Search for Dimuon Decays of a Light Scalar in Radiative Transitions Y(3S) -> gamma A0

TL;DR

We search for a light scalar $A^{0}$ in radiative $\Upsilon(3S)\rightarrow\gamma A^{0}$ decays with $A^{0}\rightarrow\mu^{+}\mu^{-}$ using $122\times10^{6}$ $\Upsilon(3S)$ decays collected by BABAR. A likelihood-based scan over $m_{A^{0}}$ (0.212–9.3 GeV) models continuum and ISR backgrounds from data and Monte Carlo, extracting signal yields from the reduced mass distribution $m_{R}$ with mass-dependent signal shapes. No significant signal is observed after accounting for a large trial factor, and 90% C.L. upper limits are set on $\mathcal{B}(\Upsilon(3S)\rightarrow\gamma A^{0})\times\mathcal{B}(A^{0}\rightarrow\mu^{+}\mu^{-})$ across the scanned range, $(0.25-5.2)\times10^{-6}$, along with a limit on $\mathcal{B}(\eta_{b}\rightarrow\mu^{+}\mu^{-})<0.8\%$; HyperCP and NMSSM/axion-like model parameter spaces are thus constrained. The results improve upon previous limits (e.g., CLEO) and provide important input for beyond-Standard-Model scenarios predicting light scalar states.

Abstract

The fundamental nature of mass is one of the greatest mysteries in physics. The Higgs mechanism is a theoretically appealing way to account for the different masses of elementary particles and implies the existence of a new, yet unseen particle, the Higgs boson. We search for evidence of a light scalar (e.g. a Higgs boson) in the radiative decays of the narrow Y(3S) resonance: Y(3S)->gamma A0, A0->mu+mu-. Such an object appears in extensions of the Standard Model, where a light CP-odd Higgs boson naturally couples strongly to b-quarks. We find no evidence for such processes in a sample of 122*10^6 Y(3S) decays collected by the BABAR collaboration at the PEP-II B-factory, and set 90% C.L. upper limits on the branching fraction product B(Y(3S)-> gamma A0)*B(A0->mu+mu-) at (0.25-5.2)*10^{-6} in the mass range 0.212<= m(A0)<=9.3 GeV. We also set a limit on the dimuon branching fraction of the eta_b meson B(eta_b->mu+mu-)<0.8% at 90% C.L. The results are preliminary.

Figures (11)

• Figure 1: Distribution of the dimuon invariant mass $m_{\mu^+\mu^-}$ in the $焇{(3S)}$ data. Black histogram shows the distribution for the selection in which only one of two muons is required to be positively identified. The peak from $e^+e^-\rightarrow\xspace\gamma_\mathrm{ISR}\rho^0(770)$, $\rho^0\rightarrow\xspace\pi^+\pi^-$, in which one of the pions is misidentified as a muon, is clearly visible. Blue (lower) histogram shows the distribution for the selection in which both muons are positively identified. The ISR-produced peaks at $J/\psi$ and $焇{(1S)}\xspace$ masses are visible.
• Figure 2: Statistical and systematic uncertainty on the product of branching fractions $\mathcal{B}(焇{(3S)}\xspace\rightarrow\xspace\gamma{A^{0}}\xspace)\times\mathcal{B}({A^{0}}\xspace\rightarrow\xspace\mu^+\mu^-)$ as a function of $m_{A^0}$, extracted from the fits to the $焇{(3S)}$ data. Statistical errors are shown as red dot-dashed line, systematic uncertainties are shown as blue dotted line, and the total uncertainty, computed as a quadrature sum of statistical and systematic errors, is the solid black line. The shaded areas show the regions around the ${J / \psi }$ and $\psi(2S)$ resonances excluded from the search.
• Figure 3: Distribution of the likelihood ratio variable $\mathcal{S}$ (with additive systematic uncertainties included for the fits to the $焇{(3S)}$ dataset. The blue curve is the Gaussian fit with fixed $\mu=0$ and $\sigma=1$.
• Figure 4: The fit for $m_{A^0}\xspace=4.940$$\mathrm{\,Ge V}$ in $焇{(3S)}$ dataset. The bottom graph shows the $m_R\xspace$ distribution (solid points), overlaid by the full PDF (solid blue line). Also shown are the contributions from the signal at $m_{A^0}\xspace=4.940$$\mathrm{\,Ge V}$ (solid red line) and the continuum background (dashed black line). The top plot shows the normalized residuals $p=(\mathrm{data}-\mathrm{fit})/\sigma(\mathrm{data})$ with unit error bars. The signal peak corresponds to the likelihood ratio variable $\mathcal{S}=3.0$, including systematics, and ${\cal B\xspace}=(1.9\pm0.7\pm0.1)\times10^{-6}$.
• Figure 5: The fit for $m_{A^0}\xspace=0.426$$\mathrm{\,Ge V}$ in $焇{(3S)}$ dataset. The bottom graph shows the $m_R\xspace$ distribution (solid points), overlaid by the full PDF (solid blue line). Also shown are the contributions from the signal at $m_{A^0}\xspace=0.426$$\mathrm{\,Ge V}$ (solid red line) and the continuum background (dashed black line). The top plot shows the normalized residuals $p=(\mathrm{data}-\mathrm{fit})/\sigma(\mathrm{data})$ with unit error bars. The signal peak corresponds to the likelihood ratio variable $\mathcal{S}=2.9$, including systematics, and ${\cal B\xspace}=(3.1\pm1.1\pm0.3)\times10^{-6}$.