Sub-GeV dark matter and multi-decay signatures from dark showers at beam-dump experiments

Abstract

In models of strongly-interacting dark sectors, the production of dark quarks at accelerators can give rise to dark showers with multiple dark mesons in the final state. If some of these dark mesons are sufficiently light and long-lived, they can be detected with searches for displaced vertices at beam-dump experiments and electron-positron colliders. In this work we focus on the case that dark quark production proceeds via effective operators, while the dark sector analogue of the $ρ^0$ meson can decay via kinetic mixing. We evaluate current constraints from NA62 and BaBar as well as sensitivity projections for SHiP and Belle II. We find that there exists a sizable parameter region where SHiP may detect several displaced vertices in a single event and thus obtain valuable information about the structure of the dark sector.

Figures (9)

• Figure 1: Multiplicity distribution of the various dark mesons predicted for the SPS beam-dump facilities. The panel shows the multiplicity distribution of neutral dark rhos (left), charged dark rhos (middle), and charged and neutral dark pions (right). The distributions are normalised to unity. Error bars represent standard Poisson uncertainties of the Monte Carlo samples.
• Figure 2: Kinematic distributions of the dark rho mesons predicted for the SPS beam dump facilities. The panels show the energy (left) and polar angle (right) distributions of the produced neutral dark rho mesons. The solid lines show the distributions for $r=1.5$ and the dotted lines for $r=1.9$. The distributions are normalised to unity. Error bars indicate the standard Poisson uncertainties of the Monte Carlo samples.
• Figure 3: Normalised two-dimensional distribution of the energy and polar angle of the dark rho mesons predicted for the SPS beam-dump facilities for two different values of ${m_{{\rho_D}}}$ and $r = 1.5$.
• Figure 4: Impact of varying the dark-shower parameters on the event yields with one (blue), two (red), and three (green) decaying dark rho mesons per event at SHiP (15-year running time is assumed; see Sec. \ref{['sec:results']} for details). Two benchmark masses are shown: $m_{\rho_D}=0.3~\mathrm{GeV}$ (left) and $1~\mathrm{GeV}$ (right). Within each colour, the six line styles denote different parameter choices: solid (baseline set-up described in the main text and App. \ref{['app:DetailsOnTheSimulatedSample']}); short-dash$(\Lambda_D=\tfrac{1}{3}\Lambda_D^{\text{default}})$; long-dash$(\Lambda_D=3\,\Lambda_D^{\text{default}})$; dash-dot (probVec$=0.33$); dotted (probVec$=0.75$); long dash-dot$(r\equiv m_{\rho_D}/m_{\pi_D}=1.9)$.
• Figure 5: Illustration of the different production modes for dark rho mesons. Panels (a)--(d) show the various dark-photon-like production processes, namely (from left to right) decays of neutral mesons $\eta^{(\prime)}\to {\rho_D}+\gamma$, proton bremsstrahlung, production via mixing with neutral mesons $V^{0}$, and Drell-Yan production. Panel (e) shows the production via dark showers specific to strongly-interacting dark sectors.