1: %&latex209
2: \documentstyle[12pt,epsfig]{article}
3: %\renewcommand{\baselinestretch}{1.65}
4:
5: \textwidth 155mm
6: \textheight 240mm
7: \columnsep 38pt
8: \topmargin -2.5cm
9: \oddsidemargin 5pt
10:
11: \def\bc{\begin{center}}
12: \def\ec{\end{center}}
13: \def\beq{\begin{equation}}
14: \def\eeq{\end{equation}}
15: \def\noi{\noindent}
16:
17: \def\al {\alpha}
18: \def\sig {\sigma}
19: \def\bet {\beta}
20: \def\ga {\gamma}
21: \def\Ga {\Gamma}
22: \def\Del {\Delta}
23: \def\kap {\kappa}
24: \def\lam {\lambda}
25:
26: \def\at#1{\left. \right|^{}_{#1}}
27: \def\hs#1{\hspace*{#1cm}}
28: \def\la{\langle}
29: \def\ra{\rangle}
30: \def\av#1{\langle {#1} \rangle}
31: \def\avr#1#2{\langle {#1} \rangle^{}_{#2}}
32: \def\etai{\sqrt{\eta_i}}
33: \def\F{{n^{}_F}}
34: \def\B{{n^{}_B}}
35: \def\pF{{p^{}_{{\rm t}F}}}
36: \def\pB{{p^{}_{{\rm t}B}}}
37: \def\ppB{{p^{2}_{{\rm t}B}}}
38: \def\Ceta{{\{\eta^{}_1,...,\eta^{}_M\}}}
39: \def\Aeta{{\eta^{}_1,...,\eta^{}_M}}
40: \def\Cn{{\{n^{}_1,...,n^{}_M\}}}
41: \def\An{{n^{}_1,...,\n^{}_M}}
42: \def\ski{\sum_{i}^{} k_i}
43: \def\sni{\sum_{i}^{} n_i}
44:
45:
46: \def\bfs{{\bf s}}
47: \def\olb{\overline{b}}
48: \def\olbe{\overline{\beta}}
49: \def\olx{\overline{x}}
50: \def\olr{\overline{r}}
51: \def\olN{\overline{N}}
52: \def\olG{\overline{G}}
53: \def\olP{\overline{P}}
54: \def\oln{\overline{n}}
55: \def\bN{\overline{N}}
56: \def\br{\overline{r}}
57: \def\bn{\overline{n}}
58:
59: \def\ol#1{\overline{#1}}
60: \def\beti{\ol{\eta}^{}_{i}}
61: \def\reta{\sum_{i}^{} \sqrt{\eta_i}}
62: \def\rbeta{\sum_{i}^{} \sqrt{\beti}}
63: \def\seta{\sum_{i}^{} \eta_i}
64:
65: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
66: % The journal macro:
67: \def\j#1#2#3#4{{#1} {\bf #2}, #4 (#3)}
68:
69: % The example of use:
70: %\bibitem{Kaidalov82}
71: %A.B.~Kaidalov and K.A.~Ter-Martirosyan,
72: %\j{\PL}{117B}{1982}{247}.
73:
74: % Some useful journal names:
75: \def\NP{Nucl. Phys.}
76: \def\PL{Phys. Lett.}
77: \def\PRL{Phys. Rev. Lett.}
78: \def\PR{Phys. Rev.}
79: \def\ZP{Z. Phys.}
80: \def\PRep{Phys. Rep.}
81: \def\JP{J. Phys.}
82: \def\EPJ{Eur. Phys. J.}
83: \def\IJMP{Int. J. of Mod. Phys.}
84: \def\NC{Nuovo Cimento}
85: \def\NIM{Nucl. Instrum. Methods}
86:
87: \def\JETP{JETP}
88: \def\JETPL{JETP Lett.}
89: \def\YF{Yad. Fiz.}
90: \def\SJNP{Sov. J. Nucl. Phys.}
91: \def\PAN{Phys. Atom. Nuclei}
92:
93: \def\ECHAYA{ECHAYA}
94:
95: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
96:
97: % title XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
98:
99: \title{
100: Cellular Approach to Long-Range\\
101: $p_t$ and Multiplicity Correlations\\
102: in the String Fusion Model
103: }
104: \author{
105: {\bf V.V. Vechernin and R.S. Kolevatov}\\
106: {\it High Energy Physics Dep., St. Petersburg State University,}\\
107: {\it 1 Ulianovskaya str., 198504 St.Petersburg, Russia}
108: }
109: \date{}
110:
111: \begin{document}
112:
113: \begin{titlepage}
114: \maketitle
115: \medskip
116:
117: % Abstract XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
118:
119: \begin{abstract}
120:
121: The long-range $p_t$ and multiplicity($n$) correlations
122: in high-energy nuclear collisions
123: are studied in the framework of a
124: simple cellular analog of the string fusion model.
125:
126: Two cases with local and global string
127: fusion is considered.
128: The $p_t$--$n$ and $n$--$n$ correlation functions and
129: correlation coefficients are calculated analytically
130: in some asymptotic cases
131: using suggested Gauss approximation.
132:
133: It's shown that at large string density
134: the $p_t$--$n$ and $n$--$n$ correlation coefficients
135: are connected and the scaling takes place.
136: The behavior of the correlations at small string density
137: is also studied.
138:
139: The asymptotic results are compared with results
140: of the numerical calculations in the framework of
141: proposed cellular approach.
142:
143: \end{abstract}
144: \vspace*{1cm}
145: \end{titlepage}
146:
147:
148: % 1 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
149:
150: \section{Introduction.}
151:
152: The colour strings approach \cite{Capella1,Kaidalov}
153: is widely applied for the description of the soft part of
154: the hadronic and nuclear interactions at high energies.
155:
156: In the frame work of this approach
157: the string fusion model was suggested in papers \cite{BP1}.
158: Later it was developed \cite{ABP1}-\cite{BPR} and applied for the
159: description of the long-range multiplicity and $p_t$ correlations
160: in relativistic nuclear collisions \cite{PRL94}-\cite{BP00}.
161:
162: In the paper \cite{vestn1} we have
163: formulated some simple cellular analog of the
164: string fusion model,
165: which enables explicit analytical calculations
166: of the correlation functions in some asymptotic cases and
167: can simplify calculations in the case of real nuclear
168: collisions.
169:
170: In that paper we have checked up the assumptions
171: of the cellular approach
172: and the validity of a suggested Gauss approximation
173: in the simplest (no fusion)
174: case when the explicit solution of the model can be found.
175:
176: In the present paper
177: in the framework of proposed cellular approach
178: we calculate
179: $p_t$--$n$ and $n$--$n$ correlation functions and
180: correlation coefficients
181: in two cases: with local and with global string
182: fusion.
183:
184: The calculations are done both
185: numerically
186: and in some asymptotic cases
187: analytically
188: using the Gauss approximation.
189: The results of both calculations are in a good agreement,
190: which proofs the validity of proposed Gauss approximation.
191:
192: At large string density $\eta \gg 1$
193: the connection between
194: the $p_t$--$n$ and $n$--$n$ correlation coefficients
195: are found in both cases: with local or global string
196: fusion.
197: At that
198: for the correlation coefficient
199: the scaling takes place.
200: It depends only on one combination $\mu_0/\sqrt{\eta}$
201: of the variables $\eta$ and $\mu_0$ ($\mu_0$ is
202: the mean multiplicity emitting by a single string).
203:
204:
205:
206: The paper organized as follows.
207: In the next section we recall
208: the formulation of the cellular approach
209: in the case with a {\it local} string fusion.
210:
211: In the section 3 we develop Gauss approximation
212: at large string density $\eta$ in the {\it local} fusion case.
213:
214: In the section 4
215: the $p_t$--$n$ and $n$--$n$ correlation coefficients
216: are calculated at large $\eta$ in the case with {\it local} string fusion
217: and connection between these two coefficients is found.
218:
219: In the section 5
220: the $p_t$--$n$ and $n$--$n$ correlation functions and
221: correlation coefficients
222: are calculated at large $\eta$ for "homogeneous" situation
223: (a constant mean density of strings)
224: in the case with {\it local} fusion.
225: It's shown that in this situation
226: the $p_t$--$n$ and $n$--$n$ correlation coefficients become equal
227: and the $\mu_0/\sqrt{\eta}$-scaling takes place.
228: The obtained results are compared with the results
229: of numerical calculations using formulas of section~2.
230:
231: In the section 6 we recall
232: the formulation of the cellular approach
233: in the case with a {\it global} string fusion.
234: The exact closed formulas for
235: the $p_t$--$n$ and $n$--$n$ correlation functions
236: in this case are obtained.
237:
238: In the section 7 we develop Gauss approximation
239: at large string density $\eta$ for the {\it global} fusion case.
240: The $p_t$--$n$ and $n$--$n$ correlation functions and coefficients
241: for this case are calculated and compared with the results
242: of numerical calculations using formulas of section~6.
243:
244: The behavior of the correlations at small string density
245: is studied in Appendix A.
246:
247:
248:
249: % 2 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
250:
251: \section{Cellular approach to the local string fusion.}
252:
253: Let us recall the formulas obtained for this case in \cite{vestn1}.
254: We consider the collision of nuclei in two stage scenario
255: when at first stage the colour strings are formed,
256: and at the second stage these strings,
257: or some other (higher colour) strings
258: formed due to fusion of primary strings, are decaying, emitting
259: observed particles.
260:
261: In principle, one can consider two types of fusion.
262: The case with a local fusion corresponds to the model,
263: where colour fields
264: are summing up only locally and the global fusion case corresponds
265: to the model, where colour fields are summing up globally -
266: all over the cluster area - into one average colour field,
267: the last case corresponds to the summing of the sources colour charges.
268: (In section 5 of \cite{vestn1}
269: we have referred to these cases as A) and B) correspondingly.)
270:
271: In the transverse plane depending on the impact
272: parameter $b$ we have some interaction area $S(b)$.
273: Let us split this area on the cells
274: of order of the transverse string size.
275: Then we'll have $M=S(b)/\sigma_0$ cells,
276: where $\sigma_0=\pi r_0^2$ is the transverse area of the string
277: and $r_0\approx 0.2 fm$ is the string radius.
278:
279: In the case with a local fusion the assumption of the model is
280: that if the number of strings
281: belonging to the $i$-th cell is $\eta_i$, then they form
282: higher colour string, which emits in average $\mu _0\sqrt{\eta _i}$ particles
283: with mean $p_t^2$ equal to $p^2\sqrt{\eta _i}$,
284: compared with $\mu _0$ particles with $\langle p_t^2 \rangle =p^2$
285: emitting by a single string.
286:
287: If we denote by $n_i$ and $\bn_i$ - the number and the average
288: number of particles emitted by the higher string from $i$-th cell
289: in a given rapidity interval, then
290: \beq
291: \bn_i=\mu _0\sqrt{\eta _i}
292: \label{fus1}
293: \eeq
294:
295: From event to event the number of strings $\eta_i$ in $i$-th cell
296: will fluctuate around some
297: average value - $\overline{\eta }_i^{}$.
298: Clear that in the case of real nuclear collisions
299: these average values $\overline{\eta }_i^{}$ will be different
300: for different cells. They will depend on the position (${\bf s}$)
301: of the $i$-th cell in the interaction area
302: (${\bf s}$ is two dimensional vector in transverse plane).
303: To get the physical answer we have to sum the contributions
304: from different cells, which corresponds to integration over
305: ${\bf s}$ in transverse plane.
306:
307: The average local density of primary strings $\overline{\eta }_i^{}$
308: in the point ${\bf s}$ of transverse plane is uniquely
309: determined by the distributions of nuclear densities and
310: the value of the impact parameter - $b$.
311: They can be calculated, for example, in Glauber approximation.
312: We'll do this later in a separate paper. In present paper
313: we consider that all $\overline{\eta }_i^{}$ are already
314: fixed from these considerations at given value
315: of the impact parameter - $b$.
316:
317: Let us introduce some quantities, which will play important
318: role in our consideration:
319: $$
320: N=\sum_{i=1}^M \eta_i , \hs{1} \bN=\sum_{i=1}^M \beti
321: $$
322: \beq
323: r=\sum_{i=1}^M \sqrt{\eta_i} , \hs{1} \br=\sum_{i=1}^M \sqrt{\beti}
324: \label{not}
325: \eeq
326: then clear that $N$ is the number of strings in the given event
327: and $\bN$ is the mean number of strings for this type of events
328: (at the fixed impact parameter $b$).
329:
330: To go to long-range rapidity correlations we have to consider
331: two rapidity windows $F$ (forward) and $B$ (backward).
332: Each event corresponds to a certain configuration $\Ceta$ of
333: strings and certain numbers of charged particles $\Cn$ emitted by
334: these strings in the forward rapidity window. Then the total
335: number of particles produced in the forward rapidity window
336: will be equal to $n_F$:
337: \beq
338: n_F=\sum_{i=1}^{M} n_i
339: \label{nF}
340: \eeq
341: The probability to detect
342: $n_F$ particles in the forward rapidity window
343: for a given configuration $\Ceta$ of strings
344: is equal to
345: \beq
346: P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F)=
347: \sum_{\{n^{}_1,...,n^{}_M\}} \delta_{\F,\sum_i n_i} \prod_{i=1}^M
348: p^{}_{\eta_i} (n_i)
349: \label{PnF}
350: \eeq
351: where $p^{}_{\eta_i} (n_i)$ is the probability of the emission
352: of $n_i$ particles by the string $\eta_i$ in the forward
353: rapidity window. By our assumption (\ref{fus1})
354: \beq
355: \bn_i\equiv\sum_{n_i=0}^{\infty} n_i p^{}_{\eta_i} (n_i) =
356: \mu_0 \sqrt{\eta_i}
357: \label{fusion}
358: \eeq
359:
360: If we denote else by $W(\Aeta)$ the probability of realization of the
361: string configuration $\Ceta$ in the given event,
362: then the average value of some quantity $O$ under condition of
363: the production of $n_F$ particles in the forward window
364: will be equal to
365: \beq
366: \avr{O}{\F}
367: =\frac{\sum_{\{\eta^{}_1,...,\eta^{}_M\}} \avr{O} {{\{\eta^{}_1,...,\eta^{}_M\}},\F} W(\Aeta) P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F)}
368: {\sum_{\{\eta^{}_1,...,\eta^{}_M\}} W(\Aeta) P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F)}
369: \label{avO}
370: \eeq
371: One has to omit in this $M$-fold sums one term, when all $\eta_i=0$,
372: which corresponds to the absence of inelastic interaction between
373: the nucleons of the colliding nuclei (see details in Appendix A).
374:
375: If the $O$ in the number of particles produced in the backward
376: rapidity window $n_B$ in the given event, then
377: we have to use for $\avr{\B}{\F}$ correlations:
378: \beq
379: \avr{\B} {{\{\eta^{}_1,...,\eta^{}_M\}},\F} =
380: \mu_0 \sum_{i=1}^{M} \sqrt{\eta_i}=\mu_0 r
381: \label{Onn}
382: \eeq
383: If the $O$ in the mean squared transverse momentum
384: of particles produced in the backward
385: rapidity window $\ppB$ in the given event, then
386: we have to use for $\avr{\ppB}{\F}$ correlations:
387: \beq
388: \avr{\ppB} {{\{\eta^{}_1,...,\eta^{}_M\}},\F}
389: = \sum_{i=1}^{M} \frac{\sqrt{\eta_i}}
390: {\sum_{i=1}^{M} \sqrt{\eta_i}} p^2 \sqrt{\eta_i}
391: = p^2 \frac{\sum_{i=1}^{M} \eta_i}{\sum_{i=1}^{M} \sqrt{\eta_i}}
392: =p^2\frac{N}{r}
393: \label{Optn}
394: \eeq
395:
396: Later we'll assume that numbers of primary strings in each cell $\eta_i$
397: fluctuate independently around some average quantities
398: $\overline{\eta }_i^{}$ uniquely
399: determined by the distributions of nuclear densities and
400: the value of the impact parameter - $b$ (see above), then
401: \beq
402: W(\Aeta) = \prod_{i=1}^M w (\eta_i), \hs{1}
403: \sum_{i=1}^M \eta_i w (\eta_i) = \beti
404: \label{WCeta}
405: \eeq
406:
407: For clearness we'll sometimes address to a simple
408: "homogeneous" case, when all $\beti$
409: (but not the $\eta_i$, which fluctuate from event to event!)
410: is equal each other in the interaction area $\beti=\eta$
411: (a constant mean density of strings).
412: The parameter $\eta$ coincides in this case with
413: the parameter $\eta$ used in the papers \cite{BPR,BPep00,BP00}
414: and has the meaning of the mean number of strings
415: per area of one string ($\eta=({\rm mean\ string\ density})\times\sigma_0$).
416: In general case the parameters $\beti$ have the same meaning,
417: but with mean string density depending on the point $\bfs$
418: in the transverse interaction plane\\
419: ($\beti=({\rm mean\ string\ density\ in\ the\ point}\ \bfs)\times\sigma_0$).
420:
421: As it was shown in \cite{vestn1}
422: if we assume else the Poissonian form of $p^{}_{\eta_i} (n_i)$
423: ($\rho^{}_{a}(x)$ is the Poisson distribution with $\olx=a$):
424: \beq
425: p^{}_{\eta_i} (n_i) = \rho^{}_{\mu_0\etai}(n_i)
426: \equiv e^{-\mu_0\etai}\frac{(\mu_0\etai)^{n_i}}{n_i!}
427: \label{Poisson}
428: \eeq
429: then we find the Poissonian distribution for
430: \beq
431: P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F) = \rho^{}_{\mu_0\reta}(\F)
432: \label{Poisso}
433: \eeq
434: with $\avr{\F} {{\{\eta^{}_1,...,\eta^{}_M\}}} = \mu_0 \reta = \mu_0 r = \avr{\F} {r}$
435: and $\sigma_{rF}^2=\avr{\F} {r}=\mu_0 r$, which we can
436: replace at large $\mu_0 \reta$ by Gauss distribution:
437: \beq
438: P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F)
439: = \frac{1}{\sqrt{2\pi}\sigma_{rF}}
440: e^{-\frac{(\F-\avr{\F} {r})^2}{2\sigma_{rF}^2}}
441: \label{pgausA}
442: \eeq
443:
444: It was also shown in the section 6 of \cite{vestn1} that
445: if we assume the Binomial form of $p^{}_{\eta_i} (n_i)$
446: then we find the Binomial distribution for $P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F)$
447: with $\avr{\F} {{\{\eta^{}_1,...,\eta^{}_M\}}} = \mu_0 \reta = \mu_0 r = \avr{\F} {r}$
448: and $\sigma_{rF}^2=\avr{\F} {r} (1-\lambda)=\mu_0 r (1-\lambda)$,
449: where $\lambda\to 0$ corresponds to the Poisson limit and
450: $\lambda\to 1$ corresponds to the case when each higher string emits
451: fixed number of particles - $n_i=\oln_i=\mu_0 \sqrt{\eta_i}$
452: in each event.
453:
454: Moreover appealing to the central limit theorem of the probability theory
455: one can state that at large $M$
456: we'll have formula (\ref{pgausA}) for any type of $p^{}_{\eta_i} (n_i)$.
457:
458: \section {Gauss approximation
459: at large string density for the {\it local} fusion.}
460:
461: Now we'll evaluate $\sum_{\{\eta^{}_1,...,\eta^{}_M\}}$ at the condition that all $\beti\gg 1$.
462: At this condition we can use Gauss approximation for each $w (\eta_i)$
463: \beq
464: w (\eta_i)
465: = \frac{1}{\sqrt{2\pi}\sigma_{\eta_i}}
466: e^{-\frac{(\eta_i-\beti)^2}{2\sigma_{\eta_i}^2}}
467: \label{wgausA}
468: \eeq
469: with $\sigma_{\eta_i}^2=\beti (1-\lambda_\eta)$,
470: where again $\lambda_\eta\to 0$ corresponds to the Poisson limit and
471: $\lambda_\eta\to 1$ corresponds to the case with a fixed number of strings
472: $N=\bN=\sum_{i=1}^M \beti$ (see the section 6 of \cite{vestn1}).
473:
474: As in \cite{vestn1} we have then
475: \beq
476: \avr{\B} {\F}
477: =\mu_0 \frac{\int d\eta^{}_1...d\eta^{}_M r \frac{1}{\sqrt{r}}e^{-\varphi(\eta_i,\F)}}
478: {\int d\eta^{}_1...d\eta^{}_M \frac{1}{\sqrt{r}}e^{-\varphi(\eta_i,\F)}}
479: \label{BF1A}
480: \eeq
481: and
482: \beq
483: \avr{\ppB} {\F}
484: =p^2 \frac{\int d\eta^{}_1...d\eta^{}_M \frac{N}{r} \frac{1}{\sqrt{r}}e^{-\varphi(\eta_i,\F)}}
485: {\int d\eta^{}_1...d\eta^{}_M \frac{1}{\sqrt{r}}e^{-\varphi(\eta_i,\F)}}
486: \label{pBF1A}
487: \eeq
488: Recall that $N=\sum_{i=1}^M \eta_i$ and $r=\sum_{i=1}^M \sqrt{\eta_i}$. Here
489: \beq
490: \varphi(\eta_i,\F)=\sum_{i=1}^M\frac{(\eta_i-\beti)^2}{2\beti(1-\lambda_\eta)}
491: +\frac{(\F-\mu_0 r)^2}{2\mu_0 r(1-\lambda)}
492: \label{phiA}
493: \eeq
494:
495: Further we take out factors before exponent
496: at the point, where $\varphi$ is minimal, after that
497: the rest integrals in the numerator and in the denominator are being reduced, and we find
498: \beq
499: \avr{\B} {\F} =\mu_0 r^*
500: \label{BF2A}
501: \eeq
502: and
503: \beq
504: \avr{\ppB} {\F} =p^2 \frac{N^*}{r^*}
505: \label{pBF2A}
506: \eeq
507: The $N^*$ and $r^*$ are the values of $N$ and $r$ in the point
508: $\{\eta^{*}_1,...,\eta^{*}_M\}$, where $\varphi(\eta_i,\F)$ is minimal:
509: \beq
510: \frac{\partial\varphi(\eta_i,\F)}{\partial\eta_i}=0
511: \label{dphiA}
512: \eeq
513: This leads to the system of equations:
514: \beq
515: \frac{\eta_i^*}{\beti}-1= \frac{\mu_0\kappa}{4\sqrt{\eta_i^*}}
516: \left( \frac{n_F^2}{\mu_0^2 r^{*2}}-1 \right)
517: \label{etistarA}
518: \eeq
519: Recall that $r^*=\sum_{i=1}^M \sqrt{\eta_i^*}$ and
520: \beq
521: \kappa=\frac{1-\lambda_\eta}{1-\lambda}
522: \label{kappa}
523: \eeq
524: For the meaning of $\kappa$ see
525: the end of the previous section and
526: the section 6 of \cite{vestn1},
527: for both Poissonian distributions $\kappa=1$
528: and $\kappa$ is the relative width of the
529: $p(n_i)$ and $w(\eta_i)$ distributions in other cases.
530: The (\ref{etistarA}) defines $\eta_i^*$ as function of $\F$.
531:
532: Introducing short notations
533: \beq
534: z_i=\sqrt{\frac{\eta_i^*}{\beti}}, \hs{1}
535: f=\frac{\F}{\mu_0 \br}=\frac{\F}{\av{\F}}, \hs{1}
536: a_i=\frac{\mu_0\kappa}{4\sqrt{\beti}}
537: \label{notA}
538: \eeq
539: we can rewrite (\ref{etistarA}) as
540: \beq
541: z_i^3-z_i^{}=a_i \left( f^2 \frac{\br^2}{r^{*2}}-1 \right)
542: \label{ziA}
543: \eeq
544: where $\br=\sum_{i=1}^M \sqrt{\beti}$, $r^*=\sum_{i=1}^M z_i\sqrt{\beti}$
545: and $N^*=\sum_{i=1}^M z_i^2 \beti$.
546: The (\ref{ziA}) defines $z_i$ as function of $f$: \
547: $z_i=z_i(f)$. Then we can calculate
548: $\avr{\B} {\F}$ and $\avr{\ppB} {\F}$ as a function of $\F$
549: using (\ref{BF2A}) and (\ref{pBF2A}).
550:
551: %XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
552:
553: \section {$p_t$--$n$ and $n$--$n$ correlation coefficients
554: at a large string density for the {\it local} fusion.}
555:
556: The correlation coefficients are defined in the same way as in
557: the section 4 of \cite{vestn1}:
558: \beq
559: b \equiv \frac{d\avr{\B} {\F}}{d\F} \at {\F=\av{\F}}
560: \label{bA}
561: \eeq
562: and
563: \beq
564: \beta \equiv \frac{d\avr{\ppB} {\F}}{d\F} \at {\F=\av{\F}}
565: \label{betaA}
566: \eeq
567: or for "relative" quantities
568: \beq
569: \ol{b} \equiv\frac{d\avr{\B} {\F}/\av{\B}}{d\F/\av{\F}} \at {\F=\av{\F}}
570: \label{bbA}
571: \eeq
572: and
573: \beq
574: \ol{\beta} \equiv\frac{d\avr{\ppB}{\F}/\av{\ppB}}{d\F/\av{\F}} \at {\F=\av{\F}}
575: \label{bbetaA}
576: \eeq
577: (Note the same definition of
578: $pt$--$n$ correlation coefficient $\ol{\beta}$
579: in \cite{Heiselberg} (see formula (44) in \cite{Heiselberg}),
580: see also remark in Appendix B).
581:
582: In short notation using (\ref{BF2A}) and (\ref{pBF2A}) we have:
583: \beq
584: \ol{b} =\frac{1}{\br}\frac{dr^*}{df} \at {f=1}
585: \label{bb1A}
586: \eeq
587: and
588: \beq
589: \ol{\beta} =\frac{\br}{\bN}\frac{d(N^*/r^*)}{df} \at {f=1}
590: \label{bbeta1A}
591: \eeq
592:
593: We can't solve the equations (\ref{ziA}) for to find $z_i=z_i(f)$ explicitly,
594: but to calculate the correlation coefficients we need to know
595: only $z'_i(1)=\frac{dz_i(f)}{df}\at {f=1}$, which can be done explicitly.
596:
597: We see that at $f=1$ (\ref{ziA}) has the obvious solution:
598: \beq
599: f=1, \hs{1} z_i=1, \hs{1} \eta^*_i=\beti, \hs{1} r^*=\br, \hs{1} N^*=\bN
600: \label{solziA}
601: \eeq
602: We need to calculate $z'_i(f)$ only at $f=1$. Differentiating
603: (\ref{ziA}) on $f$ and using then again (\ref{solziA}) we find
604: \beq
605: z'_i(1)=a_i \frac{4\br}{4\br+\mu_0\kappa M}
606: \label{solzi1A}
607: \eeq
608: with $a_i=\mu_0\kappa/(4\sqrt{\beti})$. Then
609: \beq
610: \ol{b} =\frac{1}{\br}\frac{dr^*}{df} \at {f=1}=\frac{1}{\br}\sum_{i=1}^M z'_i(1)\sqrt{\beti}
611: =\frac{\mu_0\kappa}{\mu_0\kappa+4\br/M}
612: \label{bb2A}
613: \eeq
614: and
615: \beq
616: \ol{\beta} =\frac{1}{\bN} \frac{dN^*}{df} - \frac{1}{\br} \frac{dr^*}{df} \at {f=1}
617: =\frac{1}{\bN} \frac{dN^*}{df} \at {f=1} -\ol{b}
618: \label{bbeta2A}
619: \eeq
620: Using
621: \beq
622: \frac{dN^*}{df} \at {f=1}=2\sum_{i=1}^M z'_i(1)\beti
623: =\frac{2\mu_0\kappa\br^2}{\mu_0\kappa M+4\br}
624: \label{dNdfA}
625: \eeq
626: we have
627: \beq
628: \ol{\beta}
629: = \left( \frac{2\br^2}{\bN M}-1 \right)
630: \frac{\mu_0\kappa}{\mu_0\kappa+4\br/M}
631: = \left( \frac{2\br^2}{\bN M}-1 \right) \ol{b}
632: \label{bbeta3A}
633: \eeq
634: We see the connection between $pt$--$n$ and $n$--$n$ correlation coefficients.
635: Note that due to obvious inequality:
636: \beq
637: \left( \sum_{i=1}^M \sqrt{\beti} \right)^2 \leq M \sum_{i=1}^M \beti
638: \label{ineqv}
639: \eeq
640: we have $\br^2 \leq M \bN$ and hence always $\ol{\beta}\leq\ol{b}$.
641:
642: Clear that at equal $\beti\equiv\eta$ we have
643: $\br=M\sqrt{\eta}$, $\bN=M\eta$, $\br^2 = \bN M$ and
644: \beq
645: \ol{\beta}=\ol{b}
646: =\frac{\mu_0\kappa}{\mu_0\kappa+4\sqrt{\eta}}
647: \label{bbb3A}
648: \eeq
649:
650: From (\ref{bb2A}) we see that $n$--$n$ correlation coefficient
651: is always positive.
652: Can $pt$--$n$ correlation coefficient be negative?
653: Let us consider {\it nonhomogeneous} situation when
654: $\beti=\eta_+$ at $i=1,...,M_1$,
655: $\beti=\eta_-$ at $i=M_1+1,...,M$, \ $M_1\sim M$ and $\eta_+ \gg \eta_- \gg 1$.
656: Then
657: $\br=M_1\sqrt{\eta_+}+(M-M_1)\sqrt{\eta_-}\approx M_1\sqrt{\eta_+}$, \
658: $\bN=M_1\eta_+ +(M-M_1)\eta_- \approx M_1 \eta_+$ and we have
659: \beq
660: \ol{\beta}
661: \approx \left( 2\frac{M_1}{M}-1 \right) \ol{b}
662: \label{bbeta4A}
663: \eeq
664: We see that at $M_1<M/2$ we can have $\ol{\beta} < 0 $.
665:
666:
667: %XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
668:
669: \section{The $\mu_0/\eta^{1/2}$-scaling at large string density.}
670:
671: Let us consider for clearness
672: {\it homogeneous} case, when all $\beti$
673: is equal each other in the interaction area $\beti=\eta$
674: (a constant mean density of strings).
675: In this case we can explicitly calculate
676: at large string density $\eta$
677: not only the $p_t$--$n$ and $n$--$n$ correlation coefficients,
678: but also the corresponding correlation functions
679: for the version with a local string fusion.
680:
681: We have seen in the end of previous section that in this case
682: the coefficients for $n$--$n$ and $pt$--$n$ correlations
683: defined as (\ref{bbA}) and (\ref{bbetaA})
684: are coincide:
685: \beq
686: \ol{\beta}=\ol{b}
687: =\frac{\mu_0\kappa}{\mu_0\kappa+4\sqrt{\eta}}=\frac{a}{a+1}
688: \label{bb3A}
689: \eeq
690: where $a=\mu_0\kappa/(4\sqrt{\eta})$.
691:
692: In this homogenous case $\beti\equiv\eta$ we can also calculate
693: the correlation functions $\avr{\ppB}{\F}$ and $\avr{\B}{\F}$ at any $\F$.
694: Due to symmetry, the system of equations (\ref{ziA})
695: have symmetrical solution $z_i=z$ and can be reduced to one equation
696: \beq
697: z^3-z=a \left( \frac{f^2}{z^2}-1 \right)
698: \label{zA}
699: \eeq
700: because $\br= M \sqrt{\eta}$, $r^*=z M \sqrt{\eta}$ and $N^*=z^2 M \eta$ with
701: \beq
702: f=\frac{\F}{\mu_0 M \sqrt{\eta}}=\frac{\F}{\av{\F}}, \hs{1}
703: a=\frac{\mu_0\kappa}{4\sqrt{\eta}}
704: \label{not1A}
705: \eeq
706: The (\ref{zA}) defines the function $z=z(f)$ and then
707: using (\ref{BF2A}) and (\ref{pBF2A}) we can calculate
708: \beq
709: \avr{\B}{\F} =\mu_0 r^*= \mu_0 M \sqrt{\eta} z(f)=\av{\B} z(f)
710: \label{BF3A}
711: \eeq
712: and
713: \beq
714: \avr{\ppB}{\F} =p^2 \frac{N^*}{r^*}=p^2 \sqrt{\eta} z(f)=\av{\ppB} z(f)
715: \label{pBF3A}
716: \eeq
717: So we have at any $\F=\av{\F} f $:
718: \beq
719: \frac{\avr{\B}{\F}}{\av{\B}}
720: =\frac{\avr{\ppB}{\F}}{\av{\ppB}}= z(f)
721: \label{pnA}
722: \eeq
723:
724: From (\ref{bb3A}) and (\ref{zA}) we see that
725: in this homogenous case
726: at large string density $\eta$
727: there is an remarkable {\it scaling}.
728: The $pt$--$n$ and $n$--$n$ correlation coefficients
729: and correlation functions depend only on
730: one combination $a=\mu_0\kappa/(4\sqrt{\eta})$
731: of parameters.
732:
733: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
734:
735: We present the correlation function $z(f)$ (\ref{pnA})
736: in Figs.~\ref{zf1},\ref{zf2}
737: and the correlation coefficient $\ol{\beta}=\ol{b}$ (\ref{bb3A})
738: as function of $\eta$ in Figs.~\ref{b1}-\ref{b4}
739: (the solid lines).
740: We present also in Figs.~\ref{b1}-\ref{b4}
741: the results of our direct numerical MC calculations
742: of the $pt$--$n$ and $n$--$n$ correlation coefficients
743: in the local fusion case based on formulas (\ref{avO}-\ref{Optn})
744: (empty and filled points correspondingly).
745:
746: We see that in the case with local fusion
747: at small string density we have large
748: $n$--$n$ correlations (the same as in the case without
749: string fusion \cite{vestn1}) and no $p_t$--$n$ correlations.
750: (The analysis at very small values of $\eta\leq 1/M$
751: see in Appendix A.)
752:
753: At large string density in the homogeneous case
754: the $p_t$--$n$ and $n$--$n$ correlation coefficients become equal
755: and the $\mu_0/\sqrt{\eta}$-scaling takes place.
756: We see also that in this limit our
757: Gauss asymptotic is in a good agreement with
758: results of the numerical calculations and $M$-independence
759: takes place.
760:
761: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
762:
763:
764: \section{The {\it global} fusion at large string density.
765: Exact solution.}
766:
767: In this case at first stage
768: we also have $M=S(b)/\sigma_0$ cells (like in the case with
769: local fusion) with $\eta_i$, $i=1,...,M$
770: fluctuated around $\beti$. Then (unlike the local fusion case)
771: we have to find
772: average $\eta_c=\frac{1}{M}\seta=\frac{N}{M}$ for given event,
773: and then to generate particles from one cluster
774: with average multiplicity equal to $\mu_c \sqrt{\eta_c}=$
775: $\mu_0 M \sqrt{\eta_c}$ $=\mu_0 M \sqrt{N/M}$ $=\mu_0\sqrt{MN}$.
776: The general formulae for this case was obtained in the section 5 of
777: \cite{vestn1}:
778: \beq
779: \avr{O} {\F}
780: =\frac{\sum_{\{\eta^{}_1,...,\eta^{}_M\}} \avr{O} {{\{\eta^{}_1,...,\eta^{}_M\}},\F} W(\Aeta)
781: p^{}_{\mu_c\sqrt{\frac{1}{M}\seta}} (\F)}
782: {\sum_{\{\eta^{}_1,...,\eta^{}_M\}} W(\Aeta)
783: p^{}_{\mu_c\sqrt{\frac{1}{M}\seta}} (\F)}
784: \label{avOcl}
785: \eeq
786: where $\mu_c = \mu_0 M $ with $M=S(b)/\sigma_0$.
787: The assumption $\beti>>1$ is essential, as only in this situation
788: we can consider that the transverse area
789: of the cluster $\Delta S$ is equal to all interaction area $S(b)$ ($b$-impact parameter).
790:
791: The $\avr{O} {{\{\eta^{}_1,...,\eta^{}_M\}},\F}$ is the rates of the backward production from
792: configuration ${\{\eta^{}_1,...,\eta^{}_M\}}$. We have to use for $\avr{\B}{\F}$ correlations:
793: \beq
794: \avr{\B} {{\{\eta^{}_1,...,\eta^{}_M\}},\F} =\mu_c \sqrt{\eta_c}= \mu_0 M \sqrt{\frac{1}{M}\seta}
795: \label{Onncl}
796: \eeq
797: and for $\avr{\ppB}{\F}$ correlations:
798: \beq
799: \avr{\ppB} {{\{\eta^{}_1,...,\eta^{}_M\}},\F} = p^2 \sqrt{\eta_c}=p^2 \sqrt{\frac{1}{M}\seta}
800: \label{Optncl}
801: \eeq
802: We see that the difference with the case of local fusion
803: consists in replace
804: $\frac{1}{M}\reta\to\sqrt{\frac{1}{M}\seta}$.
805: As a consequence calculations
806: in the case of global string fusion are much more simple,
807: as we can reduce all sums $\sum_{\{\eta^{}_1,...,\eta^{}_M\}}$
808: to one sum $\sum_N$, as in the no fusion case
809: (see section 3 in \cite{vestn1}).
810: So in the global fusion case we can write simple formulas:
811: \beq
812: \avr{\B}{\F}
813: =\frac{\mu_0 \sqrt{M}
814: \sum_N \sqrt{N} W(N) p^{}_{\mu_0\sqrt{M} \sqrt{N}}(\F)}
815: {\sum_N W(N) p^{}_{\mu_0\sqrt{M} \sqrt{N}}(\F)}
816: \label{BFcl}
817: \eeq
818: and
819: \beq
820: \avr{\ppB}{\F}
821: =\frac{p^2}{\sqrt{M}}
822: \frac{\sum_N \sqrt{N} W(N) p^{}_{\mu_0\sqrt{M} \sqrt{N}}(\F)}
823: {\sum_N W(N) p^{}_{\mu_0\sqrt{M} \sqrt{N}}(\F)}
824: \label{pBFcl}
825: \eeq
826: where $W(N)$ is given by the formula
827: \beq
828: W(N) = \sum_{{\{\eta^{}_1,...,\eta^{}_M\}}} \delta_{N,\seta} \prod_i w (\eta_i)
829: \label{WN}
830: \eeq
831:
832: We see that in the case of global fusion
833: (one cluster at large $\beti$ with area
834: $\Delta S$ being equal to all interaction area $S(b)$)
835: $n$--$n$ and $pt$--$n$ correlations are connected
836: \beq
837: \avr{\ppB}{\F}=\frac{p^2}{\mu_0 M} \avr{\B}{\F}
838: \hs{1} {\rm or} \hs{1}
839: \frac{\avr{\B}{\F}}{\av{\B}}
840: =\frac{\avr{\ppB}{\F}}{\av{\ppB}}
841: \label{pnB}
842: \eeq
843: Note that unlike the local fusion case
844: in this case we find this result without any assumptions
845: on the properties of $p(\F)$ and $w(\eta_i)$ distributions
846: and for arbitrary (even nonequal) $\beti$.
847:
848: Clear that in this case
849: the results can depend only on mean number of strings $\bN$
850: and on combination $\mu_M$:
851: \begin{equation}\label{muM}
852: \bN=\sum_i \beti \hs{2} \mu_M\equiv\mu_0 \sqrt{M}
853: \end{equation}
854:
855: Below we'll calculate numerically the correlation functions on
856: formulas (\ref{BFcl}) and (\ref{pBFcl}), but at first we would like
857: to find explicit formulas for global fusion case in Gauss approximation.
858: We'll see that results really depend only
859: on one combination of the variables
860: (\ref{muM}), namely on
861: \begin{equation}\label{scal}
862: \frac{\mu_M}{\sqrt{\bN}}=\frac{\mu_0}{4\sqrt{\bN/M}}
863: \end{equation}
864: and the $scaling$ takes place
865: as in the case with local fusion.
866:
867: %XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
868:
869: \section{Gauss approximation for {\it global} fusion at large
870: string density.}
871:
872: Acting as in the no fusion case in section 4 of \cite{vestn1}
873: (see also calculations in the Gauss approximation
874: in the section 3 of the present paper)
875: we find
876: \beq
877: \avr{\B}{\F} =\mu_M \sqrt{N^*}=\mu_0 \sqrt{M} \sqrt{N^*}
878: \label{BF2cl}
879: \eeq
880: or keeping in mind (\ref{pnB})
881: \beq
882: \frac{\avr{\B}{\F}}{\av{\B}}
883: =\frac{\avr{\ppB}{\F}}{\av{\ppB}}=\sqrt{\frac{N^*}{\bN}}
884: \label{BF3cl}
885: \eeq
886: The $N^*$ is the value of $N$ at which the function
887: \beq
888: \varphi(N,\F)=\frac{(N-\bN)^2}{2\bN(1-\lambda_\eta)}+
889: \frac{(\F-\mu_M \sqrt{N})^2}{2\mu_M \sqrt{N} (1-\lambda)}
890: \label{phicl}
891: \eeq
892: gains its minimum. In short notations
893: \beq
894: z\equiv\sqrt{\frac{N^*}{\bN}}, \hs{1}
895: f=\frac{\F}{\mu_M \sqrt{\bN}}=\frac{\F}{\av{\F}}, \hs{1}
896: a=\frac{\mu_M\kappa}{4\sqrt{\bN}}
897: =\frac{\mu_0\kappa}{4\sqrt{\bN/M}}
898: \label{notcl}
899: \eeq
900: we find the equation
901: \beq
902: z^3-z=a \left( \frac{f^2}{z^2}-1 \right)
903: \label{zcl}
904: \eeq
905: which defines the function $z=z(f)$ and then
906: using (\ref{BF3cl}) we can calculate correlation functions
907: \beq
908: \frac{\avr{\B}{\F}}{\av{\B}}
909: =\frac{\avr{\ppB}{\F}}{\av{\ppB}}=z\left(\frac{\F}{\av{\F}}\right)=z(f)
910: \label{BF4cl}
911: \eeq
912: and correlation coefficients for the global fusion case
913: \beq
914: \ol{\beta} =\ol{b}=z'(1)=\frac{a}{a+1}
915: =\frac{\mu_M\kappa}{\mu_M\kappa+4\sqrt{\bN}}
916: =\frac{\mu_0\kappa}{\mu_0\kappa+4\sqrt{\bN/M}}
917: \label{b4cl}
918: \eeq
919: We see again that in Gauss approximation
920: there is the same remarkable $scaling$.
921: The $pt$--$n$ and $n$--$n$ correlations depend only on
922: one combination of parameters: $a=\frac{\mu_0\kappa}{4\sqrt{\bN/M}}$.
923: Note that unlike the local fusion case
924: in this case we find this result for arbitrary (even not equal) $\beti$.
925:
926: In the homogeneous situation all $\beti=\eta$ and we have
927: \beq
928: \bN=M\eta, \hs{1}
929: a=\frac{\mu_0\kappa}{4\sqrt{\eta}}
930: \label{not1cl}
931: \eeq
932: and
933: \beq
934: \ol{\beta} =\ol{b}=
935: \frac{\mu_0\kappa}{\mu_0\kappa+4\sqrt{\eta}}=\frac{a}{a+1}
936: \label{b5cl}
937: \eeq
938:
939: We see that in the $homogeneous$ situation in Gauss approximation
940: the results with $local$ and $global$ fusion coincide
941: (as we have expected in section 5 of \cite{vestn1}).
942: We have the same equations (\ref{zA}) and (\ref{zcl})
943: with the same value of parameter $a$ (\ref{not1A}) and (\ref{not1cl}).
944: Note that in the no fusion case we have had {\it very different} equation
945: for $z(f)$ (see section 4 of \cite{vestn1}).
946:
947: Unlike the local fusion case in the global fusion case
948: we can control the validity of Gauss approximation
949: making calculations on exact formulas (\ref{BFcl}) and (\ref{pBFcl})
950: at different values of $M$.
951:
952: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
953:
954: Along with
955: the correlation function $z(f)$ (\ref{pnA}), (\ref{BF4cl}) in
956: Figs.~\ref{zf1},\ref{zf2}
957: and the correlation coefficient
958: $\ol{\beta}=\ol{b}$ (\ref{bb3A}),(\ref{b4cl})
959: in Figs.~\ref{b1}-\ref{b4}, calculated on our scaling formulas,
960: which in Gauss approximation are the same for local and global fusion
961: (the solid lines),
962: we present at the same pictures the results of exact calculations
963: in the global fusion case on formulas (\ref{BFcl}) and (\ref{pBFcl})
964: at different values of $M$ (the dotted and dashed lines).
965:
966: We present also in Figs.~\ref{b1}-\ref{b4}
967: the results of our direct numerical MC calculations
968: of the $pt$--$n$ and $n$--$n$ correlation coefficients
969: in the global fusion case based on formulas (\ref{avOcl}-\ref{Optncl})
970: (half-filled squares).
971:
972: We see that in this case the Gauss approximation works very well
973: and the $\mu_0/\sqrt{\eta}$-scaling is not an artifact of
974: this approximation.
975: More over along with the $\mu_0/\sqrt{\eta}$-scaling at large $\eta$
976: we have also $M$-independence for correlation
977: coefficients $\ol{\beta}$ and $\ol{b}$ (see Figs.~\ref{b1}-\ref{b4})
978: starting very early (from $M=4$).
979:
980: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
981:
982:
983: \section{Conclusions.}
984:
985: In conclusion let us compare the results obtained
986: in the present paper in the case with string fusion
987: with results obtained
988: in \cite{vestn1} in the case without string fusion.
989: \vskip 4mm
990:
991: We have obtained in the paper \cite{vestn1}
992: in the case {\bf without} string fusion:
993: \begin{enumerate}
994: \item {for $n$--$n$ correlations:
995: \ $\ol{b} = \frac{a}{a+1}$ with $a=\mu_0\kappa$}
996: \item {for $pt$--$n$ correlations:
997: \ $\ol{\beta} =0$}
998: \end{enumerate}
999: \vskip 2mm
1000:
1001:
1002: In the present paper in the case with the {\bf global} string fusion
1003: and in the {\bf local} fusion case
1004: for {\bf homogeneous} situation ($\beti=\eta$)
1005: we find at large $\eta$:
1006: \begin{enumerate}
1007: \item {for $n$--$n$ correlations:
1008: \ $\ol{b} = \frac{a}{a+1}$ with $a=\frac{\mu_0\kappa}{4\sqrt{\bN/M}}
1009: =\mu_0\kappa/(4\sqrt{\eta})$}
1010: \item {for $pt$--$n$ correlations:
1011: \ $\ol{\beta} =\ol{b} = \frac{a}{a+1}$ with the same $a$}
1012: \end{enumerate}
1013: We see that with fusion the $n$--$n$ correlations became weaker,
1014: but now as compensation
1015: we have the $pt$--$n$ correlations of the same strength.
1016: We see also $\mu_0/\sqrt{\eta}$-scaling in this case.
1017: \vskip 4mm
1018:
1019:
1020:
1021: For {\bf nonhomogeneous} situation (different $\beti$)
1022: in the case with {\bf local} string fusion
1023: we have find at large $\beti$:
1024: \begin{enumerate}
1025: \item {for $n$--$n$ correlations:
1026: \ $\ol{b} = \frac{a}{a+1}$ with $a=\mu_0\kappa/(4\br/M)$}
1027: \item {for $pt$--$n$ correlations:
1028: \ $\ol{\beta} =
1029: \left( \frac{2\br^2}{\bN M}-1 \right) \ol{b}$}
1030: and hence $\ol{\beta}\leq\ol{b}$
1031: \end{enumerate}
1032: $$
1033: {\rm with} \hs{1} \bN=\sum_{i=1}^M \beti \hs{1} {\rm and} \hs{1} \br=\sum_{i=1}^M \sqrt{\beti}
1034: $$
1035: As we have demonstrated above
1036: (see (\ref{bbeta2A}) and (\ref{bbeta3A})),
1037: this leads to $\ol{\beta}$ smaller than $\ol{b}$: \ $\ol{\beta}\leq\ol{b}$.
1038: It's possible situation (\ref{bbeta4A}), in which $ \ol{\beta} < 0 $.
1039: \vskip 4mm
1040:
1041:
1042:
1043: At {\bf small} string density as it's shown in Appendix A
1044: the two types of limit at $\eta\to 0$ can be studied.
1045: \begin{enumerate}
1046: \item {If one keeps $M=const$, then we have $N \to 1$
1047: (because the configurations with $N=0$, are not considered
1048: as events) and we have nor $p_t$--$n$ nor $n$--$n$ correlation.}
1049: \item {If one keeps $\olN=const$, then $M\eta=const$ and $M\to\infty$,
1050: hence the strings will be far separated in transverse plane and
1051: we'll have the same results as in the case without
1052: string fusion \cite{vestn1}.}
1053: \end{enumerate}
1054: \vskip 2mm
1055:
1056:
1057: Note that the results obtained in our cellular approach
1058: are in a good agreement with the results obtained
1059: in the framework of the real string fusion model
1060: taking into account detail geometry of strings overlapping
1061: \cite{MAB}.
1062:
1063: % 7 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
1064:
1065: \subsection*{Acknowledgments.}
1066:
1067: The authors would like to thank M.A.~Braun
1068: and G.A.~Feofilov for numerous valuable
1069: encouraging discussions.
1070: The work has been partially supported by
1071: the Russian Foundation for Fundamental Research
1072: under Grant No. 01-02-17137-a.
1073:
1074:
1075: \section*{Appendixes:}
1076: \begin{appendix}
1077: \section{Correlations at small string density.}
1078: \label{ap:a}
1079:
1080: In this appendix we calculate the correlation functions
1081: and the correlation coefficients at small string density
1082: in the case with local string fusion.
1083: (In the case with global fusion this limit has no physical
1084: sense (see discussion in the section 5 of \cite{vestn1})).
1085:
1086: For clearness we consider
1087: "homogeneous" case, when all $\beti$
1088: is equal each other in the interaction area $\beti=\eta$.
1089: Then in each cell $i$ ($i=1,...,M$) the $\eta_i$ fluctuate
1090: around this mean value according to Poisson law
1091: ($\rho^{}_{a}(x)$ is the Poisson distribution with $\olx=a$):
1092: \beq
1093: w(\eta _i)=\rho _\eta ^{}(\eta _i)\equiv
1094: e^{-\eta }\frac{(\eta )^{\eta _i}}{\eta _i!}
1095: \label{ap:w}
1096: \eeq
1097:
1098: We'll assume also the Poissonian form of $p^{}_{\eta_i} (n_i)$:
1099: \beq
1100: p^{}_{\eta_i} (n_i) = \rho^{}_{\mu_0\etai}(n_i)
1101: \equiv e^{-\mu_0\etai}\frac{(\mu_0\etai)^{n_i}}{n_i!}
1102: \label{ap:Poisson}
1103: \eeq
1104: then we have the Poissonian distribution for
1105: \beq
1106: P^{}_{\{\eta^{}_1,...,\eta^{}_M\}} (\F) = \rho^{}_{\mu_0\reta}(\F)
1107: \label{ap:Poisso}
1108: \eeq
1109: with $\avr{\F} {{\{\eta^{}_1,...,\eta^{}_M\}}} =
1110: \mu_0 \reta = \mu_0 r = \avr{\F} {r}$.
1111:
1112: Then we find from (\ref{avO}) for $n$--$n$ correlations:
1113: \beq
1114: \langle n_B^{}\rangle _{n_F^{}}^{}=
1115: \frac{\mu _0\sum'_{\{\eta _1^{},...,\eta _M^{}\}}
1116: r\left( \prod_iw(\eta _i)\right) \rho _{\mu _0r}^{}(n_F^{})}
1117: {\sum'_{\{\eta _1^{},...,\eta _M^{}\}}\left( \prod_iw(\eta _i)\right)
1118: \rho _{\mu _0r}^{}(n_F^{})}
1119: \label{ap:nn}
1120: \eeq
1121: and for $pt$--$n$ correlations:
1122: \beq
1123: \langle p_{tB}^{2} \rangle _{n_F^{}}^{}=
1124: \frac{p^2\sum'_{\{\eta _1^{},...,\eta _M^{}\}}
1125: \frac{N}{r}\left( \prod_iw(\eta _i)\right) \rho _{\mu _0r}^{}(n_F^{})}
1126: {\sum'_{\{\eta _1^{},...,\eta _M^{}\}}\left( \prod_iw(\eta _i)\right)
1127: \rho _{\mu _0r}^{}(n_F^{})}
1128: \label{ap:ptn}
1129: \eeq
1130: Recall that
1131: $$
1132: N=\sum_{i=1}^M \eta_i
1133: \hs{1}{\rm and}\hs{1}
1134: r=\sum_{i=1}^M \sqrt{\eta_i}
1135: $$
1136: The $N$ is the number of strings in the given event.
1137:
1138: One has to omit in all $M$-fold sums
1139: in (\ref{ap:nn}) and (\ref{ap:ptn}) one term, when all $\eta_i=0$,
1140: which corresponds to the absence of inelastic interactions between
1141: the nucleons of the colliding nuclei (we denote this fact
1142: by $\sum'$).
1143:
1144: The probability $P(n_F)$ to detect $n_F$ particles
1145: in the forward rapidity window, which enters denominators
1146: of (\ref{ap:nn}) and (\ref{ap:ptn}), is
1147: \beq
1148: P(n_F)=C{\sum_{\{\eta _1^{},...,\eta _M^{}\}}}^{\!\!\!\!\!\!\prime}
1149: \;\; \left( \prod_i w(\eta _i)\right)
1150: \rho _{\mu _0^{}r}^{}(n_F^{})
1151: \label{ap:PnF}
1152: \eeq
1153: where from normalization condition we have
1154: \beq
1155: C=\frac{1}{1-w^M(0)}=
1156: \frac{1}{1-e^{-M\eta}}
1157: \label{ap:C}
1158: \eeq
1159:
1160: Clear that this factor $C$ is canceling in
1161: the numerator and in the denominator
1162: of (\ref{ap:nn}) and (\ref{ap:ptn}),
1163: but if we calculate
1164: the mean number of strings $\bN$,
1165: we find
1166: \beq
1167: \olN=
1168: C{\sum_{\{\eta _1^{},...,\eta _M^{}\}}}^{\!\!\!\!\!\!\prime}
1169: \;\; \left( \prod_i w(\eta _i)\right) N =
1170: C{\sum_{\{\eta _1^{},...,\eta _M^{}\}}}^{\!\!\!\!\!\!\prime}
1171: \;\;
1172: \left( \prod_i w(\eta _i)\right)
1173: \left( \sum_i \eta _i \right) = \frac{M\eta}{1-e^{-M\eta}}
1174: \label{ap:avN}
1175: \eeq
1176: and for the $\av{\F}$
1177: at small $\eta\ll 1$
1178: we have
1179: \beq
1180: \av{\F}=\sum_{n^{}_F} n^{}_F P(n^{}_F)=
1181: C\mu^{}_{0F}{\sum_{\{\eta _1^{},...,\eta _M^{}\}}}^{\!\!\!\!\!\!\prime}
1182: \;\; \left( \prod_i w(\eta _i)\right) r =
1183: \mu^{}_{0F}\frac{M\eta}{1-e^{-M\eta}}=
1184: \mu^{}_{0F}\olN
1185: \label{ap:avF}
1186: \eeq
1187: Because for any $\omega>0$ we have $\sum_{\eta_i}^{}\eta_i^\omega
1188: w(\eta_i)=\eta+O(\eta^2)$ at $\eta\to 0$, as the main contribution
1189: comes from the term $\eta_i=1$.
1190:
1191: There are two possibilities when $\eta\to 0$:
1192: \begin{enumerate}
1193: \item {$M=const$ and then $M\eta\to 0$ and $\olN\to 1$
1194: (see (\ref{ap:avN}))}
1195: \item {$\olN=const$ and then $M\eta=const$ ((see (\ref{ap:avN}))
1196: and $M\to\infty$}
1197: \end{enumerate}
1198: We'll investigate both these possibilities.
1199:
1200: The first one means that in the limit we have $N=\olN=1$
1201: (because the configurations with $N=0$, are not considered
1202: as events). Clear that in this situation we'll have
1203: nor $p_t$--$n$ nor $n$--$n$ correlation, as we'll
1204: have no fluctuations in the number of strings
1205: (see discussion in the end of the section 4 of \cite{vestn1}).
1206: Detail calculations are presented below.
1207:
1208: In the second case with $\olN=const$ we'll have the
1209: fluctuations in the number of strings $N$,
1210: but in the limit $\eta\to 0$ we'll have $M\to\infty$,
1211: the strings will be far separated
1212: in transverse plane and the strings fusion will plays no role.
1213: So in this case we'll have the same results as in
1214: the no fusion case, considered in \cite{vestn1}:
1215: large $n$--$n$ correlation
1216: with a correlation coefficient equal to $b=\mu_0^{}/(\mu_0^{}+1)$
1217: and no $p_t$--$n$ correlation.
1218: (See calculations below.)
1219:
1220: {\bf Detail calculations.}
1221: Let us evaluate the $M$-fold sums
1222: in (\ref{ap:nn}) and (\ref{ap:ptn}) at $\eta\to 0$
1223: keeping all terms of order $(M\eta)^k$, $(M\eta)^k\eta$
1224: with any $k$
1225: and omitting all terms of order $(M\eta)^k\eta^2$ and higher.
1226:
1227: The terms of order $(M\eta)^k$ originates from the
1228: summands in (\ref{ap:nn}) and (\ref{ap:ptn})
1229: with $\eta_{i_1}=...=\eta_{i_k}=1$ and other $\eta_{i}=0$.
1230: The terms of order $(M\eta)^k\eta$ originates from the
1231: summands in (\ref{ap:nn}) and (\ref{ap:ptn})
1232: with $\eta_{i_1}=2$,
1233: $\eta_{i_2}=...=\eta_{i_k}=1$ and other $\eta_{i}=0$.
1234: Keeping this into mind we find for $P(n_F)$:
1235: \beq
1236: P(n_F)=C(G_0 + \frac{\eta}{2} G_1)
1237: \label{ap:PnF1}
1238: \eeq
1239: and for $n$--$n$ correlations:
1240: \beq
1241: \langle n_B^{}\rangle _{n_F^{}}^{}=\mu^{}_{0B}
1242: \frac{N_0 + \frac{\eta}{2} N_1}{G_0 + \frac{\eta}{2} G_1}
1243: \label{ap:nn1}
1244: \eeq
1245: For $pt$--$n$ correlations we have:
1246: \beq
1247: \langle p_{tB}^{2} \rangle _{n_F^{}}^{}=p^2
1248: \frac{P_0 + \frac{\eta}{2} P_1}{G_0 + \frac{\eta}{2} G_1}=
1249: p^2 \left( 1 + \frac{\eta}{2} \frac{P_1-G_1}{G_0} \right)
1250: \label{ap:ptn1}
1251: \eeq
1252: Here
1253: \beq
1254: G_0=P_0=\sum_{k=1}^{M} C_M^k \eta^k \rho^{}_{\mu_0 k}(n^{}_F)
1255: \label{ap:coeff1}
1256: \eeq
1257: $$
1258: N_0=\sum_{k=1}^{M} k C_M^k \eta^k \rho^{}_{\mu_0 k}(n^{}_F)
1259: $$
1260: $$
1261: G_1=\sum_{k=1}^{M} k C_M^k \eta^k \rho^{}_{\mu_0 (k+\gamma)}(n^{}_F)
1262: $$
1263: $$
1264: N_1=\sum_{k=1}^{M} (k+\gamma)k C_M^k \eta^k \rho^{}_{\mu_0 (k+\gamma)}(n^{}_F)
1265: $$
1266: $$
1267: P_1=\sum_{k=1}^{M} \frac{k+1}{k+\gamma} k C_M^k \eta^k \rho^{}_{\mu_0 (k+\gamma)}(n^{}_F)
1268: $$
1269: where $\mu^{}_0\equiv\mu^{}_{0F}$ and $\gamma=\sqrt{2}-1$.
1270:
1271: Note that at $M \gg 1$ and $M\eta=const$ we have for $P(n_F)$:
1272: \beq
1273: P(n_F)=C e^{-M\eta}\frac{\mu_0^{n^{}_F}}{n^{}_F !}
1274: (\olG_0 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma} \olG_1)
1275: \label{ap:PnF2}
1276: \eeq
1277: and for $n$--$n$ correlations:
1278: \beq
1279: \langle n_B^{}\rangle _{n_F^{}}^{}=\mu^{}_{0B}
1280: \frac{\olN_0 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma} \olN_1}
1281: {\olG_0 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma} \olG_1}
1282: \label{ap:nn2}
1283: \eeq
1284: For $pt$--$n$ correlations we have:
1285: \beq
1286: \langle p_{tB}^{2} \rangle _{n_F^{}}^{}=p^2
1287: \frac{\olP_0 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma} \olP_1}
1288: {\olG_0 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma}\olG_1}=
1289: p^2 \left( 1 + \frac{\eta}{2} e_{}^{-\mu^{}_0 \gamma}
1290: \frac{\olP_1-\olG_1}{\olG_0} \right)
1291: \label{ap:ptn2}
1292: \eeq
1293: Here
1294: \beq
1295: \olG_0=\olP_0=\sum_{k=1}^{\infty} k_{}^{n^{}_F} \frac{d^k}{k!}
1296: \label{ap:coeff2}
1297: \eeq
1298: $$
1299: \olN_0=\sum_{k=1}^{\infty} k_{}^{n^{}_F+1} \frac{d^k}{k!}
1300: $$
1301: $$
1302: \olG_1=\sum_{k=1}^{\infty} (k+\gamma)_{}^{n^{}_F} \frac{d^k}{(k-1)!}
1303: $$
1304: $$
1305: \olN_1=\sum_{k=1}^{\infty} (k+\gamma)_{}^{n^{}_F+1} \frac{d^k}{(k-1)!}
1306: $$
1307: $$
1308: \olP_1=\sum_{k=1}^{\infty} (k+\gamma)_{}^{n^{}_F-1}(k+1) \frac{d^k}{(k-1)!}
1309: $$
1310: where $d=M\eta e^{-\mu^{}_0}$.
1311:
1312: For the control of calculations
1313: we have used also the following explicit formulas at
1314: $n^{}_F=0,1,2$:
1315: \beq
1316: \olG_0(0)=\olP_0(0)=e^d-1
1317: \label{ap:coeff3}
1318: \eeq
1319: $$
1320: \olN_0(0)=\olG_1(0)=\olG_0(1)=\olP_0(1)=de^d
1321: $$
1322: $$
1323: \olN_1(0)=\olG_1(1)=(d+\sqrt{2})de^d
1324: $$
1325: $$
1326: \olP_1(0)=\sum_{k=1}^{\infty} \frac{k+1}{k+\gamma}\; \frac{d^k}{(k-1)!}
1327: $$
1328: $$
1329: \olN_0(1)=\olG_0(2)=\olP_0(2)=(d+1)de^d
1330: $$
1331: $$
1332: \olN_1(1)=\olG_1(2)=(d^2+d(1+2\sqrt{2})+2)de^d
1333: $$
1334: $$
1335: \olP_1(1)=(d+2)de^d
1336: $$
1337: $$
1338: \olN_0(2)=(d^2+3d+1)de^d
1339: $$
1340: $$
1341: \olN_1(2)=(d^3+d^23(1+\sqrt{2})+d(7+3\sqrt{2})+2\sqrt{2})de^d
1342: $$
1343: $$
1344: \olP_1(2)=(d^2+d(3+\sqrt{2})+2\sqrt{2})de^d
1345: $$
1346:
1347: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
1348:
1349: The results of the calculations using the asymptotic formulas
1350: (\ref{ap:nn1}--\ref{ap:coeff1})
1351: in the first case
1352: ($M=const$, $\eta\to 0$, $M\eta\to 0$ and $\olN\to 1$)
1353: are shown in the Figs.~\ref{sm1}-\ref{sm2} (lines) together
1354: with results of direct MC numerical calculations using
1355: formulas (\ref{ap:nn}) and (\ref{ap:ptn})(points).
1356: As we have expected in this case both
1357: $n$--$n$ and $p_t$--$n$ correlation coefficients
1358: go to zero when $\olN\sim 1$, i.e. at $\eta<1/M$.
1359: Remember that we have $M$-independence
1360: for the correlation coefficients at large $\eta$.
1361: Now we see
1362: that in this limit it disappears at $\eta \leq 1/M$.
1363: This is also the reason for nonlinear dependence
1364: of the correlation coefficients on $\eta$ in this region
1365: which one can see in Figs.~\ref{sm1}-\ref{sm2} for $\mu_0=4$.
1366:
1367: We see also that
1368: using asymptotic formulas (\ref{ap:nn1}--\ref{ap:coeff1})
1369: we can calculate
1370: $n$--$n$ correlation coefficients
1371: in wider region of small $\eta$, than
1372: $p_t$--$n$ correlation coefficients,
1373: because we have the contributions of order
1374: $(M\eta)^k$ and $(M\eta)^k\eta$ for
1375: $n$--$n$ correlations and only first non-trivial
1376: contribution of order $(M\eta)^k\eta$ for
1377: $p_t$--$n$ correlations.
1378: Note the very good agreement between results of
1379: the calculations on the asymptotic formulas
1380: (\ref{ap:nn1}--\ref{ap:coeff1})
1381: (lines in the Figs.~\ref{sm1}-\ref{sm2})
1382: and
1383: the results of direct MC numerical calculations on
1384: formulas (\ref{ap:nn}) and (\ref{ap:ptn})
1385: (points in the Figs.~\ref{sm1}-\ref{sm2}).
1386:
1387: %XXXXXX DOD XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
1388:
1389: In the second case when $\eta\to 0$
1390: we keep $\olN=const$ and then due to (\ref{ap:avN}) $M\eta=const$
1391: so $M\to\infty$. We can use formulas
1392: (\ref{ap:nn2}--\ref{ap:coeff2}) with $d=const$,
1393: then in the limit $\eta\to 0$ we find for $n$--$n$ correlations:
1394: \beq
1395: \langle n_B^{}\rangle _{n_F^{}}^{}=
1396: \mu^{}_{0B} \frac{\olN_0}{\olG_0}=
1397: \mu^{}_{0B}
1398: \frac{\sum_{k=1}^{\infty} k \rho_{M\eta}^{}(k) \rho_{\mu _0 k}^{}(n_F^{})}
1399: {\sum_{k=1}^{\infty} \rho_{M\eta}^{}(k) \rho_{\mu _0 k}^{}(n_F^{})}
1400: \label{ap:nn3}
1401: \eeq
1402: Here we multiply both the numerator and the denominator by
1403: $e^{-M\eta }\mu _0^{n_F^{}}/n_F^{}!$.
1404:
1405: Recall that
1406: $\rho^{}_{a}(x)$ is the Poisson distribution with $\olx=a$,
1407: then we see that formula (\ref{ap:nn3}) coincides with
1408: the formula for $n$--$n$ correlations obtained in the paper
1409: \cite{vestn1} in the case {\it without} string fusion
1410: (see formula (23) in \cite{vestn1}).
1411: So we'll have the same result for the $n$--$n$
1412: correlation coefficient $b=\mu_0^{}/(\mu_0^{}+1)$ as in the
1413: no fusion case (see Figs.~\ref{sm1}-\ref{sm2}).
1414:
1415: For $p_t$--$n$ correlation coefficients
1416: in this limit we find from
1417: (\ref{ap:ptn2}) and (\ref{ap:coeff2}) with $d=const$:
1418: \beq
1419: \langle p_{tB}^{2}\rangle _{n_F^{}}^{}=
1420: p^2 ( 1 + O(\eta))
1421: \label{ap:ptn3}
1422: \eeq
1423: So we have no $p_t$--$n$ correlation as in the case without
1424: string fusion \cite{vestn1}.
1425:
1426:
1427: \section{
1428: On the difference between $\av{\B}$ and $\avr{\B} {\F=\av{\F}}$.}
1429: \label{ap:b}
1430:
1431: It's possible instead of (\ref{bbA}) and (\ref{bbetaA})
1432: to use the following definitions for the correlation coefficients:
1433: \beq
1434: \ol{b} \equiv\frac{d\avr{\B}{\F}/\avr{\B} {\av{\F}}}{d\F/\av{\F}} \at {\F=\av{\F}}
1435: \label{A1}
1436: \eeq
1437: and
1438: \beq
1439: \ol{\beta} \equiv\frac{d\avr{\ppB}{\F}/\avr{\ppB} {\av{\F}}}{d\F/\av{\F}} \at {\F=\av{\F}}
1440: \label{A2}
1441: \eeq
1442: where
1443: \beq
1444: \avr{\B}{\av\F} = \avr{\B}{\F=\av\F}
1445: \label{A3}
1446: \eeq
1447: and
1448: \beq
1449: \avr{\ppB}{\av\F} = \avr{\ppB}{\F=\av\F}
1450: \label{A4}
1451: \eeq
1452: Clear that
1453: \beq
1454: \av\B = \sum_\F \olP(\F) \avr{\B}{\F} \approx \avr{\B}{\F=\av\F} \sum_\F P(\F)
1455: = \avr{\B}{\F=\av\F} \equiv \avr{\B}{\av\F}
1456: \label{A5}
1457: \eeq
1458: and the same for $\avr{\ppB}{\av\F} \equiv \avr{\ppB}{\F=\av\F} \approx \av{\ppB} $.
1459:
1460: In our Gauss approximation these two types of quantities
1461: coincide with each other.
1462: \end{appendix}
1463:
1464:
1465: % Bibl XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
1466:
1467:
1468: \begin{thebibliography}{99}
1469:
1470:
1471: \bibitem{Capella1}
1472: A.~Capella, U.P.~Sukhatme, C.--I.~Tan and J.~Tran~Thanh~Van,\\
1473: \j{\PL}{B81}{1979}{68}; \j{\PRep}{236}{1994}{225}.
1474: \bibitem{Kaidalov}
1475: A.B.~Kaidalov, \j{\PL}{116B}{1982}{459};\\
1476: A.B.~Kaidalov and K.A.~Ter-Martirosyan, \j{\PL}{117B}{1982}{247}.
1477:
1478:
1479: \bibitem{BP1} M.A.~Braun and C.~Pajares,
1480: \j{\PL}{B287}{1992}{154};\\
1481: \j{\NP}{B390}{1993}{542, 549}.
1482:
1483: \bibitem{ABP1} N.S.~Amelin, M.A.~Braun and C.~Pajares,
1484: \j{\PL}{B306}{1993}{312};\\
1485: \j{\ZP}{C63}{1994}{507}.
1486:
1487: \bibitem{BP2} N.~Armesto, M.A.~Braun, E.G.~Ferreiro and C.~Pajares,\\
1488: \j{\PRL}{77}{1996}{3736}.
1489:
1490: \bibitem{BPR}
1491: M.A.~Braun, C.~Pajares and J.~ Ranft,\\
1492: \j{\IJMP}{A14}{1999}{2689}; hep-ph/9707363.
1493:
1494:
1495:
1496: \bibitem{PRL94}
1497: N.S.~Amelin, N.~Armesto, M.A.~Braun, E.G.~Ferreiro and C.~Pajares,
1498: \j{\PRL}{73}{1994}{2813}.
1499:
1500: \bibitem{BPep00}
1501: M.A.~Braun and C.~Pajares, \j{\EPJ}{C16}{2000}{349}.
1502:
1503: \bibitem{BP00}
1504: M.A.~Braun and C.~Pajares, \j{\PRL}{85}{2000}{4864}.
1505:
1506:
1507:
1508: \bibitem{BPV00}
1509: M.A.~Braun, C.~Pajares and V.V.~Vechernin,\\
1510: \j{\PL}{B493}{2000}{54}.
1511:
1512: \bibitem{vestn1}
1513: V.V.~Vechernin and R.S.~Kolevatov, hep-ph/0304295;
1514: Russian version will appear in the journal "Vestnik SPbU".
1515:
1516:
1517: \bibitem{Heiselberg}
1518: H.Heiselberg, nucl-th/0003046, 2001.
1519:
1520: \bibitem{MAB}
1521: M.A.~Braun. Private communication.
1522:
1523:
1524: \end{thebibliography}
1525:
1526:
1527:
1528: % Figs: zf1 zf2 b1 b2 b4 sm1 sm2 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
1529:
1530: %%% BEGIN Fig.1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1531: \begin{figure}[t]
1532: % t-top, b-bottom, p-page, h-here
1533: \centerline{\epsfig{file=zf1.ps,width=10cm,angle=-90}}
1534: \caption[dummy]{\label{zf1}
1535: The $p_t$--$n$ and $n$--$n$ correlation functions.
1536: Solid line - the Gauss approximation for local and global fusion
1537: at the value of scaling variable $\mu_0/\sqrt{\eta}=1$.
1538: Dashed and dotted lines - the results of exact calculations
1539: in the global fusion case at different values of
1540: $\mu_0$, $\eta$ and $M$.
1541: The $\mu_0/\sqrt{\eta}$-scaling and $M$-independence.
1542: }
1543: \end{figure}
1544: %%% END Fig.1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1545:
1546: %%% BEGIN Fig.2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1547: \begin{figure}[t]
1548: % t-top, b-bottom, p-page, h-here
1549: \centerline{\epsfig{file=zf2.ps,width=10cm,angle=-90}}
1550: \caption[dummy]{\label{zf2}
1551: The same as in Fig.~\ref{zf1}, but at different values of
1552: $\mu_0$ and $\eta$.
1553: }
1554: \end{figure}
1555: %%% END Fig.2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1556:
1557: %%% BEGIN Fig.3%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1558: \begin{figure}[t]
1559: % t-top, b-bottom, p-page, h-here
1560: \centerline{\epsfig{file=b1.ps,width=10cm,angle=-90}}
1561: \caption[dummy]{\label{b1}
1562: The $p_t$--$n$ and $n$--$n$ correlation coefficients at $\mu_0=1$.
1563: Solid line - the Gauss approximation for local and global fusion
1564: $\olb=\olbe=\mu_0/(\mu_0+4\sqrt{\eta})$.
1565: Dashed and dotted lines - the results of exact calculations
1566: on the formulas (\ref{BFcl}-\ref{pBFcl})
1567: in the global fusion case at different values of $M$
1568: (half-filled squares - the same by means of
1569: direct numerical MC calculations on formulas (\ref{avOcl}-\ref{Optncl})).
1570: The empty and filled points -
1571: the results of direct numerical MC calculations
1572: in the local fusion case based on formulas (\ref{avO}-\ref{Optn})
1573: for the $pt$--$n$ and $n$--$n$ correlations correspondingly.
1574: The filled circle - the $n$--$n$ correlation coefficient
1575: $\olb=\mu_0/(\mu_0+1)$
1576: in the case without string fusion \cite{vestn1}.
1577: }
1578: \end{figure}
1579: %%% END Fig.3%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1580:
1581: %%% BEGIN Fig.4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1582: \begin{figure}[t]
1583: % t-top, b-bottom, p-page, h-here
1584: \centerline{\epsfig{file=b2.ps,width=10cm,angle=-90}}
1585: \caption[dummy]{\label{b2}
1586: The same as in Fig.~\ref{b1}, but at $\mu_0=2$.
1587: }
1588: \end{figure}
1589: %%% END Fig.4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1590:
1591: %%% BEGIN Fig.5%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1592: \begin{figure}[t]
1593: % t-top, b-bottom, p-page, h-here
1594: \centerline{\epsfig{file=b4.ps,width=10cm,angle=-90}}
1595: \caption[dummy]{\label{b4}
1596: The same as in Fig.~\ref{b1}, but at $\mu_0=4$.
1597: }
1598: \end{figure}
1599: %%% END Fig.5%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1600:
1601: %%% BEGIN Fig.6%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1602: \begin{figure}[t]
1603: % t-top, b-bottom, p-page, h-here
1604: \centerline{\epsfig{file=sm1.ps,width=10cm,angle=-90}}
1605: \caption[dummy]{\label{sm1}
1606: The $n$--$n$ correlation coefficient
1607: at small values of $\eta$ for $\mu_0=1$ and $\mu_0=4$.
1608: The lines - results of the calculations using the asymptotic formulas
1609: (\ref{ap:nn1}) and (\ref{ap:coeff1}) at $M=const$ ($\olN\to 1$).
1610: The points - results of direct MC numerical calculations using
1611: formula (\ref{ap:nn}).
1612: The arrows show the value of the $n$--$n$ correlation coefficient
1613: $\olb=\mu_0/(\mu_0+1)$
1614: in the case without string fusion \cite{vestn1},
1615: which corresponds to the limit $\olN=const$ ($M\to\infty$).
1616: }
1617: \end{figure}
1618: %%% END Fig.6%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1619:
1620: %%% BEGIN Fig.7%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1621: \begin{figure}[t]
1622: % t-top, b-bottom, p-page, h-here
1623: \centerline{\epsfig{file=sm2.ps,width=10cm,angle=-90}}
1624: \caption[dummy]{\label{sm2}
1625: The $p_t$--$n$ correlation coefficients
1626: at small values of $\eta$ for $\mu_0=1$ and $\mu_0=4$.
1627: The lines - results of the calculations using the asymptotic formulas
1628: (\ref{ap:ptn1}) and (\ref{ap:coeff1}) at $M=const$ ($\olN\to 1$).
1629: The points - results of direct MC numerical calculations using
1630: formula (\ref{ap:ptn}).
1631: }
1632: \end{figure}
1633: %%% END Fig.7%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1634:
1635:
1636: \end{document}