1: %\documentstyle[12pt]{article}
2: \documentclass[12pt]{article}
3: %\documentclass[12pt,a4paper]{article}
4: \usepackage{epsfig}
5: %\usepackage{showkeys}
6: \textwidth 17cm
7: \textheight 23cm
8: \voffset=-2cm
9: \hoffset=-1.7cm
10: %\renewcommand{\baselinestretch}{0.9}
11: \renewcommand{\thefootnote}{\fnsymbol{footnote}}
12:
13: % marginal note
14: \newcommand{\NB}[1]{\renewcommand{\baselinestretch}{0.5}
15: \marginpar{{\small #1 }}}
16: %\renewcommand{\baselinestretch}{0.9}}
17:
18: \newcommand{\be}{\begin{equation}}
19: \newcommand{\dlq}{\lq\lq}
20: \newcommand{\ee}{\end{equation}}
21: \newcommand{\ben}{\begin{eqnarray*}}
22: \newcommand{\een}{\end{eqnarray*}}
23: \newcommand{\stackeven}[2]{{{}_{\displaystyle{#1}}\atop\displaystyle{#2}}}
24: \newcommand{\lsim}{\stackeven{<}{\sim}}
25: \newcommand{\gsim}{\stackeven{>}{\sim}}
26: \newcommand{\un}[1]{\underline{#1}}
27: %\renewcommand{\baselinestretch}{1.0}
28: \newcommand{\as}{\alpha_s}
29: \newcommand{\tas}{{\tilde\alpha}_s}
30: \newcommand{\bas}{{\overline\alpha}_s}
31: \def\eq#1{{Eq.~(\ref{#1})}}
32: \def\fig#1{{Fig.~\ref{#1}}}
33:
34: \begin{document}
35: \title{\vspace*{1cm}{\bf Perturbative Odderon in the Dipole
36: Model\footnote{\normalsize{We dedicate this work to the memory of Jan
37: Kwieci\'nski.}}\\[1cm] }}
38: \author{
39: {Yuri V.\ Kovchegov\thanks{e-mail: yuri@phys.washington.edu}~$\, ^1$, Lech Szymanowski\thanks{e-mail: Lech.Szymanowski@fuw.edu.pl}~$\, ^2$ and Samuel Wallon\thanks{e-mail: Samuel.Wallon@th.u-psud.fr}~$\, ^3$ } \\[1cm]
40: {\it\small $^1$Department of Physics, University of Washington, Box 351560} \\
41: {\it\small Seattle, WA 98195, USA }\\[0.5cm]
42: {\it\small $^2$Soltan Institute for Nuclear Studies, Ho\.{z}a 69} \\
43: {\it\small 00-681 Warsaw, Poland}\\[0.5cm]
44: {\it\small $^3$Laboratoire de Physique Th\'eorique, Universit\'e Paris XI,} \\
45: {\it\small centre d'Orsay, b\^atiment 211, 91405 Orsay cedex,
46: France}
47: \\[1cm]
48: }
49:
50: \date{September 2003}
51: \maketitle
52:
53: \thispagestyle{empty}
54:
55: \begin{abstract}
56: We show that, in the framework of Mueller's dipole model, the
57: perturbative QCD odderon is described by the dipole model equivalent
58: of the BFKL equation with a $C$-odd initial condition. The
59: eigenfunctions and eigenvalues of the odderon solution are the same as
60: for the dipole BFKL equation and are given by the functions $E^{\,
61: n,\nu}$ and $\chi (n,\nu)$ correspondingly, where the $C$-odd initial
62: condition allows only for odd values of $n$. The leading high-energy
63: odderon intercept is given by $\alpha_{odd} - 1 = \frac{2 \, \as \,
64: N_c}{\pi} \, \chi (n=1 ,\nu=0) = 0$ in agreement with the solution
65: found by Bartels, Lipatov and Vacca. We proceed by writing down an
66: evolution equation for the odderon including the effects of parton
67: saturation. We argue that saturation makes the odderon solution a
68: decreasing function of energy.
69: \end{abstract}
70:
71:
72: \thispagestyle{empty}
73: \begin{flushright}
74: \vspace{-20cm}
75: NT@UW--03--025 \\
76: INT ACK 03--51
77: \end{flushright}
78:
79: \newpage
80:
81: \setcounter{page}{1}
82:
83:
84:
85: \section{Introduction}
86:
87:
88: The dominant contributions to the total cross sections of hadronic
89: reactions are related to the Pomeron and to the Odderon
90: exchanges. While the Pomeron exchange having the quantum numbers of
91: the vacuum represents a dominant contribution to the sum of total
92: cross sections for a given hadronic process and its crossing
93: counterpart which is even under crossing, the Odderon exchange
94: dominates the difference of these two cross sections, which is odd
95: under the crossing symmetry. Within Regge theory Pomeron and Odderon
96: are therefore natural partners. The importance of Odderon exchange for
97: the phenomenology of hadronic reactions was noted long ago
98: \cite{LukNic} and was the subject of many investigations. A
99: review of these early studies can be found in, e.g., \cite{Nic}.
100:
101:
102: With the advent of quantum chromodynamics (QCD) theoretical status of
103: the perturbative Odderon followed the development of the analogous
104: description of its Pomeron partner. Within QCD in the leading
105: logarithmic approximation (LLA), the Odderon appears as the color
106: singlet exchange of three gluons, which interactions are of a form
107: very similar to the interactions of the two gluons forming the
108: Pomeron. It is therefore not surprising that soon after the derivation
109: of the Balitsky-Fadin-Kuraev-Lipatov (BFKL) equation
110: \cite{BFKL} for the Pomeron there was derived an analogous
111: Bartels-Kwiecinski-Praszalowicz \cite{BJKP} equation for the Odderon.
112:
113:
114: The solution of the BFKL equation by Lipatov in the seminal Ref.
115: \cite{Lipatov1} stimulated similar studies of the BKP equation which,
116: because of its three-body nature, turned out to be much more difficult
117: to solve. These attempts led nevertheless to a discovery of a deep
118: connection between the BFKL equation, the BKP equation and their
119: generalizations on one side and the two dimensional conformal models
120: of spin chains \cite{Lev93} and their exact solubility
121: \cite{Lev94,Korch95} on another side. Finally they also led to
122: discovery of several solutions of the BKP equation \cite{JW},
123: \cite{KKM}, \cite{BLV} with different intercepts of the corresponding Odderon
124: Regge trajectories determining the asymptotics of the cross
125: sections. In particular, the solution found by Bartels, Lipatov and
126: Vacca (BLV) in Ref. \cite{BLV} describes the Odderon with the largest
127: intercept found to date, which is equal to $1$\footnote{Here we
128: restrict ourselves to the Odderon described by the BKP equation for
129: three gluons in the $t$-channel. In Ref. \cite{DKKM2} it was argued
130: that exchanges of many $t$-channel gluons generalizing the BKP
131: equation may also lead to the Odderon intercept asymptotically
132: approaching $1$ from below.}. Thus the present results suggest that
133: the perturbative Odderon intercept is smaller than the one for the
134: BFKL Pomeron.
135:
136: From an experimental point of view the Odderon remains a
137: mystery. Recent data on the differential elastic $pp$ scattering show
138: that one needs the Odderon to describe the cross sections in the dip
139: region \cite{dip-pp}. On the other hand, the perturbative QCD
140: prediction in the Born approximation for the cross section of
141: diffractive photoproduction of $\eta_c-$meson is rather small
142: \cite{eta}. The inclusion of evolution based on the BKP equation
143: increases the cross section by one order of magnitude
144: \cite{BBCV}. Because of that suggestions were raised to look for
145: Odderon effects in the studies of the charge asymmetries in the two
146: pion diffractive production \cite{HPST} and $\eta_c$ electroproduction
147: in the triple Regge region \cite{BBV}. A comprehensive survey of
148: searches for Odderon effects as well as its present day theoretical
149: status is given in the recent review \cite{Ewerz}.
150:
151:
152: The original derivation of the BFKL equation was based on the LLA
153: resummation of conventional Feynman diagrams in the Regge kinematics
154: typical for high energy and diffractive processes. Another approach to
155: the description of high energy processes was proposed by Mueller, who
156: constructed the dipole model \cite{dip} based on the light cone
157: perturbation theory. In the dipole model, the Pomeron, or in other
158: words the BFKL evolution emerges as an evolution of the wave function
159: of a projectile particle as viewed in the reference frame of a target
160: particle. In Ref. \cite{dipBFKL} Navelet and Wallon have shown the
161: equivalence of the color dipole model and the BFKL Pomeron. Since the
162: dipole model is actually a very popular and efficient tool for the
163: description of high energy and diffractive processes and a starting
164: point for inclusion of saturation effects \cite{glr,yuri} it appears
165: natural to ask whether the Odderon can also be incorporated in the
166: dipole model similar to how the BKP equation was derived within the
167: same approach that had earlier led to the BFKL equation. Finding the
168: answer to this question is the main motivation for the present work.
169:
170: The paper is structured as follows. In Sect. 2 we demonstrate that the
171: Odderon evolution is described in the dipole model by the same
172: evolution equation (\ref{oddeq}) as was used in \cite{dip} to describe
173: the BFKL evolution. The only difference is that to project out the
174: Odderon evolution from this equation one needs to impose $C$-odd
175: initial conditions (\ref{inc2}). The solution of \eq{oddeq} is given
176: by \eq{gensol} using the eigenfunctions (\ref{enn}) and the
177: eigenvalues (\ref{eig}) for odd $n$. The leading high energy Odderon
178: intercept is equal to $1$ as shown in \eq{oddint}. In Sect. 3 we show
179: that our solution (\ref{3gsol}) is equivalent to the BLV Odderon
180: solution \cite{BLV}. Thus, for a given three-gluon initial condition,
181: the dipole BFKL equation is equivalent to the BKP equation for three
182: $t$-channel gluons. In Sect. 4 we write down the equation
183: (\ref{oddnl}) describing the Odderon evolution including the
184: saturation effects. Saturation is likely to make the Odderon solution
185: a decreasing function of energy. We conclude in Sect. 5 by discussing
186: what we have and have not proved.
187:
188:
189:
190: \section{Evolution Equation For Perturbative QCD Odderon in the Dipole Model}
191:
192: In this Section we show that in Mueller's dipole model the odderon
193: evolution equation is given by the dipole model equivalent of the BFKL
194: equation taken with a $C$-odd initial condition. Solutions of this
195: equation for the odderon exchange amplitude are given by functions
196: $E^{n, \nu}$ with odd $n$. Corresponding eigenvalues are given by the
197: BFKL equation eigenvalues $\chi (n, \nu)$ with odd $n$, such that the
198: leading high energy intercept is given by $\alpha_{odd} - 1 = \frac{2
199: \, \as \, N_c}{\pi} \, \chi (n=1 ,\nu=0) = 0$.
200:
201: \subsection{One Step of Small-$x$ Evolution}
202:
203: Let us consider diffractive scattering of a $q\bar q$ dipole on some
204: target, which could be another dipole for the case of onium--onium
205: scattering or a proton for the case of Deep Inelastic Scattering
206: (DIS). For the scattering amplitude of a $q\bar q$ dipole on a target
207: the operation of charge conjugation corresponds to interchanging the
208: quark and the anti-quark lines. Consider a dipole consisting of a
209: quark at transverse coordinate ${\un x}_0$ and an anti-quark at
210: transverse coordinate ${\un x}_1$ carrying the light cone momentum
211: fractions $z$ and $1-z$ of the total light cone momentum of the $q\bar
212: q$ system. Operation of charge conjugation corresponds to replacing
213: \be\label{C}
214: C: \hspace*{1cm} {\un x}_0 \, \leftrightarrow \, {\un x}_1,
215: \hspace*{1cm} z \, \leftrightarrow \, 1-z.
216: \ee
217: Odderon exchange, by definition, corresponds to diffractive amplitudes
218: which are anti-symmetric under the operation shown in
219: \eq{C}:
220: \be\label{Codd}
221: {\cal O} ({\un x}_0, {\un x}_1; z, 1-z) \, = \, - \, {\cal O} ({\un
222: x}_1, {\un x}_0; 1-z, z),
223: \ee
224: where ${\cal O} ({\un x}_0, {\un x}_1; z, 1-z)$ is the forward
225: amplitude of a dipole $01$ on some target. Below we will use \eq{Codd}
226: as a definition of the odderon--mediated amplitude.
227:
228: Our goal now is to start with some $C$-odd initial condition for
229: onium--target scattering process and try to include one step of the
230: dipole evolution \cite{dip,CM} in it. The simplest initial condition
231: is given by the color-singlet three-gluon exchange diagram shown in
232: \fig{init} corresponding to the lowest order odderon exchange. Here,
233: for simplicity, we take the target to be a single quark line at the
234: bottom of \fig{init} located at transverse coordinate ${\un x} = {\un
235: 0}$. In \fig{init} the disconnected $t$-channel gluon lines imply
236: summation over all possible connections of these lines to both the
237: quark and the anti-quark in the onium state at the top of the figure,
238: as shown in \fig{impact}. When convoluted with virtual photon's wave
239: function on the left hand side and with a meson's wave function on the
240: right hand side, the diagram in \fig{init} would contribute to
241: production of pions, $\eta_c$, etc., in DIS (see
242: e.g. \cite{BLV,BBCV}).
243:
244: The three-gluon exchange diagram in \fig{init} gives the imaginary
245: part of the forward odderon-mediated amplitude ($x_i = |{\un x}_i|$)
246: \be\label{inc}
247: {\cal O}_0 ({\un x}_0, {\un x}_1) \, = \, c_0 \, \as^3 \, \ln^3
248: \frac{x_0}{x_1}
249: \ee
250: with
251: \be\label{c0}
252: c_0 = \frac{(N_c^2 - 4) \, (N_c^2 - 1)}{4 \, N_c^3}.
253: \ee
254: The color structure of the color-singlet three gluon exchange is given
255: by $d^{abc}$ for the odderon amplitude. The calculation leading to
256: \eq{inc} can be performed in covariant gauge, as well as in the $A^+ =
257: 0$ light cone gauge which we will be using throughout the paper. (The
258: onium is moving in the light cone $+$ direction.) One can easily see
259: that the amplitude in \eq{inc} satisfies the condition of \eq{Codd}
260: and is, therefore, $C$-odd. One can also verify that the integral of
261: the amplitude (\ref{inc}) over the impact parameter ${\un b} = ({\un
262: x}_0 + {\un x}_1)/2$ is zero, in agreement with the fact that the
263: odderon exchange is zero at momentum transfer $t=0$.
264:
265: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
266: \begin{figure}[h]
267: \begin{center}
268: \epsfig{file=init.eps, width=7cm}
269: \end{center}
270: \caption{Lowest order odderon exchange diagram which will be used as
271: initial condition for small-$x$ evolution. The gluon lines connect to
272: both the quark and the anti-quark in the onium above.}
273: \label{init}
274: \end{figure}
275: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
276:
277: One step of the dipole evolution \cite{dip} corresponds to emitting a
278: single $s$-channel gluon in the onium wave function. In the
279: large-$N_c$ limit in which the dipole model \cite{dip} is constructed,
280: the gluon line is represented by a double quark line, as shown in
281: \fig{one}. The diagram in \fig{one} represents a real part of the dipole
282: kernel, where the emitted gluon is present at the time of scattering
283: on the target. The $s$-channel gluon can be emitted by either quark or
284: anti-quark to the left and to the right of the interaction, which is implied
285: by the disconnected gluon (double) line. The original dipole $01$ is
286: split into two (color-singlet) dipoles $02$ and $21$. Three exchanged
287: gluons can only couple all together to either one of the dipoles $02$
288: and $21$. \fig{one} shows the interaction with the dipole $21$. The
289: only other possible interaction diagram involves one of the gluons
290: coupling to one of the dipoles with the other two gluons coupling to
291: another dipole. That diagram is zero since a color-neutral dipole can
292: not interact with the target by a single gluon exchange.
293:
294: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
295: \begin{figure}[h]
296: \begin{center}
297: \epsfig{file=one.eps, width=6cm}
298: \end{center}
299: \caption{One step of dipole evolution with the three-gluon exchange
300: initial conditions. }
301: \label{one}
302: \end{figure}
303: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
304:
305: Employing the dipole evolution kernel from \cite{dip} we obtain the
306: following expression for the first step of dipole evolution with the
307: initial condition (\ref{inc}) (${\un x}_{ij} = {\un x}_i - {\un x}_j$)
308: \be\label{step1}
309: \frac{\as \, N_c}{2 \, \pi^2} \, \int^{\mbox{min} \{z,1-z\}}_{z_{init}}
310: \frac{d z_2}{z_2} \, \int d^2 x_2 \, \frac{x_{01}^2}{x_{02}^2 \,
311: x_{12}^2} \, \left[ {\cal O}_0 ({\un x}_0, {\un x}_2) + {\cal O}_0
312: ({\un x}_2, {\un x}_1) - {\cal O}_0 ({\un x}_0, {\un x}_1) \right]
313: \ee
314: where the first and the second terms correspond to real emissions with
315: the second term pictured in \fig{one}, while the last term in
316: \eq{step1} corresponds to the virtual correction \cite{dip}. In \eq{step1}
317: $z_2$ is the fraction of onium's longitudinal (``plus'') momentum
318: carried by the gluon $2$, which, in the leading logarithmic
319: approximation, is integrated from some initial value $z_{init}$ up to the
320: smaller one of the light cone momentum fractions carried by the
321: original quark and anti-quark \cite{dip}. Substituting ${\cal O}_0$
322: from \eq{inc} into \eq{step1} yields
323: \be\label{step1'}
324: 3 \, \frac{\as \, N_c}{2 \, \pi^2} \, c_0 \, \as^3 \, \ln
325: \frac{x_0}{x_1} \ \ln \left( \frac{\mbox{min} \{z,1-z\}}{z_{init}} \right) \,
326: \int d^2 x_2 \, \frac{x_{01}^2}{x_{02}^2 \,
327: x_{12}^2} \, \ln \frac{x_0}{x_2} \, \ln \frac{x_1}{x_2}.
328: \ee
329: The amplitude in \eq{step1'} is non-zero. It changes sign under the
330: transformation of \eq{C} and is thus $C$-odd. We conclude that $C$-odd
331: initial conditions leave a dipole amplitude $C$-odd even after one
332: step of small-$x$ dipole evolution. The resulting amplitude is
333: non-zero, which means that the projection of the dipole evolution on
334: the $C$-odd channel is non-trivial.
335:
336:
337: \subsection{Evolution Equation}
338:
339: It is now straightforward to write down an evolution equation for the
340: odderon in the dipole model. We have shown above that in the dipole
341: model a single step of the small-$x$ evolution for the odderon is the
342: same as for the BFKL pomeron \cite{dip}. Subsequent steps of the
343: dipole evolution would generate more color dipoles in the onium wave
344: function. Similar to one step of evolution shown in \fig{one}, there
345: is only one way of connecting the three-gluon initial conditions to
346: the fully evolved dipole wave function having many dipoles in it: only
347: the diagrams with all three gluons interacting with the same dipole
348: survive, since a dipole can not interact via a single gluon
349: exchange. Therefore, the picture of the odderon evolution is
350: simple. The onium wave function develops the usual dipole cascade
351: described by the leading logarithmic dipole evolution
352: \cite{dip}. In the linear evolution approximation, one of the dipoles
353: generated by the evolution interacts with the target. In the case of
354: the pomeron exchange, the lowest order interaction is given by a
355: two-gluon exchange \cite{dip}. For the case of the odderon considered
356: here, the interaction should be $C$-odd and at the lowest order is
357: given by the three-gluon exchange amplitude (\ref{inc}).
358:
359: For the odderon evolution equation we thus write \cite{dip}
360: \be\label{oddeq}
361: \frac{\partial}{\partial Y} {\cal O} ({\un x}_0, {\un x}_1, Y) \, = \,
362: \frac{\as \, N_c}{2 \, \pi^2} \, \int d^2 x_2 \,
363: \frac{x_{01}^2}{x_{02}^2 \, x_{12}^2} \, \left[ {\cal O} ({\un x}_0,
364: {\un x}_2, Y) + {\cal O} ({\un x}_2, {\un
365: x}_1, Y) - {\cal O} ({\un x}_0, {\un x}_1, Y) \right]
366: \ee
367: where we changed the arguments of the dipole amplitude ${\cal O} ({\un
368: x}_{i}, {\un x}_j; z_i, z_j)$ to ${\cal O} ({\un x}_{i}, {\un x}_j,
369: Y)$ such that $Y = \ln (\mbox{min} \{z_i, z_j \} / z_{init})$ is the
370: rapidity variable. In the leading logarithmic approximation for gluon
371: evolution the dependence of the dipole amplitude on the light cone
372: momentum fractions of the quark $z_i$ and the antiquark $z_j$ comes in
373: only through the quantity $\mbox{min} \{z_i, z_j \}$ \cite{dip}, which
374: allowed us to reduce the number of arguments in ${\cal O}$.
375:
376: The initial condition for the \eq{oddeq} is given by
377: \be\label{inc2}
378: {\cal O} ({\un x}_0, {\un x}_1, Y=0) \, = \, {\cal O}_0 ({\un x}_0,
379: {\un x}_1) \, = \, c_0 \, \as^3 \, \ln^3 \frac{x_0}{x_1}.
380: \ee
381: Eqs. (\ref{oddeq}) and (\ref{inc2}) describe the perturbative QCD
382: odderon in the color dipole model. \eq{oddeq} is identical to the
383: dipole model equivalent of the BFKL equation found in \cite{dip}.
384:
385: \eq{oddeq} is illustrated in \fig{lineq}. The blob on the left hand side
386: of \fig{lineq} represents the dipole evolution and the interaction
387: with the target leading to the odderon exchange amplitude for the
388: dipole $01$. After one step of the evolution a soft gluon is
389: emitted. This is shown in the three terms on the right hand side of
390: \fig{lineq}. They all correspond to the three terms on the right of
391: \eq{oddeq}. The first two terms correspond to real emissions, where
392: the subsequent evolution continues either in dipole $02$ or in dipole
393: $21$. The last term represents virtual corrections, where the gluon
394: can be emitted and absorbed on either side of the blob.
395:
396: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 3%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
397: \begin{figure}[h]
398: \begin{center}
399: \epsfig{file=eqn.eps, width=10cm}
400: \end{center}
401: \caption{Dipole evolution equation for the odderon exchange amplitude.}
402: \label{lineq}
403: \end{figure}
404: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
405:
406: The rapidity variable does not change under $C$-parity transformation
407: of \eq{C}. Therefore, in the leading logarithmic approximation
408: considered here, charge conjugation only interchanges the transverse
409: coordinates of the quark and the anti-quark. Thus one can explicitly
410: see in \eq{oddeq} that substituting a $C$-odd amplitude into its right
411: hand side yields a $C$-odd expression, as we have seen in the previous
412: section for the case of three-gluon exchange amplitude. Evolution of
413: \eq{oddeq} preserves $C$-parity of the amplitude. When the initial
414: condition for \eq{oddeq} is given by some $C$-even amplitude, such as
415: the two-gluon exchange, it projects out the $C$-even solution of this
416: equation, which corresponds to the BFKL pomeron \cite{dip}. $C$-odd
417: initial condition projects out a $C$-odd solution, which corresponds
418: to the perturbative odderon exchange.
419:
420: A comment is in order here. In \cite{CM} a dipole model equation was
421: constructed which was equivalent to the BKP equation \cite{BJKP} for
422: a four-reggeon system. The resulting equation (48) of \cite{CM} is
423: somewhat more complicated than the dipole BFKL equation
424: (\ref{oddeq}). One may wonder why there appears to be no such increase
425: of complication in the dipole model in going from a two-reggeon
426: (pomeron) system to the three reggeon (odderon) system. The reason for
427: that is clear. To describe the four-reggeon state in the dipole model
428: the authors of \cite{CM} had to make sure that this state did not mix
429: with the single pomeron exchange, double pomeron exchange and the
430: odderon. Imposing a four-gluon exchange initial conditions in
431: \cite{CM} made sure that the state had a positive parity $C=+1$ and
432: did not mix with the odderon having $C=-1$. To avoid mixing with the
433: two-pomeron state the authors of \cite{CM} had to make sure that the
434: state had a single-cylinder color topology. Finally, to avoid mixing
435: of this single-cylinder state with the single pomeron the authors of
436: \cite{CM} constructed a coupling of four gluons to the dipole wave
437: function which preserved the single-cylinder topology without
438: connecting all four gluons to the same dipole. That required
439: introduction of color quadrupoles and resulted in a more complicated
440: kernel for the integral equation. Now the three-reggeon state
441: considered here is much simpler: the only potential problem is for it
442: to mix with the pomeron, but that can never happen due to different
443: $C$-parities. Therefore we do not have to invent any complicated
444: color-quadrupole coupling of the three-gluon state to the onium wave
445: function. While it exists, it is $N_c$-suppressed compared to the
446: dipole coupling considered above.
447:
448:
449: \subsection{Odderon Solution}
450:
451: The solution of \eq{oddeq} is easy to construct. We first note that
452: the eigenfunctions of the Casimir operators of conformal algebra
453: \be\label{enn}
454: E^{n, \nu} (\rho_0, \rho_1) \, = \, \left( \frac{\rho_{01}}{\rho_0 \,
455: \rho_1} \right)^{\frac{1+n}{2} + i \nu} \ \left(
456: \frac{\rho_{01}^*}{\rho_0^* \, \rho_1^*} \right)^{\frac{1-n}{2} + i
457: \nu}
458: \ee
459: are also the eigenfunctions of the dipole kernel of \eq{lineq} with
460: the eigenvalues
461: \be\label{eig}
462: 2 \, \bas \, \chi (n, \nu),
463: \ee
464: where
465: \be\label{chi}
466: \chi (n, \nu) \, = \, \psi (1) - \frac{1}{2} \, \psi \left(
467: \frac{1+|n|}{2} + i \nu \right) - \frac{1}{2} \, \psi \left(
468: \frac{1+|n|}{2} - i \nu \right),
469: \ee
470: \be
471: \bas \, = \, \frac{\as \, N_c}{\pi},
472: \ee
473: $\rho_{ij} = \rho_i - \rho_j$ and the asterisk in \eq{enn} denotes
474: complex conjugation. \eq{eig} follows from \eq{goal} which is derived
475: in Appendix A.
476:
477: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
478: \begin{figure}[h]
479: \begin{center}
480: \epsfig{file=oo.eps, width=3cm}
481: \end{center}
482: \caption{Onium-onium scattering.}
483: \label{oo}
484: \end{figure}
485: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
486:
487: To construct the solution of \eq{oddeq} with a three-gluon exchange
488: initial conditions let us consider onium-onium scattering, i.e. take
489: the target of the above discussion to be a $q\bar q$ pair as depicted
490: in \fig{oo}. We will work in the frame where the target onium $1'2'$ is
491: at rest, so that all the small-$x$ evolution will be included in the wave
492: function of the incoming onium $12$.
493:
494: A general solution of \eq{oddeq} is easy to write down once we know
495: that functions $E^{n, \nu}$ from \eq{enn} are the eigenfunctions of
496: its kernel. Since
497: \be
498: E^{n, \nu} (\rho_{1}, \rho_{2}) \, = \, (-1)^n \, E^{n, \nu}
499: (\rho_{2}, \rho_{1})
500: \ee
501: only the functions $E^{n, \nu} (\rho_{1}, \rho_{2})$ with odd $n$
502: satisfy the condition (\ref{Codd}). The most general $C$-odd solution
503: reads in the complex notation
504: \be\label{gensol}
505: {\cal O} (\rho_1, \rho_2 ; \rho_{1'}, \rho_{2'}; Y) \, = \,
506: \sum_{\mbox{odd n}} \, \int_{-\infty}^\infty d \nu \, \int d^2
507: \rho_0 \, e^{2 \, \bas \, \chi(n, \nu) \, Y} \, C_{n, \nu} \,
508: E^{n, \nu} (\rho_{10}, \rho_{20}) \,
509: E^{n, \nu \, *} (\rho_{1'0}, \rho_{2'0}),
510: \ee
511: where the coefficients $C_{n, \nu}$ have to be determined from the
512: initial conditions. The equivalent of \eq{inc2} for the dipole-dipole
513: scattering of \fig{oo} reads in complex notation
514: \be\label{incompl}
515: {\cal O} (\rho_1, \rho_2 ; \rho_{1'}, \rho_{2'}; Y=0) \, = \, c_0 \,
516: \as^3 \, \ln^3 \bigg| \frac{\rho_{11'} \, \rho_{22'}}{\rho_{12'} \,
517: \rho_{1'2}} \bigg|.
518: \ee
519: To determine the coefficients $C_{n, \nu}$ we rewrite the amplitude
520: (\ref{incompl}) as (see Appendix B)
521: \ben
522: {\cal O} (\rho_1, \rho_2 ; \rho_{1'}, \rho_{2'}; Y=0) \, = \, c_0 \,
523: \as^3 \, \frac{6}{\pi^2} \, \sum_{\mbox{odd n}} \, \int_{-\infty}^\infty
524: d \nu \, \int d^2
525: \rho_0 \, \frac{\nu^2 + \frac{n^2}{4}}{\left[ \nu^2 + \frac{(n+1)^2}{4}
526: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4}
527: \right]}
528: \een
529: \be\label{decom}
530: \times \, \chi (n, \nu) \, E^{n, \nu} (\rho_{10}, \rho_{20}) \,
531: E^{n, \nu \, *} (\rho_{1'0}, \rho_{2'0})
532: \ee
533: which yields
534: \be\label{coef}
535: C_{n, \nu} \, = \, c_0 \, \as^3 \, \frac{6}{\pi^2} \,
536: \frac{\nu^2 + \frac{n^2}{4}}{\left[ \nu^2 + \frac{(n+1)^2}{4}
537: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4} \right]} \, \chi (n, \nu).
538: \ee
539: Using \eq{coef} in \eq{gensol} we find the solution of \eq{oddeq} with
540: the initial condition (\ref{incompl})
541: \ben
542: {\cal O} (\rho_1, \rho_2 ; \rho_{1'}, \rho_{2'}; Y) \, = \, c_0 \,
543: \as^3 \, \frac{6}{\pi^2} \, \sum_{\mbox{odd n}} \, \int_{-\infty}^\infty
544: d \nu \, \int d^2
545: \rho_0 \, \frac{\nu^2 + \frac{n^2}{4}}{\left[ \nu^2 + \frac{(n+1)^2}{4}
546: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4}
547: \right]}
548: \een
549: \be\label{3gsol}
550: \times \, e^{2 \, \bas \, \chi(n, \nu) \, Y} \, \chi (n, \nu) \,
551: E^{n, \nu} (\rho_{10}, \rho_{20}) \,
552: E^{n, \nu \, *} (\rho_{1'0}, \rho_{2'0}).
553: \ee
554: Therefore, we conclude that the eigenfunctions and eigenvalues of the
555: dipole amplitude with the odderon exchange are given by
556: Eqs. (\ref{enn}) and (\ref{eig}) correspondingly, with only odd values of
557: $n$. Analyzing \eq{3gsol} one can easily see that the leading high
558: energy intercept of the odderon amplitude is given by
559: \be\label{oddint}
560: \alpha_{odd} - 1 \, = \, 2 \, \bas \, \chi(n=1, \nu =0) \, = \, 0
561: \ee
562: in agreement with the results of Bartels, Lipatov and Vacca (BLV)
563: \cite{BLV}.
564:
565:
566: \section{Connection to Traditional Approaches}
567:
568:
569: To make a connection between the dipole evolution equation
570: (\ref{oddeq}) and the BKP equation \cite{BJKP} for the three-reggeon
571: state we will now show that the eigenfunctions from \eq{enn} with odd
572: $n$ correspond to the odderon solution found by Bartels, Lipatov and
573: Vacca in \cite{BLV}. To find the dipole scattering amplitude generated
574: by the odderon Green function we have to connect the external gluon
575: legs of the Green function to the dipole, as shown in
576: \fig{impact}. Instead of the full odderon Green function (see Eq. (5) in
577: \cite{BBCV}), we take a single BLV eigenfunction in the form given by Eq.
578: (2) of \cite{BBCV}
579: \be\label{blveig}
580: \Psi^{n,\nu} (k_1, k_2, k_3) \, = \, c (n,\nu) \, \sum_{(123)} \,
581: \frac{({\un k}_1 + {\un k}_2)^2}{{\un k}_1^2 \, {\un k}_2^2} \,
582: E^{n, \nu} ({\un k}_1 + {\un k}_2, {\un k}_3), \hspace*{1cm}
583: \mbox{odd} \ n,
584: \ee
585: and connect in to the dipole $x_{01}$ in all possible ways as depicted
586: in \fig{impact}. $c (n,\nu)$ in \eq{blveig} is the overall
587: normalization coefficient which is not important and is given in
588: \cite{BBCV}. Summation over all permutations of $1$, $2$ and $3$ is
589: implied in \eq{blveig}. The dipole amplitude would then be given by
590: \ben
591: {\cal O}^{n, \nu} ({\un x}_0, {\un x}_1, Y) \, = \,
592: \int \left[ \prod_{i=1}^3 \, d^2 k_i \,
593: (e^{- i {\un k}_i \cdot {\un x}_0} - e^{- i {\un k}_i \cdot {\un
594: x}_1}) \right] \, \Psi^{n,\nu} (k_1, k_2, k_3)
595: \een
596: \be\label{dip1}
597: \, = \, 3 \, c (n,\nu) \,
598: \int \left[ \prod_{i=1}^3 \, d^2 k_i \,
599: (e^{- i {\un k}_i \cdot {\un x}_0} - e^{- i {\un k}_i \cdot {\un
600: x}_1}) \right] \,
601: \frac{({\un k}_1 + {\un k}_2)^2}{{\un k}_1^2 \, {\un k}_2^2} \,
602: E^{n, \nu} ({\un k}_1 + {\un k}_2, {\un k}_3).
603: \ee
604: Substituting
605: \be
606: E^{n, \nu} ({\un k}_1 + {\un k}_2, {\un k}_3) \, = \, \int \frac{d^2
607: r_1 \, d^2 r_2}{(2 \pi)^4} \, e^{i ({\un k}_1 + {\un k}_2) \cdot {\un
608: r}_1 + i {\un k}_3 \cdot {\un r}_2} \, E^{n, \nu} (r_1, r_2)
609: \ee
610: into \eq{dip1} and integrating over $k_i$'s yields
611: \ben
612: {\cal O}^{n, \nu} ({\un x}_0, {\un x}_1, Y) \, = \, 6 \, c (n,\nu) \,
613: \int d^2 r_1 \, d^2 r_2 \, E^{n, \nu} (r_1, r_2) \,
614: \left\{
615: \frac{x_{01}^2}{|{\un x}_1 - {\un r}_1|^2 \, |{\un x}_0 - {\un r}_1|^2} \,
616: \right.
617: \een
618: \be\label{dip2}
619: \times \, \left[ \delta ({\un x}_1 - {\un r}_2 ) -
620: \delta ({\un x}_0 - {\un r}_2 ) \right]
621: - \left. 2 \pi \, \ln \frac{|{\un x}_1 - {\un r}_1|}{|{\un x}_0 -
622: {\un r}_1|} \left[ \delta ({\un x}_0 - {\un r}_1 ) \, \delta ({\un
623: x}_1 - {\un r}_2 ) + \delta ({\un x}_0 - {\un r}_2 ) \, \delta ({\un
624: x}_1 - {\un r}_1 ) \right] \right\}.
625: \ee
626: Regulating $|{\un x}_1 - {\un r}_1| = 0$ and $|{\un x}_0 - {\un r}_1|
627: = 0$ singularities in the same way in the first and the second terms
628: on the right hand side of \eq{dip2} and integrating \eq{dip2} over
629: $r_2$ we rewrite it as
630: \ben
631: {\cal O}^{n, \nu} ({\un x}_0, {\un x}_1, Y) \, = \, 6 \, c (n,\nu) \,
632: \int d^2 r_1 \, \frac{x_{01}^2}{|{\un x}_1 - {\un r}_1|^2 \, |{\un
633: x}_0 - {\un r}_1|^2} \, \left[ E^{n, \nu} (x_0, r_1) + E^{n, \nu}
634: (r_1, x_1) \right.
635: \een
636: \be\label{dipeig}
637: \left. - E^{n, \nu} (x_0, x_1) \right].
638: \ee
639: Using \eq{goal} which is proved in Appendix A we rewrite \eq{dipeig}
640: as
641: \be\label{conn}
642: {\cal O}^{n, \nu} ({\un x}_0, {\un x}_1, Y) \, = \, 24 \,
643: c (n,\nu) \, \pi \, \chi(n, \nu) \, E^{n, \nu} (x_0, x_1).
644: \ee
645: \eq{conn} shows that momentum space BLV eigenfunctions in the
646: $t$-channel Green function translate into the functions $E^{n, \nu}$
647: for the dipole scattering amplitude in transverse coordinate space up
648: to an overall normalization factor. Therefore our odderon solution
649: (\ref{gensol}) is equivalent to the BLV solution
650: \cite{BLV}. As one can explicitly check, taking the odderon Green
651: function from Eq. (5) of \cite{BBCV}, connecting it to the two
652: colliding onia $12$ and $1'2'$ as shown at the lowest order in
653: \fig{oo}, and Fourier-transforming into transverse coordinate space
654: as was done in arriving at \eq{conn} would yield us the solution in
655: \eq{3gsol}. Starting with the same three-gluon exchange initial
656: condition (\ref{decom}) our evolution equation (\ref{oddeq}) and the
657: BKP equation \cite{BJKP} would give us the same answer for the
658: amplitude and are equivalent in the case of dipole scattering on a
659: target.
660: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 5%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
661: \begin{figure}
662: \begin{center}
663: \epsfig{file=impact.eps, width=13cm}
664: \end{center}
665: \caption{Dipole's impact factor.}
666: \label{impact}
667: \end{figure}
668: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
669:
670: Since the most general $C$-odd solution of \eq{oddeq} is known and is
671: given by \eq{gensol}, we have proven that for the case of a dipole
672: scattering on a target there could be no other odderon solution except
673: for the one found above, or, equivalently, except for BLV solution
674: \cite{BLV}. Any $C$-odd dipole amplitude satisfying the condition of
675: \eq{Codd} and being a solution of \eq{oddeq} can always be decomposed in
676: a series (\ref{gensol}). Therefore, for the scattering of a dipole on
677: any target the resulting cross section can not grow with center of
678: mass energy $s$ faster than $s^0$ as follows from the intercept of
679: \eq{oddint}. (The saddle point evaluation of the $\nu$ integral around
680: $\nu =0$ of the $n=1$ term in \eq{3gsol} would give an amplitude which
681: would grow slowly with energy, though the growths would be slower than
682: any power of energy.)
683:
684:
685: \section{Including Saturation Effects}
686:
687: Consider DIS on a proton or a nucleus at high enough energy so that
688: parton saturation effects \cite{glr} are becoming important. In
689: \cite{yuri,bal} (see also \cite{JKLW,FILM,Braun1}) a non-linear evolution
690: equation has been derived which resums all multiple pomeron exchanges
691: in the $C$-even forward scattering amplitude of a $q\bar q$ dipole on
692: the target. The equation is easier to visualize for a nuclear target
693: with atomic number $A$. As was discussed in \cite{yuri}, in the frame
694: where the nucleus is at rest all the small-$x$ dipole evolution takes
695: place in the incoming $q\bar q$ dipole wave function. The evolution,
696: taken in the leading logarithmic approximation, resums powers of $\as
697: Y$. The color dipoles generated by evolution rescatter on the nuclear
698: target. The rescattering in \cite{yuri} was given by the
699: Glauber-Mueller formula \cite{Mueller1}, which corresponds to
700: resummation of the parameter $\as^2 A^{1/3}$ \cite{mv,kjklw}. This
701: parameter arises if we limit rescattering of a dipole on each nucleon
702: to two-gluon exchange only \cite{kjklw}.
703:
704: To construct a similar equation for the $C$-odd (odderon) exchange
705: amplitude one has to allow one of the dipoles to interact with one
706: nucleon via a three-gluon exchange. In the high energy regime when
707: $\as Y \sim 1$ and $\as^2 A^{1/3} \sim 1$, a single three-gluon
708: exchange would bring in one power of $\as^3 A^{1/3} \sim \as$, so that
709: the corresponding odderon exchange amplitude would be of the order
710: ${\cal O} \sim \as \ll 1$. Below we limit our analysis to a single
711: odderon exchange: multiple odderon exchanges would be suppressed by
712: extra powers of $\as$. In fact, the $C$-even amplitude including a
713: pomeron splitting into two odderons which interact with the target
714: would be of the order $\as^2$, which is of the same order as the
715: so-called pomeron loop diagrams. Resummation of these diagrams is a
716: very hard unsolved problem (see the discussion in the second reference
717: in \cite{yuri}) and is beyond the scope of this paper.
718:
719:
720: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%fig 6%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
721: \begin{figure}[h]
722: \begin{center}
723: \epsfig{file=nonlin.eps, width=15cm}
724: \end{center}
725: \caption{Evolution equation for the odderon exchange amplitude ${\cal O}$
726: in the presence of gluon saturation, which comes in through the
727: $C$-even dipole (``pomeron'') amplitude $N$. }
728: \label{nl}
729: \end{figure}
730: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
731:
732: The evolution equation for the odderon amplitude which includes
733: saturation effects is depicted in \fig{nl}. Similar to
734: \fig{lineq}, in one step of evolution a gluon is emitted splitting the
735: original dipole $01$ in two dipoles $02$ and $21$. The subsequent
736: odderon evolution can take place only in one of the dipoles. The first
737: term on the right hand side of \fig{nl} schematically represents all
738: three terms on the right hand side of \fig{lineq} and \eq{oddeq}. The
739: second term on the right hand side of \fig{nl} corresponds to the case
740: where the dipole without the odderon evolution develops nonlinear
741: $C$-even evolution in it. $N ({\un x}_0, {\un x}_1, Y)$ is the
742: $C$-even forward scattering amplitude of the dipole $01$ with rapidity
743: $Y$ on the target including all multiple rescatterings and pomeron
744: exchanges \cite{yuri}. $N ({\un x}_0, {\un x}_1, Y)$ is proportional
745: to the $T$-matrix of the interaction and is normalized in such a way
746: that it goes to $0$ when interactions are weak and goes to $1$ in the
747: black limit \cite{yuri}. The high energy evolution of $N ({\un x}_0,
748: {\un x}_1, Y)$ is described by the non-linear evolution equation
749: \cite{yuri,bal}. Here we assume that $N$ has been found from that
750: evolution equation and is known to us.
751:
752: The graphical equation from \fig{nl} translates into
753: \ben
754: \frac{\partial}{\partial Y} {\cal O} ({\un x}_0, {\un x}_1, Y) \, = \,
755: \frac{\as \, N_c}{2 \, \pi^2} \, \int d^2 x_2 \,
756: \frac{x_{01}^2}{x_{02}^2 \, x_{12}^2} \, \left[ {\cal O} ({\un x}_0,
757: {\un x}_2, Y) + {\cal O} ({\un x}_2, {\un
758: x}_1, Y) - {\cal O} ({\un x}_0, {\un x}_1, Y) \right]
759: \een
760: \be\label{oddnl}
761: - \frac{\as \, N_c}{2 \, \pi^2} \, \int d^2 x_2 \,
762: \frac{x_{01}^2}{x_{02}^2 \, x_{12}^2} \, \left[ {\cal O} ({\un x}_0,
763: {\un x}_2, Y) \, N ({\un x}_2, {\un x}_1, Y) + N ({\un x}_0,
764: {\un x}_2, Y) \, {\cal O} ({\un x}_2, {\un x}_1, Y) \right].
765: \ee
766: The initial condition for the amplitude ${\cal O}$ in \eq{oddnl} is
767: given by
768: \be\label{inl1}
769: {\cal O} ({\un x}_0, {\un x}_1, Y=0) \, = \, c_0 \, \as^3 \, \rho \,
770: \int_{S_\perp} d^2 b \, T({\un b}) \, \ln^3
771: \left( \frac{{\un B} - {\un b} + \frac{1}{2} \, {\un x}_{01}}{{\un B} -
772: {\un b} - \frac{1}{2} \, {\un x}_{01}} \right) \, e^{- x_{01}^2 \,
773: Q_s^2 (x_{01}, B) /4},
774: \ee
775: where $\rho$ is the atomic number density in the nucleus, $T(\un b)$
776: is the nuclear profile function ($T(\un b) = 2 \sqrt{R^2 - b^2}$ for a
777: spherical nucleus of radius $R$) and ${\un B} = ({\un x}_0 + {\un
778: x}_1)/2$. It includes a three gluon exchange interaction with a
779: nucleon (modeled by a single quark (cf. \eq{inc2})) averaged over
780: transverse positions of the nucleon in the nucleus. (The $b$-integral
781: in \eq{inl1} is not zero since the integration is limited by
782: transverse nuclear size $S_\perp$ and does not go out to infinity.)
783: The exponential factor in Eq. (\ref{inl1}) accounts for multiple
784: $C$-even (two-gluon exchange) rescatterings of the dipole on other
785: nucleons in the nucleus
786: \cite{Mueller1,kjklw}, where
787: \be
788: Q_s^2 (x_{01}, B) = 4 \pi \, \as^2 \, \frac{C_F}{N_c} \, \rho \, T(\un B)
789: \, \ln \frac{1}{x_{01} \, \Lambda}
790: \ee
791: with $\Lambda$ some infrared cutoff.
792:
793:
794: \eq{oddnl} describes the odderon evolution including the effects of
795: saturation in the leading logarithmic approximation. While a detailed
796: analysis of \eq{oddnl} is beyond the scope of this paper, we note that
797: gluon saturation introduces negative terms on the right hand side of
798: \eq{oddnl} (the last two terms in it). Since the leading high energy
799: intercept of \eq{oddeq} corresponding to the first three terms on the
800: right hand side of (\ref{oddnl}) is zero, the negative non-zero last
801: two terms on the right hand side of (\ref{oddnl}) are likely to make
802: the odderon amplitude a decreasing function of energy/rapidity. This
803: result may be relevant to the lack of observation of QCD odderon in
804: the high energy DIS and $pp$ data.
805:
806:
807: \section{Discussion}
808:
809: In this paper we have shown that in the framework of the dipole model
810: both the Pomeron and the Odderon are described by the same evolution
811: equation (\ref{oddeq}). While the Pomeron evolution is projected out of
812: \eq{oddeq} by $C$-even initial conditions \cite{dip}, the Odderon evolution
813: is projected out by the $C$-odd ones. Since the complete set of
814: solutions to the BFKL equation is known, we have constructed the most
815: general Odderon solution for the dipole amplitude given by
816: \eq{gensol}. We have thus showed that there could be no other Odderon
817: solution for the case of a dipole scattering on any target. Our
818: solution (\ref{gensol}) is equivalent to the BLV solution \cite{BLV},
819: as shown in Sect. 3.
820:
821: In the usual Feynman diagram language the dipole model picks out a
822: part of the Odderon Green function with at least two out of three
823: $t$-channel gluons connecting to the same (anti)-quark line. Therefore
824: at least two gluons have the same transverse coordinate. This happens
825: for the Odderon exchange in DIS. However, in the case of $pp$ or
826: $p\bar p$ scattering there exist Feynman diagrams with the three
827: $t$-channel gluons coupling to different quarks at different
828: transverse coordinates. These diagrams have no equivalent in the
829: dipole model. One can not reproduce the solution of \cite{JW} in our
830: approach since it vanishes when two gluons are taken at the same
831: transverse point. While the dipole model allowed us to find the most
832: general Odderon solution for the case of dipole scattering on any
833: target, it did not produce any upper bound on the intercept of the
834: Odderon exchanged in the scattering of two three-quark states ($pp$ or
835: $p\bar p$).
836:
837:
838:
839: \section*{Acknowledgments}
840:
841: We want to thank Al Mueller for a number of critical discussions. We
842: are grateful to Ian Balitsky, Jochen Bartels, Misha Braun, Genya Levin
843: and Larry McLerran for encouraging us to study this
844: problem. L.Sz. thanks the Institute for Nuclear Theory and the Nuclear
845: Theory Group at the University of Washington for their hospitality and
846: the US Department of Energy for partial support during the beginning
847: of this work. He thanks also Oleg Teryaev for discussions and
848: hospitality at BLTP, JINR in Dubna, where this work was completed.
849:
850: The work of Yu.K. was supported in part by the U.S. Department of
851: Energy under Grant No. DE-FG03-97ER41014. The work of L.Sz. was
852: supported by the French--Polish Scientific Agreement Polonium.
853:
854:
855:
856: \renewcommand{\theequation}{A\arabic{equation}}
857: % redefine the command that creates the equation no.
858: \setcounter{equation}{0} % reset counter
859: \section*{Appendix A: Eigenfunctions of the Dipole Kernel}
860: \label{eigenftn}
861:
862: In this appendix we show that functions $E^{n, \nu} (x_0, x_1)$ are
863: eigenfunctions of the dipole kernel, such that
864: \be\label{goal}
865: \int d^2 x_2 \, \frac{x_{01}^2}{x_{02}^2 \,
866: x_{12}^2} \, \left[ E^{n, \nu} (x_0, x_2) + E^{n, \nu} (x_2, x_1) -
867: E^{n, \nu} (x_0, x_1) \right] \, = \, 4 \, \pi \, \chi(n, \nu) \,
868: E^{n, \nu} (x_0, x_1)
869: \ee
870: with $\chi(n, \nu)$ given by \eq{chi}. Our derivation below will be
871: closely following the Appendix of \cite{Lipatov1} (see also
872: \cite{DF}). In the complex notation the integral (\ref{goal}) can be
873: written as
874: \be\label{II}
875: I (\rho_0, \rho_0^*, \rho_1, \rho_1^*) \, \equiv \, \int d^2 \rho_2 \,
876: \frac{|\rho_{01}|^2}{|\rho_{02}|^2 \, |\rho_{12}|^2} \, \left[ E^{n, \nu}
877: (\rho_0, \rho_2) + E^{n, \nu} (\rho_2, \rho_1) - E^{n, \nu} (\rho_0,
878: \rho_1) \right]
879: \ee
880: where the asterisk denotes complex conjugation. Performing the
881: inversion transformation $\rho_i \rightarrow 1/\rho_i$ yields
882: \be\label{inv}
883: I (1/\rho_0, 1/\rho_0^*, 1/\rho_1, 1/\rho_1^*) \, = \, \int d^2 \rho_2
884: \, \frac{|\rho_{01}|^2}{|\rho_{02}|^2 \, |\rho_{12}|^2} \, \left[
885: \rho_{20}^h \, \rho_{20}^{* \, \bar h} + \rho_{12}^h \,
886: \rho_{12}^{* \, \bar h} - \rho_{10}^h \, \rho_{10}^{* \, \bar h} \right]
887: \ee
888: where $h = \frac{1+n}{2} + i \nu$ and ${\bar h} = \frac{1-n}{2} + i
889: \nu$. Defining a new variable
890: \be
891: R \, = \, \frac{1}{2} \, (\rho_{20} + \rho_{21})
892: \ee
893: we rewrite \eq{inv} as
894: \ben
895: I (1/\rho_0, 1/\rho_0^*, 1/\rho_1, 1/\rho_1^*) \, = \, \int d^2 R \,
896: \frac{|\rho_{01}|^2}{|R - \frac{1}{2} \rho_{01}|^2 \, |R + \frac{1}{2}
897: \rho_{01}|^2} \, \Bigg[ |R - \frac{1}{2} \rho_{01}|^{1 + 2 i \nu}
898: \een
899: \be\label{rint}
900: \times \,
901: \left( \frac{R - \frac{1}{2} \rho_{01}}{R^* - \frac{1}{2} \rho_{01}^*}
902: \right)^{n/2} + (-1)^n \, |R + \frac{1}{2} \rho_{01}|^{1 + 2 i \nu}
903: \left( \frac{R + \frac{1}{2} \rho_{01}}{R^* + \frac{1}{2} \rho_{01}^*}
904: \right)^{n/2} - \left. |\rho_{10}|^{1 + 2 i \nu} \left(
905: \frac{\rho_{10}}{\rho_{10}^*}
906: \right)^{n/2} \right].
907: \ee
908: Redefining the integration variable again such that $R = z \rho_{01}
909: /2$ we obtain
910: \ben
911: I (1/\rho_0, 1/\rho_0^*, 1/\rho_1, 1/\rho_1^*) \, = \, 4 \,
912: |\rho_{10}|^{1 + 2 i \nu} \left( \frac{\rho_{10}}{\rho_{10}^*}
913: \right)^{n/2} \, \int d^2 z \, \frac{1}{(z^2 - 1) \, (z^{* \, 2} - 1)} \,
914: \Bigg[ (-1)^n \, 2^{-1 - 2 i \nu}
915: \een
916: \be\label{zint}
917: \times \, \left. |z-1|^{1 + 2 i \nu} \, \left( \frac{z-1}{z^* -1}
918: \right)^{n/2} + 2^{-1 - 2 i \nu} \, |z+1|^{1 + 2 i \nu} \,
919: \left( \frac{z+1}{z^* +1} \right)^{n/2} - 1 \right].
920: \ee
921: Undoing the inversion transformation $\rho_i \rightarrow 1/\rho_i$ in
922: \eq{zint} gives
923: \be\label{eigen}
924: I (\rho_0, \rho_0^*, \rho_1, \rho_1^*) \, = \, C (n, \nu) \, E^{n,
925: \nu} (\rho_0, \rho_1)
926: \ee
927: with
928: \ben
929: C (n, \nu) \, = \, 4 \, \int d^2 z \, \frac{1}{(z^2 - 1) \, (z^{* \,
930: 2} - 1)} \, \left[ (-1)^n \, 2^{-1 - 2 i \nu} \, |z-1|^{1 + 2 i \nu} \,
931: \left( \frac{z-1}{z^* -1} \right)^{n/2} \right.
932: \een
933: \be\label{c1}
934: + \left. 2^{-1 - 2 i \nu} \, |z+1|^{1 + 2 i \nu} \, \left(
935: \frac{z+1}{z^* +1} \right)^{n/2} - 1 \right].
936: \ee
937: \eq{eigen} proves that functions $E^{n, \nu} (\rho_0, \rho_1)$ are,
938: indeed, eigenfunctions of the dipole kernel. To find the eigenvalue $C
939: (n, \nu)$ we follow the steps outlined in the Appendix of
940: \cite{Lipatov1}. Rewriting $z=x + i y$ with real $x$ and $y$ we perform
941: Wick rotation in the complex $y$-plane. The $y$-integral transforms
942: into an integral over $t = - i y$. Following \cite{Lipatov1} we define
943: $\alpha = x-t$ and $\beta = x+t$ obtaining
944: \ben
945: C (n, \nu) \, = \, 2 \, i \, \lim_{\sigma \rightarrow 0} \,
946: \int_{-\infty}^\infty d \alpha \, d \beta \,
947: \frac{1}{[(1-\alpha) (1-\beta) + i \epsilon ]^{1-\sigma} \,
948: [(1+\alpha) (1+\beta) + i \epsilon ]^{1-\sigma}}
949: \een
950: \ben
951: \times \, \left\{ 2^{-1 - 2 i \nu} \, (1-\alpha)^{|n|} \,
952: [(1-\alpha) (1-\beta)
953: + i \epsilon ]^{\frac{1-|n|}{2} + i \, \nu} \right.
954: \een
955: \be\label{c2}
956: \left. + 2^{-1 - 2 i \nu} \, (1+\alpha)^{|n|} \, [(1+\alpha) (1+\beta)
957: + i \epsilon ]^{\frac{1-|n|}{2} + i \, \nu} - 1 \right\},
958: \ee
959: where we introduced a dimensional regulator $\sigma$ to be able to
960: integrate the terms separately. The $\beta$-contour in \eq{c2} can be
961: distorted to give $0$ unless $-1 < \alpha <1$. Noting that the first
962: and the second terms in the curly brackets of \eq{c2} are identical we
963: rewrite it as
964: \ben
965: C (n, \nu) \, = \, 2 \, i \, \lim_{\sigma \rightarrow 0} \,
966: \int_{-1}^1 d \alpha \, \frac{1}{(1-\alpha^2)^{1-\sigma}} \,
967: \int_{-\infty}^\infty d \beta \,
968: \frac{1}{(1-\beta + i \epsilon )^{1-\sigma} \,
969: (1+\beta + i \epsilon )^{1-\sigma}}
970: \een
971: \be\label{c3}
972: \times \, \left\{ 2^{- 2 i \nu} \, (1-\alpha)^{\frac{1+|n|}{2} + i \, \nu}
973: \, (1-\beta + i \epsilon )^{\frac{1-|n|}{2} + i \, \nu} - 1 \right\}.
974: \ee
975: The $\beta$-integral in \eq{c3} can be done by distorting the contour
976: around one of the branch cuts. Performing $\alpha$-integral as well
977: and taking $\sigma \rightarrow 0$ limit of \eq{c3} yields
978: \be\label{c5}
979: C (n, \nu) \, = \, 4 \, \pi \, \chi (n, \nu).
980: \ee
981: Combining Eqs. (\ref{eigen}) and (\ref{c5}) with \eq{II} gives us
982: \eq{goal}, as desired.
983:
984:
985: \renewcommand{\theequation}{B\arabic{equation}}
986: % redefine the command that creates the equation no.
987: \setcounter{equation}{0} % reset counter
988: \section*{Appendix B: Three-Gluon Exchange Amplitude}
989: \label{dec}
990:
991: Here we decompose the function
992: \be
993: \ln^3 \bigg| \frac{\rho_{11'} \, \rho_{22'}}{\rho_{12'} \,
994: \rho_{1'2}} \bigg|
995: \ee
996: in the series of \eq{decom}. First we note that
997: \ben
998: \nabla_1^2 \, \nabla_2^2 \,
999: \ln^3 \bigg| \frac{\rho_{11'} \, \rho_{22'}}{\rho_{12'} \,
1000: \rho_{1'2}} \bigg| \, = \, 6 \, \pi \, \left[ \delta (\rho_{22'}) -
1001: \delta (\rho_{21'}) \right] \, \int d^2 \rho_3 \, \frac{|\rho_{1'2'}|^2}{
1002: |\rho_{31'}|^2 \, |\rho_{32'}|^2} \, \left[ 2 \, \delta (\rho_{31}) -
1003: \delta (\rho_{11'}) \right]
1004: \een
1005: \be\label{st1}
1006: + 6 \, \pi \, \left[ \delta (\rho_{11'}) -
1007: \delta (\rho_{12'}) \right] \, \int d^2 \rho_3 \, \frac{|\rho_{1'2'}|^2}{
1008: |\rho_{31'}|^2 \, |\rho_{32'}|^2} \, \left[ 2 \, \delta (\rho_{32}) -
1009: \delta (\rho_{22'}) \right].
1010: \ee
1011: Then, employing (see Eq. (25) in \cite{Lipatov1})
1012: \be
1013: (2 \pi)^4 \, \delta (\rho_{11'}) \, \delta (\rho_{22'}) \, = \,
1014: \sum_{n=-\infty}^\infty \, \int_{-\infty}^\infty d \nu \, \int d^2
1015: \rho_0 \, \frac{16 \left(\nu^2 + \frac{n^2}{4}\right)}{|\rho_{12}|^2 \,
1016: |\rho_{1'2'}|^2} \, E^{n, \nu} (\rho_{10}, \rho_{20}) \,
1017: E^{n, \nu \, *} (\rho_{1'0}, \rho_{2'0})
1018: \ee
1019: in \eq{st1} and remembering that (see Eqs. (21) of \cite{Lipatov1})
1020: \be
1021: \nabla_1^2 \, \nabla_2^2 \, E^{n, \nu} (\rho_{10}, \rho_{20}) \, = \,
1022: \frac{16}{|\rho_{12}|^4} \, \left[ \nu^2 + \frac{(n+1)^2}{4}
1023: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4} \right] \,
1024: E^{n, \nu} (\rho_{10}, \rho_{20})
1025: \ee
1026: yields
1027: \ben
1028: \ln^3 \bigg| \frac{\rho_{11'} \, \rho_{22'}}{\rho_{12'} \,
1029: \rho_{1'2}} \bigg| \, = \, \frac{3}{2 \, \pi^3} \, \sum_{\mbox{odd n}}
1030: \, \int_{-\infty}^\infty d \nu \, \int d^2
1031: \rho_0 \, \frac{\nu^2 + \frac{n^2}{4}}{\left[ \nu^2 + \frac{(n+1)^2}{4}
1032: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4} \, \right]} \,
1033: E^{n, \nu} (\rho_{10}, \rho_{20})
1034: \een
1035: \be\label{st2}
1036: \times \, \int d^2 \rho_3 \, \frac{|\rho_{1'2'}|^2}{
1037: |\rho_{31'}|^2 \, |\rho_{32'}|^2} \, \left[ E^{n, \nu \, *} (\rho_{1'0},
1038: \rho_{30}) + E^{n, \nu \, *} (\rho_{30}, \rho_{2'0}) -
1039: E^{n, \nu \, *} (\rho_{1'0}, \rho_{2' 0}) \right]
1040: \ee
1041: Using \eq{goal} to evaluate \eq{st2} gives the final answer
1042: \ben
1043: \ln^3 \bigg| \frac{\rho_{11'} \, \rho_{22'}}{\rho_{12'} \,
1044: \rho_{1'2}} \bigg| \, = \, \frac{6}{\pi^2} \, \sum_{\mbox{odd n}}
1045: \, \int_{-\infty}^\infty d \nu \, \int d^2
1046: \rho_0 \, \frac{\nu^2 + \frac{n^2}{4}}{\left[ \nu^2 + \frac{(n+1)^2}{4}
1047: \right] \, \left[ \nu^2 + \frac{(n-1)^2}{4} \, \right]} \,
1048: \een
1049: \be\label{st3}
1050: \times \, \chi (n, \nu) \,
1051: E^{n, \nu} (\rho_{10}, \rho_{20}) \, E^{n, \nu \, *} (\rho_{1'0},
1052: \rho_{2' 0}).
1053: \ee
1054:
1055:
1056: \begin{thebibliography}{99}
1057:
1058: \bibitem{LukNic}
1059: %\cite{Lukaszuk:1973nt}
1060: %\bibitem{Lukaszuk:1973nt}
1061: L.~Lukaszuk and B.~Nicolescu,
1062: %``A Possible Interpretation Of P P Rising Total Cross-Sections,''
1063: Lett.\ Nuovo Cim.\ {\bf 8} (1973) 405.
1064: %%CITATION = NCLTA,8,405;%%
1065:
1066:
1067: \bibitem{Nic}
1068: %\cite{Nicolescu:1990ii}
1069: %\bibitem{Nicolescu:1990ii}
1070: B.~Nicolescu,
1071: ``The Odderon Today,''
1072: IPNO/TH-90-40
1073: %\href{http://www.slac.stanford.edu/spires/find/hep/www?r=ipno\%2Fth-90-40}{SPIRES
1074: %entry}
1075: {\it Presented at 25th Rencontre de Moriond: High Energy Hadronic
1076: Interactions, Les Arcs, France, Mar 11-17, 1990}
1077:
1078:
1079: \bibitem{BFKL}
1080: E.A. Kuraev, L.N. Lipatov and V.S. Fadin, Sov. JETP {\bf 44} (1976) 443; {\it
1081: ibid} {\bf 45} (1977) 199; \\
1082: Ya.Ya. Balitsky and L.N. Lipatov, Sov. J. Nucl.
1083: Phys. {\bf 28} (1978) 822.
1084:
1085: \bibitem{BJKP}
1086: J. Bartels, Nucl. Phys. B {\bf 151} (1979) 293; {\it ibid} B {\bf 175} (1980)
1087: 413; \\
1088: T.~Jaroszewicz,
1089: %``Infrared Divergences And Regge Behavior In QCD,''
1090: Acta Phys.\ Polon.\ B {\bf 11}, 965 (1980);
1091: \\
1092: J. Kwiecinski and M. Praszalowicz, Phys. Lett. B {\bf 94} (1980) 413.
1093:
1094:
1095: %\bibitem{Lip86}
1096: %\cite{Lipatov:1985uk}
1097: \bibitem{Lipatov1}
1098: L.~N.~Lipatov,
1099: %``The Bare Pomeron In Quantum Chromodynamics,''
1100: Sov.\ Phys.\ JETP {\bf 63}, 904 (1986)
1101: [Zh.\ Eksp.\ Teor.\ Fiz.\ {\bf 90}, 1536 (1986)].
1102: %%CITATION = SPHJA,63,904;%%
1103: %L.N. Lipatov, Sov. Phys. JETP {\bf 63} (1986) 904
1104:
1105: \bibitem{Lev93}
1106: %\cite{Lipatov:1993qn}
1107: %\bibitem{Lipatov:1993qn}
1108: L.~N.~Lipatov,
1109: %``High-energy asymptotics of multicolor QCD and two-dimensional conformal
1110: %field theories,''
1111: Phys.\ Lett.\ B {\bf 309} (1993) 394.
1112: %%CITATION = PHLTA,B309,394;%%
1113:
1114: \bibitem{Lev94}
1115: %\cite{Lipatov:1993yb}
1116: %\bibitem{Lipatov:1993yb}
1117: L.~N.~Lipatov,
1118: %``High-energy asymptotics of multicolor QCD and exactly solvable lattice
1119: %models,''
1120: JETP Lett.\ {\bf 59} (1994) 596
1121: [Pisma Zh.\ Eksp.\ Teor.\ Fiz.\ {\bf 59} (1994) 571]
1122: [arXiv:hep-th/9311037].
1123: %%CITATION = HEP-TH 9311037;%%
1124:
1125:
1126: \bibitem{Korch95}
1127: %\cite{Faddeev:1994zg}
1128: %\bibitem{Faddeev:1994zg}
1129: L.~D.~Faddeev and G.~P.~Korchemsky,
1130: %``High-energy QCD as a completely integrable model,''
1131: Phys.\ Lett.\ B {\bf 342} (1995) 311
1132: [arXiv:hep-th/9404173].
1133: %%CITATION = HEP-TH 9404173;%%
1134:
1135:
1136: \bibitem{JW}
1137: %\cite{Janik:1998xj}
1138: %\bibitem{Janik:1998xj}
1139: R.~A.~Janik and J.~Wosiek,
1140: %``Solution of the odderon problem,''
1141: Phys.\ Rev.\ Lett.\ {\bf 82} (1999) 1092
1142: [arXiv:hep-th/9802100].
1143: %%CITATION = HEP-TH 9802100;%%
1144:
1145:
1146: \bibitem{KKM}
1147: %\cite{Korchemsky:2001nx}
1148: %\bibitem{Korchemsky:2001nx}
1149: G.~P.~Korchemsky, J.~Kotanski and A.~N.~Manashov,
1150: %``Compound states of reggeized gluons in multi-colour QCD as ground states
1151: %of noncompact Heisenberg magnet,''
1152: Phys.\ Rev.\ Lett.\ {\bf 88} (2002) 122002
1153: [arXiv:hep-ph/0111185].
1154: %%CITATION = HEP-PH 0111185;%%
1155:
1156:
1157: %\cite{Bartels:1999yt}
1158: \bibitem{BLV}
1159: J.~Bartels, L.~N.~Lipatov and G.~P.~Vacca,
1160: %``A new odderon solution in perturbative QCD,''
1161: Phys.\ Lett.\ B {\bf 477}, 178 (2000)
1162: [arXiv:hep-ph/9912423].
1163: %%CITATION = HEP-PH 9912423;%%
1164:
1165:
1166: \bibitem{DKKM2}
1167: %\cite{Derkachov:2002wz}
1168: %\bibitem{Derkachov:2002wz}
1169: S.~E.~Derkachov, G.~P.~Korchemsky, J.~Kotanski and A.~N.~Manashov,
1170: %``Noncompact Heisenberg spin magnets from high-energy QCD. II: Quantization
1171: %conditions and energy spectrum,''
1172: Nucl.\ Phys.\ B {\bf 645} (2002) 237
1173: [arXiv:hep-th/0204124].
1174: %%CITATION = HEP-TH 0204124;%%
1175:
1176: \bibitem{dip-pp}
1177: %\cite{Dosch:2002ai}
1178: %\bibitem{Dosch:2002ai}
1179: H.~G.~Dosch, C.~Ewerz and V.~Schatz,
1180: %``The odderon in high energy elastic p p scattering,''
1181: Eur.\ Phys.\ J.\ C {\bf 24} (2002) 561
1182: [arXiv:hep-ph/0201294].
1183: %%CITATION = HEP-PH 0201294;%%
1184:
1185:
1186: \bibitem{eta}
1187: %\cite{Czyzewski:1996bv}
1188: %\bibitem{Czyzewski:1996bv}
1189: J.~Czyzewski, J.~Kwiecinski, L.~Motyka and M.~Sadzikowski,
1190: %``Exclusive eta/c photo- and electroproduction at HERA as a possible probe
1191: %of the odderon singularity in QCD,''
1192: Phys.\ Lett.\ B {\bf 398} (1997) 400
1193: [Erratum-ibid.\ B {\bf 411} (1997) 402]
1194: [arXiv:hep-ph/9611225];
1195: %%CITATION = HEP-PH 9611225;%%
1196: \\
1197: %\cite{Engel:1997cg}
1198: %\bibitem{Engel:1997cg}
1199: R.~Engel, D.~Y.~Ivanov, R.~Kirschner and L.~Szymanowski,
1200: %``Diffractive meson production from virtual photons with odd charge-parity
1201: %exchange,''
1202: Eur.\ Phys.\ J.\ C {\bf 4} (1998) 93
1203: [arXiv:hep-ph/9707362].
1204: %%CITATION = HEP-PH 9707362;%%
1205:
1206:
1207: \bibitem{BBCV}
1208: %\cite{Bartels:2001hw}
1209: %\bibitem{Bartels:2001hw}
1210: J.~Bartels, M.~A.~Braun, D.~Colferai and G.~P.~Vacca,
1211: %``Diffractive eta/c photo- and electroproduction with the perturbative QCD
1212: %odderon,''
1213: Eur.\ Phys.\ J.\ C {\bf 20} (2001) 323
1214: [arXiv:hep-ph/0102221].
1215: %%CITATION = HEP-PH 0102221;%%
1216:
1217:
1218: \bibitem{HPST}
1219: %\cite{Hagler:2002nh}
1220: %\bibitem{Hagler:2002nh}
1221: P.~Hagler, B.~Pire, L.~Szymanowski and O.~V.~Teryaev,
1222: %``Hunting the QCD-odderon in hard diffractive electroproduction of two
1223: %pions,''
1224: Phys.\ Lett.\ B {\bf 535} (2002) 117
1225: [Erratum-ibid.\ B {\bf 540} (2002) 324]
1226: [arXiv:hep-ph/0202231].
1227: %%CITATION = HEP-PH 0202231;%%
1228: \\
1229: %\cite{Hagler:2002nf}
1230: %\bibitem{Hagler:2002nf}
1231: P.~Hagler, B.~Pire, L.~Szymanowski and O.~V.~Teryaev,
1232: %``Pomeron odderon interference effects in electroproduction of two pions,''
1233: Eur.\ Phys.\ J.\ C {\bf 26} (2002) 261
1234: [arXiv:hep-ph/0207224].
1235: %%CITATION = HEP-PH 0207224;%%
1236: \\
1237: %\cite{Ginzburg:2002zd}
1238: %\bibitem{Ginzburg:2002zd}
1239: I.~F.~Ginzburg, I.~P.~Ivanov and N.~N.~Nikolaev,
1240: %``Possible odderon discovery via observation of charge asymmetry in the
1241: %diffractive pi+ pi- production at HERA,''
1242: Eur.\ Phys.\ J.\ directC {\bf 5} (2003) 02
1243: [Eur.\ Phys.\ J.\ directC {\bf 30} (2003) 002]
1244: [arXiv:hep-ph/0207345].
1245: %%CITATION = HEP-PH 0207345;%%
1246:
1247: \bibitem{BBV}
1248: %\cite{Bartels:2003zu}
1249: %\bibitem{Bartels:2003zu}
1250: J.~Bartels, M.~A.~Braun and G.~P.~Vacca,
1251: %``The process gamma(*) + p $\to$ eta/c + X: A test for the perturbative QCD
1252: %odderon,''
1253: arXiv:hep-ph/0304160.
1254: %%CITATION = HEP-PH 0304160;%%
1255:
1256:
1257:
1258: \bibitem{Ewerz}
1259: %\cite{Ewerz:2003xi}
1260: %\bibitem{Ewerz:2003xi}
1261: C.~Ewerz,
1262: %``The odderon in quantum chromodynamics,''
1263: arXiv:hep-ph/0306137.
1264: %%CITATION = HEP-PH 0306137;%%
1265:
1266:
1267:
1268: \bibitem{dip}
1269: %\cite{Mueller:1993rr}
1270: %\bibitem{Mueller:1993rr}
1271: A.~H.~Mueller,
1272: %``Soft gluons in the infinite momentum wave function and the BFKL pomeron,''
1273: Nucl.\ Phys.\ B {\bf 415}, 373 (1994);
1274: %%CITATION = NUPHA,B415,373;%%
1275: %\cite{Mueller:1994jq}
1276: %\bibitem{Mueller:1994jq}
1277: A.~H.~Mueller and B.~Patel,
1278: %``Single and double BFKL pomeron exchange and a dipole picture of high-energy hard processes,''
1279: Nucl.\ Phys.\ B {\bf 425}, 471 (1994)
1280: [arXiv:hep-ph/9403256];
1281: %%CITATION = HEP-PH 9403256;%%
1282: %\cite{Mueller:gb}
1283: %\bibitem{Mueller:gb}
1284: A.~H.~Mueller,
1285: %``Unitarity And The Bfkl Pomeron,''
1286: Nucl.\ Phys.\ B {\bf 437}, 107 (1995)
1287: [arXiv:hep-ph/9408245].
1288: %%CITATION = HEP-PH 9408245;%%
1289:
1290:
1291: \bibitem{dipBFKL}
1292: %\cite{Navelet:1997tx}
1293: %\bibitem{Navelet:1997tx}
1294: H.~Navelet and S.~Wallon,
1295: %``Onium onium scattering at fixed impact parameter: Exact equivalence
1296: %between the color dipole model and the BFKL pomeron,''
1297: Nucl.\ Phys.\ B {\bf 522} (1998) 237
1298: [arXiv:hep-ph/9705296].
1299: %%CITATION = HEP-PH 9705296;%%
1300:
1301:
1302: %\cite{Gribov:tu}
1303: \bibitem{glr}
1304: L.~V.~Gribov, E.~M.~Levin and M.~G.~Ryskin,
1305: %``Semihard Processes In QCD,''
1306: Phys.\ Rept.\ {\bf 100}, 1 (1983);
1307: %%CITATION = PRPLC,100,1;%%
1308: %\cite{Mueller:wy}
1309: %\bibitem{mq}
1310: A.~H.~Mueller and J.~w.~Qiu,
1311: %``Gluon Recombination And Shadowing At Small Values Of X,''
1312: Nucl.\ Phys.\ B {\bf 268}, 427 (1986).
1313: %%CITATION = NUPHA,B268,427;%%
1314:
1315:
1316: %\cite{Kovchegov:1999yj}
1317: \bibitem{yuri}
1318: Y.~V.~Kovchegov,
1319: %``Small-x F2 structure function of a nucleus including multiple pomeron exchanges,''
1320: Phys.\ Rev.\ D {\bf 60}, 034008 (1999)
1321: [arXiv:hep-ph/9901281]; \\
1322: %%CITATION = HEP-PH 9901281;%%
1323: %\cite{Kovchegov:1999ua}
1324: %\bibitem{yuri2}
1325: Y.~V.~Kovchegov,
1326: %``Unitarization of the BFKL pomeron on a nucleus,''
1327: Phys.\ Rev.\ D {\bf 61}, 074018 (2000)
1328: [arXiv:hep-ph/9905214].
1329: %%CITATION = HEP-PH 9905214;%%
1330:
1331: \bibitem{CM}
1332: %\cite{Chen:1995pa}
1333: %\bibitem{Chen:1995pa}
1334: Z.~Chen and A.~H.~Mueller,
1335: %``The Dipole picture of high-energy scattering, the BFKL equation and many gluon compound states,''
1336: Nucl.\ Phys.\ B {\bf 451}, 579 (1995).
1337: %%CITATION = NUPHA,B451,579;%%
1338:
1339:
1340: \bibitem{bal}
1341: %\cite{Balitsky:1995ub}
1342: %\bibitem{Balitsky:1995ub}
1343: I.~Balitsky,
1344: %``Operator expansion for high-energy scattering,''
1345: Nucl.\ Phys.\ B {\bf 463}, 99 (1996)
1346: [arXiv:hep-ph/9509348];
1347: %%CITATION = HEP-PH 9509348;%%
1348: %\cite{Balitsky:1997mk}
1349: %\bibitem{Balitsky:1997mk}
1350: %I.~Balitsky,
1351: %``Operator expansion for diffractive high-energy scattering,''
1352: arXiv:hep-ph/9706411;
1353: %%CITATION = HEP-PH 9706411;%%
1354: %\cite{Balitsky:1998ya}
1355: %\bibitem{Balitsky:1998ya}
1356: %I.~Balitsky,
1357: %``Factorization and high-energy effective action,''
1358: Phys.\ Rev.\ D {\bf 60}, 014020 (1999)
1359: [arXiv:hep-ph/9812311].
1360: %%CITATION = HEP-PH 9812311;%%
1361:
1362: \bibitem{JKLW}
1363: %\cite{Jalilian-Marian:1997jx}
1364: %\bibitem{Jalilian-Marian:1997jx}
1365: J.~Jalilian-Marian, A.~Kovner, A.~Leonidov and H.~Weigert,
1366: %``The BFKL equation from the Wilson renormalization group,''
1367: Nucl.\ Phys.\ B {\bf 504}, 415 (1997)
1368: [arXiv:hep-ph/9701284];
1369: %%CITATION = HEP-PH 9701284;%%
1370: %\cite{Jalilian-Marian:1997gr}
1371: %\bibitem{Jalilian-Marian:1997gr}
1372: %J.~Jalilian-Marian, A.~Kovner, A.~Leonidov and H.~Weigert,
1373: %``The Wilson renormalization group for low x physics: Towards the high density regime,''
1374: Phys.\ Rev.\ D {\bf 59}, 014014 (1999)
1375: [arXiv:hep-ph/9706377];
1376: %%CITATION = HEP-PH 9706377;%%
1377: %\cite{Jalilian-Marian:1998cb}
1378: %\bibitem{Jalilian-Marian:1998cb}
1379: %J.~Jalilian-Marian, A.~Kovner, A.~Leonidov and H.~Weigert,
1380: %``Unitarization of gluon distribution in the doubly logarithmic regime at high density,''
1381: Phys.\ Rev.\ D {\bf 59}, 034007 (1999)
1382: [Erratum-ibid.\ D {\bf 59}, 099903 (1999)]
1383: [arXiv:hep-ph/9807462];
1384: %%CITATION = HEP-PH 9807462;%%
1385: %\cite{Jalilian-Marian:1997dw}
1386: %\bibitem{Jalilian-Marian:1997dw}
1387: J.~Jalilian-Marian, A.~Kovner and H.~Weigert,
1388: %``The Wilson renormalization group for low x physics: Gluon evolution at finite parton density,''
1389: Phys.\ Rev.\ D {\bf 59}, 014015 (1999)
1390: [arXiv:hep-ph/9709432];
1391: %%CITATION = HEP-PH 9709432;%%
1392: %\cite{Kovner:2000pt}
1393: %\bibitem{Kovner:2000pt}
1394: A.~Kovner, J.~G.~Milhano and H.~Weigert,
1395: %``Relating different approaches to nonlinear QCD evolution at finite gluon density,''
1396: Phys.\ Rev.\ D {\bf 62}, 114005 (2000)
1397: [arXiv:hep-ph/0004014];
1398: %%CITATION = HEP-PH 0004014;%%
1399: %\cite{Weigert:2000gi}
1400: %\bibitem{Weigert:2000gi}
1401: H.~Weigert,
1402: %``Unitarity at small Bjorken x,''
1403: Nucl.\ Phys.\ A {\bf 703}, 823 (2002)
1404: [arXiv:hep-ph/0004044].
1405: %%CITATION = HEP-PH 0004044;%%
1406:
1407: \bibitem{FILM}
1408: %\cite{Iancu:2000hn}
1409: %\bibitem{Iancu:2000hn}
1410: E.~Iancu, A.~Leonidov and L.~D.~McLerran,
1411: %``Nonlinear gluon evolution in the color glass condensate. I,''
1412: Nucl.\ Phys.\ A {\bf 692}, 583 (2001)
1413: [arXiv:hep-ph/0011241];
1414: %%CITATION = HEP-PH 0011241;%%
1415: %\cite{Iancu:2001ad}
1416: %\bibitem{Iancu:2001ad}
1417: %E.~Iancu, A.~Leonidov and L.~D.~McLerran,
1418: %``The renormalization group equation for the color glass condensate,''
1419: Phys.\ Lett.\ B {\bf 510}, 133 (2001)
1420: [arXiv:hep-ph/0102009];
1421: %%CITATION = HEP-PH 0102009;%%
1422: %\cite{Iancu:2001md}
1423: %\bibitem{Iancu:2001md}
1424: E.~Iancu and L.~D.~McLerran,
1425: %``Saturation and universality in QCD at small x,''
1426: Phys.\ Lett.\ B {\bf 510}, 145 (2001)
1427: [arXiv:hep-ph/0103032];
1428: %%CITATION = HEP-PH 0103032;%%
1429: %\cite{Ferreiro:2001qy}
1430: %\bibitem{Ferreiro:2001qy}
1431: E.~Ferreiro, E.~Iancu, A.~Leonidov and L.~McLerran,
1432: %``Nonlinear gluon evolution in the color glass condensate. II,''
1433: Nucl.\ Phys.\ A {\bf 703}, 489 (2002)
1434: [arXiv:hep-ph/0109115].
1435: %%CITATION = HEP-PH 0109115;%%
1436:
1437: %\cite{Braun:2000wr}
1438: \bibitem{Braun1}
1439: M.~Braun,
1440: %``Structure function of the nucleus in the perturbative QCD with N(c) $\to$ infinity (BFKL pomeron fan diagrams),''
1441: Eur.\ Phys.\ J.\ C {\bf 16}, 337 (2000)
1442: [arXiv:hep-ph/0001268].
1443: %%CITATION = HEP-PH 0001268;%%
1444:
1445: %\cite{Mueller:st}
1446: \bibitem{Mueller1}
1447: A.~H.~Mueller,
1448: %``Small X Behavior And Parton Saturation: A QCD Model,''
1449: Nucl.\ Phys.\ B {\bf 335}, 115 (1990).
1450: %%CITATION = NUPHA,B335,115;%%
1451:
1452: \bibitem {mv}
1453: %\cite{McLerran:1993ni}
1454: %\bibitem{McLerran:1993ni}
1455: L.~D.~McLerran and R.~Venugopalan,
1456: %``Computing quark and gluon distribution functions for very large nuclei,''
1457: Phys.\ Rev.\ D {\bf 49}, 2233 (1994)
1458: [arXiv:hep-ph/9309289];
1459: %%CITATION = HEP-PH 9309289;%%
1460: %\cite{McLerran:1993ka}
1461: %\bibitem{McLerran:1993ka}
1462: %L.~D.~McLerran and R.~Venugopalan,
1463: %``Gluon distribution functions for very large nuclei at small transverse momentum,''
1464: Phys.\ Rev.\ D {\bf 49}, 3352 (1994)
1465: [arXiv:hep-ph/9311205];
1466: %%CITATION = HEP-PH 9311205;%%
1467: %\cite{McLerran:1994vd}
1468: %\bibitem{McLerran:1994vd}
1469: %L.~D.~McLerran and R.~Venugopalan,
1470: %``Green's functions in the color field of a large nucleus,''
1471: Phys.\ Rev.\ D {\bf 50}, 2225 (1994)
1472: [arXiv:hep-ph/9402335].
1473: %%CITATION = HEP-PH 9402335;%%
1474:
1475: \bibitem{kjklw}
1476: %\cite{Kovchegov:1996ty}
1477: %\bibitem{Kovchegov:1996ty}
1478: Y.~V.~Kovchegov,
1479: %``Non-Abelian Weizsaecker-Williams field and a two-dimensional effective
1480: %color charge density for a very large nucleus,''
1481: Phys.\ Rev.\ D {\bf 54}, 5463 (1996)
1482: [arXiv:hep-ph/9605446];
1483: %%CITATION = HEP-PH 9605446;%%
1484: %\cite{Kovchegov:1997pc}
1485: %\bibitem{Kovchegov:1997pc}
1486: %Y.~V.~Kovchegov,
1487: %``Quantum structure of the non-Abelian Weizsaecker-Williams field for a very large nucleus,''
1488: Phys.\ Rev.\ D {\bf 55}, 5445 (1997)
1489: [arXiv:hep-ph/9701229];
1490: %%CITATION = HEP-PH 9701229;%%
1491: %\cite{Jalilian-Marian:1996xn}
1492: %\bibitem{Jalilian-Marian:1996xn}
1493: J.~Jalilian-Marian, A.~Kovner, L.~D.~McLerran and H.~Weigert,
1494: %``The intrinsic glue distribution at very small x,''
1495: Phys.\ Rev.\ D {\bf 55}, 5414 (1997)
1496: [arXiv:hep-ph/9606337].
1497: %%CITATION = HEP-PH 9606337;%%
1498:
1499: %\cite{Dotsenko:1984nm}
1500: \bibitem{DF}
1501: V.~S.~Dotsenko and V.~A.~Fateev,
1502: %``Conformal Algebra And Multipoint Correlation Functions In 2d Statistical Models,''
1503: Nucl.\ Phys.\ B {\bf 240}, 312 (1984);
1504: %%CITATION = NUPHA,B240,312;%%
1505: %\cite{Dotsenko:1986ca}
1506: %\bibitem{Dotsenko:1986ca}
1507: V.~S.~Dotsenko, {\it Lectures on conformal field theory}, Advanced
1508: Studies in Pure Mathematics {\bf 16}, 123 (1988).
1509: %``Lectures On Conformal Field Theory,''
1510: %RIMS-559
1511: %\href{http://www.slac.stanford.edu/spires/find/hep/www?r=rims-559}{SPIRES entry}
1512:
1513:
1514: \end{thebibliography}
1515:
1516:
1517: \end{document}
1518: