Constraining Heavy Neutral Leptons Coupled to the Tau-Neutrino Flavor at the Large Hadron Collider

TL;DR

This work investigates heavy neutral leptons (HNLs) that couple predominantly to the tau neutrino flavor by extending displaced-vertex searches at the LHC to the process $pp \to W \to \tau N$ with $N \to \ell^+\ell^-\nu_\tau$. Using generator-level Monte Carlo simulations and ATLAS-inspired displaced-vertex selections, including a piecewise invariant-mass cut in the DV plane, the authors map sensitivity in the $(m_N, Θ_τ^2)$ parameter space for Run-2 and HL-LHC scenarios. They show that Run-2 data can already improve bounds by more than an order of magnitude, with HL-LHC potentially extending reach by up to three orders of magnitude in certain mass ranges, highlighting the importance of large-radius tracking to maximize DV sensitivity. The study provides a concrete framework for reinterpretation of LHC data toward tau-focused HNLs and underscores the potential to connect neutrino mass generation with baryogenesis and dark-sector phenomenology through displaced-signature searches.

Abstract

Displaced vertex (DV) signatures at colliders offer a powerful probe of new long-lived particles beyond the Standard Model. Among the best-motivated candidates are heavy neutral leptons (HNLs) - heavier counterparts of Standard Model neutrinos - which can account for the origin of neutrino masses and potentially produce di-leptonic DV signatures. In this study, we demonstrate how existing DV searches at the LHC can be extended to probe HNLs that couple predominantly to the tau-neutrino flavor. While current search strategies rely on identifying a prompt lepton alongside a displaced vertex, we show that analyzing events without a prompt lepton enables sensitivity to the process $pp \to W \to τN$, where the tau decays hadronically and the HNL subsequently decays to a lepton pair and a neutrino. We perform detailed Monte Carlo simulations of this process with HNLs decaying to $μ^+μ^-$ or $e^+e^-$ final states, apply ATLAS-inspired selection criteria, and optimize signal sensitivity. In particular, we demonstrate that appropriate cuts in the plane of di-lepton invariant mass and DV radial position significantly enhance signal visibility. We propose several such optimized strategies and show that even with Run 2 data $139~\text{fb}^{-1}$ , existing bounds can be improved by more than an order of magnitude. Future high-luminosity runs may strengthen sensitivity by up to three orders of magnitude compared to current limits.

Figures (10)

• Figure 1: Main process. Production of an HNL through mixing with the $\tau$ flavour and its decay via a neutral current.
• Figure 2: The branching ratios of an HNL decaying into a lepton pair and $\nu_\tau$ for the flavor mixing pattern $(\Theta_e^2, \Theta_\mu^2, \Theta_\tau^{2}) = (0,0,1)$. The branching ratios are independent of the specific value of $\Theta_\tau^{2}$.
• Figure 3: Sensitivity of the combined channel as a function of the number of observed events. The signal sensitivity (for 3, 10, or 100 expected events) in the combined $e^{+}e^{-} + \mu^{+}\mu^{-}$ channel at integrated luminosities of $139\per f\barn$ (top), $300\per f\barn$ (middle), and $3000\per f\barn$ (bottom). The left column (panels a, c, and e) corresponds to the piecewise invariant mass selection, while the right column (b, d, and f) corresponds to the flat invariant mass selection. Notably, in the flat selection case, no regions with $100$ events are observed (panel f), and even the $10$-event regions disappear in panels b and d, highlighting the critical role of the DV cut choice.
• Figure 4: The signal sensitivity for $N \geq 3$ events of the $e^{+}e^{-}$ (red), $\mu^{+}\mu^{-}$ (blue), and the combined $e^{+}e^{-} + \mu^{+}\mu^{-}$ (yellow) channel at 139 fb$^{-1}$ (top), 300 fb$^{-1}$ (middle) and 3000 fb$^{-1}$ (bottom). The left column (a,c, and e) is the piecewise invariant mass cut, and the right column (b, d, and f) is the flat invariant mass cut.
• Figure 5: The fraction of HNLs that decay within the fiducial volume defined by a minimal decay radius, $r_{min}$, and a cylinder with longitudinal, $z_{max}$, and transverse, $r_{max}$, constraints, as specified in Table \ref{['tab:cut_table']}. The left panel corresponds to electrons, while the right panel corresponds to muons. The black dashed lines represent lines of constant decay length, $c\tau_N\gamma$ where $\gamma$ is the average Lorentz-factor for the HNL of a given mass.