1: \section{Quantum dynamics on a complex-time contour }
2:
3: In this section, we describe the action $S[A,\,\rho]$ which
4: enters the quantum McLerran-Venugopalan model,
5: eq.~(\ref{PART}), and derive Feynman rules for the
6: computation of $\sigma$ and $\chi$.
7: Because of the temporal non-locality of the classical equations
8: of motion (\ref{cleq1}), this action, and the
9: associated quantum dynamics, must be formulated along a
10: contour in the complex-time plane.
11: It turns out, however, that in the approximations of interest
12: the contour structure is not essential, and one can restrict
13: oneself to the dynamics in real time, as done a priori
14: in Refs. \cite{JKLW97,JKLW99a}.
15:
16:
17: \subsection{The contour action}
18:
19: By definition, the action $S[A,\,\rho]$ must be such as
20: to reproduce the classical equations of motion
21: (\ref{cleq1}) at the tree-level. Thus,
22: $S[A,\,\rho]$ has the following structure:
23: \be\labe{ACTION0}
24: S[A,\rho]\,=\,- \int d^4x \,{1 \over 4} F_{\mu\nu}^a F^{\mu\nu}_a
25: \,+\,S_W[A^-,\rho]\,,
26: \ee
27: with $S_W[A^-,\rho]$ constrained by (cf. eqs.~(\ref{JTIME}) and
28: (\ref{WLINE1}))
29: \be\labe{SWJ}
30: \frac{\delta S_W}{\delta A^-_a(x)}=-\,J^+_a(x)\,\equiv
31: \,-\,
32: \frac{1}{N_c}\,{\rm Tr}\left\{\rho(\vec x)
33: W^\dagger(x)\,T^a\,W(x)\right\}.\ee
34: However, with the retarded current $J^+(x)$ in
35: eqs.~(\ref{JTIME})--(\ref{WLINE1}),
36: the constraint (\ref{SWJ}) cannot be satisfied,
37: for any action. For instance,
38: eq.~(\ref{SWJ}) implies the following symmetry property,
39: or {\it Maxwell relation} :
40: \be
41: \frac{\delta J^+_a(x)}{\delta A^-_b(y)}
42: \,=\,\frac{-\delta^2 S_W}{\delta A^-_a(x) \delta A^-_b(y)}
43: \,=\,\frac{\delta J^+_b(y)}{\delta A^-_a(x)},\ee
44: which is however inconsistent with the current (\ref{JTIME}),
45: which rather yields
46: \be
47: \frac{\delta J^+_a(x)}{\delta A^-_b(y)}\bigg |_{A^-=0}&=&
48: \frac{ig}{N_c}\,\theta(x^+-y^+)\,\delta^{(3)}(\vec x-\vec y)\,
49: {\rm Tr}\,\Bigl\{\rho(\vec x)[T^a,T^b]\Bigr\}\nonumber\\
50: &=&-gf^{abc}\rho_c(\vec x)\,\theta(x^+-y^+)\,
51: \delta^{(3)}(\vec x-\vec y)
52: \,\equiv\,\Pi_{ab}^{R}(x,y).\ee
53: That is, the two-point polarization function $\Pi^R$,
54: and all the higher vertices generated by
55: $J^+(x)$, are necessarily retarded with respect to $x^+$,
56: while the corresponding amplitudes generated by $S_W$ should be
57: rather symmetrical (e.g., $\Pi_{ab}(x,y)= \Pi_{ba}(y,x)$).
58:
59: Thus, the non-local equation
60: (\ref{cleq1}) cannot be generated from an action.
61: This reflects the fact that this is an {\it effective}
62: equation, obtained after some degrees of freedom
63: (the fast partons) have been integrated out,
64: with retarded boundary conditions, to generate the non-local
65: current in the r.h.s. There is a standard procedure to circumvent
66: this difficulty and construct a quantum version of the non-local
67: theory (see, e.g., Ref. \cite{BI00} and Refs. therein):
68: This involves a generalization of eq.~(\ref{cleq1}) to complex
69: time variables which take values along the Schwinger-Keldysh
70: contour depicted in Fig.~\ref{CONT}.
71:
72: \begin{figure}
73: \protect \epsfxsize=16.cm{\centerline{\epsfbox{CONTX.eps}}}
74: \caption{Complex-time contour for the quantum MV model:
75: $C=C_+\cup C_-$.}
76: \labe{CONT}
77: \end{figure}
78:
79: We call $z$ the (complex) time variable along the contour,
80: and reserve the notation $x^+$ for real times.
81: The contour $C$ may be seen as the juxtaposition of two pieces:
82: $C=C_+\cup C_-$. On $C_+$, $z=x^+$ takes all the
83: real values between $x^+_0$ and $x^+_f$
84: (eventually, we let
85: $x^+_0 \to -\infty$ and $x^+_f\to \infty$).
86: On $C_-$, $z=x^+-i\eta$, with $\eta\to 0_+$
87: and $x^+$ running backward from $x^+_f$ to $x^+_0$.
88: We define a contour $\theta$-function $\theta_C\,$:
89: $\theta_C(z_1,z_2)=1$ if $z_1$ is further than $z_2$
90: along the contour (we write then $z_1 \succ z_2$),
91: while $\theta_C(z_1,z_2)=0$ if the opposite situation holds
92: ($z_1 \prec z_2$).
93: We can formalize this by introducing a real parameter $u$ which
94: is continously increasing along the contour; then, the contour $C$
95: is specified by a function $z(u)$, and $\theta_C(z_1,z_2)=
96: \theta(u_1-u_2)$. The contour $\delta$-function is defined by:
97: \be
98: \delta_C(z_1,z_2)\equiv \left(\frac {\del z}{\del u}\right)^{-1}
99: \delta(u_1-u_2).\ee
100: The quantum extension of the MV model is a quantum
101: field theory living on this contour. Specifically, we allow the
102: quantum fluctuations of the fields $A^\mu$ to take arbitrary
103: values on both sides of the contour,
104: and define a contour ``evolution operator'' by (for $z_2 \succ z_1$)
105: \be\labe{WLC0}
106: W_{z_2,z_1}(\vec x)\,\equiv\,{\rm T}_C\, \exp\left\{\,
107: ig\int_{z_1}^{z_2} dz A^-(z, \vec x) \right\},
108: \ee
109: where $\int_{z_1}^{z_2}$ is the integral running along $C$ from $z_1$
110: to $z_2$, and the operator ${\rm T}_C$
111: orders the color matrices $A^-(z)$ from right to left in
112: increasing sequence of their $u$ arguments. (Note
113: that the ordering along the contour coincides
114: with the chronological ordering on $C_+$, and with
115: antichronological ordering on $C_-$.)
116: %For instance, if both $z_1$ and $z_2$ are on
117: %$C_+$, $\theta_C(z_1,z_2)=\theta(x^+_1-x^+_2)$, while if they
118: %are both on $C_-$, $\theta_C(z_1,z_2)=\theta(x^+_2-x^+_1)$;
119: %furthermore, for $z_1\in C_+$ and $z_2\in C_-$,
120: %$\theta_C(z_1,z_2)=0$, etc.
121: In particular, for $z_1=x^+_0$ and $z_2=x^+_0-i\eta$
122: (the end points of the contour),
123: we write:
124: \be\labe{WLC}
125: W_C(\vec x)\,\equiv\,{\rm T}_C\, \exp\left\{\,
126: ig\int_C dz A^-(z, \vec x) \right\},
127: \ee
128: with $\int_C$ running all the way along the contour.
129:
130: We are now in a position to present the quantum action
131: for the non-local effective theory (below,
132: $\int_C d^4x \equiv \int_C dz\int d^3 \vec x$) :
133: \be\label{ACTION}
134: S[A,\rho]\,=\,- \int_C d^4x \,{1 \over 4} F_{\mu\nu}^a F^{\mu\nu}_a
135: \,+\,{i \over {gN_c}} \int d^3 \vec x\, {\rm Tr}\,\Bigl\{ \rho(\vec x)
136: \,W_C[A^-](\vec x)\Bigr\}\,\equiv\,S_{YM}\,+\,S_W.\,\,
137: \ee
138: This is real, as it should, since the field $A^-$ in the Wilson line
139: (\ref{WLC}) is in the adjoint representation: $A^-\equiv A^-_aT^a$.
140: In the saddle point approximation, eq.~(\ref{ACTION})
141: generates eq.~(\ref{cleq1}), as we verify now.
142: Note first that (with $x^+_0 \to -\infty$) :
143: \be\labe{SWJ1}
144: \frac{\delta S_W}{\delta A^-_a(z,\vec x)}\,=\,-\,
145: \frac{1}{N_c}\,{\rm Tr}\Bigl\{\rho(\vec x)
146: W_{-\infty-i\eta,z}(\vec x)\,T^a\,
147: W_{z,-\infty}(\vec x)\Bigr\}.\ee
148: The tree-level field $A^\mu(z)$ (the solution to
149: $\delta S/\delta A^\mu(z)=0$) takes identical values
150: on both sides of the contour:
151: $A^\mu_a(x^+) = A^\mu_a(x^+-i\eta)$ for any real $x^+$.
152: For such a field, eq.~(\ref{SWJ1})
153: reduces to the current $J^+(x)$ in eq.~(\ref{SWJ}),
154: for $z$ irrespectively on $C_+$ or on $C_-$. To verify this,
155: consider, e.g., $z=x^+\in C_+$; then,
156: $W_{x^+,-\infty}=W(x^+)$ (cf. eq.~(\ref{WLINE1})), and
157: \be
158: W_{-\infty-i\eta,x^+}\,=\,W_{-\infty-i\eta,x^+-i\eta}\,
159: W_{x^+-i\eta,\infty-i\eta}\,W_{\infty,x^+},\ee
160: where we have used the group property
161: of the contour operators (\ref{WLC0}).
162: Since the tree-level field $A^-$ is the same on both sides of the
163: contour, the forward evolution in time from
164: $x^+$ to $\infty$ is compensated by the backward
165: evolution in time from $\infty-i\eta$ to $x^+-i\eta\,$:
166: $W_{x^+-i\eta,\infty-i\eta}\,W_{\infty,x^+}=1$.
167: Similarly (with ${\tilde {\rm T}}$ denoting anti-chronological
168: ordering),
169: \be
170: W_{-\infty-i\eta,x^+-i\eta}\,=\,W^\dagger(x^+)\,
171: \equiv\,{\tilde {\rm T}}\, \exp\left\{\,
172: -ig\int_{x^+_0}^{x^+} dz^+ A^-(z^+, \vec x) \right\}.
173: \ee
174: Thus, at tree-level, and for $z=x^+\in C_+$,
175: $W_{-\infty-i\eta,z}\,T^a\, W_{z,-\infty}
176: \to W^\dagger(x^+)\,T^a\,W(x^+)$, in agreement
177: with eq.~(\ref{SWJ}). A similar conclusion holds for
178: $z=x^+-i\eta\in C_-$. In both cases, the saddle point equation
179: $\delta S/\delta A^\mu(z)=0$ turns out to be equivalent to
180: eq.~(\ref{cleq1}).
181:
182: The non-local piece of the action $S_W$
183: describes the propagation of the color source $\rho$
184: ($=$ the fast degrees of freedom of the hadron) at the
185: speed of light, in the background of the soft fields
186: $A^\mu$ (the classical field radiated by $\rho$ plus the
187: soft quantum fluctuations), and in the eikonal approximation.
188: This gives rise to vertices of the type
189: $\rho(A^-)^n$, $n\ge 1$, which are
190: non-local in time (see Fig. \ref{SWvertex}).
191: \begin{figure}
192: \protect\epsfxsize=8.cm{\centerline{\epsfbox{SWvertex.eps}}}
193: \caption{A typical $n$-point vertex (here, $n=3$)
194: generated by the action $S_W$. This is non-local in time, but local
195: in the spatial coordinates. The continuous line represents the
196: source $\rho$, while the wavy lines are $A^-$ gluon fields.}
197: \label{SWvertex}
198: \end{figure}
199:
200: %With our gauge choice here, the (LC-gauge) mean field
201: %$A^\mu$ has the property that $A^-=0$, so that the field
202: %$A^-$ in the action is a pure quantum fluctuation:
203: %$A^-\equiv \delta A^-$, cf. eq.~(\ref{fluct0}).
204: %Thus, $S_W$ generates an infinity of non-local (in time)
205: %vertices of the type $\rho(\delta A^-)^n$, as illustrated
206: %in Fig. ??.
207:
208:
209:
210: $S_W$ is invariant under the {\it periodic} gauge
211: transformations $\Omega(x)\in {\rm SU}(N)$ which satisfy
212: \be \labe{periodic}
213: \Omega(-\infty-i\eta,\vec x)\,=\,\Omega(-\infty,\vec x) \ee
214: for any $\vec x$. Indeed, under such a transformation,
215: \be\labe{gtr1}
216: A^\mu(x)&\longrightarrow& %{\vec A}^\mu(x) =
217: \Omega(x)\Bigl(A^\mu +
218: (i/g)\partial^\mu\Bigr)\Omega^{\dagger}(x)\,,\nonumber\\
219: \rho(\vec x)&\longrightarrow& %\tilde\rho(\vec x)=
220: \Omega(-\infty,\vec x) \,\rho(\vec x)\,
221: \Omega^{\dagger}(-\infty,\vec x)
222: \,,\nonumber\\
223: W_C(\vec x)&\longrightarrow& \Omega(-\infty-i\eta,\vec x)
224: \,W_C(\vec x) \,\Omega^{\dagger}(-\infty,\vec x),\ee
225: and $S_W$ is invariant because of the property (\ref{periodic}).
226: Here, we are mostly interested in time-independent gauge
227: transformations, which preserve the static character of the
228: background fields, and for which the periodicity condition
229: (\ref{periodic}) is trivially satisfied.
230:
231: The transformation law for $\rho(\vec x)$ in eq.~(\ref{gtr1})
232: reflects
233: its interpretation as the initial value of the color charge
234: $J^+$ at $x^+_0=-\infty$ (cf. Sect. 2.3)).
235: In fact, $S_W$ can be also written as
236: follows\footnote{To see this, use the identity
237: $i\del_zW_{z,-\infty}=gA^-(z)W_{z,-\infty}$ to
238: perform the integral over $z$ in eq.~(\ref{SW2}).} :
239: \be\labe{SW2}
240: S_W\,=\,-\,{1\over {N_c}} \int_C d^4x\, {\rm Tr}\,\Bigl\{A^-(z,\vec x)
241: W_{z,-\infty}(\vec x) \rho(\vec x)\Bigr\},\ee
242: which emphasizes the fact that this is a gauge-invariant
243: generalization of the linear vertex $\int d^4x \,\rho_a A^-_a$.
244:
245:
246:
247:
248: \subsection{Contour Green's functions}
249:
250: The generating functional for contour Green's functions
251: is defined as in eq.~(\ref{PARTQUANT}) where the time
252: variables now run along the contour, and the external source
253: is extended to a function $j^\mu_a(z,\vec x)$ on $C$.
254: For instance, the (connected) contour two-point function
255: is obtained as (with color indices omitted):
256: \be\labe{QcorrC}
257: iG_C^{\mu\nu}(z_1,z_2)&\equiv&\langle {\rm T}_C\,
258: A^\mu(z_1)A^\nu(z_2)\rangle_c\,=\,-\,\frac{\delta^2 \ln Z_{\Lambda}
259: [\rho,\,j]}
260: {\delta j_\mu(z_1)\delta j_\nu(z_2)}\bigg|_{j=0}\nonumber\\
261: &=&\theta_C(z_1,z_2) G^{>\,\mu\nu}(z_1,z_2)\,+\,
262: \theta_C(z_2,z_1) G^{<\,\mu\nu}(z_1,z_2),\ee
263: where, to lighten the notations, we have indicated only
264: the time variables. In the second line above, we
265: have introduced the two-point correlation, or {\it Wightman},
266: functions:
267: \be\labe{G><}
268: G^{>\,\mu\nu}_{ab}
269: (z_1,z_2)\equiv\langle A^\mu_a(z_1)A^\nu_b(z_2)\rangle_c,\qquad
270: G^{<\,\mu\nu}_{ab}
271: (z_1,z_2)\equiv\langle A_b^\nu(z_2)A_a^\mu(z_1)\rangle_c,\ee
272: which, unlike the contour propagator $G_C^{\mu\nu}(z_1,z_2)$,
273: are continuous at $z_1=z_2$. %Note the symmetry property
274: %$G^{>\,\mu\nu}_{ab}(x,y)\,=\,G^{<\,\nu\mu}_{ba}(y,x)$.
275: These are the functions which enter the quantum corrections
276: $\sigma$ and $\chi$; for instance,
277: the contribution of eq.~(\ref{drhoYM}) to $\sigma$ reads
278: (see also Sect. 4.3 below):
279: \be\label{sigmaYM}
280: \sigma_a (x)\big|_{YM}
281: %\,=\,g f^{abc} \langle(\partial^+ a_{b}^{i}(x) )a_{c}^{i}(x)\rangle
282: \,=\,g f^{abc}\partial^+_y\langle a_{b}^{i}(y)a_{c}^{i}(x)\rangle
283: \big|_{y=x}
284: \,=\,\,g f^{abc}\partial^+_yG^{<\,ii}_{cb}(x,y)\big|_{y=x}\,
285: .\ee
286: For real time variables, the
287: functions $G^>$ and $G^<$ can be used to construct the
288: retarded $(G_R)$ and
289: advanced $(G_A)$ propagators, which will also be needed:
290: \be\labe{A}
291: G_R(x,y)&\equiv &
292: i\theta(x^+-y^+)\Bigl[G^>(x,y)\,-\,G^<(x,y)\Bigr]\,,\nonumber\\
293: G_A(x,y)&\equiv & -
294: i\theta(y^+-x^+)\Bigl[G^>(x,y)\,-\,G^<(x,y)\Bigr]\,.\ee
295: %In particular, for $x^+$, $y^+\in C_+$, $G_C(x^+,y^+)$
296: %is the usual time-ordered (or Feynman) propagator.
297:
298: As a simple example, let us construct the contour propagator
299: for a free scalar field (the generalization to a free gluon field
300: in the LC gauge is straightforward). For both time arguments
301: $x^+$ and $y^+$ on $C_+$, this coincides with the usual Feynman
302: propagator, which reads (with $\vec p=(p^+,{\bf p}_\perp)$) :
303: \be\labe{GF0}
304: iG_0(x^+-y^+,\vec p)&=&\int {dp^- \over 2 \pi}\
305: e^{-ip^-(x^+-y^+)} \,\frac{i}{2p^+p^- - p_\perp^2 +i\epsilon}
306: \\&=&{1\over 2 p^+}\
307: \left \{\theta(p^+) \theta(x^+-y^+) - \theta(-p^+) \theta(y^+-x^+)
308: \right \}
309: {\rm e}^{-i {p_{\perp}^2 \over 2p^+}(x^+-y^+)},\nonumber\ee
310: from which one can identify the free Wightman functions:
311: \be
312: G^>_0(x^+,\vec p)\,=\,\frac{\theta(p^+)}{2 p^+}\,
313: {\rm e}^{-i {p_{\perp}^2 \over 2p^+}x^+},\qquad
314: G^<_0(x^+,\vec p)\,=\,-\,\frac{\theta(-p^+)}{2 p^+}\,
315: {\rm e}^{-i {p_{\perp}^2 \over 2p^+}x^+},\ee
316: or, in momentum space (with $p^2=2p^+p^- - p_\perp^2$),
317: \be\labe{G><0}
318: G^>_0(p)\,=\,2\pi\theta(p^+)\,\delta(p^2),\qquad
319: G^<_0(p)\,=\,2\pi\theta(-p^+)\,\delta(p^2),\ee
320: The free contour propagator $G^{\,0}_{C}$ is finally obtained
321: by inserting $G^>_0$ and $G^<_0$ in eq.~(\ref{QcorrC}).
322:
323: Returning to the gauge theory with action (\ref{ACTION}), the
324: corresponding contour propagator $G_C$ satisfies the
325: Dyson-Schwinger equation with time arguments on $C$:
326: \be \labe{DS0}
327: \int_C {\rm d}^4z\,\left\{
328: \frac{\delta^2 S}
329: {\delta A^\mu(x)\delta A^\nu(z)}\bigg|_{{\cal A}}\,+\,
330: \delta \Sigma_{\mu\nu}^C(x,z)\right\}G_C^{\nu\lambda}(z,y)\,=\,
331: \delta^\lambda_\mu\delta_C(x,y)
332: \,.\ee
333: Here, $\delta \Sigma_{\mu\nu}^C(x,y)$ is the (contour) self-energy,
334: and describes quantum corrections, and
335: \be
336: \frac{\delta^2 S}
337: {\delta A^\mu(x)\delta A^\nu(y)}\bigg|_{{\cal A}}
338: \,=\,{\cal D}_{\mu\nu}(x)\delta_C(x,y)
339: %\Bigl(g_{\mu\nu}{\cal D}^2-{\cal D}_\mu {\cal D}_\nu- 2ig
340: %{\cal F}_{\mu\nu}\Bigr)_x\delta_C(x,y)
341: \,+\,\delta_{\mu -}\delta_{\nu -}\Pi^C(x,y),\ee
342: where we have denoted
343: \be\labe{HATSIG}
344: {\cal D}_{\mu\nu}(x)\equiv\,g_{\mu\nu}{\cal D}^2
345: -{\cal D}_\mu {\cal D}_\nu- 2ig
346: {\cal F}_{\mu\nu}\,,\qquad
347: \Pi^C_{ab}(x,y)\equiv\,
348: \frac{\delta^2 S_W}{\delta A^-_a(x) \delta A^-_b(y)}
349: \bigg |_{{\cal A}}\,.\ee
350: In these equations, ${\cal D}^\mu=\del^\mu-ig
351: {\cal A}^\mu$, and ${\cal A}^\mu$ is the background field, to be
352: eventually identified with the classical field in
353: Sect. 2.3 (hence the notation). In particular, ${\cal A}^-=0$,
354: and a second differentiation in eq.~(\ref{SWJ1})
355: yields (with $x^+$, $y^+\in C$):
356: \be\labe{PIC}
357: \Pi_{ab}^C(x,y)
358: %&=&\frac{ig}{N_c}\,\delta^{(3)}(\vec x-\vec y)\,
359: %{\rm Tr}\,\Bigl\{\rho(\vec x)
360: %\Bigl[\theta_C(x^+,y^+)T^a T^b +\theta_C(y^+,x^+)T^b T^a\Bigr]
361: %\Bigr\}\nonumber\\
362: &=&-\,\frac{g}{2}\,f^{abc}\rho_c(\vec x)
363: \,\delta^{(3)}(\vec x-\vec y)\Bigl\{\theta_C(x^+,y^+)
364: -\theta_C(y^+,x^+)\Bigr\}\nn&\equiv&\theta_C(x^+,y^+)\Pi^>_{ab}(x,y)
365: \,+\,\theta_C(y^+,x^+)\Pi^<_{ab}(x,y),\ee
366: with time-independent $\Pi^>$ and $\Pi^<$ :
367: \be\labe{P><}
368: \Pi^>_{ab}(x,y)\,=\,-(g/2)\,f^{abc}\rho_c(\vec x)
369: \,\delta^{(3)}(\vec x-\vec y)\,=\,-\Pi^<_{ab}(x,y)\,.\ee
370: The contour self-energy admits a similar decomposition:
371: \be\labe{Sig<>}
372: \delta \Sigma^C(z_1,z_2)\,=\,\theta_C(z_1,z_2) \delta \Sigma^>(z_1,z_2)
373: +\theta_C(z_2,z_1)\delta \Sigma^<(z_1,z_2).\ee
374: The Dyson-Schwinger equation is then conveniently rewritten as
375: \be \labe{D1}
376: {\cal D}_{\mu\nu}(x)
377: %\left(g_{\mu\nu}{\cal D}^2-{\cal D}_\mu {\cal D}_\nu- 2ig
378: %{\cal F}_{\mu\nu}\right)_x
379: G_C^{\nu\lambda}(x,y)\,
380: +\int_C {\rm d}^4z\,\Sigma_{\mu\nu}^C(x,z)
381: \,G_C^{\nu\lambda}(z,y)\,=\,\delta^\lambda_\mu\delta_C(x,y)\,,\ee
382: %\,,\nn\ee
383: with the total self-energy $\Sigma_{\mu\nu}^C\equiv
384: \delta_{\mu -}\delta_{\nu -}\Pi^C+\delta \Sigma_{\mu\nu}^C$.
385:
386:
387: By choosing $x^+\in C_+$ and $y^+\in C_-$ in eq.~(\ref{D1}),
388: and using the decompositions (\ref{QcorrC})
389: and (\ref{Sig<>}),
390: we obtain, after simple manipulations, an
391: equation for $G^<(x,y)$ in {\it real} time :
392: %(also known as the Kadanoff-Baym equation \cite{KB62}) :
393: \be\labe{KBYM1}
394: {\cal D}_\mu^{\,\,\,\rho}(x)
395: %\left(g_\mu^{\,\rho}{\cal D}^2-{\cal D}_\mu {\cal D}^\rho-
396: % 2ig{\cal F}_{\mu}^{\,\rho}\right)_x
397: G^<_{\rho\nu}(x,y)\,=\,
398: \int {\rm d}^4z\,\bigl(
399: \Sigma_{R}\,G^<+\Sigma^<G_{A}\bigr)_{\mu\nu}(x,y).\ee
400: [The retarded ($\Sigma_R$) and advanced ($\Sigma_A$) self-energies
401: are defined in terms of the Wightman self-energies
402: $\Sigma^>$ and $\Sigma^<$ as in eq.~(\ref{A}).]
403: One similarly derives an equation for $G^>(x,y)$, as
404: well as the following equation for $G_R(x,y)$:
405: \be\labe{eqGR}
406: {\cal D}^{\mu}_{\,\,\rho}(x)
407: %\left(g^{\mu}_{\,\rho}{\cal D}^2-{\cal D}^\mu {\cal D}_\rho-
408: % 2ig{\cal F}^{\mu}_{\,\rho}\right)_x
409: G^{\rho\nu}_R(x,y)
410: \,-\,\int {\rm d}^4z\,\bigl(
411: \Sigma_R\,G_R\bigr)^{\mu\nu}(x,y)\,=\,g^{\mu\nu}\delta^{(4)}(x-y)\,.\ee
412: The two equations above imply a relation
413: between $G^<$ and $\Sigma^<$ :
414: \be\labe{WRA}
415: G^<_{\mu\nu}(x,y)\,=\,\int {\rm d}^4z\,{\rm d}^4u\,\Bigl(
416: G_R(x,z)\,\Sigma^<(z,u)\,G_A(u,y)\Bigr)_{\mu\nu}
417: .\ee
418: A similar relation holds between $G^>$ and $\Sigma^>$.
419:
420: Let us now apply this general formalism to the calculation
421: of the quantum corrections discussed in Sect. 3.2.1.
422: We then need the following contour Green's functions:
423:
424: ({\it a}) The two-point function
425: $G^{<\,\mu\nu}_{ab}(x,y)\equiv \langle \delta A^\nu_b(y)
426: \delta A^\mu_a(x)\rangle$ of the soft fields
427: induced by their coupling to the semi-fast gluons.
428: This is given by eq.~(\ref{WRA}) with $\Sigma^<$ replaced by
429: \be\labe{DELSIG}
430: \delta \Sigma_{\mu\nu}^<(x,y)\,=\,
431: \langle \delta J_\nu(y)\delta J_\mu(x)\rangle\,,\ee
432: where $\delta J_\mu^a$ is the quantum color current in eq.~(\ref{deltaJ}).
433: To lowest order in $\alpha_s\,$, the soft field propagators $G_R$
434: and $G_A$ in eq.~(\ref{WRA}) can be computed in the mean field
435: approximation; that is, one can neglect the quantum self-energy
436: $\delta \Sigma^{\mu\nu}_R$ when solving eq.~(\ref{eqGR}).
437: With these approximations, eq.~(\ref{WRA}) reduces
438: to eq.~(\ref{JJSIG}), which has been used in Sect. 3.2.1.
439:
440: ({\it b}) The background field propagator
441: $iG_C^{\mu\nu}(x,y)\equiv \langle {\rm T}_C\,a
442: ^\mu(x) a^\nu(y)\rangle$ of the semi-fast gluons
443: in the Gaussian, or mean field, approximation
444: (cf. eq.~(\ref{delAcorr})). % and (\ref{sigmaYM})).
445: This is given by eq.~(\ref{D1}) where the quantum self-energy
446: $\delta \Sigma^C_{\mu\nu}$ is neglected; that is,
447: \be \labe{D2}
448: {\cal D}_{\mu\nu}(x)
449: %\left(g_{\mu\nu}{\cal D}^2-{\cal D}_\mu {\cal D}_\nu- 2ig
450: %{\cal F}_{\mu\nu}\right)_x
451: G_C^{\nu\lambda}(x,y)\,+\,\delta_{\mu -}\int_C {\rm d}^4z\,
452: \Pi^C(x,z)\,G_C^{-\lambda}(z,y)
453: \,=\,\delta^\lambda_\mu\delta_C(x,y)\,.\ee
454: %\,.\nn\ee
455: Since the tree-level self-energy (\ref{PIC}) is non-local,
456: the integration over $z^+$ in this equation
457: runs along the whole contour $C$. Thus, even for real time
458: arguments $x^+$ and $y^+$, eq.~(\ref{D2}) is still
459: sensitive to the complex side $C_-$ of the contour.
460: This is to be contrasted to
461: a {\it local} field theory, where the contour structure becomes
462: irrelevant in the mean field approximation \cite{BI00}.
463:
464: However, a closer inspection reveals that, even for
465: eq.~(\ref{D2}), the contour structure is {\it not} essential,
466: in the sense that one can ignore $C_-$
467: when computing the {Feynman} propagator
468: (i.e., the propagator $G_C^{\mu\nu}(x,y)$ with both time
469: arguments on $C_+$). Indeed, for $x^+$, $y^+ \in C_+$,
470: the contribution of $C_-$ to eq.~(\ref{D2}) reads, schematically,
471: \be
472: \int_{C_-} {\rm d}^4z\,\Pi^C(x,z)\,G_C(z,y)\,=\,-
473: \int{\rm d}^4z\,\Pi^<(x,z)\,G^>(z,y)&\propto&\,\nn\propto
474: \int {\rm d}z^+ \,G^>(z^+-y^+, \vec x, \vec y)\,=\,
475: G^>(p^-=0,\vec x, \vec y)&=&0,\ee
476: where we have used eqs.~(\ref{PIC})--(\ref{P><}),
477: the time-homogeneity of the problem (i.e., the fact that
478: $\Pi^<$ is independent of time, while
479: $G^>$ depends only upon the relative time $z^+-y^+$),
480: and the fact that $G^>(p^-=0)=0$ (since, by definition,
481: this has support only for $p^-$ in the strip (\ref{strip-})).
482:
483:
484: To conclude, the background field Feynman propagator
485: $iG^{\mu\nu}(x,y)\equiv \langle {\rm T}\,a ^\mu(x) a^\nu(y)\rangle$
486: can be obtained by solving the following equation
487: %(with appropriate boundary conditions; see Sect. 5)
488: \be \labe{D3}
489: {\cal D}_{\mu\nu}(x)
490: %\left(g_{\mu\nu}{\cal D}^2-{\cal D}_\mu {\cal D}_\nu-2ig
491: %{\cal F}_{\mu\nu}\right)_x
492: G^{\nu\lambda}(x,y)\,+\,\delta_{\mu -}\int{\rm d}^4z\,
493: \Pi(x,z)\,G^{-\lambda}(z,y)
494: \,=\,\delta^\lambda_\mu\delta^{(4)}(x-y)\,,\ee
495: %\,,\nn\ee
496: where all the time variables are real, and $\Pi(x,z)$
497: is the restriction of $\Pi^C(x,z)$, eq.~(\ref{PIC}),
498: to time arguments on $C_+$. Then, the corresponding
499: Wightman functions
500: (which enter $\sigma$ and $\chi\,$; see, e.g., eq.~(\ref{sigmaYM}))
501: can be extracted from the Feynman propagator by considering
502: appropriate time orderings, as
503: we did for the free propagator in eqs.~(\ref{GF0})--(\ref{G><0}).
504:
505: Note that any explicit use of the contour has been avoided in this way.
506: This has been possible because, first, of the static
507: character of the background, and, second, of the simple nature
508: of the present, mean field, approximations.
509: For more general problems, where higher-order quantum effects
510: should be computed in the presence of inhomogeneities in
511: time, a complete use of the contour would be generally unavoidable.
512: %Moreover, the use of the complex time contour has been useful
513: %at a conceptual level, to clarify the non-local structure
514: %of the effective action (\ref{ACTION}), and also the general
515: %non-linear structure of the quantum corrections to the
516: %gluon density.
517:
518: \subsection{Feynman rules for $\sigma$ and $\chi$}
519:
520: As explained in the previous subsection, the use of the
521: contour techniques can be avoided for the calculation of
522: $\sigma$ and $\chi$.
523: It is then sufficient to consider the restriction of the quantum
524: theory to the real time axis, as defined by the following
525: action (compare to eq.~(\ref{ACTION})):
526: $S=S_{YM}+S_W$, with
527: \be\labe{ACTION1}
528: S_W\,=\,{i \over g{N_c}} \int d^3 \vec x\, {\rm Tr}\,\Bigl\{ \rho(\vec x)
529: \,W_{\infty, -\infty}[A^-](\vec x)\Bigr\}\,\ee
530: and the {\it real-time} Wilson line:
531: \be\labe{WLINE}
532: W_{\infty,-\infty}[A^-](\vec x)\, =\,{\rm T}\, \exp\left[\,
533: ig\int dx^+ A^-(x) \right].
534: \ee
535: This is the original version of the quantum MV model
536: proposed in Ref. \cite{JKLW99a}.
537: In particular, the tree-level self-energy generated by this
538: action reads:
539: \be\labe{R--}
540: \Pi^{ab}(x,y) &\equiv&
541: \frac{\delta^2 S_W}{\delta A^-_a(x) \delta A^-_b(y)}
542: \bigg |_{A^-=0} \,=\,-
543: \frac{g}{2}\,f^{abc}\rho_c(\vec x)
544: \,\epsilon(x^+-y^+)\,\delta^{(3)}(\vec x-\vec y)\nn
545: &=&g\rho^{ab} (\vec x)\,\delta^{(3)}(\vec x-\vec y)\,
546: \langle x^+ |{\rm PV}\,{1 \over i\partial^-} |y^+ \rangle\,,\ee
547: where $\rho^{ab} \equiv -if^{abc} \rho^c$, and
548: $\epsilon(x)\equiv \theta(x)- \theta(-x)$. %is the sign function.
549: Eq.~(\ref{R--}) coincides, as it should, with the restriction of the
550: contour self-energy $\Pi^C$, eq.~(\ref{PIC}),
551: to the real time axis.
552: Note the emergence of the principal value prescription in $1/p^-$
553: (this is defined as in eq.~(\ref{PV})).
554:
555: Since we are mainly interested in the strong
556: field regime, where ${\cal A}^i\sim 1/g$ and $\rho\sim 1/g$,
557: it is convenient for power counting to perform a rescaling of the
558: background fields and the color sources:
559: \be\labe{rescale}
560: {\cal A}_a^\mu \longrightarrow \frac{1}{g}\,{\cal A}_a^\mu,\qquad
561: \rho_a \longrightarrow \frac{1}{g}\,\rho_a, \qquad
562: \delta\rho_a \longrightarrow \frac{1}{g}\,\delta\rho_a\,.\ee
563: In the saturation regime, the new tree-level quantities
564: ${\cal A}^\mu$ and $\rho$ are of order 1, while the new
565: quantum corrections $\chi$ and $\sigma$ --- which are defined
566: in terms of the rescaled charge fluctuations $\delta\rho$
567: by the same equations as before (i.e.,
568: (\ref{CHIDEF}) and (\ref{JIND})) --- are of order
569: $\alpha_s$.
570: After this rescaling, the
571: coupling constant $g$ completely drops out from
572: the classical analysis in Sect. 2. On the other hand, we prefer
573: not to rescale the quantum fluctuations $a^\mu\,$; thus,
574: their propagator $G^{\mu\nu}(x,y)[{\cal A},\rho]$ remains of
575: order one, and all the factors of $g$ can be explicitly read
576: off the vertices (see below).
577:
578: The quantum charge fluctuations
579: $\delta \rho_a(x)$ are defined by eq.~(\ref{deltaJ})
580: with $\mu=-\,$.
581: After the rescaling (\ref{rescale}), this yields:
582: \begin{equation}\labe{delta12}
583: \delta \rho_a (x) =\delta \rho_a^{(1)} (x)+\delta \rho_a^{(2)} (x),
584: \end{equation}
585: where $\delta \rho^{(1)}$ is linear in $a^\mu$, and
586: $\delta \rho^{(2)}$ is quadratic:
587: \be\labe{rho10}
588: \delta \rho_a^{(1)} (x) & = &
589: -2i g{\cal F}^{+i}_{ac} (\vec x) a^{ic} (x) + \nonumber \\
590: & & +g\rho^{ac} (\vec x)
591: \int dy^+ \langle x^+ |{\rm PV}\,{1 \over i\partial^-} |y^+ \rangle
592: a^{c-}(y^+,\vec x), \\
593: \delta \rho_a^{(2)}(x)& = &
594: g^2 f^{abc} [\partial^+ a^{b}_{i}(x)
595: ]a^{c}_{i}(x)
596: \nonumber\\ &{-}& {{g^2}\over{N_c}} \,\rho^{b}({\vec x})
597: \int dy^+ a^{-c}(y^+,{\vec x}) \int dz^+ a^{-d}(z^+,{\vec x})
598: \nonumber\\ &{}& \nonumber\,\,\,\, \times\,
599: \biggl\{\theta (z^+ -y^+)
600: \theta (y^+ -x^+) {\rm Tr} \,(T^a T^c T^d T^b)
601: \\ &{}&\nonumber
602: \qquad+\ \theta (x^+ -z^+) \theta (z^+ -y^+) {\rm Tr} \,(T^a T^b T^c T^d)
603: \\ &{}&\qquad
604: +\ \theta (z^+ -x^+) \theta (x^+ -y^+) {\rm Tr}\,(T^a T^d T^b T^c) \biggr\},
605: \labe{rho2}
606: \ee
607: where all the factors of $g$ are now explicit.
608: In the right hand sides of these equations, the terms involving
609: $a^i$ come from the three-gluon vertex in
610: $S_{YM}$, while the terms involving $a^-$
611: come from the two- and three-point vertices in $S_W$.
612:
613: In eqs.~(\ref{rho10})--(\ref{rho2}) it is understood that only
614: the soft modes (with $k^+\simle b\Lambda^+$) are kept in the
615: products of fields. This has
616: been used to simplify the first contribution to $\delta \rho^{(1)}$
617: by writing:
618: \be
619: (\del^+ {\cal A}^i_{ac})\, a^{ic}\,-\,{\cal A}^i_{ac} \,\del^+a^{ic}
620: \,\approx\,2 {\cal F}^{+i}_{ac}a^{ic}\,.\ee
621: Thus, the above expressions can be used only for calculations
622: to LLA.
623:
624: The quantum corrections $\chi$ and $\sigma$ are
625: given by eq.~(\ref{CHIDEF}) (with $x^+=y^+$) and (\ref{JIND}),
626: where the average over the semi-fast fluctuations $a^\mu$
627: is defined as in eqs.~(\ref{delAcorr})--(\ref{SEXP2}).
628: This generates Wightman functions of the fields
629: $a^\mu$ (see, e.g., eq.~(\ref{sigmaYM})), which are related
630: to the corresponding time-ordered, or Feynman, propagator
631: as explained in Sect. 4.2. In particular, the equal time limit
632: of a two-point function is obtained as:
633: \be
634: \langle a_b^\nu(x^+) a_a^\mu
635: (x^+) \rangle\, \equiv \,
636: G^{<\,\mu\nu}_{\,\,\,ab}(x^+=y^+)\,=\,
637: iG^{\mu\nu}_{ab}(y^+=x^++\epsilon).\ee
638: In momentum space, this instructs us to perform the
639: integration over $p^-$ by closing the contour
640: in the upper half of the complex $p^-$ plane:
641: \be \labe{prescripW} G^< (x^+,x^+)\,=\,i
642: \int {dp^-\over 2\pi} \,{\rm e}^{ip^-\epsilon}\,G(p^-)\,.\ee
643: In practice, this prescription will play a role only for the
644: vacuum piece $G_0$ (i.e., the limit of $G$ as $\rho\to 0$,
645: which to the order of interest coincides with the free propagator
646: $G_0$; cf. Sect. 3.4). Indeed,
647: within the present approximations,
648: the induced piece $\delta G\equiv G-G_0$ is continuous at $x^+=y^+$,
649: so its equal-time limit is unambiguous.
650:
651: We are now prepared to express $\sigma$ and $\chi$
652: in terms of the Feynman propagator $G^{\mu\nu}_{ab}(x,y)$ of the
653: semi-fast gluons. For $\sigma\equiv\langle\delta \rho\rangle
654: =\langle\delta \rho^{(2)}\rangle\/$, one obtains
655: (with the ``approximately equal'' sign denoting
656: an equality which holds to LLA) :
657: \be\labe{sigma12}
658: \alpha_s\ln{1\over b}\,\sigma_a ({\bf x}_\perp)&\approx&
659: \int dx^- \, {\rm Tr } \,(T^a \hat\sigma(\vec x)),\nonumber\\
660: \hat\sigma(\vec x)&\equiv&-g^2 \partial^+_y G^{ii}\Big |_{x=y}\,
661: +\,ig^2\rho({\vec x})\Bigl\langle x\bigg|
662: {1 \over i \partial^-} \,G^{--} {1 \over i\partial^-}\bigg|x
663: \Bigr\rangle\nn
664: &\equiv&\hat\sigma_1(\vec x)\,+\,\hat\sigma_2(\vec x).
665: \ee
666: In writing $\hat\sigma_2$ as above, we have used compact but
667: formal notations for the second, non-local, contribution
668: to $\delta \rho_a^{(2)}$, eq.~(\ref{rho2}).
669: A pictorial representation of the two pieces of $\hat\sigma$
670: is given in Fig. \ref{SIGFIG}.
671: \begin{figure}
672: \protect\epsfxsize=14.cm{\centerline{\epsfbox{SIGFIG.eps}}}
673: \caption{Feynman diagrams for $\hat\sigma_1$ (a) and
674: $\hat\sigma_2$ (b,c,d). The wavy line with a blob denotes the
675: background field propagator of the semi-fast gluons;
676: the continuous line represents the source $\rho$; the precise
677: vertices can be read off eq.~(\ref{rho2}).}
678: \label{SIGFIG}
679: \end{figure}
680:
681: Concerning $\chi$, we note that to order $\alpha_s$
682: this involves $\delta \rho^{(1)}$, but not $\delta \rho^{(2)}$.
683: [This is clear by counting powers of $g$ according
684: to eqs.~(\ref{rho10}) and (\ref{rho2}).] Therefore
685: (with $y^+=x^++\epsilon$):
686: \be\labe{chi1}
687: \alpha_s\ln{1\over b}\,
688: \chi_{ab}({\bf x}_\perp,{\bf y}_\perp)\,\approx\,
689: \int dx^- \int dy^-\,\hat\chi_{ab}(\vec x,\vec y),
690: \qquad\qquad\qquad\nonumber
691: \\
692: \hat\chi_{ab}(\vec x,\vec y)\,\equiv\,g^2
693: \left\langle \left(-2i{\cal F}^{+i}
694: a^i + \,\rho {1 \over i\partial^-} a^- \right)_x^a
695: \left(
696: 2i a^i {\cal F}^{+i} +a^- {1 \over
697: i\partial^-} \rho \right)_y^b \right\rangle
698: \ee
699: where we have used also the symmetry property
700: \be
701: \langle x|{\rm PV}\, {1 \over i\partial^-} |y \rangle \, = \, -
702: \langle y|{\rm PV}\, {1 \over i\partial^-} |x \rangle.
703: \ee
704: This further yields, in matrix notations (the
705: PV prescription in $1/p^-$ is implicit) :
706: \be\labe{chi2}
707: \frac{1}{g^2}\,\hat \chi(\vec x, \vec y) &=&i\,
708: 2{\cal F}^{+i}_x\, \langle x|G^{ij}|y\rangle \,2{\cal F}^{+j}_y \, +\,
709: 2{\cal F}^{+i}_x\,\langle x|G^{i-}\,{1 \over i\partial^-}|y\rangle \, \rho_y
710: \nonumber\\&{}&\,\,\,\,-\,
711: \rho_x \, \langle x|{1 \over i\partial^-}\, G^{-i}|y\rangle \,
712: 2{\cal F}^{+i}_y\,+\,i
713: \rho_x \langle x|{1 \over i\partial^-} G^{--} {1 \over i\partial^-}
714: |y\rangle\,\rho_y.\ee
715: Diagramatically, all the above contributions to $\chi$
716: are represented by {\it tree-like} Feynman graphs (no loops),
717: with vertices proportional to $\rho$ or ${\cal F}^{+i}$ (see Fig. \ref{CHIFIG}).
718: Thus, like the tree-level source $\rho$ itself,
719: $\hat \chi(\vec x, \vec y)$ is localized near the LC, at
720: $0\simle x^-,y^- \simle 1/\Lambda^+$.
721: \begin{figure}
722: \protect\epsfxsize=12.cm{\centerline{\epsfbox{CHIFIG.eps}}}
723: \caption{Feynman diagrams for the four contributions
724: to $\chi$ given in eq.~(\ref{chi2}).}
725: \label{CHIFIG}
726: \end{figure}
727:
728:
729: It is now easy to verify, by using eqs.~(\ref{delta12}) and (\ref{rho2})
730: and straightforward power counting, that all the $n$-point correlators
731: of $\delta\rho$ beyond $n=2$ are of higher order in $\alpha_s$.
732: Indeed, each factor of $a^\mu$ in these
733: equations is accompanied by a power of $g$. Then,
734: for instance, the three-point function $\langle \delta\rho\,
735: \delta\rho\,\delta\rho\rangle$ --- which involves, at least,
736: four fields $a^\mu$ --- is of order $\alpha_s^2$.
737: This explains why, to the present accuracy, we can restrict
738: ourselves to the one- and two-point functions in
739: eqs.~(\ref{sigma12}) and (\ref{chi2}).
740: The remaining part of this paper, and also Paper II,
741: will be devoted to their explicit evaluation.
742:
743: