hep-ph0011003/Dcc.tex
1: \documentstyle[aps,epsfig]{revtex}
2: \begin{document}
3: \draft
4: %\tighten
5: \title{Interplay   of  friction  and  noise  and  enhancement  of
6: disoriented chiral condensate}
7: \author{\bf A. K. Chaudhuri \cite{byline}}
8: \address{
9: Variable Energy Cyclotron Centre\\
10: 1-AF, Bidhan Nagar, Calcutta- 700 064}
11: %\date{Started writting on 26, august 2000}
12: %Submitted to Phys. Rev. C on August, 1999
13: \maketitle
14: \begin{abstract}
15: Using  the  Langevin  equation  for the linear $\sigma$ model, we
16: have investigated the effect of  friction  and  noise  on  the
17: possible  disoriented  chiral  condensate  formation. Friction and
18: noise are supposed to suppress  longwavelength  oscillations  and
19: growth   of   disoriented   chiral  condensate  domains.  Details
20: simulation shows that for  heavy  ion  collisions,  interplay  of
21: friction  and  noise  occur  in  such  a manner that formation of
22: disoriented chiral condensate domains are enhanced.
23: 
24: \end{abstract}
25: 
26: \pacs{25.75.+r, 12.38.Mh, 11.30.Rd}
27: 
28: In recent years there is much excitement about the possibility of
29: formation  of  disoriented  chiral condensate (DCC). The idea was
30: suggested by  Rajagopal  and  Wilczek  \cite{ra93}  and  also  by
31: Bjorken  and others \cite{bj93}. In hadron-hadron or in heavy ion
32: collisions, a macroscopic region of  space-time  may  be  created
33: within  which  the  chiral order parameter is not oriented in the
34: same direction in the internal $O(4)=SU(2) \times SU(2)$ space as
35: the ordinary vacuum. The misaligned condensate has the same quark
36: content  and  quantum  numbers  as  do  pions   and   essentially
37: constitute  a  classical  pion  field.  The  system  will finally
38: relaxes to the true vacuum and in the process can  emit  coherent
39: pions.  Possibility  of  producing classical pion fields in heavy
40: ion collisions had been discussed earlier by Anslem  \cite{an91}.
41: DCC  formation in hadronic or in heavy ion collisions can lead to
42: the spectacular events that some portion of the detector will  be
43: dominated by charged pions or by neutral pions only. In contrast,
44: in  a  general  event  all  the  three pions ($\pi^+$,$\pi^-$ and
45: $\pi^0$) will be equally well produced.  This  may  then  be  the
46: natural explanation of the so called Centauro events \cite{la80}.
47: 
48: Microscopic  physics governing DCC phenomena is not well known. It
49: is in the regime of non-perturbative QCD  as  well  as  nonlinear
50: phenomena,   theoretical  understanding  of  both  of  which  are
51: limited. One thus uses some effective field  theory  like  linear
52: $\sigma$ model with various approximations to simulate the chiral
53: phase transition \cite{rand,gavi,asak}. In the linear sigma model
54: chiral  degrees  of  freedom  are  described by the the real O(4)
55: field $\Phi=(\sigma,\roarrow{\pi})$. Because of  the  isomorphism
56: between  the  groups  $O(4)$  and $SU(2) \times SU(2)$, the later
57: being the appropriate group for  two  flavour  QCD,  linear  sigma
58: model  can  effectively  model  the  low  energy  dynamic  of QCD
59: \cite{bo96}. Explicit simulation with linear sigma model, indicate
60: that DCC depends critically on the initial field  configurations.
61: With  quench like initial condition DCC domains of 4-5 fm in size
62: can form \cite{asak}. Initial conditions other than  quench  lead
63: to  much  smaller  domain  size.  Quench scenario assume that the
64: effective potential governing the evolution of long  wave  length
65: modes  immediately  after  the phase transition at $T_c$ turns to
66: classical one at zero temperature. It can happen only in case  of
67: very  rapid  cooling  and expansion of the fireball. In heavy ion
68: collisions quench like initial conditions are unlikely.
69: 
70: Very  recently, effect of external media on possible DCC is being
71: investigated   \cite{gr97,bi97,zu00,ri98,ch99a,ch99b,ch99c,ch00}.
72: Indeed,  in  heavy ion collisions, even if some region is created
73: where   chiral   symmetry   is  restored,  that  region  will  be
74: continually interacting with surrounding medium  (mostly  pions).
75: The  surrounding  medium  can be conveniently modelled by a white
76: noise source  and  one  can  use  Langevin  equation  for  linear
77: $\sigma$  model to simulate the DCC formation under the influence
78: of external media. Recently it has been shown  that  in  $\phi^4$
79: model, hard modes can be integrated out to obtain a Langevin type
80: of  equations  for  the  soft modes \cite{gr97}. Biro and Greiner
81: \cite{bi97} using a Langevin equation  for  the  linear  $\sigma$
82: model,  investigated the interplay of friction and white noise on
83: the evolution and stability of {\em zero mode}  pion  fields.  In
84: general  friction  and  noise  reduces  the amplification of zero
85: modes. But in some trajectories, large  amplification  may  occur
86: \cite{bi97}. We also obtain similar results \cite{ch99a}.
87: 
88: It is popular expectation that friction and noise will reduce the
89: DCC  formation  probability. The expectation was found to be true
90: in  one   dimensional   calculations   with   zero   modes   only
91: \cite{bi97,ch99a}.  To see how far this expectation is valid when
92: higher  modes  are  included,  in  the  present  letter  we  have
93: investigated  DCC  formation in 3+1 dimension. Two scenarios were
94: considered. In scenario I, we  use  the  equation  of  motion  of
95: linear $\sigma$ model with quenched initial condition to simulate
96: DCC  formation  without  friction  and  noise. In scenario II, we
97: solve the Langevin equation for linear $\sigma$ model,  with  the
98: same  quenched  initial  condition  to  simulate DCC formation in
99: presence of friction and noise. Comparative study  of  these  two
100: scenarios reveal that certainly the noise and friction affect DCC
101: formation   probability.  However  contrary  to  one  dimensional
102: calculations, large DCC formation is seen in scenario  II  rather
103: than  in  scenario  I.  Indeed,  it  will be shown that large DCC
104: domains are formed in scenario II only, not in scenario  I.  This
105: effect is particular to heavy ion collisions as explained below.
106: 
107: The  Langevin  equation for linear $\sigma$ model, at temperature
108: $T$ can be written as,
109: 
110: \begin{equation}
111: [\frac{\partial^2}{\partial \tau^2} +(\frac{1}{\tau}+\eta_i)
112: \frac{\partial}{\partial \tau}
113: -\frac{\partial^2}{\partial x^2} -\frac{\partial^2}{\partial y^2} -
114: \frac{1}{\tau^2} \frac{\partial^2}{\partial Y^2}
115: +\lambda (\Phi^2 - f^2_\pi -T^2/2)] \Phi
116:  = \zeta_i (\tau ,x,y,Y)\label{1}
117: \end{equation}
118: 
119: \noindent  where  $\Phi_i=(\sigma,\pi_1,\pi_2,\pi_3)$.  $\tau$ is
120: the  proper  time  and  Y  is  the  rapidity,   the   appropriate
121: coordinates to describe heavy ion collisions. $\eta_{\sigma,\pi}$
122: and  $\zeta_{\sigma,\pi}$  are  the friction coefficients and the
123: white noise for the $\sigma$ and $\pi$ fields. We  note  that  if
124: the  friction and the noise terms are dropped from eq.\ref{1} the
125: resulting equation is for the scenario I, i.e. just  the  equation
126: of motion of linear $\sigma$ model.
127: 
128: The  noise  source  $\zeta$  and  the  friction  $\eta$  are  not
129: independent. They are  related  by  the  fluctuations-dissipation
130: relation.  We  use  white  noise  source  with  zero  average and
131: correlation  as   demanded   from   the   fluctuation-dissipation
132: relation,
133: 
134: \begin{mathletters}
135: \begin{eqnarray}
136: <\zeta_a(\tau,x,y,Y)> =&&0\\
137: <\zeta_a(\tau,x,y,Y)\zeta_b(\tau^\prime,x^\prime,y^\prime,Y^\prime)>
138: =&& 2 T \eta \frac{1}{\tau} \delta(\tau-\tau^\prime) \delta(x-x^\prime)
139: \delta(y-y^\prime)
140: \delta(Y-Y^\prime) \delta_{ab}
141: %\int <\zeta_a(\tau_1,x_1,y_1,Y_1)\zeta_b(\tau_2,x_2,y_2,Y_2)> d^4x
142: %=&& 2 T \eta \delta_{ab}
143: \end{eqnarray}
144: \label{2}
145: \end{mathletters}
146: 
147: \noindent where $a,b$ corresponds to $\pi$ or $\sigma$ fields.
148: A few words are necessary about the use of temperature in the
149: noise term.  We are approximating DCC, which is a non-equilibrium
150: phenomena as a equilibrium one. Such an approximation is valid
151: when the system is not far from equilibrium. Indeed, fluctuation-dissipation
152: theorem is valid for such a system only.
153: 
154: Friction coefficients is an important ingredient for the Langevin
155: equation.  In  our  earlier  study,  we  have  assumed  that both
156: $\sigma$ and  $\pi$  evolve  under  the  influence  of  a  common
157: friction  \cite{ch99c,ch00}.  In  the  present work we treat them
158: separately. $\sigma$ and $\pi$ fields evolve under the  influence
159: of  friction,  appropriate  for  them.  Friction coefficients for
160: $\sigma$ and $\pi$ have been calculated by Rischke \cite{ri98},
161: 
162: \begin{mathletters}
163: \begin{eqnarray}
164: \eta_\pi =&& (\frac{4 \lambda f_\pi}{N})^2 \frac{m_\sigma^2}{4  \pi
165: m_\pi^3} \sqrt{1-\frac{4 m_\pi^2}{m_\sigma^2}}
166: \frac{1-exp(-m_\pi/T)}{1-exp(-m_\sigma^2/2 m_\pi T)}
167: \frac{1}{exp(m_\sigma^2-2m_\pi^2)/2m_\pi T)-1}\\
168: \eta_\sigma   =&&  (\frac{4  \lambda  f_\pi}{N})^2  \frac{N-1}{8\pi
169: m_sigma}
170: \sqrt{1-\frac{4 m_\pi^2}{m_\sigma^2}} \coth \frac{m_\sigma}{4T}
171: \end{eqnarray} \label{3}
172: \end{mathletters}
173: 
174: Solution  of  eq.\ref{1}  require  initial fields configurations.
175: They were distributed according to a random Gaussian with,
176: 
177: \begin{mathletters}
178: \begin{eqnarray}
179: <\sigma>=&&(1-f(r))f_\pi \\
180: <\pi_i>=&&0 \\
181: <\sigma^2>-<\sigma>^2 = <\pi_i^2>-<\pi_i>^2=   && f_\pi^2/6 f(r)\\
182: < \dot{\sigma}>=&& <\dot{\pi_i}>=0\\
183: <\dot{\sigma}^2>=<\dot{\pi}^2>=&& 4 f_\pi^2/6 f(r)
184: \end{eqnarray}
185: \label{4}
186: \end{mathletters}
187: 
188: The interpolation function
189: 
190: \begin{equation}
191: f(r)=[1+exp(r-r_0)/\Gamma)]^{-1}
192: \end{equation}
193: 
194: \noindent  separates  the  central  region  from  the rest of the
195: system. We have  taken  $r_0$=6.4  fm  and  $\Gamma$=.7  fm.  The
196: initial field configurations corresponds to quench like condition
197: \cite{ra93}  and  as  told  in  the  beginning are unlikely to be
198: obtained in a heavy ion collisions. We still use it as  they  are
199: the  most  favourable  initial  conditions  to  produce  DCC like
200: phenomena. In the simulation results presented below, the initial
201: time and  temperature  were  assumed  to  be  $\tau_i$=1fm/c  and
202: $T_i$=200  MeV.  Effect  of  expansion  was  included through the
203: cooling law,
204: 
205: \begin{equation}
206: T(\tau) = T_i \frac{\tau_i}{\tau}
207: \end{equation}
208: 
209: \noindent which is rather fast for heavy ion collisions. However,
210: we choose to use it as DCC formation probability increases if the
211: system cools rapidly.
212: 
213: With  the initial conditions and the cooling law most appropriate
214: for DCC formation, we have solved the Langevin equation (scenario
215: II) on a $32^3$ lattice, with lattice spacing of a=1 fm, using  a
216: time  step of a/10 fm/c. We also use periodic boundary conditions.
217: The equation of motion for linear $\sigma$ model (scenario I) was
218: solved similarly, with  identical  initial  conditions.  We  have
219: continued the evolution of the fields for about 10 fm/c.
220: 
221: We   define   a   correlation   function   at   rapidity  $Y$  as
222: \cite{asak}
223: 
224: \begin{equation}
225: C(r,\tau)  =  \frac{  \sum_{i,j} \pi(i) \dot \pi(j)}{\sum_{i,j}
226: |\pi(i)| |\pi(j)|}
227: \end{equation}
228: 
229: \noindent  where  the sum is taken over those grid points i and j
230: such that the distance between i and j is r. In  fig.1,  we  have
231: compared the correlation function in scenario I and II. Initially
232: there  is  no  correlation length beyond the lattice spacing of 1
233: fm. Correlations starts to develop at later  time.  It  increases
234: for  about  7fm/c,  then  decreases  again. Interestingly, larger
235: correlation length is  obtained  in  the  scenario  II,  than  in
236: scenario  I.  Thus at 7 fm/c, correlation length in scenario I is
237: only $\sim$ 2 fm, while that in  scenario  II  is  $\sim$  4  fm.
238: Increased  correlation  in  scenario  II  is  contrary to popular
239: expectation that friction and noise will reduce correlation.
240: 
241: Enhancement  of  correlation length in scenario II, with friction
242: and noise is corroborated in the pion field distribution also. In
243: fig.2,  we  have  shown  the  xy  contour  plot  of  the  $\pi_2$
244: component, at rapidity Y=0. Field distribution at $\tau_i$=1 fm/c
245: and  after  7  fm/c of evolution are shown. Initially there is no
246: correlation. Domain like structure is seen both in scenario I and
247: II, after 7fm/c evolution . The positive and negative  components
248: of the $\pi_2$ separates out. Here again, much larger domains are
249: formed  in  scenario  II than in scenario I. It may be noted that
250: domain like structure seen in one of the component of $\pi$ field
251: do not  necessarily  convert  into  physical  domains.  They  are
252: indication  of  larger  correlation  length only. Physical domain
253: should contain either charged  or  neutral  pion  only.  Thus  in
254: physical  domain  neutral  to  total  pion  ratio  should  differ
255: considerably from the isospin symmetric value of 1/3.
256: 
257: Assuming that the pion density is proportional to the amplitude's
258: square,  in  fig.3, we have shown the contour plot of the neutral
259: to total pion ratio, at rapidity $Y=0$.
260: 
261: \begin{equation}
262: R_3(x,y,Y,\tau)=\frac{\pi^2_3}{\pi^2_1+\pi^2_2+\pi^2_3}
263: \end{equation}
264: 
265: Very  small  or large value of the ratio over an extended spatial
266: zone   will  be  definite  indication  of  domain  formation.  As
267: expected,  initially  there  is  no  domain  like  structure.  In
268: scenario  I, we donot find any large domain like structure in the
269: ratio  $R_3$  even  after  7  fm/c  evolution. Thus while $\pi_2$
270: component of the $\pi$ field shows domain  like  structure  at  7
271: fm/c,  in terms of physical pions there is no domain structure in
272: scenario I. Large DCC domains are not formed in  linear  $\sigma$
273: model, even with quenched initial condition and fast cooling law.
274: In  scenario II however we can find some extended zone with large
275: or very small value of  the  ratio  $R_3$.  Thus  physical  pions
276: evolve  into domain like structure in scenario II, with noise and
277: friction, rather than in scenario I.
278: 
279: Present  simulation  indicate  that  friction and noise do indeed
280: enhance DCC domain formation possibility. The simulation  results
281: are contrary to expectation from one dimensional calculations. As
282: such  noise and friction are supposed to reduce the amplification
283: of long wavelength modes. We believe  that  present  results  are
284: particular to heavy ion collisions where proper time and rapidity
285: are  the  most appropriate coordinate systems. In this coordinate
286: system, correlation of the noise  decreases  with  (proper)  time
287: ($1/\tau$  dependence,  see eq.\ref{2}). Physically the source of
288: noise i.e. the medium surrounding the zone where chiral  symmetry
289: is  restored,  fly  away  with  time,  reducing  the correlation.
290: However, the friction  coefficients  $\eta_{\sigma,\pi}$  remains
291: more  or  less  same in the temperature range considered. Thus at
292: later time evolution of the fields are determined mainly
293: by the friction.
294: We also not  that  $\eta_\sigma$  is  considerably  large
295: ($\sim  3 fm^{-1}$). Then once the trajectory enters into unstable
296: regime, large friction opposes its tendency to come  out  of  the
297: instability.  Friction  forces  the  trajectory  to  remain in the
298: unstable regime for longer duration. Indeed, in one dimension, we
299: have obtained similar result  \cite{ch99a,ch99b}.  3d  simulation
300: confirms our results in one dimension.
301: 
302: Though  noise and friction enhances the DCC formation probability,
303: it may not be easy to detect it. In  fig.4,  we  have  shown  the
304: rapidity distribution of the neutral to pion ratio,
305: 
306: \begin{equation}
307: \frac{N_{\pi_0}}{N_{\pi_{total}}}=\frac{\int  \pi^2_3 dx dy}{\int
308: (\pi^2_1+\pi^2_2+\pi^2_3) dx dy}
309: \end{equation}
310: 
311: \noindent  at  different time intervals. Through out the range of
312: rapidity, the ratio fluctuate about its isospin  symmetric  value
313: of  1/3.  As expected fluctuations are larger in scenario II than
314: in scenario I. However, if we remember that the pions  that  will
315: be detected are integrated over time, then it is obvious that the
316: fluctuations   will   be   considerably   less.   Thus   even  in
317: event-by-event analysis, it will be difficult to tell  about  DCC
318: formation  from  rapidity  distribution  of pions only. Much more
319: study is needed to resolve the issue.
320: 
321: In  summary,  we  have  considered  disoriented chiral condensate
322: domain formation in two scenarios, one without noise and friction
323: (scenario I) and the other with noise and friction (scenario II).
324: In scenario I, equation of motion for the linear  $\sigma$  model
325: fields  and  in  scenario  II,  the Langevin equations for linear
326: $\sigma$ model were solved. Using the most ideal  conditions  for
327: DCC  formation, i.e. quenched initial condition and fast cooling,
328: we have evolved the fields for 10 fm/c. It was seen that in  both
329: the  scenario,  $\pi  \pi$ correlation increases with time till 7
330: fm/c, it  then  decreases.  Larger  correlation  is  obtained  in
331: scenario II than in scenario I. Evidence of increased correlation
332: is  also  obtained  from  the contour plot of pion field. $\pi_2$
333: component of the pion field shows domain like structure. Positive
334: and negative component of fields separates out at late time. Here
335: again, domain like structure is more  prominent  in  scenario  II
336: than  in  scenario  I.  Contour plot of the neutral to total pion
337: ratio suggest that though domain like structure is  seen  in  one
338: component of the $\pi$ field, there is no (large) physical domain
339: formation  in scenario I. On the contrary, large physical domains
340: are seen to be formed in scenario II after 7 fm/c  of  evolution.
341: We  have also studied the rapidity distribution of the neutral to
342: pion ratio in both the scenarios. Throughout the rapidity  range,
343: the  ratio  fluctuate  about  the isospin symmetric value of 1/3,
344: fluctuations being more  in  scenario  II  than  in  scenario  I.
345: However, it was also noted that fluctuations will be considerably
346: less  when integrated over time. Rapidity distribution of neutral
347: to total pion ratio may not be able to single out the DCC events.
348: 
349: \begin{references}
350: \bibitem[*]{byline}e-mail address:akc@veccal.ernet.in
351: \bibitem{ra93} K. Rajagopal and F. Wilczek, Nucl.Phys.B404,577,(1993).
352: K. Rajagopal and  F. Wilczek, Nucl.Phys.B399,395,(1993).
353: K. Rajagopal,  Nucl.Phys.A566,567c(1994).
354: \bibitem{bj93}J.  D.  Bjorken,  K. L. Kowalski and C. C. Taylor ,
355: SLAC-PUB-6109 (1993), K.L. Kowalski  K.  L.  and  Taylor  C.  C.,
356: CWRUTH-92-6 (1992),hep-ph/9211282.
357: J. D. Bjorken, Acta Phys. Polon. B28,2773 (1997).
358: \bibitem{an91}  A.  A.  Anselm  and  M. G. Ryskin , Phys. Lett. B
359: {\bf266},482 (1991),  A. A. Aneslm, Phys. Lett. B{\bf 217},  169(1988).
360: \bibitem{la80}C. M. G. Lattes, Y. Fujimoto and S. Hasegawa, Phys.
361: Reports, {\bf 154},247(1980).
362: \bibitem{rand}J.   Randrup,  Phys.  Rev  D55,1188  (1997).  J.
363: Randrup, Nucl. Phys. A616,531 (1997). J. Randrup and R. L. Thews,
364: Phys.  Rev.  D56,4392(1997).  J.  Schaffner-Bielich  and   Jorgen
365: Randrup,  Phys. Rev. C59,3329 (1999). J. Randrup, Heavy ion Phys.
366: 9,289 (1999).
367: \bibitem{gavi}  Gavin  S.,  Gocksch A. and Pisarski R. D., Phys.
368: Rev. Lett.{\bf72} 2143, (1994). Gavin  S.  and  Muller  B.  Phys.
369: Lett.B{\bf329} 486 (1994). S. Gavin, Nucl. Phys. A590,163c,1995
370: \bibitem{asak}M. Asakawa, Z. Huang and X.N. Wang, Nucl. Phys. A590,
371: 575c  (1995). M. Asakawa, Z. Huang and X.N. Wang, hep-ph/9408299,
372: Phys. Rev. Lett. {\bf 74} (1995) 3126. M.  Asakawa,  H.  Minakata
373: and   B.  Muller,  Phys.Rev.D58,  094011(1998).  M.  Asakawa,  H.
374: Minakata and B. Muller, Nucl. Phys. A638,443c (1998).
375: \bibitem{bo96} A. Bochkarev and J. Kapusta, Phys. Rev. D{\bf 54}
376: 4066 (1996).
377: \bibitem{gr97} C. Greiner and B. M\"{u}ller, Phys. rev. D {\bf 55}
378: 1026 (1997).
379: \bibitem{bi97}T.  S.  Biro  and C. Greiner, Phys. Rev. Lett. {\bf
380: 79}, 3138(1997).
381: \bibitem{zu00} Z. Zu and C. Greiner, hep-ph/9910562, Phys. Rev.
382: D {\bf 62} 036012 (2000).
383: \bibitem{ri98} D. H. Rischke, Phys. Rev. C, 2331 (1998)
384: \bibitem{ch99a}  A.  K.  Chaudhuri, nucl-th/9809018, Phys. Rev. D.
385: {\bf 59} 117503 (1999).
386: \bibitem{ch99b} A. K. Chaudhuri, hep-ph/9904269
387: \bibitem{ch99c} A. K. Chaudhuri, hep-ph/9908376
388: \bibitem{ch00} A. K. Chaudhuri, hep-ph/0007332
389: \end{references}
390: \eject
391: \begin{figure}
392: \centerline{\psfig{figure=corr.eps,height=10cm,width=10cm}}
393: \caption{Evolution  of the pion correlation fucntion  with  time,
394: in scenario I and II.}
395: \end{figure}
396: 
397: \begin{figure}
398: \centerline{\psfig{figure=field.eps,height=20cm,width=15cm}}
399: \caption{Contour  plot of the $\pi_2$ component of the pion field
400: at rapidity Y=0,  at  initial  time  $\tau_i$=1  fm/c  and  after
401: evolution for 7 fm/c.}
402: \end{figure}
403: 
404: \begin{figure}
405: \centerline{\psfig{figure=ratio.eps,height=20cm,width=15cm}}
406: \caption{Contour  plot  of  the  neutral  to total pion ratio, at
407: rapidity Y=0, at initial time $\tau_i$=1 fm/c and after evolution
408: for 7 fm/c.}
409: \end{figure}
410: 
411: \begin{figure}
412: \centerline{\psfig{figure=dndy.eps,height=10cm,width=10cm}}
413: %\vspace{-5cm}
414: \caption{Rapidity distribution of neutral to total pion ratio, in
415: scenario  I  and  II.  The  ratio at different time intervals are
416: shown.}
417: \end{figure}
418: \end{document}
419: 
420: 
421: 
422: 
423: 
424: \bibitem{bl92}  J. P. Blaizot ,A. Krzywicki , Phys. Rev.D{\bf46},
425: 246 (1992).
426: \bibitem{and81} I. V. Andreev, JETP Lett. 33, 367 (1981)
427: 
428: 
429: 
430: \bibitem{ka97} J. I. Kapusta and  A. P. Vischer,  Z. Phys. C75,507 (1997).
431: \bibitem{cho99}C. Chow and T. D. Cohen, nucl-th/9908013, Phys.
432: Rev. C60,054902 (1999).
433: \bibitem{cs99} L. P. Csernai, P. J. Ellis,  S.  Jeon  and  J.  I.
434: Kapusta, nucl-th/9908020,  Phys. Rev. C61,056901 (2000)
435: \bibitem{kr99}  A.  Krzywicki,  J,  Serreau, Phys. Lett. B448,257
436: (1999.)
437: \bibitem{me99}A. Z. Mekjian, Phys. Rev. C60,067902 (1999).
438: \bibitem{cho99}C. Chow and T. D. Cohen, nucl-th/9908013, Phys.
439: Rev. C60,054902 (1999).
440: \bibitem{cs99} L. P. Csernai, P. J. Ellis,  S.  Jeon  and  J.  I.
441: Kapusta, nucl-th/9908020,  Phys. Rev. C61,056901 (2000)
442: \bibitem{landau}L. D. Landau and E. M. Lifshitz, in Statistical
443: Physics, 3rd ed.Pergamon Press, 1993.
444: %\vspace{-5cm}
445: %\vspace{-5cm}
446: %\vspace{-5cm}
447: 
448: The  noise source
449: will continually heat the system while the friction  will  oppose
450: it. Equilibrium is achieved when the system is thermalised at the
451: temperature dictated by the heat bath.
452: