_e}$ at $x=1$. The lifetimes are 0.1 sec (solid), 0.2 sec (dashed) and 0.3 sec (dotted).