1: \section{Analysis Method}
2: \label{sec:Analysis}
3:
4: This measurement of the hadronic mass distribution in semileptonic \B\ decays exploits the very large sample of \B\ mesons recorded by the \babar\ detector. The analysis relies on \BB\ events in which one \B\ meson decays hadronically and is fully reconstructed ($B_{reco}$) and the semileptonic decay of the other $B$ meson ($B_{recoil}$) is identified by the presence of an electron or muon. While this approach
5: results in a low overall event selection efficiency, it allows for the determination of the momentum, charge,
6: and flavor of the $B_{reco}$ meson.
7: Furthermore, the remaining tracks and photons in the event can be used to reconstruct the hadronic mass $M_X$ of the semileptonic decay. The determination of $M_X$ is substantially improved by application of a kinematic fit to the full event.
8:
9:
10:
11: \subsection{Selection of Fully Reconstructed Hadronic \mbox{\boldmath$ B$} Decays, \mbox{\boldmath$ B \ra D Y$}}
12:
13:
14: To reconstruct a large sample of $B$ mesons, hadronic decays of the type $B_{reco}
15: \rightarrow \Db^{(*)} Y^{\pm}$ are selected. Here, $D^{(*)}$ refers to a charm meson and $Y^{\pm}$ consists of hadrons with a total charge of $\pm 1$, composed
16: of $n_1\pi^{\pm}\, n_2K^{\pm}\, n_3\KS\, n_4\piz$ with $n_1 + n_2 \leq
17: 5$, $n_3 \leq 2$, and $n_4 \leq 2$. We reconstruct $D^{*-}\ra
18: \Dzb\pi^-$ and $\Dstarzb \ra
19: \Dzb\piz, \Dzb\gamma$ and the decays $D^-\ra K^+\pi^-\pi^-$, $K^+\pi^-\pi^-\piz$, $\KS\pi^-$,
20: $\KS\pi^-\piz$, $\KS\pi^-\pi^-\pi^+$; and $\Dzb\ra K^+\pi^-$,
21: $K^+\pi^-\piz$, $K^+\pi^-\pi^-\pi^+$, $\KS\pi^+\pi^-$.
22:
23: The kinematic consistency of the $B_{reco}$ candidates
24: is checked with two variables,
25: the beam energy-substituted mass $\mes = \sqrt{s/4 -
26: \vec{p}^{\,2}_B}$ and the energy difference
27: $\Delta E = E_B - \sqrt{s}/2$. Here $\sqrt{s}$ is the total
28: energy in the \FourS\ center-of-mass frame, and $\vec{p}_B$ and $E_B$
29: denote the momentum and energy of the $B_{reco}$ candidate in the same
30: frame.
31: We require $\Delta E$ to be within three standard deviations from zero for each decay mode.
32:
33: For a given \breco\ decay mode, the purity
34: is estimated as the fraction of signal events with \mes$>
35: 5.27$\gev. For this analysis we only use decay modes for which the purity
36: exceeds 40\%.
37: In events with more than one
38: reconstructed \breco\ decay, we select the decay mode with the highest purity. For the case of two candidates being reconstructed in the same mode, the one with $\Delta E$ closest to zero is selected.
39:
40:
41: \subsection{Selection of Semileptonic Decays, \mbox{\boldmath$B \ra X \ell \nu$}}
42:
43: Semileptonic $B$ decays are identified by the presence of one and only one electron or muon above a minimum momentum $p^*_{min}$ measured in the rest frame of the $B_{recoil}$ meson recoiling against the $B_{reco}$. The lepton momentum threshold $p^*_{min}$ is varied in the range of $0.9 -1.6 \gev$.
44: The higher the lepton momentum, the lower are the residual backgrounds from secondary decays of charm particles, $\tau^{\pm}$ leptons, and from misidentified hadrons. Measurements with lower lepton momenta are more sensitive to the production of higher mass charm mesons, in the form of both resonant and non-resonant states.
45:
46: To improve the background rejection, we impose a condition on $Q_{\ell}$, the charge of the lepton from $B_{recoil}$, and $Q_b$, the charge of the $b$-quark of $B_{reco}$, namely $Q_b Q_{\ell} < 0$. This condition is fulfilled for primary leptons, except for \BzBzb\ events in which flavor mixing has occured.
47:
48: Electrons are identified using a likelihood-based algorithm, combining the track momentum with the energy, position, and shape of the shower measured in the EMC, the Cherenkov angle and the number of photons measured in the DIRC, and the specific energy loss in the DCH.
49: The efficiency of the electron selection has been measured with radiative Bhabha events and has been corrected for the higher multiplicity of \BB events using Monte Carlo (MC) simulations.
50:
51: Muons are identified using a cut-based selection for minimum ionizing tracks, relying on information from the finely segmented instrumented flux return of the magnet. The number of interaction lengths traversed by the track, the spatial width of the observed signals, and the match between the IFR hits and the extrapolated charged track are used in the selection.
52: %
53: The muon identification efficiency has been measured with $\mu^+\mu^-
54: (\gamma)$ events and two-photon production of $\mu^+\mu^-$ pairs.
55:
56: The misidentification probabilities for pions, kaons, and protons have been extracted from selected data control samples. They vary between
57: 0.05\% and 0.1\% (1.0\% and 3\%) for the electron (muon) selection.
58:
59: To further reduce backgrounds we require the total charge of the event to be $|Q_{tot}|= |Q_{Breco} + Q_{Brecoil}| \leq 1$.
60: We explicitly allow for a charge imbalance to reduce the dependence on the exact modeling of charged particle tracking in the Monte Carlo simulation, especially at low momenta, and the production of tracks from photon conversions.
61:
62:
63: \subsection{Reconstruction of the Hadronic Mass \mbox{\boldmath$M_X$}}
64: \label{subsec:kfit}
65:
66: The hadronic mass $M_X$ in the decay $B \ra X \ell \nu$ is reconstructed from charged tracks ($\pi^{\pm}$ and $K^{\pm}$) and photons that are not associated with the $B_{reco}$ candidate or identified as leptons. Specifically, we select charged tracks
67: with a minimum transverse momentum $p_t > 0.1 \gev$ and at least 12 DCH hits. For photons we require an energy sum greater than $0.05 \gev$ deposited in three or more crystals.
68: To suppress background from hadrons interacting in the calorimeter we require a minimum angular separation between the center of the shower and the nearest track impact point of 200\mrad and a shower shape consistent with a photon (lateral moment $LAT <0.5$, defined in reference~\cite{Aubert:2001tu} ).
69:
70: The measured four-momentum $P_X$ of the hadron $X$ can be written as
71: \begin{equation}
72: P_X = \sum_{i=1}^{N_{ch}} P^{ch}_i
73: + \sum_{j=1}^{N_{\gamma}} P^{\gamma}_j
74: \end{equation}
75:
76: \noindent
77: where $P$ are four-momenta and the superscripts $ch$ and $\gamma$
78: refer to the selected charged tracks and photons, respectively. Depending on particle identification the charged tracks are assigned either the $K^{\pm}$ or $\pi^{\pm}$ mass.
79:
80: The neutrino four-momentum $P_{\nu}$ is estimated from the
81: missing momentum four-vector,
82: $P_{miss} = P_{\Upsilon(4S)}-P_{B_{reco}}-P_X - P_\ell$,
83: where all momenta are measured in the laboratory frame.
84: The measured missing momentum four-vector
85: is an important indicator of the quality of the reconstruction of the total recoil system.
86: Any particle that is undetected or poorly measured, any sizable energy deposition due to charged hadron interactions or beam-generated background will impact the measurement of $M_X$ and $M_{miss}^2 = P_{miss}^2$. To reduce the impact of these experimental effects, we impose the following criteria: $E_{miss} > 0.5 \gev$, $|\vec {p}_{miss}| > 0.5 \gev$, and $|E_{miss} - |\vec {p}_{miss}|| < 0.5 \gev$.
87: In Figure~\ref{fig:datamccomp} the $E_{miss}$, $|\vec {p}_{miss}|$, and $E_{miss} - |\vec {p}_{miss}|$ distributions are shown and compared to MC simulations, normalized to the total number of events.
