TL;DR
The BABAR experiment has observed inclusive $D^{*\pm}$ production in the decay of the $\Upsilon(1S)$ meson, measuring a branching fraction of $(2.52 \pm 0.13 \pm 0.15)\%$. This result shows a significant excess over the expected rate from the standard virtual photon annihilation process, hinting at new physics or other non-standard decay mechanisms.
What Problem Does This Paper Address?
Understanding the decay modes of heavy quarkonium states like the $\Upsilon(1S)$ is crucial for testing and refining our understanding of Quantum Chromodynamics (QCD). While decays into light hadrons are well-studied, the decays of bottomonium states into open charm (mesons containing a charm quark) are less understood. Previous experiments had set limits on $D^{*\pm}$ production in $\Upsilon(1S)$ decays, but a precise measurement was lacking. This paper aims to precisely measure the inclusive $D^{*\pm}$ production rate and its momentum distribution in $\Upsilon(1S)$ decays, providing a critical test for theoretical models of these processes, including standard QED and potential non-standard QCD contributions.
Methods
The study utilizes a dataset of approximately 98.6 million $\Upsilon(2S)$ mesons collected by the BABAR detector. The $\Upsilon(1S)$ mesons are produced via the decay chain $\Upsilon(2S) \rightarrow \pi^+\pi^-\Upsilon(1S)$. $D^{*\pm}$ candidates are reconstructed using the decay chain $D^{*\pm} \rightarrow D^0\pi^{\pm}$, followed by $D^0 \rightarrow K^-\pi^+$.
Key steps in the analysis include:
- $\Upsilon(1S)$ reconstruction: Events are selected by identifying pairs of oppositely charged tracks (interpreted as pions) whose recoil mass is consistent with the $\Upsilon(1S)$ mass.
- $D^0$ reconstruction: A kaon and a pion candidate are combined to form a $D^0$ candidate, with geometric constraints and mass windows applied.
- $D^{*\pm}$ reconstruction: The $D^0$ candidate is combined with a "soft" pion of opposite charge, and the mass difference ($\Delta m$) is used for selection.
- Signal Extraction: The $D^{*\pm}$ yield is extracted by fitting the invariant mass distribution of $D^0$ candidates in different momentum bins. This involves careful subtraction of combinatorial backgrounds using recoil mass sidebands and "wrong-sign" $D^0$ combinations.
- Efficiency Determination: The reconstruction efficiency for $D^{*\pm}$ is determined as a function of its scaled momentum ($x_p$) using a simulation study.
Datasets
The primary dataset used is from the BABAR experiment, consisting of approximately $98.6 \times 10^6$ $\Upsilon(2S)$ mesons. This dataset is used to produce $\Upsilon(1S)$ mesons via the $\Upsilon(2S) \rightarrow \pi^+\pi^-\Upsilon(1S)$ decay. An additional off-resonance data sample (44.5 fb$^{-1}$) was used for background studies. A GEANT4-based simulation of the BABAR detector was also extensively used.
Key Insights and Ideas
- Excess $D^{*\pm}$ production: The measured branching fraction for $\Upsilon(1S) \rightarrow D^{*\pm}X$ is significantly higher than what is predicted by the standard virtual photon annihilation process ($\Upsilon(1S) \rightarrow \gamma^* \rightarrow c\bar{c}$). This suggests that other mechanisms are contributing to $D^{*\pm}$ production.
- Momentum distribution hints: The observed excess is particularly pronounced at lower scaled momenta ($x_p < 0.75$), deviating from the shape expected from the virtual photon process. This could point towards specific alternative production mechanisms.
- Testing QCD: This measurement provides a crucial test of QCD predictions for heavy quarkonium decays, particularly in the realm of open charm production. The discrepancy highlights areas where current theoretical models may be incomplete.
Key Results
- Branching Fraction: The inclusive branching fraction for $\Upsilon(1S) \rightarrow D^{*\pm}X$ is measured to be $\mathcal{B}[\Upsilon(1S) \rightarrow D^{*\pm}X] = (2.52 \pm 0.13 \text{ (stat)} \pm 0.15 \text{ (syst)})\%$.
- Comparison to Theory: The measured branching fraction exceeds the expected rate from the QED virtual photon process by $(1.00 \pm 0.28)\%$, corresponding to a 3.6 standard deviation difference.
- Momentum Distribution: The efficiency-corrected $D^{*\pm}$ yield as a function of scaled momentum ($x_p$) shows an excess over the virtual photon process prediction, especially for $x_p < 0.75$. The probability of the measured spectrum being consistent with the virtual photon process is very low ($1.2 \times 10^{-5}$).
- Systematic Uncertainties: The dominant sources of systematic uncertainty are the slow pion reconstruction efficiency and the $\Upsilon(1S)$ decay selection efficiency. The total systematic uncertainty is 5.9%.
Limitations and Future Work
- Discrepancy with Standard Model: The primary limitation is the significant discrepancy between the measured $D^{*\pm}$ production rate and the prediction from the dominant virtual photon annihilation process.
- Interpretation of Excess: While the paper notes that the excess is compatible with contributions from gluon splitting (predicted to be $(1.20 \pm 0.29)\%$) and disfavors the octet contribution, a definitive identification of the underlying mechanism requires further theoretical work and potentially more precise measurements.
- Future Directions: The authors suggest that this measurement provides a strong motivation for further theoretical studies of non-standard decay mechanisms in heavy quarkonium. Future experiments could aim to further refine these measurements or explore other open charm final states.
Summary updated