1: %%%%%%%%%% espcrc1.tex %%%%%%%%%%
2: %
3: % $Id: espcrc1.tex,v 1.2 2004/02/24 11:22:11 spepping Exp $
4: %
5: \documentclass[fleqn,12pt,twoside]{article}
6: % \usepackage{espcrc1}
7: % Use the option 'headings' if you want running headings
8: \usepackage[headings]{espcrc1}
9:
10: % identification
11: \readRCS
12: $Id: espcrc1.tex,v 1.2 2004/02/24 11:22:11 spepping Exp $
13: \ProvidesFile{espcrc1.tex}[\filedate \space v\fileversion
14: \space Elsevier 1-column CRC Author Instructions]
15:
16: % change this to the following line for use with LaTeX2.09
17: % \documentstyle[12pt,twoside,fleqn,espcrc1]{article}
18:
19: % if you want to include PostScript figures
20: \usepackage{graphicx}
21: % if you have landscape tables
22: \usepackage[figuresright]{rotating}
23:
24: % put your own definitions here:
25: % \newcommand{\cZ}{\cal{Z}}
26: % \newtheorem{def}{Definition}[section]
27: % ...
28: \newcommand{\ttbs}{\char'134}
29: \newcommand{\AmS}{{\protect\the\textfont2
30: A\kern-.1667em\lower.5ex\hbox{M}\kern-.125emS}}
31:
32: % add words to TeX's hyphenation exception list
33: \hyphenation{author another created financial paper re-commend-ed Post-Script}
34:
35: % set the starting page if not 1
36: % \setcounter{page}{17}
37:
38: % declarations for front matter
39: \title{Complex Plasmas as a Model for the Quark-Gluon-Plasma Liquid}
40:
41: \author{M. H. Thoma\address[MCSD]{Max-Planck-Institute for Extraterrestrial
42: Physics, \\
43: P.O. Box 1312, 85741 Garching, Germany}%
44: \thanks{Email address: thoma@mpe.mpg.de}}
45:
46: % If you use the option headings,
47: % the title is also used as the running title,
48: % and the authors are also used as the running authors.
49: % You can change that by using \runtitle and \runauthor.
50:
51: %\runtitle{1-column format camera-ready paper in \LaTeX}
52: %\runauthor{S. Pepping}
53:
54: \begin{document}
55:
56: % typeset front matter
57: \maketitle
58:
59: \begin{abstract}
60:
61: The quark-gluon plasma, possibly created in ultrarelativistic heavy-ion
62: collisions, is a strongly interacting many-body parton system. By
63: comparison with strongly coupled electromagnetic plasmas (classical and
64: non-relativistic) it is concluded that the quark-gluon plasma could be in
65: the liquid phase. As an example for a strongly coupled plasma, complex
66: plasmas, which show liquid and even solid phases, are discussed briefly.
67: Furthermore, methods based on correlation functions for confirming and
68: investigating the quark-gluon-plasma liquid are presented. Finally,
69: consequences of the strong coupling, in particular a cross section
70: enhancement in accordance with experimental observations at RHIC, are
71: discussed.
72:
73: \end{abstract}
74:
75: \section{Strongly coupled plasmas}
76:
77: Plasmas are ionized gases forming 99\% of the visible matter in the universe.
78: Plasmas can be produced by high temperatures, e.g., in stars, by electric
79: fields, e.g., in neon tubes used for illumination, or by radiation, e.g., in
80: interstellar plasmas. Plasmas can be classified according to the following
81: properties:
82:
83: 1. A plasma can be non-relativistic or relativistic. In the latter case
84: the velocity of the plasma particles is close to the speed of light.
85: Examples are the electron-positron pair plasma in a supernovae or the
86: quark-gluon-plasma (QGP) in ultrarelativistic heavy-ion collisions.
87:
88: 2. A plasma can be classical or quantum mechanical. In the latter case
89: the de Broglie wave length of the particles becomes of the order of
90: the interparticle distance or larger, which is the case, for instance,
91: for the QGP. In particular, in degenerate plasmas, such as in white dwarfs,
92: quantum effects are dominant.
93:
94: 3. A plasma can be ideal or strongly coupled. In the latter case the
95: interaction energy between the particles becomes of the order of the kinetic
96: particle energy or larger. As we will show this is the case for the QGP.
97: Another interesting example are complex plasmas, which we will discuss in the
98: following.
99:
100: Most plasmas in nature and in the laboratory are non-relativistic,
101: classical, weakly coupled (ideal) plasmas. As a matter of fact,
102: strongly coupled plasmas are hard to produce, since they require low
103: temperatures and high densities, at which a strong recombination sets in.
104:
105: For non-relativistic, classical electromagnetic plasmas the Coulomb coupling
106: parameter
107: \begin{equation}
108: \Gamma = \frac{Q^2}{dT},
109: \label{eq1}
110: \end{equation}
111: where $Q$ is the plasma particle charge, $d$ the interparticle distance,
112: and $T$ the plasma temperature, distinguishes between weakly and
113: strongly coupled plasmas. In the latter case $\Gamma \geq O(1)$ holds.
114: Most plasmas are ideal with $\Gamma <10 ^{-3}$. Examples for strongly
115: coupled plasmas are the ion component in white dwarfs \cite{ref1} or
116: short-living
117: high-density plasmas produced with heavy-ion beams on solid state targets
118: at GSI \cite{ref2}.
119:
120: Obviously, strongly coupled plasmas require a non-perturbative description,
121: e.g., molecular dynamics in the classical case. This has been used for
122: computing the equation of state of
123: one-component plasmas with a pure (repulsive) Coulomb interaction. It turned
124: out that for $\Gamma > 172$ a Coulomb crystal forms \cite{ref3},
125: i.e., it is energetically
126: more favorable if the particles arrange in a regular structure.
127:
128: \bigskip
129:
130: \includegraphics[width=9cm]{fig1.eps}
131:
132: Fig.1: Phase diagram of a strongly coupled Yukawa system from Ref.\cite{ref4}.
133:
134: \bigskip
135:
136: Under most circumstances, however, Debye screening cannot be neglected and
137: plasmas are Yukawa systems with an interaction potential
138: \begin{equation}
139: V(r)=\frac{Q}{r}\> e^{-r/\lambda_D}
140: \label{eq2}
141: \end{equation}
142: with the Debye screening length $\lambda_D$. Then it is convenient to
143: introduce a second parameter $\kappa = d/\lambda_D$. In Fig.1 the phase diagram
144: in the $\Gamma$ - $\kappa$ plane for a repulsive, one-component
145: Yukawa system is shown \cite{ref4}. One recognizes a solid state,
146: either bcc or fcc,
147: for large $\Gamma$ and small $\kappa$. Otherwise a liquid phase exists.
148: Note, however, that in the case of a purely repulsive interaction there is
149: no gas-liquid transition but only a supercritical fluid, i.e., a clear
150: distinction between the gaseous and the liquid state is not possible.
151: However, in the strongly coupled case, $\Gamma > O(1)$, the system behaves
152: more like liquid than a gas, as we will discuss below.
153:
154: \section{Complex Plasmas}
155:
156: Dusty or complex plasmas are multi component plasmas containing ions,
157: electrons, neutral gas, and microparticles, e.g., dust grains. Such
158: a situation can be studied, for example, in a low temperature noble
159: gas discharge plasma, in which micron size particles, e.g., monodisperse
160: plastic spheres with a diameter of 1 - 10 $\mu$m, are injected.
161: Due to the high mobility of the electrons, the microparticles collect
162: mostly electrons on their surface, leading to a large negative
163: charge of the microparticles between 10$^3$ and 10$^5$
164: $e$ depending on the grain size and electron temperature ($T_e= 1 -10$ eV).
165: Since the interparticle distance is typically of the order of 200 $\mu$m
166: and the kinetic energy of the particles corresponds to room temperature
167: due to friction with the neutral gas,
168: $\Gamma \gg 1$ (up to values of 10$^5$) can easily be reached.
169: Therefore, in 1986 it was predicted that a regular ordering in
170: the charged microparticle system, regarded as a massive plasma component,
171: could exist in complex laboratory plasmas \cite{ref5}. This prediction
172: triggered experimental efforts to search for this new state of matter,
173: the so called plasma crystal, which
174: was discovered for the first time in 1994 at the Max-Planck-Institute for
175: Extraterrestrial Physics \cite{ref6} and almost at the same time at
176: other places \cite{ref7,ref8}.
177:
178: An image (top view) of such a plasma crystal is shown in Fig.2.
179: Here a plasma chamber with an extension of a few centimeters is used,
180: in which a low-temperature argon plasma is ignited by a rf discharge.
181: After injecting the microparticles a plasma crystal is formed within
182: a few seconds. For observation a single
183: horizontal layer is illuminated by a laser sheet and the scattered light
184: (Mie scattering) is recorded by a CCD camera. Clearly a hexagonal structure
185: with lattice defects similar as in a real crystal can be observed.
186:
187: \bigskip
188:
189: \includegraphics[width=10cm]{fig2.ps}
190:
191: Fig.2: Top view of the plasma crystal.
192:
193: \bigskip
194:
195: Phase transitions to the liquid phase and the disordered gas phase can
196: be observed by reducing the pressure \cite{ref9}. This leads to a
197: reduction of the
198: neutral gas friction and hence an increase of the temperature of the
199: microparticle system. At the same time the electron temperature decreases
200: leading to a reduction of the charge of the microparticles.
201: Hence the Coulomb coupling parameter (\ref{eq1}) decreases and can become
202: smaller than its critical value for crystallization. Indeed
203: reducing, for instance, the pressure from about 60 Pa,
204: in which a crystal structure exists, to 30 Pa, the microparticles show
205: a liquid behavior. Going to even lower pressure, e.g., 10 Pa, the
206: velocity of the particles is increased by a factor of about 200
207: compared to the solid phase, in which the thermal lattice oscillations
208: correspond to particle velocities of the order of 0.2 mm/s. Thus at 10 Pa
209: the system resembles a gas.
210:
211: The equation of state of a complex plasma and the value of the
212: Coulomb coupling parameter can be determined by the so called
213: pair correlation function, defined as
214: \begin{equation}
215: g({\bf r}) = \frac{1}{N}\> \langle \sum_{i\neq j}^N \delta
216: ({\bf r}+{\bf r_i}-{\bf r_j}) \rangle,
217: \label{eq3}
218: \end{equation}
219: where $N$ is the particle number and ${\bf r_i}$ and ${\bf r_j}$
220: the positions of the particles.
221:
222: In the case of a regular crystal structure, where a long range order
223: exists, the pair correlation function shows pronounced peaks at the
224: locations corresponding to the nearest neighbors, next to nearest
225: neighbors and so on. Of course, due to thermal fluctuations and defects
226: the peaks have a finite width and their height decreases with increasing
227: distance.
228: In the case of a liquid, where only a short range order corresponding
229: to a fixed interparticle distance in the incompressible fluid is
230: present, the pair correlation function exhibits only one clear peak and
231: some times one or two small and broad additional peaks.
232: Finally, in a gas corresponding to a disordered system no peaks show up
233: in the pair correlation function.
234:
235: Before we turn to the QGP, let me mention that the investigation of complex
236: plasmas is hampered by the presence of gravity in the laboratory. The
237: microparticles can be suspended against gravity only by an electric field
238: (typically a few V/cm) in a small region of the plasma chamber, the
239: plasma sheath above the bottom where the field can be strong enough.
240: Hence only small systems in vertical
241: directions, e.g., quasi 2-dimensional crystals, can be built up.
242: Furthermore, the plasma
243: conditions in the plasma sheath are very complicated rendering the
244: interpretation of the results difficult. Finally, the gravitational force
245: is comparable to the interparticle force, thus strongly disturbing
246: the system, e.g., the crystal structure, and making some measurements even
247: impossible. Therefore we perform experiments with complex plasmas also under
248: microgravity conditions, i.e., in parabolic flights, in sounding rockets,
249: and on board of the International Space Station where we installed the
250: first scientific experiment (PKE-Nefedov) \cite{ref10}.
251:
252: Finally, let me discuss some applications of complex plasmas. Beside
253: investigating the plasma crystal and its melting a large variety of different
254: experiments can de conducted, e.g., phonons and plasma waves can be
255: excited, instabilities can be observed, shock waves and Mach cones can be
256: produced, shear flow can be studied, etc. In general, complex plasmas
257: are ideal model systems to study the dynamical behavior of solids,
258: liquids, and plasmas on the microscopic level in real time. In addition,
259: charged dust systems and dust-plasma interactions play an important role
260: in astrophysics (comets, interstellar clouds, planet formation, etc.),
261: in environmental research (chemical reactions involving
262: micrometeorites in the ionosphere), as well as in plasma technology
263: (dust contamination in the micro-chip production using plasma etching,
264: dust in fusion reactors, etc.).
265:
266: \section{Applications to the quark-gluon plasma}
267:
268: For applying the ideas and methods used for strongly coupled
269: electromagnetic plasmas, such as complex plasmas, to the
270: physics of the QGP, we first estimate the interaction parameter.
271: In the case of QCD it reads \cite{ref11}
272: \begin{equation}
273: \Gamma \simeq 2 \frac{C\alpha_s}{dT},
274: \label{eq4}
275: \end{equation}
276: where $C=4/3$ or $C=3$ is the Casimir invariant of quarks or gluons,
277: respectively. The factor 2 in (\ref{eq4})
278: comes from the fact that in ultrarelativistic systems the magnetic
279: interaction is as important as the electric.
280: Assuming a temperature $T=200$ MeV which corresponds to
281: a coupling constant $\alpha_s = 0.3$ to 0.5 and an interparton distance
282: $d$ of about 0.5 fm, we find $\Gamma = 1.5$ - 6. Hence the QGP is
283: a strongly coupled plasma at temperatures which can be reached
284: in accelerator experiments, and it is conceivable that it behaves more
285: like a liquid than a gas. Indeed RHIC data (elliptic flow, particle spectra)
286: \cite{ref11a},
287: which can be described
288: well by hydrodynamics with a negligible viscosity and which suggest
289: a fast thermalization, indicate such a behavior
290: \cite{ref12,ref13,ref14,ref15}.
291:
292: Therefore
293: the phase diagram of hot hadronic matter could be possibly extended
294: by another phase transition from the liquid to the gas phase
295: of the QGP at a temperature of maybe a few hundred MeV as sketched
296: in Fig.3.
297: However, this requires an attractive as well as repulsive component
298: of the potential, e.g., a Lennard-Jones type potential, which is not
299: the case in QCD where the interaction between partons in the various
300: channels is either attractive or repulsive \cite{ref16}.
301: However, it is known that in a strongly coupled plasma purely repulsive
302: forces can obtain an attractive component at certain distances due
303: to non-linear effects \cite{ref17}. Hence it might be worthwhile to look
304: for a gas-liquid transition in heavy-ion collisions, in particular at LHC.
305: Regardless whether this phase transition exists or whether there is only
306: a supercritical QGP liquid, the QGP should behave more like a liquid
307: at temperatures close to the deconfinement transition and like a gas at
308: temperatures far above this transition.
309:
310: \bigskip
311:
312: \includegraphics[width=10cm]{fig3.eps}
313:
314: Fig.3: Sketch of a phase diagram of hadronic matter with a possible
315: gas-liquid transition in the QGP phase.
316:
317: \bigskip
318:
319: For verifying and investigating the liquid state
320: quantitatively one could consider
321: experimentally as well as theoretically the so called static structure
322: function which is closely related to the Fourier transform of the pair
323: correlation function (\ref{eq3}). The static structure function is a standard tool
324: for the experimental and theoretical analysis of liquids \cite{ref18}.
325: The qualitative behavior of the static structure functions for
326: liquids and gases is shown in Fig.4. In the case of a liquid the static
327: structure function shows oscillations with decreasing amplitudes for
328: large momenta. For an interacting gas, on the other hand, the static
329: structure function increases monotonically reaching quickly a saturation
330: value.
331:
332: \bigskip
333:
334: \includegraphics[width=10cm]{fig4.eps}
335:
336: Fig.4: Qualitative behavior of the static structure functions for a liquid
337: and a gas.
338:
339: \bigskip
340:
341: In Ref.\cite{ref19} we defined the static structure function
342: for the case of the QGP and showed that it is related to the longitudinal
343: part of the QCD polarization tensor. Furthermore we
344: argued that QCD lattice simulations should
345: be able to prove the liquid behavior of the strongly coupled QGP
346: by computing the static structure function.
347: To demonstrate the use of this definition and as
348: a reference for lattice calculations, we have calculated the
349: static structure function within the Hard Thermal Loop (HTL) approximation,
350: yielding \cite{ref19}
351: \begin{equation}
352: S(p)
353: =\frac{2N_fT^3}{n}\> \frac{p^2}{p^2+m_D^2},
354: \label{eq5}
355: \end{equation}
356: where $N_f$ is the number of light quark flavors, $n$ the parton density, and
357: $m_D=1/\lambda_D$ the Debye screening mass which is proportional to
358: $gT$ in the HTL approximation. This $p$-dependence clearly belongs
359: to an interacting gas which is not surprising as the HTL approximation
360: is based on the high-temperature assumption, $T\gg T_c$.
361:
362: The pair correlation function follows from the Fourier
363: transform of $S(p)-1$ as
364: \begin{equation}
365: g(r) = -\frac{N_fT^3}{2\pi n}\> \frac{m_D^2}{r}\> e^{-m_Dr},
366: \label{eq6}
367: \end{equation}
368: showing no peaks which corresponds, of course, also to the gas phase.
369:
370: Finally, we want to point out that strongly coupled plasmas show in general
371: a cross section enhancement for the interaction of the particles
372: within the plasma. The reason is that the Coulomb radius, defined by
373: $r_c=Q^2/E$ with the particle energy $E$, in a strongly coupled plasma
374: is of the order of the Debye screening length or even larger. Hence
375: the standard Coulomb scattering theory has to be modified since due to the
376: strong interaction particles outside of the Debye sphere contribute
377: significantly and the Debye screening length cannot be used as an infrared
378: cutoff. This modification leads, for example, to the experimentally observed
379: enhancement of the so called ion drag force in complex plasmas which is
380: caused by the ion-microparticle interaction \cite{ref20}.
381:
382: In the QGP at $T\simeq 200$ MeV, the ratio $r_c/\lambda_D = 1$ - 5,
383: leading to a parton cross section enhancement in the QGP by a factor
384: of 2 - 9 \cite{ref11} compared to perturbative results. Additional
385: cross section enhancement could come from non-linear (modification of the
386: Yukawa potential to a power law potential at large distances) and
387: non-perturbative effects.
388: An enhanced parton cross section leads to a reduced mean free path $\lambda $
389: of the partons in the QGP which corresponds
390: to a small viscosity $\eta \sim \lambda$ and a fast thermalization
391: as indicated by RHIC experiments.
392:
393: A cross section enhancement of the elastic parton
394: scattering by an order of magnitude compared to perturbative results
395: has also been postulated
396: in Ref.\cite{ref20a} by considering elliptic flow data and
397: particle spectra observed at RHIC. The assumption of a strongly coupled
398: QGP, which requires an infrared cutoff smaller than the Debye mass
399: and in which non-linear and non-perturbative effects are important,
400: gives a natural explanation for this enhancement.
401:
402: Another consequence would be the enhancement of the collisional
403: energy loss
404: \begin{equation}
405: \left (\frac{dE}{dx}\right )_{\rm coll} \simeq \frac{\Delta E}{\lambda },
406: \label{eq7}
407: \end{equation}
408: where $\Delta E$ is the energy transfer per collision. This contribution
409: to the total energy loss receives an enhancement from the soft part
410: due to the mean free path reduction for low-energy partons \cite{ref20b}.
411: On the other hand,
412: it can be expected that the radiative energy loss is suppressed in
413: the strongly coupled QGP by the Landau-Pomeranchuk-Migdal effect.
414: In the case of a dense and strongly coupled plasma the emission
415: of gluons could be suppressed by this formation time effect.
416: Indeed experimental indications for explaining jet quenching with a
417: significant contribution from the collisional energy loss
418: have been discussed recently \cite{ref21,ref22}.
419:
420: \section{Conclusions}
421:
422: Within the last years strongly coupled plasmas became of increasing
423: interest in fundamental research as well as in technology applications.
424: The QGP as well as complex plasmas are important examples of strongly
425: coupled plasmas. Definitely, the QGP is the most challenging strongly coupled
426: plasma from the experimental as well as theoretical point of view.
427: Complex plasmas, on the other hand, can easily be produced and studied.
428: Therefore they can be used as ideal models to investigate fundamental
429: aspects of other many-body systems, such as the dynamic evolution of phase
430: transitions on the microscopic level. Complex plasmas
431: can also be used to learn some interesting lessons about the QGP
432: by analogy, such as the existence of a possible liquid phase and
433: a gas-liquid transition, the use of correlation functions for investigating
434: the equation of state, or cross section enhancement in a strongly coupled
435: plasma.
436:
437: \begin{thebibliography}{9}
438: \bibitem{ref1} S. Ichimaru, Rev. Mod. Phys. 54 (1982) 1017.
439: \bibitem{ref2} E. Dewald et al., IEEE Trans. Plasma Sc. 31 (2003) 221.
440: \bibitem{ref3} W.L. Slattery, G.D. Doolen, and H.E. Witt, Phys. Rev. A
441: 21 (1980) 2087.
442: \bibitem{ref4} S. Hamaguchi, R.T. Farouki, D.H.E. Dubin, Phys. Rev. E
443: 56 (1997) 4671.
444: \bibitem{ref5} H. Ikezi, Phys. Fluids 29 (1986) 1764.
445: \bibitem{ref6} H.M. Thomas et al., Phys. Rev. Lett. 73 (1994) 652.
446: \bibitem{ref7} J.H. Chu and I Lin, Phys. Rev. Lett. 72 (1994) 4009.
447: \bibitem{ref8} Y. Hayashi and K. Tachibana, Jpn. J. Appl. Phys. 33 (1994)
448: L804.
449: \bibitem{ref9} H.M. Thomas and G.E. Morfill, Nature 379 (1996) 806.
450: \bibitem{ref10} A.P. Nefedov et al., New J. Phys. 5 (2003) 2633.
451: \bibitem{ref11} M. H. Thoma, J. Phys. G 31 (2005) L7 and Erratum J. Phys.
452: G 31 (2005) 539
453: \bibitem{ref11a} K. Adcox et. al. (Phenix collaboration), Nucl. Phys.
454: A 757 (2005) 184.
455: \bibitem{ref12} E.V. Shuryak, J. Phys. G 30 (2004) S122.
456: \bibitem{ref13} M. Gyulassy and L. McLerran, Nucl. Phys. A 750 (2005) 30.
457: \bibitem{ref14} U. Heinz, AIP Conf. Proc. 739 (2005) 163.
458: \bibitem{ref15} W. Cassing and A. Peshier, Phys. Rev. Lett. 94 (2005) 172301.
459: \bibitem{ref16} Discussion contribution by K. Rajagopal after this talk.
460: \bibitem{ref17} V.N. Tsytovich, JETP Lett. 78 (2003) 1283.
461: \bibitem{ref18} J.-P. Hansen and I.R. McDonald, Theory of Simple Liquids,
462: Academic Press, London, 1986.
463: \bibitem{ref19} M.H. Thoma, {\it hep-ph/0504163}.
464: \bibitem{ref20} V. Yaroshenko et al., Phys. Plasmas 12 (2005) 093503.
465: \bibitem{ref20a}D. Molnar and M. Gyulassy, Nucl. Phys. A 697 (2002) 495.
466: \bibitem{ref20b}W. Cassing, A. Peshier, and M.H. Thoma, work in progress.
467: \bibitem{ref21} M.G. Mustafa and M.H. Thoma, Acta Phys. Hung. A 22 (2005) 93.
468: \bibitem{ref22} P. Jacobs and M. van Leeuwen, these proceedings.
469:
470: \end{thebibliography}
471:
472:
473: \end{document}
474:
475:
476:
477:
478:
479:
480: