The Search for the QGP: A Critical Appraisal

Over the past 15 years, an extensive program of high energy nuclear collisions at BNL and CERN was devoted to the experimental search for the quark-gluon plasma predicted by QCD. The start of RHIC this year will increase the highest available collision energy by a factor 10. This seems a good time for a critical assessment: what have we learned so far and what can we hope to learn in the coming years? Invited talk at Lattice 2000: XVIII International Symposium on Lattice Field Theory, August 17-22, 2000, Bangalore/India; Proceedings to appear in Nucl. Phys. B. Also presented at the RHIC and AGS Users Meeting, August 7-8, 2000, Brookhaven National Laboratory, Upton/NY, USA, and at Bologna 2000: Structure of the Nucleus at the Dawn of the Century, May 29-June 3, 2000, Bologna/Italy. “What’s the good of Mercator’s North Poles and Equators, Tropics, Zones and Meridian Lines?” So the Bellman would cry: and the crew would reply “They are merely conventional signs!” Lewis Carroll, The Hunting of the Snark 1. Expecting the Unexpected QCD predicts that at high energy density, hadronic matter will turn into a plasma of deconfined quarks and gluons. Eighteen years ago, theorists and experimentalists met in Bielefeld to chart the course for the study of quark-gluon plasma formation in heavy ion collisions. It was noted then: “In the past decades, we have investigated the strong interaction of elementary particles at higher and higher energies. So far, we have not been able to study in parallel strongly interacting macroscopic systems at higher and higher [energy] density. The thermodynamics of very dense matter is largely an unexplored field for terrestial experimental physics” [1]. In 1986/1987, experimental studies of high energy nuclear collisions began at the BNL-AGS and the CERN-SPS. The AGS started with Si beams (at 15 GeV/nucleon) and later went on to Au beams (at 12 GeV/nucleon) incident on heavy targets. At CERN, O and S beams (at 200 GeV/nucleon) were followed in 1995 by Pb beams (at 160 GeV/nucleon). While both labs have by now provided a wealth of data on the hadronic final states produced in the different collisions, the AGS program did not include any experiment able to reach back into the early ‘hot’ stages of the collision evolution. Moreover, estimates in general indicate that with the AGS beam energy one can at best barely reach the initial energy densities required for quark-gluon plasma formation. For these reasons, most of our present knowledge about a possible onset of deconfinement in nuclear collisions has come from the CERN-SPS. At the beginning, the search for colour deconfinement in high energy nuclear collisions was very much an exploratory endeavor. As in most explorations, the enthusiasm of the explorers was greater than their knowledge of what to expect or look for. There was a general feeling that if the quark-gluon plasma was indeed produced, it would manifest itself in a variety of unknown but dramatic ways, including the end of the world; this, however, was ruled out [2]. So the basic idea in the planning of experiments was to cover all possibilities, hoping that in this way the QGP would be observed somehow, somewhere, preferably everywhere. Today, we are somewhat wiser. Changing conditions generally lead to a change in the related observables; if this happens in a monotonic way, it is indeed merely a conventional sign and has nothing to do with any phase transition. What we are really looking for is something like an ‘order parameter’ in statistical mechanics: an observable which remains constant over a range of different conditions, but which then at a certain point starts changing. Such non-analytic behaviour could be the consequence of a critical or transition point. It is not yet known if an experimentally accessible order parameter for deconfinement really exists.