1: \documentstyle[epsfig,12pt]{article}
2: \textheight=20cm
3: %-------------------------------------------------------------
4: \tolerance=10000
5: %-------------------------------------------------------------
6: \newcommand{\beq}{\begin{equation}}
7: \newcommand{\eeq}{\end{equation}}
8: \newcommand{\bea}{\vspace{0.25cm}\begin{eqnarray}}
9: \newcommand{\eea}{\end{eqnarray}}
10:
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12:
13: \newcommand{\gb}{\mbox{{\boldmath
14: $\gamma$}}}
15: \newcommand{\ab}{\mbox{{\boldmath
16: $\alpha$}}}
17: \newcommand{\r}{\mbox{{\boldmath
18: $\rho$}}}
19: \newcommand{\ta}{\mbox{{\boldmath
20: $\tau$}}}
21: \newcommand{\qb}{\mbox{{\bf
22: q}}}
23: \newcommand{\pb}{\mbox{{\bf
24: p}}}
25: \newcommand{\rb}{\mbox{{\bf
26: r}}}
27: \newcommand{\kb}{\mbox{{\bf
28: k}}}
29: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30:
31: \setlength{\topmargin}{-0.5in}
32: \setlength{\oddsidemargin}{0in}
33: \setlength{\textheight}{9.0in}
34: \setlength{\textwidth}{6.3in}
35:
36: \def\lsim{\mathrel{\rlap{\lower4pt\hbox{\hskip1pt$\sim$}}
37: \raise1pt\hbox{$<$}}} %less than or approx. symbol
38: \def\gsim{\mathrel{\rlap{\lower4pt\hbox{\hskip1pt$\sim$}}
39: \raise1pt\hbox{$>$}}} %greater than or approx. symbol
40: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
41: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
42: \begin{document}
43: %\thispagestyle{empty}
44: \vspace*{-2cm}
45:
46: \bigskip
47: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
48: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
49:
50:
51: \begin{center}
52:
53: \renewcommand{\thefootnote}{\fnsymbol{footnote}}
54:
55: {\Large\bf
56: Radiative parton energy loss and
57: jet quenching in high-energy heavy-ion collisions
58: \footnote{Supported by DFG-grant Schi 189/6-1}
59: \\
60: \vspace{.7cm}
61: }
62: \renewcommand{\thefootnote}{\arabic{footnote}}
63: \medskip
64: {\large
65: B.G.~Zakharov
66: \bigskip
67: \\
68: }
69: {\it
70: Fakult\"at f\"ur Physik, Universit\"at Bielefeld\\
71: D-33501 Bielefeld, Germany\\
72: %{\begin{center} and \end{center}}\\
73: and\\
74: L.D.~Landau Institute for Theoretical Physics,
75: GSP-1, 117940,\\ Kosygina Str. 2, 117334 Moscow, Russia
76: \vspace{1.7cm}\\}
77:
78:
79: {\bf
80: Abstract}
81: \end{center}
82: {
83: \baselineskip=9pt
84: We study within the light-cone path integral approach \cite{Z1}
85: the effect of the induced gluon radiation
86: on high-$p_{T}$ hadrons in high-energy heavy-ion collisions.
87: The induced gluon spectrum is represented in a new form which
88: is convenient for numerical simulations. For the first time,
89: computations are performed with a realistic parametrization
90: of the dipole cross section.
91: The results are in reasonable agreement with
92: suppression of high-$p_{T}$ hadrons in $Au+Au$ collisions
93: at $\sqrt{s}=200$ GeV observed at RHIC.
94: \vspace{.5cm}
95: \\
96: }
97:
98: %\pagebreak
99: %-------------------------------------------------------------
100: %\vspace{.2cm}
101: \noindent{\bf 1.} One of the most interesting results
102: obtained at RHIC is the suppression of high-$p_{T}$ hadrons in
103: $Au+Au$ collisions (for a review of the data, see \cite{RHIC_data}).
104: It is widely believed that parton energy loss due to
105: the induced gluon radiation caused by multiple scattering in the
106: quark-gluon
107: plasma (QGP) produced in the initial stage of nucleus-nucleus collisions
108: plays a major role in this phenomenon (usually called
109: jet quenching) \cite{BDMPS,Z1,Z2,BDMS1,GLV1,W1} (for a review,
110: see \cite{BSZ}).
111: %
112: The most general approach to the induced gluon emission
113: is the light-cone path integral (LCPI) approach developed in
114: \cite{Z1} (see also \cite{Z_YAF,Z3,BSZ}).
115: It accurately treats the mass and finite-size effects, and applies
116: at arbitrary strength of the Landau-Pomeranchuk-Migdal
117: (LPM) effect \cite{LP,Migdal}. Other available approaches
118: have limited domains of applicability, and can only be used either in
119: the regime of strong (the BDMPS formalism \cite{BDMPS,BDMS1})
120: or weak (the GLV formalism \cite{GLV1}) LPM suppression (the GLV
121: approach \cite{GLV1}, in addition, is restricted to the emission of soft
122: gluons). For this reason they can not be used for an
123: accurate analysis of jet quenching
124: for RHIC (and LHC) conditions.
125:
126:
127: The LCPI approach expresses the gluon spectrum
128: through the solution of a two-dimensional Schr\"odinger
129: equation with an imaginary potential in the impact parameter plane.
130: The imaginary potential is
131: proportional to the cross section of interaction
132: of the $\bar{q}qg$ system (for $q\rightarrow gq$ transition) with a
133: particle in the medium, $\sigma_{3}(\rho)$ (here $\rho$ is the
134: transverse distance between quark and gluon, the antiquark is
135: located at the center of mass of the $qg$-system).
136: The $\sigma_{3}(\rho)$ can be written as
137: $\sigma_{3}(\rho)=C(\rho)\rho^{2}$. The factor $C(\rho)$
138: has a smooth (logarithmic) dependence on $\rho$ for
139: $\rho\ll 1/\mu_{D}$ (hereafter, $\mu_{D}$ is
140: the Debye screening mass).
141: If one replaces $C(\rho)$ by $C(\rho_{eff})$, where $\rho_{eff}$ is the typical
142: value of $\rho$, the Hamiltonian takes the oscillator form.
143: This approximation, which greatly simplifies the calculations,
144: was employed in several analyses
145: \cite{Z2,W1,OA} (it was also used in the BDMPS approach \cite{BDMPS,BDMS1}).
146: However, the oscillator approximation turns out to be too crude and
147: unsatisfactory. First of all,
148: for RHIC and LHC conditions, the dominating $\rho$ scale is not
149: small enough and the results depend strongly on the choice
150: of $\rho_{eff}$. Another reason why the oscillator approximation
151: is unsatisfactory is more serious.
152: In the high energy limit the gluon formation length, $L_{f}$, becomes
153: larger than the quark pathlength in the QGP, and finite-size effects
154: become important. In this regime $\rho_{eff}\ll 1/\mu_{D}$, and
155: one might naively expect
156: that the oscillator approximation should work
157: very well. However, one can show \cite{Z_OA}
158: that in this regime the dominating $N=1$ rescattering contribution
159: (and any odd rescattering)
160: evaluated in the oscillator approximation simply vanishes.
161: For RHIC and LHC conditions, the finite-size effects play a very
162: important role and the oscillator approximation can lead to
163: uncontrolled errors. For this reason, one has to use an accurate
164: parametrization of the three-body cross section. It requires
165: numerical calculations for solving the Schr\"odinger equation.
166:
167: In the present paper we represent the induced
168: gluon spectrum in a new form which is convenient for numerical
169: computations. We, for the first time, calculate the
170: induced gluon emission and the nuclear modification factor
171: for RHIC conditions using a realistic imaginary potential.
172:
173: \vspace{.2cm}
174: \noindent {\bf 2.} We consider a quark with energy $E$ produced
175: in a medium at $z=0$ (we chose the $z$-axis along the quark momentum).
176: The induced gluon spectrum in the gluon fractional longitudinal
177: momentum $x$ reads \cite{Z1}
178: \beq
179: \frac{d P}{d
180: x}=2\mbox{Re}\!
181: \int\limits_{0}^{\infty}\! d
182: z_{1}\!
183: \int\limits_{z_{1}}^{\infty}d
184: z_{2}
185: g(x)\left[{K}(z_{2},\r_{2}|z_{1},\r_{1})
186: -{K}_{v}(z_{2},\r_{2}|z_{1},\r_{1})\right]
187: \Big|_{\r_{1}=\r_{2}=0}\,.
188: \label{eq:10}
189: \eeq
190: Here $K$
191: is
192: the Green's
193: function
194: for the Hamiltonian (acting in the transverse
195: plane)
196: \beq
197: {H}=
198: -\frac{1}{2M(x)}\,
199: \left(\frac{\partial}{\partial \r}\right)^{2}
200: +v(\r,z) +\frac{1}{L_{f}}\,,
201: \label{eq:20}
202: \eeq
203: %
204: \beq
205: v(\r,z)=-i\frac{n(z)\sigma_{3}(\rho)}{2}\,,
206: \label{eq:30}
207: \eeq
208: and
209: \beq
210: {K}_{v}(z_{2},\r_{2}|z_{1},\r_{1})=\frac{M(x)}{2\pi i(z_{2}-z_{1})}
211: \exp\left[\frac{iM(x)(\r_{2}-\r_{1})^{2}}{2(z_{2}-z_{1})}-
212: \frac{i(z_{2}-z_{1})}{L_{f}}\right]
213: \label{eq:40}
214: \eeq
215: is the Green's
216: function
217: for the Hamiltonian (\ref{eq:20}) with
218: $v(\r,z)=0$.
219: In
220: (\ref{eq:20}), the Schr\"odinger mass is
221: $M(x)=Ex(1-x)$,
222: $L_{f}={2Ex(1-x)}/{[m_{q}^{2}x^{2}+m_{g}^{2}(1-x)]}\,\,$
223: is the gluon formation length,
224: here
225: $m_{q}$ and $m_{g}$ are the quark and gluon masses that play
226: the role of the infrared cutoffs at $x\sim 1$
227: and $x\sim 0$ (in the QGP $m_{q,g}$ are given by the quark and gluon
228: quasiparticle masses).
229: %, and $m_{g}$ is the mass of the
230: %
231: In (\ref{eq:30}), $n(z)$ is the number density of QGP, and $\sigma_{3}$
232: is
233: the above mentioned cross section of
234: the color singlet $q\bar{q}g$ system with a particle in the medium.
235: Summation over triplet (quark) and octet (gluon) color states is
236: implicit in (\ref{eq:30}).
237: The $\sigma_{3}$ may depend on $z$ (through the Debye screening mass),
238: however, below we will use $z$-independent $\mu_{D}$.
239: %
240: The vertex factor $g(x)$, entering
241: (\ref{eq:10}), reads
242: \beq
243: g(x)=
244: \frac{\alpha_{s}P(x)}{2M^{2}(x)}\,
245: \frac{\partial}{\partial \r_{1}}\cdot\frac{\partial}{\partial \r_{2}}\,,
246: \label{eq:50}
247: \eeq
248: where $P(x)=C_{F}[1+(1-x)^{2}]/x$ is the splitting function for
249: the $q\rightarrow gq$ transition ($C_{F}$ is the quark Casimir factor).
250: Note that we neglect the spin-flip $q\rightarrow qg$ transition,
251: which gives a small contribution to the quark energy loss.
252:
253:
254: The three-body cross section entering the potential (\ref{eq:30}) can be
255: written as \cite{NZZ,Z1}
256: \beq
257: \sigma_{3}(\rho)=\frac{9}{8}
258: [\sigma_{q\bar{q}}(\rho)+
259: \sigma_{q\bar{q}}((1-x)\rho)]-
260: \frac{1}{8}\sigma_{q\bar{q}}(x\rho)\,,
261: \eeq
262: where
263: \beq
264: \sigma_{q\bar{q}}(\rho)=\alpha_{s}^{2}C_{T}C_{F}\int d\qb
265: \frac{[1-\exp(i\qb\r)]}{(q^{2}+\mu^{2}_{D})^{2}}\,
266: \label{eq:60}
267: \eeq
268: is the dipole cross section for the color singlet $q\bar{q}$ pair
269: ($C_{T}$ is the color Casimir for the thermal parton
270: (quark or gluon)).
271:
272:
273: The spectrum (\ref{eq:10}) can be rewritten as ($L$ is the quark pathlength
274: in the medium)
275: \beq
276: \frac{d P}{d
277: x}=
278: \int\limits_{0}^{L}\! d z\,
279: n(z)
280: \frac{d
281: \sigma_{eff}^{BH}(x,z)}{dx}\,,
282: \label{eq:70}
283: \eeq
284: \beq
285: \frac{d
286: \sigma_{eff}^{BH}(x,z)}{dx}=\mbox{Re}
287: \int\limits_{0}^{z} dz_{1}\int\limits_{z}^{\infty}dz_{2}\int d\r\,
288: g(x)K_{v}(z_{2},\r_{2}|z,\r)
289: \sigma_{3}(\rho)
290: K(z,\r|z_{1},\r_{1}){\Big|}_{\r_{1}=\r_{2}=0}\,.
291: \label{eq:80}
292: \eeq
293: $d\sigma^{BH}_{eff}/dx$ (\ref{eq:80}) can be viewed as an
294: effective Bethe-Heitler
295: cross section, which accounts for the LPM and finite-size effects.
296: %
297: One can represent (\ref{eq:80}) as
298: \beq
299: \frac{d
300: \sigma_{eff}^{BH}(x,z)}{dx}=-\frac{\alpha_{s}P(x)}{\pi M(x)}\mbox{Im}
301: \int\limits_{0}^{z} d\xi
302: \left.\frac{\partial }{\partial \rho}
303: \left(\frac{F(\xi,\rho)}{\sqrt{\rho}}\right)
304: \right|_{\rho=0}\,\,,
305: \label{eq:90}
306: \eeq
307: where the function $F$ is the solution to the radial Schr\"odinger
308: equation for the azimuthal quantum number $m=1$
309: \beq
310: i\frac{\partial F(\xi,\rho)}{\partial \xi}=
311: \left[-\frac{1}{2M(x)}\left(\frac{\partial}{\partial \rho}\right)^{2}
312: -i\frac{n(z-\xi)\sigma_{3}(\rho)}{2}+
313: \frac{4m^{2}-1}{8M(x)\rho^{2}}
314: +\frac{1}{L_{f}}
315: \right]F(\xi,\rho)\,.
316: \label{eq:100}
317: \eeq
318: The boundary condition for $F(\xi,\rho)$ reads
319: $F(\xi=0,\rho)=\sqrt{\rho}\sigma_{3}(\rho)
320: \epsilon K_{1}(\epsilon \rho)$, where
321: $\epsilon=[m_{q}^{2}x^{2}+m_{g}^{2}(1-x)^{2}]^{1/2}$, and
322: $K_{1}$ is the Bessel function.
323: In deriving (\ref{eq:90}), we used the relations \cite{Z_YAF}
324: $$
325: \frac{\partial}{\partial \r_{1}}\cdot\frac{\partial}{\partial \r_{2}}
326: =\frac{1}{2}
327: \left[
328: \left(\frac{\partial}{\partial x_{1}}-i\frac{\partial}{\partial y_{1}}\right)
329: \cdot
330: \left(\frac{\partial}{\partial x_{2}}+i\frac{\partial}{\partial y_{2}}\right)
331: +
332: \left(\frac{\partial}{\partial x_{1}}+i\frac{\partial}{\partial y_{1}}\right)
333: \cdot
334: \left(\frac{\partial}{\partial x_{2}}-i\frac{\partial}{\partial y_{2}}\right)
335: \right]\,,
336: $$
337: %\beq
338: $$
339: \left(\frac{\partial}{\partial x_{2}}\pm i\frac{\partial}{\partial y_{2}}\right)
340: \left.\int\limits_{z}^{\infty}dz_{2}K_{v}(z_{2},\r_{2}|z,\r)
341: \right|_{\r_{2}=0}=
342: \pm\frac{M(x)}{i\pi}\exp(\pm i\phi) K_{1}(\epsilon \rho)\,.
343: %\eeq
344: $$
345: The time variable $\xi$ in (\ref{eq:90}), in terms of the variables $z$
346: and $z_{1}$
347: of equation (\ref{eq:80}), is given by $\xi=z-z_{1}$; i.e.,
348: contrary to the Schr\"odinger equation for the Green's functions
349: entering (\ref{eq:10}), (\ref{eq:90}) represents the spectrum
350: through the solution
351: to the Schr\"odinger equation, which describes evolution of the $q\bar{q}g$
352: system back in time. It allows one to have a smooth boundary condition,
353: which is convenient for numerical calculations.
354:
355: \vspace{.2cm}
356: \noindent {\bf 3.}
357: The jet quenching is
358: usually characterized by the nuclear modification factor
359: (we consider the central rapidity region $y\sim 0$ and
360: suppress the explicit $y$-dependence)
361: \beq
362: R_{AA}(p_{T})=
363: \frac{d\sigma^{AA}(p_{T})/dydp_{T}^{2}}
364: {N_{bin}d\sigma^{pp}(p_{T})/dydp_{T}^{2}}\,,
365: \label{eq:110}
366: \eeq
367: where $d\sigma^{AA}/dydp_{T}^{2}$ and
368: $d\sigma^{pp}/dydp_{T}^{2}$ are the inclusive cross section for $A+A$ and
369: $p+p$ collisions, and
370: $N_{bin}$ is the number of the binary nucleon-nucleon collisions.
371: The effect of the parton energy loss on the high-$p_{T}$ hadron
372: production in $A+A$ collisions can approximately be described in
373: terms of effective
374: hard partonic cross sections, which account for the induced gluon emission
375: \cite{BDMS_quenching}.
376: Using the power-low parametrization for cross section of quark
377: production in $p+p$ collisions $\propto 1/p_{T}^{n(p_{T})}$
378: one can obtain
379: \beq
380: R_{AA}(p_{T})\approx P_{0}(p_{T})+\frac{1}{J(p_{T})}
381: \int\limits_{0}^{1} dz z^{n(p_{T})-2}
382: D_{q}^{h}(z,\frac{p_{T}}{z})
383: \int\limits_{0}^{1} dx(1-x)^{n(p_{T}/{z})-2}
384: \frac{dI(x,\frac{p_{T}}{z(1-x)})}{dx}\,,
385: \label{eq:120}
386: \eeq
387: \beq
388: J(p_{T})=\int\limits_{0}^{1}dz z^{n(p_{T})-2} D_{q}^{h}(z,
389: \frac{p_{T}}{z})\,,
390: \label{eq:121}
391: \eeq
392: where $P_{0}$ is the probability of quark propagation without induced
393: gluon emission,
394: $dI(x,p_{T})/dx$ is the probability distribution in
395: the quark energy loss for a quark with $E=p_{T}$,
396: $D_{q}^{h}(z,p_{T}/z)$ is the quark fragmentation function.
397: Note that, since $n(p_{T})\gg 1$, the $z$-integrands
398: in (\ref{eq:120}),~(\ref{eq:121}) are sharply
399: peaked at $z\approx \bar{z}$ ($\bar{z}$ is the value of $z$
400: at which
401: $z^{n(p_{T}-2} D_{q}^{h}(z,p_{T}/z)$ has a maximum). For this reason
402: (\ref{eq:120}) to quite good accuracy can be approximated as
403: \beq
404: R_{AA}(p_{T})\approx P_{0}(p_{T})+
405: \int\limits_{0}^{1} dx(1-x)^{n(p_{T}/\bar{z})-2}
406: \frac{dI(x,\frac{p_{T}}{\bar{z}(1-x)})}{dx}\,.
407: \label{eq:122}
408: \eeq
409: %
410: We take the $P_{0}$ and spectrum in the radiated energy entering
411: (\ref{eq:120}) in the form
412: \beq
413: P_{0}(E)=\exp\left(-\int_{x_{min}}^{1}dx
414: \frac{dP(x,E)}{dx}\right)\,,
415: \label{eq:123}
416: \eeq
417: %
418: \beq
419: \frac{dI(x,E)}{dx}=\frac{dP(x,E)}{dx}\cdot\exp\left(-\int_{x}^{1}dy
420: \frac{dP(y,E)}{dy}\right)
421: \,,
422: \label{eq:130}
423: \eeq
424: where $x_{min}\approx m_{g}/E$, and
425: it is assumed that the spectrum equals zero at $x\le x_{min}$.
426: Formula (\ref{eq:130}) corresponds to the leading order
427: term of the series in $L/L_{rad}$ (here $L_{rad}$ is the radiation
428: length corresponding to the energy loss
429: $\Delta E\sim E$) of the spectrum derived in
430: \cite{Z_SLACSPS} for the photon emission and, strictly speaking,
431: is only valid for $\Delta E\ll E$.
432: However, even for the more broad domain $\Delta E\lsim E$
433: (which is interesting from the point of view of
434: jet quenching at
435: RHIC) (\ref{eq:130})
436: reproduces the energy loss spectrum evaluated assuming independent gluon
437: radiation to a reasonable accuracy.
438: An accurate calculation of the nuclear modification factor
439: accounting for the higher order terms in $L/L_{rad}$
440: in the approximation of independent gluon emission
441: \cite{BDMS_quenching} does not make sense because this
442: approximation itself does not have any theoretical justification.
443: Note that our spectrum is automatically normalized to unity.
444:
445: The effective exponent $n(p_{T})$ for quark production
446: entering (\ref{eq:120}) is close to that for hadron production,
447: $n_{h}(p_{T})$.
448: A small difference between these quantities (stemming from
449: the $p_{T}$ dependence of the integral (\ref{eq:121}))
450: is given by $n(p_{T})-n_{h}(p_{T})=d\ln{J(p_{T})}/d\ln{p_{T}}$.
451: For $Au+Au$ collisions at $\sqrt{s}=200$ GeV
452: we use $n_{h}(p_{T})=np_{T}/(p_{T}+b)$ with $n=9.99$ and $b=1.219$
453: corresponding to the parametrization $d\sigma/dydp_{T}^{2}=A/(p_{T}+b)^{n}$
454: obtained in \cite{power} for $\pi^{0}$
455: production in $p+p$ collisions. The above procedure allows one
456: to avoid uncertainties of the pQCD calculations
457: of the partonic cross sections.
458:
459:
460: \vspace{.2cm}
461: \noindent {\bf 4}.
462: To fix the $m_{q,g}$ and $\mu_{D}$ we use the results of the
463: analysis of the lattice calculations within the quasiparticle model
464: \cite{LH}.
465: For the relevant range of temperature of the plasma $T\sim (1-3)T_{c}$
466: ($T_{c}\approx 170$ MeV is the
467: temperature of the deconfinement phase transition) the analysis
468: \cite{LH} gives for the quark and gluon quasiparticle masses
469: $m_{q}\approx 0.3$ and $m_{g}\approx 0.4$ GeV.
470: With the above value of
471: $m_{g}$ from the perturbative relation $\mu_{D}=\sqrt{2}m_{g}$ one
472: obtains $\mu_{D}\approx 0.57$ GeV.
473: To study the infrared sensitivity of our results we also perform
474: computations for $m_{g}=0.75$ GeV (with $\mu_{D}=0.57$ GeV).
475: This value of the infrared cutoff for gluon emission in parton-nucleon
476: interaction has been obtained from the analysis of the
477: low-$x$ proton structure function within the dipole BFKL equation
478: \cite{NZZ,NZ_HERA}. It seems to be reasonable for gluon emission in
479: the developed mixed phase and for fast gluons with $L_{f}\gsim L$.
480: This value agrees well with the natural infrared cutoff
481: for gluon emission in the vacuum $m_{g}\sim 1/R_{c}$, where
482: $R_{c}\approx 0.27$ fm is the gluon correlation radius in the QCD vacuum
483: \cite{Shuryak1}.
484: The above two values of $m_{g}$ give reasonable
485: lower and upper limits of the infrared cutoff for the induced gluon
486: emission for RHIC and LHC conditions.
487:
488: We perform numerical calculations for fixed and running $\alpha_{s}$.
489: In the first case we take $\alpha_{s}=0.5$
490: for gluon emission from light quarks and gluons, and $\alpha_{s}=0.4$
491: for the case of $c$-quark. For the running $\alpha_{s}$,
492: we use the parametrization (with $\Lambda_{QCD}=0.3$ GeV)
493: frozen at $\alpha_{s}=0.7$ at low momenta.
494: This parametrization is consistent with the integral
495: of $\alpha_{s}(Q)$ in the interval $0<Q<2$ GeV obtained from the analysis
496: of the heavy quark energy loss (in vacuum) \cite{DKT}. Previously
497: this parametrization was used to describe successfully the HERA data
498: on the low-$x$ proton structure function within the dipole BFKL
499: approach \cite{NZZ,NZ_HERA}.
500: To incorporate the running $\alpha_{s}$ in
501: our formalism, we include $\alpha_{s}$ in the integrand on the right-hand
502: side of (\ref{eq:90}) and take for virtuality $Q^{2}=aM(x)/\xi$.
503: The parameter $a$
504: was adjusted to reproduce the $N=1$ rescattering contribution
505: evaluated in the ordinary diagrammatic approach \cite{Z_kinb}. It gives
506: $a\approx 1.85$. For the dipole cross section (\ref{eq:60}) we take
507: $Q^{2}=\qb^{2}$.
508:
509:
510:
511: We assume the Bjorken \cite{Bjorken} longitudinal expansion of the QGP with
512: $T^{3}\tau=T_{0}^{3}\tau_{0}$ which gives $n(z)\propto 1/z$ for $z>\tau_{0}$.
513: We use the initial conditions suggested
514: in \cite{FMS}:
515: $T_{0}=446$ MeV and $\tau_{0}=0.147$ fm for RHIC, and
516: %
517: $T_{0}=897$ MeV and $\tau_{0}=0.073$ fm for LHC.
518: For RHIC, the above condition were obtained from
519: the charged particle pseudorapidity density
520: $dN/dy\approx 1260$ measured by the PHOBOS experiment
521: \cite{PHOBOS} in $Au+Au$ collisions at $\sqrt{s}=200$ GeV
522: assuming an isentropic expansion and rapid thermolization at
523: $\tau_{0}\sim 1/3T_{0}$.
524: The LHC parameters correspond to $dN/dy\approx 5625$ at $\sqrt{s}=5.5$ TeV,
525: which was estimated in
526: \cite{KMS}. The above initial conditions for RHIC (translated into
527: $\tau_{0}=0.6$ fm) agree with those used in successful hydrodynamic description
528: of $Au+Au$ collisions at RHIC \cite{HeinzKolb}.
529: %
530: Note that, since the dominating $\rho$-scale in (\ref{eq:80})
531: $\propto \sqrt{z}$ for $z\ll L_{f}$,
532: our results are not very sensitive to $\tau_{0}$ (for a given entropy).
533: %
534: The maximum parton pathlength in the hot QCD medium is restricted by
535: the life-time of the QGP (and mixed) phase, $\tau_{max}$.
536: %For the upper limit of the $z$-integration in (\ref{eq:70}) we take
537: \footnote[1]{For our choice of the initial conditions the life-time of QGP is
538: $\sim 3$ fm for RHIC. However,
539: in the interval $\tau\sim 3-6$ fm the density of the
540: mixed phase is practically the same as that for the pure QGP phase.}.
541: We take $\tau_{max}\sim R_{A} \sim 6$ fm.
542: This seems to be a reasonable value for central heavy-ion collisions,
543: since,
544: due to the transverse expansion, the hot QCD matter
545: should cool quickly at $\tau\gsim R_{A}$ \cite{Bjorken}.
546:
547:
548:
549: In Fig.~1 we show the induced gluon spectrum for
550: $q\rightarrow gq$
551: transition for RHIC conditions for the quark pathlength
552: $L=6$ fm obtained
553: with $m_{q}=0.3$ and $m_{q}=1.5$ GeV for $m_{g}=0.4$ and $m_{g}=0.75$ GeV.
554: In Fig.~1 we also show the Bethe-Heitler spectrum (dashed line).
555: One sees that the LPM and finite-size effects strongly suppress
556: the gluon emission. The gluon emission from $c$-quark is suppressed
557: in comparison with light quark due to larger mass, which leads
558: to decreasing of the dominating $\rho$ scale (note that
559: the spectrum is not sensitive to the light quark mass,
560: except for $x\sim 1$). One can see from Fig.~1
561: that, although the Bethe-Heitler spectrum differs strongly for two values of
562: $m_{g}$, the difference becomes relatively small for
563: the spectrum, which accounts for the LPM and finite-size effects.
564: It is connected with the fact that, due to the multiple scattering and
565: finite-size effects, the dominating $1/\rho$-scale becomes larger
566: than $m_{g}$; namely this in-medium scale plays the role of the infrared
567: cutoff at high energies \cite{Z2} (however, of course, for not very
568: high $p_{T}$ the value of $m_{g}$ is still important).
569: We do not show the spectra for
570: running $\alpha_{s}$.
571: They are similar in form (but somewhat suppressed at moderate
572: fractional momenta).
573: The LPM suppression for LHC is considerably stronger than for RHIC, but
574: the spectra are similar in form, and we do
575: not show them as well.
576:
577: In Fig.~2 we plot the quark energy loss
578: $
579: \Delta E=E\int_{x_{min}}^{1} dx x{dP}/{dx}
580: $
581: evaluated for fixed (solid line) and running (dashed line) $\alpha_{s}$
582: for RHIC and LHC conditions for $L=6$ fm with $m_{g}=0.4$ GeV (thick lines)
583: and $m_{g}=0.75$ GeV (thin lines).
584: The results for $\alpha_{s}=0.5$ agree
585: roughly with that for running $\alpha_{s}$ for $E\lsim 10$ GeV
586: but at higher energies the energy dependence is steeper for fixed
587: $\alpha_{s}$. This says that the typical $\rho$-scale becomes
588: smaller with increasing energy. It is also seen from the decrease
589: of the relative difference between the curves for $m_{g}=0.4$ and $0.75$ GeV.
590:
591: In Fig.~3a we compare the nuclear modification factor (\ref{eq:120})
592: for $T_{0}=446$ MeV
593: calculated using the NLO KKP fragmentation functions \cite{KKP}
594: for running $\alpha_{s}$ with that obtained at
595: RHIC \cite{RHIC_data} for central
596: $Au+Au$ collisions at $\sqrt{s}=200$ GeV.
597: For illustration of the dependence on $T_{0}$, in Fig.~3b we also present
598: the results for $T_{0}=375$ MeV.
599: The theoretical curves were obtained
600: for $L=4.9$ fm. It is the typical parton pathlength in
601: the QGP (and mixed) phase for $\tau_{max}=6$ fm. We present the
602: results for $m_{g}=0.4$ and $0.75$ GeV. For $p_{T}\lsim 15$ the results for
603: $\alpha_{s}=0.5$ are close to that for running $\alpha_{s}$
604: and we do not plot them.
605: The results for the quark (solid line)
606: and gluon (dashed line) jets are shown separately
607: (note that for $\sqrt{s}=200$ GeV the quark and gluon contributions
608: are comparable). The suppression is somewhat stronger for gluon jets.
609: One can see from Fig.~3a that the theoretical $R_{AA}$ for
610: $m_{g}=0.4$ is in reasonable
611: agreement with the experimental one.
612: %
613: One should bear in mind, however, that our calculations neglect the
614: collisional energy loss \cite{Bjorken2}.
615: For $p_{T}\sim 5-15$ GeV the collisional energy loss may increase
616: the total energy loss by $\sim 30-40$\%. In this case (if one takes
617: the initial conditions \cite{FMS}) the value $m_{g}=0.75$ GeV would be
618: more preferable for agreement with the RHIC data.
619: As mentioned previously, this value is reasonable
620: for the mixed phase and for gluons with $L_{f}\gsim L$.
621: Since, for $\sqrt{s}=200$ GeV, the medium spends about half
622: of its time in the mixed phase, the effective infrared cutoff may be
623: larger than the gluon quasiparticle mass in the QGP.
624: For this reason, the collisional energy loss may be included without
625: using an unrealistic infrared cutoff for the induced energy loss.
626: The possible remaining small disagreement with the data may be avoided
627: by taking a somewhat smaller value of $T_{0}$
628: (or $\alpha_{s}$). In any case, it is clear that, for such a complicated
629: phenomenon, it is hardly possible to expect a perfect agreement with
630: experiment and the agreement found in the present paper is surprisingly
631: good.
632:
633: The above estimate for the collisional energy
634: loss has been obtained for the pQCD plasma.
635: Presently, there are some indications \cite{Shuryak2} that the medium
636: produced at RHIC may be a strongly coupled QGP.
637: The radiative energy loss should not be very sensitive to the dynamics of
638: the QGP (for the same number density of the QGP).
639: However, it may be important for the collisional
640: energy loss. Unfortunately, the corresponding calculations have not
641: been made yet.
642: %
643: It is interesting that our results give support for the scenario with
644: strongly coupled QGP. Indeed, this scenario requires
645: $\alpha_{s}\gsim 0.5$ \cite{Shuryak2} for thermal partons.
646: The $R_{AA}$ is sensitive to the radiation of soft gluons with an
647: energy of about a few $\mu_{D}$.
648: One can expect that, for such gluons, $\alpha_{s}$ should be close to that
649: for thermal partons. We obtained agreement with the data with
650: $\alpha_{s}$, which is frozen at a value of $0.7$ at low momenta.
651: If $\alpha_{s}$ is frozen at a value below 0.4-0.5, the theoretical
652: $R_{AA}$ strongly disagrees
653: with that observed at RHIC.
654:
655:
656: \vspace{.2cm}
657: \noindent {\bf 5}.
658: In summary, we have represented, within the LCPI approach \cite{Z1},
659: the induced gluon spectrum in a new form convenient for numerical
660: calculations
661: and carried out computations of the induced
662: gluon emission from fast partons in the expanding QGP for
663: RHIC and LHC conditions. The calculations
664: for, the first time, have been performed with a realistic parametrization
665: of the dipole cross section.
666: The theoretical nuclear modification factor calculated for the initial
667: conditions obtained from the charged particle rapidity density
668: observed at RHIC \cite{FMS} and the hydrodynamic simulation of the
669: RHIC data \cite{HeinzKolb}
670: is in a reasonable agreement with that observed at RHIC.
671:
672:
673:
674:
675: \bigskip
676: \noindent {\large \bf Acknowledgements.}
677: I thank R.~Baier and D.~Schildknecht for discussions and their kind
678: hospitality at the University
679: of Bielefeld, where this work was completed.
680: I am also grateful to the High Energy Group of
681: the ICTP for their kind hospitality during my
682: visit to Trieste, where part of this work was done.
683:
684:
685: \begin{thebibliography}{99}
686:
687: \bibitem{RHIC_data}
688: %QUANTUM CHROMO (MANY-BODY) DYNAMICS PROBED IN THE HARD SECTOR AT RHIC.
689: %By David d'Enterria (Nevis Labs, Columbia U.),. Jun 2004. 8pp.
690: %Invited overview talk at 39th Rencontres de Moriond on QCD and High-Energy Hadronic Interactions,
691: %La Thuile, Italy, 28 Mar - 4 Apr 2004.
692: %e-Print Archive: nucl-ex/0406012
693: D. d'Enterria,
694: Invited overview talk at 39th Rencontres de Moriond on QCD
695: and High-Energy Hadronic Interactions,
696: La Thuile, Italy, 28 Mar - 4 Apr 2004;
697: nucl-ex/0406012.
698:
699:
700: \bibitem{BDMPS}
701: R.~Baier, Y.L.~Dokshitzer, A.H.~Mueller, S.~Peign\'e and D.~Schiff,
702: Nucl.\ Phys.\ B{\bf 483}, 291 (1997); {\it ibid.} B{\bf 484}, 265 (197).
703:
704: \bibitem{Z1}
705: B.G. Zakharov, JETP\ Lett. {\bf 63}, 952 (1996).
706:
707: \bibitem{Z2}
708: B.G.~Zakharov, JETP\ Lett. {\bf 65}, 615 (1997).
709:
710: \bibitem{BDMS1}
711: R.~Baier, Y.L.~Dokshitzer, A.H.~Mueller and D.~Schiff,
712: Nucl.\ Phys.\ B{\bf 531}, 403 (1998).
713:
714:
715: %REACTION OPERATOR APPROACH TO NONABELIAN ENERGY LOSS.
716: \bibitem{GLV1}
717: M.~Gyulassy, P.~L\'evai and I.~Vitev,
718: Nucl.\ Phys. B{\bf 594}, 371 (2001).
719: %(hep-ph/0006010);
720:
721: %JET QUENCHING VERSUS JET ENHANCEMENT: A QUANTITATIVE STUDY OF
722: %THE BDMPS-Z GLUON RADIATION SPECTRUM.
723: \bibitem{W1}
724: U.A.~Wiedemann,
725: Nucl.\ Phys.\ A{\bf 690}, 731 (2001).
726: %[HEP-PH 0008241]
727:
728: \bibitem{BSZ}
729: R. Baier, D. Schiff and B.G. Zakharov,
730: Ann.\ Rev.\ Nucl.\ Part. {\bf 50}, 37 (2000).
731: %{\sl hep-ph/0002198} (2000).
732:
733: \bibitem{Z_YAF}
734: B.G.~Zakharov,
735: Phys.\ Atom.\ Nucl. {\bf 61}, 838 (1998).
736:
737: \bibitem{Z3}
738: B.G.~Zakharov, JETP\ Lett. {\bf 70}, 176 (1999).
739:
740:
741: \bibitem{LP}
742: L.D.~Landau and I.Ya.~Pomeranchuk,
743: Dokl.\ Akad.\ Nauk\ SSSR {\bf 92}, 535, 735 (1953).
744:
745: \bibitem{Migdal}
746: A.B.~Migdal, Phys.\ Rev. {\bf 103}, 1811 (1956).
747:
748: \bibitem{OA}
749: C.A.~Salgado and U.A.~Wiedemann,
750: Phys.\ Rev.\ Lett.\ {\bf 89}. 092303 (2002);
751: %e-Print Archive: hep-ph/0204221
752: C.A.~Salgado and U.A.~Wiedemann,
753: Phys.\ Rev.\ D {\bf 68}, 014008 (2003).
754:
755:
756: \bibitem{Z_OA}
757: B.G.~Zakharov, JETP Lett. {\bf 73}, 49 (2001).
758:
759:
760: \bibitem{NZZ}
761: N.N.~Nikolaev, B.G.~Zakharov and V.R.~Zoller,
762: Phys. Lett. B{\bf 328}, 486 (1994).
763:
764: \bibitem{BDMS_quenching}
765: R.~Baier, Yu.L.~Dokshitzer, A.H.~Mueller and
766: D.~Schiff, JHEP {\bf 0109}, 033 (2001);
767: hep-ph/0106347 (2001).
768:
769: \bibitem{Z_SLACSPS}
770: B.G.~Zakharov,
771: Phys. Atom. Nucl. {\bf 62}, 1008 (1999);
772: JETP Lett. {\bf 78}, 759 (2003).
773:
774: \bibitem{power}
775: S.S. Adler {\it et al.}
776: [PHENIX Collaboration],
777: Phys.\ Rev.\ Lett. {\bf 91}, 241803 (2003).
778: %hep-ex/0304038.
779:
780:
781: \bibitem{LH}
782: P.~L\'evai and U.~Heinz,
783: Phys.\ Rev.\ C{\bf 57}, 1879 (1998).
784:
785: \bibitem{NZ_HERA}
786: %DEEP INELASTIC SCATTERING AT HERA AND THE BFKL POMERON.
787: N.N.~Nikolaev and B.G.~Zakharov,
788: Phys. Lett. B{\bf 327}, 149 (1994).
789:
790:
791: \bibitem{Shuryak1}
792: E.V.~Shuryak, Rev.\ Mod.\ Phys. {\bf 65}, 1 (1993).
793:
794:
795: \bibitem{DKT}
796: Yu.L.~Dokshitzer, V.A.~Khoze and S.I.~Troyan,
797: Phys.\ Rev. D{\bf 53}, 89 (1996).
798:
799:
800: \bibitem{Z_kinb}
801: B.G.~Zakharov, JETP Lett. {\bf 80}, 67 (2004).
802:
803:
804: \bibitem{Bjorken}
805: J.D.~Bjorken,
806: Phys.\ Rev. D{\bf 27}, 140 (1983).
807:
808: \bibitem{FMS}
809: R.J.~Fries, B.~M\"uller and D.K.~Srivastava,
810: Phys.\ Rev.\ Lett. {\bf 90}, 132301 (2003).
811: %nucl-th/0208001 (2003).
812:
813:
814: \bibitem{PHOBOS}
815: B.B.~Back {\it et al}. [PHOBOS Collaboration],
816: Phys. Rev. C{\bf 65}, 061901 (2002).
817:
818:
819: \bibitem{KMS}
820: J.~Kapusta, L.D.~McLerran and D.K.~Srivastava,
821: Phys. Lett. B{\bf 283}, 145 (1992).
822:
823:
824: \bibitem{HeinzKolb}
825: U.~W. Heinz and P.~F. Kolb,
826: Nucl.\ Phys. A{\bf 702}, 269 (2002).
827:
828:
829: \bibitem{KKP}
830: B.~A. Kniehl, G. Kramer and B. Potter,
831: Nucl.\ Phys. B{\bf 582}, 514 (2000).
832:
833:
834:
835: \bibitem{Bjorken2}
836: J.D.~Bjorken,
837: Fermilab preprint Pub-82/59-THY (1982).
838:
839: \bibitem{Shuryak2}
840: E.V.~Shuryak,
841: hep-ph/0405066;
842: E.V.~Shuryak and I. Zahed,
843: hep-ph/0403127.
844:
845:
846: \end{thebibliography}
847:
848: %\end{document}
849: \newpage
850:
851: %------------------------------------------------------------------
852: \begin{center}
853: {\Large \bf Figures}
854: \end{center}
855: %------------------------------------------------------------------
856: \begin{figure}[h]
857: \begin{center}
858: \epsfig{file=FIG1.epsi,height=13.9cm}
859: \end{center}
860: \caption[.]{
861: The induced gluon spectrum (solid line) for $q\rightarrow gq$ transition
862: versus the gluon momentum $k$ for
863: RHIC conditions for $E=5,$ 10 and 20 GeV, $L=6$ fm obtained
864: using (\ref{eq:70}),~(\ref{eq:90}) with $\alpha_{s}=0.5$ for
865: $m_{q}=0.3$ GeV (a,b) and $m_{q}=1.5$ GeV (c,d);
866: $m_{g}=0.4$ GeV (left) and 0.75 GeV (right).
867: The Bethe-Heitler spectrum is shown by the dashed line.
868: }
869: \end{figure}
870:
871: \begin{figure}[h]
872: %\begin{center}
873: \epsfig{file=FIG2.epsi,height=8.cm}
874: %\end{center}
875: \caption[.]{
876: The energy dependence of the quark energy loss
877: for RHIC (a) and LHC (b)
878: for $L=6$ fm obtained with
879: $\alpha_{s}=0.5$ (solid line) and running $\alpha_{s}$ (dashed line),
880: $m_{g}=0.4$ GeV (thick lines) and $m_{g}=0.75$ GeV (thin lines),
881: $m_{q}=0.3$ GeV.
882: }
883: \end{figure}
884:
885:
886: \begin{figure}[t]
887: \begin{center}
888: \epsfig{file=FIG3.epsi,height=9.cm}
889: \end{center}
890: \caption[.]{
891: The nuclear modification factor (\ref{eq:120})
892: for central $Au+Au$ collisions at $\sqrt{s}=200$ GeV
893: for quark (solid line) and
894: gluon (dashed line)
895: jets obtained with $m_{g}=0.4$ GeV (thick lines) and $m_{g}=0.75$ GeV
896: (thin lines) for running $\alpha_{s}$.
897: The experimental points (from \cite{RHIC_data}) are for the following:
898: circle - $Au+Au\rightarrow \pi^{0}+X$ (0-10\% central)
899: [PHENIX Collaboration],
900: %
901: square - $Au+Au\rightarrow h^{\pm}+X$ (0-10\% central)
902: [PHENIX Collaboration],
903: star - $Au+Au\rightarrow h^{\pm}+X$ (0-5\% central)
904: [STAR Collaboration].
905: }
906: \end{figure}
907:
908: \end{document}
909: