1: % version 28.05.01/07.08.01- accepted for publication in Nucl. Phys. A
2: %--------------------------------------------------------------------
3: \documentstyle[12pt,epsfig]{article}
4: %\def\baselinestretch{1.7}
5: \setlength{\textwidth}{15cm}
6: \setlength{\textheight}{23cm}
7: \setlength{\evensidemargin}{-0.5cm}
8: \setlength{\oddsidemargin}{0.5cm}
9: \topmargin=-1cm
10: \parindent = 1.5em
11: \newcommand{\be}{\begin{equation}}
12: \newcommand{\ee}{\end{equation}}
13: \newcommand{\bea}{\begin{eqnarray}}
14: \newcommand{\eea}{\end{eqnarray}}
15:
16: \begin{document}
17: \title{Antibaryon production in hot and dense nuclear matter
18: \thanks{Supported by GSI Darmstadt.} }
19: \author{ W. Cassing \\
20: Institut f\"{u}r Theoretische Physik, Universit\"{a}t Giessen \\
21: D-35392 Giessen, Germany}
22: \maketitle
23: \begin{abstract}
24: The production of antibaryons is calculated in a microscopic
25: transport approach employing multiple meson fusion reactions
26: according to detailed balance relations with respect to
27: baryon-antibaryon annihilation. It is found that the abundancies
28: of antiprotons as observed from peripheral to central collisions
29: of $Pb + Pb$ at the SPS and $Au + Au$ at the AGS can approximately
30: be described on the basis of multiple interactions of 'formed'
31: hadronic states which drive the system to chemical equilibrium by
32: flavor exchange or quark rearrangement reactions.
33: \end{abstract}
34:
35: \vspace{2cm} \noindent PACS: 24.10.-i; 24.10.Cn; 24.10.Jv;
36: 25.75.-q; 14.65.-q
37:
38: \noindent Keywords: Nuclear reaction models and methods; Many-body
39: theory; Relativistic models; Relativistic heavy-ion collisions;
40: Quarks
41:
42: \newpage
43:
44: \section{Introduction}
45: Ever since the first observation of antiproton production in
46: proton-nucleus \cite{chamberlain,elioff,dorfan} and
47: nucleus-nucleus collisions \cite{JINR,BEVALAC1,BEVALAC2,KEK,GSI}
48: the production mechanism has been quite a matter of debate. Especially
49: in nucleus-nucleus collisions at subthreshold energies
50: traditional cascade calculations, that employ free $NN$ production
51: and $\bar{p}N$ annihilation cross sections, essentially fail in
52: describing the high cross sections seen from 1.5 - 2.1 A$\cdot$GeV
53: \cite{Batko91,Cass92,Faess1}. Thus multiparticle nucleon
54: interactions \cite{Danielewicz90,Weise} have been suggested as a
55: possible solution to this problem. On the other hand it has been
56: pointed out that the quasi-particle properties of the nucleons
57: and antinucleons might be important for the $\bar{p}$ production
58: process which become more significant with increasing nuclear
59: density. Schaffner et al. \cite{Schaffner91} found in a static
60: thermal relativistic model based on scalar and vector self
61: energies-- assuming kinetic and chemical equilibrium -- that the
62: $\bar{p}$-abundancy might be dramatically enhanced when assuming
63: the antiproton self energy in the medium to be given by charge conjugation of
64: the nucleon self energy. This assumption implies strong attractive
65: vector self energies for the antiprotons which lead to a
66: reduction of the necessary energy to produce $p\bar{p}$ pairs in the medium by
67: binary quasi-particle interactions. On the other hand, such self
68: energies will lead to different spectral slopes of protons and
69: antiprotons as pointed out in Ref. \cite{Koch}.
70:
71: First nonequilibrium calculations within a fully relativistic
72: transport model for antiproton production -- including $\bar{p}$
73: annihilation as well as the change of the quasi-particle
74: properties in the medium -- have been performed in Ref.
75: \cite{Cass92}. There it was found that according to the reduced
76: antinucleon energy in the medium the threshold for
77: $\bar{p}$-production is shifted to lower energy and the antiproton
78: cross section prior to annihilation becomes enhanced e.g. for
79: $Si+Si$ at 2.1 GeV$\cdot$GeV by approximately a factor 70 as
80: compared to a relativistic cascade calculation where no in-medium
81: effects are incorporated. Later on, a couple of relativistic
82: transport calculations have been performed
83: \cite{Teis94,Ko93,RQMD,sibirtsev} essentially pointing out that
84: all the low energy data from proton-nucleus and nucleus-nucleus
85: collisions are compatible with attractive $\bar{p}$ self energies
86: (at normal nuclear matter density $\rho_0$) in the order of -100
87: to -150 MeV \cite{sibirtsev,Cass99,Ko96}. However, it had been
88: stressed at that time that the high antiproton yield might also be
89: attributed to mesonic production channels \cite{ko1,ko2,Wittmann}
90: since $p \bar{p}$ annihilation leads to multi-pion final states
91: with an average abundancy of 5 pions \cite{Dover}, which e.g.
92: might stem from an intermediate state of 2 $\rho$-mesons and a pion.
93:
94: With new data coming up on antibaryon production from
95: nucleus-nucleus collisions at the AGS \cite{AGSall,E877,AGSnew}
96: and SPS \cite{NA49,Na49b,NAxx,NA57,Andersen} the $\bar{p},
97: \bar{\Lambda}$ enhancement factors seen experimentally were no
98: longer that dramatic as at SIS energies, however, traditional
99: cascade calculations employing free production and annihilation
100: cross sections again were not able to reproduce the measured
101: abundancies and spectra \cite{AGS1,AGS2,Kahana,Bleicher,WA97c}
102: especially for $\Xi, \bar{\Xi}$ and $\Omega, \bar{\Omega}$. Here
103: additional collective mechanisms in the entrance channel have been
104: suggested such as color rope formation \cite{Sorge} or hot plasma
105: droplet formation \cite{Werner}. In another language this has been
106: addressed also as string fusion \cite{Carlos,Carlos2}, a precursor
107: phenomenon for the formation of a quark-gluon plasma (QGP).
108:
109: The intimate connection of antibaryon abundancies with the
110: possible observation of a new state of the strongly interacting
111: hadronic matter, i.e. the quark-gluon plasma, has been often
112: discussed since the early suggestion in Ref. \cite{Rafelski} that
113: especially the enhanced yield of strange antibaryons --
114: approximately in chemical equilibrium with the other hadronic
115: states -- should be a reliable indicator for a new state of
116: matter. In fact, the data on strange baryon and antibaryon
117: production from the NA49 and WA97 Collaborations show an
118: approximate chemical equilibrium \cite{BM,becca,Redlich00,Red01}
119: with an enhancement of the $\Omega^-, \Omega^+$ yield in central
120: $Pb + Pb$ collisions (per participant) relative to $p Be$
121: collisions at the same invariant energy per nucleon by a factor
122: $\sim $ 15. As pointed out in Ref. \cite{Redlich2} the data on
123: multi-strange antibaryons at the SPS seem compatible with a
124: canonical ensemble in chemical equilibrium. At AGS energies of
125: 11.6 A$\cdot$GeV/c, furthermore, a high ratio of
126: $\bar{\Lambda}$/$\bar{p}$ of $3.6^{+4.7}_{-1.8}$ has been reported
127: \cite{AGSnew} for central collisions of $Au + Au$, that is not
128: described by any approach so far. Strange flavor exchange
129: reactions \cite{Capella} help in creating multistrange
130: antibaryons, however, the latter strangeness enhancement factors
131: could not be described within traditional transport or cascade
132: simulations without additional assumptions such as enhanced string
133: tensions or reduced quark masses \cite{Bleicher}.
134:
135: In a more recent paper Rapp and Shuryak have taken up again the
136: idea of multi-meson production channels for baryon-antibaryon
137: pairs \cite{Rapp} to describe the antiproton abundancies in
138: central $Pb+Pb$ collisions at the SPS by introducing additionally
139: a finite pion chemical potential which helps in enhancing the
140: multi-pion collision rate. Later on, Greiner and Leupold
141: \cite{Carsten} have applied the same concept for the
142: $\bar{\Lambda}$ production by a couple of mesons including a $K^+$
143: or $K^0$ (for the $\bar{s}$ quark). However, such estimates remain
144: schematic unless fully microscopic multi-particle calculations
145: support or disprove such suggestions. The problem here is that
146: most of the transport models include only binary reactions in the
147: entrance channel whereas the final channel of an energetic
148: collision may well consist of many hadrons emerging from the decay
149: of strings that are excited in the initial reaction. Thus, as has
150: been pointed out quite often \cite{Rapp,Brat,Bravina}, detailed
151: balance is not included on the many-particle level leading to an
152: improper equilibrium state for large times ($t \rightarrow
153: \infty$).
154:
155: In this work we will address two separate questions: the first one
156: is of more formal nature and related to a transport approach that
157: properly takes into account reactions of 2 hadrons $\rightarrow n$
158: hadrons and vice versa employing detailed balance (Section 2). The
159: second one addresses a suitable covariant scheme for the
160: calculation of such multi-particle reactions in transport models.
161: The method and its implementation in the hadron-string-dynamics
162: (HSD) transport approach \cite{Cass99,Ehehalt} will be described
163: in Section 3. A first application of this novel approach is
164: devoted to the problem of antibaryon production in relativistic
165: nucleus-nucleus collisions by multi-meson reaction channels.
166: Respective calculations and studies at SPS and AGS energies --
167: with a focus on antiproton abundancies -- will be presented in
168: Section 4 whereas Section 5 concludes this work with a summary.
169:
170:
171: \section{Generalized transport equations}
172: In this Section a brief description of the relativistic transport
173: model is given with emphasis on a new development, i.e. the
174: multi-particle reaction dynamics. First we summarize (or review) the relevant
175: equations determining the dynamics of baryons and mesons and then
176: discuss a flavor rearrangement model for baryon-antibaryon
177: annihilation and production in the extended
178: HSD transport approach.
179:
180: \subsection{Hadron transport and multi-particle transitions}
181:
182: Since the covariant transport approach for binary reactions has
183: been extensively discussed in Refs. \cite{KLW1} and in the reviews
184: \cite{Cass99,Mal,Cassing90c} we only recall the basic equations
185: that are relevant for a proper understanding of the results to be
186: reported in this study.
187:
188:
189: For the discussion of the general collision terms the hadron self energies
190: will be discarded for transparency\footnote{The actual transport calculations,
191: however, include hadron self energies which optionally can be switched-off.},
192: such that the transport equation in the cascade limit reduces to
193: \begin{equation}
194: \label{vlasov}
195: p_\mu \partial^\mu_x F_i(x,p) = ( p_0 \partial_t +
196: {\vec p} \cdot {\vec \partial}_r) F_i(x,p) = I_{coll}^i ,
197: \end{equation}
198: where $F_i(x,p)$ is the Lorentz covariant 8 dimensional
199: phase-space distribution function for an off-shell hadron with quantum
200: numbers $i$, i.e.
201: \begin{equation}
202: \label{spectral}
203: F(x,p) = A(x,p) f(x,p),
204: \end{equation}
205: where $A(x,p)$ denotes the hadron spectral function while $f(x,p)$
206: describes the occupation probability in phase-space.
207: The off-shell propagation of hadrons leads to additional terms
208: in the l.h.s. of (1) that describe the change of the spectral
209: function $A_i(x,p)$ during the propagation. These terms are omitted here
210: since the actual calculations will be performed in the on-shell limit.
211: For details on the off-shell propagation of hadrons in
212: the medium the reader is
213: referred to Refs. \cite{Casju,Leupold}.
214:
215: We now turn to a discussion of the collision term $I_{coll}^i$,
216: which includes the new elements to be presented below. In the most
217: general case it is a sum of collision integrals involving $n
218: \leftrightarrow m$ reactions,
219: \begin{equation}
220: \label{icoll}
221: I_{coll} = \sum_n \sum_m \ I_{coll} [n \leftrightarrow m].
222: \end{equation}
223: The general form for off-shell fermions with spectral functions
224: $A_i(x,p_i)$ in case of 2-body interactions is given by
225: \cite{Casju}
226:
227: \bea \lefteqn{ I_{coll}^i [2 \leftrightarrow 2] = } \nonumber \\
228: && \frac{1}{2} \sum_j \sum_{k,l} \frac{1}{(2 \pi)^{12}} \int d^{4} p_{2} \,
229: d^{4} p_{3} \,
230: d^{4} p_4 \ A_i(x,p) A_j(x,p_2) A_k(x,p_3)
231: A_l(x,p_4)
232: \nonumber \\
233: && \times W_{2,2}(p, p_2; i,j \mid p_3,p_4; k, l) \
234: (2\pi)^4 \ \delta^{4}(p^{\mu} + p^{\mu}_2 - p_3^{\mu} - p_4^{\mu} )
235: \nonumber \\
236: && \times [ f_k(x,p_3) f_l(x,p_4)(1 - f_i(x,p))(1 -
237: f_j(x,p_2))
238: \nonumber \\
239: && - f_i(x,p) f_j(x,p_2)(1 - f_k(x,p_3))(1-f_l(x,p_4)) ].
240: \label{icoll2} \eea
241: %
242: This collision integral describes the change in the 8 dimensional
243: phase-space distribution function $F_i(x,p)$ due to the collisions
244: of two baryons with momenta $p^{\mu}$ and $p^{\mu}_2$ and
245: discrete quantum numbers $i$ and $j$, respectively, whereas the
246: two fermions in the final state of the reaction are labeled by
247: their momenta $p_3$ and $p_4$ and discrete quantum numbers $k$ and
248: $l$. The $\delta-$function guarantees energy and momentum
249: conservation in the individual collision while $ W_{2,2}(p, p_2;
250: i,j \mid p_3, p_4; k,l)$ denotes the transition probability (or
251: transition amplitude squared) for this reaction which in case of
252: fermions includes the antisymmetrization.
253:
254: The generalization of (\ref{icoll2}) to $n \leftrightarrow m$ interactions is
255: straight forward and reads: \bea \lefteqn{I_{coll}^i [n
256: \leftrightarrow m] = } \nonumber \\ && \frac{1}{2} N_n^m \sum_\nu
257: \sum_{\lambda} \left( \frac{1}{(2 \pi)^{4}} \right)^{n+m-1} \int
258: \left ( \prod_{j=2}^n d^{4} p_j \,\ A_j(x,p_j)\right ) \left (
259: \prod_{k=1}^m d^{4} p_{k} \,
260: A_k(x,p_k) \right ) \
261: \nonumber \\
262: && \times A_i(x,p) \ W_{n,m}(p, p_j; i,\nu \mid p_k; \lambda) \
263: (2\pi)^4 \ \delta^{4}(p^{\mu} + \sum_{j=2}^n p^{\mu}_j - \sum_{k=1}^m p_k^{\mu})
264: \nonumber \\
265: && \times [\tilde{f}_i(x,p) \prod_{k=1}^m f_k(x,p_k) \prod_{j=2}^n \tilde{f}_j(x,p_j)
266: - f_i(x,p) \prod_{j=2}^n f_j(x,p_j) \prod_{k=1}^m \tilde{f}_k(x,p_k)].
267: \label{icollm}
268: \eea
269: In Eq. (\ref{icollm}) the quantities $\tilde{f}$ denote Pauli-blocking
270: or Bose-enhancement factors as
271: \begin{equation}
272: \tilde{f} = 1 + \eta f
273: \end{equation}
274: with $\eta$ = 1 for bosons and $\eta = -1$ for fermions,
275: respectively. The indices $\nu$ and $\lambda$
276: stand for the set of discrete quantum numbers in the initial
277: (except for particle $i$) and final states, respectively, and $N_n^m$
278: denotes a statistical factor that takes into account the number of
279: identical fermions and bosons in the initial and final states. In
280: the following we will assume that the transition probabilities
281: $W_{n,m}$ are evaluated with respect to asymptotically free
282: antisymmetrized fermion many-body
283: states and symmetrized meson states such that $N_n^m$ = 1.
284:
285: In the on-shell quasi-particle limit the spectral functions,
286: furthermore, are given by\footnote{In general the spectral
287: functions for fermions differ from that of bosons \cite{Casju};
288: here baryons and antibaryons with different spins are treated as
289: independent Klein-Gordon particles.}
290: \be
291: \label{spec} A_i(x,p) = 2 \pi \ \delta(p^2 - M_i^2) \ee which
292: leads to the 2-body collision integral for particle $i$ (in case of fermions) as:
293: \bea
294: \lefteqn{ I_{coll}^i [2 \leftrightarrow 2] = \frac{1}{2} \sum_j \sum_{k,l}
295: \frac{1}{(2 \pi)^{9}} \int \frac{d^{3} p_{2}}{2 E_2} \,
296: \frac{d^{3} p_{3}}{2 E_3} \, \frac{d^{3} p_4}{2 E_4} } \nonumber
297: \\
298: && \times W_{2,2}(p, p_2; i,j \mid p_3,p_4; k, l) \
299: (2\pi)^4 \ \delta^{4}(p^{\mu} + p^{\mu}_2 - p_3^{\mu} - p_4^{\mu} )
300: \nonumber \\
301: && \times [ f_k(x,p_3) f_l(x,p_4)(1 - f_i(x,p))(1 - f_j(x,p_2))
302: \nonumber \\
303: && - f_i(x,p) f_j(x,p_2)(1 - f_k(x,p_3))(1-f_l(x,p_4))],
304: \label{icoll2a} \eea where the energy $p^0_k$ is rewritten as
305: $E_k$. Note that the $p_0=E_1$ integration over the spectral
306: function $A_i(x,p)$ appears on both sides of the transport
307: equation and cancels out in the limit $\Gamma_i \rightarrow$ 0.
308: The on-shell version of the collision integral
309: (\ref{icollm}) then reads
310:
311: \bea \lefteqn{I_{coll}^i [n \leftrightarrow m] = \frac{1}{2} \sum_\nu
312: \sum_{\lambda} \left( \frac{1}{(2 \pi)^{3}} \right)^{n+m-1} \int
313: \prod_{j=2}^n \frac{d^{3} p_j}{2 E_j} \, \prod_{k=1}^m
314: \frac{d^{3} p_{k}}{2E_k} \, } \nonumber \\
315: && \times W_{n,m}(p, p_j;i,\nu \mid p_k; \lambda) \
316: (2\pi)^4 \ \delta^{4}(p^{\mu} + \sum_{j=2}^n p^{\mu}_j - \sum_{k=1}^m p_k^{\mu})
317: \nonumber \\
318: && [\tilde{f}_i(x,p) \prod_{k=1}^m f_k(x,p_k) \prod_{j=2}^n \tilde{f}_j(x,p_j)
319: - f_i(x,p) \prod_{j=2}^n f_j(x,p_j) \prod_{k=1}^m \tilde{f}_k(x,p_k)].
320: \label{icollma}
321: \eea
322: For large times $(t \rightarrow \infty$) all collision integrals
323: vanish, which implies that 'gain' and 'loss' terms become equal in
324: magnitude.
325:
326: The number of reactions in the covariant 4-volume $d^3r dt$ = $dV
327: dt$ is obtained by dividing the gain and loss terms in the
328: collision integrals by the energy $p_0=E_1$, integrating over
329: $d^3p/(2\pi)^3$ and summing over the discrete quantum numbers $i$.
330: For the case of fermion two-body collisions this gives (using $p=p_1$) for
331: the 'loss' term
332:
333: \bea \lefteqn{ \frac{d N_{coll} [2 \rightarrow 2]}{dt dV} =
334: \sum_{i,j} \sum_{k,l} \frac{1}{(2 \pi)^{12}} \int \frac{d^{3}
335: p_{1}}{2 E_1} \
336: \frac{d^{3} p_{2}}{2 E_2} \,
337: \frac{d^{3} p_{3}}{2 E_3} \, \frac{d^{3} p_4}{2 E_4} }
338: \nonumber \\
339: && \times W_{2,2}(p_1, p_2; i,j \mid p_3,p_4; k, l) \
340: (2\pi)^4 \ \delta^{4}(p^{\mu}_1 + p^{\mu}_2 - p_3^{\mu} - p_4^{\mu} )
341: \nonumber \\
342: && \times [ f_i(x,p_1) f_j(x,p_2)(1 - f_k(x,p_3))(1-f_l(x,p_4))].
343: \label{icoll2c} \eea
344: In case of $n \rightarrow m$ processes this leads to
345: \bea
346: \lefteqn{\frac{d N_{coll} [n \rightarrow m]}{dt dV} =
347: \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2 \pi)^{3}}
348: \right)^{n+m} \int \prod_{j=1}^n \frac{d^{3} p_j}{2 E_j} \,
349: \prod_{k=1}^m \frac{d^{3} p_{k}}{2E_k} \, } \nonumber \\
350: && \times W_{n,m}( p_j; i,\nu \mid p_k; \lambda) \
351: (2\pi)^4 \ \delta^{4}( \sum_{j=1}^n p^{\mu}_j - \sum_{k=1}^m p_k^{\mu})
352: ( \prod_{j=1}^n f_j(x,p_j) \prod_{k=1}^m \tilde{f}_k(x,p_k))
353: \label{icollmc} \eea
354: and in case of $m \rightarrow n$ processes to
355: \bea \lefteqn{\frac{d N_{coll} [m \rightarrow n]}{dt dV} =
356: \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2
357: \pi)^{3}} \right)^{n+m} \int \prod_{j=1}^n \frac{d^{3} p_j}{2
358: E_j} \, \prod_{k=1}^m \frac{d^{3} p_{k}}{2E_k} \, } \nonumber
359: \\
360: && \times W_{n,m}( p_j; i,\nu \mid p_k; \lambda) \
361: (2\pi)^4 \ \delta^{4}( \sum_{j=1}^n p^{\mu}_j - \sum_{k=1}^m p_k^{\mu})
362: ( \prod_{k=1}^m f_k(x,p_k) \prod_{j=1}^n
363: \tilde{f}_j(x,p_j)) .
364: \label{icollmd} \eea For the phase-space configurations of
365: interest in this study the Pauli-blocking or Bose-enhancement
366: terms $\tilde{f_k}$ are $\approx 1$, which implies to replace the
367: quantum statistical ensembles by classical ones. In this limit
368: the integrals over the final momenta can be carried out
369: provided that the transition probabilities $W_{n,m}$ do not
370: sensitively depend on the final momenta $p_k$. Employing the
371: definition of the $n-body$ phase-space integrals for total
372: 4-momentum $P^\mu$ \cite{Byckling},
373: \be
374: \label{RN} R_n(P^\mu;M_1,..,M_n) = \left( \frac{1}{(2 \pi)^{3}}
375: \right)^{n} \int \prod_{k=1}^n \ \frac{d^{3} p_{k}}{2E_k} (2\pi)^4
376: \ \delta^{4}(P^{\mu} - \sum_{j=1}^n p^{\mu}_j ) , \ee one obtains
377: the recursion relation \cite{Byckling}
378: \be
379: R_n(P^\mu,M_1,..,M_n) = \frac{1}{(2 \pi)^{3}} \int
380: \frac{d^{3} p_{n}}{2E_n} R_{n-1}(P^{\mu}- p^\mu_n; M_1,..,M_{n-1}). \ee
381: Note, that the phase-space integrals are of dimension
382: GeV$^{2n-4}$ or (1/fm)$^{2n-4}$. Inserting (\ref{RN})
383: this gives in case of $n \rightarrow m$
384: processes \bea \lefteqn{\frac{d N_{coll} [n \rightarrow m]}{dt dV}
385: = \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2 \pi)^{3}}
386: \right)^{n} \int \left ( \prod_{j=1}^n \frac{d^{3} p_j}{2 E_j}
387: \right ) \, } \nonumber \\
388: && \times W_{n,m}(P) \
389: R_m(P^{\mu}=\sum_{j=1}^n p^{\mu}_j;M_1,..,M_m)
390: \prod_{j=1}^n f_j(x,p_j) \nonumber \\
391: && = \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2 \pi)^{3}}
392: \right)^{n} \int \left ( \prod_{j=1}^n d^{3} p_j \right ) \,
393: P(n \rightarrow m)_{i,\nu}^\lambda \
394: \left ( \prod_{j=1}^n f_j(x,p_j) \right ) ,
395: \label{icollm4} \eea where $M_1,..,M_m$ stand for the masses in
396: the final state, and in case of $m \rightarrow n$ processes \bea
397: \lefteqn{\frac{d N_{coll} [m \rightarrow n]}{dt dV} =
398: \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2
399: \pi)^{3}} \right)^{m} \int \left ( \prod_{k=1}^m \frac{d^{3}
400: p_{k}}{2E_k} \right ) \, } \nonumber \\
401: && \times W_{n,m}(P) \
402: R_n(P^{\mu} =\sum_{k=1}^m p_k^{\mu};M_1,...,M_n)
403: \left ( \prod_{k=1}^m f_k(x,p_k) \right ) \nonumber \\
404: && =
405: \sum_{i,\nu} \sum_{\lambda} \left( \frac{1}{(2
406: \pi)^{3}} \right)^{m} \int \left (\prod_{k=1}^m d^{3}
407: p_{k} \right ) \ P(m \rightarrow n)_\lambda^{i,\nu} \
408: \left ( \prod_{k=1}^m f_k(x,p_k) \right ) .
409: \label{icollm5} \eea For fixed sets of quantum numbers $(i,\nu)$
410: and $\lambda$ in the initial and final states this relates the
411: integrands $P(n \rightarrow m)$ in (\ref{icollm4}) and $P( m
412: \rightarrow n)$ in (\ref{icollm5}) for individual scatterings as
413: (dropping the indices for quantum numbers):
414: \be
415: \label{Ratio} \frac{P(m \rightarrow n)}{P (n \rightarrow m)} =
416: \left[ \prod_{k=1}^m \frac{1}{2E_k}\right] \, \left[ \prod_{j=1}^n
417: {2 E_j}\right] \frac{R_n(P^{\mu} =\sum_{k=1}^n
418: p_k^{\mu};M_1,...,M_n)}{ R_m(P^{\mu}=\sum_{j=1}^m
419: p^{\mu}_j;M_1,..,M_m)}, \ee if $W_{n,m}$ essentially depends only
420: on the invariant energy $\sqrt{s} = \sqrt{P^2}$. Note, that the
421: r.h.s. of (\ref{Ratio}) is in units of GeV$^{3(n-m)}$ or
422: fm$^{3(m-n)}$ such that a factor $(dV)^{n-m}$ is needed to
423: interpret the quantities as relative 'probabilities'.
