Long-lived particle production through the PRISM

TL;DR

The paper addresses the challenge of distinguishing production mechanisms for sub-GeV long-lived particles (LLPs) at accelerator-based neutrino facilities. By focusing on the DUNE near detector and the DUNE-PRISM off-axis capability, the authors analyze benchmark models (Dark Higgs, Dark Photon, and a Leptophilic gauge boson) and implement a model-generic framework that treats $m_X$, the LLP lifetime $c\tau$, and production fractions as independent. Using Monte Carlo simulations of LLP production and decay, and a $\chi^2$ analysis that incorporates both the energy spectrum and the transverse-displacement observable $r_T$, they demonstrate that combining energy and spatial information substantially improves discrimination between production channels, including two-body meson decays and proton bremsstrahlung; ternary (three-channel) discrimination is also feasible with modest event counts. The work highlights the practical impact of detector geometry and off-axis sampling for LLP interpretation, providing a broadly applicable methodology for disentangling production mechanisms in fixed-target LLP searches and informing future detector design and analysis strategies.

Abstract

Accelerator-based neutrino experiments offer a competitive environment to search for long-lived particles with sub-GeV masses. Yet, many theoretical models involving such particles predict very similar phenomenology and nearly identical final-state signatures. In view of this, we study the capabilities of upcoming experiments -- specifically the DUNE near detectors -- to distinguish between different classes of long-lived particles and the mechanisms by which they are produces. We expound how the experiment's excellent energy resolution, combined with the possibility to move the detector off-axis (the DUNE-PRISM concept), work in tandem to improve the discrimination power.

Figures (14)

• Figure 1: Approximate reach of DUNE to discover model-specific long-lived particles, dark Higgs (left) and dark photons (right), assuming ten years of data collection. Sensitivity from different production mechanisms corresponds to different color contours -- the left panel shows charged kaon decay (blue) and proton bremsstrahlung of a scalar boson (red), where the right panel shows neutral $\eta$ meson decay (purple) and proton bremsstrahlung of a vector boson (green). Contours of lifetimes of the particles are shown by various gray lines, and gray shaded regions indicate parameter space excluded by existing searches Batell:2022dpxGori:2022vri.
• Figure 2: Kinematics of LLPs produced in the DUNE/LBNF beamline in the energy-vs-transverse momentum fraction plane).
• Figure 3: Top Panel : Area-normalized LLP fluxes at the DUNE near detector as a function of LLP energy. We show results for the dark Higgs ($S$), dark photon ($A'$), and leptophilic gauge boson ($V$), and for several different production channels (left panel: kaon decays, right panel: bremsstrahlung). In all cases, we have assumed an LLP mass of $m_X = 200MeV$. Bottom panel : Area-normalized fluxes with respect to transverse radius for five long-lived particle production methods, all of which having $m_X = 200$ MeV.
• Figure 4: Normalized event rates with respect to particle energy for five long-lived particle production methods, all of which having $m_X = 200$ MeV, where we fix the lifetime to be $c\tau = 10^5$ m. The left panel shows meson-decay production: two-body decay of a charged kaon (blue), two-body decay of an $\eta$ meson (purple), and three-body radiative decay of a charged kaon (black). The right panel contrasts two proton-bremsstrahlung production modes: that of a vector (green) vs. that of a scalar (red). For comparison, a smaller-lifetime ($10m$) version of the expected event-rate distrubtion is shown in \ref{['fig:Event_rate_2_body_shortlifetime']}.
• Figure 5: Minimum likelihood comparison between the channels $K^\pm \to \pi^\pm S$ and $\eta \to \gamma A'$. The horizontal axis shows the assumed production fraction from $\eta$ decays, while the vertical axis gives the minimum $\chi^2$ value obtained when testing a hypothesis against the true one. Curves in purple correspond to the case where the true hypothesis is $\eta$ decay, and curves in blue correspond to kaon decay. Solid lines indicate a two-dimensional comparison that includes both energy and radial distributions, whereas dashed lines represent a one-dimensional comparison based only on the energy spectrum. The three dashed horizontal lines corresponds to values of $1\sigma, \,2\sigma$ and $3\sigma$ for reference. The 3 panels corresponds to different choices of LLP lifetimes, viz., 1 m, 10 m and 100 m.