Abstract
The first measurement of $B(D^0 \to \phi X^0)$ and an upper limit for $B(D^+ \to \phi X^+)$ are determined from 22.3 pb$^{-1}$ of $e^+e^-$ annihilation data at a c.m. energy of 4.03 GeV. The data were recorded by the Beijing Spectrometer (BES) at BEPC. A recoil charge method is applied to charm threshold data to determine the charge of the $D$ meson in the recoil from fully reconstructed $D$ mesons. The branching fractions $B(D^0 \to \phi X^0)$ and $B(D^+ \to \phi X^+)$ are determined from 10 events with a reconstructed $D$ and a recoiling $\phi \to K^+K^-$. In addition, a 90% C.L. upper limit of $B(D^+ \to \phi e^+ X^0) < 1.6\%$ is determined from a search for semileptonic decays of the $D^+$.
Introduction
While exclusive decays of charm mesons—those into specific, fully identified final states—are well-documented, knowledge of inclusive decays remains limited. Understanding the inclusive decay $D \to \phi X$ is critical for two primary reasons. First, it serves as an independent check on decay dynamics; a discrepancy between the inclusive rate and the sum of known exclusive rates would indicate missing physics or undiscovered decay modes. Second, accurate $D \to \phi$ branching fractions are essential for reducing systematic uncertainties in $B$ meson studies, such as searches for gluonic penguins ($b \to sg$) or $B_s^0$ oscillations, where $\phi$ mesons from intermediate charm decays constitute a significant background.
Experimental Methodology
The analysis utilized 22.3 pb$^{-1}$ of $e^+e^-$ annihilation data collected at a center-of-mass energy of 4.03 GeV using the Beijing Spectrometer (BES) at the Beijing Electron Positron Collider (BEPC). This energy is just above the threshold for $D\bar{D}$ pair production, providing a clean environment where $D$ mesons are produced nearly at rest.
The researchers employed a novel "recoil charge method" to identify the flavor of the recoiling $D$ meson without the efficiency loss associated with full double-tagging. In this approach, one $D$ meson (the "tag") is fully reconstructed. The charge of the recoiling system is then determined by summing the charges of all other detected particles. Due to charge conservation in the initial $e^+e^-$ state, the recoil charge is highly correlated with the charge of the second $D$ meson produced in the event.
As shown in Figure 1, $D^0$ recoils typically exhibit a total charge near 0, while $D^+$ recoils (associated with a $D^-$ tag) tend toward a charge of +1. This statistical separation allowed the team to categorize the recoil systems before searching for $\phi \to K^+K^-$ signals.
Results and Discussion
From a sample of $9054 \pm 309 \pm 416$ reconstructed tag $D$ mesons, the analysis identified 10 events containing a recoiling $\phi$ meson candidate. The invariant mass distributions for these $K^+K^-$ pairs are shown in Figure 6, separated by the identified recoil flavor.
The primary quantitative results are as follows:
- $B(D^0 \to \phi X^0) = (1.71^{+0.76}_{-0.71} \pm 0.17)\%$
- $B(D^+ \to \phi X^+) < 1.8\%$ (at 90% confidence level)
- $B(D^+ \to \phi e^+ X^0) < 1.6\%$ (at 90% confidence level)
These measurements are consistent with the sum of previously measured exclusive $\phi$ decay modes, such as $D^0 \to \phi \bar{K}^0$ and $D^0 \to \phi \pi^0$. This consistency suggests that there are no major undiscovered $D$ decay channels involving $\phi$ mesons at the current level of sensitivity.
Conclusion
This study provides the first direct measurement of the inclusive $D^0 \to \phi X^0$ branching fraction. While the recoil charge method significantly improved the statistical reach of the BES data, the results remain limited by a small signal sample, leading to statistical uncertainties of approximately 45%. These values establish a necessary baseline for charm physics; however, high-precision constraints will require the significantly larger datasets expected from future high-luminosity experiments at the charm threshold or $B$-factories.
Summary updated