424:
425: Thus, once the transition probabilities $W_{n,m}$ are known as a
426: function of $\sqrt{s}$ for a given set of quantum numbers, the
427: integrand $P(n \rightarrow m)$ in (\ref{icollm4}) is determined by
428: phase space and the backward reactions in (\ref{icollm5}) are
429: fixed by Eq. (\ref{Ratio}).
430:
431:
432: \subsection{Antibaryon annihilation and recreation}
433: In the following the processes $B \bar{B} \leftrightarrow m \
434: mesons$ are discussed, which are of relevance for annihilation of
435: antibaryons on baryons and the recreation of $B\bar{B}$ pairs by
436: $m \ meson$ interactions. The 4-differential collision rate for
437: baryon-antibaryon annihilation $(1+2 \rightarrow 3, ..,m+2)$ then
438: is given by \bea \lefteqn{\frac{d N_{coll} [B\bar{B} \rightarrow m
439: \ mesons]}{dt dV} = \sum_{i,j} \sum_{\lambda_m} \left(
440: \frac{1}{(2 \pi)^{3}} \right)^{2} \int \frac{d^{3} p_1}{2 E_1}
441: \, \frac{d^{3} p_2}{2 E_2} \,} \nonumber \\
442: && \times W_{2,m}(P= p_1+p_2;i,j;\lambda_m) \
443: R_m(P^\mu;M_3,..,M_{m+2})
444: f_i(x,p_1) f_j(x,p_2).
445: \label{ic1}
446: \eea
447: The integrand is related to the annihilation cross section
448: $\sigma_{ann.}(\sqrt{s})$ for a baryon-antibaryon pair with quantum numbers $i,j$ as
449: \cite{Byckling}
450: $$\sum_m \sum_{\lambda_m} \ W_{2,m}(p_1 + p_2;i,j;\lambda_m) \
451: R_m(p_1^\mu+ p_2^\mu;M_3,..,M_{m+2}) $$
452: \be
453: \label{cross}
454: = 2 \sqrt{\tilde{\lambda}(s,M_1^2,M_2^2)}\ \sigma_{ann.}(\sqrt{s}) =
455: 4 E_1 E_2 \ v_{rel} \ \sigma_{ann.}(\sqrt{s})
456: \ee
457: with the Lorentz-invariant relative velocity
458: \cite{Byckling,Lang93}
459: \be
460: \label{vrel} v_{rel}= \frac{\sqrt{\tilde{\lambda}
461: (s,M_1^2,M_2^2)}}{2 E_1 E_2}, \ee involving
462: \be
463: \tilde{\lambda}(x,y,z) = (x-y-z)^2 - 4 yz.
464: \ee
465: In (\ref{cross}) the sum runs over the final meson multiplicity
466: ($m$ $\approx$ 2, ... , 9) in the
467: final state and over $\lambda_m$ which denotes all discrete quantum
468: numbers of the final mesons for given multiplicity $m$.
469:
470: Note, that by summing (\ref{ic1}) additionally over $m$, but
471: keeping the quantum numbers $i,j$ fixed, one arrives at
472: \bea
473: \frac{d N_{coll}^{i,j}}{dt dV} =
474: \frac{1}{(2 \pi)^{6}}
475: \int d^{3} p_1\, d^{3} p_2 \ v_{rel}(p_1,p_2) \
476: \sigma_{ann}(\sqrt{s}) \
477: f_i(x,p_1) f_j(x,p_2).
478: \label{ic1b} \eea If the product of the relative velocity and the
479: cross section, i.e. $v_{rel} \sigma_{ann}$, is approximately constant (see below) the
480: integrals over the momenta in (\ref{ic1b}) give the classical
481: Boltzmann limit \bea \frac{d N_{coll}^{i,j}}{dt dV} = \ < v_{rel}
482: \ \sigma_{ann} > \ \rho_i(x) \rho_j(x), \label{ic1c} \eea where
483: $\rho_i(x)$ is the density of the hadron with quantum numbers $i$.
484:
485: The number of reactions per volume and time for the back processes
486: is then given by ($\lambda_m= k_1,..,k_m)$
487: \bea \lefteqn{\frac{d
488: N_{coll} [m \ mesons \rightarrow B\bar{B}]}{dt dV} = \sum_{i,j}
489: \sum_{\lambda_m} \left( \frac{1}{(2 \pi)^{3}} \right)^{m} \int
490: \left (\prod_{k=3}^{m+2} \frac{d^{3} p_{k}}{2E_k} \right ) \, }
491: \nonumber
492: \\
493: && \times W_{2,m}(\sqrt{s};i,j,\lambda_m) \
494: R_2(P^\mu=\sum_{k=3}^{m+2} p_k^\mu;M_1,M_2)
495: \left ( \prod_{k=3}^{m+2} f_k(x,p_k) \right ) ,
496: \label{ic5}
497: \eea
498: assuming $W_{2,m}(\sqrt{s};i,j,\lambda_m)$ to depend only on the available
499: energy $\sqrt{s}$ and conserved quantum numbers.
500:
501: To proceed further, some simplifying assumptions have to be
502: invoked to lead to a tractable problem for antibaryon annihilation
503: and production. Experimentally, the differential multiplicity in the pions from
504: $p \bar{p}$ annihilation at low $\sqrt{s}$ above threshold can be described as
505: \be
506: \label{pion5} P(N_\pi) \approx \frac{1}{\sqrt{2\pi} D} \exp(-
507: \frac{(N_\pi - <N_\pi >)^2}{2 D^2}) \ee with an average pion
508: multiplicity of $<N_\pi> \approx$ 5 and $D^2$ = 0.95 \cite{Dover}.
509:
510: This observation is reminiscent of flavor rearrangement processes
511: in the $B\bar{B}$ annihilation reaction to vector mesons and
512: pseudoscalar mesons, e.g. $\rho + \rho + \pi$ or $\omega + \omega
513: + \pi^0$, where the $\rho$ and $\omega$ 'later' decay to 2 or 3
514: pions, respectively. In this picture the $\rho + \rho + \pi$ final
515: channel in $p\bar{p}$ annihilation is the dominant process leading
516: finally to 5 pions. Alternatively, the $\omega + \omega + \pi^0$
517: channel leads to 7 pions in the final channel which will appear on
518: the scale of the $\omega$-meson lifetime. Three pions are obtained
519: in the direct 3 pion decay which, however, is substantially
520: suppressed at higher $\sqrt{s}$ due to spin multiplicities (see
521: below).
522:
523: For the problem of interest we thus employ a quark rearrangement
524: model for $B\bar{B}$ annihilation to 3 mesons as illustrated in
525: Fig. \ref{bild1}, where the final mesons $M_i$ may be pseudoscalar
526: or vector mesons, i.e ($\pi, K, \eta$) or $(\rho, \omega, K^*,
527: \phi)$, respectively. In this model there is no creation of an
528: $s\bar{s}$ pair for nucleon-antinucleon annihilation due to the
529: conservation of constituent quark flavors. As can be extracted
530: from the detailed experimental cross sections given in Ref.
531: \cite{LB}, such processes are suppressed by more than an order of
532: magnitude. In principle, such channels can additionally be
533: included in the model, however, the backward reactions then are
534: also suppressed by the same relative factor according to detailed
535: balance. For the study of interest such 'small' channels are
536: discarded.
537:
538:
539: In the following, the quantum numbers denoted by $\lambda_m$ will
540: be separated into different channels $c$, that can be
541: distinguished by their mass decomposition, and degenerate quantum
542: numbers such as spin multiplicities and isospin projections. In
543: the latter sense the sum over the final quantum numbers
544: $\lambda_m$ in (\ref{ic1}), (\ref{ic5}) then includes a sum over
545: the mass partitions $c=(M_3,M_4,M_5)$, a sum over the spins of
546: the mesons and a sum over all isospin quantum numbers, that
547: are compatible with charge conservation in the transition. The
548: probability for a channel
549: $c= (M_3,M_4,M_5)$ then reads
550: \be \label{model}
551: P_c(\sqrt{s};M_3,M_4,M_5) = N_3(\sqrt{s})\ R_3(\sqrt{s};M_3,M_4,M_5)\
552: N_{fin}^c,
553: \ee
554: where the number of 'equivalent' final states in the channel $c$ is given by
555: \be
556: \label{model1} N_{fin}^c = (2s_3+1)(2s_4+1)(2s_5+1)
557: \frac{F_{iso}}{N_{id}!} . \ee
558: In (\ref{model1})
559: $s_j$ denote the spins of the final mesons, $F_{iso}$ is the
560: number of isospin projections compatible with charge conservation
561: while $N_{id}$ is the number of identical mesons in the final
562: channel (e.g. $N_{id}$ = 3 for the $\pi \pi \pi$ final channel).
563: This combinatorial problem for the final number of states
564: $N_{fin}^c$ is of finite dimension and easily tractable
565: numerically. For each mass partition $c=(M_3,M_4,M_5)$ the decay
566: probability then is given by the 3-body phase space and the
567: allowed number of final states $N_{fin}^c$ since the absolute
568: normalization -- described by $N_3(\sqrt{s})$ -- is fixed by the
569: constraint $\sum_c P_c = 1$.
570:
571: As an example let us consider the problem of nucleon-antinucleon
572: annihilation, where the following final meson channels contribute,
573: \be \label{index} (1) \ \pi \pi \pi \ (2) \ \pi \pi \rho \ (3) \
574: \pi \pi \omega \ (4)\ \pi \rho \rho \ (5)\ \pi \rho \omega \ (6) \
575: \pi \omega \omega , \ee excluding 3 vector mesons in the final
576: channel. According to (\ref{model}) the distribution in the final
577: number of pions (including the explicit vector meson decays to
578: pions) can be evaluated as a function of $\sqrt{s}$ since it only
579: depends on the phase space and the number of possible final states
580: $N_{fin}^c$ in each channel $c$. The numerical results are
581: displayed in Fig. \ref{bild2}
582: for 2.3 GeV $\leq
583: \sqrt{s} \leq$ 4 GeV in comparison to the parametrization
584: (\ref{pion5}) (solid line). Here the horizontal bars indicate the
585: range of $N_\pi$-pion probabilities
586: when varying $\sqrt{s}$ from 2.3 to 4 GeV. Obviously, the simple
587: phase-space model (\ref{model})
588: is in a fair agreement with the experimental observation. For
589: related or more extended models for $p\bar{p}$ annihilation the
590: reader is referred to Ref. \cite{Dover}.
591:
592: For the backward reactions, i.e. the 3 meson fusion to a
593: $B\bar{B}$ pair, the quarks and antiquarks are redistributed in a
594: baryon and antibaryon, respectively, incorporating the baryons $N,
595: \Delta, \Lambda, \Sigma, \Sigma^*, \Xi, \Xi^*, \Omega$ as well as
596: their antiparticles. In line with (\ref{model}) the relative
597: population of states (with the same quark content) is determined
598: by phase space, i.e. $$ P_{c'}(\sqrt{s};M_1,M_2) = N_2(\sqrt{s}) \
599: R_2(\sqrt{s};c'=(M_1,M_2))\ (2s_1+1)(2s_2+1) = $$ \be \label{back}
600: N_2(\sqrt{s}) \ R_2(\sqrt{s};M_1,M_2) \ N_{B}^{c'}, \ee where
601: $N_B^{c'}$ now denotes the number of final states for the
602: particular mass channel $c'$ in the backward reaction. The
603: absolute normalization $N_2(\sqrt{s})$ is fixed again by the
604: constraint $\sum_{c'} P_{c'}$ = 1.
605:
606: As an example consider the reactions $\pi^- \pi^+ \pi^-$ or $\pi^- \rho^+ \pi^-$
607: or $\pi^- \rho^+ \rho^-$ (and isospin combinations), i.e.
608: $\bar{u}d + \bar{d}u + \bar{u}d \rightarrow (\bar{u} \bar{u}
609: \bar{d}) + (u d d)$: here the final states may be either $\bar{p}
610: + n$, $\bar{\Delta}^- + n$, $\bar{p} + \Delta^0$ or $\bar{\Delta}^- + \Delta^0$
611: within the Fock space considered. Note,
612: that the final states with a $\Delta$-resonance are favored due to the spin
613: factors in (\ref{back}), however, somewhat suppressed by the 2-body
614: phase-space integral $R_2(\sqrt{s})$ for low $\sqrt{s}$.
615:
616:
617: One is thus left with the $B\bar{B}$ annihilation problem
618: \bea \lefteqn{ \frac{d N_{coll}
619: [B\bar{B} \rightarrow 3 \ mesons]}{dt dV} = \sum_{c}
620: \sum_{c'} \frac{1}{(2 \pi)^{6}} \int
621: \frac{d^{3} p_1}{2 E_1} \, \frac{d^{3} p_2}{2 E_2} \ W_{2,3}(\sqrt{s})} \nonumber
622: \\
623: && \times N_3(\sqrt{s}) \
624: R_3(p_1 + p_2;c=(M_3,M_4,M_5))\ N_{fin}^c
625: \ f_i(x,p_1) f_j(x,p_2),
626: \label{ic6} \eea where $(M_1, M_2)$ denote the baryon and
627: antibaryon masses in the channel $c'$ and $(M_3,M_4,M_5)$ the
628: final meson masses in the channel $c$. Eq. (\ref{ic6}) can be
629: rewritten as \bea \lefteqn{\frac{d N_{coll} [B\bar{B} \rightarrow
630: 3 \ mesons]}{dt dV} = } \nonumber \\ && \sum_{c} \sum_{c'}
631: \frac{1}{(2 \pi)^{6}} \int d^3p_1 \ d^3p_2 \
632: P^{2,3}_{cc'}(\sqrt{s}) \ f_i(x,p_1) f_j(x,p_2) \label{ic6b}
633: \eea with the channel probabilities
634: \be
635: P^{2,3}_{cc'}(\sqrt{s}) = \frac{1}{4E_1 E_2} \,
636: W^{2,3}(\sqrt{s}) \ N_3(\sqrt{s}) \
637: R_3(p_1 + p_2;(M_3,M_4,M_5))\ N_{fin}^c .
638: \label{p2}
639: \ee
640: Note, that by construction we have
641: \be
642: \sum_c P^{2,3}_{cc'}(\sqrt{s}) = \frac{1}{4E_1 E_2} \,
643: W^{2,3}(\sqrt{s}) = v_{rel} \
644: \sigma_{ann}(\sqrt{s})_{c'},
645: \label{p3}
646: \ee
647: where $v_{rel}$ denotes the relative velocity (\ref{vrel}) and
648: $\sigma_{ann}(\sqrt{s})_{c'}$ is the total annihilation cross
649: section for $B\bar{B}$ pairs of channel $c'$.
650:
651: The backward invariant collision rate is given by
652: \bea \lefteqn{\frac{d N_{coll} [3 \ mesons \rightarrow B\bar{B}]}{dt dV} =
653: \sum_{c} \sum_{c'} \frac{1}{(2
654: \pi)^{9}} \int \left (\prod_{k=3}^5 \frac{d^{3} p_{k}}{2E_k}
655: \right ) \ W_{2,3}(\sqrt{s})} \nonumber \\
656: && \times N_2(\sqrt{s}) \
657: R_2(\sum_{k=3}^5 p_k;c'=(M_1,M_2)) \ N_B^{c'}
658: \left ( \prod_{k=3}^5 f_k(x,p_k) \right ) .
659: \label{ic7}
660: \eea
661: Using Eq. (\ref{Ratio}), the relation (\ref{cross}) for 3 mesons in the final state
662: and (\ref{p3}) one arrives at
663:
664: \bea \lefteqn{\frac{d N_{coll} [3 \ mesons \rightarrow B\bar{B}]}{dt dV} =
665: \sum_{c} \sum_{c'} \frac{1}{(2
666: \pi)^{9}} \int
667: \frac{d^{3} p_{3}}{2E_3} \, \frac{d^{3} p_{4}}{2E_4} \, \frac{d^{3} p_{5}}{2E_5} \,
668: 4 E_1 E_2 \ }
669: \nonumber \\
670: && \times v_{rel}\ \sigma(\sqrt{s})_{c'} \
671: \frac{N_2(\sqrt{s})}{N_3(\sqrt{s})}
672: \frac{R_2(P^\mu;c'=(M_1,M_2))}{R_3(P^\mu ;c=(M_3,M_4,M_5))} \frac{N_B^{c'}}{N_{fin}^c} \nonumber \\
673: && \hspace{3cm} \times f_3(x,p_3) f_4(x,p_4) f_5(x,p_5)
674: \label{final}
675: \eea
676: for the backward reaction $3+4+5 \rightarrow 1+2$.
677: Eq. (\ref{final}) can now be rewritten as \bea \lefteqn{\frac{d
678: N_{coll} [3 \ mesons \rightarrow B\bar{B}]}{dt dV} = } \nonumber
679: \\ &&
680: \sum_{c} \sum_{c'} \frac{1}{(2
681: \pi)^{9}} \int d^3 p_3 \ d^3p_4 \ d^3p_5 \
682: P^{3,2}_{cc'}(\sqrt{s}) \
683: f_3(x,p_3) f_4(x,p_4) f_5(x,p_5)
684: \label{finalb}
685: \eea
686: with the 'transition integrand'
687: \be
688: P^{3,2}_{cc'}(\sqrt{s}) =
689: \frac{ E_1 E_2}{2 E_3 E_4 E_5} v_{rel}\ \sigma(\sqrt{s})_{c'} \ \frac{N_2(\sqrt{s})}{N_3(\sqrt{s})}
690: \frac{R_2(P;c'=(M_1,M_2))}{R_3(P;c=(M_3,M_4,M_5))} \ \frac{N_B^{c'}}{N_{fin
691: }^c},
692: \label{finalc}
693: \ee
694: which is of dimension GeV$^{-3}$ or fm$^3$.
695:
696:
697:
698: \section{Numerical implementation}
699: For a reformulation of the 'transition integrands' (specified in (\ref{finalc})) in a
700: test-particle representation one has to recall
701: that the average density of a meson with quantum numbers $k$ is obtained
702: by integration over momentum as:
703: \be
704: n_k(x) = \frac{1}{(2 \pi)^{3}} \int d^3p \ f_k(x,p), \ee where
705: e.g. charge, strange flavor content, total spin and spin
706: projection are specified by the discrete quantum number $k$. The
707: conversion formula thus reads:
708: \be
709: \label{conv}
710: \frac{1}{(2 \pi)^{3}} \int d^3p \ f_k(x,p) \rightarrow
711: \frac{1}{dV} \sum_{i \ \epsilon \ dV},
712: \ee
713: where $dV$ is a (small) finite volume and the sum runs over all
714: test particles in the volume $dV$ with quantum numbers $k$. The
715: number of $B \bar{B}$ annihilations in the volume $dV$ during the
716: time $dt$ is thus given by \cite{Lang93}
717: \be
718: \label{twobody}
719: N_{B \bar{B}} = \frac{dt}{dV} \sum_{i,j \ \epsilon \ dV}
720: v_{rel}(i,j) \sigma_{ann}(\sqrt{s}_{i,j})
721: \ee
722: with the invariant energy squared
723: \be
724: s_{i,j} = (p_1 + p_2)^2, \ee where $p_1, p_2$ denote the 4-momenta
725: of the colliding $B\bar{B}$ pair. The relative velocity
726: $v_{rel}(i,j)$ is given by (\ref{vrel}) while the annihilation
727: cross section $\sigma_{ann}(\sqrt{s})$, furthermore, has to be
728: specified for all baryon-antibaryon pairs. This cross section is
729: rather well known for nucleon-antinucleon reactions
730: \cite{Dover,PDG,LB}, however, the channels involving $\Lambda,
731: \Sigma, \Sigma^*, \Xi, \Xi^*, \Omega^-$ baryons or their
732: antiparticles are not available experimentally by now and have to
733: be modeled to some extent.
734:
735: For guidance we recall that the product $v_{rel} \
736: \sigma_{ann}(\sqrt{s}) \approx$ 50 mb for a wide range of
737: energies $\sqrt{s}$ in case of $p\bar{p}$ annihilation. This is
738: demonstrated explicitly in Fig. \ref{bild1b}, where the
739: experimental annihilation cross section for $p\bar{p}$ from Ref.
740: \cite{LB} is compared to the approximation
741: \begin{equation}
742: \label{appann} \sigma_{ann} (\sqrt{s}) = \frac{50 [mb]}{v_{rel}} ,
743: \end{equation}
744: which holds well in the dynamical range of interest.
745: We thus can adopt the Boltzmann limit (\ref{ic1c}) to estimate the
746: $B\bar{B}$ annihilation time at nucleon density $\rho$ as
747: \begin{equation}
748: \label{estimate2} \tau_{ann.} \approx (5 fm^2 \rho)^{-1} \approx
749: 1.2 \ \frac{\rho_0}{\rho}\ [fm/c].
750: \end{equation}
751:
752: In this work we will proceed with dynamical calculations in the
753: strangeness sector $S$=0, thus essentially addressing the
754: $\bar{p}$ abundancies in relativistic nucleus-nucleus collisions.
755: In this case the approach formulated above does not involve any
756: new parameter or cross section; it is just an extension of the HSD
757: approach \cite{Cass99,Ehehalt,Geiss} to include the $B\bar{B}
758: \leftrightarrow$ 3 meson reactions by detailed balance.
759:
760: The number of backward reactions by 3 mesons in the test-particle
761: picture in the volume $dV$ and time $dt$ according to
762: (\ref{final}) for a given mass channel $c'$ is given by $$N_{3
763: meson} = \frac{dt}{dV dV} \sum_{i,j,k \ \epsilon \ dV} \frac{E_1
764: E_2}{2 E_i E_j E_k} v_{rel}(1,2) \sigma(\sqrt{s})_{c'} $$ \be
765: \label{trans}
766: \times \frac{N_2(\sqrt{s})}{N_3(\sqrt{s})}
767: \frac{R_2(\sqrt{s};c'=(M_1,M_2))}{R_3(\sqrt{s};c=(M_i,M_j,M_k))}
768: \frac{N_B^{c'}}{N_{fin}^c} = \sum_{i,j,k \ \epsilon \ dV} \
769: P_{ijk}, \ee where the channel $c$ is defined by the colliding
770: mesons (cf. (\ref{ic5})) and the outgoing channel $c'$ by the
771: $B\bar{B}$ pair with masses $M_1$ and $M_2$ and energies $E_1$ and
772: $E_2$, respectively. In (\ref{trans}) the summation over the
773: mesons in the volume $dV$ is restricted to $i < j < k$ in case of
774: 3 identical mesons (e.g. 3 pions) and to $i < j$ in case of 2
775: identical mesons $i,j$ in order to account for the statistical
776: factor $N_{id}!$ in Eq. (\ref{model1}).
777:
778: Eqs. (\ref{twobody}) and (\ref{trans}) are well suited for a Monte Carlo decision
779: problem, i.e. a transition is accepted if the probability
780: $P_{ijk}$ is larger than some random number in the interval [0,1].
781: One has to assure only, that all $P_{ijk}$ are smaller than 1,
782: which -- for a fixed volume $dV$ -- can easily be achieved by
783: adjusting the time-step $dt$. This evaluation of scattering
784: probabilities is Lorentz-invariant and does not suffer from
785: geometrical collision criteria as in the standard approaches
786: \cite{Wolf90,Batko3}, that imply a different sequence of
787: collisions when changing the reference frame by a Lorentz
788: transformation \cite{kodama}. For $2 \leftrightarrow 2$
789: transitions it has first been employed and tested by Lang et al.
790: in Ref. \cite{Lang93}; this method is also implemented in the HSD
791: approach, where it can be used optionally instead of the standard
792: geometrical collision criteria as described e.g. in Refs.
793: \cite{Cassing90,Wolf90,Bertsch88}. The present implementation in
794: this respect is a straight forward generalization of the concept
795: in Ref. \cite{Lang93} to $2 \leftrightarrow 3$ reactions. It is
796: worth to point out that this numerical implementation is a
797: promising way to treat $n \leftrightarrow m$ transitions in
798: transport theories without violating covariance or causality. In
799: case of infinitesimal volumes $dV$ and time steps $dt$ it gives
800: the correct solution to the many-particle Boltzmann equation.
801:
802:
803: For the actual numerical calculations a dynamical time-step size
804: $dt$ is employed which on average amounts to $dt \approx
805: 0.5/\gamma_{cm}$ [fm/c], where $\gamma_{cm}$ is the Lorentz factor
806: in the nucleus-nucleus cms, i.e. $\gamma_{cm} \approx$ 9.3 for
807: collisions of $Pb + Pb$ at 160 A$\cdot$GeV. The volume $dV$ is
808: chosen to be $dV = A \ dz$ with the transverse area $A = 9 fm^2$
809: and $dz = 3/\gamma_{cm}$ [fm]. Variations of these parameters
810: within a factor of 3 do not change the numerical results to be
811: presented below. In order to avoid numerical artefacts, that are
812: due to the finite volume $dV$, a $B\bar{B}$ pair, that has been
813: produced by meson fusion, is not allowed to annihilate on each
814: other again without performing an additional collision in between.
815: On the other hand, the mesons stemming from a particular
816: $B\bar{B}$ annihilation are not allowed to fuse again with the
817: same partners for the backward reaction, if no intermediate extra
818: collision has occured.
819:
820: As a numerical test the number of collisions in a single box of
821: volume 10 fm$^3$ during the time $dt$ =1 fm/c has been calculated
822: with spatially uniform phase-space distributions given by a
823: classical system of hadrons in thermal and chemical equilibrium,
824: i.e.
825: \be
826: \label{equil} f_k(p) = \frac{(2s+1)(2I+1)}{(2\pi)^3}
827: {\exp(-E_k(p)/T)} \ee with $s$ and $I$ denoting spin and
828: isospin, respectively. The particles taken into account are $N,
829: \Delta$ and their antiparticles and $\pi, \rho, \omega$ on the
830: meson side in the strangeness sector $S$=0. The numerical results
831: for the number of $B\bar{B}$ annihilation collisions ($\rightarrow
832: \pi \rho \rho$) are shown in Fig. \ref{bild3} in terms of the
833: dashed line as a function of $\sqrt{s}$, which corresponds to the
834: invariant energy in an individual collision. As can be seen from
835: Fig. \ref{bild3} the dashed line very well coincides with the
836: solid line that corresponds to the energy differential number of
837: $\pi \rho \rho$ collisions for the backward reactions. Thus the
838: numerical scheme employed well reproduces the detailed balance
839: relation in thermal equilibrium for a given channel combination
840: $cc'$. Without explicit representation we mention that the
841: detailed balance relation is fulfilled for all channel
842: combinations $cc'$ specified above.
843:
844:
845: \section{Nucleus-nucleus collisions}
846: \subsection{SPS energies}
847: The most complete set of data on antibaryon production in
848: nucleus-nucleus collisions is available from the NA44 \cite{NA44},
849: NA49 \cite{Na49b} and WA97 \cite{NAxx,NA57} Collaborations for
850: $Pb+Pb$ collisions at SPS energies of 160 A$\cdot$GeV, that allow
851: for stringent tests of the dynamics proposed. Since the extension
852: of the HSD transport approach is described in the previous Section
853: and no new parameters enter into the calculations, we proceed with
854: the actual results.
855:
856: \subsubsection{Cascade calculations}
857: Though antiproton self energies have been found to be important in
858: nucleus-nucleus collisions at subthreshold energies
859: \cite{sibirtsev,Cass99,Ko96}, we start with cascade calculations
860: because the initial invariant energy per nucleon is large compared
861: to the $B\bar{B}$ threshold. However, before coming to the
862: antibaryon abundancies the performance of the HSD transport
863: approach has to be tested in comparison to experimental data for
864: baryons and mesons. Since detailed differential
865: spectra for protons, hyperons, pions and kaons have been presented
866: in Ref. \cite{Geiss} in comparison to the experimental data for
867: central collisions of $Pb+Pb$ at 160 A$\cdot$GeV, we concentrate
868: here on particle abundancies as a function of 'centrality'. The
869: latter is defined by the number of participants $A_{part}$, which
870: is extracted from the transport calculation at impact parameter
871: $b$ as
872: \be
873: A_{part}(b) = 2 A - N_0(b), \ee
874: where $N_0(b)$ is the number of
875: nucleons that were not involved in any hard scattering process.
876:
877: The average number of charged pions $<\pi> = (<\pi^+ + \pi^->)/2$
878: (divided by $A_{part}$) is shown in Fig. 5 as a function of $A_{part}$
879: in comparison to the data
880: from Ref. \cite{Na49b} for $Pb+Pb$ at 160 A$\cdot$GeV. The number
881: of pions per participant nucleon is found to be approximately
882: constant within 10 \% as a function of centrality; there is a
883: slight trend in the data as well as in the HSD calculations (solid
884: line) for an increase of $<\pi>/A_{part}$ for central collisions.
885: However, the HSD transport calculations overestimate the pion
886: abundancy by about 20 \%. Such an overprediction of pion
887: multiplicities in heavy systems occurs in other transport
888: approaches as well \cite{Bravina,Bass98} -- or is even higher --
889: and is not well understood so far. At SIS energies of 1--2
890: A$\cdot$GeV it has been argued in Ref. \cite{Larionov} that a
891: quenching of nucleon resonances at baryon densities $\rho \ge
892: \rho_0$ might be responsible for the relative pion suppression
893: seen experimentally in heavy systems, but it is not yet clear if
894: such a mechanism will also explain the discrepancies at SPS
895: energies. We thus have to keep in mind this overprediction of
896: pions especially when comparing particle ratios (see below).
897:
898: The effect of antibaryon annihilation and reformation by means of
899: meson fusion channels is shown in Fig. 6, where the
900: $<\bar{p}>/<\pi>$ ratio is displayed for $Pb+Pb$ at 160
901: A$\cdot$GeV as a function of $A_{part}$. The dashed line shows a
902: calculation without including annihilation channels in the
903: transport calculation, the dotted line gives the results including
904: the annihilation to mesons while the solid line is obtained when
905: including both, annihilation and meson fusion channels by detailed
906: balance. The first point in Fig. 6 gives the numerical result for
907: $NN$ interactions at $T_{lab}$ = 160 GeV ($A_{part}$ = 2). As
908: shown in Table 1 of Ref. \cite{Geiss}, the description of the HSD
909: approach for $pp$ interactions is well in line with the data on
910: $\pi^+, \pi^0, \pi^-, K^+, K^-, K^0_s$, $\Lambda+\Sigma^0$,
911: $\bar{\Lambda} + \bar{\Sigma}^0$ as well as $p$ and $\bar{p}$
912: multiplicities at SPS energies, such that this point may serve for
913: reference to the particle abundancies in the elementary $NN$
914: interaction. As seen from Fig. 6 the $<\bar{p}>/<\pi>$ ratio
915: slightly drops with centrality for peripheral reactions, however,
916: stays approximately constant for $A_{part} \ge $ 150 when
917: neglecting annihilation. Thus antiprotons in central $Pb+Pb$
918: collisions are produced less frequent than pions by about 30 \%
919: relative to the $NN$ interaction in vacuum. The effect of
920: annihilation (dotted line) sets in already for peripheral
921: reactions and amounts to a factor $\sim$6--7 suppression for
922: central collisions. The latter suppression is sensitive to the
923: formation time $\tau_F$ of the antibaryons and becomes larger
924: (smaller) for shorter (longer) $\tau_F$. In the actual
925: calculations we have used $\tau_F$ = 0.8 fm/c for all hadrons
926: \cite{Geiss}. When including annihilation as well as meson fusion
927: channels by detailed balance, the $<\bar{p}>/<\pi>$ ratio (solid
928: line) becomes again close to the calculation that does not include
929: annihilation nor meson fusion reactions to $B\bar{B}$ (dashed
930: line). Thus on average the annihilation channels are almost
931: compensated by the reproduction channels.
932:
933:
934: In order to get some idea about the dynamical origin of the
935: approximately constant $\bar{p}/\pi$ ratio shown in Fig.
936: \ref{6new} by the solid line the reaction rate $B+\bar{B}
937: \rightarrow mesons$ (dashed histogram in Fig. \ref{bild5} ) is
938: compared to the backward reaction rate (solid histogram in Fig.
939: \ref{bild5}) for a central collision of $Pb+Pb$ at 160
940: A$\cdot$GeV. Fig. \ref{bild5} demonstrates that both rates are
941: comparable within the statistics. Thus an approximate local
942: chemical equilibrium is established very fast between the
943: nonstrange antibaryon degrees of freedom and the nonstrange mesons
944: $\pi, \rho$ and $\omega$. The latter fact is supported by the
945: absolute number of reactions $B\bar{B} \rightarrow mesons$ which
946: is about 4-5 times higher than the final number of antibaryons per
947: event. On the other hand, the number of backward reactions is
948: $\sim$ 96\% of the number of annihilation reactions leading to a
949: small net absorption of the antibaryons produced initially by
950: baryon-baryon ($\sim$ 73\%) or meson-baryon ($\sim$ 27\%)
951: inelastic collisions. Only $\sim 12$\% of the final antibaryons
952: stem from 'hard' baryon-baryon or meson-baryon reactions; the
953: dominant amount of final $\bar{p}$'s ($\sim$ 88\%) are from the 3
954: meson fusion reactions indicating that the memory with respect to
955: the initial 'hard' collision phase is practically lost.
956:
957: One might worry about the sensitivity of these results to the
958: annihilation cross section $\sigma_{ann}(\sqrt{s})$ that so far
959: has been taken to be the free cross section (in the
960: parametrization from Ref. \cite{sibirtsev}). However, this
961: quantity might change in the medium due to screening effects. It
962: is clear that any enhancement of this cross section will lead to
963: an even faster equilibration in the light flavor degrees of
964: freedom and to a more perfect chemical equlibrium. Thus numerical
965: calculations have been performed for central $Pb+Pb$ collisions at
966: 160 A$\cdot$GeV by assuming that $\sigma_{ann}$ is reduced by a
967: factor of 2. This leads to a reduction of the total number of
968: annihilation reactions (and backward reactions) by a factor of 2
969: -- which essentially reduces the numerical statistics -- but
970: leaves the conclusions unchanged. Thus the findings from Figs.
971: \ref{6new} and \ref{bild5} are robust against 'reasonable'
972: modifications of the in-medium transition rates.
973:
974: The numerical abundancies for $K^\pm$ mesons and antiprotons
975: relative to the average charged pion multiplicity $<\pi> = <\pi^+
976: + \pi^->/2$ are displayed in Fig. \ref{bild4} for $Pb+Pb$ at 160
977: A$\cdot$GeV as a function of $A_{part}$. The comparison of the
978: calculations with the data from Ref. \cite{Na49b} indicates that
979: the dependence of the $K^\pm$ multiplicities on $A_{part}$ is
980: roughly met -- except a single point for $K^-$ at rather
981: peripheral reactions -- in line with the earlier analysis in Refs.
982: \cite{Cass99,Geiss}, which concentrated on central collisions in
983: this system. However, the $\bar{p}/\pi$ ratio is lower by about
984: a factor of 2 compared to the data of the NA49 Collaboration that
985: also include the feeddown from $\bar{\Lambda}$ and
986: $\bar{\Sigma}^0$ due to the weak interaction. Since the latter
987: contribution is presently unknown one might either speculate that
988: the $\bar{\Lambda}+ \bar{\Sigma}^0$ abundancy is comparable to the
989: antiproton abundancy or that antiproton self energy effects might
990: be responsible for the experimental observation.
991:
992: \subsubsection{Antiproton self energies}
993: As shown in Refs. \cite{Teis94,sibirtsev,Cass99,Ko96} the
994: production of antiprotons at SIS energies of 1.4--2.1 A$\cdot$GeV
995: as well as in $p+A$ reactions is described by adopting attractive
996: self energies for the antiprotons in the range of -100 to -150 MeV
997: at normal nuclear matter density $\rho_0$. Especially in $p+A$
998: reactions the backward production channels by a couple of mesons
999: are statistically irrelevant as can be easily checked by the
1000: transport model described above. The question thus arises if such
1001: 'established' antiproton self energies for densities 1--3 $\rho_0$
1002: might also be responsible for an enhancement of the $\bar{p}$
1003: yield at SPS energies in the $Pb+Pb$ system.
1004:
1005: To examine this possibility we show in Fig. \ref{bild6} the time
1006: evolution of the baryon-density in a central cylinder of
1007: transverse radius $R_T$ = 5 fm, that has moving boundaries with
1008: the expanding hadronic system in longitudinal direction. The solid
1009: line denotes the average density of 'formed' baryonic states
1010: whereas the dashed line represents the net quark density
1011: $\rho_q/3$ which merges with the baryon density in the later
1012: expansion phase. Both densities have been evaluated in the cms
1013: rapidity interval $|\Delta y|_{cm} \leq$ 1 in order to exclude
1014: spectator nucleons and to gate on midrapidity physics. The
1015: difference between the solid and dashed line in Fig. \ref{bild6}
1016: has to be attributed to 'non-hadronic' states, which in the HSD
1017: approach are quarks and diquarks (as well as their antiparticles)
1018: that constitute the ends of 'strings' or continuum excitations of
1019: the hadrons. As can be seen from Fig. \ref{bild6} the
1020: 'non-hadronic' phase lasts about 2.2 fm/c which is roughly the
1021: diameter of the target ($2 R$) divided by the Lorentz-factor
1022: $\gamma_{cm} \approx $ 9.3 plus the hadron formation time $\tau_F$
1023: = 0.8 fm/c \cite{Cass99},
1024: \be
1025: \label{nonh} \tau_{nonhad.} \approx \frac{2 R}{\gamma_{cm}} +
1026: \tau_F . \ee Then a mixed phase of 'partons' and 'formed' hadrons
1027: comes up which practically ends around 6 fm/c where all continuum
1028: excitations have merged to hadrons or hadronic resonant states.
1029: Quite remarkably, the density of 'formed' baryons is about $2
1030: \rho_0$ at the beginning of the pure hadronic expansion phase. Now
1031: the $B\bar{B}$ annihilation rate (in Fig. \ref{bild5}) starts with
1032: a maximum around 3 fm/c -- when the 'baryons' also start to
1033: hadronize -- along with the
1034: meson fusion reactions that last up to about 8 fm/c. The
1035: characteristic time scale for the decrease of the annihilation/fusion rate is
1036: $\tau_{prod} \approx $1.6 fm/c for central $Pb+Pb$ reactions at
1037: the SPS as extracted from Fig. \ref{bild5} using an exponential ansatz.
1038: During the $B\bar{B}$ production
1039: phase by meson fusion from 3--8 fm/c
1040: the density of baryons drops from $\sim 2.5 \rho_0$ to $\rho_0$.
1041: Employing (\ref{estimate2}) the time scale for
1042: annihilation at these densities changes from 0.5 -- 1.2 fm/c,
1043: which is considerably shorter than the characteristic production
1044: time scale of 1.6 fm/c. The approximate chemical equilibration between mesons and
1045: antibaryons thus is no surprise according to these simple 'classical'
1046: estimates.
1047: On the other hand, the densities of 2.5 $\rho_0$ -- $\rho_0$ are
1048: very similar to the densities probed in heavy-ion
1049: reactions at SIS energies from 1--2 A$\cdot$GeV \cite{Ko96,Cassing90}.
1050:
1051: In order to investigate if $\bar{p}$ self energies might be
1052: responsible for the experimental antiproton yield, a scalar
1053: attractive $\bar{p}$ potential of the form
1054: \be
1055: \label{pot} U_{\bar{p}}(\rho) = - \alpha \frac{\rho}{\rho_0} \ee
1056: is assumed with $\alpha$ = 100 MeV, where $\rho$ denotes the
1057: density of 'formed' baryons. This amounts to produce and propagate
1058: antiprotons with a density-dependent mass $M^* = M_0 +U(\rho)$.
1059: For more technical details the reader is referred to Refs.
1060: \cite{Teis94,sibirtsev}. Note, that when introducing self energies
1061: for particles by averaging the latter over many events one no
1062: longer performs {\it microcanonical} simulations since the field
1063: energy is shared between different events. However, quark flavor
1064: conservation still holds exactly in each event such that the
1065: physical system corresponds to a {\it canonical} ensemble.
1066:
1067: The numerical results for the $\bar{p}/<\pi>$ ratio in $Pb + Pb$
1068: collisions at 160 A$\cdot$GeV are displayed in Fig. \ref{bild7} by
1069: the solid line in comparison to the cascade calculation (dashed
1070: line) and the experimental data from NA49 \cite{Na49b} (full
1071: triangles). It is seen that with increasing centrality or
1072: $A_{part}$ there is a slight enhancement of the $\bar{p}$ yield
1073: for the potential (\ref{pot}), which vanishes for very peripheral
1074: reactions where the average baryon density is very small. The
1075: experimental data, however, are still underestimated
1076: significantly. This also holds true when accounting for the 20\%
1077: overprediction of pions in the transport approach (cf. Fig. 5).
1078:
1079:
1080: \subsubsection{Extrapolation for antihyperon production}
1081: In view of the approximate chemical equilibrium achieved for the
1082: $u,d$ quark sector between mesons and antibaryons -- even for
1083: reduced annihilation cross sections -- one may proceed with
1084: speculations on the strangeness ($s\bar{s}$-quark) sector, where
1085: no experimental data on the annihilation cross sections are
1086: available. However, in case of similar transition matrix elements
1087: squared ($\approx$ 5 fm$^2$) the $Y \bar{N}$ and $\bar{Y} N$
1088: channels, where $Y$ stands for the hyperons ($\Lambda, \Sigma,
1089: \Sigma^*$), will achieve chemical equilibrium with the 3 meson
1090: system as before, however, with a $\pi$ or $\rho, \omega$
1091: exchanged by a $K$ or $K^*,\phi$, respectively (cf. Ref.
1092: \cite{Carsten} for the case of 5 meson reaction channels). A
1093: further step then consists in replacing another pion or $\rho,
1094: \omega$ by a $K$ or $K^*,\phi$ which will bring the $\Xi, \Xi^*$
1095: and $\bar{\Xi}, \bar{\Xi}^*$ in chemical equilibrium with the
1096: $SU(3)_{flavor}$ meson system. Moreover, in case of flavor
1097: rearrangement reactions with 3 strange mesons $K, K^*, \phi$ (or
1098: any combinations of those) the $\Omega, \bar{\Omega}$ system might
1099: also achieve chemical equilibrium. This will imply for the
1100: antibaryon to baryon ratios:
1101: \be
1102: \label{chemr} \frac{\bar{p}}{p} \approx \frac{K^-}{K^+} \
1103: \frac{\bar{\Lambda}}{\Lambda} \approx \left ( \frac{K^-}{K^+}
1104: \right )^2 \ \frac{\bar{\Xi}}{\Xi} \approx \left ( \frac{K^-}{K^+}
1105: \right )^3 \ \frac{\bar{\Omega}}{\Omega} . \ee The relations
1106: (\ref{chemr}) are easily obtained for a {\it grand canonical}
1107: ensemble at 'high' temperature $T$ where Fermi and Bose
1108: distributions coincide with the Boltzmann distribution. In this
1109: limit the ratios of particles to antiparticles (apart from the
1110: temperature $T$) only depend on the chemical potential $\mu_q$ for
1111: light quarks and $\mu_s$ for strange quarks, i.e. $$
1112: \frac{\bar{p}}{p} = \frac{\exp(-(E_{\bar{p}} +3
1113: \mu_q)/T)}{\exp-(E_p -3 \mu_q)/T)} = \exp(-6 \mu_q/T), $$ $$
1114: \frac{\bar{K}}{K} = \frac{\exp(-(E_{\bar{K}} - \mu_s +
1115: \mu_q)/T)}{\exp(-(E_K + \mu_s - \mu_q)/T)} = \exp(2(\mu_s-
1116: \mu_q)/T), \ $$ $$ \frac{\bar{\Lambda}}{\Lambda} =
1117: \frac{\exp(-(E_{\bar{\Lambda}} + \mu_s + 2
1118: \mu_q)/T)}{\exp(-(E_\Lambda - \mu_s - 2 \mu_q)/T)} = \exp(-(2
1119: \mu_s+ 4 \mu_q)/T) \ $$ $$ \frac{\bar{\Xi}}{\Xi} =
1120: \frac{\exp(-(E_{\bar{\Xi}} + 2 \mu_s + \mu_q)/T)}{\exp(-(E_\Xi - 2
1121: \mu_s - \mu_q)/T)} = \exp(-(4 \mu_s+ 2 \mu_q)/T) \ $$
1122: \be
1123: \frac{\bar{\Omega}}{\Omega} = \frac{\exp(-(E_{\bar{\Omega}} + 3
1124: \mu_s )/T)}{\exp(-(E_\Omega - 3 \mu_s )/T)} = \exp(-6 \mu_s/T), \
1125: \ee
1126: where $E_X, E_{\bar{X}}$ denote the energies of particles and
1127: antiparticles, respectively, that are the same when neglecting
1128: self energies. Note, that the relations (\ref{chemr}) also result
1129: from the quark condensation model in Refs.
1130: \cite{Zimanyi1,Zimanyi2}, however, involve a slightly different
1131: physical picture. In the latter approach the mesons, baryons and
1132: antibaryons emerge from an equilibrium QGP state by condensation
1133: under the constraint of quark flavor conservation. In the
1134: microcanonical transport approach discussed here, the relations
1135: (\ref{chemr}) come about due to the strong annihilation of
1136: antibaryons with baryons and the backward flavor rearrangement
1137: channels by detailed balance, i.e. by purely hadronic reaction
1138: channels in the expansion phase of the system.
1139:
1140: It should be pointed out, that the canonical statistical approach
1141: of Ref. \cite{Redlich2} -- involving an additional parameter of
1142: dimension fm$^3$ -- well describes the strange and
1143: multi-strange baryon and antibaryon abundancies in $Pb+Pb$
1144: collisions at SPS energies. Thus the concept of chemical
1145: equilibrium also on the strangeness sector $S$= $\pm 1,\pm 2,\pm 3$ appears
1146: compatible with the experimental observations. However, it is
1147: argued here that the relations (\ref{chemr}) are a consequence of
1148: the strong flavor rearrangement reactions between formed hadrons and
1149: do not signal the presence of a QGP state.
1150:
1151:
1152:
1153: Assuming chemical equilibrium also for baryons and antibaryons
1154: with strangeness then (by Eq. (\ref{chemr})) the $\bar{\Lambda}$
1155: multiplicity is related to the antiproton multiplicity as
1156: \be
1157: \label{antihyp} \bar{\Lambda} \approx \left ( \frac{K^+}{K^-}
1158: \times \frac{\Lambda}{p}\right ) \ \bar{p}. \ee All quantities on
1159: the r.h.s. of (\ref{antihyp}) are known from the transport
1160: calculation, however, only the interacting number of protons have
1161: to be counted in this case since proton spectators have to be
1162: excluded in this balance. A rather save way is to take into
1163: account only particles at midrapidity for $|\Delta y| \leq$ 1,
1164: which then excludes spectator baryons. Thus counting only
1165: particles at midrapidity for the ratio in the brackets in
1166: (\ref{antihyp}) the $\bar{\Lambda}$ abundancy is entirely
1167: determined by particle ratios that have sufficient statistics in
1168: the transport calculation. The resulting $(\bar{p} +
1169: \bar{Y})/<\pi>$ ratio for $Pb + Pb$ at 160 A$\cdot$GeV is shown in
1170: Fig. \ref{bild7} by the open circles with error bars that are due
1171: to particle statistics in the transport calculation. In this case
1172: the experimental multiplicity is reproduced rather well from
1173: peripheral to central collisions (within the error bars)
1174: suggesting the ratio $\bar{\Lambda}/\bar{p} \approx $ 1 for a wide
1175: range of impact parameters. However, a precise experimental
1176: separation of antiprotons from antihyperons will be necessary to clarify the present
1177: ambiguities.
1178:
1179: \subsection{AGS energies}
1180: The experimental information on antibaryon production at AGS
1181: energies is rather scarce and limited to specific rapidity
1182: intervals. Since antibaryon yields are also reduced substantially
1183: as compared to SPS energies this imposes severe constraints on the
1184: statistics in nonperturbative transport calculations. Thus, before
1185: addressing any comparison to experimental data, we show in Fig.
1186: \ref{bild8} the density of 'formed' baryons in a central expanding
1187: cylinder of transverse radius $R_T$ = 5 fm for $Au + Au$ at 11
1188: A$\cdot$GeV in comparison to the net quark density $\rho_q/3$
1189: (dashed line). When comparing to Fig. \ref{bild6} for $Pb + Pb$ at
1190: 160 A$\cdot$GeV roughly the same maximum net quark density is
1191: found for $|\Delta y| \leq$ 1, however, the nonhadronic phase
1192: characterized by (\ref{nonh}), i.e. $\tau_{nonhad.} \approx 6.2$
1193: fm/c, lasts much longer due to the lower Lorentz $\gamma$-factor
1194: $\gamma_{cm} \approx 2.6$. Furthermore, the density of 'formed'
1195: baryons is lower than at SPS energies since most of these hadronic
1196: states rescatter again on baryons in the medium since the
1197: formation time $\tau_F$ = 0.8 fm/c is small compared to the
1198: reaction time roughly given by $2 R/\gamma_{cm}$, where $R$
1199: denotes the radius of the $Au$ nucleus. Thus 'formed' baryons are
1200: reexcited to strings for a couple times during the nucleus-nucleus
1201: collision. This phenomenon has been addressed as 'string matter'
1202: in Ref. \cite{Sahu00} and should not be interpreted as a state of
1203: quark-gluon plasma (QGP).
1204:
1205: A comparison of the HSD transport calculations with the
1206: experimental data of the E866 Collaboration at midrapidity
1207: \cite{AGSall} is presented in Fig. \ref{bild9} as a function of
1208: the participating protons $N_{pp}$ for $Au + Au$ at 11.6
1209: A$\cdot$GeV/c. Whereas the proton to $\pi^+$ ratio is rather well
1210: described as a function of centrality (lower left part) the
1211: $K^+/\pi^+$ and $K^-/\pi^+$ ratios are underestimated
1212: systematically for all centralities when discarding self energies
1213: for the strange hadrons (cf. Refs. \cite{Cass99,Geiss,Cass00a}).
1214: The calculated $K^+/K^-$ ratio from the HSD calculation is 5$\pm
1215: 0.3$ for all centralities rather well in line with the
1216: experimental observation at midrapidity. Since strangeness
1217: conservation is exactly fulfilled in the calculations this
1218: demonstrates that the net production of $s\bar{s}$ quarks by
1219: 'hard' baryon-baryon, baryon-meson and meson-meson reactions is
1220: underestimated in the transport approach as discussed in more
1221: detail in Refs. \cite{Cass99,Geiss}. On the other hand the
1222: antiproton to $\pi^+$ ratio (lower right part) is compatible with
1223: the experimental ratios within the error bars indicating an
1224: approximately constant value of $\sim 2.5-3 \times 10^{-4}$. This
1225: roughly constant $\bar{p}/\pi^+$ ratio is a consequence of an
1226: approximate chemical equilibration as demonstrated in Fig.
1227: \ref{bild10} for a central collision of $Au+Au$ at 11.6
1228: A$\cdot$GeV/c. Here the solid histogram corresponds to the
1229: annihilation rate of antibaryons whereas the dashed histogram
1230: stands for the backward 3 meson fuse rate. Though the statistics
1231: are very limited in this case a net absorption of antibaryons,
1232: i.e. the difference in the time integrals of the annihilation rate
1233: and 3 meson production rate, is still present in the calculations
1234: which amounts to about 20\% for central reactions and about 30\%
1235: for very peripheral reactions of the total number of antibaryons
1236: produced in baryon-baryon or meson-baryon reactions. This
1237: relative net absorption, however, can only be extracted from
1238: transport calculations and is not a measurable quantity
1239: experimentally.
1240:
1241: The E877 Collaboration, furthermore, has observed a sizeable
1242: anti-flow of $\bar{p}$'s in $Au + Au$ reactions at 11.6
1243: A$\cdot$GeV/c \cite{E877} which either indicates a strong
1244: absorption of antiprotons on baryons or the current of comoving
1245: mesons, that fuse to $B\bar{B}$ pairs, and are anticorrelated to
1246: the proton current themselves. The present transport calculations
1247: reproduce these correlations, however, suffer from large
1248: statistical error bars (similar in size to those of the data) such
1249: that an explicit comparison is discarded in this work.
1250:
1251: As mentioned in the introduction, a high ratio of $\bar{\Lambda}$
1252: to $\bar{p}$ of $3.6^{+4.7}_{-1.8}$ has been reported by the E917
1253: Collaboration \cite{AGSnew} for central collisions of $Au + Au$ at
1254: 11.7 A$\cdot$GeV/c that is not understood so far. These data, furthermore,
1255: are in a qualitative agreement with the measurements from the E864/E878
1256: Collaboration \cite{E864n}. To estimate the
1257: $\bar{\Lambda}/\bar{p}$ ratio within the present transport
1258: approach as a function of the centrality of the collision we employ
1259: again the relation (\ref{antihyp}) and invoke
1260: the particle multiplicities at midrapidity $|\Delta y_{cm}| \leq
1261: 1$. The results of these calculations are displayed in Fig.
1262: \ref{bild11} for $Au+Au$ at 11.6 A$\cdot$GeV/c as a function of
1263: the number of participating protons indicating a steady rise of
1264: the ratio with centrality. The hatched area in Fig. \ref{bild11}
1265: demonstrates the uncertainty in the $\bar{\Lambda}/\bar{p}$ ratio
1266: due to the limited statistics. However, the calculations for the
1267: most central collisions do not suggest ratios above 1.4 which is
1268: still slightly out of the range of the number quoted in Ref.
1269: \cite{AGSnew} of $3.6^{+4.7}_{-1.8}$. Thus either new theoretical
1270: concepts and/or much refined data are necessary to unravel this
1271: puzzle.
1272:
1273:
1274:
1275: \section{Summary}
1276: In this work the conventional transport approach for two-body
1277: induced reactions has been extended on the formal level to n-body
1278: $n \leftrightarrow m$ reaction channels employing the principles
1279: of detailed balance. As a specific example of current interest the
1280: baryon-antibaryon annihilation problem in relativistic
1281: nucleus-nucleus collisions has been addressed at AGS and SPS
1282: energies where the meson densities are comparable (at the AGS) or
1283: much larger than the baryon densities (at the SPS) such that
1284: multiple meson fusion reactions as suggested in Refs.
1285: \cite{ko1,Rapp,Carsten} become very probable.
1286:
1287: In order to employ the many-body detailed balance relations (as
1288: addressed in Section 2) a simple phase-space model for antibaryon
1289: annihilation has been presented that is based on flavor
1290: rearrangement channels to pseudoscalar and vector mesons (cf. Fig.
1291: \ref{bild1}). Furthermore, a suitable covariant scheme for the
1292: calculation of such multi-particle reactions has been presented in
1293: Section 3 which is a straight forward extension of the concept
1294: proposed by Lang et al. in Ref. \cite{Lang93}. The method and its
1295: implementation in the HSD transport approach \cite{Cass99,Ehehalt}
1296: has been tested for a homogenuous system of nonstrange hadrons in
1297: thermal and chemical equilibrium (cf. Fig. \ref{bild3}).
1298:
1299: Actual transport calculations have been performed for $Pb + Pb$ at
1300: 160 A$\cdot$GeV and $Au+Au$ at 11.6 A$\cdot$GeV/c, i.e. the most
1301: prominant reactions at the SPS and the AGS, where a couple of
1302: experimental data are available to control the dynamics. It is
1303: found that at both energies the meson fusion reactions are by far
1304: the most dominant production channel for the final antibaryons
1305: (seen experimentally) and that the antibaryons and baryons -- at
1306: least at midrapidity -- come close to local chemical equilibrium
1307: with the mesons. As a consequence the $\bar{p}/\pi$ ratio is
1308: practically independent on the centrality of the collision, i.e.
1309: as a function of $A_{part}$, in line with the experimental
1310: observation at both energies.
1311:
1312: On the other hand, the approximate
1313: chemical equilibration allows to perform rather reliable
1314: extrapolations for the strange and multistrange antibaryon
1315: abundancies on the basis of particle ratios (cf. (\ref{chemr})),
1316: that follow from chemical equilibrium for the mesons and antibaryons.
1317: The latter ratios can be calculated with better statistics in the
1318: nonperturbative approach than the direct strange antibaryon abundancies.
1319: Here a roughly constant
1320: $\bar{\Lambda}/\bar{p}$ ratio of $\approx$ 1 is predicted for
1321: semi-peripheral to central collisions of $Pb+Pb$ at the SPS that
1322: will be controlled soon by experimental data from the NA49
1323: Collaboration. Moreover, a separation of antiprotons from
1324: antihyperons will also allow to investigate the question of
1325: antiproton self energies that enhance the $\bar{p}/\pi$ ratio with
1326: centrality when adopting attractive scalar potentials in line with
1327: the analysis performed at SIS energies of $\sim$ 2 A$\cdot$GeV
1328: \cite{sibirtsev}. At AGS energies of 11.6 A$\cdot$GeV/c the
1329: situation is less clear due to the rather low statistics for
1330: antibaryons, both theoretically and experimentally. Recall that
1331: the $\bar{p}/\pi^+$ ratio is only about $2.5-3 \times 10^{-4}$ at
1332: this energy. The HSD transport calculations for the most central
1333: collisions of $Au+Au$ at the AGS -- assuming (\ref{chemr}) to hold --
1334: give an upper limit of 1.4 for
1335: the $\bar{\Lambda}/\bar{p}$ ratio, which is still slightly out of
1336: the range of the number quoted by the E917 Collaboration
1337: \cite{AGSnew} of $3.6^{+4.7}_{-1.8}$ or related results from Ref. \cite{E864n}.
1338: Thus either new theoretical
1339: concepts and/or much refined data will be necessary to unravel the
1340: antihyperon puzzle at AGS energies.
1341:
1342: \vspace{0.5cm} The author likes to acknowledge continuous and
1343: valuable discussions with C. Greiner. Furthermore, he is indepted
1344: to E. L. Bratkovskaya and C. Greiner for critical suggestions and
1345: a careful reading of the manuscript.
1346:
1347:
1348: \begin{thebibliography}{99}
1349: \bibitem{chamberlain} O. Chamberlain et al., Nuovo Cimento { 3} (1956) 447.
1350: \bibitem{elioff} T. Elioff et al., Phys. Rev. { 128} (1962) 869.
1351: \bibitem{dorfan} D. Dorfan et al., Phys. Rev. Lett. { 14} (1965) 995.
1352: \bibitem{JINR} A. A. Baldin et al., JETP Lett. { 47} (1988) 137.
1353: \bibitem{BEVALAC1} J. B. Carroll et al., Phys. Rev. Lett. { 62} (1989) 1829.
1354: \bibitem{BEVALAC2} A. Shor, V. Perez-Mendez, K. Ganezer,
1355: Phys. Rev. Lett. { 63} (1989) 2192.
1356: \bibitem{KEK} J. Chiba et al., Nucl. Phys. { A 553} (1993) 771c.
1357: \bibitem{GSI} A. Schr\"oter et al., Nucl. Phys. { A 553} (1993) 775c.
1358: \bibitem{Batko91} G. Batko, W. Cassing, U. Mosel, K. Niita,
1359: Phys. Lett. { B 256} (1991) 331.
1360: \bibitem{Cass92} W. Cassing, A. Lang, S. Teis, K. Weber,
1361: Nucl. Phys. { A 545} (1992) 123c.
1362: \bibitem{Faess1} S. W. Huang, G. Q. Li, T. Maruyama, A. Faessler,
1363: Nucl. Phys. { A 547} (1992) 653.
1364: \bibitem{Danielewicz90} P. Danielewicz, Phys. Rev. { C 42} (1990)
1365: 1564.
1366: \bibitem{Weise} E. Hernandez, E. Oset, W. Weise,
1367: Z. Phys. A 351 (1995) 99.
1368: \bibitem{Schaffner91} J. Schaffner, I. N. Mishustin, L. M. Satarov,
1369: H. St\"ocker, W. Greiner, Z. Phys. { A 341} (1991) 47.
1370: \bibitem{Koch} V. Koch, G. E. Brown, C. M. Ko, Phys. Lett. B 265 (1991) 29.
1371: \bibitem{Teis94} S. Teis, W. Cassing, T. Maruyama, U. Mosel,
1372: Phys. Rev. C 50 (1994) 388.
1373: \bibitem{Ko93} G. Q. Li, C. M. Ko, X. S. Fang, Y. M. Zheng,
1374: Phys. Rev. C 49 (1994) 1139.
1375: \bibitem{RQMD} C. Spieles, A. Jahns, H. Sorge, H. St\"ocker,
1376: W. Greiner, Mod. Phys. Lett. A 27 (1993) 2547.
1377: \bibitem{sibirtsev} A. Sibirtsev, W. Cassing, G. I. Lykasov,
1378: M. V. Rzjanin, Nucl. Phys. A 632 (1998) 131.
1379: \bibitem{Cass99} W. Cassing, E. L. Bratkovskaya,
1380: Phys. Rep. 308 (1999) 65.
1381: \bibitem{Ko96} C. M. Ko, G. Q. Li, J. Phys. G 22 (1996) 1673.
1382: \bibitem{ko1} C. M. Ko, X. Ge, Phys. Lett. { B 205} (1988) 195.
1383: \bibitem{ko2} C. M. Ko, L. H. Xia, Phys. Rev. { C 40} (1989) R1118.
1384: \bibitem{Wittmann} R. Wittmann, PhD thesis, Univ. of Regensburg,
1385: 1995; R. Wittmann, U. Heinz, {\it hep-ph}/9509328.
1386: \bibitem{Dover} C. B. Dover, T. Gutsche, M. Maruyama, A. Faessler,
1387: Prog. Part. Nucl. Phys. 29 (1992) 87.
1388: \bibitem{AGSall} L. Ahle et al., Nucl. Phys. A 610 (1996) 139c.
1389: \bibitem{E877} J. Barrette et al., Phys. Lett. B 485 (2000) 319.
1390: \bibitem{AGSnew} B. B. Back et al., {\it nucl-ex}/0101008.
1391: \bibitem{NA49} G. I. Veres and the NA49 Collaboration,
1392: Nucl. Phys. A 661 (1999) 383c.
1393: \bibitem{Na49b} J. B\"achler et al., Nucl. Phys. A 661 (1999) 45c.
1394: \bibitem{NAxx} F. Antinori et al., Nucl. Phys. A 661 (1999) 130c;
1395: R. Caliandro et al., J. Phys. G 25 (1999) 171.
1396: \bibitem{NA57} F. Antinori et al., Nucl. Phys. A 681 (2001) 165c;
1397: {\it hep-ex/0105049}.
1398: \bibitem{Andersen} E. Andersen et al., Phys. Lett. B 449 (1999)
1399: 401.
1400: \bibitem{AGS1} A. Jahns, H. St\"ocker, W. Greiner, H. Sorge,
1401: Phys. Rev. Lett. 68 (1992) 2895.
1402: \bibitem{AGS2} S. H. Kahana, Y. Pang, T. Schlagel, C. B. Dover,
1403: Phys. Rev. C 47 (1993) 1356.
1404: \bibitem{Kahana} Y. Pang, D. E. Kahana, S. H. Kahana, H. Crawford,
1405: Phys. Rev. Lett. 78 (1997) 3418.
1406: \bibitem{Bleicher} M. Bleicher et al., Phys. Lett. B 485 (2000) 133.
1407: \bibitem{WA97c} F. Antinori et al., Eur. Phys. J. C 11 (1999) 79.
1408: \bibitem{Sorge} H. Sorge, Z. Phys. C 67 (1995) 479;
1409: Phys. Rev. C 52 (1995) 3291.
1410: \bibitem{Werner} K. Werner, J. Aichelin, Phys. Lett. B 308 (1993) 372.
1411: \bibitem{Carlos} M. A. Braun, C. Pajares, Nucl. Phys. B 390 (1993) 542;
1412: N. Armesto, M. A. Braun, E. G. Ferreiro, C. Pajares,
1413: Phys. Lett. B 344 (1995) 301; M. A. Braun, C. Pajares, J. Ranft,
1414: Int. Jour. Mod. Phys. A 14 (1999) 2689.
1415: \bibitem{Carlos2} N. S. Amelin, N. Armesto, C. Pajares, D. Sousa,
1416: {\it hep-ph}/0103060.
1417: \bibitem{Rafelski} P. Koch, B. M\"uller, J. Rafelski,
1418: Phys. Rep. 142 (1986) 167.
1419: \bibitem{BM} P. Braun-Munzinger, I. Heppe, J. Stachel,
1420: Phys. Lett. B 465 (1999) 15.
1421: \bibitem{becca} F. Becattini et al., Phys. Rev. C 64 (2001) 024901.
1422: \bibitem{Redlich00} F. Becattini, J. Cleymans, A. Keranen, E. Suhonen,
1423: K. Redlich, {\it hep-ph}/0011322.
1424: \bibitem{Red01} K. Redlich, {\it hep-ph}/0105104.
1425: \bibitem{Redlich2} S. Hamieh, K. Redlich, A. Pounsi,
1426: Phys. Lett. B 486 (2000) 61.
1427: \bibitem{Capella} A. Capella, Nucl. Phys. A 661 (1999) 502c.
1428: \bibitem{Rapp} R. Rapp, E. Shuryak, Phys. Rev. Lett. 86 (2001) 2980.
1429: \bibitem{Carsten} C. Greiner, S. Leupold, nucl-th/0009036.
1430: %
1431: \bibitem{Brat} E. L. Bratkovskaya, W. Cassing, C. Greiner et al.,
1432: Nucl. Phys. A 675 (2000) 661.
1433: \bibitem{Bravina} L. V. Bravina et al., J. Phys. G 25 (1999) 351;
1434: Phys. Rev. C 62 (2000) 064906.
1435: \bibitem{Ehehalt} W. Ehehalt, W. Cassing, Nucl. Phys. A 602 (1996) 449.
1436: \bibitem{KLW1} K. Weber et al., Nucl. Phys. { A 539} (1992) 713;
1437: Nucl. Phys. { A 552} (1993) 571; T. Maruyama et al., Nucl.
1438: Phys. A 573 (1994) 653.
1439: \bibitem{Mal} W. Botermans, R. Malfliet, Phys. Rep. 198 (1990) 115.
1440: \bibitem{Cassing90} W. Cassing, V. Metag, U. Mosel, K. Niita,
1441: Phys. Rep. { 188} (1990) 363.
1442: \bibitem{Cassing90c} W. Cassing, U. Mosel, Prog. Part. Nucl. Phys. 25 (1990) 235.
1443: \bibitem{Casju} W. Cassing, S. Juchem, Nucl. Phys. A 665 (2000) 377;
1444: Nucl. Phys. A 672 (2000) 417; Nucl. Phys. A 677 (2000) 445.
1445: \bibitem{Leupold} S. Leupold, Nucl. Phys. A 672 (2000) 475.
1446: \bibitem{Byckling} E. Byckling, K. Kajantie, {\em Particle Kinematics}
1447: (John Wiley and Sons, London, 1973).
1448: \bibitem{Lang93} A. Lang, H. Babovsky, W. Cassing, U. Mosel, H.-G. Reusch,
1449: K. Weber , J. Comp. Phys. { 106} (1993) 391.
1450: \bibitem{LB} H. Schopper (Editor), Landolt-B\"ornstein, New
1451: Series, Vol. I/12, Springer-Verlag, 1988.
1452: \bibitem{PDG} Particle Data Group, Eur. Phys. J. C 15 (2000) 1.
1453: \bibitem{Geiss} J. Geiss, W. Cassing, C. Greiner,
1454: Nucl. Phys. A 644 (1998) 107.
1455: \bibitem{Wolf90} Gy. Wolf et al., Nucl. Phys. A 517 (1990) 615;
1456: Nucl. Phys. A 552 (1993) 549.
1457: \bibitem{Batko3} G. Batko, J. Randrup, T. Vetter,
1458: Nucl. Phys. A 536 (1992) 786; Nucl. Phys. A 546 (1992) 761.
1459: \bibitem{kodama} T. Kodama, S. B. Duarte, K. C. Chung et al.,
1460: Phys. Rev. C 29 (1984) 2146.
1461: \bibitem{Bertsch88} G. F. Bertsch, S. Das Gupta, Phys. Rep. 160 (1988) 189.
1462: \bibitem{Bass98} S. Bass et al., Prog. Part. Nucl. Phys. 41 (1998)
1463: 225.
1464: \bibitem{Larionov} A. B. Larionov, W. Cassing, S. Leupold, U.
1465: Mosel, {\it nucl-th}/0103019, Nucl. Phys. A, in press.
1466: \bibitem{NA44} I. G. Bearden et al., Nucl. Phys. A 610 (1996) 175c;
1467: M. Caneta et al., J. Phys. G 23 (1997) 1865.
1468: \bibitem{Zimanyi1} J. Zimanyi, in: S. Bass et al.,
1469: Nucl. Phys. A 661 (1999) 205c.
1470: \bibitem{Zimanyi2} J. Zimanyi, T. S. Biro et al., {\it hep-ph}/9904501.
1471: \bibitem{Sahu00} P. K. Sahu, W. Cassing, U. Mosel, A. Ohnishi,
1472: Nucl. Phys. A 672 (2000) 376.
1473: \bibitem{Cass00a} W. Cassing, E. L. Bratkovskaya, S. Juchem, Nucl.
1474: Phys. A 674 (2000) 249.
1475: \bibitem{E864n} T. A. Armstrong et al., Phys. Rev. C 59 (1999)
1476: 2699.
1477: \end{thebibliography}
1478: %
1479: \newpage
1480: %----------------------------------------------------------------------
1481: \begin{figure}[h]
1482: \centerline{\psfig{file=bbbar_1.ps,height=10cm}}
1483: \caption{Illustration of the flavor rearrangement model for
1484: $B\bar{B}$ annihilation to 3 mesons and vice versa. The mesons
1485: $M_i$ may be either pseudo-scalar or vector mesons, respectively.}
1486: \label{bild1}
1487: \end{figure}
1488: %
1489: \begin{figure}[h]
1490: \centerline{\psfig{file=npi.eps,width=15cm}}
1491: \vspace*{-5cm}
1492: \caption{The distribution in the final number of pions $P(N_\pi)$
1493: for $p\bar{p}$ annihilation at invariant energies 2.3 GeV $\leq
1494: \sqrt{s} \leq$ 4 GeV (short lines). The solid line is the gaussian
1495: parametrization (\protect\ref{pion5}) that is fitted to the experimental data.}
1496: \label{bild2}
1497: \end{figure}
1498:
1499: \begin{figure}[h]
1500: \centerline{\psfig{file=sppbar.eps,height=20cm}} \caption{The
1501: annihilation cross section $\sigma_{ann.}$ for the $p\bar{p}$
1502: reaction as a function of the laboratory momentum $p_{lab}$ from
1503: Ref. \protect\cite{LB} in comparison to the approximation
1504: (\ref{appann}) (solid line). }
1505: \label{bild1b}
1506: \end{figure}
1507:
1508: %
1509: \begin{figure}[h]
1510: \centerline{\psfig{file=dndst170.eps,width=15cm}}
1511: \vspace*{-5cm}
1512: \caption{The number of $N\bar{N} \rightarrow \rho \rho \pi$ reactions as a function
1513: of the invariant energy $\sqrt{s}$ for a system in thermal and chemical equilibrium
1514: at temperature $T$ = 170 MeV and $\mu_q$ = 0. The solid line denotes the differential number
1515: in the backward ($\rho \rho \pi$) collisions, respectively.}
1516: \label{bild3}
1517: \end{figure}
1518: %
1519: \begin{figure}[h]
1520: \centerline{\psfig{file=pi_apart.eps,width=15cm}}
1521: \vspace*{-5cm}
1522: \caption{The calculated average number of charged pions $<\pi>$ divided by $A_{part}$
1523: as a function of centrality for $Pb+Pb$ at 160 A$\cdot$GeV in comparison to the data from Ref.
1524: \cite{Na49b}.}
1525: \label{fig4}
1526: \end{figure}
1527: %
1528: \begin{figure}[h]
1529: \centerline{\psfig{file=rpbar_pi.eps,width=15cm}}
1530: \vspace*{-5cm}
1531: \caption{The ratio of antiprotons to the average number of charged pions $<\pi>$
1532: as a function of centrality for $Pb+Pb$ at 160 A$\cdot$GeV within different approximations.
1533: The dashed line reflects calculations without annihilation of
1534: antibaryons, the dotted line includes the annihilation channels
1535: while the solid line stands for the calculations with both, the
1536: annihilation and meson fusion channels.}
1537: \label{6new}
1538: \end{figure}
1539: %
1540: %
1541: \begin{figure}[h]
1542: \centerline{\psfig{file=dndt160.eps,width=15cm}}
1543: \vspace*{-5cm}
1544: \caption{The annihilation rate $B\bar{B} \rightarrow 3 mesons$ (dashed histogram)
1545: for a central $Pb+Pb$ collision at 160 A$\cdot$GeV as a function
1546: of time in comparison to the backward reaction rate (solid
1547: histogram) within the HSD transport approach.}
1548: \label{bild5}
1549: \end{figure}
1550: %
1551: \begin{figure}[h]
1552: \centerline{\psfig{file=ratio.eps,width=15cm}} \vspace*{-5cm}
1553: \caption{The calculated $K^+/\pi$ (dashed), $K^-/\pi$ (dotted) and
1554: $\bar{p}/\pi$ (solid) ratio for $Pb + Pb$ at 160 A$\cdot$GeV as a
1555: function of the number of participating nucleons $A_{part}$. The
1556: experimental data are taken from Ref. \cite{Na49b}; the latter
1557: data for antiprotons include also the feeddown from $\bar{\Lambda}$ and
1558: $\bar{\Sigma}^0$, respectively. The stars for $A_{part}$ = 2 denote the
1559: transport results for $K^+/\pi$, $K^-/\pi$ and $\bar{p}/\pi$ ratios in case of
1560: $NN$ reactions at $T_{lab}$ = 160 GeV.} \label{bild4}
1561: \end{figure}
1562: %
1563: \begin{figure}[h]
1564: \centerline{\psfig{file=dens160.eps,width=15cm}}
1565: \vspace*{-5cm}
1566: \caption{The baryon density in a central expanding cylinder (see text) for $|\Delta y|_{cm} \leq$
1567: 1 in case of a central collision of $Pb+Pb$ at 160 A$\cdot$GeV in the HSD approach as a function of time.
1568: The solid line shows the density of 'formed' baryonic states
1569: while the dashed line stands for the net quark density
1570: $\rho_q/3$, while merges with the baryon density in the expansion
1571: phase. The dotted lines at $\rho_0$ and $2 \rho_0$ are drawn to guide the eye.}
1572: \label{bild6}
1573: \end{figure}
1574: %
1575: \begin{figure}[h]
1576: \centerline{\psfig{file=rat_pbar.eps,width=15cm}} \vspace*{-5cm}
1577: \caption{The antiproton to charged pion ratio (see text) as a
1578: function of the number of participating nucleons $A_{part}$ for
1579: $Pb+Pb$ at 160 A$\cdot$GeV. The dotted line denoted by $U=0$
1580: corresponds to the cascade result (cf. Fig. \ref{bild4}) while the
1581: solid line denoted by $U= - 100$ MeV corresponds to a calculation
1582: with the attractive scalar antiproton potential (\ref{pot}) of
1583: -100 MeV at baryon density $\rho_0$. The full triangles represent
1584: the data from Ref. \cite{Na49b} while the open circles correspond
1585: to a cascade calculation including the feeddown from antihyperons
1586: according to Eq. (\ref{antihyp}). The errorbars are due to the
1587: limited statistics in the transport calculation. The star for
1588: $A_{part}$ = 2 denotes $\bar{p}/\pi$ ratio for $NN$ collisions at
1589: $T_{lab}$ = 160 GeV. } \label{bild7}
1590: \end{figure}
1591: %
1592: %
1593: \begin{figure}[h]
1594: \centerline{\psfig{file=dens11.eps,width=15cm}} \vspace*{-5cm}
1595: \caption{The baryon density in a central expanding cylinder (see
1596: text) for $|\Delta y|_{cm} \leq$
1597: 1 in case of a central collision of $Au+Au$ at 11.6 A$\cdot$GeV/c in the HSD approach
1598: as a function of time.
1599: The solid line shows the density of 'formed' baryonic states
1600: while the dashed line stands for the net quark density
1601: $\rho_q/3$, while merges with the baryon density in the expansion
1602: phase. }
1603: \label{bild8}
1604: \end{figure}
1605: %
1606: \begin{figure}[h]
1607: \centerline{\psfig{file=rat_pi.eps,width=15cm}} \vspace*{-5cm}
1608: \caption{The $K^+/\pi^+$, $K^-/\pi^+$, $p/\pi^+$ and
1609: $\bar{p}/\pi^+$ ratio at midrapidity for $Au + Au$ at 11.6
1610: A$\cdot$GeV/c as a function of the number of participating protons
1611: $N_{pp}$. The solid line correspond to the HSD transport
1612: calculation in the cascade mode for mesons and antibaryons (open
1613: circles) while the experimental data (full symbols) are taken
1614: from \cite{AGSall}.} \label{bild9}
1615: \end{figure}
1616: %
1617: %
1618: \begin{figure}[h]
1619: \centerline{\psfig{file=dndt_11.eps,width=15cm}} \vspace*{-5cm}
1620: \caption{The annihilation rate $B\bar{B} \rightarrow 3 mesons$
1621: (dashed histogram) for a central $Au+Au$ collision at 11.6
1622: A$\cdot$GeV/c as a function of time in comparison to the backward
1623: reaction rate (solid histogram) within the HSD transport
1624: approach.} \label{bild10}
1625: \end{figure}
1626: %
1627: \begin{figure}[h]
1628: \centerline{\psfig{file=lamb_pb.eps,width=15cm}} \vspace*{-5cm}
1629: \caption{The $\bar{\Lambda}/\bar{p}$ ratio as a function of
1630: centrality for $Au+Au$ at 11.6 A$\cdot$GeV/c in the HSD approach
1631: within the approximation
1632: (\ref{antihyp}). The shaded area corresponds to the uncertainty in
1633: the statistics of the transport calculations for the different
1634: particle abundancies at midrapidity.} \label{bild11}
1635: \end{figure}
1636:
1637: \end{document}
1638:
1639: #!/bin/csh -f
1640: # Uuencoded gz-compressed .tar file created by csh script uufiles
1641: # For more info (11/95), see e.g. http://xxx.lanl.gov/faq/uufaq.html
1642: # If you are on a unix machine this file will unpack itself: strip
1643: # any mail header and call resulting file, e.g., figures.uu
1644: # (uudecode ignores these header lines and starts at begin line below)
1645: # Then say csh figures.uu
1646: # or explicitly execute the commands (generally more secure):
1647: # uudecode figures.uu ; gunzip figures.tar.gz ;
1648: # tar -xvf figures.tar
1649: # On some non-unix (e.g. VAX/VMS), first use editor to change filename
1650: # in "begin" line below to figures.tar-gz , then execute
1651: # uudecode figures.uu
1652: # gzip -d figures.tar-gz
1653: # tar -xvf figures.tar
1654: #
1655: uudecode $0
1656: chmod 644 figures.tar.gz
1657: gunzip -c figures.tar.gz | tar -xvf -
1658: rm $0 figures.tar.gz
1659: exit
1660:
1661: