A search for the production of direct photons in S+Au collisions at 200 A GeV has been carried out in the CERN-WA80 experiment. For central collisions the measured photon excess at each pT , averaged over the range 0.5 GeV/c pT 2.5 GeV/c, corresponded to 5.0% of the total inclusive photon yield with a statistical error of stat =0.8% and a systematic error of syst =5.8%. Upper limits on the invariant yield for direct photon production at the 90% C.L. are presented. Possible implications for the dynamics of high-energy heavy-ion collisions are discussed. (Submitted to Physical Review Letters) 1) Gesellschaft f ur Schwerionenforschung, D-64291 Darmstadt, Germany. 2) Kurchatov Institute of Atomic Energy, Moscow 123182, Russia. 3) Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831. 4) University of M unster, D-48149 M unster, Germany. 5) Lawrence Berkeley Laboratory, Berkeley, California 94720. 6) University of Lund, S-22362 Lund, Sweden. 7) Brookhaven National Laboratory, Upton, New York 11973. 8) University of Tennessee, Knoxville, Tennessee 37996. 9) Kernfysisch Versneller Instituut, University of Groningen, NL-9747 AA Groningen, Netherlands. ) Deceased Directly radiated thermal photons have long been considered an interesting penetrating probe with which to study the early phase of the hot and dense matter produced in ultra-relativistic nucleus-nucleus collisions. Single \direct" photons are expected at high transverse momentum, pT , from well-known hard QCD processes, but also possibly in the pT region below several GeV/c due to thermal radiation from the hot dense matter [1]. Since the mean free path of the produced photons is considerably larger than the size of the nuclear volume, photons produced throughout all stages of the collision will be observable in the nal state. Thus, it is believed that the emitted photons should provide information about the initial conditions of the hot dense system and thereby provide evidence for the possible formation of a Quark Gluon Plasma (QGP). The search for direct photon production in ultra-relativistic nucleus-nucleus collisions has been a major emphasis of the WA80 experiment at CERN. First results from WA80 found no excess photon yield beyond that attributable to resonance decays in central collisions of O+Au at 200 A GeV, setting an upper limit of = 0 < 15% [2]. The preliminary results of the 1990 WA80 S+Au photon analysis showed no signi cant excess in peripheral collisions, while an excess at about the 2 level was seen in central collisions [3]. Although preliminary, these results have generated a great deal of theoretical interest [4, 5, 6, 7, 8]. In this Letter we report the nal results of the WA80 S+Au direct photon analysis, we compare the nal results to theoretical calculations, and discuss the implications towards the possible formation of a QGP. The WA80 experimental setup for the 1990 run period with 200 A GeV S beams was upgraded from that used for the previous run periods with O and S beams [2, 3, 9]. The direct photon sensitivity for this data set, relative to the O data [2], was improved by several factors [3] including an increased data sample, an increased detector coverage, a coverage closer to mid-rapidity, and improved analysis techniques. The WA80 photon spectrometer consisted of a nely segmented electromagnetic calorimeter composed of 3798 lead-glass modules with photomultiplier tube readout. The lead-glass was arranged into three independently calibrated arrays, of roughly equal size. Two of the arrays consisted of TF1 lead-glass of 4 cm x 4 cm x 40 cm (15 X0) [10] deployed as towers to the left and right of the beam axis. The third array, located below the beam axis, was the SAPHIR lead-glass detector [11] used in the WA80 O run period [2] which consisted of SF5 lead-glass modules of 3.5 cm x 3.5 cm x 46 cm (18 X0). The entire photon spectrometer provided coverage of from 1/10 to 1/2 of full over the rapidity range of 2:1 y 2:9. Immediately in front of the photon spectrometer was a double-layer charged-particle veto (CPV) counter which covered the lead-glass region of acceptance. Each layer of the CPV consisted of streamer tubes with charge-sensitive pad readout, with pads of dimension similar to the lead-glass modules [12]. For the direct photon analysis the total event sample of 6:27 10 events was divided into various centrality classes based on the measured transverse energy. The total transverse energy was measured in the WA80 mid-rapidity calorimeter [13] which had full coverage over the pseudo-rapidity range 2:9 5:5 and partial coverage extending to 2:4 . In this Letter, results are presented for the most peripheral events corresponding to 31% mb and the most central events corresponding to 7:4% mb, with mb = 3600 mb [15]. This central event class corresponds to the complete geometrical overlap of the S nucleus with the Au target, with an average of 107 participating nucleons (to be compared to an average of 5.6 participating nucleons for the peripheral event class), in contrast to the less restrictive centrality condition of 25% mb used in the preliminary analysis [3, 14, 15]. In the WA80 experiment, the 0 and yields have been measured simultaneously

Correlations between protons are studied in the target fragmentation region of reactions of protons and16O with C, Cu, Ag, Au and of32S with Al and Au at 200A GeV. The emitted protons were measured with the Plastic Ball detector in the WA80 experiment at the CERN SPS. The comparison of the correlation function with calculations, assuming a spherical, gaussian shaped source with a lifetime τ=0 fm/c, allows the extraction of radius parameters. The values are very close to those expected from the geometry of the target nuclei and increase with the target mass as αA Target 1/3 . Even in proton induced reactions the whole target nucleus is involved. The dependence of the radii on centrality, polar angleθlab, and energy, and their relation to measured proton yields are presented.

Minimum Bias production cross sections of η mesons have been measured in 200 AGeV/c S+Au and S+S collisions at the CERN SPS by reconstructing the η→γγ decay. The measurements have been made over the rapidity range 2.1 ≤ y ≤ 2.9 using the leadglass spectrometer of WA80. Within the statistical and systematical uncertainties the spectral shapes of π^0 and η mesons yields are identical when their invariant differential cross section is plotted as a function of the transverse mass. The relative normalization of the η to π^0 transverse mass spectra is found to be 0.53 ± 0.07 for S+Au and 0.43 ± 0.15 for S+S reactions. Extrapolation to full phase space leads to an integrated cross section ratio of η to π^0 mesons of 0.147 ± 0.017 (stat.)± 0.015 (syst.), and 0.120 ± 0.034 (stat.)± 0.022 (syst.) for S+Au and S+S collisions, respectively.

The investigation of highly excited and compressed nuclear matter created in relativistic heavy ion collisions requires probes sensitive to the different stages of the development of the system. Among the characteristic signals listed in table 1, hadrons mainly probe the late stage of the reaction and their final spectra are influenced by their last scatterings. Electromagnetic probes, on the other hand, escape from the reaction zone without rescattering and probe the very early stage of the reaction, where temperatures and densities are highest. Electromagnetic probes are therefore particularly useful to study the possible formation of a Quark-Gluon Plasma, which is expected only at extreme values of temperature and density. The disadvantage of electromagnetic probes is that the signals are usually small and have to be disentangled from a tremendous physical background requiring a very high experimental precision and detailed treatment of the various background sources. In the case of photons this background consists mainly of photons from γ decaying hadrons like π0 and η.

Correlations between protons are studied in the target fragmentation region of reactions of protons and O-16 with C, Cu, Ag, Au and of S-32 With Al and Au at 200 A GeV. The emitted protons were measured with the Plastic Ball detector in the WA80 experiment at the CERN SPS. The comparison of the correlation function with calculations, assuming a spherical, gaussian shaped source with a lifetime tau = 0 fm/c, allows the extraction of radius parameters. The values are very close to those expected from the geometry of the target nuclei and increase with the target mass as proportional to A(Target)(1/3) . Even in proton induced reactions the whole target nucleus is involved. The dependence of the radii on centrality, polar angle theta(lab), and energy, and their relation to measured proton yields are presented.

Direct thermal photons in the pT range of 0–5 GeV/c are expected to provide a sensitive probe of the hot dense matter formed in the early stage of relativistic heavy ion collisions. The production of single photons in 200 A GeV S + Au reactions has been investigated using the 3800 element Pbglass calorimeter of CERN experiment WA80. Neutral π0 and η cross sections have been measured via their two-photon decay branch yields. In a first analysis of the WA80 results, a slight excess photon yield above that which may be accounted for by hadronic decays was observed for central collisions. A report on the status of the reanalysis of this data is presented.

Correlations of positive pions and protons measured with the Plastic Ball detector in ultrarelativistic nucleus-nucleus collisions are studied. Source parameters are extracted for various projectile-target combinations. While the proton source can be explained by geometry, the pion source shows more subtle effects, which may be related to a very large source component. A comparison with the RQMD model for 200 AGeV32S + Au reactions reveals discrepancies for the pion source.

Neutral pi0 and eta spectra have been obtained over a wide transverse momentum range for various centralities of S + Au collisions at 200 GeV/nucleon. The analysis was done of the full statistics of the WA80 experiment at CERN using the two-photon invariant mass spectra. Compared to the previous analysis [1, 2] the pi0 and eta spectra were obntained over an extended p(t) range of 0.2 < p(t)(pi0) < 4.4 GeV/c and 0.4 < p(t)(eta) < 3.0 GeV/c.

Results from the experimental program with light ion beams and heavy target nuclei at the CERN SPS could demonstrate the occurrence of an unprecedented state of high density in hadronic matter. The thermal nature of the hadronic system has been investigated by analyzing spectra and production ratios of hadrons which reveal a large degree of rescattering of primary and secondary hadrons.Thermal photons from elementary quark-gluon interactions are considered a promising signal for the occurrence of a phase transition to the quark-gluon plasma. The predictions for thermal photons from elementary parton interactions are discussed and compared to the thermal emission rate of photons from a hot hadronic gas. Recent results from the photon spectrometers in heavy ion experiments are presented. Production cross sections of pi0 and eta mesons are determined and the projectile and target mass dependence is discussed. An upper limit for the single photon yield was determined for central O+Au reactions. Recent S+Au reactions exhibit an excess of photons over the yield expected from hadronic decays. The spectral shape of the expected single photon signal is discussed which might reveal the temperature of hot matter and indicate a phase transition.

We have studied one- and two-dimensional scaled factorial moments in S-32+S and S-32+Au collisions at 200 GeV/nucleon in a high statistics electronic measurement at the CERN SPS using pad-readout streamer tubes. We observe no intermittency signal beyond that produced by folding the FRITIOF event generator with a detailed model of our detector. The systematic effects of detector response, two-track separation, and finite statistics in a factorial moment analysis are discussed in detail. Even though the observed signal contains measurable distortions due to these experimental effects, we show that we are sensitive to intermittency. As an alternative method, a two-particle correlation function analysis was applied to the same data to measure correlated particle production at small scales. We show that this method does not suffer as much as the factorial moment analysis does from distortions due to the limited two-track resolution of the detector. The correlation functions also agree with the predictions of FRITIOF filtered through our detector simulation, down to the limit of the two-track resolution. Since FRITIOF models nucleus-nucleus collisions by the superposition of nucleon-nucleon collisions, we conclude that there is no evidence in our data of the kinds of collective behavior predicted to give strong intermittency in heavy ion collisions.

The construction and operation of light emitting multistep avalanche chambers will be described. Based on a preliminary analysis of data taken during the first Pb ion run at CERN the performance and operation of the tracking system will be discussed.

Measurements of slow, singly charged fragments in the target rapidity region have been performed for proton and pion induced reactions with various nuclei at 200 GeV/c. Multiplicity, angular and energy distributions are examined and used to study the effects of rescattering in the nuclear medium. Data are compared to a “geometric cascade model” and to simulations with the VENUS 3.11 and the FRITIOF 1.7 Monte Carlo codes.

Results on pseudorapidity distributions, azimuthal correlations, and on two particle interferometry for identified protons and pions measured in the target fragmentation region of high energy proton-nucleus and nucleus-nucleus collisions are presented. Full participation of the whole target nucleus is observed even for 200 GeV p + Au collisions. Incorporation of strong rescattering phenomena into string models, including pion absorption effects in excited target matter, provides a good description of most experimental data.

Using a 3790 module lead glass photon spectrometer π° and η mesons have been measured in 200 A · GeV S + Au collisions by invariant mass reconstruction. Centrality selected transverse momentum π° spectra have been obtained for 0.5 ≤ pT ≤ 4.0 GeV/c displaying a much flatter slope for central compared to peripheral data. The problems of extracting single photon to π° yields from the data are discussed and the total error of their determination is given for present and future experiments. It is concluded that an accuracy of about 7 % can be obtained in the γπ° ratio.

Multiplicity distributions of slow singly charged target fragments from 16O induced reactions on C, Cu, Ag and Au are studied at 60 and 200 A GeV. The distributions from the two energies are essentially the same. Energy and angular distributions for slow protons are revealing the important role of cascading and rescattering in the spectator matter. The target dependence of the multiplicity of slow fragments exceeds AT23 for the heaviest targets.

Correlations between positive pions are investigated in the target fragmentation region of 200A GeV16O+nucleus collisions. The pions are measured with the Plastic Ball detector in the WA80 experiment at the CERN SPS. The target mass dependence of the radii and the correlation strength extracted by interferometry is studied. A new approach to the fit of the correlation function is introduced. The correlation strength and both invariant and transverse radii increase with decreasing target mass. The transverse radius for16O+C reactions appears to be much larger than the geometrical radius of the nuclei involved. For the Au target only a small fraction of the measured pions contributes to the apparent correlation. Hints for a much larger second component in16O+Au reactions are observed. Rescattering phenomena may provide a clue to understand these phenomena.