hep-ph0404161/BtoX.tex
1: %\documentclass[preprint,tightenlines,aps,prd,groupedaddress,showpacs]
2: \documentclass[twocolumn,aps,prl,groupedaddress,showpacs]{revtex4}
3: % \documentclass[preprint,tightenlines,aps,prd,showpacs]{revtex4}
4: %\documentclass[preprint,aps,prd,showpacs]{revtex4}
5: \usepackage{graphicx}
6: \usepackage{amsmath}
7: \usepackage{bm}
8: 
9: 
10: \begin{document}
11: 
12: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
13: %Title of paper
14: \title{\mbox{}\\[10pt]
15: Production of the $X(3872)$ in $B$ Meson Decay\\
16: by the Coalescence of Charm Mesons}
17: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18: 
19: \author{Eric Braaten and Masaoki Kusunoki}
20: %\email[]{Your e-mail address}
21: %\homepage[]{Your web page}
22: %\thanks{}
23: %\altaffiliation{}
24: \affiliation{
25: Physics Department, Ohio State University, 
26: Columbus, Ohio 43210, USA}
27: \author{Shmuel Nussinov}
28: \affiliation{
29: School of Physics and Astrononmy,
30: Tel Aviv University, Ramat Aviv, 
31: Tel Aviv 69978, Israel}
32: 
33: 
34: \date{\today}
35: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36: \begin{abstract}
37: % insert abstract here
38: If the recently-discovered charmonium state $X(3872)$ is a 
39: loosely-bound S-wave molecule of the charm mesons 
40: $\bar D^0 D^{*0}$ or $\bar D^{*0} D^0$, it can be 
41: produced in $B$ meson decay by the coalescence of charm mesons.
42: If this coalescence mechanism dominates,
43: the ratio of the differential rate for $B^+ \to \bar D^0 D^{*0} K^+$
44: near the $\bar D^0 D^{*0}$ threshold and the rate for
45: $B^+ \to X K^+$ is a function of the $\bar D^0 D^{*0}$ 
46: invariant mass and hadron masses only.  
47: The identification of the $X(3872)$ as a
48: $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ molecule can be confirmed 
49: by observing an enhancement in the 
50: $\bar D^0 D^{*0}$ invariant mass distribution near the threshold. 
51: An estimate of the branching fraction 
52: for $B^+ \to X K^+$ is consistent with observations 
53: if $X$ has quantum numbers $J^{PC} = 1^{++}$ and if 
54: $J/\psi \; \pi^+ \pi^-$ is one of its major decay modes.
55: \end{abstract}
56: 
57: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
58: % insert suggested PACS numbers in braces on next line
59: \pacs{12.38.-t, 12.38.Bx, 13.20.Gd, 14.40.Gx}
60: % 12.38.-t   Quantum chromodynamics
61: % 12.38.Bx   Perturbative calculations
62: % 13.20.Gd  Decays of J/psi, Upsilon, and other quarkonia
63: % 14.40.Gx   Mesons with S=C=B=0, mass > 2.5 GeV (including quarkonia)
64: 
65: 
66: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
67: % insert suggested keywords - APS authors don't need to do this
68: %\keywords{}
69: 
70: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
71: %\maketitle must follow title, authors, abstract, \pacs, and \keywords
72: \maketitle
73: 
74: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
75: % body of paper here - Use proper section commands
76: % References should be done using the \cite, \ref, and \label commands
77: 
78: The recent unexpected discovery 
79: of a narrow charmonium resonance near 3.87 GeV 
80: challenges our understanding of heavy quarks and QCD.
81: This mysterious state $X(3872)$
82: was discovered by the Belle collaboration in electron-positron 
83: collisions through the $B$-meson decay $B^\pm  \to K^\pm X$
84: followed by the decay $X  \to J/\psi \; \pi^+ \pi^-$ \cite{Choi:2003ue}.
85: The discovery was confirmed by the CDF collaboration using
86: proton-antiproton collisions \cite{Acosta:2003zx}.
87: The $X$ is much narrower than all other charmonium states  
88: above the threshold for decay into a pair of charm mesons.
89: Its mass is also extremely close to the threshold for decay
90: into the charmed mesons $\bar D^0 D^{*0}$ or $\bar D^{*0} D^0$.
91: 
92: The proposed interpretations of the $X(3872)$
93: include a D-wave charmonium state with 
94: quantum numbers $J^{PC}= 2^{--}$ or $2^{-+}$,
95: an excited P-wave charmonium state 
96: with $J^{PC}= 1^{++}$ or $1^{+-}$,
97: a ``hybrid charmonium'' state in which a gluonic mode 
98: has been excited, and a $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$  
99: molecule 
100: \cite{Tornqvist:2004qy,Close:2003sg,Pakvasa:2003ea,Voloshin:2003nt,%
101: Yuan:2003yz,Wong:2003xk,Braaten:2003he,Barnes:2003vb,Swanson:2003tb,%
102: Eichten:2004uh,Quigg:2004nv}. 
103: The possibility that charm mesons might form molecular states was 
104: considered some time ago \cite{Voloshin:ap,Nussinov:1976fg,Tornqvist:1991ks}.  
105: If the binding is due to pion exchange, the most favorable channels
106: are S-wave with quantum numbers $J^{PC} = 1^{++}$ or P-wave with $0^{-+}$
107: \cite{Tornqvist:2004qy}.
108: The proximity of the mass of $X$ to the $\bar D^0 D^{*0}$ threshold
109: indicates that it is extremely loosely bound.
110: If $X$ is an S-wave $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ molecule, 
111: the tiny binding energy introduces a new length scale,
112: the $\bar D^0 D^{*0}$ scattering length $a$,
113: that is much larger than other QCD length scales.
114: As a consequence, certain properties of the 
115: $X$/$\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ system are determined by $a$ 
116: and are insensitive to the shorter distance scales of QCD.
117: This phenomenon is called {\it low-energy universality}. 
118: 
119: A challenge for any interpretation of the $X(3872)$
120: is to explain its production rate. 
121: This could be problematic for the identification of $X$ 
122: as an S-wave $\bar{D}^0D^{*0}$/$D^0\bar{D}^{*0}$ molecule, 
123: because it can readily dissociate due to its tiny binding energy.
124: One way to produce $X$ is to produce $\bar{D}^0$ 
125: and $D^{*0}$ with small enough relative momentum that they 
126: can coalesce into $X$. 
127: An example is the decay 
128: $\Upsilon(4S) \to X h h'$, where $h$ and $h'$ are light hadrons,
129: which can proceed through the coalescense into $X$ of charm mesons 
130: from the 2-body decays of a virtual $B$ and a virtual $\bar B$. 
131: Remarkably, low-energy universality determines
132: the decay rate for this process in terms of hadron masses 
133: and the width $\Gamma_B$ of the $B$ meson \cite{Braaten:2004rn}.  
134: Unfortunately, the rate is supressed by a factor of 
135: $(\Gamma_B/m_B)^2$ and is many orders of magnitude too small 
136: to be observed.
137: 
138: In this paper, we apply low-energy universality to the 
139: discovery mode $B^+ \to X K^+$ and to the process 
140: $B^+ \to \bar D^0 D^{*0} K^+$. 
141: We point out that the interpretation of $X$ as an S-wave 
142: $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ molecule can be confirmed 
143: by observing a peak in the $\bar D^0 D^{*0}$ invariant mass 
144: distribution near the $\bar D^0 D^{*0}$ threshold
145: in the decay $B^+ \to \bar D^0 D^{*0} K^+$.
146: We also estimate the branching fraction for $B^+ \to X K^+$.
147: The estimate is compatible with observations 
148: if $X$ has quantum numbers $J^{PC} = 1^{++}$ and if 
149: $J/\psi \; \pi^+ \pi^-$ is one of its major decay modes.
150: 
151: The mass of the $X$ has been measured to be 
152: $m_X = 3872.0 \pm 0.6 \pm 0.5$ MeV by Belle \cite{Choi:2003ue} and
153: $3871.4 \pm 0.7 \pm 0.4$ MeV by CDF \cite{Acosta:2003zx}.
154: It is extremely close to the $\bar D^0 D^{*0}$ threshold 
155: $3871.2 \pm 0.7$ MeV.
156: The binding energy is $E_b = -0.5 \pm 0.9$ MeV.  
157: If the state is bound, $E_b$ is positive, so
158: it is likely to be less than 0.4 MeV.
159: This is the smallest binding energy of any
160: S-wave two-hadron bound state.
161: The next smallest is the deuteron, 
162: a proton-neutron state with binding energy 2.2 MeV.
163: For two hadrons whose low-energy interactions are mediated by 
164: pion exchange, the natural scale for the binding energy of 
165: a molecule is $m_\pi^2/(2 \mu)$,
166: where $\mu$ is the reduced mass of the two hadrons.  
167: For a  $\bar D^0 D^{*0}$ molecule, this scale is 
168: about 10 MeV, so $E_b$ is at least an order
169: of magnitude smaller than the natural low-energy scale.
170: 
171: If the binding energy of $X$ is so small, low-energy universality 
172: implies that the $X$/$\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ system
173: has properties that are determined by the 
174: $\bar D^0 D^{*0}$ scattering length $a$ and are 
175: insensitive to the shorter distance scales of QCD.  
176: The universal binding energy of the molecule is 
177: %
178: \begin{eqnarray}
179: E_b \equiv m_D+m_{D^*} - m_X \simeq \left( 2 \mu a^2 \right)^{-1},
180: \label{B2}
181: \end{eqnarray}
182: %
183: where $\mu = m_D m_{D^*}/(m_D+m_{D^*})$
184: is the reduced mass of the $\bar D^0$ and $D^{*0}$. 
185: The universal normalized momentum-space wavefunction 
186: at relative momentum $k \ll m_\pi$,
187: %
188: \begin{eqnarray}
189: \psi(k) \simeq (8\pi/a)^{1/2} (k^2+1/a^2)^{-1} ,
190: \label{psi}
191: \end{eqnarray}
192: %
193: was used by Voloshin 
194: to calculate the momentum distributions for the decays 
195: $X \to \bar D^0 D^0 \pi^0$ and $X \to \bar D^0 D^0 \gamma$ 
196: \cite{Voloshin:2003nt}.
197: The universal $\bar{D}^0 D^{*0}$ elastic scattering
198: amplitude at relative momentum $k_{\rm cm} \ll m_\pi$ is
199: %
200: \begin{equation}
201: {\cal A}[\bar{D}^0 D^{*0} \to \bar{D}^0 D^{*0}] \simeq 
202: \frac{8\pi m_D m_{D^*}}{\mu \left(-1/a-i k_{\rm cm} \right)},
203: \label{eq:scat}
204: \end{equation}
205: %
206: where $k_{\rm cm} \approx [2\mu(E - m_D - m_{D^*})]^{1/2}$
207: and $E$ is the total energy in the center-of-momentum frame.
208: The amplitude ${\cal A}[\bar D^{*0} D^0 \to \bar{D}^0 D^{*0}]$ 
209: for scattering to the CP conjugate state  differs by 
210: the charge conjugation $C = \pm$ of the channel 
211: with the large scattering length.
212: Another consequence of low-energy universality is that as 
213: the binding energy $E_b$ decreases, the probabilities for 
214: components of the wavefunction other than
215: $\bar D^0 D^{*0}$ and $\bar D^{*0} D^0$ decrease as $E_b^{1/2}$ 
216: \cite{Braaten:2003he}.
217: In the limit $E_b \to 0$, the state becomes 
218: $(|\bar D^{*0} D^0\rangle \pm |\bar D^0 D^{*0}\rangle )/\sqrt{2}$
219: if  $C = \pm$.
220: The rates for decays that do not correspond to the decay of a 
221: constituent $D^{*0}$ or $\bar D^{*0}$ also decrease as $E_b^{1/2}$.
222: This suppression may explain the surprisingly narrow width of the $X$.
223: 
224: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
225: \begin{figure}
226: \includegraphics[width=7cm]{./Bfig1.eps}
227: \caption{Feynman diagram for 
228: $B^+ \rightarrow X K^+$ via the first pathway. 
229: \label{fig:B-XK}}
230: \end{figure}
231: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
232: 
233: The decay $B^+ \to X K^+$ proceeds through the weak decay
234: $\bar b \to \bar c c s$ at very short distances.
235: The subsequent formation of $X K^+$ is a QCD process
236: that involves momenta $k$ as low as $1/a$. The contributions
237: from $k \sim 1/a$ are constrained by low-energy universality, 
238: but those from $k \gtrsim m_\pi$ involve the full
239: complications of low-energy QCD. 
240: We analyze the decay $B^+ \to X K^+$ by separating 
241: short-distance effects involving  $k \gtrsim m_\pi$ 
242: from long-distance effects involving  $k \sim 1/a$.  
243: The decay can proceed via the short-distance
244: 3-body decay $B^+ \to \bar{D}^0  D^{*0} K^+$ 
245: followed by the long-distance coalescence process 
246: $\bar{D}^0 D^{*0} \to X$. It can
247: also proceed through a second pathway consisting of 
248: $B^+ \to \bar D^{*0} D^0  K^+$ followed by $D^0  \bar{D}^{*0} \to X$. 
249: The amplitude for the first pathway can be expressed as
250: %
251: \begin{eqnarray}
252: &&{\cal A}_1[B^+\to X K^+] 
253: =  - i  {\sum} \int \! \! \frac{d^4\ell}{(2\pi)^4} \,  
254: {\cal A}[B^+\to  \bar D^0 D^{*0} K^+] 
255: \nonumber 
256: \\
257: && \hspace{0.5cm} \times 
258: D(q+\ell,m_D) \, D(q_*-\ell,m_{D^*}) \, {\cal A}[ \bar D^0 D^{*0}\to X ],
259: \label{eq:amp1}
260: \end{eqnarray}
261: %
262: where $q = (m_D/m_X)Q$ and $q_* = (m_{D^*}/m_X)Q$
263: are 4-momenta that add up to the 4-momentum $Q$ of $X$ 
264: and $D(p,m) = (p^2 - m^2 + i \epsilon)^{-1}$.
265: The sum is over the spin states of the $D^{*0}$.
266: This amplitude can be represented by the Feynman diagram 
267: with meson lines shown in Fig.~\ref{fig:B-XK}.
268: We constrain the loop integral to the small-momentum region 
269: by imposing a cutoff $|\mbox{\boldmath  $\ell$}| < \Lambda$ 
270: in the rest frame of the virtual $D^0$ and $\bar D^{*0}$.
271: The natural scale for the cutoff is $\Lambda \sim m_\pi$.
272: The amplitude for  $\bar D^0 D^{*0}$ to coalesce into $X$ 
273: is determined by the $\bar D^0 D^{*0}$ scattering length $a$:
274: %
275: \begin{eqnarray}
276: &&{\cal A}[\bar D^0 D^{*0}\to X ] 
277: \nonumber
278: \\
279: && \hspace{1cm}
280: = \left( 16\pi Z m_X m_D m_{D^*}/\mu^2 a \right)^{1/2} 
281: \epsilon_X^* \cdot \epsilon,
282: \label{eq:ampx}
283: \end{eqnarray}
284: %
285: where $\epsilon_X$ and $\epsilon$ are the polarization vectors 
286: of $X$ and $D^{*0}$ and $Z$ is the probability for the $X$
287: to be in a $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ state.  
288: At the $\bar D^0 D^{*0}$ threshold, the amplitude for 
289: $B^+ \to \bar D^0 D^{*0} K^+$ 
290: is constrained by Lorentz invariance to have the form
291: %
292: \begin{eqnarray}
293: {\cal A}[B^+ \to \bar D^0 D^{*0} K^+] 
294: = c_1 \, P \cdot \epsilon^*,
295: \label{eq:A-DD*K}
296: \end{eqnarray}
297: %
298: where $P$ is the 4-momentum of the $B$ meson and $c_1$ is a constant. 
299: The amplitude for 
300: $B^+ \to \bar D^{*0} D^0 K^+$ has the same form with $c_1$ replaced by 
301: a constant $c_2$.
302: In the $\bar D^0 D^{*0}$ rest frame, the integral over $\ell_0$ of the two 
303: propagators in (\ref{eq:amp1}) is proportional 
304: to the momentum-space wavefunction of $X$.
305: The subsequent integral over $\bm \ell$ is linear in the
306: ultraviolet cutoff $\Lambda$ for the low-momentum region:
307: %
308: \begin{eqnarray}
309: \int \!\! \frac{d^4\ell}{(2\pi)^4}\,  
310: D(q+\ell,m_D) \, D(q'-\ell,m_{D^*}) =
311: {i \mu \Lambda \over 4 \pi^2 m_D m_{D^*}} .
312: \end{eqnarray}
313: %
314: The total amplitude from the two pathways is 
315: %
316: \begin{eqnarray}
317: {\cal A}[B^+\to X K^+] = 
318: - \left( Z m_X/\pi^3 m_D m_{D^*} a \right)^{1/2}
319: \nonumber
320: \\
321: \times (c_1 \pm c_2) \Lambda \, P \cdot \epsilon_X^*.
322: \label{eq:ampsum}
323: \end{eqnarray}
324: %
325: The sign $\pm$ corresponds to the charge conjugation
326: $C = \pm$ of $X$.
327: Heavy-quark spin symmetry implies $c_1 = c_2$
328: up to corrections suppresssed by a factor $\Lambda_{\rm QCD}/m_B$.
329: The interference is constructive if $C = +$
330: and destructive if  $C = -$.
331:  The dependence of the loop amplitude (\ref{eq:ampsum}) on $\Lambda$ 
332: is cancelled by a tree diagram with a $B-XK$ contact interaction 
333: whose coefficient therefore depends linearly on $\Lambda$.  
334: If the $X$ is predominantly a $\bar D D^*$ molecule,
335: there must be some value $\Lambda_1$ of the ultraviolet cutoff 
336: for which the loop amplitude dominates over the tree amplitude. 
337: Squaring the amplitude, summing over spins, and integrating 
338: over phase space, the final result for the decay rate is
339: %
340: \begin{eqnarray}
341: \Gamma[B^+ \to X K^+] =
342: {Z \lambda^{3/2}(m_B,m_X,m_K)  \over 64 \pi^4 m_B^3 m_X^2 \mu a} 
343: 	|c_1 \pm c_2|^2 \Lambda_1^2 ,
344: \label{eq:B-XK}
345: \end{eqnarray}
346: %
347: where $\lambda(x,y,z) = x^4 + y^4 + z^4 - 2 (x^2y^2+y^2z^2+z^2x^2)$. 
348: Due to the factor $1/a$, the decay rate scales like $E_b^{1/2}$
349: as $E_b \to 0$. 
350: 
351: If another hadronic state $H$ is
352: close enough to the $\bar D^0 D^{*0}$ threshold that $X$ 
353: has a nonnegligible probability $Z_H$ of being in the state $H$,
354: the decay can also proceed through a short-distance 2-body decay 
355: $B^+ \to H K^+$.  In this case, there is an additional 
356: term ${\cal A}[B^+ \to H K^+] Z_H^{1/2}$ in (\ref{eq:ampsum}).
357: Its contribution to the decay rate also scales 
358: like $E_b^{1/2}$ as $E_b \to 0$, 
359: because $Z_H$ scales like $E_b^{1/2}$ \cite{Braaten:2003he}.
360: If $C = +$, one possibility for such a state 
361: is the excited P-wave charmonium state $\chi_{c1}(2P)$.
362: Recent coupled-channel calculations of the charmonium spectrum
363: suggest that $\chi_{c1}(2P)$ is likely to be well above
364: the $\bar D^0 D^{*0}$ threshold \cite{Eichten:2004uh}.
365: We will henceforth assume that $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$
366: is the only important component of the wavefunction 
367: and set $Z \approx 1$.
368: 
369: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
370: \begin{figure}
371: \includegraphics[width=7cm]{./Bfig2.eps}
372: \caption{Feynman diagram for 
373: $B^+ \rightarrow \bar D D^{*0} K^+$ via the first pathway. 
374: \label{fig:BDD*K}}
375: \end{figure}
376: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
377: 
378: We can calculate the differential decay rate 
379: for $B^+ \to \bar D^0 D^{*0} K^+$ in the same way.
380: There are again two pathways:
381: the short-distance decay $B^+ \to \bar{D}^0  D^{*0} K^+$ 
382: followed by the long-distance scattering
383: $\bar{D}^0 D^{*0} \to \bar{D}^0 D^{*0}$
384: and $B^+ \to \bar D^{*0} D^0 K^+$ 
385: followed by $\bar D^{*0} D^0 \to \bar{D}^0 D^{*0}$. 
386: The amplitude for the first pathway can be represented 
387: by the Feynman diagram with meson lines shown in Fig.~\ref{fig:BDD*K}.
388: The calculation of the amplitude is similar to that for 
389: $B^+ \to X K^+$ except that 
390: it involves the scattering amplitude (\ref{eq:scat})
391: instead of the coalescence amplitude (\ref{eq:ampx}).
392: In the loop amplitude for $B^+ \to \bar D^0 D^{*0} K^+$,
393: we keep only the term (\ref{eq:A-DD*K}) that is nonzero at the 
394: $\bar D^0 D^{*0}$ threshold.  
395: There must be some value $\Lambda_2$ of the ultraviolet cutoff 
396: for which the loop amplitude dominates over the tree amplitude. 
397: The factor $c_1 \pm c_2$ cancels in the ratio between the
398: amplitudes for $B^+ \to \bar D^0 D^{*0} K^+$ and $B^+ \to X K^+$. 
399: Our final expression for the differential decay rate is
400: %
401: \begin{eqnarray}
402: && \frac{d\Gamma}{d M_{\bar D D^*}}[B^+\to \bar D^0 D^{*0} K^+]
403: \nonumber
404: \\
405: && \hspace{1cm}
406: = \Gamma[B^+\to X K^+] \; {\Lambda_2^2 \over \Lambda_1^2} \;
407: {\mu a^3 k_{\rm cm} \over \pi (1 + a^2 k_{\rm cm}^2)} ,
408: \label{eq:B-DD*K}
409: \end{eqnarray}
410: %
411: where  $M_{\bar D D^*}$ is the $\bar D^0 D^{*0}$ invariant mass
412: and $k_{\rm cm}$ is the relative momentum
413: in the $\bar D^0 D^{*0}$ rest frame:
414: %
415: \begin{eqnarray}
416: k_{\rm cm} = 
417: \lambda^{1/2}(M_{\bar D D^*}, m_D, m_{D^*})/(2 M_{\bar D D^*}).
418: \end{eqnarray}
419: %
420: In (\ref{eq:B-DD*K}), we have neglected 
421: terms suppressed by $k_{\rm cm}^2/m_D^2$.
422: The invariant mass distribution is illustrated in Fig.~\ref{fig:MDD*}
423: for several values of the binding energy $E_b$.
424: The distributions are normalized to 1 at $k_{\rm cm} = m_\pi$.
425: As the binding energy is tuned toward 0, 
426: the peak value scales like $E_b^{-1/2}$ and the position of the peak 
427: in $M_{\bar D D^*} - (m_D + m_{D^*})$ scales like $E_b$.
428: The observation of such an enhancement near the $\bar D^0 D^{*0}$
429: threshold would confirm the interpretation of $X$ as a 
430: $\bar D^0 D^{*0}$/$\bar D^{*0} D^0$ molecule.
431: 
432: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
433: \begin{figure}
434: \includegraphics[width=7cm,angle=270]{./Bfig3.ps}
435: \caption{The $\bar D^0 D^{*0}$ invariant mass distribution for 
436: $B^+ \rightarrow \bar D^0 D^{*0} K^+$ for three different values 
437: of the binding energy of $X$.
438: The distributions are normalized to 1 at $k_{\rm cm} = m_\pi$. 
439: \label{fig:MDD*}}
440: \end{figure}
441: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
442: 
443: The Babar collaboration has recently measured the branching fractions 
444: for $B^+$ to decay into $\bar{D}^0 D^0 K^+$,
445: $\bar{D}^0 D^{*0}K^+$, $\bar D^{*0} D^0 K^+$, and $\bar D^{*0} D^{*0} K^+$
446: to be $(0.19 \pm 0.03)\%$, $(0.47 \pm 0.07)\%$, $(0.18 \pm 0.07)\%$,
447: and $(0.53 \pm 0.11)\%$, respectively \cite{Aubert:2003jq}.
448: We use these measurements to estimate 
449: the branching fraction for $B^+ \rightarrow X K^+$.
450: We make the simplifying assumption that the decay 
451: amplitude factors into currents $\bar c \gamma^\mu (1-\gamma_5) b$ and 
452: $\bar s \gamma^\mu (1-\gamma_5) c$.  Heavy quark symmetry can then 
453: be used to express the 3-body double-charm decay amplitudes 
454: in terms of two functions $G_1(q^2)$ and $G_2(q^2)$,
455: where $q^2$ is the invariant mass of the hadrons produced by the 
456: $\bar s \gamma^\mu (1-\gamma_5) c$ current \cite{Manohar-Wise}.   
457: For example, the 
458: amplitudes for decays into $\bar{D}^0 D^{*0} K^+$ 
459: and $\bar D^{*0} D^0 K^+$ are
460: %
461: \begin{eqnarray}
462: && {\cal A}[B^+ \to \bar D^0 D^{*0} K^+] = 
463: -i G_1 \epsilon^* \cdot (V + v)
464: \nonumber
465: \\
466: &&  \hspace{0.5cm}
467: -i( G_2/m_B) \epsilon^*_\nu 
468: \left[ v_* \cdot k (V + v)^\nu 
469: 	- v_* \cdot (V + v) k^\nu \right.
470: \nonumber
471: \\
472: && \hspace{2.5cm} \left.
473: - i \epsilon^{\nu \mu \alpha \beta} (V + v)_\mu v_{* \alpha} k_\beta \right] ,
474: \label{A-Dbar*D}
475: \\
476: && {\cal A}[B^+ \to \bar D^{*0} D^0 K^+] = 
477: i (G_1 v_\mu + G_2 k_\mu/m_B) \epsilon^*_\nu 
478: \nonumber
479: \\
480: &&  \hspace{0.5cm}
481: \times 
482: \left[ (1+ v_* \cdot V) g^{\mu \nu} 
483: 	- v_*^\mu V^\nu
484: 	-   i \epsilon^{\mu\nu \alpha \beta} v_{* \alpha} V_\beta \right] ,
485: \label{A-DbarD*}
486: \end{eqnarray}
487: %
488: where $k$ is the 4-momentum of the $K^+$ and $V$, $v_*$, and $v$ 
489: are the 4-velocities of the $B^+$, $\bar D^{*0}$ or $D^{*0}$, 
490: and $D^0$ or $\bar D^0$, respectively.
491: As a further simplification, we approximate $G_1$ and $G_2$
492: by constants.  The resulting expressions for the 
493: 3-body double-charm decay rates are
494: %
495: \begin{eqnarray}
496: && \Gamma[B^+ \to \bar D^0 D^0 K^+] = 10^{-3} {\rm MeV} 
497: \nonumber 
498: \\
499: && \times 
500: \left( 178.9\, |G_1|^2 + 51.8\, {\rm Re}(G_1^*G_2) + 4.37\, |G_2|^2 \right),
501: \\
502: && \Gamma[B^+ \to \bar D^0 D^{*0} K^+]  = 10^{-3} {\rm MeV} 
503: \nonumber 
504: \\
505: && \times 
506: \left( 49.6\, |G_1|^2 + 2.61\, {\rm Re}(G_1^*G_2) + 3.49\, |G_2|^2 \right),
507: \\
508: && \Gamma[B^+ \to \bar D^{*0} D^0 K^+]  = 10^{-3} {\rm MeV} 
509: \nonumber 
510: \\
511: && \times 
512: \left( 52.5\, |G_1|^2 + 1.87\, {\rm Re}(G_1^*G_2) + 2.31\, |G_2|^2 \right),
513: \\
514: && \Gamma[B^+ \to \bar D^{*0} D^{*0} K^+]  = 10^{-3} {\rm MeV} 
515: \nonumber 
516: \\
517: && \times 
518: \left( 221.5\, |G_1|^2 + 74.8\, {\rm Re}(G_1^*G_2) + 11.58\, |G_2|^2 \right).
519: \end{eqnarray}
520: %
521: We obtain a good fit to the Babar branching fractions
522: with $G_1 = 3.2\times 10^{-6}$ and $G_2 = (-14.6 + 9.6i) \times 10^{-6}$.
523: In the corner of phase space where the 4-velocities of $\bar D^0$ 
524: and $D^{*0}$ are equal, the amplitudes
525: (\ref{A-Dbar*D}) and (\ref{A-DbarD*}) reduce to the form 
526: on the right side of (\ref{eq:A-DD*K}) with coefficients
527: $c_1 = c_2 = -i G_1/m_B + i G_2 (m_B+m_D+m_{D^*})/m_B^2$.
528: If $X$ has charge conjugation 
529: $C = +$, the estimate (\ref{eq:B-XK}) reduces to
530: %
531: \begin{eqnarray}
532: {\cal B}[B^+ \rightarrow X K^+] \approx 
533: \left( 2.6 \times 10^{-5} \right) {\Lambda_1^2 \over m_\pi^2}
534: \left( {E_b \over 0.4 \, {\rm MeV}} \right)^{1/2}.
535: \label{B-DDK:est}
536: \end{eqnarray}
537: %
538: If $C = -$, the branching fraction would be significantly smaller 
539: because of destructive interference between $c_1$ and $c_2$.  
540: We could get a more reliable result for the numerical prefactor 
541: in (\ref{B-DDK:est}) by relaxing the factorization assumption 
542: and carrying out a Dalitz plot analysis of the 3-body decays.
543: Since the result depends quadratically on the ultraviolet cutoff
544: $\Lambda_1$, the best we can do is obtain an order-of-magnitude
545: estimate of the branching fraction by setting 
546: $\Lambda_1 \approx m_\pi$. 
547: 
548: The Belle collaboration measured the product of the branching fractions
549: ${\cal B}[B^+ \to X K^+]$ and ${\cal B}[X \to J/\psi \; \pi^+ \pi^-]$
550: to be $(1.3 \pm 0.3)\times 10^{-5}$ \cite{Choi:2003ue}.
551: Our estimate of ${\cal B}[B^+ \rightarrow X K^+]$ 
552: is compatible with this result
553: if $J/\psi \; \pi^+ \pi^-$ is one of the major decay modes of $X$.
554: The experimental upper bound on the width of $X(3872)$ is
555: $\Gamma_X < 2300$ keV.
556: The sum of the widths for decay into $\bar D^0 D^0 \pi^0$ 
557: and $\bar D^0 D^0 \gamma$ approaches $\Gamma[D^{*0}] \approx 50$ 
558: keV in the limit $E_b \to 0$ \cite{Voloshin:2003nt}.
559: The remaining partial widths scale as $E_b^{1/2}$.
560: Using a coupled-channel calculation in a model in which 
561: $X$ mixes with $J/\psi \;\rho$, the decay rate for $J/\psi \; \pi^+ \pi^-$
562: has been estimated to be 1290 keV for $E_b = 0.7$ MeV \cite{Swanson:2003tb}.
563: Thus it is at least plausible that $J/\psi \; \pi^+ \pi^-$ 
564: is one of the major decay modes.
565: Other possible decay channels are $\eta_c \pi \pi$,
566: radiative transitions to charmonium states,
567: and $c \bar c$ annihilation decays.
568: 
569: We have calculated the decay rate for
570: $B^+ \to X K^+$ and the differential
571: decay rate for $B^+ \to \bar{D}^0 D^{*0}K^+$
572: near the $\bar{D}^0 D^{*0}$ threshold
573: under the assumption that $X(3872)$ is a loosely-bound S-wave
574: $\bar{D}^0 D^{*0}$/$D^{0} \bar{D}^{*0}$ molecule
575: and that its production rate is dominated by the 
576: coalescense of charm mesons.
577: Observation of a sharp peak in the $\bar D^0 D^{*0}$ 
578: invariant mass distribution near threshold in the decay 
579: $B^+ \to \bar{D}^0 D^{*0}K^+$ would confirm the 
580: interpretation of $X$ as a $\bar D^0 D^{*0}$ molecule.
581: Our order-of-magnitude estimate of the branching fraction 
582: for $B^+ \to X K^+$ is compatible with observations 
583: if $X(3872)$ has quantum numbers $J^{PC} = 1^{++}$ and if 
584: $J/\psi \; \pi^+ \pi^-$ is one of its major decay modes.
585: 
586: This research was supported in part by the Department of Energy under
587: grant DE-FG02-91-ER4069.  
588: 
589: 
590: 
591: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
592: % Create the reference section using BibTeX:
593: %----------------------------------------------------------------------
594: 
595: \begin{thebibliography}{}
596: 
597: %\cite{Choi:2003ue}
598: \bibitem{Choi:2003ue}
599: S.~K.~Choi {\it et al.}  [Belle Collaboration],
600: %``Observation of a new narrow charmonium state in exclusive B+- $\to$
601: 	%K+- pi+
602: %pi- J/psi decays,''
603: Phys.\ Rev.\ Lett.\  {\bf 91}, 262001 (2003).
604: % [arXiv:hep-ex/0309032].
605: 
606: %\cite{Acosta:2003zx}
607: \bibitem{Acosta:2003zx}
608: D.~Acosta {\it et al.}  [CDF II Collaboration],
609:  %``Observation of the narrow state X(3872) $\to$ J/psi pi+ pi- in
610: 	% anti-p p collisions at s**(1/2) = 1.96-TeV,''
611: arXiv:hep-ex/0312021.
612: 
613: %\cite{Tornqvist:2004qy}
614: \bibitem{Tornqvist:2004qy}
615: N.~A.~Tornqvist,
616: %``Isospin breaking of the narrow charmonium state of Belle at 3872-MeV as a
617: %deuson,''
618: Phys.\ Lett.\ B {\bf 590}, 209 (2004);
619: %[arXiv:hep-ph/0402237].
620: %\cite{Tornqvist:2003na}
621: %\bibitem{Tornqvist:2003na}
622: %N.A.~Tornqvist,
623: %%``Comment on the narrow charmonium state of Belle 
624: %%	at 3871.8-MeV as a deuson,''
625: arXiv:hep-ph/0308277.
626: 
627: %\cite{Close:2003sg}
628: \bibitem{Close:2003sg}
629: F.~E.~Close and P.~R.~Page,
630: %``The D*0 anti-D0 threshold resonance,''
631: Phys.\ Lett.\ B {\bf 578}, 119 (2004).
632: % [arXiv:hep-ph/0309253].
633: 
634: %\cite{Pakvasa:2003ea}
635: \bibitem{Pakvasa:2003ea}
636: S.~Pakvasa and M.~Suzuki,
637: %``On the hidden charm state at 3872-MeV,''
638: Phys.\ Lett.\ B {\bf 579}, 67 (2004).
639: % [arXiv:hep-ph/0309294].
640: 
641: %\cite{Voloshin:2003nt}
642: \bibitem{Voloshin:2003nt}
643: M.~B.~Voloshin,
644:  %``Interference and binding effects in decays of possible molecular
645: 	%component
646: %of X(3872),''
647: Phys.\ Lett.\ B {\bf 579}, 316 (2004).
648: % [arXiv:hep-ph/0309307].
649: 
650: %\cite{Yuan:2003yz}
651: \bibitem{Yuan:2003yz}
652: C.~Z.~Yuan, X.~H.~Mo and P.~Wang,
653: %``The upper limit of the e+ e- partial width of X(3872),''
654: Phys.\ Lett.\ B {\bf 579}, 74 (2004).
655: %[arXiv:hep-ph/0310261].
656: 
657: %\cite{Wong:2003xk}
658: \bibitem{Wong:2003xk}
659: C.-Y.~Wong, 
660: %``Molecular States of Heavy Quark Mesons,''
661: Phys.\ Rev.\ C {\bf 69}, 055202 (2004).
662: %[arXiv:hep-ph/0311088].
663:     
664: %\cite{Braaten:2003he}
665: \bibitem{Braaten:2003he}
666: E.~Braaten and M.~Kusunoki,
667: %``Low-energy universality and the new charmonium resonance at 3870-MeV,''
668: Phys.\ Rev.\ D {\bf 69}, 074005 (2004).
669: % [arXiv:hep-ph/0311147].
670: 
671: %\cite{Barnes:2003vb}
672: \bibitem{Barnes:2003vb}
673: T.~Barnes and S.~Godfrey,
674: %``Charmonium options for the X(3872),''
675: Phys.\ Rev.\ D {\bf 69}, 054008 (2004).
676: % [arXiv:hep-ph/0311162].
677: 
678: %\cite{Swanson:2003tb}
679: \bibitem{Swanson:2003tb}
680: E.~S.~Swanson,
681: %``Short range structure in the X(3872),''
682: Phys.\ Lett.\ B {\bf 588}, 189 (2004).
683: %[arXiv:hep-ph/0311229].
684: 
685: %\cite{Eichten:2004uh}
686: \bibitem{Eichten:2004uh}
687: E.~J.~Eichten, K.~Lane and C.~Quigg,
688: %``Charmonium levels near threshold and the narrow state $X(3872) \to
689: %\pi^{+}\pi^{-}\jpsi$,''
690: Phys.\ Rev.\ D {\bf 69}, 094019 (2004).
691: %[arXiv:hep-ph/0401210].
692: 
693: %\cite{Quigg:2004nv}
694: \bibitem{Quigg:2004nv}
695: C.~Quigg,
696: %``Quarkonium: New developments,''
697: arXiv:hep-ph/0403187.
698: 
699: %\cite{Voloshin:ap}
700: \bibitem{Voloshin:ap}
701: M.~B.~Voloshin and L.~B.~Okun,
702: %``Hadron Molecules And Charmonium Atom,''
703: JETP Lett.\  {\bf 23}, 333 (1976).
704: %[Pisma Zh.\ Eksp.\ Teor.\ Fiz.\  {\bf 23}, 369 (1976)].
705: 
706: %\cite{Nussinov:1976fg}
707: \bibitem{Nussinov:1976fg}
708: S.~Nussinov and D.~P.~Sidhu,
709: %``Loosely Bound States Near The Charm Threshold: Charm Molecules,''
710: Nuovo Cim.\ A {\bf 44}, 230 (1978).
711: 
712: %\cite{Tornqvist:1991ks}
713: \bibitem{Tornqvist:1991ks}
714: N.~A.~Tornqvist,
715: %``Possible large deuteron - like meson meson states bound by pions,''
716: Phys.\ Rev.\ Lett.\  {\bf 67}, 556 (1991).
717: 
718: %\cite{Braaten:2004rn}
719: \bibitem{Braaten:2004rn}
720: E.~Braaten and M.~Kusunoki,
721: %``Production of the X(3870) at the Upsilon(4S) by the coalescence of charm
722: %mesons from B decays,''
723: arXiv:hep-ph/0402177.
724: 
725: %\cite{Aubert:2003jq}
726: \bibitem{Aubert:2003jq}
727: B.~Aubert {\it et al.}  [BABAR Collaboration],
728: %``Measurement of the branching fractions for the exclusive decays of B0  and B+
729: %to anti-D(*) D(*) K,''
730: Phys.\ Rev.\ D {\bf 68}, 092001 (2003).
731: %[arXiv:hep-ex/0305003].
732:   
733: \bibitem{Manohar-Wise}
734: A.V.~Manohar and M.B.~Wise, 
735: {\bf Heavy Quark Physics}
736: (Cambridge University Press, 2000).
737: 
738: 
739: \end{thebibliography}
740: 
741: \end{document}
742: