1: \section{ANALYSIS METHOD}
2: \label{sec:Analysis}
3:
4: The $\Bz$ meson is reconstructed in the following four decay modes:
5: $\BtoDstppbarpi$, $\BtoDcppbarpi$, $\BtoDszppbar$, and $\BtoDzppbar$.
6: The $\Dstarm$ and $\Dstarzb$ are reconstructed as $\Dzpi$ and
7: $\Dzpiz$, respectively. The $\Dm$ candidates are
8: reconstructed in the decay mode $\Kpipi$. The $\Dzb$ candidates
9: are reconstructed in the $\Kpi$ mode.
10:
11:
12: \subsection{$\Bz$ SELECTION}
13: \label{sec:BReco}
14:
15: The $\Bz$ reconstruction proceeds as follows. First we reconstruct the
16: $\Dm$ and $\Dzb$ candidates in the decay modes noted above.
17: For all decay modes, charged kaons are distinguished
18: from the pions and protons with energy-loss ($dE/dx$) information in the tracking system and
19: the Cherenkov angle and the number of photons measured by the DIRC.
20: For the $\BtoDszppbar$ decay mode only, the
21: pion from $\Dzb \to \Kpi$ decay must not be identified as either
22: an electron, muon, kaon or proton. The mass of the reconstructed $\Dzb$ or $\Dm$
23: candidates must be within 3 standard deviations of the fitted mean of
24: reconstructed $\Dzb$ or $\Dm$ mass for each decay mode. Finally, the
25: daughter tracks from the $\Dzb$ or $\Dm$ are required to be consistent with
26: originating from a common vertex.
27:
28: For the decay modes $\BtoDstppbarpi$ and $\BtoDszppbar$, $\Dstarm$ and
29: $\Dstarzb$ candidates are reconstructed by combining the $\Dzb$
30: candidate with a soft track or a soft $\piz$, respectively.
31: The momentum of the soft track or soft $\piz$
32: in the CM frame must be less than 0.45 \gev/c.
33: The $\piz$ candidate is required to have a two-photon invariant mass between
34: 0.116--0.150 \gev/$c^2$, and its daughter photon candidates must
35: have a minimum cluster energy of 30 \mev.
36: The mass difference between $D^{*}$ and $D^{0}$ ($\Delta M
37: =m_{D^{*}}-m_{D^{0}}$) is required to
38: be less than $0.1486~\gev/ c^2$ and $0.1453~\gev/c^2$ for $\Dstarm$ and
39: $\Dstarzb$ candidates, respectively. The $\Delta M$ requirement removes
40: crossfeed from $\BtoDszppbar$ into $\BtoDzppbar$.
41: For each track except for the soft pion from $\Dstarm$ decays, the
42: transverse momentum $p_T$ must be larger than 0.1~\gev/c in order to improve
43: the quality of the vertex fit.
44:
45: To reconstruct the candidate $\BtoDstppbarpi$ and $\BtoDcppbarpi$ decay modes,
46: both proton and anti-proton candidates are distinguished
47: from pions and kaons on the basis of energy-loss ($dE/dx$) information in the
48: tracking system and the Cherenkov angle and the number of photons measured by
49: DIRC. The reconstructed $\Dstarm$ and $\Dm$, respectively,
50: are combined with an identified proton and anti-proton pair and a track.
51: We require that the pion candidate track
52: must not be identified as either an electron, a muon,
53: a kaon or a proton. For the decay modes $\BtoDszppbar$ and
54: $\BtoDzppbar$, the $\Dstarzb$ and $\Dzb$ candidates, respectively,
55: are combined with an identified
56: proton and anti-proton pair. All the daughters of the $\Bz$ must originate from a common vertex.
57:
58: Two additional kinematic variables are used to identify the reconstructed $\Bz$
59: candidates~\cite{ref:babar}. The first is the beam-energy-substituted mass,
60: $m_{ES}=[(E^2_{CM}/2+\mbox{\bf p}_i\cdot \mbox{\bf p}_B)^2/E^2_i-{\mbox{\bf p}^2_B}]^{1/2}$,
61: where $E_{CM}$ is the total CM energy of the $e^+e^-$ collision,
62: ($E_i, \mbox{\bf p}_i$) is
63: the four-momentum of the initial $e^+e^-$ system
64: and ${\bf p}_B$ is the momentum of the
65: reconstructed $B$ candidate, both measured in the laboratory frame.
66: The second variable is $\Delta E=E^*_B-E_{CM}/2$, where
67: $E^*_B$ is the $B$-candidate energy in the CM frame.
68: The \B meson candidates are defined by requiring:
69: $5.20 \gev/c^2 <\mes<5.29\,\gev/c^2$ and $\left|\DeltaE\right|<0.12\,\gev$.
70: These two variables are then used in a maximum likelihood fit to
71: extract signal and background yields.
72:
73: If more than one suitable $\Bz$ candidate is
74: reconstructed in an event, then for each $\Bz$ decay mode one
75: best candidate is selected.
76: The best candidate selection algorithm is based
77: on a $\chi^2$-like quantity constructed from the difference between the $D$
78: mass and/or mass difference $\Delta M$ for the candidate and the nominal
79: value~\cite{ref:pdg2002}. For a given event,
80: the candidate with the lowest value of $\chi^2$ is selected for each $\B^0$
81: decay mode. For the $\BtoDstppbarpi$ and
82: $\BtoDszppbar$ decays, the best candidate is selected based on
83: the mass of the $\Dzb$ candidate and $\Delta M$. For the $\BtoDcppbarpi$
84: and $\BtoDzppbar$ decay modes, the reconstructed $D$ meson
85: mass is used for best candidate selection.
86:
87:
88: To suppress background from two-jet-like $\epem \to$
89: \qq~continuum, variables that characterize the event topology are
90: used. We require $\cos{\theta_{thr}}<0.9$, where $\theta_{thr}$ is the angle
91: between the thrust axis of the $\Bz$ candidate and that of the
92: rest of the event. This requirement eliminates 63\% of the
93: continuum background and retains 88\% of the signal events.
94: For further continuum background suppression we
95: require that the ratio of the second to zeroth
96: Fox-Wolfram moments~\cite{r2definision} is less than 0.35.
97:
98: The signal efficiency in each $B$ decay mode
99: after applying all selection criteria is shown in
100: Table~\ref{tab:effi-yields}. The efficiencies listed in the
101: table are obtained from signal MC simulation assuming pure phase-space for
102: the \B decay model.
103:
104:
105: \begin{table}[h]
106: \begin{center}
107: \caption{ Summary of selection efficiencies obtained from MC simulation
108: based on a phase-space
109: model and fitted yields from data for the \B decay modes in this report.
110: The error on the efficiency
111: comes from the Monte Carlo statistics. The statistical significance is
112: calculated as $\sqrt{2\Delta \log{\cal L}}$, where $\Delta \log{\cal L}$ is
113: the log-likelihood difference between a signal hypothesis corresponding to
114: the yield and that corresponding to a null yield.}
115: \label{tab:effi-yields}
116: \begin{tabular}{lccc}
117: \hline\hline
118: Mode & Efficiency, \% & Signal yield & Statistical significance \\ \hline
119: \BtoDstppbarpi & 6.97 $\pm$ 0.10 & 130$\pm$14 & 18$\sigma$\\
120: \BtoDcppbarpi & 5.87 $\pm$ 0.10 & 238$\pm$22 & 17$\sigma$\\
121: \BtoDszppbar & 5.53 $\pm$ 0.09 & 13$\pm$ 4 & 5$\sigma$\\
122: \BtoDzppbar & 16.53$\pm$ 0.15 & 96 $\pm$ 11 & 17$\sigma$ \\ \hline \hline
123: \end{tabular}
124: \end{center}
125: \end{table}
126:
127: \input{MLfit}
128:
129: