April 1996 OCR Output

BI·'I`P 96/ 15

CERN-TH/96-118

Talk given at the XXXI° Rencontre de Moriond, Les Arcs / France, 23. · 30. 3. 1996

results from Pb-—Pb interactions at CERN seem quite affirmative in both cases.

provide an apparently unambiguous probe for the onset of colour deconfinement. First

how the suppression of J / 1/1 and zL·' production, after removal of pre—resonance absorption,

strongly interacting matter in thermal and chemical equilibrium. Subsequently, we show

emitted in high energy nuclear collisions can be used to probe if such collisions produce

We discuss how the momentum distributions and the relative abundances of hadrons

Abstract:

Theory Division, CERN, CH—1211 Geneva 23, Switzerland

and

Fakultat fiir Physik, Universitat Bielefeld, D—33501 Bielefeld, Germany

Helmut Satz

HADROSYNTHESIS AND COLOUR DECONFINEMENT*

CERN-TH-96-1 18lllllllllllllllllllllllllltlllllllllll B"`? 96/15

CERN-_TH/96_118CERN LIBRARIES, GENEVA

\ »·—’* " i lv/I (k·‘l<.l`~J · = 6 · *l—»

region. In the longitudinal direction. along the beam axis. we assume the density of fireballs OCR Outputfireball in the nuclear fragmentation region .will presumably differ from that in the centralthe incident nuclei will in general pass fully through each other. the baryon density of aat SPS energy, we expect this assumption to be reasonable. but at very high energies, when

such fireballs. We shall here assume that the condensation trail consists of identical fireballs:

distribution for secondaries from a nucleus-nucleus collision. we have to superimpose many

its three-momentum: li] denotes the spatial volume of the fireball. To obtain the momentum

where pg = (pz + m?)1/2 is the energy of the secondary in the rest frame of the fireball. p

dE ll)%=%¤xp{—<p¤—¤B>/T}.3 _ d

is given by

The momentum distribution of hadrons of species i emitted from a fireball at freeze—out

potential ug; the latter determines the baryon density of the system.

interaction era. at freeze-out, are determined by its temperature T and its baryochemicalIt would then be something like a mini-universe, whose properties at the end of the strong

and chemical equilibrium before expanding so much that it freezes out into free hadrons.

produce in such collisions. Consider a single such “fireball” and assume it to reach thermalof deposited energy in this trail are candidates for the thermal systems which we hope tothrough each other, leaving behind a “condensation trail” of excited vacuum. The droplets

ln a high energy collision, the two interacting nuclei will interpenetrate and partially pass

Hadrosynthesis

of the third topic is given in the report of C. de los Heros at this meeting.My talk will deal only with the first two of these topics. A nice overview of the present status

resonances undergo a change in mass and decay width?

Does the production of low mass dileptons indicate that in a dense medium, hadronic

that there is an onset of colour deconfinement?

Can one conclude from the suppression of J / if and 1// production in nuclear collisions

collisions indicate thermal hadron formation?

Do the momentum spectra and the relative abundances of hadrons produced in nuclearThere are three topics which today seem of particular interest:is amusing to note that also the first light ion results were presented at Moriond, in 1987.

. data expected in the course of this year, with a first look to be provided at this meeting. It

An additional timely question thus is what we can hope to learn from the Pb and Au beam

heavy nuclear beams (‘°7Au, 2°8Pb), so that we can now study truly heavy ion collisions.ion beams (160, 28Si, 325) incident on heavy target nuclei. Since a year or two, we haveThe experimental study of high energy nuclear collisions began ten years ago, with light

is this matter in its early stages made up of deconlined quarks and gluons?

thermal and chemical equilibrium? and

do high energy nuclear collisions indeed produce matter, i.e., large scale systems in local

(QGP). The basic questions for such a program are therefore

temperature or baryon density, it will turn into a plasma of deconfined quarks and gluonsSuch matter is of particular interest because statistical QCD predicts that at sufficiently high

The main aim of high energy nuclear collisions is the study of strongly interacting matter.

predictions in Fig. 2 for the transverse and the longitudinal momentum distributions of pions OCR Outputhave completely determined predictions for the forthcoming Pb - Pb data. `We show thesebe taken into account. The p — W data give us 60 2 0.2: using this value in Eq. (4). we then(5). By extending our considerations to resonance freeze-out..as we will do below, this could

deviationsiat very small pion transverse momenta, which are due to production via A decay

we can fix 60. The p — W data [4] in Fig. 1 show that this indeed the case, apart fromof the transverse. momentum distribution is in accord with that found in p — A data, so that

Before turning to nucleus·nucleus collisions, we have to check if the resulting functional form

is also the transition temperature from QGP to hadronic matter obtained in lattice QCD.

describes the transverse mass distributions of secondaries from proton-proton interactions: itA reasonable guess for T is the Hagedorn temperature T 2 0.14 GeV, which correctly

baryochemical potential enters only in fixing the relative abundances of the different species.rameter needed for normalised momentum distributions is the freeze-out temperature T; thethis description is developed. With 60 determined from p — A interactions, the only pafor a more precise formulation of the somewhat generic expressions (2) — (5), see [3], wherebutions of hadrons emitted from a superposition of isotropically decaying thermal sources:

We thus obtain a complete description of the transverse and longitudinal momentum distri

(5)638 as (A+ B- 2) ag.1/3 m

the average broadening 63 per collision. For AB interactions, we then have

(4)6§A¤(.4—1)6§,1/3

experiments to determine through

this effect occurs already in p — A collisions, we can use the measured spectra from such

a broadening of the transverse momentum distribution of the observed secondaries. Since

random walk. The rotation of the collision axes underlying the random walk thus leads to

The dispersion 62 of the Gaussian in Eq. (3) depends on the number of collisions in the

(3)~ dpr ¢><p{—pi—/6} (y.pr — p)2%gig; f

transverse momentum distribution of secondaries becomes

assume this to result in a random walk in the transverse rapidity p of the fireball, so that the

axis, a nucleon once scattered will now interact along a rotated collision axis, and so on. We[2]: while the iirst nucleon-nucleon interaction in a nuclear collision occurs along the beam

In the transversedirection, the fireballs will acquire non·zero momenta by a Cronin-like effectanother proton, we take Y 2 |yi,,| — 1, with (yi,.] for the cms rapidity of the incident protons.the collision. Since a proton looses typically about one unit of rapidity in a collision withwhere the integration limits are determined by the least average energy loss of nucleons in

1/ -Y dy(2)r Y rr 9% ~ f an gi-h(Yr — y).

system. The resulting longitudinal momentum distribution thus becomes

to be boost·invariant so that the nature of the fireball trail is the same in each reference

Fig. 1: Transverse momentum distributions in p — T/Y collisions [4] OCR Output

pT [GeV]

0.5 1.s

ig I

QIOHS

5pw =i 0.3

T = 0.15 GGV

¤-W/ p-p

pT [GeV]

0.5 1.5

1 |- pIOflS

10l

102

T = 0.15 GeV, 5Dw = 0.3

d d =1 y pT Y=YA103p ` W

dN

Fig. 26: Transverse momentum distributions in Pb —- Pb collisions [6] OCR Output

mT - m [GeV]

0 0.2 0.4 0.6 0.6 1 1.2 1.4 1.6 1.8 210*

nucleons

T = 0.15 GGV

101

bpbpb = 0.436

NA49 preliminarymTclmT¤Y102

Pb — PbdN

mr ‘ m IGGVJ

0 0.2 0.4 0.6 0.6 1 1.2 1.4 1.6 1.8 210"

plOl’lS

101

T = 0.15 GQV

bpbpb = 0.436102

NA49 preliminary10** mTdmTdY

Pb - Pb_ dN

though we know that at least some particle ratios. e.g. K+/w`*` [8], still change considerably OCR Outputthese data are not yet available, and so we have to take recourse to light·heavy collisions. even

lt would be ideal to now apply this formalism to Au —- Au or P b—P b data. Unfortunately.

of T and pg: all other ratios are then predicted.

T, and pg. For example. we can use the measured rr“"/p and I{*/7:+ ratios to fix the values(rr/p. K/rr, K/K, p/rr. rp/vv. Y/p. ...) are given in terms of the two remaining parametersstate. lf there is a one·stage freeze·out of all thermal hadrons, then all production ratiosfixed by requiring the charge/baryon ratio of the final state to be equal to that of the initialstrangeness of the system to vanish. Similarly, the chemical potential for the charge, pq, isand pq. The chemical potential for the strangeness, pg, is fixed by requiring the overall(T) determines all thermal properties of the system in terms of the four parameters T, p B, p Swith d, denoting the spin degeneracy. mi the mass of hadron species i. The partition function

(8).i= ;;K2(@>1 vr T2 d.n,l0T

obtained by integrating the thermal form (1),as »\_<; = exp(pg/T), and that for charge as »\q ; exp(pq/T). The phase space factors arein question. The baryonic fugacity is defined as AB = exp( p 8/ T), that for the strangenessticles), with Si, B, and Q, denoting the strangeness, baryon number and charge of the hadronHere W, is the phase space factor for hadrons of species i (mesons, baryons and their antipar

(7)1¤Z(T,#B,#s»MQ) = 2/lg/\§‘/\g‘

branching ratios The corresponding partition function can be written asall possible hadronic resonances, which after freeze·out decay according to their empirical

The predictions of the different production ratios are determined from an ideal gas of

of the nucleosynthesis at the end of the strong interaction era in the big bang evolution.thermal parameters T and p B atfreeze-out. Such a hadrosynthesis is the little bang analogy

ratio N§(T , p) /Nj (T, p B). The relative abundance of all the species is thus fixed by the twoso that the production ratio for two species, N , /N j, is just given by the corresponding fireball

(6)N, 2 2YN(§(T,pB),

superposition of identical fireballs implies for the integrated abundance of particle species i

their production is in accord with emission from sources in chemical equilibrium as well. The

Next we consider the relative abundances of the different hadron species, to see whether

increased freeze-out temperature nor collective fiow of the final state medium.initial state collision axis rotations of the Cronin type. Such broadening requires neither an

that the observed broadening of the transverse momentum spectra is fully accounted for byically decaying thermal sources of temperature T 2 0.14 GeV. In particular, we emphasizeenergy nuclear collisions seem to be in accord with emission from a superposition of isotrop·

We thus conclude that the momentum distributions of the hadrons produced in highthese data agree quite well with the predicted behaviour.figures the (still preliminary) data presented at this meeting by P. Seyboth EvidentlyOCR Outputand nucleons. In this written version of my talk. I have taken the liberty to include in the

aim is not to just find these constituents in the debris from nuclear collisions. but to check OCR Outputdeconfined quarks and gluons. All matter is presumably made of quarks and gluons - so the

The search for colour deconfinement means looking for a large scale svstem made up of

Colour Deconfinement

Fig. 3: The T — pg regions determined by the indicated particle ratios

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

Q I I

UB [Eek/11</:<

i °'

A/N \

0.1

\-lx

K/n'" 1 {1

r M ·1 x/I //j

x z’/ A \rr`/P

T[6eVJ 1 / i

0.2

variety of hadron species produced in heavy ion collisions.

but still wait with our Bnal conclusion a few more weeks or months for data on a sumcient

as promising indications that nuclear collisions may well also lead to chemical equilibrium,really be excluded at this time [10]. For the moment we should therefore take these resultsThere is, however, another region around T 2 160 MeV and p g 2 600 MeV, which cannotcommon region, with T 2 110 d: 10 MeV and pg 2 550 zb 20 MeV as freeze·out values [9,10].and hyperons), as produced in S i — Au collisions at \/S 2 5 GeV They do indeed define aFig. 3, the result is shown for the most abundant species (pions, nucleons, kaons, antikaonssuch paths would have to cross at one point (or in one small region) in the T — p g plane. In

errors. If all hadrons were emitted from a thermal system of the same T and p g, then all

a region (“path”) in the T — p g plane; it would be a line if the experimental ratio had no

T and p g. Any measured production ratio of two diferent hadron species then traces out[10]. From Eq. (T). we can calculate the production rate for any hadron species in terms ofhave led to very promising conclusions. Let me illustrate this for the mentionediAGS databetween Si -— Au and Au — Au. Nevertheless, recent studies of AGS Si — Au data°[9,10]

hadron has a momentum of some 30 to 40 Ge\`. The hadron-J/1: dissociation cross section OCR Output1/5 to 1/6 ofthe hadron momentum. and that does not allow it. to break up a J/ui unless theby the gluon. with lc 2 3 for mesons. 4 for baryons. As a result. the gluon gets only aboutis approximately (1 — .1*)*. where r = pg /ph is the fraction of the hadron momentum carriedby hadrons is quite different. The momentum distribution of gluons confined inside a hadronit, like X-rays. The result is a sharp peak around kg 2 1 GeV. The dissociation cross sectiontoo much above this value, sirce very hard gluons will just pass the J / tl- without disturbingto be above the threshold fixed by the J/ui binding energy, eu. 2 0.65 GeV, but it cannot be

the photo-effect, in which a photon breaks up a hydrogen atom. The gluon momentum hasinteraction [21]. The break-up of a J / tb by a "fre€" gluon is essentially the QCD version of

In Fig. 4, we show the resulting cross sections for J/tb dissociation by gluon and by pion

Fig. 4: J / 212 dissociation by gluon and by pion interaction

0.5 1 2 5 10 20 s0

P [GeV]

TV " lilQ " by

o [mb]

With that, all elements needed for the calculation are fixed.calculable in terms of the gluon distribution function determined in deep inelastic scattering.perturbatively. The emission of such a hard gluon from a hadron is non—perturbative, but it isstates, this gluon has to be hard, and hence its interaction with the cc or bb can be calculated

and the hadron exchange a gluon. Because of the large binding energy of quarkonium groundrules based on the operator product expansion [19-20]. In this calculation, the quarkoniumnormal light hadron can be calculated in short distance QCD, essentially by solving sum

The cross section for the dissociation of a J / tb (or of an T) in an interaction with acollisions [15j18]. This question was in fact answered during the last two years.medium [14], all the more so since a considerable suppression was already found in p — AIt was only natural now to ask whether such a suppression could also occur in a hadronicfirst results were reported by C. Gerschel at the mentioned 1987 Rencontre de Moriond [13].the NA38 collaboration at CERN, in their study of O — U and S — U collisions [12]; theshould lead to J/tb suppression [11]. A strong suppression was soon afterwards found by

Ten years ago, T. Matsui and I had argued that the presence of a quark-gluon plasmaquark resonances (quarkonia) as a probe.that they are deconfined. I want to show in this section how that can be done using heavy

nucleus. In Fig. 5 we compare the predicted suppressioni(9) and the corresponding form for OCR Outputto become a physical J/z;· or u·'. Hence we have LA 2 (3/4)R_.; after averaging over thethe cE—g subsequently traverses the entire nucleus before absorbing its accompanying gluonthe target nucleus: for present incident energies and measuredcharmonium momenta [16-18],nucleus oniwhich it was produced. VVe here assume the cc- g to be formed at the center of

where ng = 0.17 fm"3 is normal nuclear density and LA the path of the cE—g through the

(9)S0 = exp]-n00cé.gLA},

production in p — p collisions, to be suppressed by a factor

threshold on. As a result, we expect J/1b and 1b' production in p — A interactions, relative tocross section odag can be estimated and is found to be about 6 - 7 mb essentially from the[26]. This in turn gives them a common cross section for the dissociation by hadrons. Thissize. however, is determined by confinement constraints and hence becomes the same for all

resonance ancestors of the different charmonia (tb, 1b' , XC) are in principle different; theirto take into account their possible absorption in the pre-resonance cE—g state. The pre

Before we can use charmonia as deconiinement probes in nuclear collisions, we then have

specific form of colour evaporation [27]a bb-—g for bottonia as well. Production through higher Fock space states thus provides a

evolution from a pre-resonance cE—g stage takes place for the other charmonia, and from

some 0.3 fm or so, this state converts to the basic cc and becomes a physical J/1,/1. The same

here proceeds through an intermediate higher Fock space state [24-26]. In a proper time ofcollisions, however, this state is the most readily accessible, and so charmonium formation

just constitute a relativistic correction to the J/dz wave function; in production by hadronic

state of J / rb quantum numbers. Normally, a higher Fock space state such as the cc- g wouldTo become colourless, it combines itself with a collinear gluon, forming a colour—neutral cc-gformed in gluon fusion is coloured and does not have the right quantum numbers for a J / zjx

new Fermilab data on J / d· and zf·' production at high transverse momenta [23]. The cc pairThe first part of this question appeared recently as central point also in the understanding of

to a real physical J/1/2, and we have to know what can happen to it in its pre-resonance stage.

the evolution of the ci formed in a nucleon-nucleon collision (predominantly by gluon fusion)

or nuclear collisions is not immediately such a physical resonance. We therefore have to know

talking about the break-up of fully formed physical J / zlfs, and what is produced in hadronic

There is, however, a catch to this quarkonium probe of deconfinement. We are here

In contrast to J / 2/fs, ¢"s can be broken up by hadrons as well as by deconiined gluons.carry out a direct experimental test of the transparency of confined matter to J / z!¤’s [22].states indeed provide an unambiguous deconiinement test [20]. It is moreover possible toJ/1/fs (or T’s) will be suppressed if and only if there is colour deconfinement, so that these

We can thus today strengthen the original claim [11]: in strongly interacting matter,essentially no threshold effect in either case.

only 60 MeV, so that it can easily be dissociated by "hadrons as well as by free gluons, with'I"s, which have 1.2 GeV binding energy. The 1/J', on the other hand, has a binding energy ofJ /zl· dissociation by hadrons at present energies. The same holds of course even more forIn summary: the large J /2li binding energy and the gluon distribution in hadrons preventfor hadrons of the full incident SPS energy, to break up a J / zb produced at mid—rapidity [20].therefore shows a very strong threshold suppression, making it effectively impossible, even

Fig. 6: Quarkoniuni rssoxiaiicc suppression in p — .-l and S — [` collisions [26] OCR Output

1210

0 NA as ir L ll"`]

0.2 A E 772 T

J/°’0.4 Q NA 38 I Evvz l

0.6

0.8

1.0

1.2

Fig. 5: Quarkonium suppression in p — A collisions [26]

0 1 2 3 4 5 6 7IV I

LA [fm]0.5

A T E 772

| J / my E 772

Q J! xy NA 38

0.6

0.7

0.8

::.9 5 § L I

1 1 T—'

some kind of deconfinement onset. OCR Output

presented by P. Bordalo show a much stronger and ET-dependent suppression. thus indicatingif also here pre-resonance cc- g absorption is the only effect. The beautiful NA50 results

a suppression which is approximately ET-independent, _roughly the same suppression as in central S — U collisions. and

should show

remains approximately constant. As a consequence. J/11* production in Pb — Pb collisionsa considerable range of impact parameters b. from central collisions (b = 0) to b 2 Rp],. L

central S — U collisions. Moreover, a closer look at the nuclear geometry shows that overis L 2 *2 (3/4) Rp], 2 10.0 fm. which is approximately the same as the path length inthe combined path length from target and projectile nuclei, averaged over the nuclear size,

only pre-resonance suppression, i.e., if there is no deconfinement. For a central collision.me first summarize what these data should look like if J / di production again would suffer

What can we now expect from the forthcoming data from Pb — Pb interactions? Let[28].first the effect seen in p- A interactions must be removed, as had been emphasized previouslynomenological level, it says that before looking for J /1b suppression due to deconfinement,statement can now be made in a quantitative and conceptually clear way. On a more phe

a J / tb suppression beyond what can be accounted for in terms of cE——g absorption. This

illustrates how charmonia can be used as confinement probes. Deconfinement then requires

Our conclusion, though negative as far as the QGP search is concerned, nevertheless

it must consist of confined quarks and gluons [26].after the nuclei have passed. Since this "com0ver” medium affects only v,l·"s and not J / vJ·.`s.collisions an additional suppression, indicating that there is indeed a further medium left

all values of L, from p — A to S — U interactions. In contrast, the di' experiences in S — U

we see that the pre-resonance absorption fully accounts for the observed J/1/2 suppression athadron energy Eg- and the path length leads to corresponding values of SCE., [18]. In Fig. 6,the associated cc-g suppression. A relation between the measured associated transverse

of the larger U nucleus. For non-central collisions, the path length decreases, and hence alsoboth nuclei, corresponding to a central collision, with the smaller S incident on the center

We have here used the form L 2 (3 / 4)RS+ RU 2 9.6 fm as the combined path length through

(10)5,,,;..9 = exp{—n0ac,;.g[(3/4)R5 + RU]}.

J/1/¤ survival probability 5,), by the cE—g suppression factor

due to the transition of the two nuclei. We therefore divide the experimentally measuredis deconfined, we have to check whether there the J / tb shows more suppression than thatprojectile and target nuclei “fly by". To see whether the matter remaining after this passageformed at mid·rapidity in a S — U collision will during its pre-resonance stage see both

remove the suppression which we know must come from pre-resonance suppression. A ccTurning now to the corresponding production in nucleus-nucleus collisions, we want to

in p ~ A interactions.

states [24-26] thus leads very naturally to the observed features of quarkonium productionboth equally.] as is in fact observed [16]. Charmonium evolution through higher Fock spacethe same size for the two resonances, the nucleus cannot tell what passed and must suppress

the nucleus never sees a physical J / dr or zb', but only the pre-resonance cc-—g. Since this is ofT`s with recent p — A data [16»18]. The excellent agreement shows that in p — A collisions

collisions. OCR Output

crucial in order to achieve the rather sudden onset of strong 'J/zu suppression in Pb — Pbexistence of a rather sharp deconfinement threshold in density. however. seems to be reallytoo central to reach the matching point with S -— Ci collisions at larger impact parameter. Thebetween central S - U and central Pb- Pb collisions. The present NASD data are appearentlythe density of sufficiently hardgluons surpasses the threshold for deconfinement somewherenumber density 11, we therefore expect a behaviour of the form shown in Fig. 8, provided

collisions will be the same as that for a central S — U interaction. As function of the collision

interaction decreases the number of collisions, so that for large enough b the number of

the collision numbers listed in Table 1. Increasing the impact parameter b in a Pb — Pbinelastic nucleon·nucleon cross section. Averaging for b = 0 over the nuclei involved. we getwith RA and RB for the radii of the two nuclei, if we take ro 2 1 fm, corresponding to the

(11)um) z [w§2R;4n0][¤w§2RBn0],

i.e., for impact parameter b = O. The number of inelastic collisions in this tube isin a tube of radius ru, along the beam axis and through the center of the two colliding nuclei.

We therefore compare for a central A — B interaction the number of nucleon-nucleon collisions

GeV) gluons, and this depends on the number of nucle0n—nucleon collisions per unit volume.possible deconiined neighbourhood. The crucial factor here is the density of hard (kg ~ 1first step in the direction of a QGP is the internetted process shown in Fig. 7, the smallestQGP formation, the J / tj: would find itself in a large-scale medium of such hard gluons. Thewhich then, in a deconfined state, i.e., before hadronizing, dissociates the J /1b. In case ofnucleon collision produces a J / ab. Another nucleon-nucleon collision produces a hard gluon,

How does deconfinement start? `We give a schematic illustration in Fig. 7. One nucleon

Fig. T: J/zb dissociation by local deconfinement

-7/tb

Fig. 9: Measured J/is resonance suppression as function of path length L ~ e [29]. OCR Output

10 12

L (fm)

`6’ 0.5

-·------ »---¢-·----- ---— -·—— -——E 1 =1’*=T~.g.=*=_·—¢·•£=

Zi 1.5

• P b — P b

I S ·— U

A p - W , p — U

Fig. 8: Schematic J / zj» resonance suppression as function of collision number

302010

• Pb - Pb

n S - U

0.5 { A p - A

S\ls(V)

298. Nov. 1995. rev. version April 1996. OCR Output

10) J. Cleymans et al.. "Thermal Hadron Production in Si-Al] Collisions", CERN-TH/95P. Braun·Munzinger ret al., Phys. Lett. B344 (1955) 43.

T. Abbott et al.. Nucl. Phys. A544 (1992) 237.

J. Cleymans and H. Satz. Z. Phys. C57 (1993) 135.

P. Seyboth, XXX? Rencontre de Moriond, March 1996.See e.g. T. K. Hemmick. Nucl. Phys. A566 (1994) 435c.

H. R. Schmidt and J. Schukraft, J. Phys. G 19 (1993) 1705.

T. Akeson et al., Z. Phys. C46 (1990) 361:Preprint CERN·TH/96-71, April 1996.

A. Leonidov, M. Nardi and H. Satz. "Hadron Spectra from Nuclear Collisions", CERN

2) J. W. Cronin et al., Phys. Rev. D11 (1975) 3105.

J. D. Bjorken, Phys. Rev. D27 (1983) 140.

References

of L. Kluberg in just looking for the facts.C. Gerschel, from numerous discussions with C. Lourenco, and from the inspiring insistence

D. Kharzeev; we have both benefitted much from the constructive criticism of A. Capella and

with J. Stachel and R. Stock. The entire section on colour deconfinement is joint work with

A. Leonidov, M. Nardi and R. L. Thews; it has been much stimulated by spirited discussions

The hadrosynthesis section of this survey is largely based on joint work with J. Cleymans.

Acknowledgements

targets, and a look at Pb — U for a possible saturation.

matter [22], a more detailed study of how Sd,/Sc;..g begins to deviate from unity for diHerentsome further experimental studies seem essential: a check of the J / ;b—transparency of confined

Pb — Pb interactions. To provide as firm as possible a basis for the onset of deconfinement,

at present no known conventional explanation for the J/tb suppression observed in the SPS

My conclusion for this section therefore agrees fully with that of Paula Bordalo: there isseen in Fig. 6.

and quite different from the gradually increasing 1,/J' absorption in confined (hadronic) matter

by P. Bordalo, is shown in Fig. 9. The sudden onset of the suppression is indeed strikingperhaps use L ~ e as a measure. The result, including the new NA50 data presented here

the measured ET into e involves detailed nuclear geometry; until this is carried out, we canstudy the resonance suppression 5,;,/5,:,;.9 as function of the energy density e. Converting

If we have more than just a local precursor of deconnnement, it would be better to

Table 1: The number of collisions in a, tube of 1 fm radius

1/(ro) ] 1.0 ] 6.0 [ 21.1 [ 28.3

p-p |p—C’ |S—U fPb—Pb

29) P. Bordalo, XXXI° Rencontre de Moriond, March 1996. OCR Output

28) C. Gerschel and J. Hiifner, Z. Phys. C 56 (1992) 171.27) For a recent review, see e.g., R. V. Gavai et al., Int. J. Mod. Phys. 10 (1995) 3043.

26) D. Kharzeev and H. Satz, Phys. Lett. B 366 (1996) 316.

25) E. Braaten and S. Fleming. Phys. Rev. Lett. 74 (1995) 3327.

24) G. T. Bodwin, E. Braaten and G. P. Lepage, Phys. Rev. D 51 (1995) 1125

lab-Conf-95-137 E, May 1995.

L. Markosky (D0), “Measurements of Heavy Quark Production at D0”, Preprint Fermi—Fermilab-Conf-95-128 E, March 1995;

23) See e.g. V. Papadimitriou (CDF), "P1·0duction of Heavy Quark States at CDF”,· Preprint

22) D. Kharzeev and H. Satz, Phys. Lett. B 356 (1995) 365.

Singapore 1996.

21) D. Kharzeev and H. Satz, in Quark·Gluon Plasma 2, R. C. Hwa (Ed.), World Scientific,20) D._ Kharzeev and H. Satz, Phys. Lett; B 334 (1994) 155.

G. Bhanot and M. E. Peskin, Nucl. Phys. B156 (1979) 391.

19) M. E. Peskin, Nucl. Phys. B156 (1979) 365.

Portugal, January 1995.

periment”, CERN-PPE/95-72, May 1995, and Thesis, Universidade Técnica de Lisboa,18) C. Lourenco (NA38/51), “Recent Results on Dimuon Production from the NA38 Ex17) L. Fredj (NA38/51), Thesis, Université de Clermont-Ferrand. France, September 1991.

16) D. M. Alde et al., Phys. Rev. Lett. 66 (1991) 133 and 2285.

15) J. Badier etal., Z. Phys. C20 (1983) 101.

World Scientific, Singapore 1990.

14) See e.g., J.·P. Blaizot and J.-Y. Ollitrault, in Quark-Gluon Plasma, R. C. Hwa (Ed.),Thanh Van (Ed.), Eds. Frontieres,'1987, p. 443.

13) C. Gerschel, in Haclrons, Quarks and Gluons, XXIIme Recontre de Moriond, J. Tran

12) C. Baglin et al., Phys. Lett. B220 (1989) 471; B251 (1990) 465, 472; B225 (1991) 459.

11) T. Matsui and H. Satz, Phys. Lett. 178B (1996) 416.