Dipole portal to heavy neutral leptons

We consider generic neutrino dipole portals between left-handed neutrinos, photons, and right-handed heavy neutral leptons (HNL) with Dirac masses. The dominance of this portal significantly alters the conventional phenomenology of HNLs. We derive a comprehensive set of constraints on the dipole portal to HNLs by utilizing data from LEP, LHC, MiniBooNE, LSND as well as observations of Supernova 1987A and consistency of the standard big bang nucleosynthesis. We calculate projected sensitivities from the proposed high-intensity SHiP beam dump experiment, and the ongoing experiments at the Short-Baseline Neutrino facility at Fermilab. Dipole mediated Primakoff neutrino upscattering and Dalitz-like meson decays are found to be the main production mechanisms in most of the parametric regime under consideration. Proposed explanations of LSND and MiniBooNE anomalies based on HNLs with dipole-induced decays are found to be severely constrained, or to be tested in the future experiments.

Figure 1

Overview of projected sensitivities (95% CL) and constraints obtained from SHiP, LHC, LEP, Supernova 1987A and experiments at the Short-Baseline Neutrino facility at Fermilab. We also show previously calculated favored regions of interest (ROI) in parameter space for MiniBooNE and LSND, and constraints from NOMAD. Limits are shown for the dimension 5 ($\gamma$ mediator) and dimension 6 ($\gamma +Z$ mediators) extensions. See Table 2 for an explanation of the labels. Each curve is discussed and presented in the paper.

Figure 2

Dipole portal processes for $N$: (a) Production of $N$ from off-shell neutrinos arising from weak meson decays (e.g., from $\pi$, $K\to {\mu }^{+}{\nu }^{*}$); (b) Production of $N$ from off-shell photons arising from Dalitz-like meson decays (e.g., from ${\pi }^0$, $\eta \to {\gamma }^{*}\gamma$); (c) Production of $N$ from on-shell neutrinos via Primakoff-type upscattering (via photon exchange with the nucleus); (d) Decays of $N$ to single photon final states (the main signal studied in this paper). Processes (a) and (b) are important for production of low mass $N$ at neutrino experiments. Process (c) dominates production in supernovas at lower $N$ masses, and at neutrino experiments. Process (d) is relevant for energy injection at BBN, for neutrino beam dump experiments, and controls the escape probabilities in supernova for large $N$ masses.

Figure 3

Loop level contribution to the $\nu$ mass mixing matrix in the presence of a Majorana mass term for the heavy neutral lepton $N$. With only Dirac masses, such diagrams will not be generated.

Figure 4

Tree level neutrino scattering process with a final state photon, arising from dipole portal to HNL. We work in the narrow width approximation, and assume the above diagram factorizes.

Figure 5

95% CL limits for HNL particles using MiniBooNE and LSND ${\nu }_e\mathrm{CC}$ measurements. In light of the experimental anomaly, background option 1 (Bkg 1) uses the data and backgrounds as is, option 2 includes an alternative stronger $\mathrm{\Delta }\to \gamma N$ background estimate [18], and option 3 includes only neutrino energies in the anomaly-free region ($E_{{\nu }_e}>470\mathrm{MeV}$). We also overlay regions of interest (ROI) from the MiniBooNE and LSND anomalies (see text).

Figure 6

Projected 95% CL sensitivities at Fermilab’s upcoming Short-Baseline Neutrino program [23]. Results for electron (black) and muon (red) dipole couplings are shown for the SBND near detector (solid) and the MicroBooNE middle detector (dotted). Backgrounds are calculated based on expected lifetime single photons (see text).

Figure 7

Projected 95% CL sensitivities at SHiP for muon neutrino dipole moments. Solid (dotted) lines indicate the main (ECC) detector, and black (red) lines represent 10 (1000) background events during the lifetime of the experiment. We also overlay existing constraints [44, 45] from NOMAD.

Figure 8

Projected 95% CL sensitivities at SHiP for electron (black curve) and tau (red curve) neutrino dipole moments. Solid (dotted) lines indicate the main (ECC) detector. In this plot, we assume 100 background events.

Figure 9

95% CL sensitivities at LHC and LEP. Limits are shown for the dimension 5 ($\gamma$ mediator) and dimension 6 ($\gamma$, $Z$ and $W^{\pm }$ mediators) extensions. For the LHC 8 TeV results involving a photon and charged lepton final state, we consider various relations between the production ($d_{\gamma W}^a$) and decay ($d_{\gamma }$) couplings. See Table 2 for an explanation of the plot labels.

Figure 10

Radial profiles of the number density, temperature, and chemical potentials at one second after the bounce from the simulation of an $18M_{\odot }$ progenitor [65].

Figure 11

Emissivity and optical depth constraints (red) from supernovae SN 1987A, and parameter space facilitating its conversion to a neutron star (green). We also show lines of constant HNL lifetimes to gauge where BBN might be affected. Two radii of production $r_0$ are plotted for comparison, with one at the hottest densest radius $r_0=10\mathrm{km}$ and one closer to the edge of the high density region $r_0=14\mathrm{km}$. The gravitational trapping becomes significant for HNLs with mass above the vertical gray line, labeled “Gravity”.

Figure 12

Efficiency of imposing energy and angular cuts on the photon produced by a 50 MeV heavy neutral lepton.