Relativistic heavy ion collisions in the central regime are
characterized by the formation of dense, hot hadronic matter or
perhaps, with luck, droplets of quark-gluon plasma. This zone of
dense matter hadronizes into a large multiplicity of particles: in
addition to an abundant supply of pions, one also finds antiparticles
and strange particles 
in the
collision debris. Occasionally, composite objects can be formed, such
as antinuclei 
and hadronic
systems with multiple units of strangeness S (strangelets, the
H-dibaryon, strange chiral solitons, etc.), assuming that the latter are
indeed stable with respect to strong decay. Such objects are formed
in only a tiny fraction of all heavy ion collisions, so an experiment
of very high sensitivity is required to detect their presence, or to
extract meaningful constraints on theoretical models from their
absence. E864 is such an experiment, with unprecedented sensitivities
in the range 
to 
per collision.
Why heavy ions? If we focus on the excitation of the strangeness
degree of freedom in hadronic collisions, heavy ion collisions offer
the only practical method of creating systems with multiple
strangeness S, say 
. Reactions induced with
strange meson beams, for instance 
or 
processes, can be used to produce single- and double- 
hypernuclei, but only heavy ions enable us to go beyond the production
of S=-1,-2 objects. In a heavy ion collision, each of the
independent nucleon-nucleon collisions can lead to the creation of an
strange quark-antiquark pair. It is already known from the
first round of AGS experiments with Si beams at 15 GeV/nucleon
that substantial numbers of strange particles are produced in
central collisions [14]. For Au-Au collisions at the
AGS, we anticipate that 10 - 15 's will be produced in a
typical central collision, permitting the exploration of a significant
domain in strangeness and baryon number in experiment E864. We return to some
more detailed theoretical estimates later.
E864 is an experiment with strong interdisciplinary aspects, with
important implications in both nuclear and elementary particle
physics. A fundamental question concerns the stability of strange
quark matter. The strangelets proposed in Refs. [3] and
[7] are stable in a certain domain of the
underlying parameters of the Bag Model of quantum chromodynamics
(QCD), namely the strange quark current mass 
, the bag constant
B, and the QCD coupling constant 
. The effective
parameters 
determined by a fit to baryon
(A=1) and meson (A=0) spectra cannot be reliably used to predict
the binding energy of systems of larger (A,S), as emphasized by Ref.
[7]. In particular, it is difficult to predict the
minimum values of A and S for which strangelets are stable,
even if one is convinced of the stability of such objects in the bulk
limit (large A, with 
). The question of the existence of
strangelets, both large and small A, is one that must be
resolved experimentally. This is a task that E864 can accomplish. Thus a
dominant theme among the particle physics aspects of E864 is
non-perturbative QCD: the limits we obtain on the production of
multi-strange objects can provide significant constraints on
effective models for the non-perturbative regime. As an example,
consider dibaryons with A=2, S=-2. The six-quark Bag Model [1]
leads to the prediction of a stable H-dibaryon
in the SU(3) limit, while a version of the chiral
soliton model [15] leads to a stable
bound state (I=2). Both of these models are
consistent with the observed spectroscopy of strange baryons.
This illustrates the difficulty of extrapolating
effective models of non-perturbative QCD, even by one unit in baryon
number! As discussed in detail later, the expected production rates
of the putative bound state is well within
the sensitivity of E864, and a meaningful limit on H production may be
possible within the E864 setup.
The stability of strange quark matter is related to other problems in astrophysics and cosmology. For instance, if strangelets exist, they could be a candidate for ``dark matter,'' as originally envisioned in Ref. [3]. The astrophysical problem of the strangeness content of the cores of neutron stars is also a closely related one, involving the equation of state of high density, low temperature hadronic matter.
Thus far we have mentioned only some specific motivations for E864
based on various speculative theoretical models. However, one should
also emphasize the flexible and global nature of the open geometry,
``wide band'' approach adopted here, which is capable of accommodating
the unexpected. E864 can be regarded as a general, high sensitivity
search for new neutral or charged particles. We will
provide limits on the production of fractionally-charged free quarks,
as well as unanticipated neutral particles. For instance, there is the
possibility, however remote, that SU(3) color symmetry is slightly broken
in such a way that free gluons are produced rather than free quarks. Slansky
et al. suggest that SU(3) is spontaneously broken to
SU(2) 
U(1) [16], and Saly et al. report that free
gluons may arise as a result of dynamical symmetry breaking [17].
The phenomenological aspects of these suggestions were considered by Rinfret
and Watson [18], and Berezinsky et
al. [19]. The experimental signature of the hypothetical
ninth gluon would be distinctive: it would appear as an massive neutral
hadron.
One might argue that the
hot, dense hadronic soup resulting from a central collision of very
heavy ions provides enhanced prospects for producing such new particles.
The sensitivity of this experiment offers significant
discovery potential.
E864 will also address a number of questions of fundamental interest
in nuclear physics, such as the production rates or limits for a number of
light nuclei,
some rather ordinary ones 
and some unusual ones 
. Excellent limits can be obtained
on the production of neutron rich nuclei, for
instance 
or 
,
which have been searched for in numerous other experiments and not
found, presumably because they are unstable with respect to strong
emission of neutrons. The rates for central production of light nuclei
provide a stringent test of coalescence models [20], in
which such composite objects are formed at a late stage of the
reaction process (``freezeout'') from baryons which overlap in phase
space. Such a coalescence picture works rather well at BEVALAC
energies of 0.4 - 2 GeV/A [21], and it is
important to establish whether this success extends to AGS energies,
or whether additional cluster production mechanisms enter due to the
formation of a hot and dense intermediate state of hadronic matter or
even a quark-gluon plasma.
The rates for production of antinuclei
in heavy ion collisions are a sensitive probe of the time dependent
reaction dynamics. There is some preliminary evidence for the
production of 
's in heavy ion collisions at AGS
energies [22], at a rate below that expected on the basis of the
coalescence model. Antimatter is strongly absorbed in nuclear matter,
due to annihilation processes such as 
's or 
's. Thus , 
, and
abundances will be very sensitive to hadron densities, a
``formation time'' for 
's, and other features of the
dynamics. Due to the very high sensitivity, E864 has a fair
chance to observe and .
The rates for light nuclei formed in central collisions (as
contrasted with beam fragmentation) are interesting in their own right as
tests of models for heavy ion reaction dynamics, but they also serve
to illuminate the coalescence production mechanism which is so important in
understanding the meaning of the strangelet search. From the
A dependence of production rates, one deduces a ``penalty
factor'' for the addition of another nucleon to a cluster. From
theoretical estimates or eventual measurement of cross sections for
or 
hypernuclear production (E864 is not
sensitive to these because of their short lifetime 
ns), one obtains a similar ``penalty factor'' for the addition of a
unit of strangeness to a cluster. Thus one can use measured rates for
nuclear clusters as a baseline for coalescence estimates of
strangelet production. We return to this point later.
We summarize the main areas of emphasis of E864:
ns)
multi-strange quark matter,
We now provide a more detailed discussion of a), b), and c).