Search for the charmonium semi-leptonic weak decay $J/ψ ightarrow D_s^-e^+ν_e+c.c.$

Using a data sample of $(10087 \pm 44) \times 10^6$ $J/ψ$ events collected with the BESIII detector at a centre-of-mass energy of $\sqrt{s}=3.097\ \textrm{GeV}$, a dedicated search for the charmonium semileptonic weak decay $J/ψ\rightarrow D_s^-e^+ν_e + \text{c.c.}$ is performed. No significant signal is observed. An upper limit on the branching fraction is set at $\mathcal{B}(J/ψ\rightarrow D_s^- e^+ ν_e + \text{c.c.}) < 1.0 \times 10^{-7}$ at the 90% confidence level. This result improves upon previous constraints by an order of magnitude, representing the most stringent experimental limit to date. It thus provides a critical test of Standard Model predictions and new physics scenarios in heavy-quark dynamics.

💡 Research Summary

The BESIII Collaboration performed the first high‑sensitivity search for the rare weak decay J/ψ → Dₛ⁻ e⁺ νₑ (and its charge‑conjugate) using the full data set of (10087 ± 44) × 10⁶ J/ψ events collected at a centre‑of‑mass energy of √s = 3.097 GeV. This decay proceeds via the Cabibbo‑favoured c → s e⁺ νₑ transition and is predicted in the Standard Model (SM) to have a branching fraction (BF) of order 10⁻⁸–10⁻⁹, depending on the theoretical framework (lattice QCD, QCD sum rules, covariant light‑front quark model, etc.). Certain beyond‑SM scenarios (top‑color, R‑parity‑violating SUSY, two‑Higgs‑doublet models) could enhance the BF up to 10⁻⁵, making an experimental limit a powerful probe of new physics.

The analysis reconstructs the Dₛ⁻ meson in four hadronic channels: Dₛ⁻ → K_S⁰ K⁻, Dₛ⁻ → K⁺K⁻π⁻, Dₛ⁻ → K⁺K⁻π⁻π⁰, and Dₛ⁻ → K_S⁰ K⁻π⁺π⁻. Charged tracks are required to be within |cosθ| < 0.93 and to originate close to the interaction point. Electron identification combines MDC dE/dx, TOF timing, and EMC energy‑to‑momentum ratio (0.92 < E/p < 1.03) with a likelihood requirement L(e) > 0.001 and L(e)/(L(e)+L(π)+L(K)) > 0.8. Kaon and pion candidates are distinguished by likelihood comparisons. K_S⁰ candidates are built from π⁺π⁻ pairs with a vertex fit and a mass window of ±12 MeV/c² around the nominal K_S⁰ mass; π⁰ candidates are reconstructed from photon pairs with an invariant mass between 115 and 150 MeV/c² and a χ² < 200 mass‑constrained fit.

A one‑constraint (1C) kinematic fit forces the reconstructed Dₛ⁻ mass to its nominal value, and the combination with the smallest χ² is retained. Additional selections suppress backgrounds: the missing momentum magnitude |p_miss| (computed as the vector sum of the beam momentum minus the Dₛ⁻ and electron momenta) must exceed 0.05 GeV/c; the sum |p_e| + |p_miss| is required to lie within mode‑dependent intervals; the opening angle between the electron and any charged pion is forced to be larger than 20°, which removes most events where the electron originates from π⁰ → e⁺e⁻γ conversions. Mode‑specific invariant‑mass vetoes further reduce backgrounds involving extra neutral pions.

The key discriminating variable is U_miss = E_miss − |p_miss|, where E_miss = 2E_beam − E_Ds − E_e. For signal events, the undetected neutrino leads to U_miss ≈ 0, while background processes produce broader distributions. Signal and background probability density functions (PDFs) are derived from dedicated signal Monte Carlo (MC) and inclusive J/ψ MC samples, respectively. The signal PDFs are smoothed using kernel density estimation and then convolved with a Gaussian whose mean (μ) and width (σ) are fixed from control samples (ψ(2S) decays).

A simultaneous unbinned maximum‑likelihood fit to the U_miss spectra of all four Dₛ⁻ modes is performed. The fit includes constraints linking the yields of each mode to a common total signal yield N_sig via the known product of branching fractions (including the Dₛ⁻ decay BF) and detection efficiencies (ε_i). The efficiencies, obtained from signal MC, range from ≈5 % to ≈7 % depending on the mode. The fit returns N_sig = 5.7 ± 4.5 (statistical), consistent with the background‑only hypothesis.

Systematic uncertainties are evaluated for tracking, particle identification, K_S⁰ and π⁰ reconstruction, signal MC modeling, background shape, and the external branching fractions used in the efficiency calculation. Their combined effect is incorporated into the final limit using a Bayesian approach with a flat prior for the BF. The resulting 90 % confidence‑level upper limit is

  𝔅(J/ψ → Dₛ⁻ e⁺ νₑ + c.c.) < 1.0 × 10⁻⁷.

This limit improves the previous BESII bound (𝔅 < 1.3 × 10⁻⁶) by an order of magnitude, making it the most stringent constraint on this decay to date. While the limit is still above the SM expectation, it excludes a wide class of new‑physics models that would enhance the branching fraction to the 10⁻⁵ level. The analysis demonstrates the power of large‑statistics J/ψ samples and multi‑mode reconstruction techniques in probing ultra‑rare weak decays. Future data sets, possibly from upgraded BESIII or next‑generation τ‑charm factories, could push the sensitivity down to the SM predicted range, providing a direct test of the c → s e⁺ νₑ transition in the charmonium system.