88: The agreement reflects our
89: understanding of the detector performance and improvements in the suppression of track and photon background which have been studied extensively. It should be noted that the branching fractions for the individual decay modes used in the MC can also contribute to differences between data and MC. However, the extraction method for the moment measurements is not sensitive to branching fraction variations.
90:
91: To further improve the resolution on the measurement of the hadronic mass $M_X$, we exploit the available kinematic information from the full event, namely the $B_{reco}$ and the $B_{recoil}$ candidate, by performing a kinematic fit with two constraints that imposes four-momentum conservation, the equality of the masses of the two $B$ mesons, $M_{recoil}=M_{reco}$, and forces $M_{miss}^2 = 0$. The fit takes into account event-by-event measurement errors for all individual particles and the four-vector of the measured missing momentum. The resulting $M_X$ resolution is 350 \mev.
92:
93:
94: \subsection{Subtraction of \mbox{\boldmath$B_{reco}$} Background}
95: \label{subsec:mesfit}
96:
97: Figure ~\ref{fig:mes} shows the \mes distribution for the $B_{reco}$ candidates for all events for which the $B_{recoil}$ meson meets all selection criteria for a semileptonic decay.
98: To estimate the combinatorial background in the signal region
99: ($\mes > 5.27 \gev$), the measured \mes distribution is fitted to a sum of a background function (introduced by the ARGUS Collaboration~\cite{Albrecht:1987nr}) and a signal function (first used by the Crystal Ball Collaboration~\cite{Skwarnicki:1986xj}).
100: For $\mes > 5.27 \gev$, we find a total of $7114$ signal events above a background of $2102$ events.
101: In the analysis, the \breco\ background subtraction is carried out for several regions in $M_X$ to account for changes in the signal-to-background ratio as a function of $M_X$.
102: \unitlength1.0cm % coordinates in cm
103:
104: \begin{figure}[t]
105: \begin{center}
106: \begin{picture}(15.,6.)
107: \put(-1.0,0.0){\mbox{\epsfig{file=data_mc_emiss_new.eps,scale=0.28}}}
108: \put(4.9,0.0){\mbox{\epsfig{file=data_mc_pmiss_new.eps,scale=0.28}}}
109: \put(10.8,0.0){\mbox{\epsfig{file=data_mc_e-pmiss_new.eps,scale=0.28}}}
110: \put(2.1,-0.2){$E_{miss} (\gev)$}
111: \put(7.8,-0.2){$|\vec {p}_{miss}| (\gev)$}
112: \put(12.3,-0.2){$E_{miss}-|\vec {p}_{miss}| (\gev)$}
113: \put(-1.3,3.0){\mbox{\begin{turn}{90}{$N/ 200 \mev$}\end{turn}}}
114: \put(4.5,3.0){\mbox{\begin{turn}{90}{$N/ 160 \mev$}\end{turn}}}
115: \put(10.5,3.0){\mbox{\begin{turn}{90}{$N/ 100 \mev$}\end{turn}}}
116: \end{picture}
117: \end{center}
118: %\vspace{-1.0cm}
119: \caption{\em Data - MC comparison of missing energy $E_{miss}$ (left), missing momentum $|\vec {p}_{miss}|$ (middle) and $E_{miss}-|\vec {p}_{miss}|$ (right) for $\psm =0.9~\gev$. The MC predictions for the total distribution as well as the residual background are indicated.
120: \label{fig:datamccomp} }
121: \end{figure}
122:
123:
124: \begin{figure}[t]
125: \begin{center}
126: \begin{picture}(15.,7.5)
127: \put(2.0,0.0){\mbox{\epsfig{file=mes_0.9_mx3_B2102_S7114.eps,scale=0.7}}}
128: \end{picture}
129: \end{center}
130: \vspace{-1.0cm}
131: \caption{\em The $m_{ES}$ distribution for $B_{reco}$ candidates in events for which the $B_{recoil}$ meets all selection criteria for a semileptonic decay. The lepton momentum threshold is $p^*_{min}=0.9 \gev$.
132: \label{fig:mes} }
133: \end{figure}
134: