1: %
2: % This is a example LaTeX file for documents submitted to the SLAC Summer
3: % Institute Proceedings
4: % Revised 10/2001 LJD
5: %
6: % Times style required to increase readability of PDF file (type 1 font)
7: %
8: \documentclass[12pt]{article}
9: \usepackage{ssi}
10: \usepackage{times}
11: \usepackage[dvips]{graphicx}
12: \usepackage{epsf} % for embedding postscript figures
13: \usepackage{epsfig} % for those that use epsfig instead
14: %\usepackage{graphicx}
15: \usepackage{relsize}
16: \def\babar{\mbox{\slshape B\kern-0.1em{\smaller A}\kern-0.1em
17: B\kern-0.1em{\smaller A\kern-0.2em R}}}
18: \newcommand{\psfiletwo}[3]{ % (arguments: full filenames, width/textwidth)
19: \begin{minipage}{\linewidth}
20: \parbox[b]{.49\linewidth}{%
21: \begin{center}
22: \setlength{\epsfxsize}{#3\linewidth}\leavevmode\epsfbox{#1}
23: \end{center}
24: }
25: \hfill
26: \parbox[b]{.49\linewidth}{%
27: \begin{center}
28: \setlength{\epsfxsize}{#3\linewidth}\leavevmode\epsfbox{#2}
29: \end{center}
30: }
31: \end{minipage}
32: }
33: \newcommand{\psfiletwoBB}[5]{ % (arguments: (bounding box, full
34: % filename, i=1,2), width/linewidth)
35: \begin{minipage}{\linewidth}
36: \parbox[b]{.49\linewidth}{%
37: \begin{center}
38: \setlength{\epsfxsize}{#5\linewidth}\leavevmode\epsfbox[#1]{#2}
39: \end{center}
40: }
41: \hfill
42: \parbox[b]{.49\linewidth}{%
43: \begin{center}
44: \setlength{\epsfxsize}{#5\linewidth}\leavevmode\epsfbox[#3]{#4}
45: \end{center}
46: }
47: \end{minipage}
48: }
49: \begin{document}
50:
51: \title{Measurements of Rare B Decays at \babar\ }
52:
53: \author{Paul C. Bloom\thanks{Supported by DOE Contract DE-FG03-95ER40894.
54: %\thanks{Supported by DOE Contract xxx.
55: % Footnotes in ten-point font.
56: \vskip 0.5in
57: \noindent
58: \copyright\ 2002 by Paul C. Bloom.}\\
59: University of Colorado, Boulder \\
60: bloom@slac.stanford.edu\\[0.4cm]
61: %and \\[0.4cm]
62: %Jane Eyre\thanks{Supported by NSF Contract xxx.}\\
63: %Washington University, Seattle, Washington xxxxx \\[0.4cm]
64: Representing the \babar\ Collaboration
65: }
66:
67:
68: \maketitle
69: \begin{abstract}%
70: \baselineskip 16pt
71:
72: We present the results of searches for rare B meson decays. The measurements
73: use all or part of a data sample of about 88 million $\Upsilon(4S)\rightarrow
74: B{\bar B}$ decays collected between 1999 and 2002 with the \babar\
75: detector at the PEP-II
76: asymmetric energy B Factory at the Stanford Linear Accelerator Center.
77: We study a variety of decays dominated
78: by electromagnetic, electroweak and gluonic penguin transitions, and report
79: measurements of branching fractions and other quantities of interest.
80:
81: \end{abstract}
82:
83: \section{Introduction}
84:
85: Measurements\cite{babar_kstargamma,babar_rhogamma,babar_semibsg,babar_fullbsg,babar_kll,babar_ll,babar_knunu,babar_2gamma,babar_hphm,babar_hpi0,babar_hpi02,babar_2pi0,babar_phi,babar_omega,babar_etap,babar_3h}
86: of rare B meson branching
87: fractions have been performed using the
88: \babar\cite{babarnim} detector. B decays in which CKM favored amplitudes are
89: suppressed or forbidden
90: are sensitive to penguin amplitudes and hence to possible
91: non-Standard Model effects arising from new particles participating in
92: internal loops. In addition to probes for new physics, many of these modes
93: are also crucial to the full constraint of the ``Unitarity Triangle''.
94: As the definition implies, rare decays typically have branching fractions of
95: less than $10^{-4}$. The present data sample of roughly 88 million
96: $B{\bar B}$
97: pairs allows for measurements or stringent limits on many such modes.
98:
99: \subsection{Flavor and the Quark Sector of the Standard Model}
100:
101: The complex CKM\cite{ckm} matrix describes the coupling of the charged weak
102: transition $q\rightarrow W^{*+}q^\prime$, which is proportional to
103: $V^{*}_{qq^\prime}$. The
104: non-diagonality of this matrix expresses the fact that the Weak isospin doublet
105: members $(b^\prime, s^\prime, d^\prime)$ are states of mixed flavor. We can
106: thus view the CKM matrix as the transformation between the mass and
107: flavor eigenstates of the quarks
108: \begin{equation}
109: \left ( \begin{array}{c}
110: d^\prime \\ s^\prime \\ b^\prime \end{array} \right)
111: = \left ( \begin{array}{ccc}
112: V_{ud} & V_{us} & V_{ub} \\
113: V_{cd} & V_{cs} & V_{cb} \\
114: V_{td} & V_{ts} & V_{tb} \end{array} \right)
115: \left ( \begin{array}{c}
116: d \\ s \\ b \end{array} \right). \label{eqn1}
117: \end{equation}
118: The unitarity condition implies that there are four
119: free parameters in this matrix,
120: one of which is a phase. It is through this phase that the Standard Model
121: can accommodate CP violation. In particular, the orthogonality requirement
122: between the first and third columns requires
123: \begin{equation}
124: V_{ud}V_{ub}^*+V_{cd}V_{cb}^*+V_{td}V_{tb}^*=0 , \label{eqn2}
125: \end{equation}
126: which can be expressed geometrically as the so called ``Unitarity Triangle''
127: shown in Figure~\ref{fig:unit}. Information about each side and angle is
128: accessible through a variety of measurements in the B meson system. The angles
129: are measured through time-dependent decay rate asymmetries, and the sides via
130: direct or indirect measurements of the CKM matrix elements.
131: Measurement of
132: all the components of the Unitarity Triangle over-constrains the triangle, and
133: thus provides a test of the Standard Model (SM).
134: \begin{figure}[t]
135: \begin{center}
136: \begin{picture}(350,200)(-175,-130)
137: %\multiput(-175,-130)(0,10){21}{\line(1,0){350}}
138: %\multiput(-175,-130)(10,0){36}{\line(0,1){200}}
139: %
140: \thicklines
141: \put(-120,-80){\line(1,0){240}}
142: \put(-120,-80){\line(4,5){96}}
143: \put(+120,-80){\line(-6,5){144}}
144: %
145: \put(-65,-25){$V_{ud}V_{ub}^*$}
146: \put(+15,-25){$V_{td}V_{tb}^*$}
147: \put(-30,-75){$V_{cd}V_{cb}^*$}
148: %
149: \put(-25,+20){$\alpha$}
150: \put(+80,-70){$\beta$}
151: \put(-100,-70){$\gamma$}
152: %
153: \put(-130,-70){\vector(1,0){28}}
154: \put(-130,-70){\line(-1,-2){8}}
155: \put(-175,-130){\shortstack[l]{$B_S \rightarrow \rho K^0_S$ \\
156: $B^+ \rightarrow D^0_{CP} K^+$ \\
157: $B \rightarrow \pi\pi, K\pi$}}
158: \put(130,-70){\vector(-1,0){35}}
159: \put(130,-70){\line(1,-2){8}}
160: %**%\put(95,-120){\shortstack[l]{$B^0 \rightarrow%
161: \put(80,-120){\shortstack[l]{$B^0 \rightarrow%
162: (\psi^{(\prime)},\phi,\eta^\prime ) K^0_S$ \\
163: $B^0 \rightarrow D^{(*)+} D^{(*)-}$}}
164: \put( 5, 28){\line(-1,3){8}}
165: \put( 5, 28){\vector(-3,-1){18}}
166: \put(-40, 55){$B \rightarrow \pi^+\pi^-, \rho^+\pi^- $}
167: \put(-30,-120){$b \rightarrow c\ell\nu$}
168: \put(-140,0){\shortstack[l]{$b \rightarrow u\ell\nu$\\
169: $B\rightarrow (\pi,\rho, \omega)\ell\nu$}}
170: \put(75,0){\shortstack[l]{$B^0_d$ Mixing \\
171: $B \rightarrow (\rho, \omega)\gamma$}}
172: \end{picture}
173: \end{center}
174: \caption{The Unitarity Triangle determined from the
175: orthogonality of the first and third columns of the CKM matrix.
176: Also shown are B meson processes which yield information about
177: each side and angle.}
178: \label{fig:unit}
179: \end{figure}
180:
181: \section{The \babar\ Detector}
182:
183: A detailed description of the \babar\ detector can be found
184: elsewhere\cite{babarnim}. Charged particle momenta are measured
185: in a tracking system that consists of a 5-layer double-sided
186: silicon micro-strip vertex tracker (SVT) and a 40-layer drift chamber (DCH)
187: filled with an (80:20) mixture of helium and isobutane. The
188: tracking volume is contained within the 1.5T magnetic
189: field of a superconducting solenoid.
190: The combined track momentum resolution is $\sigma_{p_{T}}/p_{T}
191: =0.13\%\times p_{T} + 0.45\%$. The primary charged hadron identifiation device
192: is a detector of internally reflected Cerenkov radiation (DIRC). The typical
193: separation of kaons and pions due to their measured Cerenkov angle $\theta_C$
194: varies
195: from 8$\sigma$ at 2 GeV/c to 2.5$\sigma$ at 4 GeV/c, where $\sigma$ is the
196: average $\theta_C$ resolution. Specific ionization energy loss (dE/dx)
197: measurements in the DCH and SVT also contribute to charged hadron
198: identification for particle momenta less than 0.7 GeV/c.
199: Photons are detected in an electromagnetic
200: calorimeter (EMC) consisting of 6580 Thallium doped CsI crystals arranged in
201: barrel and forward end-cap sub-detectors. The $\pi^0$ mass resolution in
202: on average about 7 MeV/c$^2$. Muons and long-lived neutral hadrons are
203: detected within the instrumentation of the solenoid flux return (IFR) which
204: consists of alternating layers of iron and resistive plate chambers.
205:
206: \section{Common Analysis Features}
207:
208: \subsection{Data sample}
209:
210: The analyses described here use all or part of a data sample consisting of
211: approximately 88 million pairs of $\Upsilon (4S)\rightarrow B{\bar B}$ decays,
212: corresponding to a detector exposure of about 81 fb$^{-1}$.
213: An additional sample of
214: 9.6 fb$^{-1}$ taken about 40 MeV below the peak of the $\Upsilon (4S)$
215: resonance
216: (``off-resonance'') is used by many analyses to study $e^+e^-\rightarrow
217: q{\bar q}$ ``continuum'' backgrounds.
218:
219: \subsection{B Meson Reconstruction}
220:
221: B mesons produced from $\Upsilon (4S)$ decays are identified via their
222: unique kinematics. Because the mass of the $B$ meson pair is nearly that of
223: the $\Upsilon (4S)$, they are produced nearly at rest
224: ($p^*_B\approx 325$ MeV/c). Use of the beam energy in constraining the
225: kinematics serves to reduce the resolution of theses variables.
226:
227: The conservation of energy can be expressed as:
228: \begin{equation}
229: \Delta E = E^*_B - E^*_{beam} , \label{eqn3}
230: \end{equation}
231: where $E^*_{beam}$ is the single beam energy in the center-of-mass (CM)
232: frame.
233: $E^*_B$ is the measured energy of the $B$ candidate in the CM.
234: Correctly reconstructed $B$ candidates have $\Delta E$ distributed around
235: zero with a resolution ranging from 15 to 80 MeV. The energy resolution of
236: the $B$ decay products dominates the resolution of this variable. Continuum
237: background in this variable is well described by a monotonically decreasing
238: low order polynomial.
239: Figure~\ref{fig:de} shows the $\Delta E$ distribution for a typical rare mode
240: after all selection criteria have been applied (except that on $\Delta E$).
241: \begin{figure}[ht]
242: \begin{center}
243: %\leavevmode
244: \includegraphics[angle=-90,keepaspectratio=true,width=4.5in]{3k_de.eps}
245: \end{center}
246: \caption{$\Delta E$ distribution for charged $B$ decays to three charged kaons.
247: All candidate selection criteria have been applied except that on $\Delta E$.
248: The solid line shows the expected continuum background level.}
249: \label{fig:de}
250: \end{figure}
251:
252: We express momentum conservation as:
253: \begin{equation}
254: m_{ES} = \sqrt{E^{*2}_{beam} - {\vec p}^{*2}_B} . \label{eqn4}
255: \end{equation}
256: Here $m_{ES}$ is the ``beam-energy substituted mass'', with ${\vec p}^*_B$
257: the $B$-candidate momentum in the CM. Correctly reconstructed $B$ candidates
258: have $m_{ES}$ equal to the $B$ meson mass, with a resolution of about
259: 2.5-3.0 MeV/c$^2$, which is dominated by the beam energy spread. The continuum
260: background shape in $m_{ES}$ is parameterized by a threshold
261: function\cite{argus} with a fixed endpoint given by the average beam energy.
262: Figure~\ref{fig:mes} shows the $M_{ES}$ distribution for a typical rare mode
263: after all other selections have been applied.
264: \begin{figure}[ht]
265: \begin{center}
266: %\leavevmode
267: \includegraphics[angle=-90,keepaspectratio=true,width=4.5in]{3k_mes.eps}
268: \end{center}
269: \caption{$m_{ES}$ distribution for charged $B$ decays to three charged kaons.
270: All candidate selection criteria have been applied except that on $m_{ES}$.
271: The solid line shows the expected continuum background level.}
272: \label{fig:mes}
273: \end{figure}
274:
275: In addition to the kinematics of the $B$ meson, signal events are selected
276: by making requirements on the decay products. $B$ daughter resonances are
277: required to have invariant masses within a restricted range typically
278: determined by resolution and the need to leave sufficient sideband to determine
279: background levels. Particle identification requirements are made to
280: select some particles and veto sources of background.
281:
282: \subsection{Background Suppression}
283:
284: All rare analyses suffer from substantial backgrounds, and a variety of
285: techniques are employed to reduce this to manageable levels. In general,
286: backgrounds from other $B$ decays are small. Decays resulting from CKM
287: favored $b\rightarrow c$ transitions have heavier daughters and higher
288: multiplicity final states than do CKM suppressed decays. In order to wrongly
289: reconstruct such a decay as a rare signal, one must typically lose a particle
290: from the true $B$ decay, resulting is a substantial shift in the candidate's
291: $\Delta E$. The only exception to this is for rare modes with high final state
292: multiplicities, in which indications of some background have been observed.
293: While $B$ decays from other CKM suppressed transitions have similar kinematics
294: and multiplicities to that of a rare signal, such modes are rare themselves,
295: and require only limited suppression in most analyses. Where $B$ backgrounds
296: are present, they typically populate the sidebands of the $\Delta E$
297: distribution, but have tails that reach into the signal region as illustrated
298: in Figure~\ref{fig:mesvsde}
299: \begin{figure}[ht]
300: \begin{center}
301: %\leavevmode
302: \includegraphics[width=4.5in]{mesvsde2.ps}
303: \end{center}
304: \caption{Typical $m_{ES}$ vs $\Delta E$ distribution after event selection.
305: The signal populates the region around $m_{ES}=5.280$ and $\Delta E = 0$.
306: Continuum background populates the entire plane. $B$
307: background populates the $\Delta E$ sideband.}
308: \label{fig:mesvsde}
309: \end{figure}
310:
311: For all the modes
312: discussed here, the primary background is due to random particle combinations
313: arising from continuum quark-antiquark production. Although the probability
314: for any given continuum event to satisfy a signal selection is quite small,
315: the numbers favor the continuum. The total production cross section
316: for light quarks (including charm) under the $\Upsilon (4S)$ is about 3.5 nb,
317: but only 1 nb for the $\Upsilon (4S)$ itself. For a mode with an expected
318: branching fraction of order 10$^{-6}$ this means that continuum events are
319: produced at a rate well in excess of 10$^6$ times that of the signal.
320:
321: In order to control continuum backgrounds, one typically exploits the
322: fact that while the $B$ meson pairs are produced near threshold in
323: $\Upsilon (4S)$ decays, the light and charm quark
324: pairs which comprise the continuum
325: are produced with a great deal of excess energy. The result is that
326: for true $B$ meson decays, the decay products entering the detector are
327: distributed isotropically in the CM, while the continuum background
328: exhibits a ``jet-like'' topology, with a strong
329: correlation between the $B$ candidate decay and jet axes.
330:
331: The first topological variable typically employed is the angle $\theta_T$
332: between the thrust axes of the $B$ candidate and the remaining particles in
333: the event. The sphericity axes may be used almost interchangeably. The
334: distribution of $|cos \theta_T|$ is nearly uniform for true $B$ mesons, but is
335: strongly peaked near 1 for continuum background as is illustrated in
336: Figure~\ref{fig:thrust}. If additional background rejection is required, one
337: may consider the remaining event shape information, such as the angles between
338: the $B$ thrust and decay axes and the beam as well as the angular energy flow
339: in the event, and combine it into an optimized quantity using a neutral network
340: or a Fisher discriminant. An example of the the separation power of a
341: Fisher discriminant after a thrust cut has been made can be found in
342: Figure~\ref{fig:fisher}.
343: \begin{figure}[t]
344: \begin{center}
345: %\leavevmode
346: \includegraphics[angle=0,keepaspectratio=true,width=4.5in]{thrust.eps}
347: \end{center}
348: \caption{Signal and background distributions of $|cos\theta_{Thrust}|$ for
349: a typical $B$ meson decay. The signal distribution (solid histogram) is
350: uniform, reflection the random orientation between the $B$ candidate thrust
351: axis and the thrust axis of the rest of the event. The background
352: (open points)
353: is strongly peaked at one due to the strong correlation between these two
354: axes in fake $B$ candidates arising from random particle combinations in the
355: continuum.}
356: \label{fig:thrust}
357: \end{figure}
358: \begin{figure}[ht]
359: \begin{center}
360: %\leavevmode
361: \includegraphics[angle=-90,keepaspectratio=true,width=4.5in]{fisher.eps}
362: \end{center}
363: \caption{Distribution of Fisher discriminant for data control mode (solid
364: points), control mode signal Monte Carlo (solid
365: histogram), continuum data (open points) and continuum Monte Carlo
366: (dashed histogram) after a cut on
367: $|cos\theta_{Thrust}|$. The Fisher and Thrust angle are strongly correlated,
368: thus the separation will depend strongly on the thrust cut made.}
369: \label{fig:fisher}
370: \end{figure}
371:
372: \subsection{Signal Extraction}
373:
374: All rare analyses at \babar\ are performed ``blind'', meaning that signal
375: yields are hidden from the analyzer until the analysis has been peer
376: reviewed and determined to be in a final form. These steps are taken to
377: avoid experimenter's bias.
378:
379: There are two primary methods in which signal event yields are extracted; the
380: event counting analysis, and the maximum likelihood fit. Detection
381: efficiencies determined from signal Monte Carlo simulations and data control
382: samples are used to convert yields into branching fraction measurements. Equal
383: production of charged and neutral $B$'s from the $\Upsilon (4S)$ decay
384: are assumed throughout.
385:
386: In the event
387: counting analysis, a set of selection criteria are defined to select a signal
388: region of the parameter space. The criteria are optimized with respect to
389: expected signal and background yields to produce a measurement of the
390: greatest possible statistical significance. Systematic uncertainties may
391: figure into this process, but they are usually negligible for rare modes.
392: The selection are applied to the data, and the population of signal region is
393: counted. An estimated background is subtracted to determine the signal event
394: yield. The background yield is typically determined by measuring the density
395: of events in a sideband region and projecting that density into the signal
396: region.
397:
398: In the maximum likelihood fit method signal yields are determined by an
399: unbinned extended maximum likelihood fit to a set of observables. These
400: typically include $m_{ES}$, $\Delta E$, event shape variables and, where
401: appropriate, $B$ daughter resonance invariant
402: masses and particle identification information. The
403: probability
404: ${\cal P}_{i}({\vec x}_{j};{\vec \alpha}_{i})$ for a given hypothesis $i$
405: is the
406: product of probability density functions (PDFs) for each of the variables
407: ${\vec x}_{j}=(m_{ES}, \Delta E, {\rm Fisher}, ...)$ given
408: the set of parameters
409: ${\vec \alpha}_{i}$.
410: The hypotheses $i$ are signal, continuum background and
411: sometimes $B$ background for each final state in the fit.
412: The likelihood function is given by a product over all
413: events $N$ and the signal and background components:
414: \begin{equation}
415: {\cal L} = \frac{e^{-\sum_{i}n_i}}{N!}\prod_{j=1}^{N}{\cal L}_{j},\;\;\;
416: {\cal L}_{j} = \sum_{i}n_i{\cal P}_i({\vec x}_{j};{\vec \alpha}_{i}) .
417: \label{eqn5}
418: \end{equation}
419: The $n_i$ are the numbers of events for each hypothesis. The values of the
420: yields (and any free parameters in the PDFs) are taken as those which maximize
421: the likelihood function. Unit change in $-2 ln {\cal L}$ defines the one
422: standard deviation statistical uncertainties on the free parameters in the fit.
423: The statistical significance of the signal yield is determined from the
424: change in $-2 ln {\cal L}$ when the signal yield is forced to zero. If
425: no statistically significant signal is found (more than 4 standard deviations),
426: a 90\% confidence level upper limit may be obtained by requiring:
427: \begin{equation}
428: \frac{\int_{0}^{n_{UL}} {\cal L}(n)dn}{\int_{0}^{\infty}{\cal L}(n)dn} = 0.9.
429: \label{eqn6}
430: \end{equation}
431: Full and toy Monte Carlo simulations are used to verify that the fit is
432: unbiased.
433:
434: The accuracy with which the PDFs describe the data is of utmost importance in
435: the likelihood fit. Background PDFs are determined by fits to off-resonance and
436: sideband data. Signal PDFs are determined primarily from signal Monte Carlo
437: simulations, but ultimately rely on data control samples to verify their
438: validity.
439:
440:
441: \section{Electromagnetic Penguins}
442:
443: Electromagnetic penguins consist of the class of amplitudes in which an
444: external photon is emitted by one of the virtual particles
445: participating in the loop through which the $b\rightarrow s(d)$
446: transition proceeds.
447: This is illustrated in Figure~\ref{fig:empenguin}.
448: Such diagrams are relatively clean from a theoretical perspective, and a
449: variety of information can be gather from measurements of decays dominated
450: by these amplitudes. The decay $B\rightarrow K^*\gamma$ was the first
451: penguin to be
452: observed\cite{kstar_cleo}. Measurements of its branching fraction
453: provides a test of QCD, and direct CP violation in this mode would be an
454: indication of new physics. The decay rate ratio of $B\rightarrow\rho\gamma$
455: to $B\rightarrow K^*\gamma$ is sensitive to the ratio of
456: $|\frac{V_{td}}{V_{ts}}|$. The photon energy spectrum from measurements of
457: $b\rightarrow s\gamma$ provides information on the mass and Fermi motion
458: of the $b$ quark within the $B$ meson.
459:
460: \begin{figure}[t]
461: \begin{center}
462: %\leavevmode
463: \includegraphics[angle=0,keepaspectratio=true,width=3.5in]{EMpenguin2.ps}
464: \end{center}
465: \caption{Feynman diagram for an electromagnetic penguin amplitude.}
466: \label{fig:empenguin}
467: \end{figure}
468:
469: In the analysis of these modes, in each case there is a requirement
470: of a high energy isolated photon. The calorimeter cluster is
471: required to have a profile consistent with an electromagnetic shower,
472: and the candidate photon must not be consistent with having originated from
473: a $\pi^0$ or $\eta$ decay. Further details and results of each analysis are
474: presented below.
475:
476: \subsection{Measurement of $B\rightarrow K^*\gamma$}
477:
478: The analysis of $B\rightarrow K^*\gamma$\cite{babar_kstargamma} has been
479: performed on a data sample corresponding to approximately 22 million
480: $B{\bar B}$ pairs recorded in 1999-2000.
481: The $K^*\gamma$ final state is reconstructed in all four $K^*$ decay modes.
482: Stringent identification requirements are placed on charged kaons.
483: Invariant mass
484: requirements are placed on both $K^*$ and $K^0_S$ candidates. The $K^0_S$ is
485: also required to have a decay vertex displaced from the $e^+e^-$ interaction
486: point. Since the $B$ meson is a pseudoscalar, angular momentum conservation
487: requires that the $K^*$ is polarized. The absolute value of the cosine of
488: the $K^*$ helicity angle is required to be less than 0.75.
489: Continuum background is suppressed with cuts on the absolute values of the
490: cosines of the thrust and $B$ flight angles, both of 0.80.
491:
492: After cutting on $\Delta E$, the signal yield is determined from an
493: unbinned maximum likelihood fit to the $m_{ES}$ distribution, shown for each
494: $K^*$ decay mode in Figure~\ref{fig:kstar_mes}. Branching fraction and
495: direct CP asymmetry results are shown in
496: Table~\ref{tab:kstar_results}\cite{charge}.
497:
498: \begin{table}[hb]
499: \begin{center}
500: %\hspace*{-1.5cm}
501: \begin{tabular}{|l|c|c|c|}
502: \hline
503: &
504: ${\cal B}(B^0\rightarrow K^{*0}\gamma)$ &
505: ${\cal B}(B^+\rightarrow K^{*+}\gamma)$ &
506: $A_{CP}$ \\
507: \hline
508: \hline
509: Theory\cite{kstar_theory1,kstar_theory2,kstar_theory3} &
510: $7.5 \pm 3.0$ &
511: $7.5 \pm 3.0$ &
512: $|A_{CP}|<0.005$ \\
513: \hline
514: \babar &
515: $4.23 \pm 0.40 \pm 0.22$ &
516: $3.83 \pm 0.62 \pm 0.22$ &
517: $-0.17 < A_{CP} < 0.08$ \\
518: & & & @90\% CL \\
519: \hline
520: \end{tabular}
521: \end{center}
522: \caption{Results of the branching fraction and direct asymmetry analysis
523: of $B\rightarrow K^{*}\gamma$. In each result, the first uncertainty is
524: statistical, the second systematic. Branching fractions are in units of
525: $10^{-5}$.}
526: \label{tab:kstar_results}
527: \end{table}
528:
529: \begin{figure}[ht]
530: \begin{center}
531: %\leavevmode
532: \includegraphics[angle=0,keepaspectratio=true,totalheight=4.5in]{kstargamma.eps}
533: \end{center}
534: \vspace{-0.25in}
535: \caption{The $m_{ES}$ distribution for $B\rightarrow K^*\gamma$ in each
536: of the $K^*$ decay modes. All selection criteria have been applied.
537: The solid curve is the combined signal and background
538: PDF shape, the dashed is the background only.}
539: \label{fig:kstar_mes}
540: \end{figure}
541:
542: \subsection{Search for $B\rightarrow\rho\gamma$ and $B\rightarrow\omega\gamma$}
543:
544: The analysis of the $\rho\gamma$ and $\omega\gamma$ final
545: states\cite{babar_rhogamma} is
546: significantly more challenging than that of $K^*\gamma$. The predicted
547: branching fractions are about 50 times smaller than for $K^*\gamma$. Both
548: the $\rho$ and the $\omega$ have significantly more background under
549: their peaks than does the $K^*$, and the rho is much broader. In addition
550: to continuum background, these modes also potentially suffer
551: cross-feed background from
552: $K^*\gamma$, other $b\rightarrow s\gamma$ processes and
553: from $B\rightarrow\rho\pi^0$.
554:
555: A neutral network containing information from event shape, $\Delta t$ and
556: flavor tagging is used to control continuum background. $K^*\gamma$
557: feed-across is vetoed using particle identification. After these selection
558: criteria are applied, the signal yield for
559: each final state is extracted using a unbinned maximum likelihood fit to
560: $m_{ES}$, $\Delta E$, and the $\rho / \omega$ invariant mass. Studies of
561: generic $B{\bar B}$ Monte Carlo show that the expected $B$ background is
562: quite small, so the fit includes components only for signal and continuum
563: background. Background from $B$ decays is considered as a systematic
564: uncertainty.
565:
566: The results of this analysis applied to a sample of 84 million $B{\bar B}$
567: pairs can be found in table~\ref{tab:rhogamma_results}. If isospin symmetry
568: is assumed, all three modes can be combined to produce an upper limit of
569: ${\cal B}(B\rightarrow\rho\gamma)<1.9\times 10^{-6}$ @ 90\% CL. This result
570: can be used to place on upper limit on CKM parameters
571: $|\frac{V_{td}}{V_{ts}}| < 0.036$ @ 90\% CL. A discussion of theoretical
572: errors can be found in Ali and Parkhomonko\cite{kstar_theory3}.
573:
574: \begin{table}[h]
575: \begin{center}
576: %\hspace*{-1.5cm}
577: \begin{tabular}{|l|c|c|c|}
578: \hline
579: &
580: ${\cal B}(B^0\rightarrow \rho^{0}\gamma)$ &
581: ${\cal B}(B^+\rightarrow \rho^{+}\gamma)$ &
582: ${\cal B}(B^0\rightarrow \omega\gamma)$ \\
583: \hline
584: \hline
585: Theory\cite{kstar_theory3} &
586: $0.5-0.75$ &
587: $0.8-1.5$ &
588: $0.5-0.75$ \\
589: \hline
590: \babar &
591: $< 1.4$ &
592: $< 2.3$ &
593: $< 1.2$ \\
594: \hline
595: \end{tabular}
596: \end{center}
597: \vspace{-0.4cm}
598: \caption{Results of the branching fraction analysis
599: of $B\rightarrow \rho\gamma$ and $\omega\gamma$.
600: Branching ratios are in units of $10^{-6}$. Upper limits are at 90\% CL.}
601: \label{tab:rhogamma_results}
602: \end{table}
603:
604: \subsection{Semi-Inclusive Measurement of $b\rightarrow s\gamma$}
605:
606: This analysis\cite{babar_semibsg} of 22 million $B{\bar B}$ pairs
607: is a study of a collection of
608: exclusive final states with a kaon plus up to four pions, no more than one of
609: which may be neutral.
610: Because $b\rightarrow s\gamma$ is a two-body decay
611: process, the photon energy $E_{\gamma}$ in the $B$ rest frame is related to
612: the recoil hadronic mass, $M_{Had}$:
613: \begin{equation}
614: E_{\gamma} = \frac{M_B^2-M_{Had}^2}{2M_B}.
615: \label{eqn7}
616: \end{equation}
617: Fits to the measured spectra of both of these quantities can be used to
618: determine the total branching ratio for
619: $B\rightarrow X_s\gamma$\cite{bsg_theory1}. In addition to constraining
620: new physics contributions to the underlying amplitude, parameters associated
621: with heavy quark effective theory (HQET) are also extracted in the analysis.
622: These parameters are critical to reducing theory errors in the extraction of
623: $V_{ub}$ and $V_{cb}$.
624:
625: Measured branching fractions as a function of $M_{Had}$ and $E_{\gamma}$ can
626: be found in Figure~\ref{fig:semi_bsgspectra}. Analysis of these spectra
627: yield results:
628: \begin{eqnarray}
629: {\bar \Lambda} & = & 0.37 \pm 0.09 (stat) \pm 0.07 (syst) \pm 0.10 (model)
630: \;{\rm GeV/c^2} \nonumber \\
631: m_b & = & 4.79 \pm 0.08 (stat) \pm 0.10 (syst) \pm 0.10 (model)
632: \;{\rm GeV/c^2} \nonumber \\
633: \lambda_1 & = & -0.24^{+0.03}_{-0.04}\; (stat)
634: \pm 0.02 (syst)\; ^{+0.15}_{-0.21}
635: (model)
636: \;{\rm GeV/c^2} \nonumber \\
637: {\cal B}(b\rightarrow s\gamma) & = & 4.3 \pm 0.5 (stat) \pm 0.8 (syst)
638: \pm 1.3 (model) \times 10^{-4}.
639: \label{eqn11}
640: \end{eqnarray}
641:
642: \begin{figure}[ht]
643: \begin{center}
644: %\leavevmode
645: \psfiletwo{semibsg_had.eps}{semibsg_egam.eps}{1.0}%
646: \end{center}
647: \caption{Hadronic recoil mass and photon energy spectra for the semi-inclusive
648: analysis of $b\rightarrow s\gamma$. Errors are statistical only.}
649: \label{fig:semi_bsgspectra}
650: \end{figure}
651:
652: \subsection{Fully-Inclusive Measurement of $b\rightarrow s\gamma$}
653:
654: Much of the uncertainty in the semi-inclusive $b\rightarrow s\gamma$ analysis
655: arises from theoretical errors. HQET implies a duality between the quark and
656: hadron level of an interaction, which implies that parton level rate for
657: $b\rightarrow s\gamma$ is the same as the inclusive rate for
658: $B\rightarrow X_s\gamma$. These two issues motivate the fully inclusive
659: analysis technique.
660:
661: This analysis\cite{babar_fullbsg}
662: is performed on a sample of 60 million $B{\bar B}$ pairs. Photons in the
663: range $1.5 < E^*_\gamma < 3.5$ GeV are analyzed.
664: These photons are required to meet
665: the selection criteria described above. To suppress continuum background
666: the event is required to have a lepton flavor tag which strongly selects
667: true $B{\bar B}$ decays. Such a selection induces no model dependency into the
668: analysis as it only applies to the ``other'' $B$ in the decay.
669: In addition to particle identification criteria,
670: fake leptons are further rejected by requiring large missing energy in the
671: event, which is normally associated with semi-leptonic $b\rightarrow c$
672: transitions. To obtain additional discrimination, the angular separation
673: between the lepton and the photon is required not to be small. Event topology
674: in the form of the Fox-Wolfram moments in also employed to reduce background
675: from the continuum. Backgrounds are estimated using off-resonance data and
676: $B{\bar B}$ Monte Carlo.
677:
678: The $E_\gamma$ spectra for on-resonance data and the predicted background are
679: shown in Figure~\ref{fig:inclbsg_egam}.
680: \begin{figure}[ht]
681: \begin{center}
682: %\leavevmode
683: \includegraphics[angle=0,keepaspectratio=true,totalheight=3.25in]
684: {incbsg_egam.eps}
685: \end{center}
686: \vspace{-0.3in}
687: \caption{The $E^*_\gamma$ distribution of on-resonance data (solid points)
688: and background expectations for the fully inclusive analysis
689: of $B\rightarrow X_s\gamma$. Errors are statistical only.}
690: \label{fig:inclbsg_egam}
691: \end{figure}
692: The photon energy range
693: $2.1 < E^*_\gamma < 2.7$ GeV is considered to reduce model dependencies. The
694: branching fraction for $B\rightarrow X_s\gamma$ is measured in this region and
695: then extrapolated to the full spectrum:
696: \begin{equation}
697: {\cal B}(B\rightarrow X_s\gamma) = 3.88 \pm 0.36 (stat) \pm 0.37 (syst)
698: ^{+0.43}_{-0.23} (model) \times 10^{-4}.
699: \label{eqn12}
700: \end{equation}
701:
702: \section{Electroweak Penguins}
703:
704: As the name suggests, Electroweak penguins are amplitudes that proceed via
705: loops involving photons, $W$ or $Z$ bosons. Such processes are strongly
706: suppressed in the Standard Model, and as such, are excellent windows onto
707: potential new physics. Well controlled theoretical uncertainties aid in this
708: sensitivity. We discuss the analysis of four such final states
709: here: $K^{(*)}\ell^+\ell^-$\cite{babar_kll},
710: $\ell^+\ell^-$\cite{babar_ll},
711: $K\nu{\bar \nu}$\cite{babar_knunu},
712: and $\gamma\gamma$\cite{babar_2gamma}.
713:
714: \subsection{Measurement of $B\rightarrow K^{(*)} \ell^+\ell^-$}
715:
716: The flavor-changing neutral current decays $B\rightarrow K \ell^+\ell^-$
717: and $B\rightarrow K^* \ell^+\ell^-$ have predicted branching fractions
718: on the order $10^{-6}-10^{-7}$\cite{kll_theory1}.
719: The leading diagrams for this decay are electroweak penguin and box diagrams,
720: and can be found in Figure~\ref{fig:kll_diagram}.
721:
722: \begin{figure}[ht]
723: \begin{center}
724: %\leavevmode
725: \includegraphics[angle=0,keepaspectratio=true,width=4.5in]
726: {kll_diagrams.eps}
727: \end{center}
728: %\vspace{-0.3in}
729: \caption{Leading Electroweak penguin and box diagrams for the decay
730: $B\rightarrow K^{(*)} \ell^+\ell^-$.}
731: \label{fig:kll_diagram}
732: \end{figure}
733:
734: The decay rate for $B\rightarrow K^{(*)} \ell^+\ell^-$ is rather sensitive to
735: the presence of new physics. In particular, certain extensions to the SM can
736: vary the rate by more than a factor of two. In addition to the decay rate,
737: kinematic distributions accessible with higher statistics, such as the boson
738: $q^2$ distribution ($m_{\ell\ell}^2$) and the
739: forward-backward asymmetry in the $K^*$ channel are of considerable
740: interest as they are also quite
741: sensitive to non-SM physics, and are less model dependent than the overall
742: rate.
743:
744: The experimental challenge in this analysis is to control the various sources
745: of background. Background from $B\rightarrow$ charmonium decays which have the
746: same final state particles are control by vetoing regions in the
747: $\Delta E$ vs $m_{\ell\ell}$ plane. Continuum background is reduced
748: using a Fisher discriminant which in addition to event shape information
749: includes information on the $K\ell$ invariant mass, which serves to veto
750: $D\rightarrow K\ell\nu$. Combinatorics from semi-leptonic $B$ decays
751: are rejected using a B-likelihood built from the missing energy in the event,
752: vertex information, and the $B$ production angle. Finally, peaking backgrounds
753: from particle mis-identification are reduced by vetoing the $K^{(*)}\pi$ mass
754: in the region of the $D$ mass.
755:
756: After background rejection and particle identification criteria are applied,
757: the signal is extracted with a likelihood fit to $m_{ES}$ and $\Delta E$.
758: The results of this analysis on a sample of 88.4M $B{\bar B}$ pairs are:
759: \begin{equation}
760: {\cal B}(B\rightarrow K\ell^+\ell^-) = (0.78^{+0.24}_{-0.20} (stat)
761: ^{+0.11}_{-0.18} (syst) ) \times 10^{-6}
762: \label{eqn13}
763: \end{equation}
764: with a significance (including systematics) of $4.4\sigma$, and
765: \begin{equation}
766: {\cal B}(B\rightarrow K^*\ell^+\ell^-) = (1.68^{+0.68}_{-0.58} (stat)
767: \pm 0.28 (syst) ) \times 10^{-6}
768: \label{eqn14}
769: \end{equation}
770: with a significance of $2.8\sigma$. Since the $K^*$ result
771: is not significant, we report a 90\% CL upper limit:
772: \begin{equation}
773: {\cal B}(B\rightarrow K^*\ell^+\ell^-) < 3.0 \times 10^{-6}.
774: \label{eqn15}
775: \end{equation}
776: Combined projections of $m_{ES}$ and $\Delta E$ are shown in
777: Figure~\ref{fig:kll_proj}.
778: %\vspace{-0.4in}
779: \begin{figure}[ht]
780: \begin{center}
781: %\leavevmode
782: \includegraphics[angle=0,keepaspectratio=true,height=4.5in]
783: {kll_proj.eps}
784: \end{center}
785: \vspace{-0.65in}
786: \caption{Projections of $m_{ES}$ and $\Delta E$ for the combined
787: $K^{(*)}\ell^+\ell^-$ final states. The solid line is the combined signal
788: plus background PDF, the histogram are the data.}
789: \label{fig:kll_proj}
790: \end{figure}
791:
792: \subsection{Search for $B^0\rightarrow \ell^+\ell^-$}
793:
794: The decay of a $B$ meson to a pair of leptons is highly suppressed within the
795: SM by factors resulting from CKM, internal quark annihilation and helicity.
796: Leading diagrams are shown in Figure~\ref{fig:ll_diagrams}. Within the SM,
797: predicted branching fractions are $1.9\times 10^{-15}$ and $8.0\times 10^{-11}$
798: for the $e^+e^-$ and $\mu^+\mu^-$ channels respectively\cite{ll_theory1}.
799: The $e\mu$ channel is forbidden by lepton number conservation. New physics can
800: significantly alter these predictions\cite{ll_theory2}.
801:
802: \begin{figure}[ht]
803: \begin{center}
804: %\leavevmode
805: \includegraphics[angle=0,keepaspectratio=true,width=4.5in]
806: {ll_diagrams.eps}
807: \end{center}
808: %\vspace{-0.3in}
809: \caption{Leading diagrams for the decay
810: $B\rightarrow \ell^+\ell^-$.}
811: \label{fig:ll_diagrams}
812: \end{figure}
813:
814: In this analysis, the primary sources of background are from real lepton
815: production from continuum $c{\bar c}$ decays, pions which are mis-identified as
816: muons, and two photon processes. Continuum background is suppressed using the
817: thrust magnitude, and the angle between the thrust axes of the $B$ candidate
818: and the rest of the event. A track multiplicity cut serves to reject
819: two photon processes. The signal is selected by requiring two high momentum
820: leptons of opposite charge and good vertex information. Particle
821: identification requirements for both leptons are made. The signal yield is
822: determined by counting events in a signal region of $m_{ES}$ and $\Delta E$,
823: and subtracting an estimated background determined from the scaled population
824: of the $m_{ES}$ vs $\Delta E$ plane. The results of this analysis
825: applied to a sample of approximately 60 million $B{\bar B}$ pairs are
826: presented in Table~\ref{tab:ll_results}.
827:
828: \begin{table}[ht]
829: \begin{center}
830: %\hspace*{-1.5cm}
831: \begin{tabular}{|l|c|c|c|c|}
832: \hline
833: & $N_{GSB}$ & $N_{SigBox}$ & $N_{BG}$ & 90\% CL Upper Limit \\
834: \hline
835: \hline
836: ${\cal B}(B^0\rightarrow e^+e^-)$ & 25 & 1 & $0.60 \pm 0.24$
837: & $3.3 \times 10^{-7}$ \\
838: \hline
839: ${\cal B}(B^0\rightarrow \mu^+\mu^-)$ & 26 & 0 & $0.49 \pm 0.19$
840: & $2.7\times 10^{-7}$ \\
841: \hline
842: ${\cal B}(B^0\rightarrow \mu^+\mu^-)$ & 26 & 0 & $0.49 \pm 0.19$
843: & $2.7\times 10^{-7}$ \\
844: \hline
845: \end{tabular}
846: \end{center}
847: \caption{Results of the search for
848: $B\rightarrow\ell^+\ell^-$. $N_{GSB}$ is the population of the
849: $m_{ES}$ vs $\Delta E$ sideband and $N_{SigBox}$ the population of the signal
850: region after all selection criteria have been applied. $N_{BG}$
851: is the expected
852: background in the signal region based on the sideband population.}
853: \label{tab:ll_results}
854: \end{table}
855:
856: \subsection{Search for $B^+\rightarrow K^+ \nu{\bar \nu}$}
857:
858: Within the SM, the decay $b\rightarrow s \nu{\bar \nu}$ is a pure
859: electroweak flavor changing neutral current. The final state is nearly free
860: of strong interaction uncertainties, and hence the theoretical errors
861: associated with this decay are small. While the inclusive analysis is not
862: currently feasible, it is possible to search for the exclusive decay
863: $B^+\rightarrow K^+ \nu{\bar \nu}$. Summing over all neutrino species,
864: the SM prediction for this branching fraction\cite{babar_physbook} is
865: \begin{equation}
866: {\cal B}(B^+\rightarrow K^+\nu{\bar \nu}) = 3.8^{+1.2}_{-0.6} \times 10^{-6}.
867: \label{eqn16}
868: \end{equation}
869:
870: The presence of two neutrinos in the final state makes this analysis difficult,
871: as there are no kinematic constraints which may be applied to the signal $B$.
872: Instead, the strategy is to fully reconstruct the other $B$ from the
873: $\Upsilon (4S)$ decay, and compare the remaining particles in the event with
874: the signature expected from the signal. The ``tag'' $B$ is required to be
875: fully reconstructed as either $B^-\rightarrow D^0\ell^-{\bar \nu}$ or
876: $B^-\rightarrow D^{*0}\ell^-{\bar \nu}$. The $D^0$ is reconstructed in the
877: $K^-\pi^+$, $K^-\pi^+\pi^-\pi^+$ and $K^-\pi^+\pi^0$ modes, which results in a
878: total of about 0.5\% of all charged $B$s being reconstructed as tags. To
879: select signal events, a high momentum charged kaon is required in the
880: recoil of the tagged $B$. Additional requirements are made on the neutral
881: energy in the recoil and on the angle between the kaon and the tag side lepton.
882:
883: \begin{figure}[ht]
884: \begin{center}
885: %\leavevmode
886: \includegraphics[angle=0,keepaspectratio=true,width=4in]
887: {knunu_data.eps}
888: \end{center}
889: %\vspace{-0.3in}
890: \caption{Distribution of $K^+\nu{\bar \nu}$ candidates in the plane
891: defined by the electromagnetic energy not from the tag $B$, $E_{Left}$,
892: and the resolution scaled difference between the measured and
893: mean measured $D$ mass. The signal region is the bounded area at the
894: left of the plot.}
895: \label{fig:knunu_data}
896: \end{figure}
897:
898: Events are counted in a signal region in the
899: plane defined by the electromagnetic energy in the tag $B$ recoil, and
900: the difference between the reconstructed and
901: mean fitted $D$ mass, scaled by the fitted $D$ mass resolution. The
902: expected background, determined by scaling the sideband
903: population into the signal region, is subtracted from the signal region
904: population to determine the signal yield. This is illustrated in
905: Figure~\ref{fig:knunu_data}. In a sample of 60 million $B{\bar B}$ pairs,
906: the expected background in the signal region is 2.2 events, and there are two
907: events observed. The 90\% confidence level upper limit
908: on the branching fraction, including systematics, is
909: \begin{equation}
910: {\cal B}(B^+\rightarrow K^+\nu{\bar \nu}) < 9.4 \times 10^{-5}.
911: \label{eqn17}
912: \end{equation}
913:
914: \subsection{Search for $B^0\rightarrow \gamma\gamma$}
915:
916: The decay $B^0\rightarrow \gamma\gamma$ is an example of electroweak
917: annihilation. The SM expectation for this decay is small, with predictions
918: ranging from 0.1 to $2.3\times 10^{-8}$\cite{2gamma_theory1}. As with the
919: other modes discussed in this section, physics beyond the SM can result in
920: significant enhancements to this rate\cite{2gamma_theory2}.
921:
922: In this analysis, selection criteria are placed on the ratio of the 2nd to 0th
923: Fox-Wolfram moments, the cosine of the angle between one of the photons
924: (chosen at random) and the thrust axis of the rest of the event, and the $B$
925: production angle to suppress continuum background. Selected photons are
926: required not to be consistent with having come from a $\pi^0$ or $\eta$ decay.
927: The signal yield is determined by counting events in a signal region of the
928: plane defined by $m_{ES}$ and $\Delta E$, and subtracting the expected
929: background determined by scaling the the sideband population into the signal
930: region. The result for a sample of 22 million $B{\bar B}$ pairs is
931: \begin{equation}
932: {\cal B}(B^0\rightarrow\gamma\gamma) < 1.7 \times 10^{-6},
933: \label{eqn18}
934: \end{equation}
935: at the 90\% confidence level, including systematic uncertainties.
936:
937: \section{Gluonic Penguins (Charmless Hadronic B Decays)}
938:
939: Charmless hadronic $B$ decays proceed through a combination of CKM suppressed
940: tree ($b\rightarrow u$) and gluonic penguin ($b\rightarrow d,s$) amplitudes.
941: There are about 70 possible combinations of two-body decays in the lowest
942: pseudoscalar and vector nonets. These may be further broken into two groups;
943: two-body decays in which both $B$ daughters are kaons or pions, and
944: quasi-two-body decays in which at least one of the $B$ daughters is a
945: short-lived resonance. The two-body modes can be analyzed for information
946: on the CP phases $\alpha$ and $\gamma$, and have been found to have significant
947: penguin
948: contributions in addition to CKM allowed tree amplitudes. Several of the
949: quasi-two-body modes are sensitive to the CP phase $\beta$.
950: In addition to yielding
951: information about the Unitarity Triangle, decays in which penguin amplitudes
952: are dominant are sensitive to new physics. Our study of three-body $B$ decays
953: has thus far been limited to combinations of three charged kaons or pions.
954:
955: All of these modes share some common features. The primary source of
956: background is random particle combinations in the continuum, although modes
957: with large final state multiplicities or significant neutral energy may suffer
958: from non-negligible $B{\bar B}$ backgrounds. All the final states are
959: ultimately composed of high momentum kaons and pions, so the ability to
960: distinguish been these particles at high momenta is crucial.
961:
962:
963: \subsection{Two-Body Decays}
964:
965: Two body $B$ decays to kaons and pions are sensitive to the angle $\alpha$
966: of the Unitarity Triangle through the time-dependent CP violating asymmetry
967: in the decay $B\rightarrow\pi^+\pi^-$ and
968: to the the angle $\gamma$ through branching fractions and direct CP-violating
969: asymmetries of decays to various $\pi\pi$ and $K\pi$ final states. Because
970: there are substantial penguin amplitudes which contribute to the $\pi^+\pi^-$
971: final state in addition to the tree amplitude, the time-dependent asymmetry in
972: that mode does not directly measure $\alpha$. An isospin analysis of the rates
973: for all the $B\rightarrow\pi\pi$ decays is required to fully unfold the
974: effects of
975: the penguin contributions and determine the relationship between what is
976: measured from the $\pi^+\pi^-$ analysis ($\alpha_{eff}$) and $\alpha$.
977: Interference between
978: penguin and tree amplitudes may also lead to substantial direct
979: (time-independent) CP asymmetries in the $K\pi$ final states.
980:
981: In each mode, the signal is extracted using an unbinned extended maximum
982: likelihood fit, using the $m_{ES}$, $\Delta E$, a Fisher discriminant, and
983: where appropriate, Cerenkov angle residuals. Groups of related decays are fit
984: simultaneously. For example, the $\pi^+\pi^-$, $K^+\pi^-$ and
985: $K^+K^-$ yields are determined from a single fit. In these cases, $\Delta E$
986: and the Cerenkov angle residuals separate the signal modes from each other.
987: Branching fraction results for all two-body modes based on a sample of
988: 88 million $B{\bar B}$ pairs can be found in
989: Table~\ref{tab:2body}.
990:
991: \begin{table}[ht]
992: \begin{center}
993: %\hspace*{-1.5cm}
994: \begin{tabular}{|c|c|c|c|}
995: \hline
996: Decay & $N_{Signal}$ & ${\cal B}\times 10^{-6}$ & $A_{CP}$\\
997: \hline
998: \hline
999: $B^0\rightarrow\pi^+\pi^-$ & $157 \pm 19 $ & $4.7 \pm 0.6 \pm 0.2$ & \\
1000: \hline
1001: $B^0\rightarrow K^{\pm}\pi^{\mp}$ & $589 \pm 30 $ & $17.9 \pm 0.9 \pm 0.7$ &
1002: $-0.102 \pm 0.050 \pm 0.016$\\
1003: \hline
1004: $B^0\rightarrow K^+K^-$ & $1 \pm 8 $ & $<0.6$ & \\
1005: \hline
1006: $B^+\rightarrow\pi^+\pi^0$ & $125 \pm 22 $ & $5.5 \pm 1.0 \pm 0.6$ &
1007: $-0.03 \pm 0.18 \pm 0.02$ \\
1008: \hline
1009: $B^+\rightarrow K^+\pi^0$ & $239 \pm 22 $ & $12.8 \pm 1.2 \pm 1.0$ &
1010: $-0.09 \pm 0.09 \pm 0.01$ \\
1011: \hline
1012: $B^+\rightarrow K^0\pi^0$ & $86 \pm 13 $ & $10.4 \pm 1.5 \pm 0.8$ &
1013: $0.03 \pm 0.36 \pm 0.09$ \\
1014: \hline
1015: $B^0\rightarrow\pi^0\pi^0$ & $23 \pm 10 $ & $<3.6 \;\; (1.6^{+0.7\; +0.6}_{-0.6\; -0.3})$ & \\
1016: \hline
1017: $B^+\rightarrow K^0\pi^+$ & $172 \pm 17$ & $17.5 \pm 1.8 \pm 1.3$ &
1018: $-0.17 \pm 0.10 \pm 0.02$ \\
1019: \hline
1020: $B^+\rightarrow K^0 K^+$ & $<10$ & $<1.3$ & \\
1021: \hline
1022: \end{tabular}
1023: \end{center}
1024: \caption{Results of two-body branching fraction analyses\cite{babar_hphm,babar_hpi0,babar_hpi02,babar_2pi0}. The $\pi^0\pi^0$
1025: result has a statistical significance of $2.5\sigma$. The results for
1026: $K^0\pi^+$ and $K^0 K^+$ are based on 60 million $B{\bar B}$ pairs.
1027: Upper limits are at the
1028: 90\% confidence level.}
1029: \label{tab:2body}
1030: \end{table}
1031:
1032: Despite no central measurement of the $\pi^0\pi^0$ final state,
1033: it is still possible to
1034: place limits on the relationship between the measured parameter $\alpha_{eff}$
1035: and the Unitarity Triangle parameter $\alpha$. Using the bound of Grossman and
1036: Quinn\cite{grossman_quinn} and our measured values, we set an upper limit of
1037: $|\alpha_{eff}-\alpha| < 51^{\circ} $ at 90\% CL.
1038:
1039: \subsection{Quasi-Two-Body Decays}
1040:
1041: Quasi-two-body decays proceed through resonant intermediate states. The
1042: analysis of such modes is very similar to true two-body decays, but there are
1043: additional variables that provide separation between the signal and background,
1044: such as the resonance invariant mass and polarization (if the final
1045: state is a pseudoscaler-vector combination). We present the analyses of
1046: three groups of related quasi-two-body decays;
1047: $B\rightarrow\phi K^{(*)}$\cite{babar_phi},
1048: $B\rightarrow\omega h$\cite{babar_phi,babar_omega} and
1049: $B\rightarrow\eta^{(\prime)} K^{(*)}$\cite{babar_etap}.
1050:
1051: \subsubsection{$B\rightarrow\phi K^{(*)}$}
1052:
1053: The decay $b\rightarrow s{\bar s}s$ is CKM forbidden, thus the decay
1054: $B\rightarrow\phi K^{(*)}$ is a nearly pure gluonic penguin, as shown in
1055: Figure~\ref{fig:phi_diagram}. New physics might not only manifest itself
1056: as a deviation from the SM prediction for the decay rate, but since the mode
1057: $\phi K^0_S$ is the strange analog to $J/\psi K^0_S$, the CP phase one measures
1058: in a time-dependent analysis could be altered from it's SM value of $\beta$.
1059:
1060: \begin{figure}[t]
1061: \begin{center}
1062: %\leavevmode
1063: \includegraphics[angle=0,keepaspectratio=true,height=1.5in]
1064: {phi_diagram.eps}
1065: \end{center}
1066: %\vspace{-0.3in}
1067: \caption{Leading diagram for the decay $B\rightarrow\phi K^{(*)}$.}
1068: \label{fig:phi_diagram}
1069: \end{figure}
1070:
1071: The signal yield in each of these modes is determined from an extended unbinned
1072: maximum likelihood fit to $m_{ES}$, $\Delta E$, a Fisher discriminant and the
1073: $\phi\;(K^+K^-)$ invariant mass.
1074: For the $\phi K^0_S$, $\phi K^+$ and $\phi\pi^+$
1075: final states, the $\phi$ polarization is included in the fit, as is the
1076: Cerenkov angle residual for the charged states. For the $\phi K^*$ final
1077: states, the $K^*$ invariant mass is included in the fit.
1078: Significant signals are observed for both charged and
1079: neutral $B$ decays to $K$ and $K^*$ final states. The results of this
1080: analysis on a sample of 60 million $B{\bar B}$ pairs are
1081:
1082: \begin{eqnarray}
1083: {\cal B}(B^+\rightarrow \phi K^+)
1084: & = & (9.2 \pm 1.0 \pm 0.8) \times 10^{-6} \nonumber\\
1085: {\cal B}(B^+\rightarrow \phi K^0)
1086: & = & (8.7^{+1.7}_{-1.5} \pm 0.9) \times 10^{-6} \nonumber\\
1087: {\cal B}(B^+\rightarrow \phi K^{*+})
1088: & = & (9.7^{+4.2}_{-3.4} \pm 1.7) \times 10^{-6} \nonumber\\
1089: {\cal B}(B^+\rightarrow \phi K^{*0})
1090: & = & (8.7^{+2.5}_{-2.1} \pm 1.1) \times 10^{-6} \nonumber\\
1091: {\cal B}(B^+\rightarrow \phi\pi^+)
1092: & < & 0.56 \times 10^{-6} \; @ \; 90\%\; {\rm CL}.
1093: \label{eqn20}
1094: \end{eqnarray}
1095: A stringent limit is also placed on the decay
1096: $B^+\rightarrow\phi\pi^+$, which is both CKM and color suppressed.
1097:
1098: \subsubsection{$B\rightarrow\omega h\;(h=K,\pi)$}
1099:
1100: $B$ decays involving an omega and either a kaon or a pion proceed through
1101: a mixture of CKM suppressed $b\rightarrow u$ tree and CKM forbidden
1102: $b\rightarrow d,s$ penguin amplitudes. The analysis method is identical to
1103: that described for $\phi K^{(*)}$. The results in Table~\ref{tab:omega}
1104: are based on a sample
1105: of 22 million $B{\bar B}$, except for the $\omega K^0_S$ analysis, which
1106: was performed on a sample of 60 million $B{\bar B}$ pairs, and is a first
1107: observation.
1108:
1109: \begin{table}[ht]
1110: \begin{center}
1111: %\hspace*{-1.5cm}
1112: \begin{tabular}{|c|c|c|c|}
1113: \hline
1114: Final State & $N_{Signal}$ & $S(\sigma)$ & ${\cal B}\times 10^{-6}$ \\
1115: \hline
1116: \hline
1117: $\omega K^+$ & $6.4^{+5.6}_{-4.4}$ & 1.3 & $<4\;(1.4^{+1.3}_{-1.0}\pm 0.3)$ \\
1118: \hline
1119: $\omega K^0$ & $26.6^{+7.7}_{-6.6}$ & 6.6 & $5.9^{+1.7}_{-1.5}\pm 0.9$ \\
1120: \hline
1121: $\omega \pi^+$&$27.6^{+8.8}_{-7.7}$ & 4.9 & $6.6^{+2.1}_{-1.8}\pm 0.7$ \\
1122: \hline
1123: $\omega \pi^0$&$-0.9^{+5.0}_{-3.2}$ & - & $<3\;(-0.3\pm 1.1\pm 0.3)$ \\
1124: \hline
1125: \end{tabular}
1126: \end{center}
1127: \vspace{-0.5cm}
1128: \caption{Results of the branching fraction analysis of
1129: $B\rightarrow \omega h$. S is the statistic significance of the result.
1130: Upper limits are at the 90\% confidence level.}
1131: \label{tab:omega}
1132: \end{table}
1133:
1134:
1135: \subsubsection{$B\rightarrow\eta^{(\prime)} K^{(*)}$}
1136:
1137: $B$ decays to $\eta$ and $\eta^\prime$ with a kaon or $K^*$ proceed
1138: predominantly through penguins, although there is some $b\rightarrow u$ tree
1139: contribution as well. The decays $B\rightarrow\eta^\prime K$ and
1140: $B\rightarrow\eta K^*$ were the first gluonic penguins to be
1141: observed\cite{cleo_etap}, and the rates are much larger than initially
1142: expected. The best present conjecture\cite{etapk_theory} is that the tree
1143: and penguin amplitudes interfere in such a way as to enhance $\eta^\prime K$
1144: and $\eta K^*$ but suppress $\eta^\prime K^*$ and $\eta K$. Because of it's
1145: relatively large rate and nearly pure penguin content,
1146: $B\rightarrow\eta^\prime K^0_S$
1147: is also of considerable interest for measurements of time-dependent CP
1148: asymmetries, which within the SM should probe the angle $\beta$.
1149:
1150: Signals for these modes are extracted as described above for
1151: $\omega$ and $\phi$. The $\eta^\prime$ is reconstructed in two decay chains;
1152: $\eta(\gamma\gamma)\pi^+\pi^-$ and $\rho^0\gamma$. The $\eta$ is reconstructed
1153: as $\eta\rightarrow\gamma\gamma$ and $\eta\rightarrow\pi^+\pi^-\pi^0$. The
1154: results of these analyses are displayed in Table~\ref{tab:etap}. The
1155: data samples used for the $\eta$ and $\eta^\prime$ analyses are 22 and 60
1156: million $B{\bar B}$ pairs respectively.
1157:
1158: \begin{table}[ht]
1159: \begin{center}
1160: %\hspace*{-1.5cm}
1161: \begin{tabular}{|c|c|c|}
1162: \hline
1163: Final State & $N_{Signal}$ & ${\cal B}\times 10^{-6}$ \\
1164: \hline
1165: \hline
1166: $\eta^\prime K^+$ & $445 \pm 26 $ & $67 \pm 5 \pm 5$ \\
1167: \hline
1168: $\eta^\prime K^0$ & $135 \pm 15 $ & $46 \pm 6 \pm 4$ \\
1169: \hline
1170: $\eta^\prime K^{*0}$ & $5.2 \pm 3.4 $ & $<13\;(4.0^{+3.5}_{-2.4}\pm 1.0)$ \\
1171: \hline
1172: $\eta K^+$ & $12.9 \pm 5.7 $ & $<6.4\;(3.8^{+1.8}_{-1.5}\pm 0.2)$ \\
1173: \hline
1174: $\eta \pi^+$ & $8.0 \pm 5.9 $ & $<5.2\;(2.2^{+1.8}_{-1.6}\pm 0.1)$ \\
1175: \hline
1176: $\eta K^0$ & $5.7 \pm 3.3 $ & $<12\;(6.0^{+3.8}_{-2.9}\pm 0.4)$ \\
1177: \hline
1178: $\eta K^{*0}$ & $20.5 \pm 6.3 $ & $19.8^{+6.5}_{-5.6}\pm 1.5$ \\
1179: \hline
1180: $\eta K^{*+}$ & $14.3 \pm 6.6 $ & $22.1^{+11.1}_{-9.2}\pm 3.2$ \\
1181: \hline
1182: \end{tabular}
1183: \end{center}
1184: \vspace{-0.5cm}
1185: \caption{Results of the branching fraction analyses of
1186: $B\rightarrow \eta^{(\prime)} K^{(*)}$. Upper limits are at the
1187: 90\% confidence level.}
1188: \label{tab:etap}
1189: \end{table}
1190:
1191: \subsection{Three-Body Decays}
1192:
1193: We describe here the analysis of
1194: $B^+\rightarrow h^+h^-h^+$\cite{babar_3h},
1195: where $h$ is either a charged kaon or pion. An event counting analysis
1196: is performed over the full three particle dalitz plot. All final states are
1197: measured simultaneously, and unfolded to obtain branching fractions for each
1198: combination. Continuum background is suppressed using the thrust angle
1199: and a Fisher discriminant. In addition to continuum background, the open
1200: nature of the dalitz plot also admits background in some regions from
1201: $B^+\rightarrow J/\Psi K^+$ and $B^+\rightarrow D\pi^+ / DK^+$.
1202: These regions of the dalitz plot are vetoed.
1203: Charged particle identification is
1204: crucial to this analysis, and along with tracking, is the primary source
1205: of systematic uncertainty. Figure~\ref{fig:dalitz} shows the dalitz plots
1206: for $B^+\rightarrow K^+K^-K^+$ and
1207: $B^+\rightarrow K^+\pi^-\pi^+$. Results of this analysis on a sample of 56
1208: million $B{\bar B}$ pairs are
1209: \begin{eqnarray}
1210: {\cal B}(B^{\pm}\rightarrow \pi^{\pm}\pi^{\mp}\pi^{\pm})
1211: & < & 15 \times 10^{-6} \; @ \; 90\%\; {\rm CL} \; (8.5\pm 4.0\pm 3.6) \nonumber\\
1212: {\cal B}(B^{\pm}\rightarrow K^{\pm}\pi^{\mp}\pi^{\pm})
1213: & = & (59.2 \pm 4.7 (stat) \pm 4.9 (sys))\times 10^{-6} \nonumber\\
1214: {\cal B}(B^{\pm}\rightarrow K^{\pm}K^{\mp}\pi^{\pm})
1215: & < & 7 \times 10^{-6} \; @ \; 90\% \; {\rm CL} \; (2.1\pm 2.9\pm 2.0) \nonumber\\
1216: {\cal B}(B^{\pm}\rightarrow K^{\pm}K^{\mp}K^{\pm})
1217: & = & (34.7 \pm 2.0 (stat) \pm 1.8 (sys))\times 10^{-6}.
1218: \label{eqn19}
1219: \end{eqnarray}
1220:
1221: \begin{figure}[ht]
1222: \begin{center}
1223: %\leavevmode
1224: \psfiletwo{k2pidalitz.eps}{3kdalitz.eps}{1.0}%
1225: \end{center}
1226: \vspace{-1cm}
1227: \caption{Unbinned dalitz plots for
1228: $K^+\pi^-\pi^+$ (left) and $K^+K^-K^+$
1229: (right) for events in the
1230: signal region. No efficiency corrections have been applied to the dalitz
1231: plots, and the charm contributions have not been removed.}
1232: \label{fig:dalitz}
1233: \end{figure}
1234:
1235:
1236: \section{Conclusion and Outlook}
1237:
1238: We have presented a number of results for rare $B$ meson
1239: decays using all or part of a sample of approximately 88 million $B{\bar B}$
1240: pairs collected by the \babar\ detector. Updates of many of these analyses
1241: to the full data set are in progress. These results represent only a part of
1242: the
1243: spectrum of possible measurements of rare decays. The larger data sets
1244: that will be available in the coming years will allow us to more fully
1245: exploit rare decays to test the self consistency of the flavor sector of the
1246: Standard Model, and will perhaps offer the first glimpse of new physics which
1247: lies beyond.
1248:
1249: \section{Acknowledgments}
1250:
1251: We are grateful for the excellent luminosity and machine conditions provided
1252: by our PEP-II colleagues, and for the substantial dedicated effort from the
1253: computing organizations that support \babar. The author's work was performed
1254: under the auspices of the U.S. Department of Energy by the University of
1255: Colorado under Contract DE-FG03-95ER40894.
1256:
1257: \begin{thebibliography}{10}
1258:
1259: \bibitem{babar_kstargamma}
1260: B.~Aubert {\it et al.} [BABAR Collaboration],
1261: %``Measurement of B $\to$ K* gamma branching fractions and charge
1262: % asymmetries,''
1263: Phys.\ Rev.\ Lett.\ {\bf 88}, 101805 (2002)
1264: [arXiv:hep-ex/0110065].
1265: %%CITATION = HEP-EX 0110065;%%
1266: \bibitem{babar_rhogamma}
1267: B.~Aubert {\it et al.} [BABAR Collaboration],
1268: %``Search for the exclusive radiative decays B $\to$ rho gamma and
1269: %B0 $\to$ omega gamma,''
1270: arXiv:hep-ex/0207073.
1271: %%CITATION = HEP-EX 0207073;%%
1272: \bibitem{babar_semibsg}
1273: B.~Aubert {\it et al.} [BABAR Collaboration],
1274: %``b $\to$ s gamma using a sum of exclusive modes,''
1275: arXiv:hep-ex/0207074.
1276: %%CITATION = HEP-EX 0207074;%%
1277: \bibitem{babar_fullbsg}
1278: B.~Aubert {\it et al.} [BABAR Collaboration],
1279: %``Determination of the branching fraction for inclusive decays
1280: %B $\to$ X/s gamma,''
1281: arXiv:hep-ex/0207076.
1282: %%CITATION = HEP-EX 0207076;%%
1283: \bibitem{babar_kll}
1284: B.~Aubert {\it et al.} [BABAR Collaboration],
1285: %``Evidence for the flavor changing neutral current decays B $\to$ K l+ l-
1286: %and B $\to$ K* l+ l-,''
1287: arXiv:hep-ex/0207082.
1288: %%CITATION = HEP-EX 0207082;%%
1289: \bibitem{babar_ll}
1290: B.~Aubert {\it et al.} [BABAR Collaboration],
1291: %``Search for decays of B0 mesons into pairs of leptons,''
1292: arXiv:hep-ex/0207083.
1293: %%CITATION = HEP-EX 0207083;%%
1294: \bibitem{babar_knunu}
1295: B.~Aubert {\it et al.} [BABAR Collaboration],
1296: %``A search for B+ $\to$ K+ nu anti-nu,''
1297: arXiv:hep-ex/0207069.
1298: %%CITATION = HEP-EX 0207069;%%
1299: \bibitem{babar_2gamma}
1300: B.~Aubert {\it et al.} [BABAR Collaboration],
1301: %``Search for the decay B0 $\to$ gamma gamma,''
1302: Phys.\ Rev.\ Lett.\ {\bf 87}, 241803 (2001)
1303: [arXiv:hep-ex/0107068].
1304: %%CITATION = HEP-EX 0107068;%%
1305: \bibitem{babar_hphm}
1306: B.~Aubert {\it et al.} [BABAR Collaboration],
1307: %``Measurements of branching fractions and CP-violating asymmetries in B0 $\to$ pi+ pi-, K+ pi-, K+ K- decays,''
1308: Phys.\ Rev.\ Lett.\ {\bf 89}, 281802 (2002)
1309: [arXiv:hep-ex/0207055].
1310: %%CITATION = HEP-EX 0207055;%%
1311: \bibitem{babar_hpi0}
1312: B.~Aubert {\it et al.} [BABAR Collaboration],
1313: %``Measurements of branching fractions and direct CP asymmetries in
1314: %pi+ pi0, K+ pi0 and K0 pi0 B decays,''
1315: arXiv:hep-ex/0207065.
1316: %%CITATION = HEP-EX 0207065;%%
1317: \bibitem{babar_hpi02}
1318: B.~Aubert {\it et al.} [BABAR Collaboration],
1319: %``A study of the rare decays B0 $\to$ D/s(*)+ pi- and B0 $\to$ D/s(*)- K+,''
1320: arXiv:hep-ex/0207053.
1321: %%CITATION = HEP-EX 0207053;%%
1322: \bibitem{babar_2pi0}
1323: B.~Aubert {\it et al.} [BABAR Collaboration],
1324: %``A search for the decay B0 $\to$ pi0 pi0,''
1325: arXiv:hep-ex/0207063.
1326: %%CITATION = HEP-EX 0207063;%%
1327: \bibitem{babar_phi}
1328: A. Bevan, to appear in {\it Proceedings of the 31st International
1329: Conference on High Energy Physics}.
1330: \bibitem{babar_omega}
1331: B.~Aubert {\it et al.} [BABAR Collaboration],
1332: %``Measurements of the branching fractions of exclusive charmless B meson
1333: %decays with eta' or omega mesons,''
1334: Phys.\ Rev.\ Lett.\ {\bf 87}, 221802 (2001)
1335: [arXiv:hep-ex/0108017].
1336: %%CITATION = HEP-EX 0108017;%%
1337: \bibitem{babar_etap}
1338: P. Bloom, ``Branching Fractions for $B\rightarrow\eta^{(\prime)}K^{(*)}$'',
1339: 2002 Meeting of the APS Division of Particles and Fields, May 24, 2002. \\
1340: http://dpf2002.velopers.net/talks\_pdf/361talk.pdf
1341: \bibitem{babar_3h}
1342: B.~Aubert {\it et al.} [BABAR Collaboration],
1343: %``Measurements of the branching fractions of charmless three-body
1344: %charged B decays,''
1345: arXiv:hep-ex/0206004.
1346: %%CITATION = HEP-EX 0206004;%%
1347: \bibitem{babarnim}
1348: B.~Aubert {\it et al.} [BABAR Collaboration],
1349: %``The BaBar detector,''
1350: Nucl.\ Instrum.\ Meth.\ A {\bf 479}, 1 (2002)
1351: [arXiv:hep-ex/0105044].
1352: %%CITATION = HEP-EX 0105044;%%
1353: \bibitem{ckm}
1354: N.~Cabibbo,
1355: %``Unitary Symmetry And Leptonic Decays,''
1356: Phys.\ Rev.\ Lett.\ {\bf 10}, 531 (1963).
1357: %%CITATION = PRLTA,10,531;%%
1358: M.~Kobayashi and T.~Maskawa,
1359: %``CP Violation In The Renormalizable Theory Of Weak Interaction,''
1360: Prog.\ Theor.\ Phys.\ {\bf 49}, 652 (1973).
1361: %%CITATION = PTPKA,49,652;%%
1362: \bibitem{argus}
1363: H.~Albrecht {\it et al.} [ARGUS Collaboration],
1364: %``Search For Hadronic B $\to$ U Decays,''
1365: Phys.\ Lett.\ B {\bf 241}, 278 (1990).
1366: %%CITATION = PHLTA,B241,278;%%
1367: \bibitem{kstar_cleo}
1368: T.~E.~Coan {\it et al.} [CLEO Collaboration],
1369: %``Study of exclusive radiative B meson decays,''
1370: Phys.\ Rev.\ Lett.\ {\bf 84}, 5283 (2000)
1371: [arXiv:hep-ex/9912057].
1372: %%CITATION = HEP-EX 9912057;%%
1373: \bibitem{charge} Except as noted explicitly, we use a particle name to denote
1374: either member of a charge conjugate pair.
1375: \bibitem{kstar_theory1}
1376: S.~W.~Bosch and G.~Buchalla,
1377: %``The radiative decays B $\to$ V gamma at next-to-leading order in QCD,''
1378: Nucl.\ Phys.\ B {\bf 621}, 459 (2002)
1379: [arXiv:hep-ph/0106081].
1380: %%CITATION = HEP-PH 0106081;%%
1381: \bibitem{kstar_theory2}
1382: M.~Beneke, T.~Feldmann and D.~Seidel,
1383: %``Systematic approach to exclusive B $\to$ V l+ l-, V gamma decays,''
1384: Nucl.\ Phys.\ B {\bf 612}, 25 (2001)
1385: [arXiv:hep-ph/0106067].
1386: %%CITATION = HEP-PH 0106067;%%
1387: \bibitem{kstar_theory3}
1388: A.~Ali and A.~Y.~Parkhomenko,
1389: %``Branching ratios for B $\to$ rho gamma decays in next-to-leading order in alpha(s) including hard spectator corrections,''
1390: Eur.\ Phys.\ J.\ C {\bf 23}, 89 (2002)
1391: [arXiv:hep-ph/0105302].
1392: %%CITATION = HEP-PH 0105302;%%
1393: \bibitem{bsg_theory1}
1394: A.~L.~Kagan and M.~Neubert,
1395: %``{QCD} anatomy of B $\to$ X/s gamma decays,''
1396: Eur.\ Phys.\ J.\ C {\bf 7}, 5 (1999)
1397: [arXiv:hep-ph/9805303].
1398: %%CITATION = HEP-PH 9805303;%%
1399: \bibitem{kll_theory1}
1400: A.~Ali, P.~Ball, L.~T.~Handoko and G.~Hiller,
1401: %``A comparative study of the decays B $\to$ (K,K*) l+ l- in standard model and supersymmetric theories,''
1402: Phys.\ Rev.\ D {\bf 61}, 074024 (2000)
1403: [arXiv:hep-ph/9910221].
1404: %%CITATION = HEP-PH 9910221;%%
1405: \bibitem{ll_theory1}
1406: A.~Ali, C.~Greub and T.~Mannel,
1407: %``Rare B decays in the Standard Model,''
1408: DESY-93-016
1409: %\href{http://www.slac.stanford.edu/spires/find/hep/www?r=desy-93-016}{SPIRES entry}
1410: {\it To be publ. in Proc. of ECFA Workshop on the Physics of a B Meson Factory, Eds. R. Aleksan, A. Ali, 1993}
1411: \bibitem{ll_theory2}
1412: M.~Gronau and D.~London,
1413: %``New physics in CP asymmetries and rare B decays,''
1414: Phys.\ Rev.\ D {\bf 55}, 2845 (1997)
1415: [arXiv:hep-ph/9608430].
1416: %%CITATION = HEP-PH 9608430;%%
1417: \bibitem{babar_physbook}
1418: P.~F.~Harrison and H.~R.~Quinn [BABAR Collaboration],
1419: %``The BaBar physics book: Physics at an asymmetric B factory,''
1420: SLAC-R-0504
1421: %\href{http://www.slac.stanford.edu/spires/find/hep/www?r=slac-r-0504}{SPIRES entry}
1422: {\it Papers from Workshop on Physics at an Asymmetric B Factory (BaBar Collaboration Meeting), Rome, Italy, 11-14 Nov 1996, Princeton, NJ, 17-20
1423: Mar 1997, Orsay, France, 16-19 Jun 1997 and Pasadena, CA, 22-24 Sep 1997}
1424: \bibitem{2gamma_theory1}
1425: G.~G.~Devidze and G.~R.~Dzhibuti,
1426: %``Double-Photon Decays Of B/S And B/D Mesons In The Mssm,''
1427: Phys.\ Lett.\ B {\bf 429}, 48 (1998).
1428: %%CITATION = PHLTA,B429,48;%%
1429: \bibitem{2gamma_theory2}
1430: T.~M.~Aliev and G.~Turan,
1431: %``B(S) $\to$ Gamma Gamma Decay In The Two Higgs Doublet Model,''
1432: Phys.\ Rev.\ D {\bf 48}, 1176 (1993).
1433: %%CITATION = PHRVA,D48,1176;%%
1434: \bibitem{grossman_quinn}
1435: Y.~Grossman and H.~R.~Quinn,
1436: %``Bounding the effect of penguin diagrams in a(CP)(B0 $\to$ pi+ pi-),''
1437: Phys.\ Rev.\ D {\bf 58}, 017504 (1998)
1438: [arXiv:hep-ph/9712306].
1439: %%CITATION = HEP-PH 9712306;%%
1440: \bibitem{cleo_etap}
1441: S.~J.~Richichi {\it et al.} [CLEO Collaboration],
1442: %``Two-body B meson decays to eta and eta': Observation of B $\to$ eta K*,''
1443: Phys.\ Rev.\ Lett.\ {\bf 85}, 520 (2000)
1444: [arXiv:hep-ex/9912059].
1445: %%CITATION = HEP-EX 9912059;%%
1446: \bibitem{etapk_theory}
1447: M.~Beneke and M.~Neubert,
1448: %``Flavor-singlet B decay amplitudes in QCD factorization,''
1449: Nucl.\ Phys.\ B {\bf 651}, 225 (2003)
1450: [arXiv:hep-ph/0210085].
1451: %%CITATION = HEP-PH 0210085;%%
1452: \end{thebibliography}
1453:
1454: \end{document}
1455: