phi-meson in Nuclear Matter
Extracted source text ยท arXiv nucl-th/9507017
%Paper: nucl-th/9507017
%From: Tetsuo Hatsuda <hatsuda@nucl.ph.tsukuba.ac.jp>
%Date: Tue, 11 Jul 95 16:06:00 JST
\documentstyle[12pt]{article}
\def\baselinestretch{1.2}
\parindent 30pt \textheight 9in
\topmargin -.5in \textwidth 6in
\oddsidemargin 0.25in
\evensidemargin 0.0in
\newcommand{\beq}{\begin{eqnarray}}
\newcommand{\eeq}{\end{eqnarray}}
\begin{document}
\rightline{UTHEP-307}
\rightline{June, 1995}
\vskip 2cm
\vspace{5pt}
\centerline{\Large{\bf $\phi$-meson in Nuclear Matter}}
\vspace{1.0cm}
\centerline{H. Kuwabara and T. Hatsuda}
\vspace{0.5cm}
\noindent
\centerline{\em Institute of Physics, University
of Tsukuba, Tsukuba, Ibaraki 305, Japan}
\vspace{2cm}
\centerline{Abstract}
$\phi$-meson mass in nuclear matter ($m_{\phi}^*$)
is investigated using an effective Lagrangian of $\phi$ interacting
with octet baryons.
$m_{\phi}^*$ decreases by a few \% in nuclear matter due to the
current conservation and effective nucleon/hyperon masses.
Its implication to the p-A and A-A collisions are briefly discussed.
\vspace{2cm}
\noindent
Submitted to Prog. Theor. Phys.
\newpage
\setcounter{equation}{0}
\renewcommand{\theequation}{\arabic{equation}}
In recent years, the effecitve meson masses in
hadronic matter at finite density
($\rho$) and temperature ($T$)
acquire wide interests both theoretically and experimentally
(see the recent reviews, \cite{yukawa}.)
In particular, the $\phi$-meson,
which is a $\bar{s}s$ resonance in $J^P = 1^-$ channel
with a narrow width
($m_{\phi} =1019.4$ MeV
and $\Gamma_{\phi} = 4.4$ MeV), is a unique probe for
partial restoration of chiral symmetry
in hot/dense hadronic matter \cite{finiteT,LH,theory}.
Detection of $\phi$ through the dacays
$\phi \rightarrow K \bar{K}, e^+e^-, \mu^+ \mu^-$
in nucleous-nulceous (A-A) and
proton-nucleous (p-A) collisions could give experimental
information on the spectral change of $\phi$ in matter:
preliminary data on $\phi \rightarrow K^+K^-$
in $A-A$ colisions at AGS-BNL has been recently reported \cite{AGS},
and an experiment using p-A reactions is planned at KEK \cite{Enyo}.
In this letter, we will report our recent study
on the $\phi$-meson mass in nuclear matter at zero $T$.
Our starting point is an effective hadronic model
composed of $\phi$-meson, nucleon and
hyperons. This is a generalization of the recent works
by Shiomi and Hatsuda \cite{SH} and others \cite{others,french,PW}
who studied the
effect of nucleon-loops to the proporties of
rho and omega mesons in nuclear matter.
Let's start with the vector coupling of $\phi$ with
octet baryons
($B \equiv N, \Lambda^0, \Sigma^{\pm}, \Sigma^{0}$);
\beq
L_{int} = \sum_B g_{\phi B} \ \bar{B} \gamma_{\mu}B\ \phi^{\mu} ,
\eeq
where $g_{\phi B}$ being the $\phi$-baryon coupling constant listed in
Table 1.
\vspace{0.8cm}
\begin{center}
\begin{tabular}{|c|c|c|c|} \hline
Baryons & $g_{\sigma B}$ & $g_{\omega B}$ &
$g_{\phi B}$ \\ \hline \hline
N &8.7 &10.6 &4.2$^*$ \\ \hline
$\Lambda$ &5.2 &6.9 &2.3 \\ \hline
$\Sigma$ &5.2 &6.9 &2.3 \\ \hline
\end{tabular}\\
\vspace{0.5cm}
\end{center}
\noindent
Table1:
$g_{\sigma B}$, $g_{\omega B}$ and $g_{\phi B}$ denote
$\sigma$-$B$ scalar coupling,
$\omega$-$B$ vector coupling,
and
$\phi$-$B$ vector coupling, respectively.
$g_{\sigma B}$ and $g_{\omega B}$
are taken from \cite{gled}. The number with $^*$
should be considered as an upper bound.
\vspace{0.8cm}
Some remarks are in order here:
(i) $\phi - \Lambda$ and $\phi - \Sigma$ couplings
do not break the OZI rule, since the quark lines at the vertices are
connected. On the other hand, the $\phi - N$ coupling is OZI violating.
(ii) $\Xi$ is neglected, since its effect to the
$\phi$ self-energy is doubly
suppressed by the mass of $\Xi$ and by the OZI violating
nature of $\phi-\Xi$ coupling.
(iii) If one relies on the quark counting rule \cite{QM},
the $\phi$-hyperon couplings are related to the
$\omega$-hyperon couplings as
$g_{\phi \Lambda (\phi \Sigma)} = g_{\omega \Lambda (\phi \Sigma)}/3$ with
$g_{\omega \Lambda (\phi \Sigma)}$ being determined by the
fit of the hypernuclear levels \cite{gled}. This is assumed in Table 1.
(iv) $\phi$-nucleon coupling, which is OZI violating,
is not known experimentally.
However, a study of the electromagnetic form-factors of the nucleon
yields an upper bound of its strength \cite{jaffe}.
Using the notation of ref.\cite{jaffe},
$g_{\phi N}/g_{\omega N}= (\sin \epsilon + \cos \epsilon \tan \eta_1)
/(\cos \epsilon - \sin \epsilon \tan \eta_1) \simeq \epsilon +
\tan \eta_1 < 0.4 $.
We will consider only the $N=Z$ non-strange nuclear matter in this letter.
In this case, effects of the
hyperons to the $\phi$-meson self-energy arise only through
hyperon$-$anti-hyperon loops. Effective masses of hyperons
$M_{\Lambda,\Sigma}^*$ in nuclear matter
give density dependence of the self-energy.
Nucleon contribution to the self energy
has both $N-\bar{N}$ loop (polarization of the Dirac sea)
and the scattering with nucleons in nuclear
matter (polarization of the Fermi sea) \cite{others,PW,SH}.
The one-loop self energy from hyperon and nucleon contributions
reads
\beq
\label{polarization}
\Pi^{\mu \nu}_B (\omega, {\bf q};\rho)
=- ig_{\phi B}^2 \int {d^4k \over (2\pi)^4}\
{\rm Tr}[\gamma^\mu G(k+q) \gamma^\nu G(k)] \ \ ,
\eeq
where the four momentum of $\phi$ is
$q^{\mu} = (\omega, {\bf q})$, ``Tr'' is for the Dirac indices,
and
$G(k)$ denotes baryon propagator in nuclear matter
which depends on the effective mass of the nucleon and hyperons
$M_B^*$ ($B= N,\Lambda, \Sigma)$. (See \cite{SH} for
the explicit form of $G(k)$.)
Although one can calculate the density dependence
of $M^*_B$ within the framework of quantum hadrodynamics (QHD)
\cite{QHD}, we adopt the following forms to study
correlations between $M^*_B$ and $m_{\phi}^*$ in a qualitative manner:
\beq
\label{baryon}
g_{\sigma \Lambda (\sigma\Sigma)}/g_{\sigma N}
& = & ( M_{\Lambda (\Sigma)} -M_{\Lambda (\Sigma)}^*)
/ (M_N - M_N^*) \\
\label{nucleon}
M_N^*/M_N & \simeq & 1 - 0.15 \rho/\rho_0 ,
\eeq
where $\rho_0$ is the normal nuclear-matter density,
and $g_{\sigma \Lambda (\sigma \Sigma)}$ and $g_{\sigma N}$ are
given in Table 1. Eq.(\ref{baryon}) is an
universal relation in the relativistic mean field theory
\cite{gled}
and Eq.(\ref{nucleon}) is a standard parametrization
for the the effective nucleon mass at $\rho < 2 \rho_0$ \cite{french}.
In Fig. 1, effective masses of $N$, $\Lambda$ and $\Sigma$
parametrized by eqs. (\ref{baryon},\ref{nucleon})
are shown as a function of baryon density.
The effective $\phi$-meson mass $m^*_{\phi}$
at rest ($\omega \neq 0$, ${\bf q}=0$)
is obtained as a solution of the dispersion relation
\beq
\label{dispersion}
\omega^2 - m_{\phi}^2 + \sum_B
\tilde{\Pi}_B (\omega, {\bf 0};\rho) =0,
\eeq
where
$\tilde{\Pi}_B (\omega, {\bf 0};\rho) \equiv -
\tilde{\Pi}_B^{\mu\mu}(\omega, {\bf 0};\rho)/ 3\omega^2 $, and
$m_{\phi}$ is the $\phi$-meson mass in the vacuum.
$\tilde{\Pi}_B^{\mu \mu}$ denotes the
density dependent part of ${\Pi}_B^{\mu \mu}$:
the density independent logarithemic divergence
is subtracted out in the dimensional regularization scheme
following the procedure
given in \cite{french,SH}, namely
$ \tilde{\Pi}_B^{\mu \mu}(\omega, {\bf 0};\rho)
\equiv \Pi_B^{\mu \mu}(\omega, {\bf 0};\rho) -
\Pi_B^{\mu \mu}(\omega, {\bf 0};0)$.
On should note here that
similar ``renormalized'' self-energy in the relativistic $\sigma-\omega$
model has been used to study the electron-nucleous scattering, and
it has been shown that the renormalized Dirac-sea polarization
gives rather good agreement with experiments on
the quenching of the Coulomb sum values \cite{KS88}.
The solid line in Fig.2 show
the ratio $m_{\phi}^*/m_{\phi}$ calculated in the above
subtraction procedure with hyperon-loops only.
The hyperon contribution is less ambiguous comapred to the
nucleon contribution, since the absolute value of
$g_{\phi N}$ in the latter case is not known.
$m_{\phi}^*$ decreases by $2-3 \% $ in the range
$\rho_0 < \rho < 2 \rho_0$.
Note that
the OZI rule is preserved for $\phi$-hyperon vertices, while it is
violated in the
self-energy $ \tilde{\Pi}_B^{\mu \mu}(\omega, {\bf 0};\rho)$.
This is because the self-energy
respresents interaction
of $\phi$ ($s\bar{s}$ pair) with {\em non-strange} nuclear matter.
Similar phenomena are known in two-step decay processes
such as $\phi \rightarrow K\bar{K} \rightarrow \rho \pi$,
$f' \rightarrow K\bar{K} \rightarrow \pi \pi$, and
$J/\psi \rightarrow D\bar{D} \rightarrow \rho \pi$, where
each vertex preserves the rule while
the whole amplitude violates the OZI rule \cite{lipkin}.
The solid line in Fig.3 shows
the ratio $m_{\phi}^*/m_{\phi}$ calculated in the above
subtraction procedure with nucleon-loops only.
We have used $g_{\phi N}/g_{\omega N}
= 0.32$ in Fig.3 which is close to the upper bound
given in Table 1: thus the
resultant decrease of $m_{\phi}^*$ in Fig.3 should be considered as
an upper
limit originating from the nucleon-loop. Note here that
the nucleon contribution contains the polarization of Dirac sea
and fermi sea.
The fomer (latter) tends to decreases
(increases) the effective mass.
The nagitive mass shift in Fig.2 and Fig.3 is
a direct consequence of the current conservation
($\partial_{\mu}(\bar{B}\gamma^{\mu}B)=0)$ and
$M^*_B/M_B < 1$,
which was first discussed in \cite{SH,PW} for rho and omega
mesons. For the $\phi$-meson,
the current conservation implies that
the propagator of $\phi$ (without a small Fermi-sea polarization)
has a form $D(q) \simeq 1/(Z^{-1}q^2 - m_{\phi}^2)$
with $Z$ being a {\em finite} wave-function renormalization in medium.
$M_B^*/M_B < 1$ implies that $\phi$ is more dressed by
baryonic clouds in medium, which leads to $Z < 1$.
Thus, one arrives
at the conclusion $m_{\phi}^*/m_{\phi} \equiv Z < 1$.
This mechanism is quite general and does not depend on the
details of the interaction and on the virtual particles running
in the loop;
for example, a similar decrease of $m_{\phi}^*$ should be seen even when
one replaces the baryonic loops by the
constituent-quark loop.
To see the effect of the ultraviolet cutoff on the {\em finite
part} of the loop
integral in (\ref{polarization}),
let us define
$ \tilde{\Pi}_B^{\mu \mu}(\omega, {\bf 0};\rho,\Lambda_{cut})
\equiv \Pi_B^{\mu \mu}(\omega, {\bf 0};\rho,\Lambda_{cut}) -
\Pi_B^{\mu \mu}(\omega, {\bf 0};0,\Lambda_{cut})$ and use this
in (\ref{dispersion}). We take covariant cutoff for
$\Lambda_{cut}$ for simplicity.
When $\Lambda_{cut} \rightarrow \infty$,
$ \tilde{\Pi}_B^{\mu \mu}(\omega, {\bf 0};\rho,\Lambda_{cut})$
reduces to $\tilde{\Pi}_B^{\mu \mu}(\omega, {\bf 0};\rho)$.
The dashed lines in Fig.2 and Fig.3
are the results of such calculations for three cases,
$\Lambda_{cut}$ = 1, 2, 10 GeV.
Although the cutoff dependence is not negligible,
the qualitative picture we draw in the above is not affected.
Also, ``renormalized'' curve (solid line) is more favorable
from the phenomenological point of view, since it can explain
the quenching of the Coulomb sum values as we have mentioned before.
We have considered only the nucleon and hyperon loops in the $\phi$
self-energy. Another possible contribution is the kaon-loop in medium,
which was studied by Ko et al. \cite{theory}.
They found that the kaon-loop also has a tendency to decrease
$m_{\phi}^*$ at low densities provided that the
effective kaon mass $\bar{m}_K^* = ( m^*_{K^{-}}+ m^*_{K^{+}})/2$
decreases in medium.
However, it is still controvertial whether $\bar{m}_K^*$ really
decreases
in nuclear matter or not (see e.g. \cite{Yabu}).
In QCD sum rules (QSR), $m_{\phi}^*$ is shown to decrease as a
result of the
partial restoration of chiral symmetry in nuclear matter, in
particular by
the medium modification of the strangeness condensate
$\langle \bar{s}s \rangle$ \cite{LH}.
Unfortunatelly, it is hard to make a solid connection of
this result with that in this letter, since the kinamatical region
to extract $m_{\phi}^*$ in two approaches are quite different (
deep Euclidian region in QSR versus on-shell region in the approach here).
Recently, Enyo et al. have proposes an
experiment to create $\phi$-meson in heavy nuclei using the proton-nucleus
reaction and to detect lepton pairs and kaon pairs from $\phi$
decaying in the nucleous \cite{Enyo}. A possible signal
in this experiment is a double $\phi$-meson peak in the $e^+e^-$
spectrum and also a large change of the branching ratio
$\Gamma(\phi \rightarrow e^+e^-)/\Gamma(\phi \rightarrow K^+K^-)$.
Also, E859 at BNL-AGS has recently reported
a possible spectral change of the $\phi$-peak in
$K^+K^-$spectrum in heavy ion collisions \cite{AGS}.
In such experiments, a shoulder structure of the $\phi$-peak
should be a possible signal of the mass shift of $\phi$.
\vspace{3cm}
The authors thank H. Shiomi for useful discussions and helps.
We also thank Y. Akiba for suggesting us to study
the cutoff dependence, and H. Hamagaki, Y. Miake and K. Yagi
for discussions on E859 data at AGS.
\newpage
\begin{thebibliography}{99}
\bibliographystyle{unsrt}
%\footnotesize{
\bibitem{yukawa}
T. Hatsuda, {\em Hadron Structure and the QCD Phase Transition},
hep-ph/9502345 (1995); R. Pisarski, {\em Applications of Chiral Symmetry},
hep-ph/9503330 (1995);
G. E. Brown and M. Rho, {\em Chiral Restoration in Hot and/or
Dense Matter}, hep-ph/9504250 (1995).
\bibitem{finiteT}
D. Lissauer and E. V. Shuryak, Phys. Lett. {\bf B253} (1991) 15;\\
P. Bi and J. Rafelski, Phys. Lett. {\bf B262} (1991) 485;\\
J. P. Blaizot and R. Mendez Galain, Phys. Lett. {\bf B271} (1991) 32.
\bibitem{LH}
T. Hatsuda and S. H. Lee, Phys. Rev. {\bf C46} (1992) R34.
\bibitem{theory}
C. M. Ko, P. Levai, X. J. Qiu and C. T. Li, Phys. Rev. {\bf C45}
(1992) 1400.\\
K. L. Haglin and C. Gale, Nucl. Phys. {\bf B421} (1994) 613.\\
M. Asakawa and C. M. Ko, Phys. Rev. {\bf C50} (1994) 3064;
Nucl. Phys. {\bf A572} (1994) 732.
\bibitem{AGS} B. Cole, in Proceedings of {\em Quark Matter '95},
Nucl. Phys. {\bf A} (1995) in press.
\bibitem{Enyo} H. Enyo, in {\em Properties and Interactions
of Hyperons}, ed. B. Gibson, P. Barnes and K. Nakai
(World Scientific, 1994, Singapore); KEK-PS proposal (1994).
\bibitem{SH} H. Shiomi and T. Hatsuda, Phys. Lett. {\bf B334} (1994) 281.
\bibitem{others}
K. Saito, T. Maruyama and K. Soutome, Phys. Rev. {\bf C40} (1989)
407;\\
H. Kurasawa and T. Suzuki, Prog. Theor. Phys. {\bf 84} (1990)
1030;\\
K. Tanaka, W. Bentz, A. Arima and F. Beck, Nucl. Phys. {\bf A528}
(1991) 676;\\
M. Jaminon and G. Ripka, Nucl. Phys. {\bf A564}
(1993) 505.
\bibitem{french} J. C. Caillon and J. Labasouque, Phys. Lett.
{\bf B311} (1993) 19
\bibitem{PW}
H. -C. Jean, J. Piekarewicz and A. G. Williams, Phys. Rev. {\bf C49}
(1994) 1981.
\bibitem{QM} See e.g., J. J. J. Kokkedee, {\em The Quark Model},
(Benjamin, 1969, New York).
\bibitem{gled} N. K. Glendenning et al., Phys. Rev. {\bf C48}
(1993) 889.
\bibitem{jaffe} R. L. Jaffe, Phys. Lett. {\bf B229} (1989) 275.
\bibitem{QHD} B.D. Serot and J.D. Walecka, Adv. Nucl. Phys. {\bf 16} (1986).
\bibitem{KS88} H. Kurasawa and T. Suzuki,
Nucl. Phys. {\bf A490} (1988) 571.
\bibitem{lipkin} H. Lipkin, Int. J. of Mod. Phys. {\bf E1} (1992) 603.
\bibitem{Yabu} H. Yabu, F. Myhrer and K. Kubodera, Phys. Rev.
{\bf D50} (1994) 3549, and references therein.
\end{thebibliography}
\vspace{2cm}
\centerline{\bf{Figure captions}}
\vspace{1cm}
\noindent
Fig.1: Ratio of baryon mass in matter $M_{B}^*$ and that in vacuum
$M_{B}$ ($B=N, \Lambda, \Sigma$) as a function of $\rho/\rho_0$.
\vspace{0.8cm}
\noindent
Fig.2: Ratio of the $\phi$-meson mass in matter $m_{\phi}^*$ and that
in vacuum $m_{\phi}$ as a function of $\rho/\rho_0$.
Only hyperon contributions are included in the $\phi$-meson self-energy.
\vspace{0.8cm}
\noindent
Fig. 3: Same with Fig.2 except that
only the nucleon contribution is included.
\end{document}