Search for dark sector at BESIII

TL;DR

This BESIII study probes light dark-sector states that couple to charm and light quarks by examining multiple final states across charmonium- and hyperon-rich data. The analyses cover massless dark photons in D0 decays, invisible K0_S decays, dark baryons in Ξ decays, massless particles in Σ^+ decays, and axion-like particles in J/ψ decays, employing methods such as double-tagging and recoil techniques. No significant signals are observed in any channel, but BESIII sets competitive upper limits on branching fractions and couplings (e.g., |C|^2+|C5|^2 < 8.2×10^-17 GeV^-2; B limits down to 1.1×10^-5, 2.0×10^-6, and 3.2×10^-5 in various channels, plus ALP g_{aγγ} limits up to ~10^-4 GeV^-1). These results tighten the low-mass dark-sector parameter space and demonstrate BESIII’s continued sensitivity to light NP scenarios, guiding future searches in the GeV-scale frontier.

Abstract

BESIII experiment has collected a large data sample of charmonium, charm mesons, hyperons, and other light mesons. These data provide a unique opportunity to explore the dark sector beyond the Standard Model, particularly for dark sectors that couple to charm quarks or other light quarks, and for dark sectors with masses in the $τ-c$ energy region. We present recent dark sector search results from the BESIII experiment, including the search for massless dark photons in $D^0\toωγ'$ and $D^0\toγγ'$, the search for invisible decays of $K^0_S$, the search for dark baryon particles in $Ξ^-\toπ^-χ$, the search for massless particles in $Σ^+\to p + \rm{invisible}$, and the search for axion-like particles in $J/ψ\toγa$ with $a\toγγ$.

Figures (5)

• Figure 1: (a) and (b) display the $M^2_{\rm{miss}}$ distributions for the $D^0 \to \omega \gamma'$ and $D^0 \to \gamma \gamma'$ decay candidates, respectively. The black points represent the observed data, while the other lines correspond to various background and signal components. (c) illustrates the constraints on the $cu \gamma'$ coupling. The black solid lines indicate the observed ULs from different processes, and the red shaded regions denote the allowed parameter space from DM and VS.
• Figure 2: (a) Illustration of the event displays for the $K^0_S \to \text{invisible}$ signal process (left) and a corresponding background process (right) obtained from Monte Carlo simulations. The EMC sub-detector is depicted by the light blue barrel, while the purple blocks indicate energy deposits within the EMC. (b) Distribution of the deposited energy for the selected $K^0_S \to \text{invisible}$ candidate events. The black data points represent the observed events, and the other lines or histograms correspond to various background and signal components.
• Figure 3: (a) Distribution of the deposited energy for the selected $\Xi^- \to \pi^- \chi$ candidate events with $m_\chi = 1.10\,\rm{GeV}$. The black data points represent the observed events, while the other lines or histograms correspond to various background and signal components. (b) ULs on the BFs of $\Xi^- \to \pi^- \chi$ at the 95% C.L.. The black points indicate the observed ULs obtained from the real data samples, the black dashed line and filled color regions represent the expected ULs, and the red and blue lines show the ULs recast from Ref. Alonso-Alvarez:2021oaj.
• Figure 4: (a) Distribution of the deposited energy for the selected $\Sigma^+ \to p +\rm{invisible}$ candidate events with $m_{\rm{invisible}} = 0$. The black data points represent the observed events, while the other lines or histograms correspond to various background and signal components. (b) The constraints on the effective decay constants for the vector coupling term and axial coupling term of QCD axion from different experiments.
• Figure 5: (a) Distribution of the di-photon invariant mass in the ALP search. The black points represent the observed data, while the histogram shows contributions from various background processes. (b) Constraints on the ALP-photon coupling. The red-filled regions indicate the excluded parameter space based on the latest BESIII results, whereas the other filled regions correspond to previously excluded parameter spaces.