1: \documentclass[12pt]{article}

2: \usepackage{epsfig}

3: 

4: 

5: \bibliographystyle{unsrt}    % for BibTeX - sorted numerical labels by order of

6: 

7: \tolerance=10000

8: \pagenumbering{arabic}

9: \textheight 22.cm

10: \textwidth 16.2 cm

11: \oddsidemargin 0.5cm

12: \evensidemargin 0.5cm

13: \topmargin=-1.cm

14: \hoffset -0.5cm

15: 

16: \newcommand{\be}{\begin{eqnarray}}

17: \newcommand{\ee}{\end{eqnarray}}

18: 

19: \begin{document}

20: 

21: 

22: \bibliographystyle{unsrt}

23: \footskip 1.0cm

24: 

25: 

26: %\pagestyle{empty}  % no page number at all

27: \thispagestyle{empty}

28: \begin{flushright}

29: INT--PUB 04--03

30: \end{flushright}

31: \vspace{0.3in}

32: 

33: \begin{center}{\Large \bf {Forward Rapidity Hadron Production 

34: in Deuteron Gold Collisions from Valence Quarks}}\\

35: 

36: \vspace{1.8in}

37: {\large  Jamal Jalilian-Marian}\\

38: 

39: \vspace{.2in}

40: {\it Institute for Nuclear Theory, University of Washington, 

41: Seattle, WA 98195\\}

42: 

43: \end{center}

44: 

45: \vspace*{25mm}

46: 

47: 

48: \begin{abstract}

49: 

50: \noindent We consider hadron production in deuteron gold collisions at 

51: RHIC in the forward rapidity region. Treating the target nucleus as 

52: a Color Glass Condensate and the projectile deuteron as a dilute

53: system of valence quarks, we obtain good agreement with the BRAHMS

54: minimum bias data on charged hadron production in the forward rapidity 

55: ($y=3.2$) and low $p_t$ region. We provide predictions for neutral 

56: pion production in minimum bias deuteron gold collisions in the 

57: forward rapidity region, $y=3.8$, measured by the STAR collaboration 

58: at RHIC.

59: 

60: \end{abstract}

61: \newpage

62: 

63: \section{Introduction}

64: 

65: The recent observation of the suppression of the charged hadron spectra

66: in the forward rapidity region \cite{brahms} at RHIC, and its centrality 

67: dependence has generated a lot of interest and excitement in the high 

68: energy heavy ion community. The suppression of particle spectra and 

69: disappearance of the Cronin effect at forward rapidity had been 

70: predicted \cite{kkt,aaksw} in the Color Glass Condensate formalism 

71: \cite{cgc}, unlike the 

72: more conventional models which predicted a stronger enhancement of

73: the spectra in the forward rapidity as compared with mid rapidity \cite{iv}.

74: The forward rapidity data strongly suggest that the high gluon density 

75: region of QCD phase space, the Color Glass Condensate, has been 

76: observed at RHIC. While the Color Glass Condensate has been successful 

77: in predicting some global features of the data, such

78: as multiplicities and their energy, rapidity and centrality dependence

79: in gold gold and deuteron gold collisions \cite{kln} as well as 

80: qualitative predictions \cite{kkt,aaksw} of the suppression of the 

81: hadron spectra and their centrality dependence, there has not been a 

82: quantitative analysis of the forward rapidity hadron spectra using the

83: Color Glass Condensate formalism. Here, for the first time, we provide a 

84: limited, but quantitative analysis of the low $p_t$, forward rapidity 

85: RHIC data, using the Color Glass Condensate formalism.

86: 

87: 

88: As emphasized in \cite{adjjm}, the forward rapidity region at RHIC

89: is the best kinematic region to look for the signatures of the Color

90: Glass Condensate since this is the region where one probes the smallest

91: $x$ in the target nucleus so that the Color Glass Condensate will be

92: manifest more strongly in this kinematic region. Also, forward rapidity 

93: deuteron gold collisions are ideal since one does not have the final 

94: state (Quark Gluon Plasma) interactions, present in mid rapidity heavy 

95: ion collisions. 

96: 

97: It is important to note that at very forward rapidities ($y > 3$), one

98: is probing the small $x$ region of the nucleus and the {\it large} $x$

99: region of the deuteron projectile. For example, at $y \sim 3.2$ and 

100: $p_t \sim 2$, the deuteron wave function at $x \sim 0.25$ is probed. 

101: This is the region where valence quarks dominate over gluons and sea 

102: quarks in the deuteron wave function while the relevant $x$ for the 

103: target nucleus is $x \sim 4\times 10^{-4}$ so that gluons are the 

104: dominant parton species in the target nucleus. 

105: 

106: In this work, we concentrate in the very forward

107: rapidity region ($y \ge 3.2$) so that the target nucleus is treated as

108: a Color Glass Condensate while only the valence quarks in the projectile 

109: deuteron are included. We use the results of \cite{adjjm} for the scattering

110: of valence quarks on a Color Glass Condensate to calculate charged hadron

111: and neutral pion $p_t$ spectra as well as the nuclear modification factor

112: $R_{dA}$ in the forward rapidity region at RHIC.

113: 

114: 

115: \section{Scattering of quarks on a Color Glass Condensate}

116: 

117: In \cite{adjjm}, scattering of quark on a target described as a Color

118: Glass Condensate was considered and the scattering cross section was

119: calculated. The incoming quark is taken to be massless and carries zero

120: transverse momentum. The scattering cross section is given by

121: 

122: \begin{eqnarray}

123: q^- {d\sigma^{qA\rightarrow qX} \over d^2q_t dq^-} =- {1 \over(2\pi)^2}\,

124: q^- \, \delta(q^--p^-) \int d^2b_t \, d^2 r_t e^{i q_t\cdot r_t}

125: \sigma_{dipole}(r_t,b_t,x_g)

126: \label{eq:qAcs}

127: \end{eqnarray}

128: 

129: Here, $p^-$ is the light cone energy of the incoming quark while $q^-$

130: is the light cone energy of the outgoing quark with transverse momentum

131: $q_t$ and  $\sigma_{dipole}$ is the cross section for scattering of a 

132: quark anti-quark dipole on a target described as a Color Glass Condensate.

133: The dipole cross section satisfies the non-linear JIMWLK \cite{jimwlk}

134: (BK \cite{bk} at large $N_c$)

135: evolution equation. To relate this quark-nucleus (proton) target 

136: scattering cross section to hadron production in deuteron gold 

137: collisions, we convolute this

138: cross section with the quark distribution function in a deuteron and 

139: quark-hadron fragmentation function. We get  

140: \begin{eqnarray}

141: &&{d\sigma^{dA\rightarrow h(y,k_t)X} \over dyd^2k_t}= - 

142: {1\over (2\pi)^2}\sqrt{k_t^2 \over s}\,e^y\,\int_{z_{min}}^1 dz\,

143: q_d\left(x_q\right)\,D_{q/h}(z) \nonumber\\

144: &&\qquad\times 

145: \int d^2 b_t \, d^2 r_t e^{i k_t \cdot r_t/z}

146: \sigma_{dipole}(r_t,b_t,x_g)

147: \label{eq:csrap}

148: \end{eqnarray}

149: where we have used the following kinematical relations $x_q=k_\perp

150: e^y/z\sqrt{s}$, $x_g=k_\perp e^{-y}/z\sqrt{s}$, $z_{\rm min}= k_\perp

151: e^y/\sqrt{s}$ ($\sqrt{s}$ is the center of mass energy) while $y$ and

152: $k_t$ are the rapidity and transverse momentum of the measured hadron.

153: Both the quark distribution and fragmentation functions depend on a 

154: factorization scale $Q_f^2$ which is not written out explicitly. In our

155: calculation, we set $Q_f = k_t$ where $k_t$ is the transverse momentum

156: of the observed particle. Eq. (\ref{eq:csrap}) is the formula used in 

157: this work to calculate the hadron spectra at forward rapidities in 

158: deuteron gold collisions.

159: 

160: We use the LO GRV98 quark distribution function \cite{grv98} and the 

161: LO KKP quark hadron fragmentation function \cite{kkp}. It should be noted 

162: that there is no available fragmentation function for negatively charged 

163: hadrons so therefore we use

164: the fragmentation function for $(h^+ + h^ -)/2$. However, in the

165: transverse momentum range considered here ($p_t < 2 GeV$), this should 

166: not make a sizable difference. Since we are sensitive only to the 

167: large $x_q$ ($ \sim 0.2-0.5$) region of the deuteron wave function, there

168: is practically no sensitivity to the choice of the quark distribution 

169: function. Also, the nuclear modification of valence quarks 

170: in the deuteron wave function is minimal in this $x_q$ range and is 

171: therefore neglected here.

172: 

173: To proceed further, we need to know the dipole cross section 

174: $\sigma_{dipole}$ which satisfies the JIMWLK equation. This is a 

175: very complicated functional equation \cite{hw} which simplifies 

176: in the large $N_c$ limit, known as the BK equation. The BK equation 

177: for the dipole cross section has been numerically solved by various 

178: people in different limits \cite{solbk}. Iancu, Itakura and

179: Munier proposed a parameterization of the dipole cross section which 

180: has all the properties of the solution to the BK equation and used it 

181: to fit the HERA data on the proton structure function $F_2$ \cite{iim}. 

182: This is 

183: a very simple and economical parameterization (it basically has three 

184: free parameters, in addition to 

185: the light quark mass) which does an excellent job of describing the HERA 

186: data at $x < 0.01$. Therefore, we use this parameterization in this work. 

187: The dipole cross section is given by 

188: 

189: \be

190: \int d^2 b_t \, \sigma_{dipole}(x_g,b_t,r_t) \equiv 

191: 2 \pi R^2  \,\,{\cal N} \,(x_g, r_t Q_s)

192: \label{eq:cs_iim}

193: \ee

194: where 

195: \be

196: {\cal N} \,(x_g, r_t Q_s) = 1 - e^{-a \ln^2 b\,r_t Q_s}  \,\,\,\,\,\,\,\,\,\,

197: r_t Q_s > 2 

198: \nonumber

199: \ee

200: and 

201: \be

202: {\cal N} \,(x_g, r_t Q_s) = {\cal N}_0 \exp \Bigg\{ 2 \ln ({r_t Q_s \over 2})

203: \bigg [\gamma_s + {\ln 2/r_t Q_s \over \kappa \lambda \ln 1/x_g} \bigg] \Bigg\}

204: \,\,\,\,\,\,\,\,\,\,

205: r_t Q_s < 2 

206: \label{eq:cs_param}

207: \ee

208: The constants $a, b$ are determined by matching the solutions at $r_t Q_s =2$

209: and $\gamma_s = 0.63$ and $\kappa = 9.9$ are determined from LO BFKL.

210: The form of the proton saturation scale $Q_s^2$ is taken to be 

211: $Q_s^2 \equiv (x_0/x)^{\lambda} \, GeV^2$ with $x_0, \lambda, {\cal N}_0$

212: determined from fitting the HERA data on proton structure function $F_2$.

213: We refer the reader to \cite{iim} for details of the fit. In case of a

214: nucleus, we have $R_{A} = 1.1 \, A^{1/3} R_{p}$ and 

215: use the value of the minimum bias saturation scale given

216: in \cite{kln} ($Q^2_{s, {\it min \, bias}} = 0.95 \,GeV^2$ at $x = 0.01$), 

217: with the same $x$ dependence as in a proton given above.

218: 

219: In Fig. (\ref{fig:cs_pp_32}), we show our results for charged hadron 

220: production at $y=3.2$ in proton proton collisions in

221: arbitrary units while in Fig. (\ref{fig:cs_dA_32}), our results

222: for charged hadron production in deuteron gold collisions is shown. 

223: Both figures are for minimum bias events. The proton-proton cross section

224: is normalized to the data at $k_t = 1.1$ GeV by multiplying by a $K$

225: factor of $2.57$. To get the normalization of the $dA$ spectra, we again 

226: multiply by a $K$ factor which is slightly less than the $K=1.8$ factor 

227: for the proton-proton case. This is not unreasonable since Next to Leading 

228: Order corrections are typically large at low $k_t$ and that the $K$ factors

229: used can be different for nuclei \cite{ina}. However, the ratio $R_dA$ is 

230: calculated {\it without} any $K$ factor. The agreement with 

231: the slope of the spectra for $k_t < 2$ GeV is quite reasonable specially 

232: since there are {\it no free parameters} in this calculation. 

233: 

234: \begin{figure}[htp]

235: \centering

236: \setlength{\epsfxsize=16cm}

237: \centerline{\epsffile{pp_ch_32.eps}}

238: \caption{The invariant yield of charged hadrons at $y=3.2$ at RHIC in 

239: proton-proton collisions at $\sqrt{s}=200$ GeV. The normalization 

240: is a fit to the data.}

241: \label{fig:cs_pp_32}

242: \end{figure}

243: 

244: \begin{figure}[htp]

245: \centering

246: \setlength{\epsfxsize=16cm}

247: \centerline{\epsffile{dA_ch_32.eps}}

248: \caption{The invariant yield of charged hadrons at $y=3.2$ at RHIC in 

249: minimum bias deuteron-nucleus collisions at $\sqrt{s}=200$ GeV.}

250: \label{fig:cs_dA_32}

251: \end{figure}

252: 

253: In Fig. (\ref{fig:R_dA_32}), we show the nuclear modification factor $R_{dA}$ 

254: for hadron production in deuteron gold collisions at $y=3.2$. We define

255: \be

256: R_{dA}\equiv {{d\sigma^{dA \rightarrow h X} \over dy d^2k_t} \over

257: 2\,A\, {d\sigma^{pp \rightarrow h X} \over dy d^2k_t}}

258: \label{eq:R_dA}

259: \ee

260: Again, the agreement with the data at low $k_t$ is quite good but the higher 

261: $k_t$ points start to show a deviation. This is discussed in more detail 

262: later. We emphasis the point that the ratio $R_dA$ is calculated without

263: multiplying by any $K$ factor and with no free parameters.

264: 

265: \begin{figure}[hbp]

266: \centering

267: \setlength{\epsfxsize=16cm}

268: \centerline{\epsffile{R_dA_ch_32.eps}}

269: \caption{The nuclear modification factor for charged hadrons at $y=3.2$ 

270: in minimum bias deuteron-nucleus collisions at RHIC, $\sqrt{s}=200$ GeV.}

271: \label{fig:R_dA_32}

272: \end{figure}

273: 

274: It should be noted that the BRAHMS data is for negatively charged

275: hadrons while our calculations are done for the average of positively

276: and negatively charged hadrons since fragmentation functions for

277: negatively charged hadrons are not available. Recently, 

278: Guzey et al. \cite{guzey} investigated the dependence of the suppression

279: and showed that isospin symmetry considerations can make a huge effect

280: on the observed suppression. This effect is most pronounced in the 

281: $k_t > 2$ GeV and can affect our results by $(15-20) \%$ in the kinematic

282: region we cover.

283: 

284: Finally, since the STAR collaboration has measured neutral pions at rapidity

285: $y=3.8$ in deuteron gold collisions, we show our predictions for neutral

286: pion nuclear modification factor $R_{dA}$ at $y=3.8$ in 

287: Fig. (\ref{fig:R_dA_38}). A slightly stronger suppression of 

288: $R_{dA}$ is seen at lowest $k_t$ as compared to charged hadrons at $y=3.2$.

289: 

290: \begin{figure}[hbp]

291: \centering

292: \setlength{\epsfxsize=16cm}

293: \centerline{\epsffile{R_dA_pi0_38.eps}}

294: \caption{The nuclear modification factor for neutral pions at $y=3.8$ 

295: in minimum bias deuteron-nucleus collisions at RHIC, $\sqrt{s}=200$ GeV.}

296: \label{fig:R_dA_38}

297: \end{figure}

298: 

299: 

300: \section{Discussion}

301: 

302: In deuteron gold collisions in the forward rapidity and low $k_t$ region 

303: at RHIC considered in this work, the valence quarks are the most abundant 

304: parton species in the deuteron. They scatter on the target nucleus, 

305: which has its wave function fully developed (evolved in $x_g$ as much as

306: allowed by the kinematics) and is characterized by the nucleus saturation 

307: scale $Q_s^A(x_g)$. The valence quark gets a transverse momentum kick of 

308: order $Q_s^A$ and is then ``produced''. 

309: 

310: To get a rough idea of the scales involved and estimate where our approach

311: should break down, we note that hadrons carry $z$ fraction of the parent 

312: parton energy and that $<z> \sim 0.7-0.8$ in this rapidity (for hadron 

313: production in $pp$ collisions). The saturation scale of the nucleus is around 

314: $Q_s^A \sim 1.5 - 1.8$ GeV (minimum bias) at rapidity of $y=3.2$.

315: The geometric scaling region \cite{ks,iimc} extends to a little higher 

316: momentum $Q_{es} \equiv Q^2_s/Q_{s0}$ and can be as high as $2.5$ GeV. 

317: This means that our formalism should describe hadron

318: production in {\it minimum bias} deuteron gold collisions in the forward 

319: rapidity region up to $k_t \sim z \, Q_{es} \sim 2$ GeV. We emphasize

320: that these estimates are for {minimum bias} events only and the Color

321: Glass Condensate formalism is

322: expected to be valid at higher $k_t$ for more central collisions.

323: 

324: As one goes to higher $k_t$, gluon radiation becomes important \cite{ykam} 

325: and will eventually dominate the hadron production cross section. This has 

326: not been included here and would presumably improve the high $k_t$ 

327: behavior of the spectra. Another caveat of our approach is the use of 

328: the dipole parameterization advocated in \cite{iim}. This parameterization 

329: does not have the Cronin effect \cite{gjm,jmnr,bkw}. This may be partly 

330: responsible for the deviation of calculated $R_{dA}$ from the data.

331: Also, this parameterization does not have the right high $k_t$ behavior

332: since the double log limit is not built into it. This is the reason for 

333: the decrease of the $R_{dA}$ at higher $k_t$ which is not expected from pQCD.  

334: However, none of the other simple parameterizations of the dipole 

335: model \cite{gbw} which have been used to fit the HERA data have the 

336: right (BFKL) anomalous dimension. Therefore, we use this parameterization

337: because it has the right anomalous dimension and we are staying in a limited

338: kinematic region and since it is experimentally known that the Cronin effect 

339: goes away in the forward rapidity region. 

340: 

341: In order to extend this formalism to mid rapidity, one needs to include

342: gluon production as considered in \cite{ykam}. The contribution of gluons

343: to hadron production in mid rapidity is more important than the

344: contribution of valence quarks since at mid rapidity, one probes small

345: values of $x$ in the deuteron wave function where there are a lot more

346: gluons than quarks. Nevertheless, a practical problem with inclusion of 

347: gluons is that there is no available parameterization of the gluon-gluon 

348: dipole cross section which has been tested in other processes unlike

349: the quark anti-quark dipole cross section which is probed in DIS processes. 

350: 

351: An important observable which has not been considered here is the

352: centrality dependence of the hadron suppression factor in the forward

353: rapidity region, recently shown by the BRAHMS collaboration in the

354: Quark Matter $2004$ \cite{brahms}. It will be very interesting to see whether

355: our formalism can also describe the centrality dependence of the 

356: data. This would require the application of the Monte Carlo Glauber 

357: method to our formalism since this is the method used by experimentalists to

358: extract centrality dependence of the data. This is beyond the scope of 

359: this work and will be pursued later.

360: 

361: 

362: The agreement of our calculations with the forward rapidity data 

363: becomes even more significant due to the fact we have not used any

364: free parameters (for $R_{dA}$ and the slope of the spectra) in this 

365: calculation. All the necessary ingredients,

366: the dipole cross section (for a proton) and the value of the nucleus 

367: saturation scale (for minimum bias events) have already been known

368: for a while and used by various authors \cite{kln,iim}. The fact that 

369: our simple and {\it parameter free} calculation based on the Color 

370: Glass Condensate formalism can reproduce the experimental data at 

371: low $k_t$ is a strong indication that the physics of forward rapidity 

372: region at RHIC is that of high gluon density QCD and the Color Glass 

373: Condensate.

374: 

375: 

376: 

377: 

378: \leftline{\bf Acknowledgments} 

379: 

380: We would like to thank F. Gelis, Y. Kovchegov and R. Venugopalan for useful 

381: discussions. We also thank F. Gelis for providing us with his Fortran code

382: for Fourier transforming the dipole cross section. This work is supported by 

383: DOE under grant number DOE/ER/41132.

384: 

385: \leftline{\bf References}

386: 

387: 

388: \renewenvironment{thebibliography}[1]

389:         {\begin{list}{[$\,$\arabic{enumi}$\,$]}  % {\arabic{enumi}.}

390:         {\usecounter{enumi}\setlength{\parsep}{0pt}

391:          \setlength{\itemsep}{0pt}  \renewcommand{\baselinestretch}{1.2}

392:          \settowidth

393:         {\labelwidth}{#1 ~ ~}\sloppy}}{\end{list}}

394: 

395: \begin{thebibliography}{99}

396: 

397: \bibitem{brahms}

398: R. Debbe for BRAHMS collaboration, Quark matter 2004 and DNP 2003, 

399: nucl-ex/0403005.

400: 

401: \bibitem{kkt}

402: D.~Kharzeev, Y.~V.~Kovchegov and K.~Tuchin.

403: Phys.\ Rev.\ D {\bf 68}, 094013 (2003). 

404: %%CITATION = HEP-PH 0307037;%%

405: 

406: \bibitem{aaksw}

407: J.~L.~Albacete, N.~Armesto, A.~Kovner, C.~A.~Salgado and U.~A.~Wiedemann, 

408: hep-ph/0307179.

409: %%CITATION = HEP-PH 0307179;%%

410: 

411: \bibitem{cgc}

412: L.~V.~Gribov, E.~M.~Levin and M.~G.~Ryskin,

413: Phys.\ Rept.\  {\bf 100} (1983) 1; 

414: %%CITATION = PRPLC,100,1;%%

415: A.~H.~Mueller and J.~W.~Qiu,

416: Nucl.\ Phys.\ B {\bf 268}, 427 (1986);

417: %%CITATION = NUPHA,B268,427;%%

418: L.~D.~McLerran and R.~Venugopalan,

419: Phys.\ Rev.\ D {\bf 49}, 2233 (1994), 

420: %%CITATION = HEP-PH 9309289;%%

421: Phys.\ Rev.\ D {\bf 49}, 3352 (1994), 

422: %%CITATION = HEP-PH 9311205;%%

423: Phys.\ Rev.\ D {\bf 59}, 094002 (1999); 

424: %%CITATION = HEP-PH 9809427;%%

425: Y.~V.~Kovchegov,

426: Phys.\ Rev.\ D {\bf 54}, 5463 (1996), 

427: %%CITATION = HEP-PH 9605446;%%

428: Phys.\ Rev.\ D {\bf 55}, 5445 (1997).

429: %%CITATION = HEP-PH 9701229;%%

430: 

431: \bibitem{iv}

432: I.~Vitev,

433: Phys.\ Lett.\ B {\bf 562}, 36 (2003).

434: %%CITATION = NUCL-TH 0302002;%%

435: 

436: \bibitem{kln}

437: D.~Kharzeev, E.~Levin and M.~Nardi, hep-ph/0111315, 

438: %%CITATION = HEP-PH 0111315;%%

439: Nucl.\ Phys.\ A {\bf 730}, 448 (2004); 

440: %%CITATION = HEP-PH 0212316;%%

441: J.~Jalilian-Marian, nucl-th/0212018.

442: %%CITATION = NUCL-TH 0212018;%%

443: 

444: \bibitem{adjjm}

445: A.~Dumitru and J.~Jalilian-Marian,

446: Phys.\ Lett.\ B {\bf 547}, 15 (2002), 

447: %%CITATION = HEP-PH 0111357;%%

448: Phys.\ Rev.\ Lett.\  {\bf 89}, 022301 (2002).

449: %%CITATION = HEP-PH 0204028;%%

450: 

451: \bibitem{jimwlk}

452: A.~Ayala, J.~Jalilian-Marian, L.~D.~McLerran and R.~Venugopalan,

453: Phys.\ Rev.\ D {\bf 52}, 2935 (1995),

454: %%CITATION = HEP-PH 9501324;%%

455: Phys.\ Rev.\ D {\bf 53}, 458 (1996); 

456: %%CITATION = HEP-PH 9508302;%%

457: J.~Jalilian-Marian, A.~Kovner, L.~D.~McLerran and H.~Weigert,

458: Phys.\ Rev.\ D {\bf 55}, 5414 (1997); 

459: %%CITATION = HEP-PH 9606337;%%

460: J.~Jalilian-Marian, A.~Kovner, A.~Leonidov and H.~Weigert,

461: Nucl.\ Phys.\ B {\bf 504}, 415 (1997), 

462: %%CITATION = HEP-PH 9701284;%%

463: Phys.\ Rev.\ D {\bf 59}, 014014 (1999),

464: %%CITATION = HEP-PH 9706377;%%

465: Phys.\ Rev.\ D {\bf 59}, 034007 (1999),

466: [Erratum-ibid.\ D {\bf 59}, 099903 (1999)];

467: %%CITATION = HEP-PH 9807462;%%

468: J.~Jalilian-Marian, A.~Kovner and H.~Weigert,

469: Phys.\ Rev.\ D {\bf 59}, 014015 (1999);

470: %%CITATION = HEP-PH 9709432;%%

471: A.~Kovner, J.~G.~Milhano and H.~Weigert,

472: Phys.\ Rev.\ D {\bf 62}, 114005 (2000);

473: %%CITATION = HEP-PH 0004014;%%

474: A.~Kovner and J.~G.~Milhano,

475: Phys.\ Rev.\ D {\bf 61}, 014012 (2000);

476: %%CITATION = HEP-PH 9904420;%%

477: E.~Iancu, A.~Leonidov and L.~D.~McLerran,

478: Nucl.\ Phys.\ A {\bf 692}, 583 (2001),

479: %%CITATION = HEP-PH 0011241;%%

480: hep-ph/0202270, 

481: %%CITATION = HEP-PH 0202270;%%

482: Phys.\ Lett.\ B {\bf 510}, 133 (2001); 

483: %%CITATION = HEP-PH 0102009;%%

484: E.~Iancu and L.~D.~McLerran,

485: Phys.\ Lett.\ B {\bf 510}, 145 (2001);

486: %%CITATION = HEP-PH 0103032;%%

487: E.~Ferreiro, E.~Iancu, A.~Leonidov and L.~McLerran,

488: Nucl.\ Phys.\ A {\bf 703}, 489 (2002); 

489: %%CITATION = HEP-PH 0109115;%%

490: E.~Iancu and R.~Venugopalan, hep-ph/0303204.

491: %%CITATION = HEP-PH 0303204;%%

492: 

493: \bibitem{bk}

494: Y.~V.~Kovchegov, Phys.\ Rev.\ D {\bf 60}, 034008 (1999), 

495: %%CITATION = HEP-PH 9901281;%%

496: Phys.\ Rev.\ D {\bf 61}, 074018 (2000); 

497: %%CITATION = HEP-PH 9905214;%%

498: I.~Balitsky, Nucl.\ Phys.\ B {\bf 463}, 99 (1996).

499: %%CITATION = HEP-PH 9509348;%%

500: 

501: \bibitem{grv98}

502: M.~Gluck, E.~Reya and A.~Vogt,

503: Eur.\ Phys.\ J.\ C {\bf 5}, 461 (1998).

504: %%CITATION = HEP-PH 9806404;%%

505: 

506: \bibitem{kkp}

507: B.~A.~Kniehl, G.~Kramer and B.~Potter,

508: Nucl.\ Phys.\ B {\bf 582}, 514 (2000).

509: %%CITATION = HEP-PH 0010289;%%

510: 

511: \bibitem{hw}

512: K.~Rummukainen and H.~Weigert, hep-ph/0309306.

513: %%CITATION = HEP-PH 0309306;%%

514: 

515: \bibitem{solbk}

516: K.~Golec-Biernat and A.~M.~Stasto,

517: Nucl.\ Phys.\ B {\bf 668}, 345 (2003); 

518: %%CITATION = HEP-PH 0306279;%%

519: E.~Gotsman, M.~Kozlov, E.~Levin, U.~Maor and E.~Naftali, hep-ph/0401021. 

520: %%CITATION = HEP-PH 0401021;%%

521: 

522: \bibitem{iim}

523: E.~Iancu, K.~Itakura and S.~Munier, hep-ph/0310338.

524: %%CITATION = HEP-PH 0310338;%%

525: 

526: \bibitem{ina}

527: S.~Jeon, J.~Jalilian-Marian and I.~Sarcevic,

528: Nucl.\ Phys.\ A {\bf 723}, 467 (2003),  

529: %%CITATION = HEP-PH 0207120;%%

530: Phys.\ Lett.\ B {\bf 562}, 45 (2003).

531: %%CITATION = NUCL-TH 0208012;%%

532: 

533: \bibitem{guzey}

534: V.~Guzey, M.~Strikman and W.~Vogelsang, hep-ph/0407201.

535: %%CITATION = HEP-PH 0407201;%%

536: 

537: \bibitem{ks}

538: J.~Kwiecinski and A.~M.~Stasto, Phys.\ Rev.\ D {\bf 66}, 014013 (2002).

539: %%CITATION = HEP-PH 0203030;%%

540: 

541: \bibitem{iimc}

542: E.~Iancu, K.~Itakura and L.~McLerran,

543: Nucl.\ Phys.\ A {\bf 708}, 327 (2002), 

544: %%CITATION = HEP-PH 0203137;%%

545: hep-ph/0205198. 

546: 

547: \bibitem{ykam}

548: Y.~V.~Kovchegov and A.~H.~Mueller,

549: Nucl.\ Phys.\ B {\bf 529}, 451 (1998); 

550: %%CITATION = HEP-PH 9802440;%%

551: A.~Dumitru and L.~D.~McLerran,

552: Nucl.\ Phys.\ A {\bf 700}, 492 (2002).

553: %%CITATION = HEP-PH 0105268;%%

554: 

555: \bibitem{gjm}

556: F.~Gelis and J.~Jalilian-Marian,

557: Phys.\ Rev.\ D {\bf 66}, 014021 (2002),  

558: %%CITATION = HEP-PH 0205037;%%

559: Phys.\ Rev.\ D {\bf 66}, 094014 (2002), 

560: %%CITATION = HEP-PH 0208141;%%

561: Phys.\ Rev.\ D {\bf 67}, 074019 (2003).

562: %%CITATION = HEP-PH 0211363;%%

563: 

564: \bibitem{jmnr}

565: J.~Jalilian-Marian, Y.~Nara and R.~Venugopalan,

566: Phys.\ Lett.\ B {\bf 577}, 54 (2003).

567: %%CITATION = NUCL-TH 0307022;%%

568: 

569: \bibitem{bkw}

570: R.~Baier, A.~Kovner and U.~A.~Wiedemann,

571: Phys.\ Rev.\ D {\bf 68}, 054009 (2003).

572: %%CITATION = HEP-PH 0305265;%%

573: 

574: \bibitem{gbw}

575: K.~Golec-Biernat and M.~Wusthoff,

576: Phys.\ Rev.\ D {\bf 59}, 014017 (1999), 

577: %%CITATION = HEP-PH 9807513;%%

578: Phys.\ Rev.\ D {\bf 60}, 114023 (1999); 

579: %%CITATION = HEP-PH 9903358;%%

580: J.~Bartels, K.~Golec-Biernat and H.~Kowalski,

581: Phys.\ Rev.\ D {\bf 66}, 014001 (2002);

582: %%CITATION = HEP-PH 0203258;%%

583: H.~Kowalski and D.~Teaney,

584: Phys.\ Rev.\ D {\bf 68}, 114005 (2003).

585: %%CITATION = HEP-PH 0304189;%%

586: 

587: \end{thebibliography}

588: 

589: \end{document}

590: 

591: