Particle ratios at SPS, AGS and SIS
Extracted source text ยท arXiv nucl-th/9809031
%123456789012345678901234567890123456789012345678901234567890123456789012345678
% 1 2 3 4 5 6 7
%
\documentclass[10pt,a4paper]{article}
\usepackage{epsfig}
\textwidth 160mm
\textheight 230mm
\pagenumbering{arabic}
\topmargin -25mm
\oddsidemargin -4mm
\begin{document}
%
%------------SPECIAL COMMANDS--------------------------------------------------
%
%UCT-TP 217/94
%\hoffset=-0.9cm
%\topmargin=-2cm
\date{\today}
\title{Particle ratios at SPS, AGS and SIS.}
%
%
%
%
\author{Jean Cleymans$^1$, Helmut Oeschler$^2$
and Krzysztof Redlich$^{3,4}$\\[0.5cm]
$^1$Department of Physics, University of Cape Town,\\
Rondebosch 7701, South Africa\\
$^2$Institut f\"ur Kernphysik, Technische Universit\"at Darmstadt, \\
D-64289 Darmstadt, Germany\\
$^3$Gesellschaft f\"ur Schwerionenforschung, D-64291 Darmstadt, Germany\\
$^4$Institute of Theoretical Physics, University of Wroc\l aw,
Wroc\l aw, Poland\\}
\maketitle
%
%
\begin{abstract}
Ratios of integrated particle yields
provide the best method for
determining the temperature
and the chemical potential.
The chemical freeze-out parameters obtained at CERN/SPS,
BNL/AGS and GSI/SIS energies all correspond to a unique value
of 1 GeV per hadron in the local rest frame of the system,
independent of the beam energy and of the target and beam
particles.
\end{abstract}
%
%
%
%
\section{Particle Ratios}
%
%
%
Ratios of integrated particle yields provide the best way to determine
the freeze-out values of the temperature and baryon chemical potential
of hadronic matter produced in relativistiv heavy ion
collisions.
Various
effects (e.g. flow) which can severely distort
the momentum
spectra of the particles produced cancel out in
such ratios (see e.g. \cite{jaipur}).
The analysis of particle ratios
is therefore the best method to obtain reliable information on
the chemical freeze-out parameters of the hadronic final state.
Below we briefly summarize arguments supporting this
statement.
\begin{enumerate}
\item Excluded volume corrections cancel out
in particle ratios. In most models one simply has the following
relations (see e.g. \cite{crss,greiner})
\begin{eqnarray}
\displaystyle
{N_i\over N_j} &=&
{{N_i^0\over 1 + \sum_l N_l^0V_0}
\over {N_j^0\over 1 + \sum_l N_l^0V_0}} ,\nonumber\\
&=& {N_i^0\over N_j^0},
\end{eqnarray}
%
%
where $N_i^0$ refers to the distribution from a fireball at
rest which in
the Boltzmann approximation is given by
%
\begin{eqnarray}
N_i^0 &=& g\int {d^3p\over (2\pi)^3} e^{-E/T}e^{\mu/T} \nonumber \\
&=& gm_i^2TK_2(m_i/T)e^{\mu/T} .
\end{eqnarray}
%
%
Equation (1) shows that effects due to excluded volume corrections
cancel out in particle ratios. As a word of caution : there exist
models where these
corrections do not cancel exactly as above, see e.g.
\cite{stachel}.
%
%
\item If the hadronic gas is made up of a
superposition of fireballs having different rapidities
but the same temperature and chemical potential then
the particle ratios are as if there were only one
fireball:
\begin{eqnarray}
{N_i\over N_j} &=&
{\int_{-\infty}^{\infty}dy
\int_{-Y}^YdY_{FB}~\rho(Y_{FB})
{dN_i^0\over dy}(y-Y_{FB})
\over \int_{-\infty}^{\infty}dy
\int_{-Y}^Y dY_{FB}~\rho(Y_{FB})
{ dN_i^0\over dy}(y-Y_{FB})} ,\nonumber\\
&=&{N_i^0\over N_j^0}~{\int_{-Y}^Y~dY_{FB}~\rho (Y_{FB})
\over\int_{-Y}^Y~dY_{FB}~\rho (Y_{FB})} ,\nonumber\\
&=& {N_i^0\over N_j^0}.
\end{eqnarray}
Hence effects due to a superposition of similar fireballs cancel out.
It is of course a severe limitation that the fireballs must all
have the same temperature, this has been lifted partially by
Becattini recently \cite{becattini}.
%
%
%
%
%
\item Transverse flow (or random walk \cite{random}) effects
with instantaneous
freeze-out cancel out if one considers
particle yields that have been integrated over transverse
momentum.
The discussion about transverse flow
needs the following two integrals:
\begin{equation}
\int_0^\infty dp_T
m_T^i~~K_1\left({m_T^i\over T}\cosh y_T\right)
I_0\left({ p_T\over T}\sinh y_T\right)
=m_i^2K_2({m^i\over T})\cosh y_T ,
\end{equation}
and
\begin{equation}
\int_0^\infty dp_T
p_T~K_0\left({m_T^i\over T}\cosh y_T\right)
I_1\left({ p_T\over T}\sinh y_T\right)
=m_i^2K_2({m^i\over T})\sinh y_T .
\end{equation}
%
%
%
For instantaneous freeze-out
and constant transverse velocity, one has
\begin{eqnarray}
{N_i\over N_j} &=&
{\int_0^\infty dp_T
m_T^i~~K_1\left({\displaystyle m_T^i\over T}\cosh y_T\right)
I_0\left({\displaystyle p_T\over T}\sinh y_T\right)
\over
\int_0^\infty dp_T
m_T^j~~K_1\left({\displaystyle m_T^j\over T}\cosh y_T\right)
I_0\left({\displaystyle p_T\over T}\sinh y_T\right)} ,
\nonumber\\
&=& {m_i^2~~K_2\left({\displaystyle m^i\over T}\right)
\cosh y_T
\over
m_j^2~~K_2\left({\displaystyle m^j\over T}\right)\cosh y_T} ,
\nonumber\\
&=& {N_i^0\over N_j^0} .
\end{eqnarray}
i.e. effects due to transverse flow cancel out if the particle
yields have been integrated over all transverse momenta.
%
\item For a longitudinal expansion
\`a la Bjorken accompanied by a transverse
expansion, the effects
also cancel if one considers particle ratios integrated
over transverse momenta as shown below. The distribution
is, in this case, given by
%
\begin{eqnarray}
\left( {dN_i\over dy m_Tdm_T}\right)_{y=0}
& & = {g\over \pi} \int_\sigma r~dr~\tau_F(r) \nonumber\\
& &\left\{ m_T^iI_0 \left( {p_T\sinh y_T\over T} \right)\right.
K_1 \left( {m_T^i\cosh y_T\over T} \right) \nonumber \\
-& &\left( {\partial\tau_F\over\partial r} \right) p_T
I_1 \left( {p_T\sinh y_T\over T} \right)
\left. K_0 \left( {m_T^i\cosh y_T\over T} \right) \right\}
\end{eqnarray}
where $\tau_F(r)$ is the proper freeze-out time which is a
function of the radial distance, $r$.
After integration over $m_T$ one is left with
%
%
\begin{eqnarray}
\left( {dN_i\over dy }\right)_{y=0}
= & &{g\over \pi} \int_\sigma r~dr~\tau_F(r) \nonumber \\
& & \left\{ \cosh(y_T) - \left({\partial\tau_F\over\partial r} \right)
\sinh (y_T) \right\} m_i^2 T K_2 \left( {m_i\over T} \right)
\end{eqnarray}
%
%
%
\bigskip
Since the freeze-out parameters are the same for all particles
it follows that
\begin{equation}
{dN_i/dy\over dN_j/dy} = {N_i^0\over N_j^0}
\end{equation}
This relation shows that in a model where the momentum
distribution shows a plateau in rapidity space and where the
flow pattern could be completely change the transverse
momentum distribution, the ratio of particles is still
as if one would have a single fireball at rest.
\end{enumerate}
%
%
%
\begin{center}
\begin{figure}
\epsfig{file=figure1.eps,scale=0.65}
\caption{
Lines corresponding to fixed particle ratios
in the $T-\mu_B$ plane. The indicated values were obtained at
the GSI/SIS accelerator for $Au-Au$ collisions at 1
A$\cdot$GeV.}
\end{figure}
\end{center}
%
%
\section{Results}
%
We show in
Fig. 1 the results obtained from an analysis \cite{oeschler}
of $Au-Au$ collisons at an energy of 1 A$\cdot$GeV.
As one can see from an inspection of this figure there is a
consistency with a freeze-out temperature of about 50 MeV and a
a freeze-out baryon chemical potential of about 850 MeV. A more
complete analysis of the GSI/SIS data can be found in
\cite{oeschler} and in \cite{ceks,averbeck,PBM}. A notable
exception is the $\eta/\pi^0$ ratio which does not fit the
expectations of the thermal model.
%
%
Combining the results from different accelerators (for a review
and references see Sollfrank \cite{sollfrank})
it can be seen that
a unified description of the hadronic
abundances produced in heavy ion collisions at the CERN/SPS,
the BNL/AGS and the GSI/SIS accelerators
is possible \cite{prl}. This
description covers a range in beam energies from 200
A$\cdot$GeV to below 1 A$\cdot$GeV. As it turns out the
same description can also be applied to the hadronic abundances
in LEP and in $p-p$ and $\bar{p}-p$ collisions
%new
with slightly different treatment of strangeness sector which
accounts for strangeness under-saturation.
The
result
%
can be summarized in a surprisingly simple way: the
hadronic composition of the final state is determined by an
energy per hadron being approximately 1 GeV per hadron in the
rest frame of the produced system. This generalizes an
observation made a long time ago by Hagedorn \cite{hagedorn}
for $p-p$
collisions, namely, as one increases the
beam energy, the available energy is used to produce more
particles, but not to increase the temperature of the system.
This led Hagedorn to the idea of a limiting
temperature. For heavy ion collisions one has to take into
account not only the temperature but also
the finite baryon density
of the system, which is described by the baryon
chemical potential
$\mu_B$.
%new
%
This, as it was first indicated by P. Braun-Munzinger and J. Stachel
\cite{stachel},
leads to a freeze-out curve in the $T,\mu_B$ plane. In
Fig. 2 the values of the freeze-out parameters are shown, as
obtained by various groups (a summary can be found in
\cite{sollfrank}).
%
The solid line corresponds to 1 GeV
per hadron, the dashed line corresponds to 0.94 GeV per hadron.
This energy corresponds to the chemical freeze-out stage,
namely, before the hadrons decay into
the stable hadrons.
%
%We will now discuss in more detail how Fig. 2 was obtained.
Such an analysis, relying as much as possible on fully
integrated particle multiplicities was carried out
for BNL/AGS and for CERN/SPS data.
In Fig. 2 the SPS points are indicated by open squares
\cite{stachel,bgs} while the AGS points are indicated by open
circles \cite{stachel,cets,bgs}.
Data using $Ni$ and $Au$ beams at energies between 0.8 and 1.9
A$\cdot$GeV have become available recently from the GSI/SIS
accelerator. These data have attracted considerable interest
due to the surprisingly large number of $K^-$ mesons being
produced below threshold. A very detailed and extensive
discussion of these results in the framework of thermal models
has been presented in \cite{ceks,averbeck,oeschler}.
The results for the
freeze-out parameters
for $Ni-Ni$ at 1.9 A$\cdot$GeV is shown as an open triangle
in Fig. 2.
The
points with the lowest temperature correspond to $Au-Au$
collisions at 0.8 and 1.0 A$\cdot$GeV and $Ni-Ni$ collisions at
1.0 and 1.8 A$\cdot$GeV.
and are also shown as open triangles.
%
%
%
\begin{center}
\begin{figure}
\epsfig{file=figure2.eps,scale=0.5}
\caption{
Freeze-out values obtained from hadronic abundances at
CERN/SPS, BNL/AGS and GSI/SIS. Also indicated are the points
obtained from observed hadronic abundances at LEP and in $p-p$
collisions at CERN. The smooth curves correspond to a fixed
energy per hadron in the hadronic gas model (from \cite{prl}). }
\end{figure}
\end{center}
%
%
%
A similar analysis has been performed in
\cite{becattini} for $e^+e^-$
annihilation into hadrons at LEP.
Since no baryons are involved here this corresponds to zero
baryon chemical potential, $\mu_B=0$. An impressive fit has
been obtained here since no less than 29 different hadronic
abundances can be reproduced. It is our view that such a good
agreement cannot simply be a coincidence. This analysis was
subsequently extended \cite{becattini} to $p-p$ and
$\bar{p}-p$ reactions at CERN.
In this case,
one reproduces the Hagedorn temperature obtained many
years ago.
In the underlying hadronic gas model all these points
can be described by a single
curve corresponding to a fixed energy per particle, $\epsilon/n$,
which has
approximately the value of 1 GeV per particle in the hadronic gas.
{\it This value characterizes all the final states produced by
beams having
1 A$\cdot$GeV all the way up to 200 A$\cdot$GeV.}
Thus, the only modification one needs to make to the concept
of Hagedorn's limiting temperature it that there exist
a "limiting" - freeze-out energy per particle of 1 GeV at which
hadrons are formed in a collision.
This observation leads to a considerable
unification in the description of the
hadronic final states produced in high
energy collisions.
%
%
\subsection*{Acknowledgment}
\hspace*{\parindent}
We acknowledge stimulating discussions with P. Braun-Munzinger,
B. Friman, W. N\"orenberg, H. Satz and W. Weinhold.
%
\begin{thebibliography}{99}
%
%
%
\bibitem{jaipur} J. Cleymans, in ``Physics and Astrophysics of
Quark-Gluon Plasma'',
B.C. Sinha, D.K. Srivastava and Y.P. Viyogi (Eds.); Narosa Publishing
House, New Delhi (1998).
%
\bibitem{crss} J.~Cleymans, K.~Redlich, H.~Satz and E.~Suhonen,
Z. f. Physik C{\bf 33}, 151 (1986).
%
%
%
\bibitem{greiner} M.I. Gorenstein, H. St\"ocker, G.D. Yen, S.N.
Yang and W. Greiner, University of Frankfurt preprint
nucl-th/9711055 (1997).
%
\bibitem{stachel} P. Braun-Munzinger, J. Stachel,
Wessels, and N. Xu, Phys. Lett B{\bf 365}, 1 (1996);
P. Braun-Munzinger, and J. Stachel, Nucl. Phys. A{\bf 606}, 320
(1996);
nucl-ex/9903015, Nucl. Phys. A (1998) {\it in print}.
%
\bibitem{becattini} F. Becattini,
Z. Physik C{\bf 69}, 485 (1996); F. Becattini and U.~Heinz,
Z. Physik C{\bf 76}, 269 (1997); see also T.F. Hoang and B.
Cork, Z. Physik C{\bf 38}, 603 (1988).
%
\bibitem{random} A. Leonidov, M. Nardi and H. Satz, Nucl.
Phys. A{\bf 610} (1996) 124c; Z. Physik C{\bf 74} (1997) 535.
%
%
\bibitem{oeschler} J. Cleymans, H. Oeschler and K. Redlich,
GSI-preprint September 1998.
%
\bibitem{ceks} J. Cleymans, D. Elliott, A. Kera\"anen and E. Suhonen,
Phys. Rev. C(1998).
%
%
\bibitem{averbeck} R. Averbeck et al., (TAPS Collaboration)
TAPS workshop, St.~Odile, Sept. 1997 nucl-ex/9803001.
%
\bibitem{PBM} P. Braun-Munzinger, private communication.
%
%
\bibitem{sollfrank} J. Sollfrank, J. of Physics G{\bf 23}, 1903 (1997).
%
\bibitem{prl} J. Cleymans and K. Redlich, nucl-th/9808030,
GSI-preprint-98-43.
%
\bibitem{hagedorn} R. Hagedorn, Nuovo Cimento Suppl. {\bf 3},
147 (1965);
see also R. Hagedorn and J. Rafelski, Phys. Lett. B{\bf 97},
136 (1980).
\bibitem{cets} J. Cleymans, D. Elliott, R.L. Thews and H. Satz,
Z. f. Physik C{\bf 74} 319 (1997).
%
\bibitem{bgs} F. Becattini, M.Ga\'zdzicki and J. Sollfrank,
hep-ph/9710529 (1997).
%
%
\end{thebibliography}
%
%
\end{document}