1: %% ****** Start of file template.aps ****** %

2: %%

3: %%

4: %%   This file is part of the APS files in the REVTeX 4 distribution.

5: %%   Version 4.0 of REVTeX, August 2001

6: %%

7: %%

8: %%   Copyright (c) 2001 The American Physical Society.

9: %%

10: %%   See the REVTeX 4 README file for restrictions and more information.

11: %%

12: %

13: % This is a template for producing manuscripts for use with REVTEX 4.0

14: % Copy this file to another name and then work on that file.

15: % That way, you always have this original template file to use.

16: %

17: % Group addresses by affiliation; use superscriptaddress for long

18: % author lists, or if there are many overlapping affiliations.

19: % For Phys. Rev. appearance, change preprint to twocolumn.

20: % Choose pra, prb, prc, prd, pre, prl, prstab, or rmp for journal

21: %  Add 'draft' option to mark overfull boxes with black boxes

22: %  Add 'showpacs' option to make PACS codes appear

23: %  Add 'showkeys' option to make keywords appear

24: %%\documentclass[10pt,aps,prd,groupedaddress,showpacs]{revtex4}%

25: %%               ,draft,showpacs,showkeys]{revtex4}

26: 

27: %\documentclass[preprint,tightenlines,aps,prd,groupedaddress,showpacs]

28: %              {revtex4}%

29: \documentclass[preprint,aps,prd,groupedaddress,showpacs,nofootinbib]{revtex4}%}

30: \usepackage{graphicx}

31: \usepackage{amsmath}

32: \usepackage{bm}

33: \usepackage{epsfig}

34: %\documentclass[aps,prl,preprint,superscriptaddress]{revtex4}

35: %\documentclass[aps,prl,twocolumn,groupedaddress]{revtex4}

36: % You should use BibTeX and apsrev.bst for references

37: % Choosing a journal automatically selects the correct APS

38: % BibTeX style file (bst file), so only uncomment the line

39: % below if necessary.

40: %%%%%%\bibliographystyle{apsrev}

41: 

42: \setlength\arraycolsep{2pt}

43: 

44: \begin{document}

45: 

46: % Use the \preprint command to place your local institutional report

47: % number in the upper righthand corner of the title page in preprint mode.

48: % Multiple \preprint commands are allowed.

49: % Use the 'preprintnumbers' class option to override journal defaults

50: % to display numbers if necessary

51: \preprint{ANL-HEP-PR-06-57}

52: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

53: %Title of paper

54: \title{\mbox{}\\[10pt]

55: Exclusive Two-Vector-Meson Production

56: from $\bm{e^+ e^-}$ Annihilation}

57: 

58: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

59: % repeat the \author .. \affiliation  etc. as needed

60: % \email, \thanks, \homepage, \altaffiliation all apply to the current

61: % author. Explanatory text should go in the []'s, actual e-mail

62: % address or url should go in the {}'s for \email and \homepage.

63: % Please use the appropriate macro foreach each type of information

64: % \affiliation command applies to all authors since the last

65: % \affiliation command. The \affiliation command should follow the

66: % other information

67: % \affiliation can be followed by \email, \homepage, \thanks as well.

68: % \altaffiliation{}

69: 

70: \author{Geoffrey T. Bodwin}

71: %\email[]{Your e-mail address}

72: %\homepage[]{Your web page}

73: %\thanks{}

74: %\altaffiliation{}

75: \affiliation{

76: High Energy Physics Division,

77: Argonne National Laboratory,

78: 9700 South Cass Avenue, Argonne, Illinois 60439}

79: 

80: \author{Eric Braaten}

81: %\email[]{Your e-mail address}

82: %\homepage[]{Your web page}

83: %\thanks{}

84: %\altaffiliation{}

85: \affiliation{

86: Physics Department, Ohio State University, Columbus, Ohio 43210}

87: 

88: \author{Jungil Lee and Chaehyun Yu}

89: %\email[]{Your e-mail address}

90: %\homepage[]{Your web page}

91: %\thanks{}

92: %\altaffiliation{}

93: \affiliation{

94: Department of Physics, Korea University,

95: Seoul 136-701, Korea

96: }

97: 

98: 

99: %Collaboration name if desired (requires use of superscriptaddress

100: %option in \documentclass). \noaffiliation is required (may also be

101: %used with the \author command).

102: %\collaboration can be followed by \email, \homepage, \thanks as well.

103: %\collaboration{}

104: %\noaffiliation

105: 

106: \date{\today}

107: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

108: \begin{abstract}

109: % insert abstract here

110: The exclusive production of pairs of vector mesons with $J^{PC}  =

111: 1^{--}$ in $e^+ e^-$ collisions can proceed through $e^+ e^-$

112: annihilation into two virtual photons. At energies much greater than

113: the meson masses, the cross section is dominated by the independent

114: fragmentation of the virtual photons into the vector mesons. The

115: fragmentation approximation is used to calculate the cross sections and

116: angular distributions for pairs of vector mesons that can be produced at

117: the $B$ factories. The predicted cross sections for $\rho^0 + \rho^0$ and

118: $\rho^0 + \phi$ production agree with recent measurements by the BaBar

119: Collaboration. For the production of two charmonium vector mesons, the

120: nonfragmentation corrections to the cross sections are calculated by

121: using the NRQCD factorization formalism. The predicted cross sections

122: for $J/\psi + J/\psi$ and $J/\psi + \psi(2S)$ production are compatible

123: with upper limits set by the Belle Collaboration.

124: \end{abstract}

125: 

126: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

127: % insert suggested PACS numbers in braces on next line

128: \pacs{13.60.Le,12.40.Vv,13.66.Bc,14.40.Gx}

129: % 12.38.-t Quantum chromodynamics

130: % 12.38.Bx Perturbative calculations

131: % 12.40.Vv Vector-meson dominance

132: % 13.60.Le Meson production

133: % 13.66.Bc Hadron production in e-e+ interactions

134: % 14.40.Gx Mesons with S=C=B=0, mass>2.5 GeV (including quarkonia)

135: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

136: % insert suggested keywords - APS authors don't need to do this

137: %\keywords{}

138: 

139: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

140: %\maketitle must follow title, authors, abstract, \pacs, and \keywords

141: \maketitle

142: 

143: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

144: % body of paper here - Use proper section commands

145: % References should be done using the \cite, \ref, and \label commands

146: 

147: 

148: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

149: \section{Introduction

150:  \label{intro}}

151: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

152: % Put \label in argument of \section for cross-referencing

153: %\section{\label{}}

154: 

155: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

156: % Surround figure environment with turnpage environment for landscape

157: % figure

158: % \begin{turnpage}

159: %\begin{figure}

160: %\includegraphics[height=12cm,angle=-90]{double_fig1.ps}

161: %\caption{\label{fig1}%

162: %QCD diagrams that can contribute to the color-singlet process

163: %$\gamma^* \to c \bar c_1 + c \bar c_1$.  }

164: %\end{figure}

165: % \end{turnpage}

166: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

167: 

168: The production of hadrons in $e^+ e^-$ collisions 

169: at energies well below the mass of the $Z^0$ proceeds 

170: predominantly through the annihilation of $e^+ e^-$ into a 

171: virtual photon.  This process can only produce final states 

172: with charge conjugation quantum number $C=-1$.

173: Final states with $C=+1$   can be produced by the annihilation 

174: of $e^+ e^-$ into two virtual photons, but the cross sections 

175: for these processes are suppressed by a factor of

176: $\alpha^2$, where $\alpha \simeq 1/137$ is the QED fine 

177: structure constant.  Final states with $C=+1$   can also be produced 

178: by the annihilation of $e^+ e^-$ into a virtual $Z^0$, 

179: but the cross sections for these processes are suppressed by 

180: a factor of $(s/M_Z^2)^2$ if the $e^+ e^-$ center-of-mass energy 

181: $\sqrt{s}$ is small compared to $M_Z$.  Thus, unless there are 

182: enhancement factors to compensate for the factors   of $\alpha^2$

183: or $(s/M_Z^2)^2$, the production   cross sections at $B$-factory 

184: energies   for $C=+1$   states 

185: are orders of magnitude smaller than those for $C=-1$   states.

186: 

187: In $e^+ e^-$ collisions with large center-of-mass energy $\sqrt{s}$,

188: factors of  $\sqrt{s}/m$, where $m$ is a hadronic mass scale,

189: can compensate for the suppression factor of $\alpha^2$. 

190: An example of such

191: compensation can be seen in the exclusive double-charmonium

192: production process $e^+ e^- \to J/\psi + J/\psi$

193: \cite{Bodwin:2002fk,Bodwin:2002kk}. At the beam energy of the $B$

194: factories, $\sqrt{s}/2=5.29$ GeV, the 

195: leading-order prediction for the production cross section for

196: the $C=+1$ final state $J/\psi + J/\psi$ has roughly the same order of

197: magnitude as that for the corresponding $C=-1$ final state $J/\psi +

198: \eta_c$~\cite{Braaten:2002fi}. The production cross section for

199: $J/\psi + \eta_c$ scales as $\alpha^2 \alpha_s^2 m_c^6/s^4$, where

200: $m_c$ is the charm-quark mass. In contrast, the production cross

201: section for $J/\psi + J/\psi$ scales as $\alpha^4/s$ because there is a

202: photon-fragmentation contribution that corresponds to the annihilation

203: process $e^+ e^- \to \gamma^* + \gamma^*$, followed by independent

204: fragmentation processes $\gamma^* \to J/\psi$. The enhancement

205: factor of $(s/m_c^2)^3$ in the production cross section for $J/\psi +

206: J/\psi$ relative to that for $J/\psi + \eta_c$ partially compensates 

207: for the suppression factor of $(\alpha/\alpha_s)^2$.  

208: 

209: A similar enhancement mechanism is present   in the exclusive

210: production of a pair of light vector mesons $V_1 + V_2$,   both of

211: which have quantum numbers $J^{PC} = 1^{--}$.  The production process

212: proceeds   predominantly through the annihilation process $e^+ e^- \to

213: \gamma^* + \gamma^*$,   followed by the   independent fragmentation

214: processes $\gamma^* \to V_i$. We can compare the production cross

215: section for $V_1 + V_2$ to that for $V + P$, where $P$ is a light

216: pseudoscalar meson. In the ratio of the former cross section to the

217: latter, there is an enhancement factor of $(s/\Lambda_{\rm QCD}^2)^3$

218: that compensates for the suppression factor of $(\alpha/\alpha_s)^2$.  

219: In this paper, we calculate the cross sections and angular distributions

220: for the exclusive production of pairs of vector mesons with $J^{PC}  =

221: 1^{--}$.

222: 

223: The remainder of this paper is organized as follows. In Sec.~\ref{ampl},

224: we discuss the various contributions to the amplitudes for the

225: production of two vector mesons and estimate their relative sizes. In

226: Sec.~\ref{sec:differential}, we give expressions for the differential

227: cross sections for the production of two vector mesons in the

228: fragmentation approximation and give numerical results for the cross

229: sections. Sec.~\ref{sec:nonfrag} contains a computation of the

230: nonfragmentation corrections to the cross sections for the production of

231: two charmonium vector mesons. In Sec.~\ref{sec:comparison}, we compare

232: our results with experimental measurements and with previous theoretical

233: calculations. The Appendix contains the expressions for the

234: nonfragmentation corrections to the differential cross sections for two

235: charmonium vector mesons.

236: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

237: \section{Production amplitudes

238:  \label{ampl}}

239: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

240: 

241: The lowest-order   QED diagrams that correspond to the creation of the

242: quarks in the two vector mesons $V_1$ and $V_2$ are shown in

243: Figs.~\ref{fig1} and \ref{fig2}. If the two vector mesons have different

244: quark contents $q \bar q$ and $q' \bar q'$, then only the diagrams in

245: Fig.~\ref{fig1} are present.  Once the quarks are created, the formation

246: of the vector mesons involves the exchange of arbitrarily many gluons

247: between the quark lines.  Gluons can be exchanged between the quark and

248: antiquark that form a given meson or between quarks or antiquarks that

249: form different mesons.  However, factorization theorems and the

250: scaling of various contributions yield a simplification of this picture

251: in the limit $E_{\rm beam}\to \infty$. In what follows, we take the beam

252: energy $E_{\rm beam}= \sqrt{s}/2$ to be much greater than the masses

253: $m_{V_i}$ of the vector mesons, which implies that $E_{\rm beam}$ is

254: also much greater than $\Lambda_{\rm QCD}$.

255: 

256: Let us first consider the exchange of gluons only between the $q$ and

257: $\bar q$ within a given meson. Then, the diagrams in

258: Fig.~\ref{fig2} are suppressed in comparison to those in

259: Fig.~\ref{fig1}. In the diagrams of Fig.~\ref{fig1}, the virtual-photon

260: propagators are $1/m_{V_i}^2$, and the two hadronic electromagnetic

261: currents each give a factor of order $\Lambda_{\rm QCD}$ for light

262: mesons and $m_c$ for charmonium mesons.  In the diagrams of

263: Fig.~\ref{fig2}, the virtual-photon propagators are of order $1/E_{\rm

264: beam}^2$, and the two hadronic electromagnetic currents each give factors

265: of order $E_{\rm beam}$. This leads to the following suppression factors

266: for the diagrams of Fig.~\ref{fig2} relative to those of

267: Fig.~\ref{fig1}: $\Lambda_{\rm QCD}^2/ E_{\rm beam}^2$ for two light

268: mesons and $m_c^2/E_{\rm beam}^2$ for two charmonium mesons. We have

269: taken the light-meson masses to be of order $\Lambda_{\rm QCD}$ and the

270: charmonium masses to be of order $m_c$. (The diagrams of Fig.~\ref{fig2}

271: do not contribute in the case of a light meson and a charmonium meson.)

272: 

273: Next, let us consider the exchange of gluons between quarks or

274: antiquarks in different mesons, with some of those gluons soft or, in

275: the case of a light meson, collinear. In the case of two light mesons,

276: one can use standard methods for proving factorization theorems

277: \cite{Collins:1989gx} to show that the soft contributions cancel up to

278: corrections of relative order $\Lambda_{\rm QCD}^4/E_{\rm beam}^4$ and

279: the collinear contributions cancel up to corrections of relative order

280: $\Lambda_{\rm QCD}^2/E_{\rm beam}^2$.\footnote{The results that we quote

281: here apply to the case in which one sums over all polarizations of the

282: final-state mesons. If one observes the polarization of one meson, then

283: the soft contributions cancel up to terms of relative order

284: $\Lambda_{\rm QCD}^3/E_{\rm beam}^3$, and the collinear contributions

285: cancel up to terms of relative order $\Lambda_{\rm QCD}/E_{\rm beam}$.

286: If one observes the polarization of both mesons, then the soft

287: contributions cancel up to corrections of relative order $\Lambda_{\rm

288: QCD}^2/E_{\rm beam}^2$ and the collinear contributions cancel up to

289: corrections of relative order $\Lambda_{\rm QCD}/E_{\rm beam}$. Similar

290: reductions of the suppression factors for soft and collinear

291: contributions occur in the cases in which one or more charmonium mesons

292: are produced.} Factorization theorems are not well established for the

293: production of heavy-quarkonium mesons. However, on the basis of existing

294: factorization technology, it seems plausible, in the case of one light

295: meson and one charmonium meson, that soft contributions cancel up to

296: terms of relative order $\Lambda_{\rm QCD}^2(m_cv)^2/E_{\rm beam}^4$ and

297: that collinear contributions cancel up to terms of relative order

298: $\Lambda_{\rm QCD}^2/E_{\rm beam}^2$. Here $v$ is the typical velocity

299: of the charm quark or antiquark in the charmonium rest frame.

300: ($v^2\approx 0.3$.) In the case of two charmonium mesons, there are no

301: collinear singularities. It is plausible that soft contributions cancel

302: up to terms of relative order $(m_cv)^4/E_{\rm beam}^4$ in this case.

303: 

304: Finally, let us consider the exchange of hard gluons between quarks or

305: antiquarks in different mesons. (Hard gluons have energies and momenta

306: of order $E_{\rm beam}$.) In the case of the diagrams of

307: Fig.~\ref{fig2}, exchanges of hard gluons are suppressed as

308: $\alpha_s(E_{\rm beam})$. In the case of the diagrams of Fig.~\ref{fig1}, 

309: it is necessary to exchange at least two gluons between the mesons 

310: in order to keep the $q \bar q$ pair in a color-singlet state. Hence, 

311: such exchanges are suppressed as $\alpha_s^2(E_{\rm beam})$. They are 

312: also suppressed by a kinematic factor that arises as follows. In the

313: diagrams of Fig.~\ref{fig1} without hard-gluon exchange, the

314: virtual-photon propagators are $1/m_{V_i}^2$ and the two hadronic

315: electromagnetic currents each give a factor of order $\Lambda_{\rm QCD}$

316: for light mesons and $m_c$ for charmonium mesons. In the diagrams of

317: Fig.~\ref{fig1} with hard-gluon exchanges, the virtual-photon

318: propagators are of order $1/E_{\rm beam}^2$, and the two hadronic

319: electromagnetic currents each give factors of order $E_{\rm beam}$. This

320: leads to the following additional suppression factors for hard-gluon

321: exchanges in the diagrams of Fig.~\ref{fig1}: $\Lambda_{\rm QCD}^2/

322: E_{\rm beam}^2$ for two light mesons, $m_c\Lambda_{\rm QCD}/ E_{\rm

323: beam}^2$ for a light meson and a charmonium meson, and $m_c^2/E_{\rm

324: beam}^2$ for two charmonium mesons. Again, we have taken the light-meson

325: masses to be of order $\Lambda_{\rm QCD}$ and the charmonium masses to

326: be of order $m_c$. 

327: 

328: 

329: 

330: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

331: % Fig 1

332: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

333: \begin{figure}

334: \begin{tabular}{lcr}

335: \epsfig{file=fig1a.ps,width=7cm}&\qquad\qquad\qquad&

336: \epsfig{file=fig1b.ps,width=7cm}

337: \end{tabular}

338: \vspace*{8pt}

339: \caption{\label{fig1}%

340: QED fragmentation diagrams that contribute to exclusive 

341: two-vector-meson production in $e^+ e^-$ collisions if the two 

342: vector mesons have quark contents $q \bar q$ and $q' \bar q'$. 

343: }

344: \end{figure}

345: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

346: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

347: 

348: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

349: % Fig 2

350: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

351: \begin{figure}

352: \begin{tabular}{lcr}

353: \epsfig{file=fig2a.ps,width=7cm}&\qquad\qquad\qquad&

354: \epsfig{file=fig2b.ps,width=7cm}

355: \end{tabular}

356: \vspace*{8pt}

357: \caption{\label{fig2}%

358: QED nonfragmentation diagrams that contribute to exclusive 

359: two-vector-meson production in $e^+ e^-$ collisions if the two 

360: vector mesons have the same quark content $q \bar q$. }

361: \end{figure}

362: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

363: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

364: 

365: Let us now summarize the results of the preceding analyses.  We conclude

366: that we need only consider the fragmentation diagrams in

367: Fig.~\ref{fig1}, up to terms of relative order $\Lambda_{\rm QCD}^2/

368: E_{\rm beam}^2$ for two light mesons and $m_c^2/E_{\rm beam}^2$ for two

369: charmonium mesons. (The diagrams of Fig.~\ref{fig2} do not contribute in

370: the case of a light meson and a charmonium meson.) We also conclude that

371: we need only consider the QCD corrections to the diagrams of

372: Fig.~\ref{fig1} that correspond to the exchange of gluons between the

373: $q$ and $\bar q$ in each meson, up to terms of relative order

374: $\Lambda_{\rm QCD}^2/E_{\rm beam}^2$ for two light mesons, $\Lambda_{\rm

375: QCD}^2/E_{\rm beam}^2$ or $\alpha_s^2(E_{\rm beam}) \, (m_c\Lambda_{\rm

376: QCD}/ E_{\rm beam}^2)$ for a light meson and a charmonium meson, and

377: $(m_cv)^4/E_{\rm beam}^4$ or $\alpha_s^2(E_{\rm beam}) \, (m_c^2/E_{\rm

378: beam}^2)$ for two charmonium mesons.\footnote{

379: There are several recent papers in which the rate for the process

380: $\gamma^* \gamma^* \to \rho^0 \rho^0$ is calculated in the high-energy

381: and diffractive limit $s \gg -t$, where $s$ and $t$ are the

382: Mandelstam variables 

383: \cite{Pire:2005ic,Enberg:2005eq,Ivanov:2005gn,Pire:2006ik}. 

384: In this limit, the dominant

385: contributions to this process involve the exchange of gluons between the

386: quarks and antiquarks that form different mesons.  There are analogous

387: contributions to the process $e^+ e^- \to \rho^0 \rho^0$, since it also

388: proceeds through $\gamma^* \gamma^* \to \rho^0 \rho^0$. However, as we

389: have discussed above, these contributions are suppressed, and the

390: dominant mechanism in $e^+ e^- \to \rho^0 \rho^0$ is independent photon

391: fragmentation. 

392: Independent photon fragmentation is not considered in

393: Refs.~\cite{Pire:2005ic,Enberg:2005eq,Ivanov:2005gn,Pire:2006ik}

394: because it corresponds in the process

395: $\gamma^* \gamma^* \to \rho^0 \rho^0$ to a disconnected diagram.

396: }

397: 

398: The QCD corrections that involve the exchange of gluons between

399: the $q$ and $\bar q$ in each meson are precisely those that correspond

400: to the formation of a vector meson $V$ from a $q \bar q$ pair that has

401: been created at a point. They can be expressed in terms of a QCD

402: matrix element of the electromagnetic current:

403: %-------------------

404: \begin{equation}

405: %-------------------

406: J^\mu(x) = e_u\bar u(x) \gamma^\mu u(x) 

407: + e_d\bar d(x) \gamma^\mu d(x) 

408: + e_s\bar s(x) \gamma^\mu s(x) 

409: + e_c\bar c(x) \gamma^\mu c(x),

410: \label{current}

411: %-------------------

412: \end{equation}

413: %-------------------

414: where $e_u = e_c = +\frac{2}{3}$ and $e_d = e_s = -\frac{1}{3}$.

415: The matrix element can be used to define a vector-meson-photon

416: coupling constant $g_{V \gamma}$:

417: %-------------------

418: \begin{equation}

419: %-------------------

420: \langle V(\lambda) | J^\mu(x=0) | 0 \rangle = 

421: g_{V \gamma} \, \epsilon^\mu(\lambda)^*,

422: \label{GV-def}

423: %-------------------

424: \end{equation}

425: %-------------------

426: where $\lambda$ is the helicity of the vector meson.

427: The coupling constant $g_{V\gamma}$ can be determined from the 

428: electronic width of the vector meson:

429: %-------------------

430: \begin{equation}

431: \Gamma[V \to e^+ e^-] = \frac{4\pi\alpha^2 g^2_{V\gamma}}{3m_V^3}.

432: \label{gamma-v-ee}

433: \end{equation}

434: %-------------------

435: Using the measured electronic widths that are given in Ref.~\cite{PDG2006},

436: we obtain the values of $g_{V\gamma}$ that are shown in Table~\ref{constants}.

437: The stated uncertainties arise 

438: from the electronic widths. The uncertainties in the masses are at 

439: most $0.06\%$ and can be neglected.  

440: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

441: \begin{table}[t]

442: \caption{\label{constants}% 

443: Masses, electronic decay widths, and coupling constants $g_{V\gamma}$

444: for vector mesons $V$. All of the data except for those of $\rho^0$ are

445: taken from Ref.~\cite{PDG2006}. The data for $\rho^0$ are taken from

446: Ref.~\cite{Akhmetshin:2003zn} in order to maintain consistency between

447: the narrow-width approximation and the nonzero-width parametrization. The

448: uncertainties shown for $g_{V\gamma}$ are those that arise from the

449: uncertainties in the electronic widths of the vector mesons.}

450: \begin{ruledtabular}

451: \begin{tabular}{l|ccccc}

452: %-------------------------------------------------------------------------

453: $V$ & Mass~(MeV) & Width~(keV) & $g_{V\gamma}$~(GeV$^2$)\\

454: %-------------------------------------------------------------------------

455: \hline

456: %-------------------------------------------------------------------------

457: $\rho^0$&

458: $ 775.65  \pm 0.64 \pm 0.50  $& $7.06\pm 0.11 \pm 0.05$& 

459: $0.122 \pm 0.001$

460: \\

461: $\omega$&

462: $ 782.65 \pm 0.12 $& $0.60\pm 0.02$& $0.036 \pm 0.001$

463: \\

464: $\phi$&

465: $1019.46 \pm 0.019$& $1.27\pm 0.04$& $0.078 \pm 0.001$

466: \\

467: $J/\psi$&

468: $3096.916\pm 0.011$& $5.55\pm 0.14$& $0.860 \pm 0.011$

469: \\

470: $\psi(2S)$&

471: $3686.093\pm 0.034$& $2.48\pm 0.06$& $0.746 \pm 0.009$

472: %-------------------------------------------------------------------------

473: \end{tabular}

474: \end{ruledtabular}

475: \end{table}

476: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

477: 

478: 

479: 

480: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

481: \section{Cross sections in the fragmentation 

482: approximation}\label{sec:differential}

483: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

484: 

485: We express our results for the cross section for $e^+e^-\to V_1+V_2$ 

486: in terms of dimensionless 

487: variables $r_{V_1}$ and $r_{V_2}$, which are defined by

488: %--------------

489: \begin{eqnarray}

490: %--------------

491: r_V &=& \frac{ m_V }{ \sqrt{s}}.

492: %--------------

493: \label{r-def}

494: \end{eqnarray}

495: %--------------

496: If we take $\sqrt{s}$ to be $10.58$ GeV, the center-of-mass energy

497: of the $B$ factories, then $r_V^2$ is small for charmonium

498: ($r_{J/\psi}^2= 0.086$) and very small for light vector mesons

499: ($r_\rho^2 = 0.0054$). The differential cross sections

500: $d\sigma/d\cos\theta$ for each of the helicity states of $V_1$ and $V_2$

501: are

502: %--------------

503: \begin{eqnarray}

504: %--------------

505: \frac{d\sigma}{dx}[V_1(\lambda_1) + V_2(\lambda_2)] &=&

506: \frac{16\pi^3 \alpha^4 g_{V_1\gamma}^2 g_{V_2\gamma}^2

507:  \lambda^{1/2}(1,r_1^2,r_2^2) F_{\lambda_1,\lambda_2}(r_1,r_2,x)}

508:      {s^5 \, r_1^4 r_2^4

509:  \big[ (1-x^2) \lambda(1,r_1^2,r_2^2) + 4r_1^2r_2^2 \big]^2},

510: %--------------

511: \label{dsigmadx:VV}

512: \end{eqnarray}

513: %--------------

514: where $x=\cos \theta$,  $r_i = r_{V_i}$, and  

515: %--------------

516: \begin{eqnarray}

517: %--------------

518: \lambda(a,b,c) &=& a^2+b^2+c^2-2ab -2 bc-2ca.

519: %--------------

520: \end{eqnarray}

521: %--------------

522: The functions $F_{\lambda_1,\lambda_2}(r_1,r_2,x)$, 

523: which depend on the helicities of the vector mesons, are given by

524: %--------------

525: \begin{subequations}

526: \label{F10}

527: \begin{eqnarray}

528: %--------------

529: F_{\pm 1,\mp 1}(r_1,r_2,x)&=& f_{+-}(r_1,r_2,x)+ f_{+-}(r_1,r_2,-x),

530: \label{Fpm}

531: \\

532: F_{\pm 1,0}(r_1,r_2,x)&=&f_{+0}(r_1,r_2,x)+ f_{+0}(r_1,r_2,-x),

533: \\

534: F_{0,\pm 1}(r_1,r_2,x)&=&f_{+0}(r_2,r_1,x)+ f_{+0}(r_2,r_1,-x),

535: \\

536: F_{\pm 1,\pm 1}(r_1,r_2,x)&=& f_{++}(r_1,r_2,x)+ f_{++}(r_1,r_2,-x),

537: \\

538: F_{0,0}(r_1,r_2,x)&=& 16 r_1^2 r_2^2 x^2 (1-x^2),

539: %--------------

540: \end{eqnarray}

541: \end{subequations}%

542: %--------------

543: where

544: %--------------

545: \begin{subequations}

546: \label{ff}

547: \begin{eqnarray}

548: %--------------

549: f_{+-}(r_1,r_2,x)&=&

550: \mbox{$\frac{1}{2}$}(1+x)(1-x)^3(1-r_1^2-r_2^2)^2,

551: \\

552: f_{+0}(r_1,r_2,x)&=&

553: r_2^2(1-x)^2[ (1-r_2^2)(1+x)-r_1^2(1-x) ]^2,

554: \\

555: f_{++}(r_1,r_2,x)&=&

556: \mbox{$\frac{1}{2}$}(1-x^2)[ (1+x)r_2^2(1-r_2^2)-(1-x)r_1^2(1-r_1^2) 

557:          + 2 r_1^2r_2^2 x ]^2.

558: %--------------

559: \end{eqnarray}

560: \end{subequations}%

561: %--------------

562: Since $r_1$ and $r_2$ are small, the cross section is largest 

563: if the two vector mesons are transversely polarized with opposite 

564: helicities: $\lambda_1 = - \lambda_2 = \pm 1$.

565: If the vector meson $V_i$ is longitudinally polarized,

566: then the cross section is suppressed as $r_i^2$.

567: If the vector mesons have the same helicities, then the cross section 

568: is suppressed as four powers of $r_1$ or $r_2$.

569: 

570: After summing over the helicities $\lambda_1$ and $\lambda_2$ of

571: the vector mesons, we obtain

572: \begin{eqnarray}

573: \frac{d\sigma}{dx}(m_{V_1},m_{V_2})&=&

574: \frac{32\pi^3\alpha^4 g^2_{V_1\gamma} g^2_{V_2\gamma}

575:       \lambda^{1/2}(1,r_1^2,r_2^2)}

576:       {s^5 [r_1 r_2 (1-r_1^2-r_2^2)]^4 [1-(1-\Delta)x^2]^2}

577: \bigg[

578: 16 r_1^2 r_2^2 (r_1^2+r_2^2)\nonumber\\

579: &&\qquad +2(1-x^2)(1+r_1^4+r_2^4)\lambda(1,r_1^2,r_2^2)

580: -(1-x^2)^2\lambda^2(1,r_1^2,r_2^2)

581: \bigg],

582: \label{sigma-hel-sum}

583: \end{eqnarray}

584: where $\Delta = 4r_1^2 r_2^2/(1-r_1^2 -r_2^2)^2$.

585: The numerical value of $\Delta$ is very small for both charmonia and

586: light vector mesons. For example, $\Delta=0.043$ for $J/\psi + J/\psi$

587: and $\Delta = 0.00012$ for $\rho^0 + \rho^0$. In the arguments on the

588: left side of Eq.~(\ref{sigma-hel-sum}), we have indicated the

589: dependences on the masses of the vector mesons, as they will play a

590: r\^ole in a subsequent discussion of the effect of the width of the

591: $\rho$ meson. The total cross section $\sigma$ is obtained by

592: integrating over $x$ from $-1$ to $+1$ if the two vector mesons are

593: distinct. If the two vector mesons are identical, then the integration

594: range of $x$ is taken to be $0$ to $1$ in order to avoid double

595: counting. The integration formulas that are required  to evaluate the

596: definite integrals over $x$ from $-a$ to $a$ analytically are

597: %--------------

598: \begin{subequations}

599: \label{integrals}

600: \begin{eqnarray}

601: %--------------

602: \int_{-a}^a dx \frac{1}{[1-(1-\Delta)x^2]^2}&=& 

603:   \frac{a}{1-a^2+a^2\Delta}+\frac{1}{2\sqrt{1-\Delta}}

604:   \ln \frac{1+a\sqrt{1-\Delta}}{1-a\sqrt{1-\Delta}},

605: \\

606: \int_{-a}^a dx \frac{1-x^2}{[1-(1-\Delta)x^2]^2}&=& 

607:    -\frac{a\Delta}{(1-\Delta)(1-a^2+a^2\Delta)}

608: \nonumber\\

609:   &&

610:   +\frac{2-\Delta}{2(1-\Delta)^{3/2}}

611:   \ln \frac{1+a\sqrt{1-\Delta}}{1-a\sqrt{1-\Delta}},

612: \\

613: \int_{-a}^a dx \frac{(1-x^2)^2}{[1-(1-\Delta)x^2]^2}&=& 

614:   \frac{1}{(1-\Delta)^2}

615:   \left( 2a

616:         +\frac{a\Delta^2}{1-a^2+a^2\Delta}

617:   \right)

618:   \nonumber\\

619:   &&

620:   +\frac{\Delta(\Delta-4)}{2(1-\Delta)^{5/2}}

621:   \ln \frac{1+a\sqrt{1-\Delta}}{1-a\sqrt{1-\Delta}}.

622: %--------------

623: \end{eqnarray}

624: \end{subequations}%

625: %--------------

626:   

627: The cross sections for two vector mesons with $J^{PC} = 1^{--}$ have

628: also been calculated recently in Ref.~\cite{Davier:2006fu}. As was

629: pointed out in Ref.~\cite{Davier:2006fu}, at the level of precision of

630: this work, it is necessary to take into account the nonzero width of the

631: $\rho$ meson. Following Ref.~\cite{Davier:2006fu}, we note that

632: Eq.~(\ref{sigma-hel-sum}) can be generalized to the case of a continuous

633: mass spectrum:

634: \begin{equation}

635: \left(\frac{d\sigma}{dx}\right)_{\rm cont}

636: =\left(\frac{1}{4\pi^2\alpha}\right)^2\int 

637: d\tilde{m}_1^2 d\tilde{m}_2^2

638: \,\sigma_{e^+e^-\to H}(\tilde{m}_1) \,\sigma_{e^+e^-\to H}(\tilde{m}_2)

639: \, \frac{d\sigma_{\gamma_1^*\gamma_2^*}}{dx}(\tilde{m}_1,\tilde{m}_2),

640: \label{sigma-cont}

641: \end{equation}

642: where $\sigma_{e^+e^-\to H}(m)$ is the total cross section for the process

643: $e^+e^-\to \hbox{hadrons}$ at an energy $m$ in the $e^+e^-$ rest frame, 

644: and $d\sigma_{\gamma_1^*\gamma_2^*}/dx$ is the cross section for 

645: $e^+e^-$ annihilation into two massive photons. 

646: $d\sigma_{\gamma_1^*\gamma_2^*}/dx$ can be related to $d\sigma/dx$ in 

647: Eq.~(\ref{sigma-hel-sum}):

648: \begin{equation}

649: \frac{d\sigma_{\gamma_1^*\gamma_2^*}}{dx}(\tilde{m}_1,\tilde{m}_2)=

650: \left(

651: \frac{\tilde{m}_1^2 \tilde{m}_2^2}

652:      {4\pi\alpha g_{V_1\gamma}g_{V_2\gamma}}

653: \right)^2

654: \frac{d\sigma}{dx}(\tilde{m}_1,\tilde{m}_2).

655: \label{sigma-cont2}

656: \end{equation}

657: %

658: The ranges of integration of $\tilde{m}_1$ and $\tilde{m}_2$ in

659: Eq.~(\ref{sigma-cont}) are determined by the physical region for the

660: two-meson final state: $\tilde{m}_1\leq \sqrt{s}-m_{V_i}$ for the case

661: of a $\rho^0$ meson and another meson $V_i$ and

662: $\tilde{m}_1+\tilde{m}_2\leq \sqrt{s}$ for the case of two $\rho^0$

663: mesons. One can recover the narrow-width approximation by writing

664: \begin{equation}

665: \sigma_{e^+e^-\to V_i}^{\rm NW}(\tilde{m})=

666: \left(

667: \frac{4\pi\alpha g_{V_i\gamma}}{m_{V_i}^2}

668: \right)^2

669: \pi \delta(\tilde{m}^2-m_{V_i}^2).

670: \label{narrow-width}

671: \end{equation}

672: 

673: We parametrize the contribution of the $\rho^0$ meson to 

674: $\sigma_{e^+e^-\to H}(\tilde{m})$, following 

675: Ref.~\cite{Akhmetshin:2001ig}, as

676: %

677: \begin{equation}

678: \sigma_{e^+e^-\to \rho}(\tilde{m})=\frac{8\pi\alpha^2}{3\tilde{m}^5}

679: p_\pi^3(\tilde{m}^2)\theta(\tilde{m}-2m_\pi)|F_\pi(\tilde{m}^2)|^2,

680: \end{equation}

681: %

682: where $m_\pi$ is the pion mass and $p_\pi(s)$ is the pion momentum:

683: %

684: \begin{equation}

685: p_\pi(s)=\sqrt{s/4-m_\pi^2}.

686: \end{equation}

687: %

688: The form factor $F_\pi(s)$, which depends on the parameters

689: $M_\rho$, $\Gamma_\rho$, and $\beta$, is

690: %

691: \begin{equation}

692: F_\pi(s)=

693: \frac{(1+\beta)^{-1}M_\rho^2(1+d\,\Gamma_\rho/M_\rho)}

694: {M_\rho^2-s+f(s)-iM_\rho\Gamma_\rho(s)},

695: \end{equation} 

696: %

697: where

698: %

699: \begin{equation}

700: d=\frac{3m_\pi^2}{\pi p^2_\pi(M_\rho^2)}

701: \log\frac{M_\rho+2p_\pi(M_\rho^2)}{2m_\pi}+\frac{M_\rho}{2\pi p_\pi(M_\rho^2)}

702: -\frac{m_\pi^2M_\rho}{\pi p_\pi^3(M_\rho^2)},

703: \end{equation}

704: %

705: $\Gamma_\rho(s)$ is the energy-dependent width of the $\rho$,

706: %

707: \begin{equation}

708: \Gamma_\rho(s)=\Gamma_\rho\left[\frac{p_\pi(s)}{p_\pi(M_\rho^2)}\right]^3

709: \left[\frac{M_\rho^2}{s}\right]^{1/2},

710: \end{equation}

711: %

712: and the real part of the correction to the denominator is given by

713: %

714: \begin{subequations}

715: \begin{eqnarray}

716: f(s)&=&\Gamma_\rho\frac{M_\rho^2}{p_\pi^3(M_\rho^2)}

717: \left\{p_\pi^2(s)[h(s)-h(M_\rho^2)]+(M_\rho^2-s)p_\pi^2(M_\rho^2)

718: h'(M_\rho^2)\right\},

719: \\

720: h(s)&=&\frac{2}{\pi}\frac{p_\pi(s)}{\sqrt{s}}

721: \log\frac{\sqrt{s}+2p_\pi(s)}{2m_\pi},

722: \\

723: h'(s)&=&\frac{dh(s)}{ds}.

724: \end{eqnarray}

725: \end{subequations}%

726: %

727: The parameters $M_\rho$ and $\Gamma_\rho$ are obtained from a fit to the

728: data for the $e^+e^-\to \pi^+\pi^-$ cross section

729: \cite{Akhmetshin:2003zn}: $M_\rho=775.65\pm 0.64\pm 0.50$~MeV and

730: $\Gamma_\rho=143.85 \pm 1.33 \pm 0.80$~MeV. (Similar results have been

731: obtained in Ref.~\cite{Achasov:2006vp}.) The parameter $\beta$ is not

732: given in Ref.~\cite{Akhmetshin:2003zn}. We infer it from the value 

733: $\Gamma[\rho\to e^+e^-]=7.06 \pm 0.11 \pm 0.05$~MeV that is given in 

734: Ref.~\cite{Akhmetshin:2003zn} and the formula

735: for the electronic width that is given in Ref.~\cite{Akhmetshin:2001ig}:

736: \begin{equation}

737: \Gamma[\rho\to e^+e^-]=

738: \frac{2\alpha^2 p_\pi^3(M_\rho^2)}{9M_\rho\Gamma_\rho}

739: \frac{(1+d\,\Gamma_\rho/M_\rho)^2}{(1+\beta)^2}.

740: \label{lept-width}

741: \end{equation}

742: Using the values for $M_\rho$ and $\Gamma_\rho$ from 

743: Ref.~\cite{Akhmetshin:2003zn}, we obtain $\beta=-0.0815234$. 

744:   

745: 

746: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

747: \begin{table}[t]

748: \caption{\label{crosssection}% 

749: Cross sections in units of fb for $e^+ e^- \to V_1 + V_2$ at $E_{\rm

750: beam}=5.29$ GeV, calculated by using the fragmentation approximation.

751: The uncertainties that are shown are only those that arise from the

752: uncertainties in the electronic widths of the vector mesons. The first

753: five rows are calculated in the narrow-width approximation. The last row

754: is calculated by taking into account the nonzero width of the $\rho$

755: meson, as is described in the text.}

756: \begin{ruledtabular}

757: \begin{tabular}{l|ccccc}

758: %-------------------------------------------------------------------------

759: $V_1$ $\backslash$ $V_2$ & $\rho^0$         &   $\omega$

760:            &   $\phi$  &   $J/\psi$       &   $\psi(2S)$      \\

761: %-------------------------------------------------------------------------

762: \hline

763: %-------------------------------------------------------------------------

764: $\rho^0$

765: & $139.61\pm 4.82$

766: & $23.47\pm 0.88$

767: & $35.93\pm 1.29$

768: & $41.67\pm 1.27$

769: & $15.60\pm 0.46 $ \\

770: %-------------------------------------------------------------------------

771: $\omega$

772: &

773: & $\phantom{3}0.99\pm 0.07 $

774: & $\phantom{3}3.02\pm 0.14 $

775: & $\phantom{3}3.50\pm 0.15 $

776: & $\phantom{3}1.31\pm 0.05  $ \\

777: %-------------------------------------------------------------------------

778: $\phi$

779: &

780: &

781: & $\phantom{3}2.30\pm 0.15 $

782: & $\phantom{3}5.20\pm 0.21 $

783: & $\phantom{3}1.94\pm 0.08  $ \\

784: %-------------------------------------------------------------------------

785: $J/\psi$

786: &

787: &

788: &

789: & $\phantom{3}2.52\pm 0.13 $

790: & $\phantom{3}1.81 \pm 0.06  $ \\

791: %-------------------------------------------------------------------------

792: $\psi(2S)$

793: &

794: &

795: &

796: &

797: & $\phantom{3}0.32 \pm 0.02  $ \\

798: %-------------------------------------------------------------------------

799: \hline

800: %-------------------------------------------------------------------------

801: $\rho^0$

802: & $126.08\pm 4.36$

803: & $22.31\pm 0.84$

804: & $34.18\pm 1.23$

805: & $39.83\pm 1.22$

806: & $14.92\pm 0.44$ \\

807: %-------------------------------------------------------------------------

808: \end{tabular}

809: \end{ruledtabular}

810: \end{table}

811: 

812: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

813: In Table~\ref{crosssection}, we give our results for the integrated cross

814: sections, calculated in the narrow-width approximation, for the

815: production of $V_1+V_2$ for the vector mesons $V_i=\rho^0$, $\omega$,

816: $\phi$, $J/\psi$, and $\psi(2S)$. We also give integrated cross sections

817: for the production of $\rho^0+V$ in which the nonzero width of the

818: $\rho$ has been taken into account. The uncertainties that are shown

819: are only those that arise from the uncertainties in the $V-\gamma$

820: coupling constants that are given in Table~\ref{constants}. The effect

821: of the nonzero width of the $\rho$ meson is to decrease the cross

822: sections for $\rho^0 + \rho^0$ by about 10\%, for $\rho^0 + \omega$ and

823: $\rho^0 + \phi$ by about 5\%, and for $\rho^0 + J/\psi$ and $\rho^0

824: + \psi(2S)$ by about 4\%.

825: 

826: We note that the differences between the zero-width and nonzero-width

827: results arise mostly from the fact that the quantity \begin{equation}

828: I_\rho=\int_{4m_\pi^2}^\infty

829: d\tilde{m}^2\sigma_{e^+e^-\to\rho}(\tilde{m}) \end{equation} differs

830: from the coefficient of the $\delta$~function in the narrow-width

831: approximation of Eq.~(\ref{narrow-width}). The quantity $I_\rho$ is

832: about $6\%$ smaller than the coefficient of the $\delta$~function in

833: Eq.~(\ref{narrow-width}). (The effect of the nonzero width of the $\rho$

834: meson itself is to actually {\it increase} the cross sections.) The

835: coefficient of the $\delta$~function in Eq.~(\ref{narrow-width}) derives

836: from the electronic width of the $\rho$ meson, which, in turn, is

837: calculated in Ref.~\cite{Akhmetshin:2003zn} by using

838: Eq.~(\ref{lept-width}). Eq.~(\ref{lept-width}) is derived from the

839: vector-meson-dominance model \cite{Akhmetshin:2001ig}. Hence, the

840: experimental electronic width of the $\rho$ meson that is given in

841: Ref.~\cite{Akhmetshin:2003zn} depends on the assumptions of that model.

842: An alternative definition of the electronic width of the $\rho$ meson can

843: be obtained by equating $I_\rho$ to the coefficient of the

844: $\delta$~function in Eq.~(\ref{narrow-width}) and using

845: Eq.~(\ref{gamma-v-ee}) to relate $g_{V_i\gamma}^2$ to $\Gamma[\rho \to

846: e^+e^-]$. This approach leads to the result

847: %

848: \begin{equation}

849: \Gamma[\rho \to e^+e^-]=\frac{m_\rho}{12\pi^2}I_\rho.

850: \label{lept-width-alt}

851: \end{equation}

852: %

853: It can be shown that the definition in Eq.~(\ref{lept-width-alt}) 

854: differs from the one in Eq.~(\ref{lept-width}) by terms of order 

855: $\Gamma_\rho^2/m_\rho^2$. 

856: 

857: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

858: % Fig 3

859: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

860: \begin{figure}

861: \begin{tabular}{cc}

862: \epsfig{file=fig3a.ps,width=8.0cm}&

863: \epsfig{file=fig3b.ps,width=8.0cm}

864: \end{tabular}

865: \vspace*{2pt}

866: \caption{\label{fig3}%

867: Differential cross sections $d\sigma/d|x|$ for $e^+ e^-$ annihilation

868: into $V_1+V_2$ at $E_{\rm beam}= 5.29$ GeV, where $V_i= \rho^0$, $\phi$,

869: or $J/\psi$. The areas under the curves are the total cross sections.

870: }

871: \end{figure}

872: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

873: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

874: In Fig.~\ref{fig3}, we show the differential cross sections

875: $d\sigma/d|x|$ for the production of the identical vector mesons

876: $\rho^0+\rho^0$, $\phi +\phi$, and $J/\psi + J/\psi$ and for the 

877: production of the distinct vector mesons $\rho^0+\phi$, $\rho^0+J/\psi$,

878: and $\phi+J/\psi$. The differential cross sections peak sharply near

879: $|x|=1$ before falling to zero at the endpoint. For example, the maximum

880: values of the differential cross sections are $7.1 \times 10^4$ fb

881: at $x=0.99994$ for $\rho^0+\rho^0$ and $13.3$ fb at $x=0.993$ for

882: $J/\psi + J/\psi$.

883: 

884: We now discuss the theoretical uncertainties in the predictions for the cross

885: sections in Table~\ref{crosssection}. The scaling of the theoretical

886: uncertainties was determined in Section~\ref{ampl}. In the case of two

887: light vector mesons, the leading theoretical uncertainties arise from

888: the fragmentation approximation and from exchanges of collinear gluons

889: between the mesons. These uncertainties both scale as $(\Lambda_{\rm

890: QCD}/E_{\rm beam})^2$. If we take $\Lambda_{\rm QCD} \approx 0.5$ GeV,

891: then the estimated fractional uncertainty is 0.9\%. In the case

892: of a light vector meson and a charmonium meson, there is an additional

893: theoretical uncertainty that arises from the exchange of hard

894: gluons between the mesons that scales as $\alpha_s^2(E_{\rm beam}) \,

895: (m_c\Lambda_{\rm QCD}/ E_{\rm beam}^2)$. If we take $\alpha_s^2(E_{\rm

896: beam})=0.25$ and $m_c = 1.4$ GeV, then the estimated fractional

897: uncertainty is 0.2\%. In the case of two charmonium mesons, the

898: leading theoretical uncertainty arises from the fragmentation

899: approximation and scales as $m_c^2/E_{\rm beam}^2$. If we take $m_c =

900: 1.4$ GeV, then the estimated fractional uncertainty is 7\%. This

901: uncertainty can be reduced by calculating the contributions to the

902: cross section from nonfragmentation diagrams. We discuss this

903: calculation in Sec.~\ref{sec:nonfrag}.

904: 

905: \section{Nonfragmentation corrections to the production of two

906: charmonia}\label{sec:nonfrag}

907: 

908: In this section, we calculate corrections to the production cross

909: sections for two charmonium mesons that arise from the nonfragmentation

910: contributions to the amplitudes. The contributions from the

911: nonfragmentation diagrams in Fig.~\ref{fig2} can be calculated using the

912: NRQCD factorization formalism \cite{Bodwin:1994jh}. This formalism

913: allows QCD radiative corrections and relativistic corrections to be

914: taken into account systematically. For example, the expression for the

915: $V-\gamma$ coupling constant defined by Eq.~(\ref{dsigmadx:VV}),

916: including the leading QCD radiative correction and the leading

917: relativistic correction, is

918: %--------------

919: \begin{eqnarray}

920: %--------------

921: g_{V \gamma} = 

922: e_c \sqrt{2 m_V}

923: \left(1 - \frac{8\alpha_s}{3\pi} - \frac{1}{6} \langle v^2 \rangle_V \right)

924: \langle V(\lambda) |\psi^\dagger \vec{\sigma} \chi | 0 \rangle 

925: \cdot  \vec{\epsilon}(\lambda) .

926: %--------------

927: \label{g-NRQCD}

928: \end{eqnarray}

929: %--------------

930: Here $\psi^\dagger$ and $\chi$ are the two-component Pauli operators in

931: the NRQCD formalism that create a heavy quark and antiquark,

932: respectively. $\langle v^2 \rangle_V$ is proportional to a ratio of

933: matrix elements of NRQCD operators between the state $V$ and the

934: vacuum.   The factor $e_c$ comes from the electromagnetic current in

935: Eq.~(\ref{current}).  The factor $\sqrt{2 m_V}$ takes into account the

936: difference between the standard relativistic and nonrelativistic

937: normalizations of the state $|V(\lambda)\rangle$.  

938: 

939: The NRQCD factorization formalism was used in

940: Refs.~\cite{Bodwin:2002fk,Bodwin:2002kk} to calculate the cross sections

941: for double-charmonium production from $e^+e^-$ annihilation into two

942: virtual photons. The calculation of the cross sections and the estimates

943: of theoretical errors were carried out in a way that was as close as

944: possible to a previous calculation of the cross sections for

945: double-charmonium production from $e^+e^-$ annihilation into a single

946: virtual photon \cite{Braaten:2002fi}. The cross sections were expressed

947: in terms of the coupling constants $\alpha$ and $\alpha_s$, the pole

948: mass $m_c$ of the charm quark, and a factor $\langle {\cal O}_1

949: \rangle_V$ that is   related to the NRQCD matrix element in

950: Eq.~(\ref{g-NRQCD}):

951: %--------------

952: \begin{eqnarray}

953: %--------------

954: \langle {\cal O}_1 \rangle_V = 

955: \frac{1}{3} \sum_\lambda 

956: \left| \langle V(\lambda) |\psi^\dagger \vec{\sigma} \chi | 0 \rangle \right|^2 .

957: %--------------

958: \label{O-NRQCD}

959: \end{eqnarray}

960: %--------------

961: The fragmentation terms in the cross sections in

962: Ref.~\cite{Bodwin:2002kk} can be recovered by replacing $g_{V\gamma}^2$

963: in Eq.~(\ref{dsigmadx:VV}) with $16 m_c \langle {\cal O}_1 \rangle_V/9$

964: and by replacing $m_V$ with $2 m_c$.  (In the calculation of

965: Ref.~\cite{Bodwin:2002kk}, the relative momentum of the $c$ and $\bar c$

966: that form the charmonium and their binding energy were neglected, and so

967: the invariant mass of the $c \bar c$ pair was taken to be $2 m_c$.) In

968: Table~II of Ref.~\cite{Bodwin:2002kk}, the cross section for

969: $J/\psi+J/\psi$  was given as $6.65 \pm 3.02$ fb, where the uncertainty

970: comes only from varying the charm quark mass over the range

971: $1.2~\hbox{GeV}\leq m_c\leq 1.6~\hbox{GeV}$.   This result includes both

972: the fragmentation diagrams in Fig.~\ref{fig1} and the nonfragmentation

973: diagrams in Fig.~\ref{fig2}. It does not include QCD radiative

974: corrections or relativistic corrections, which are known only for the

975: fragmentation term. As is described in Ref.~\cite{Bodwin:2002kk}, if

976: the known corrections to the fragmentation term in the cross section are

977: applied to the entire cross section, then the central value for the

978: $J/\psi+J/\psi$ cross section decreases to about $2$~fb.  This value is

979: reasonably close to the result in Table~\ref{crosssection}, but there 

980: are large uncertainties that arise from the uncertainty in $m_c$. 

981: Furthermore, this procedure suffers from the deficiency that the 

982: radiative and relativistic corrections to the fragmentation term in the 

983: cross section are not necessarily valid for the entire cross section.

984: 

985: A more correct procedure would be to apply the known corrections to the

986: fragmentation amplitude to that amplitude alone in the calculation of

987: Ref.~\cite{Bodwin:2002kk}. However, this approach would still be

988: hampered by a large uncertainty in the fragmentation amplitude that

989: arises from the uncertainty in $m_c$. Furthermore, there are higher-order

990: radiative and relativistic corrections to the fragmentation amplitude

991: that would not be included in this approach.

992: 

993: Both of these drawbacks can be eliminated by writing the fragmentation

994: amplitude in terms of $g_{V \gamma}$ and $m_V$ instead of 

995: $\langle {\cal O}_1 \rangle_V$ and $m_c$. The resulting

996: expression for the fragmentation term in the differential cross section

997: is Eq.~(\ref{dsigmadx:VV}), which has no explicit dependence on the

998: charm-quark mass. Note that this procedure cannot be used to account

999: correctly for the radiative and relativistic corrections to the

1000: nonfragmentation amplitude. These corrections to the nonfragmentation

1001: amplitude do not necessarily correspond to those that are contained in

1002: $g_{V \gamma}$. Furthermore, it would be inappropriate to set $m_c$ equal

1003: to $m_V/2$ in the nonfragmentation amplitude, since the vector-meson

1004: mass has little to do with the propagation of a charm quark at short

1005: distances of order $1/E_{\rm beam}$. In the case of production of

1006: $J/\psi + \psi(2S)$, no consistent choice would be possible because

1007: $m_{J/\psi} \ne m_{\psi(2S)}$.

1008: 

1009: We now explain precisely how we calculate improved cross sections for

1010: two charmonium vector mesons that take into account the nonfragmentation

1011: contributions. For each pair of vector meson helicities $\lambda_1$ and

1012: $\lambda_2$, the cross section can be expressed as the product of a flux

1013: factor $1/(2 s)$, the square of a T-matrix element, and a phase-space

1014: factor $\lambda^{1/2}(1, r_1^2, r_2^2)/(8 \pi)$. We use the physical

1015: vector meson masses in the phase-space factor. The T-matrix element is

1016: the sum of a fragmentation amplitude and a nonfragmentation amplitude.

1017: The cross section is the sum of a fragmentation contribution, 

1018: a nonfragmentation contribution, and an interference contribution. 

1019: Our expression for the fragmentation amplitude is proportional to 

1020: $g_{V_1\gamma}g_{V_2\gamma}$, with a coefficient that is a function 

1021: of the meson masses $m_{V_1}$ and $m_{V_2}$. Hence, 

1022: the fragmentation contribution to the cross section is given by

1023: Eq.~(\ref{dsigmadx:VV}). Our expression for the nonfragmentation

1024: amplitude is proportional to $(4 m_{V_1} m_{V_2} \langle{\cal

1025: O}\rangle_{V_1} \langle{\cal O}\rangle_{V_2})^{1/2}$, with a coefficient

1026: that is a function of the quark mass $m_c$. The factors $(2

1027: m_{V_i})^{1/2}$ arise from the relativistic normalizations of the meson

1028: states. Aside from these normalization factors, the square of the

1029: nonfragmentation contribution to the T-matrix is identical to that in

1030: Ref.~\cite{Bodwin:2002kk}. The interference contribution to the square of the

1031: T-matrix element depends on both the meson masses and the quark mass. We

1032: calculate the cross sections for each helicity combination and then add

1033: them. This approach is convenient for taking into account the difference

1034: between the polarization vector for a meson with mass $m_{V_i}$ and the

1035: polarization vector for a quark pair with invariant mass $2 m_c$. The

1036: interference and nonfragmentation contributions to the cross section are

1037: given in Eqs.~(\ref{int-hel-sum}) and (\ref{nonfrag-hel-sum}) in the 

1038: Appendix.

1039: 

1040: Next, let us describe how we estimate the residual theoretical

1041: uncertainties. The uncertainties are obtained by adding five theoretical

1042: uncertainties in quadrature. The only significant uncertainties in the

1043: fragmentation amplitudes are those that arise from the electronic widths

1044: of the vector mesons, which enter through the overall factor of $g_{V_1

1045: \gamma} g_{V_2 \gamma}$. These uncertainties affect the fragmentation

1046: and interference terms in the cross sections. The most significant

1047: uncertainties in the nonfragmentation amplitudes arise from the NRQCD

1048: matrix elements, QCD radiative corrections, relativistic corrections,

1049: and the charm-quark mass. These uncertainties affect the interference

1050: and nonfragmentation contributions to the cross sections. We estimate

1051: the error associated with the charm-quark mass by varying $m_c$ in the

1052: interference and nonfragmentation terms in the cross section over the

1053: range $m_c = 1.4 \pm 0.2$ GeV. Our estimates of the uncertainties in the

1054: NRQCD matrix elements $\langle{\cal O}_1\rangle_V$ are described below.

1055: We assume that the QCD radiative corrections to the nonfragmentation

1056: amplitude are of relative size $\alpha_s(2m_c)=0.25$. We assume that the

1057: relativistic corrections are of relative size $(\langle

1058: v^2\rangle_{V_1}+\langle v^2\rangle_{V_2})/2$. Radiative corrections to

1059: the fragmentation amplitude that involve the exchange of hard gluons

1060: between the two mesons are suppressed as $\alpha_s^2(E_{\rm beam}) \,

1061: (m_c^2/E_{\rm beam}^2)\approx 0.4\%$ and can be neglected. Corrections

1062: that involve the exchange of soft gluons between the two mesons are

1063: suppressed as $(m_cv)^4/E_{\rm beam}^4 \approx 0.04\%$ and can also be

1064: neglected.

1065: 

1066: We compute the NRQCD matrix elements 

1067: $\langle {\cal O}_1 \rangle_{J/\psi}$ and 

1068: $\langle {\cal O}_1 \rangle_{\psi(2S)}$ from the electronic widths

1069: of the charmonia by making use of the formula

1070: \begin{equation}

1071: \Gamma[V \to e^+e^-]=\frac{8e_c^2\pi\alpha^2}{3}

1072: \frac{\langle{\cal 

1073: O}_1\rangle_V}{m_{V}^2}\left(1-\frac{8}{3}\frac{\alpha_s}{\pi}

1074: -\frac{1}{6}\langle v^2\rangle_V\right)^2,

1075: \label{lept-width-nrqcd}

1076: \end{equation}

1077: which follows from Eq.~(\ref{g-NRQCD}). Note that

1078: Eq.~(\ref{lept-width-nrqcd}) differs from the NRQCD factorization

1079: formula that is given in Ref.~\cite{Bodwin:1994jh}: $m_c$ has been

1080: replaced with $m_V/2$ and, consequently, the relativistic correction has

1081: changed from $-\frac{2}{3}\langle v^2\rangle_V$ to $-\frac{1}{6}\langle

1082: v^2\rangle_V$. The formula (\ref{lept-width-nrqcd}) is less subject to

1083: uncertainties in the value of $\langle v^2\rangle_V$ than the standard

1084: NRQCD formula and, through its dependence on $m_V$, resums some

1085: corrections of higher order in $v$. In our calculation, we take $\langle

1086: v^2\rangle_{J/\psi}=0.25\pm 0.09$ (Ref.~\cite{Bodwin:2006dn}),

1087: $\langle v^2\rangle_{\psi(2S)}= 0.45 \pm 0.19$

1088: (Ref.~\cite{kang}), and $\alpha_s(2m_c)\approx 0.25$. We obtain

1089: \begin{eqnarray}

1090: \langle{\cal O}_1\rangle_{J/\psi}&=& 0.482 \pm 0.049~\hbox{GeV}^3,

1091: \nonumber\\

1092: \langle{\cal O}_1\rangle_{\psi(2S)}&=& 0.335\pm 0.080~\hbox{GeV}^3.

1093: \label{ME-values}

1094: \end{eqnarray}

1095: The error bars are obtained by combining in quadrature the uncertainties

1096: from the electronic widths, the uncertainties from the values of $\langle

1097: v^2\rangle_{V_i}$, and the estimated errors from uncalculated radiative

1098: and relativistic corrections, which we assume to be of relative size

1099: $\alpha_s^2$ and $\langle v^2\rangle_V^2$ in the rate, respectively.

1100: Note that the values of the NRQCD matrix elements in

1101: Eq.~(\ref{ME-values}) are considerably larger than those that were used

1102: in Refs.~\cite{Bodwin:2002fk,Bodwin:2002kk}.

1103: 

1104: Our results for the cross sections for the production of $J/\psi +

1105: J/\psi$, $J/\psi + \psi(2S)$, and $\psi(2S) + \psi(2S)$ are given in

1106: Table~\ref{crosssectioncharmonium}. Note that the error bars for the

1107: total cross section are less than the error bars for the fragmentation,

1108: interference, and nonfragmentation contributions added in quadrature

1109: because the interference uncertainties are 100\% anticorrelated with the

1110: fragmentation and nonfragmentation uncertainties. The cross section for

1111: $J/\psi + J/\psi$ is $1.69\pm 0.35$~fb. The central value is $33\%$

1112: smaller than the value in the fragmentation approximation

1113: (Table~\ref{crosssection} or the first row of

1114: Table~\ref{crosssectioncharmonium}). The $33\%$ correction to the

1115: fragmentation approximation is much larger than the estimate

1116: $m_c^2/E_{\rm beam}^2\approx 7\%$ that was given at the end of

1117: Sec.~\ref{sec:differential}. The reason for the large relative size of

1118: this correction is that radiative and relativistic corrections to the

1119: fragmentation contribution to the cross section, which are contained

1120: implicitly in the quantities $g_{V_i\gamma}$ and $m_{V_i}$, decrease

1121: the fragmentation contribution by about a factor of three from

1122: its value without radiative and relativistic corrections.  We are able

1123: to obtain a reasonably accurate result for the $J/\psi + J/\psi$ total

1124: cross section, in spite of the large relative errors in the

1125: nonfragmentation and interference terms, because the total cross section

1126:  is dominated by the fragmentation term.

1127:  

1128: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1129: \begin{table}[t]

1130: \caption{\label{crosssectioncharmonium}% 

1131: Cross sections in units of fb for $e^+ e^- \to V_1 + V_2$ at $E_{\rm

1132: beam}=5.29$ GeV for charmonium vector mesons, calculated by including

1133: both the fragmentation and nonfragmentation amplitudes, as is described

1134: in the text. The four rows give the fragmentation, interference,

1135: nonfragmentation, and total contributions to the cross sections. The

1136: uncertainties are obtained by combining five uncertainties in

1137: quadrature, as is described in the text.}

1138: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1139: \begin{ruledtabular}

1140: \begin{tabular}{l|ccc}

1141: %-------------------------------------------------------------------------

1142: cross section & $J/\psi+J/\psi$ & $J/\psi+\psi(2S)$ & $\psi(2S)+\psi(2S)$

1143: \\

1144: \hline

1145: fragmentation

1146: & $ 2.52 \pm 0.13$

1147: & $ 1.81 \pm 0.06$

1148: & $ 0.32 \pm 0.02$

1149: \\

1150: interference

1151: & \hspace{-2ex}$ -0.98 \pm 0.48$

1152: & \hspace{-2ex}$ -1.09 \pm 0.60$

1153: & \hspace{-2ex}$ -0.30 \pm 0.19$

1154: \\

1155: nonfragmentation

1156: & $ 0.15 \pm 0.16$

1157: & $ 0.23 \pm 0.29$

1158: & $ 0.09 \pm 0.14$

1159: \\

1160: total

1161: & $1.69 \pm 0.35$

1162: & $0.95 \pm 0.36$

1163: & $0.11 \pm 0.09$

1164: %-------------------------------------------------------------------------

1165: \end{tabular}

1166: \end{ruledtabular}

1167: \end{table}

1168: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1169: 

1170: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1171: \section{Comparison with experiment and  previous theoretical

1172: results}\label{sec:comparison}

1173: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1174: 

1175: The BaBar Collaboration has recently measured the cross sections for 

1176: production of $\rho^0+\rho^0$ and $\rho^0+\phi$ in $e^+ e^-$ collisions at 

1177: energy $E_{\rm beam}=5.29$ GeV~\cite{Aubert:2006we}: 

1178: %--------------

1179: \begin{subequations}

1180: \label{sigmaBabar}

1181: \begin{eqnarray}

1182: %--------------

1183: \sigma[e^+ e^- \to \rho^0+\rho^0]  &=& 

1184: 20.7 \pm 0.7_{\rm stat} \pm 2.7_{\rm syst} \ {\rm fb},

1185: \\

1186: \sigma[e^+ e^- \to \rho^0+\phi] &=& 

1187: 5.7 \pm 0.5_{\rm stat} \pm 0.8_{\rm syst} \ {\rm fb}.

1188: %--------------

1189: \end{eqnarray}

1190: \end{subequations}%

1191: %--------------

1192: The measured cross 

1193: sections are subject to the cuts $|\cos \theta| < 0.8$, 

1194: $0.5~\hbox{GeV}<m_\rho<1.1~\hbox{GeV}$, and 

1195: $1.008~\hbox{GeV}<m_\phi<1.035~\hbox{GeV}$.

1196: 

1197: In Table~\ref{crosssectionwithcut}, we give our predictions for the

1198: cross sections that involve light mesons, integrated over the region

1199: $|\cos \theta| < 0.8$. In Table~\ref{charmcrosssection-cut}, we give

1200: our predictions for the charmonium-charmonium cross sections, integrated

1201: over the same angular region. The cross sections in

1202: Tables~\ref{crosssectionwithcut} and \ref{charmcrosssection-cut} are

1203: much smaller than those in Table~\ref{crosssection} and

1204: Table~\ref{crosssectioncharmonium}, respectively, because the cut on

1205: $\theta$ excludes most of the peak near $|x|=1$. In the last row of

1206: Table~\ref{crosssectionwithcut}, the $m_\rho$ cut is implemented. We can

1207: calculate the effect of the $m_\phi$ cut as follows. The cross section to

1208: produce a $\phi$ meson and one of the other vector mesons is a constant,

1209: to within $1.5\%$, over the range

1210: $1.008~\hbox{GeV}<m_\phi<1.035~\hbox{GeV}$. Hence, we can calculate the

1211: fractional correction from the $m_\phi$ cut, with errors of less than

1212: $1\%$, by calculating the integral with respect to $m_\phi^2$ of the

1213: $\phi$ line shape over the range

1214: $1.008~\hbox{GeV}<m_\phi<1.035~\hbox{GeV}$ and comparing it with the

1215: integral of the line shape from threshold to infinity. Using a simple

1216: Breit-Wigner line shape and taking the values of $m_\phi$ and

1217: $\Gamma_\phi$ in Table~\ref{constants}, we find that the fraction 0.898

1218: of the integral of the line shape is contained in the range

1219: $1.008~\hbox{GeV}<m_\phi<1.035~\hbox{GeV}$. Therefore, in order to

1220: compare the results in Table~\ref{crosssectionwithcut} with the BaBar

1221: results, one should multiply each entry in

1222: Table~\ref{crosssectionwithcut} by a factor 0.898 for each $\phi$ meson

1223: in the final state. Specifically, we obtain $\sigma[e^+e^- \to

1224: \rho^0+\phi]=5.04\pm 0.18$~fb with the mass cuts on both $m_\rho$ and

1225: $m_\phi$.  Our predictions for $\rho^0+\rho^0$ and $\rho^0+\phi$ in

1226: Table~\ref{crosssectionwithcut} agree with the BaBar results in

1227: Eq.~(\ref{sigmaBabar}) to within the experimental and theoretical

1228: uncertainties.

1229: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1230: \begin{table}[t]

1231: \caption{\label{crosssectionwithcut}% 

1232: Cross sections in units of fb for $e^+ e^- \to V_1 + V_2$ at $E_{\rm

1233: beam}=5.29$ GeV, calculated using the fragmentation approximation. The

1234: angular cut $|\cos \theta| < 0.8$, which is used in

1235: Ref.~\cite{Aubert:2006we}, has been applied. The uncertainties shown are

1236: only those that arise from the uncertainties in the electronic widths of

1237: the vector mesons. The first three rows are calculated in the

1238: narrow-width approximation. The last two rows are calculated by taking

1239: into account the nonzero width of the $\rho$ meson, as is described in

1240: the text. In the last row, the cut

1241: $0.5~\hbox{GeV}<m_\rho<1.1~\hbox{GeV}$, which is used in

1242: Ref.~\cite{Aubert:2006we}, has been applied. }

1243: \begin{ruledtabular}

1244: \begin{tabular}{l|ccccc}

1245: %-------------------------------------------------------------------------

1246: $V_1$ $\backslash$ $V_2$ & $\rho^0$         &   $\omega$

1247:            &   $\phi$  &   $J/\psi$       &   $\psi(2S)$      \\

1248: %-------------------------------------------------------------------------

1249: \hline

1250: %-------------------------------------------------------------------------

1251: $\rho^0$

1252: & $23.30\pm 0.80$

1253: & $3.93\pm 0.15$

1254: & $6.43\pm 0.23$

1255: & $10.92\pm 0.33$

1256: & $4.44 \pm 0.13 $ \\

1257: %-------------------------------------------------------------------------

1258: $\omega$

1259: &

1260: & $0.17\pm 0.01 $

1261: & $0.54\pm 0.02 $

1262: & $0.92\pm 0.04 $

1263: & $0.37\pm 0.02 $ \\

1264: %-------------------------------------------------------------------------

1265: $\phi$

1266: &

1267: &

1268: & $0.44\pm 0.03 $

1269: & $1.50\pm 0.06 $

1270: & $0.61\pm 0.02 $ \\

1271: %-------------------------------------------------------------------------

1272: \hline

1273: %-------------------------------------------------------------------------

1274: $\rho^0$ (no mass cut)

1275: & $20.50\pm 0.71$

1276: & $3.68\pm 0.14$

1277: & $6.03\pm 0.22$

1278: & $10.24\pm 0.31$

1279: & $4.17\pm 0.12$ \\

1280: %-------------------------------------------------------------------------

1281: \hline

1282: %-------------------------------------------------------------------------

1283: $\rho^0$ (mass cut)

1284: & $17.71\pm 0.61$

1285: & $3.42 \pm 0.13$

1286: & $5.61 \pm 0.20$

1287: & $9.52\pm 0.29$

1288: & $3.87\pm 0.11$ \\

1289: %-------------------------------------------------------------------------

1290: \end{tabular}

1291: \end{ruledtabular}

1292: \end{table}

1293: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1294: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1295: \begin{table}[t]

1296: \caption{\label{charmcrosssection-cut}% 

1297: Cross sections in units of fb for $e^+ e^- \to V_1 + V_2$ at $E_{\rm

1298: beam}=5.29$ GeV for charmonium vector mesons, calculated by including

1299: both the fragmentation and nonfragmentation amplitudes, as is described

1300: in the text. The angular cut $|\cos\theta|<0.8$ has been applied. The

1301: four rows give the fragmentation, interference, nonfragmentation, and

1302: total  contributions to the cross sections. The uncertainties are

1303: obtained by combining five uncertainties in quadrature, as is described

1304: in the text.}

1305:  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1306:  \begin{ruledtabular}

1307:  \begin{tabular}{l|ccc}

1308:  %-------------------------------------------------------------------------

1309:  cross section & $J/\psi+J/\psi$ & $J/\psi+\psi(2S)$ & $\psi(2S)+\psi(2S)$

1310:  \\

1311:  \hline

1312:  fragmentation

1313:  & $ 1.20 \pm 0.06$

1314:  & $ 0.94 \pm 0.03$

1315:  & $ 0.18 \pm 0.01$

1316:  \\

1317:  interference

1318:  & \hspace{-2ex}$ -0.72 \pm 0.36$

1319:  & \hspace{-2ex}$ -0.81 \pm 0.45$

1320:  & \hspace{-2ex}$ -0.22 \pm 0.15$

1321:  \\

1322:  nonfragmentation

1323:  & $ 0.13 \pm 0.14$

1324:  & $ 0.20 \pm 0.25$

1325:  & $ 0.08 \pm 0.12$

1326:  \\

1327:  total

1328:  & $0.60 \pm 0.24$

1329:  & $0.33 \pm 0.24$

1330:  & $0.04 \pm 0.06$

1331:  %-------------------------------------------------------------------------

1332:  \end{tabular}

1333:  \end{ruledtabular}

1334:  \end{table}

1335:  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1336: 

1337: Our results in Tables~\ref{crosssection} and \ref{crosssectionwithcut}

1338: differ somewhat from those in Ref.~\cite{Davier:2006fu}. Some of the

1339: differences arise because we are using the 2006 compilation of the

1340: Particle Data Group \cite{PDG2006}, rather than the 2004 compilation

1341: \cite{Eidelman:2004wy}, for the meson masses and widths. However, once

1342: these differences in the input data are taken into account,

1343: discrepancies still remain. The largest discrepancy is between the

1344: results for $\rho^0 +\psi(2S)$ production with nonzero $\rho$-meson width

1345: and $|\cos\theta|<0.8$. This discrepancy is about $27\%$.  Other

1346: discrepancies for cross sections computed with the cut

1347: $|\cos\theta|<0.8$ are approximately $4\%$ or less. For cross sections

1348: computed with $|\cos\theta|<1.0$, the discrepancies are as large as

1349: $10\%$.

1350: 

1351: The production cross sections for two vector-meson charmonium states 

1352: were also calculated in the fragmentation and narrow-width 

1353: approximations in Ref.~\cite{Luchinsky:2003yh}. In this paper, somewhat 

1354: different values for the electronic widths of the charmonium states were 

1355: used, and the resulting production cross sections are lower than ours by

1356: about $10$--$30\%$.

1357: 

1358: The Belle Collaboration has set an upper limit on the cross 

1359: section for $J/\psi+J/\psi$ in $e^+ e^-$ collisions at

1360: energy $E_{\rm beam} \approx 5.29$ GeV~\cite{Abe:2004ww}: 

1361: %--------------

1362: \begin{subequations}

1363: \begin{eqnarray}

1364: %--------------

1365: \sigma[e^+ e^- \to J/\psi+J/\psi] \times {\cal B}_{>2}[J/\psi] &<& 

1366: 9.1 \ {\rm fb} \hspace{1cm} (90\% \ {\rm C.L.}),

1367: \label{sigmaBelle}

1368: \\

1369: \sigma[e^+ e^- \to J/\psi+\psi(2S)] \times {\cal B}_{>2}[\psi(2S)] &<& 

1370: 5.2 \ {\rm fb} \hspace{1cm} (90\% \ {\rm C.L.}),

1371: %--------------

1372: \end{eqnarray}

1373: \end{subequations}%

1374: %--------------

1375: where ${\cal B}_{>2}[V]$ is the branching fraction of $V$ 

1376: into final states with  more than two charged tracks.

1377: These upper limits are compatible with the predictions 

1378: in Table~\ref{crosssectioncharmonium}.

1379: By adding up exclusive branching fractions for $J/\psi$ decays

1380: \cite{PDG2006}, one can show that the 

1381: branching fraction in Eq.~(\ref{sigmaBelle}) satisfies

1382: $13\% < {\cal B}_{>2}[J/\psi] < 80\%$.

1383: It should be possible to measure ${\cal B}_{>2}[J/\psi]$ 

1384: and ${\cal B}_{>2}[\psi(2S)]$ at CLEOc or BESIII.

1385: The upper limit in Eq.~(\ref{sigmaBelle}) was obtained with 

1386: a data sample of 155 fb$^{-1}$ at or near the $\Upsilon(4S)$.  

1387: The combined data samples of the Belle and BaBar experiments 

1388: now exceed 1000 fb$^{-1}$.  Our prediction for the cross 

1389: section for $e^+ e^- \to J/\psi+J/\psi$ indicates that 

1390: there is a possibility that this process can be observed 

1391: at the $B$ factories. 

1392: 

1393: 

1394: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1395: % If in two-column mode, this environment will change to single-column

1396: % format so that long equations can be displayed. Use

1397: % sparingly.

1398: %\begin{widetext}

1399: % put long equation here

1400: %\end{widetext}

1401: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1402: 

1403: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1404: % Surround figure environment with turnpage environment for landscape

1405: % figure

1406: % \begin{turnpage}

1407: % \begin{figure}

1408: % \includegraphics{}%

1409: % \caption{\label{}}

1410: % \end{figure}

1411: % \end{turnpage}

1412: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1413: 

1414: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1415: % tables should appear as floats within the text

1416: % Here is an example of the general form of a table:

1417: % Fill in the caption in the braces of the \caption{} command. Put the label

1418: % that you will use with \ref{} command in the braces of the \label{} command.

1419: % Insert the column specifiers (l, r, c, d, etc.) in the empty braces of the

1420: % \begin{tabular}{} command.

1421: % The ruledtabular enviroment adds doubled rules to table and sets a

1422: % reasonable default table settings.

1423: % Use the table* environment to get a full-width table in two-column

1424: % Add \usepackage{longtable} and the longtable (or longtable*}

1425: % environment for nicely formatted long tables. Or use the the [H]

1426: % placement option to break a long table (with less control than 

1427: % in longtable).

1428: % \begin{table}%[H] add [H] placement to break table across pages

1429: % \caption{\label{}}

1430: % \begin{ruledtabular}

1431: % \begin{tabular}{}

1432: % Lines of table here ending with \\

1433: % \end{tabular}

1434: % \end{ruledtabular}

1435: % \end{table}

1436: % Surround table environment with turnpage environment for landscape

1437: % table

1438: % \begin{turnpage}

1439: % \begin{table}

1440: % \caption{\label{}}

1441: % \begin{ruledtabular}

1442: % \begin{tabular}{}

1443: % \end{tabular}

1444: % \end{ruledtabular}

1445: % \end{table}

1446: % \end{turnpage}

1447: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1448: %\newpage

1449: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1450: % Specify following sections are appendices. Use \appendix* if there

1451: % only one appendix.

1452: %\appendix

1453: %\section{}

1454: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1455: 

1456: \appendix

1457: \section{Interference and nonfragmentation contributions to the 

1458: cross sections}

1459: 

1460: In this appendix, we give the interference and nonfragmentation

1461: contributions to the differential cross section for $e^+ e^- \to V_1

1462: (\lambda_1) + V_2 (\lambda_2)$, where $V_1$ and $V_2$ are charmonium

1463: vector mesons, and $\lambda_1$ and $\lambda_2$ are their helicities. The

1464: differential cross section for the fragmentation contribution can be

1465: found in Eq.~(\ref{dsigmadx:VV}). The interference contribution to the

1466: differential cross section is

1467: %--------------

1468: \begin{eqnarray}

1469: %--------------

1470: \frac{d \sigma^\textrm{int}}{d x} [V_1(\lambda_1) + V_2(\lambda_2)]&=&

1471: -\frac{1024 \pi^3 \alpha^4 

1472:      g_{V_1\gamma} g_{V_2\gamma}

1473:     (4e_c^4m_{V_1}m_{V_2}

1474:     \langle O \rangle_{V_1}\langle O \rangle_{V_2})^{1/2}

1475:     \lambda^{1/2}(1,r_1^2, r_2^2)}

1476:      {3 s^5 r^2 r_1^2 r_2^2

1477:     (1-r_1^2-r_2^2)^2 [ 1-(1-\Delta) x^2]}\nonumber\\

1478: &&\qquad\times F^\textrm{int}_{\lambda_1,\lambda_2}(r_1,r_2,r,x),

1479: \label{sig-int}

1480: %--------------

1481: \end{eqnarray}

1482: %--------------

1483: where $r=4m_c/\sqrt{s}$. (Note that $r_1$ and $r_2$ reduce to $r/2$ in

1484: the nonrelativistic limit.) In Eq.~(\ref{sig-int}), the numerator

1485: factor in parentheses corresponds to the expression for $g_{V_1\gamma}

1486: g_{V_2\gamma}$ at leading order in $\alpha_s$ and $v$

1487: [Eq.~(\ref{g-NRQCD})]. Similar factors appear in subsequent equations

1488: for cross sections in this Appendix. The functions

1489: $F^\textrm{int}_{\lambda_1,\lambda_2}(r_1,r_2,r,x)$ are given by

1490: %--------------

1491: \begin{subequations}

1492: \label{inthelicity}

1493: \begin{eqnarray}

1494: %--------------

1495: F^\textrm{int}_{\pm 1, \pm 1}(r_1,r_2,r,x) &=& r^2 x^2(1-x^2) [

1496:       (1-r_1^2 -r_2^2)-\lambda(1,r_1^2,r_2^2) ], \\

1497: F^\textrm{int}_{\pm 1, \mp 1}(r_1,r_2,r,x) &=& (1-r_1^2 -r_2^2) (1-x^4), \\

1498: F^\textrm{int}_ {\pm 1, 0 }(r_1,r_2,r,x) &=& 

1499:    r r_2 [ (1-r_1^2 -r_2^2) (1-x^2)(1-2 x^2)

1500:     +4 r_1^2 x^4 ],\\

1501: F^\textrm{int}_{0, \pm 1}(r_1,r_2,r,x) &=& 

1502:    r r_1 [ (1-r_1^2 -r_2^2) (1-x^2)(1-2 x^2)

1503:     +4 r_2^2 x^4 ],\\

1504: F^\textrm{int}_{0, 0}(r_1,r_2,r,x) &=& 4 r_1 r_2 (1+ r^2) x^2 (1-x^2). 

1505: %--------------

1506: \end{eqnarray}

1507: \end{subequations}%

1508: %--------------

1509: The nonfragmentation contribution to the differential cross section is

1510: %--------------

1511: \begin{equation}

1512: %--------------

1513: \frac{d \sigma^\textrm{nf}}{d x} [V_1(\lambda_1) + V_2(\lambda_2)] =

1514: \frac{8192 \pi^3 \alpha^4 

1515:       (4e_c^4m_{V_1}m_{V_2}

1516:       \langle O \rangle_{V_1} \langle O \rangle_{V_2}) 

1517:       \lambda^{1/2}(1,r_1^2, r_2^2)

1518:       F^\textrm{nf}_{\lambda_1,\lambda_2}(r,x)}

1519:      {9 s^5 r^4 },

1520: %--------------

1521: \end{equation}

1522: %--------------

1523: where

1524: %--------------

1525: \begin{subequations}

1526: \label{nonfraghelicity}

1527: \begin{eqnarray}

1528: %--------------

1529: F^\textrm{nf}_{\pm 1, \pm 1}(r,x) &=& 2 r^4 x^2 (1-x^2), \\

1530: F^\textrm{nf}_{\pm 1, \mp 1}(r,x) &=& 2 (1-x^4), \\

1531: F^\textrm{nf}_{\pm 1, 0 }(r,x) &=& F^\textrm{nf}_{0, \pm 1}(r,x) =

1532:     r^2 (1-3 x^2 + 4 x^4 ), \\

1533: F^\textrm{nf}_{0, 0}(r,x) &=& 2 (1+ r^2)^2 x^2 (1-x^2). 

1534: %--------------

1535: \end{eqnarray}

1536: \end{subequations}%

1537: %--------------

1538: After summing over helicity states, one finds that the interference and 

1539: nonfragmentation contributions to the differential 

1540: cross section become

1541: %--------------

1542: \begin{eqnarray}

1543: %--------------

1544: \frac{d\sigma^{\textrm{int}}}{dx}(m_{V_1},m_{V_2},m_c)

1545: &=&

1546: -\frac{2048\pi^3\alpha^4g_{V_1\gamma} g_{V_2\gamma}

1547:    (4e_c^4m_{V_1}m_{V_2}

1548:    \langle O \rangle_{V_1} \langle O \rangle_{V_2})^{1/2}

1549:       \lambda^{1/2}(1,r_1^2,r_2^2)}

1550:        {3s^5 r^2 r_1^2 r_2^2 (1-r_1^2-r_2^2)^2

1551:         [1-(1-\Delta)x^2]}\nonumber\\

1552: &&\times\bigg\{

1553: 4r r_1 r_2(r_1+r_2)

1554: +(1-x^2)

1555: \big[ r^2(r_1+r_2)^2(1-r_1+r_2)(1+r_1-r_2)\nonumber\\

1556:  &&\qquad -r(r_1+r_2)(1-r_1^2+8r_1r_2-r_2^2)

1557: + 2(1 - r_1^2 + r_1r_2 - r_2^2)\big]\nonumber\\

1558: &&\qquad-(1-x^2)^2[1-r(r_1+r_2)]^2

1559: (1-r_1+r_2)(1+r_1-r_2)

1560: \bigg\}\label{int-hel-sum}

1561: %--------------

1562: \end{eqnarray}

1563: %--------------

1564: and

1565: \begin{eqnarray}

1566: \frac{d\sigma^{\textrm{nf}}}{dx}(m_{V_1},m_{V_2},m_c)

1567: &=&

1568: \frac{16384\pi^3\alpha^4

1569:       (4e_c^4m_{V_1}m_{V_2}

1570:       \langle O\rangle_{V_1}\langle O\rangle_{V_2})

1571:       \lambda^{1/2}(1,r_1^2,r_2^2)}

1572:        {9s^5r^4} \nonumber\\

1573:   &&\times \big[ 4r^2 + (1-x^2)(3r^4 -8r^2 + 5)

1574:      - 3(1-x^2)^2(1-r^2)^2 \big].\label{nonfrag-hel-sum}

1575: %--------------

1576: \end{eqnarray}

1577: %--------------

1578: 

1579: 

1580: 

1581: 

1582: \begin{acknowledgments}

1583: % put your acknowledgments here.

1584: \noindent

1585: We thank Michael Peskin and Kai Yi for helpful discussions.

1586: Work by G.~T.~Bodwin in the

1587: High Energy Physics Division at Argonne National Laboratory is supported

1588: by the U.~S.~Department of Energy, Division of High Energy Physics, under

1589: Contract No.~W-31-109-ENG-38. 

1590: The work of E.~Braaten was supported in part by

1591: the U.~S.~Department of Energy, Division of High Energy Physics, 

1592: under grant No.~DE-FG02-91-ER4069.

1593: The work of J.~Lee was supported by the Second Brain Korea 21 Project and

1594: the Basic Research Program of the Korea Science and Engineering

1595: Foundation (KOSEF) under grant No.~R01-2005-000-10089-0.

1596: The work of C.~Yu was supported by the

1597: Korea Research Foundation Grant funded by Korea Government 

1598: (MOEHRD, Basic Research Promotion FUND) (KRF-2005-075-C00008).

1599: \end{acknowledgments}

1600: 

1601: 

1602: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1603: % Create the reference section using BibTeX:

1604: 

1605: \begin{thebibliography}{}

1606: 

1607: %----------------------------------------------------------------------

1608: %\cite{Bodwin:2002fk}

1609: \bibitem{Bodwin:2002fk}

1610:   G.~T.~Bodwin, J.~Lee, and E.~Braaten,

1611:   %``e+ e- annihilation into J/psi + J/psi,''

1612:   Phys.\ Rev.\ Lett.\  {\bf 90}, 162001 (2003)

1613:   [Erratum-ibid.\ {\bf 95}, 239901 (2005)]

1614:   [arXiv:hep-ph/0212181].

1615:   %%CITATION = HEP-PH 0212181;%%

1616: 

1617: %----------------------------------------------------------------------

1618: %\cite{Bodwin:2002kk}

1619: \bibitem{Bodwin:2002kk}

1620:   G.~T.~Bodwin, J.~Lee, and E.~Braaten,

1621:   % ``Exclusive double-charmonium production from e+ e- annihilation into two

1622:   % virtual photons,''

1623:   %

1624:   Phys.\ Rev.\ D {\bf 67}, 054023 (2003)

1625:   [Erratum-ibid.\ D {\bf 72}, 099904 (2005)]

1626:   [arXiv:hep-ph/0212352].

1627:   %%CITATION = HEP-PH 0212352;%%

1628: 

1629: %----------------------------------------------------------------------

1630: %\cite{Braaten:2002fi}

1631: \bibitem{Braaten:2002fi}

1632:   E.~Braaten and J.~Lee,

1633:   % ``Exclusive double-charmonium production in e+ e- annihilation,''

1634:   %

1635:   Phys.\ Rev.\ D {\bf 67}, 054007 (2003)

1636:   [Erratum-ibid.\ D {\bf 72}, 099901 (2005)]

1637:   [arXiv:hep-ph/0211085].

1638:   %%CITATION = HEP-PH 0211085;%%

1639: 

1640: %----------------------------------------------------------------------

1641: %\cite{Collins:1989gx}

1642: \bibitem{Collins:1989gx}

1643:   See, for example, J.~C.~Collins, D.~E.~Soper and G.~Sterman,

1644:   %``Factorization Of Hard Processes In QCD,''

1645:   Adv.\ Ser.\ Direct.\ High Energy Phys.\  {\bf 5}, 1 (1988)

1646:   [arXiv:hep-ph/0409313] and references therein.

1647:   %%CITATION = HEP-PH 0409313;%%

1648: 

1649: %----------------------------------------------------------------------

1650: %\cite{Pire:2005ic}

1651: \bibitem{Pire:2005ic}

1652:   B.~Pire, L.~Szymanowski, and S.~Wallon,

1653:   %``Double diffractive rho-production in gamma* gamma* collisions,''

1654:   Eur.\ Phys.\ J.\ C {\bf 44}, 545 (2005)

1655:   [arXiv:hep-ph/0507038].

1656:   %%CITATION = HEP-PH 0507038;%%

1657: %----------------------------------------------------------------------

1658: %\cite{Enberg:2005eq}

1659: \bibitem{Enberg:2005eq}

1660:   R.~Enberg, B.~Pire, L.~Szymanowski, and S.~Wallon,

1661:   %``BFKL resummation effects in gamma* gamma* --> rho rho,''

1662:   Eur.\ Phys.\ J.\ C {\bf 45}, 759 (2006)

1663:   [arXiv:hep-ph/0508134].

1664:   %%CITATION = HEP-PH 0508134;%%

1665: %----------------------------------------------------------------------

1666: %\cite{Ivanov:2005gn}

1667: \bibitem{Ivanov:2005gn}

1668:   D.~Y.~Ivanov and A.~Papa,

1669:   % ``Electroproduction of two light vector mesons in the next-to-leading

1670:   %approximation,''

1671:   Nucl.\ Phys.\ B {\bf 732}, 183 (2006)

1672:   [arXiv:hep-ph/0508162].

1673:   %%CITATION = HEP-PH 0508162;%%

1674: %----------------------------------------------------------------------

1675: %\cite{Pire:2006ik}

1676: \bibitem{Pire:2006ik}

1677:   B.~Pire, M.~Segond, L.~Szymanowski, and S.~Wallon,

1678:   %``QCD factorizations in gamma* gamma* --> rho0(L) rho0(L),''

1679:   Phys.\ Lett.\ B {\bf 639}, 642 (2006)

1680:   [arXiv:hep-ph/0605320].

1681:   %%CITATION = HEP-PH 0605320;%%

1682: 

1683: %----------------------------------------------------------------------

1684: \bibitem{PDG2006}

1685: W.-M. Yao \textit{et al.} [Particle Data Group],

1686: J.\ of Phys.\ G \textbf{33}, 1 (2006).

1687: 

1688: %----------------------------------------------------------------------

1689: \bibitem{Akhmetshin:2003zn}

1690:   R.~R.~Akhmetshin {\it et al.}  [CMD-2 Collaboration],

1691:   %``Reanalysis of hadronic cross section measurements at CMD-2,''

1692:   Phys.\ Lett.\ B {\bf 578}, 285 (2004)

1693:   [arXiv:hep-ex/0308008].

1694:   %%CITATION = HEP-EX 0308008;%%

1695: %----------------------------------------------------------------------

1696: %\cite{Davier:2006fu}

1697: \bibitem{Davier:2006fu}

1698:   M.~Davier, M.~Peskin, and A.~Snyder,

1699:   %``Two-photon exchange model for production of neutral meson pairs in e+ e-

1700:   % annihilation,''

1701:   arXiv:hep-ph/0606155.

1702:   %%CITATION = HEP-PH 0606155;%%

1703: %----------------------------------------------------------------------

1704: \bibitem{Akhmetshin:2001ig}

1705:   R.~R.~Akhmetshin {\it et al.}  [CMD-2 Collaboration],

1706:   %``Measurement of e+ e- --> pi+ pi- cross section with CMD-2 around rho

1707:   %meson,''

1708:   Phys.\ Lett.\ B {\bf 527}, 161 (2002)

1709:   [arXiv:hep-ex/0112031].

1710:   %%CITATION = HEP-EX 0112031;%%

1711: %----------------------------------------------------------------------

1712: %\cite{Achasov:2006vp}

1713: \bibitem{Achasov:2006vp}

1714:   M.~N.~Achasov {\it et al.},

1715:    %``Update of the e+ e- --> pi+ pi- cross section measured by SND detector in

1716:    %  the energy region 400-MeV < s**(1/2) < 1000-MeV,''

1717:   arXiv:hep-ex/0605013.

1718:   %%CITATION = HEP-EX 0605013;%%

1719: 

1720: %----------------------------------------------------------------------

1721: %\cite{Bodwin:1994jh}

1722: \bibitem{Bodwin:1994jh}

1723:   G.~T.~Bodwin, E.~Braaten, and G.~P.~Lepage,

1724:   % ``Rigorous QCD analysis of inclusive annihilation and production of heavy

1725:   % quarkonium,''

1726:   %

1727:   Phys.\ Rev.\ D {\bf 51}, 1125 (1995)

1728:   [Erratum-ibid.\ D {\bf 55}, 5853 (1997)]

1729:   [arXiv:hep-ph/9407339].

1730:   %%CITATION = HEP-PH 9407339;%%

1731: 

1732: %----------------------------------------------------------------------

1733: %\cite{Bodwin:2006dn}

1734: \bibitem{Bodwin:2006dn}

1735:   G.~T.~Bodwin, D.~Kang, and J.~Lee,

1736:   %``Potential-model calculation of an order-v**2 NRQCD matrix element,''

1737:   Phys.\ Rev.\ D {\bf 74}, 014014 (2006)

1738:   [arXiv:hep-ph/0603186].

1739:   %%CITATION = HEP-PH 0603186;%%

1740: 

1741: %----------------------------------------------------------------------

1742: \bibitem{kang}

1743: G.~T.~Bodwin, D.~Kang, and J.~Lee, in preparation.

1744: 

1745: %----------------------------------------------------------------------

1746: %\cite{Aubert:2006we}

1747: \bibitem{Aubert:2006we}

1748:   B.~Aubert  [BABAR Collaboration],

1749:   % ``Observation of e+ e- annihilations into the C = +1 hadronic final states

1750:   % rho0 rho0 and Phi rho0,''

1751:   %

1752:   Phys.\ Rev.\ Lett.\  {\bf 97}, 112002 (2006)

1753:   [arXiv:hep-ex/0606054].

1754:   %%CITATION = HEP-EX 0606054;%%

1755: 

1756: %----------------------------------------------------------------------

1757: %\cite{Eidelman:2004wy}

1758: \bibitem{Eidelman:2004wy}

1759:   S.~Eidelman {\it et al.}  [Particle Data Group],

1760:   %``Review of particle physics,''

1761:   Phys.\ Lett.\ B {\bf 592}, 1 (2004).

1762:   %%CITATION = PHLTA,B592,1;%%

1763: 

1764: %----------------------------------------------------------------------

1765:   %\cite{Luchinsky:2003yh}

1766: \bibitem{Luchinsky:2003yh}

1767:   A.~V.~Luchinsky,

1768:    %``On double 1-- charmonium production through two-photon e+ e-  annihilation

1769:    %  at s**(1/2) = 10.6-GeV,''

1770:   arXiv:hep-ph/0301190.

1771:   %%CITATION = HEP-PH 0301190;%%

1772: 

1773: %----------------------------------------------------------------------

1774: %\cite{Abe:2004ww}

1775: \bibitem{Abe:2004ww}

1776:   K.~Abe {\it et al.}  [Belle Collaboration],

1777:   % ``Study of double charmonium production in e+ e- annihilation at s**(1/2)

1778:   % approx. 10.6-GeV,''

1779:   %

1780:   Phys.\ Rev.\ D {\bf 70}, 071102 (2004)

1781:   [arXiv:hep-ex/0407009].

1782:   %%CITATION = HEP-EX 0407009;%%

1783: 

1784: \end{thebibliography}

1785: 

1786: \end{document}

1787: