Quark Matter Quark Matter Proceedings of the Sixth International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions - Quark Matter 1987 Nordkirchen, FRG, 24-28 August 1987 Editors: H. Satz, H.J. Specht, and R. Stock With 320 Figures Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Professor Dr. Helmut Satz Fakultat fiir Physik, Universitat Bielefeld, Postfach 8640 D-4800 Bielefeld, Fed. Rep. of Germany Professor Dr. Hans Joachim Specht 1. Physikalisches Institut, Universitat Heidelberg, Philosophenweg 12 D-6900 Heidelberg, Fed. Rep. of Germany Professor Dr. Reinhard Stock Institut fiir Kernphysik, Universitat Frankfurt, August-Euler-Str. 6 D-6000 Frankfurt 90, Fed. Rep. of Germany This book originally appeared as the journal Zeitschrift f1ir Physik C - Particles and Fields, Volume 38, Number 1/2 (ISSN 0170-9739) © Springer-Verlag Berlin, Heidelberg 1988 ISBN-13: 978-3-642-83526-1 e-ISBN-13: 978-3-642-83524-7 DOl: 10.1007/978-3-642-83524-7 Library of Congress Cataloging-in-Publication Data. International Conference on Ultra-Relativistic Nucleus Nucleus Collisions (6th: 1987 : Nordkirchen, Germany) Quark matter: proceedings of the Sixth International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions: quark matter, 1987, Nordkirchen, FRG, 24-28 August 1987/ editors, H. Satz, H.J. Specht, R. Stock. Published also as v. 38, no. 1/2 of the Zeitschrift fur Physik C-Particles and fields. Bibiography: p. 1. Quarks-Congresses. 2. Hadrons-Congresses. 3. Heavy ion colli sions-Congresses. I. Satz, H. II. Specht, H.J. (Hans Joachim), 1936-. III. Stock, R. (Reinhard), 1938-. IV. Title. QC793.5.Q252I57 1987 539.7'21-dc 19 88-20011 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only per mitted under the provisions of the German Copyright Law of September 9, 1965, in its version of June 24, 1985, and a copyright fee must always be paid. Violations fall under the prosecution act of the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1988 Softcover reprint of the hardcover 1st edition 1988 The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting, printing and binding: Universitatsdruckerei H. Sturtz AG, D-8700Wurzburg 2155/3150-543210 - Printed on acid-free paper Contents Preface. By R. Santo. H. Satz. H.J. Specht. R. Stock . 1 Oxygen-induced reactions at 60A GeVand 200A GeVstudied by calorimetry By S.P. Sorensen et a!.. WA80 Collaboration (With 6 Figures) . 3 Transverse energy distributions in 160-nucleus collisions By F. Corriveau et al.. HELlOS Collaboration (With 2 Figures) . 15 Study of the energy flow in 160-nucleus collisions at 60 and 200 GeV/nucleon ByW. Heck et al.. NA35 Collaboration (With 14 Figures) . 19 Measurement of energy and charged particle emission in the central rapidity region from O+A and p+A collisions at 14.5 GeV/c per nucleon and preliminary results from Si +A collisions. By L.P. Remsberg et al.. E 802 Collaboration (With 12 Figures) . 35 Transverse energy distributions in Si-nucleus collisions at 10 Ge V/nucleon By P. Braun-Munzinger et al.. E 814 Collaboration (With 7 Figures) . 45 Charged particle spectra in 160 induced nuclear collisions at the CERN SPS By 1. Lund et a!.. WA 80 Collaboration (With 8 Figures) . 51 Measurement of multiplicity distributions in oxygentungsten collisions at 200 Ge V per nucleon. By J. Schukraft (With 6 Figures) . 59 Review of high energy heavy ion reactions in emulsion. By I. Otterlund (With 13 Figures) . 65 First results from the hybrid emulsion experiment on 160-nucleus collisions at 200 Ge V/N By L. Ramello (With 7 Figures) . 73 Pion interferometry with ultrarelativistic heavy-ion collisions from the NA 35 experiment ByT.J. Humanic et al.. NA 35 Collaboration (With 5 Figures) . 79 Spectra of negative particles and photons in collisions of p-W and loO_W at 200 Ge V/u By H.-W. Bartels (With 3 Figures) . 85 Negative particle production in nuclear collisions at 60 and 200 Ge V/nucleon By H. Strbbele et al.. NA 35 Collaboration (With 11 Figures) . 89 Neutral transverse momentum spectra in 60 and 200A· GeVl60 + nucleus and proton + nucleus reactions. By H. Lbhneret al .. WA 80 Collaboration (With 8 Figures) . 97 Transverse momentum systematics in proton-proton and light ion collisions at the ISR By H.G. Fischer (With 7 Figures) . 105 Target fragmentation in proton-nucleus and 160-nucleus reactions at 60 and 200 GeV/nucleon By H.R. Schmidt et al.. WA 80 Collaboration (With 12 Figures) . 109 The production of JIIJf in 200 GeV IA oxygen-uranium interactions By A. Bussiere et al.. NA38 Collaboration (With 11 Figures) . 117 First results on strangeness production in 60 and 200 Ge V/nucleon heavy ion reactions from the NA 35 streamer chamber By G. Vesztergombi et a!.. NA 35 Collaboration (With 5 Figures) . . 129 Astudy of 1f and K production in proton-uranium and oxygen-uranium interactions at 200 GeVIA using decay muons By P. Sonderegger et al.. NA 38 Collaboration (With 18 Figures). 133 Preliminary spectrometer results from E-802 ByY. Miake et al.. E802 Collaboration (With 5 Figures) . 135 The QCD vacuum and quark-gluon plasma. By E.V. Shuryak (With 5 Figures). 141 Lattice QCD at finite temperature: a status report. By F. Karsch (With 10 Figures) . 147 Connection between perturbation theory and lattice QCD. By K. Kajantie (With 3 Figures) 157 On a non-perturbative pressure effect in lattice QCD By M.l. Gorenstein. O.A. Mogilevsky (With 2 Figures). 161 The '"instanton liquid"". By E. V Shuryak (With is Figures) . 165 Quarks and gluons in colour fields. By A. Bialas. W. Czyz (With 6 Figures) . 173 A unified approach for hadronic and nuclear collisions: the dual parton model By A. Capella. J. Tran Thanh Van (With 16 Figures) . 177 The materialization phase in the colour rope picture. By J. Knoll (With 4 Figures) . 187 Ultra-relativistic heavy ion collisions in a multi-string model. By K. Werner (With 6 Figures) 193 Multiplicity distributions in hadronic and heavy ion reactions. By R.M. Weiner 199 The approach to equilibrium in a quark-gluon plasma. By U. Heinz. 203 Gluon transport equations. plasma oscillations. and screening. By H.-Th. Elze . 211 Central region in relativistic heavy ion collisions; results from hydrodynamic calculations and cascade simulation. By P. V Ruuskanen (With 12 Figures) . 219 Initial state formation in the hydrodynamical theory of nucleon-nucleon collisions By E.L. Feinberg. 229 Rapidity distributions and energy densities from hydrodynamical studies at 5-200 GeV/n ByT. Rentzsch. G. Graebner. J.A. Maruhn. H. Stocker. W. Greiner (With 14 Figurcs) . 237 JllJf Suppression by plasma formation. ByT. Matsui (With 1 Figurc) . 245 Mass shifts of charmoniums and electromagnetic signals ByT. Hashimoto. K. Hirose. T. Kanki. O. Miyamura (With 3 Figures) 251 Searching photon signal of quark-gluon plasma formation ByVV. Goloviznin. A.M. Snigirev. G.M. Zinovjev. 255 Quark-matter diagnostics: dileptons. photons and (J/lfI) suppression By B. Sinha (With 6 Figures) . 259 Perturbative aspects of dilcpton production at finite temperature By R. Baier. B. Pire. D. Schiff (With 1 Figure) . 265 Strong interacting probes for quark gluon plasma. By P. Koch (With 3 Figures) . 269 /1(/1) Longitudinal polarization: a signature for the formation of a quark-gluon plasma in heavy ion collisions. By M. Jacob. 273 Enhanced multiplicity fluctuation as a possible signature of quark matter. By R.C. Hwa . 277 Emission of droplets of strange quark matter in RHIC By C. Greiner. D. Rischke. H. Stocker. P. Koch (With 10 Figures) . 283 Dense nuclear matter: supernovae and heavy ions. By G.E. Brown (With 10 Figures) 291 Quark matter in astrophysics and cosmology. By A. V Olin to (With 3 Figures) 303 Hadron-quark phase transition in dense stars. By F. Grassi (With 4 Figures) . 307 The EMC effect - status and perspectives. By K. Rith (With 6 Figures) . 317 The ion programme at CERN: a brief outline. By P. Darriulat. 325 Lead beam experiments at the CERN SPS. By A. Sandoval (With 10 Figures) . 329 The time projection chamber for heavy-ion collisions: trends and perspectives By F. Sauli (With 22 Figures) . 339 The relativistic heavy ion collider project: a status report By T. W. Ludlam. N .P. Samios (With 6 Figures) . 353 Quark Matter 1987: concluding remarks. By M. Gyulassy . 361 Preface This volume contains the papers which have been presented at the Sixth International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions, Quark Matter 1987, held in Nordkirchen from August 24-28, 1987. The conference was attended by about 250 scientists, representing 75 research institutions around the world - the scientific community engaged in experimental and theoretical studies of high energy nuclear collisions. The central theme of the meeting was the possibility of achieving extreme energy densities in extended systems of strongly interacting matter - with the ultimate aim of creating in the laboratory a deconfined state of matter, a state in which quarks and gluons become the active degrees of freedom. High energy accelerator beams and cosmic radiation projectiles provide the experimental tools for this endeavor; on the theoretical side, it is intimate ly connected to recent developments in the non-perturbative study of quantum chro modynamics. Phase transitions between hadronic matter and quark-gluon plasma are of basic interest also for our understanding of the dynamics of the early universe; they may be relevant for the behaviour of neutron stars, or even for the structure of certain exotic astrophysical objects, such as Cygnus X. A very special aspect of this Sixth Quark Matter Conference was the advent of the first experimental results from dedicated accelerator studies. These were con ducted during 1986/87 at the AGS of Brookhaven National Laboratory, with beams of 160 and 28Si at 14 GeV/nucleon, and at the CERN SPS, with 160 at 60 and 200 GeV / nucleon. An intense discussion of these data formed the main activity of the meeting. Despite some hints in the right direction, it is still too early to reach any conclusions about actual quark-gluon plasma formation. It did become clear, however, that the experimental techniques are suitable for measuring the relevant observables, and it seems possible to attain in nuclear collisions energy densities sufficient for plasma production. For the convenience of the reader, we have grouped the written versions of the talks according to their subject matter, and not in the order in which they were presented at Nordkirchen. We are grateful to all contributors for their efforts in providing the manuscripts - in particular to the experimental groups who were left with little time between first data taking and publication. Some of the contribu tions therefore still retain a certain preliminarity; but perhaps this will co very to the reader some of the excitement felt by the participants as they had their first look at a so far unexplored area of physics. In closing we want to thank the members of our International Advisory Commit tee for their assistance, and the entire organisational staff of the meeting who worked so hard in making Quark Matter 1987 a success. March 1988 R. Santo, H. Satz, H.J. Specht, R. Stock (Organizing Committee) Oxygen-induced reactions at 60 A GeV and 200 A GeV * studied by calorimetry S.P. Sorensen University of Tennessee, Knoxville, and Oak Ridge National Laboratory, Oak Ridge, USA W A 80 Collaboration R. Albrecht,3 T.e. Awes/ e. Baktash/ P. Beckmann,4 F. Berger,4 R. Bock,3 G. Claesson,3 L. Dragon,4 R.L. Ferguson,2 A Franz,5 S. Garpmann,6 R. Glasow,4 H.A Gustafsson,6 H.H. Gutbrod,3 K.H. Kampert,4 B.W. Kolb,3 P. Kristiansson,5 LY. Lee,2 H. Lohner,4 L Lund,3 F.E. Obenshain,1.2 A Oskarsson,6 L Otterlund,6 T. Peitzmann,4 S. Persson,6 F. PlasiI,2 AM. Poskanzer,s M. Purschke,4 H.G. Ritter,5 R. Santo,4 H.R. Schmidt,3 T. Siemiarczuk,3.a S.P. Sorensen,1,2 E. Stenlund,6 G.R. Young 2 1 University of Tennessee, Knoxville, TN 37996, USA 2 Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA 3 Gesellschaft fUr Schwerionenforschung (GSI), D-6100 Darmstadt, Federal Republic of Germany 4 University of Munster, D-4400 Munster, Federal Republic of Germany 5 Lawrence Berkeley Laboratory, Berkeley, CA 94720, USA 6 University of Lund, S-22362 Lund, Sweden Abstract. Results based on calorimetric measurements trum of possible plasma signatures has been suggested are presented from reactions of 60 A Ge V and 200 A [1,2]. An unfortunate common characteristic of most GeV 160 projectiles with C, Cu, Ag, and Au nuclei. of these signatures is the necessity to distinguish them Minimum-bias cross sections are discussed. Energy from the background created by nonplasma events. spectra measured at zero degrees and transverse-ener A thorough understanding of the reaction mecha gy distributions for the pseudo rapidity range 2.4 ~ fJ ::s;; nisms in ultrarelativistic heavy ion collisions is, there 5.5 are shown. An analysis of the average transverse fore, an important prerequisite in any QGP search. energy in terms of the number of participating nucle This paper presents some of the first steps taken in ons and the number of binary nucleon-nucleon col this direction by the W A 80 collaboration. The focus lisions is presented. Estimates of nuclear stopping and will be on data obtained with our calorimeters, with of attained energy densities are made. special emphasis on the results which can be extracted from forward calorimetry. In an effort to isolate characteristic features of nucleus-nucleus collisions (e.g., collective phenomena) from those that may be expected on the basis of a 1 Introduction linear superposition of nucleon-nucleon collisions, we The study of heavy ion collisions at ultrarelativistic will compare measured quantities with calculations energies is a new and exciting field on the borderline that reproduce data from nucleon-induced reactions between nuclear and high energy physics. Presently, and that make predictions for nucleus-nucleus reac the primary goal is to verify the existence of the postu tions. While several models are available for this pro lated quark-gluon plasma (QGP) [1]. A broad spec- cedure, none has yet been demonstrated to have clear advantages over the others. Consequently, we have Presented at the 6th International Conference on Ultra-Relativistic Nucleus-Neucleus Collisions - Quark Matter 1987, 24--28 August chosen to make comparisons with the Lund model 1987, Nordkirchen, Federal Republic of Germany for high-energy nucleus-nucleus interactions (FRI a On leave of absence from the Institute for Nuclear Studies, War TIOF) [3]. Effects of detector acceptance and of trig saw, Poland ger bias have been included in all FRITIOF simula- 4 tions shown in this work except for the total reaction to I] > 6.0. The ZDC is both a key component of the cross section calculations. In describing the nuclear trigger system and an important measuring device density, the radius parameter ro = 1.10 fm has been from which the total energy of projectile spectators used for the Woods-Saxon distributions. This results and/or of the leading particles is obtained. The resolu in somewhat larger sizes of the light nuclei as com tion of the ZDC is 2.5% at 3.2 TeV and 4.5% at pared to the standard FRITIOF parameter selection 0.96 TeV. but is in better agreement with results for the nuclear All data presented in this paper were obtained density distribution from electron scattering [4]. under the minimum-bias condition. This condition Another theme in our current analysis is the ex is defined by the requirements that: (a) less than 88% traction of results from the data with minimal use of the full projectile energy is measured by the ZDC of complicated models. Some reliance on models is, and (b) at least one charged particle is recorded by however, needed in any interpretation of experimental the multiplicity arrays in the interval 1.3 < I] < 4.4. data. An analysis of the transverse energy generated in terms of the number of participating nucleons and 3 Reaction cross sections the number of binary nucleon-nucleon collisions will be presented in Sect. 6. Although the extraction of Usually the first quantities to be measured in a new these numbers is not model-independent, it is based field of nuclear physics are the reaction cross sections. primarily on simple geometrical aspects of the nuclear We will not deviate from this practice and present collisions. in Table 1 the measured minimum-bias cross sections, Finally results on nuclear stopping and energy together with two sets of FRITIOF calculations: the densities are presented in Sect. 7 and 8, respectively. total reaction cross section based on the requirement that at least one nucleon from the projectile collides with a target nucleon, and the minimum-bias cross 2 Experimental setup sections, based on simulated trigger constrains in the The measurements described in this paper were per FRITIOF-generated events. Systematic errors on the formed with the W A 80 experimental arrangement [5, absolute cross sections are estimated to be less than 6J at the CERN SPS. The setup includes two calorim 10%. FRITIOF reproduces both the energy and the eters: the Mid-Rapidity Calorimeter (MIRAC) and target dependence of the measured minimum-bias the Zero-Degree Calorimeter (ZDC) [7]. MIRAC con cross sections also to within 10%. sists of 30 stacks with each stack subdivided into six 20 x 20 cm2 towers. Each tower consists of a lead/scin tillator electromagnetic section of 15.6 radiation Table 1. Reaction cross sections. Systematic errors on the experi- lengths (0.8 absorption lengths) and an iron/scintillator mental cross sections are estimated to less than 10%. Please refer hadronic section of 6.1 absorption lengths. MIRAC is to the next for the definition of the minimum-bias condition. ro organized into five groups of six stacks, called six and b are fitted Bradt-Peters parameters packs, each with dimensions of 1.3 x 1.2 m2• Four ofthe six-packs are arranged in a wall around the beam E1ab=60 A GeV min bias min bias reaction axis at a distance of 6.5 m from the target and with a O'EXP (JFRITIOF (JFRITIOF [mb] [mb] [mb] 7.5 x 7.5 cm2 hole in the center to allow the beam to reach the ZDC. The MIRAC wall has full azimuth 160+ 12C 650 650 1100 al coverage in pseudo rapidity, 1], from 2.4 to 5.5 with 160+65CU 1650 1440 2100 160+ 108Ag 2250 1950 2630 partial coverage extending down to 2.0. The fifth six 160+197Au 2900 2660 3360 pack of MIRAC is placed next to the MIRAC wall, ro [fm] 1.46 1.32 1.26 where it covers approximately 10% of the azimuthal b[fm] -2.34 -1.82 -0.08 angles in the pseudo rapidity interval from 1.6 to 2.4 The measured (J/E resolutions of the calorimeter are 14.2% for 10 GeV/c charged pions and 5.1% for E1ab = 200 A GeV min bias min bias reaction 10 GeV/ c electrons [7]. O'EXP (JFRITIOF O'FRITIOF [mb] [mb] [mb] The ZDC is a 60 x 60 cm2 uranium/scintillator calorimeter divided into an electromagnetic section 160+12C 450 460 1100 of 20.5 radiation lengths and a hadronic section of 160+65CU 1350 1270 2100 9.6 absorption lengths. The ZDC is located 11 m from 160+ 108Ag 1800 1660 2630 160+197Au 2450 2290 3360 the target and measures the energy of particles that pass through the beam hole in MIRAC. This hole ro [fm] 1.44 1.41 1.26 b[fm] -3.02 -2.92 -0.08 has an inscribed cone angle of 0.3 deg, corresponding 5 The target dependence is given approximately by metrical considerations can be used as a key for a the sum of the transverse areas of the nuclei as ex qualitative understanding of the ZDC energy spectra pressed in the Bradt-Peters parametrization [8J shown as filled circles in Fig. 1. At 200 A GeV, the 160 + 12C reaction has essentially no cross section (1) for events depositing a small amount of energy in the ZDC because, in a simple participant spectator The radius parameter, ro, and the overlap parameter picture, even in the most central collisions, several b extracted from the At dependence of the cross sec projectile spectator nucleons, each with an energy of tions are also shown in Table 1. It is noteworthy that 200 GeV, proceed in the beam direction. In contrast, the overlap parameter for the FRITIOF total reaction a pronounced peak is seen at small ZDC energies cross sections is practically zero, whereas the mini in the spectrum from the 160 + 197 Au reaction. In mum-bias cross sections show a large negative value this case, events with low ZDC energies result from of b. This indicates that b is to a larger extent a mea central collisions in which the oxygen projectile is sure of the trigger bias in this experiment than an engulfed by the massive Au nucleus, resulting in the estimate of the overlap between two nuclei in reac emission of only a few leading particles at angles less tions where at least one nucleon-nucleon collision than 0.3 deg. Furthermore, in this case, a wide range takes place. of impact parameters gives rise to collisions in which While the calculated reaction cross sections show the entire projectile interacts with a nearly constant no energy dependence, the minimum-bias cross sec number of target nucleons, thus producing the peak tions show a strong energy dependence. This depen at low ZDC energies. dence originates from different acceptances of the The effects of the more restricted acceptance of minimum-bias trigger at the two beam energies and the ZDC in going from 200 A GeV to 60 A GeV, can be understood from the following argument. as discussed above, is clearly seen in the 60 A GeV In a reaction with a given impact parameter the 160 + 197 Au spectrum, which has an even more pro ratio between the energy deposited in the Zero-De nounced peak at the lowest energies. In the 60 A GeV gree Calorimeter, EZDO and the beam energy, Ebeam, 160 + 12C reaction there are many more events with can be expressed as low ZDC energies as compared to the 200 A GeV case. These events may originate from collisions in (2) which the oxygen projectile fragments so violently that one or more of the projectile spectators has a pseudorapidity lower than 6 and is thereby inter where Es p is the energy carried by the projectile spec cepted by MIRAC. tators, and E is the reaction energy or, equivalent reac The predictions of the FRITIOF model are shown ly, the energy carried by the projectile participants, as histograms in Fig. 1. The agreement with the data and E (1J > 6) is the reaction energy accepted by reac is generally better at 200 A GeV than at 60 A GeV. the 0.30 beam hole in MIRAC. In going from 60 A There is a clear tendency for FRITIOF to underesti GeV to 200 A GeV, the nucleon-nucleon CM rapidity increases from 2.4 to 3.0. As a consequence, if there mate the cross section for small values of EZDC• This could indicate that FRITIOF does not include a suffi is no significant change in the reaction mechanism, ciently large degree of longitudinal momentum E (1J> 6) will constitute a larger fraction of E reac reac transfer in binary nucleon-nucleon collisions, but a at the higher bombarding energy and Esp will, there final conclusion on this point will have to wait until fore, have to constitute a smaller fraction of E beam the nuclear density prescription in FRITIOF has been in order to fulfill the most stringent of the trigger requirements, namely EZDC < 0.88 * Ebeam. Smaller further evaluated. values of Esp/Ebeam implie smaller impact parameters and, thereby, smaller total minimum-bias cross sec 5 Transverse energy distributions tions at the higher beam energy. The transverse energy produced in the reaction is measured on an event-by-event basis in MIRAC. The transverse energy is calculated as ET = I: Ei sin(Oi), 4 Zero-degree energy distributions where Ei and 0i are the observed energy and the effec An important aspect of high energy nucleus-nucleus tive angle of each element i of MIRAC, respectively. collisions is the nuclear collision geometry [9-11 J as The estimated systematic error in the transverse ener determined by the relative sizes of the target and pro gy scale is less than 10%. Based on measurements jectile nuclei, the overlap volume in the collision, and of the response of the calorimeter to electrons, pions, the impact parameter. As a consequence, simple geo- and protons of known energies between 2 and 6 60 A GeV 200 A GeV c C 0.8 2.4 1.2 0.0 3 Cu Cu 1.5 :> 2 1.0 Q) -0- -.ED 1 .... ... ..+ ::!,"""'~~ 0.5 0 f+:.!.! :. .L---1.-L-.....L.---l'--'----1.-+-J 0.0 5 Ag Ag () ... 1.5 ~4 ... W -- 3 1.0 '0 ....... ........ \:) 2 '0 0.5 o 8 Au 1.5 6 1.0 4 ... Fig. 1. Energy spectra measured in the Zero-Degree 0.5 Calorimeter (filled circles) in 160 induced reactions . 2 Histograms give the results of the FRITIOF model o '-'---'---'--'--'--'----'---'.-.J 0.0 [3]. The vertical error bars represent statistical errors o 200 400 600 800 o 1000 2000 3000 E ZDC (GeV) 50 GeV, an iterative procedure has been developed a nearly constant number of target nucleons. As the by means of which the nonprojective features of the target becomes smaller, the peak and the plateau be calorimeter response are corrected for. The method come less pronounced. For 160 + 12C, the ET spectra is described in detail elsewhere [7, 12]. have shapes similar to those of the ET spectra mea The transverse energy distributions for sured in proton-induced reactions [13], whereas the 2.4 < '1 < 5.5 are shown as filled circles in Fig. 2. As heavy target spectra are similar, both in shape and in the case of the ZDC spectra, the shapes of the energy scale, to the ET spectra for 200 A GeV 160 ET spectra are dominated by effects of the nuclear + Pb of the NA 35 collaboration [10]. geometry. The spectra for the heaviest nuclei, Ag and At 60 A GeV, the high-energy tails of the ET distri Au, show a large "plateau" extending out to 80- butions for Cu, Ag, and Au targets almost coincide 100 GeV at 200 A GeV beam energy and to 40- with one another at a value of approximately 60 GeV. 45 GeV at 60 A GeV. The Au spectra have a broad This phenomenon could be caused by "complete peak at the high-energy end of the plateau. This peak stopping" as discussed by the E802 collaboration for is closely correlated with the low-energy peak in the data obtained at 15 A GeV [14]. However, at our Au ZDC spectra [9]. This correlation demonstrates beam energies, this finding is more likely to be due that the peak in the ET distribution, corresponding to a combination of two opposing effects. As the tar to low ZDC energies, originates from the most central get mass or number of target participants increases, collisions, in which the entire projectile interacts with the maximum transverse energy increases. At the