Nuclear collisions at relativistic energies are the experimental tool developed in the past 20 years to form, study, and explore a locally `color-melted' space-time domain. At this time we are assembling comprehensive experimental evidence that a locally color deconfined state indeed is formed in present day experiments. I present here my personal assessment of the recent developments, current status, and a view of near-future opportunities in searching for, and studying the quark-gluon plasma (QGP) using hadronic flavor observables, strangeness in particular: the study of strange and also charmed hadronic particle production in nuclear relativistic collisions offers an opportunity to explore the physical properties of the deconfined quark-gluon phase. I survey here the development of the subject, recent accomplishments and future directions of this research program.
There are further and considerable research opportunities, as in the march towards higher energies where the deconfinement is nearly a sure bet, we may have jumped the collision energy domain in which the effects related to critical behavior could be studied at the conditions which just suffice to melt the strongly interacting vacuum structure. Where in energy this situation arises is presently still a hotly debated topic.
The physical ideas on which this research program depends are rather simple and hence dependable: if color deconfinement occurs, the quark-gluon soup will in due course equilibrate the flavor abundance between the light quarks u,d, that are brought into the collision and the initially absent heavier, on laboratory time scale short lived s quarks. However, given the short hadronic interaction scale of the lifespan of strangeness, is of course infinite, as is the somewhat shorter lifespan of charm. Different mechanisms of flavor pair production have been explored and in the deconfined region heavy quarks are primarily produced in gluon fusion reactions . For each s-quark made, there is its anti-s-partner, since in strong interactions only quark pairs can be produced. At QGP break-up, ready availability of already prepared strange and antistrange quark reservoir allows quark clustering into multi-strange final state hadrons without `penalty'. This collective effect, where key components of a final state particle are made in earlier reactions, is the source of our expectation that we can unravel the structure of the hadronic ball of hot and dense matter using strongly interacting observables.
One of the keys to the dense matter is the role of gluons, the color charged companions of photons. Gluons provide not only as the source of strangeness in QGP the key distinction between the deconfined quark-gluon phase and the confined hadron gas. Gluons play a major role in the dynamics of the QGP hadronization since they carry much of the QGP phase properties. Because gluons can be created and annihilated easily in diverse QCD based strong interactions involving other gluons and light quarks, the gluon density is most capable of all QGP components to follow the evolution of the exploding/flowing matter closely, maintaining the chemical equilibrium abundance. The formation time of QGP: t = 0.2--0.8 fm/c, is widely defined as the time needed for the gluon gas to reach initial absolute chemical equilibrium. Strangeness production by gluon fusion sets in at that time, and is studied assuming chemical gluon equilibrium.
Strange antibaryon production has been first explored in the early 1980's in central proton-proton interactions at CERN-ISR (Intersecting Storage Ring) by the AFS (Axial Field Spectrometer) collaboration where the available N--N CM-energy was four times greater than it is now under study in nuclear collision experiments at SPS. A few years later a similar study was performed at yet much higher CM-energies at the CERN-SppS collider by the UA5 collaboration, with relative yields remaining consistently small. Especially in view of these emerging much higher energy proton-proton and proton-antiproton experimental results, such an anomaly attained at much lower collision energies seemed either impossible or, to others, implied that deconfinement would occur at energies above and beyond studied so far in N--N reactions. In any case, if the theory was correct, and deconfinement probable, there was a hadronic signature worth to look for experimentally in the nucleus-nucleus interactions: larger volumes of deconfined hadronic matter when collided even at lower available energies should produce particle abundances with unexpected features, qualitatively different from the `elementary' interactions.
Strongly interacting flavor probes of dense matter have been from the beginning placed on the project menu of the CERN-SPS experimental nuclear collision program during the past decade. Several experiments, today referred to as NA57/WA97/WA94/WA85 and NA49/NA35, explored strange hadron production, including strange antibaryons, first in a 200 A GeV Sulphur (S) beam interactions with laboratory stationary targets, including the symmetric S--S reactions, and Sulphur collisions with `heavy' Silver (Ag), Gold (Au), Tungsten (W) or Lead (Pb) nuclei at 200 A GeV, and, more recently, moving on to the Pb--Pb collisions at 158 A GeV. These results obtained at available energy of 8.6--9.2 GeV per participating nucleon will be soon complemented by strange antibaryon data from 11 GeV Au-induced reactions at the BNL-AGS. Analysis of the global hadronic yields shows a great similarity in the behavior of dense matter at the AGS and SPS energies, despite 4 times higher available energy at the SPS. The key difference between the SPS and the AGS experiments is that hadronic particles produced at the lower AGS energies show features characteristic of confined matter chemical equilibration, such as is a finite non-negligible strange quark chemical potential. In principle, this is not in contradiction to possible formation of deconfined baryon rich quark matter at AGS energies, since hadronic abundance equilibration could be just a final state effect. Moreover, since at AGS energies the specific entropy per baryon expected in confined and deconfined matter is very similar to the entropy of QGP, there is to this date no clear evidence for or against QGP formation in the dense baryon rich matter fireballs seen at AGS. In principle measurement of strange antibaryon production should resolve the issue in the near future.
Measurement of strange antibaryons is not easy and experiments need to be designed for the task as the relatively rarely produced strange antibaryons are literally buried in much more abundant mesons. Ways have to be designed to eliminate that background without loosing the acceptance which is required to observe mesons emerging from the self-analyzing weak strange (anti)baryon decays. The degree of difficulty is seen in the (preliminary) NA49 result for Pb--Pb collisions which has a precision of 30%, so within 3s.d. (standard deviations) one could claim that the ratio of anti-lambdas to antiprotons is negligible, even though quite evidently in Pb--Pb collisions it is significantly greater than unity, defying the normal hadron production rules.
On the other hand a strangeness enhancement effect has been studied now in considerable detail and there is a well reported evidence assembled over the past 10 years in all the pertinent BNL and CERN experiments. Without reviewing these results in every detail, I note that in several slightly varying definitions, this enhancement is reported to be a factor two or three, as it has been expected if QGP is formed. Moreover, this enhancement is consistently rising with the multi-strangeness of hadrons. Moreover, there is enhancement of single-strange hadrons (factor 1.5 --- 3), and of negative hadrons (factor 1.5). This enhancement is understood as an increase of entropy production, an effect expected to occur when deconfienemnt sets in, and color bonds are broken. But the crucial result is that the yield increase of multi-strange hadrons is in general much stronger, reaching factor 4.5 for doubly (anti)strange anti-cascades and above 10 for the triple strange Omegas. This effect is the anticipated and specific signal of formation and breakup of a deconfined space-time region, and has so far eluded models that rely on confinement.
The AGS experiments were initially focused on other `strange' issues, and in particular many resources have been vested to seek strange nuclear matter. Only in recent time the study of strange antibaryons as a possible signature of the deconfined state has commenced, for there are strong indications e.g. in the E802/E859/E866 experiment series that the yield of anti-lambdas exceeds the yield of anti-protons at AGS energies. There is a continued and considerable effort devoted in experiments E917 and E895 to measure systematically strange particle yields as a function of energy up to 11A GeV. It is an interesting race against time, for the completion and availability of RHIC is likely to overpower AGS experimental efforts in the not too distant future.
To proceed we need to clarify the understanding of the different local chemical equilibria. We distinguish two cases for their different reaction time scales. The slower absolute chemical equilibrium, which requires that the particles are completely filling the available phase space; the faster, relative chemical equilibrium, here only the distribution of some property (e.g., strange quark flavor) among different carriers (particles) according to the relative phase space size is required. Several alternative evolution scenarios of dense matter are studied today:
There is the caveat that since the phase space density of strangeness in QGP and HG phases is different, the phase space populations must be appropriately adjusted to infer from the observed final state yield of strange quarks per baryon number the conditions in the earlier QGP phase --- naturally in such an approach the assumption is made that the baryon number is conserved. The important point here is that once this procedure is carried out, in general this leads to significant overpopulation of strange particle equilibrium phase space abundance in the presumed QGP. However, even more important is the second observation, namely that the most chemical analysis performed to-date, which reported that strangeness phase space was saturated up to about 75% in the final hadronic phase, have indeed been reporting the ratio of strange to non-strange phase-space saturation. The non-strange quark occupancy is at 150--200% of the equilibrium, as is noted by the excess of mesons (excess entropy). Allowing for this effect we see that the strangeness phase space is indeed overpopulated in the HG phase. Thus it appears as if the chemical strangeness equilibrium were approached from above and not from below, as would be the case in a microscopic confined hadron production process. This implicates again a new mechanism of strangeness formation, and indeed it is virtually impossible to imagine other mechanisms that could lead to a strong inversion of a population of `heavy' flavor, but deconfinement. It is important to repeat that to see this effect the light quark abundance must not be assumed implicitly to be at the chemical equilibrium reference point. Overpopulation of the light quark phase space is also expected on different grounds: the glue degree of freedom and some quarks must fragment in the hadronization to preserve the source entropy enhanced by the melting color degrees of freedom.
As particle emerge from the QGP hadronization not only their abundance but also their spectra are probably as unusual as they are experimentally reported. Indeed, some bold conclusions about the origin of the strange antibaryons can be inferred from the remarkable fact that the transverse mass spectra are so similar for particle-antiparticle pairs
There are almost too many different theoretical approaches to this vast research field and the question really is not which method is better or correct, but why do we need so many different theoretical approaches? Nobody doubts that fully kinetic microscopic reaction models are superior to statistical models (and hence this is what we all should be doing?), but some of us also realize that, in practice, microscopic approaches suffer critically from the need to understand and be able to model all relevant and accessible reaction mechanisms, including the novel phenomena that are yet to be discovered. Thus practice calls for compromise solutions, in which some aspects are considered to be precisely described in a simpler (statistical) manner. Moreover, the degree to which the reaction is treated as a quantum mechanical process, as compared to classical two particle cascades is not understood in principle. In fact, even if we had much greater computing power available allowing us to compute within fundamental theoretical approaches such as lattice-QCD, we would not know how to treat the dynamics of the collision, the transition from quantum to classical dynamics, the non-equilibrium aspects and many other issues. So the multitude of approaches really reflects on the explorative character of the research we are engaged in.
Each approach has its pros and cons, and clearly one has to be aware of this when addressing the flavor observables. The thermal models suffer from the perception that kinetic equilibrium is introduced `deus ex machina', without proper understanding of the dynamics and time scales involved. On the other hand, even the best microscopic models cannot be used to interpret results without prolonged fine tuning the various implicit and explicit reaction assumptions made. By comparison, the reaction dynamics in local equilibrium approaches are usually relatively simple and often allow to come to model-independent conclusions about observables. The simplicity of the statistical model allows to connect several measured objects together and thus a test of the physical understanding is rapidly at hand. The issue here is that we must scrupulously distinguish which quantities can be placed in equilibrium, and which require a non-equilibrium description, if we wish to take advantage of the powerful methods of quantum statistical mechanics. If a (near) equilibrium description is possible, it is the better way of approach.
Given the phase transition the community is undergoing at present: the exploratory fixed target programs at SPS and AGS are coming to an end, while dedicated measurements of the properties of the deconfined phase are likely to begin at RHIC, where in fact the situation is completely unknown. However, the discussion we have presented provides for a clear shopping list which we will need to fill in next few years of theoretical research in this field:
for more details see:
and:
submitted to J. Phys. G
and
Publications after1991
Conference Proceedings after1991
October 14, 1998