20 years of J suppression at the CERN SPS

1
20 years of J/? suppressionat the CERN SPS

• Results from experiments
• NA38, NA51, NA50 and NA60

Louis Kluberg LLR Ecole Polytechnique/CNRS-IN2P3
in honour of our friend and colleague Helmut
Satz without whom all this would not exist.
2
The very beginning of the story

• was, in fact, NOT Helmuts fault at all !
• Experiment NA38 was proposed in March 1985
• to study thermal dimuon production in AA
  collisions
• using the existing NA10 muon spectrometer
• without even mentioning J/? production
• From the abstract of the proposal
• Shuryak (1980), Kajantie/Miettinen (1982),
  Hwa/Kajantie (1985), Mc Lerran/Toimela (1985)
• ...Thermal dimuons are expected to be emitted
  from a quark-gluon plasma at a reasonable rate in
  the 1-3 GeV/c2 transverse mass range, and to
  differ from ordinary dimuons by their pT and
  rapidity distributions But then

3
(No Transcript)
4
But thencame Matsui and Satz (1986)

• and spoiled our original, already SPSC
  approved,
• scientific goals.
• From their abstract (Phys. Lett. B
  178 (1986) 416. )
• If high energy heavy ion collisions lead to
  the formation of a hot quark-gluon plasma, then
  colour screening prevents ccbar binding in the
  deconfined interior of the interaction region /
  It is concluded that J/? suppression in nuclear
  collisions should provide an unambiguous
  signature of quark-gluon plasma formation
• The most brilliant concept of this abstract was
  the word
• U N A M B I G U O U S

5

• Had this prediction (not postdiction ) not
  existed
• NA38/50 might have foundthermal dimuons (????)
• (long before NA60)
• comovers, for sure, would have not been
  invented and would probably still be unknown
  particles
• PHENIX (RHIC) and ALICE (LHC) would look quite
  different and
• Many theoreticians might have 50 less (or quite
  different) publications

6
The NA10/38/51/5060 muon spectrometer
Kinematical coverage Acceptances

• 0 ycm 1 ( 2.92 ylab 1 )
• cos ?cs lt 0.5
• Acc (J/?) 12.5
• Acc (DY) 13.8
• (for 2.9 lt Mµµ lt 4.5
  GeV/c2)

7
The muon pair mass spectrum

• In the beginningthere was no Drell-Yan.
• and there was no anticipated normal J/?
  behaviour either
• and we had to live without
• as shown in the next slides
• for our first muon pair mass in 200 GeV OU
  reactions
• Was J/? production in 200 GeV OU abnormal ????

8
(No Transcript)
9
(No Transcript)
10
NA38 first results

• OU at 200 GeV/c
• Factor 2 suppression
• but including
• normal nuclear absorption !!!
• IMR charm-like excess !!! (fit starts
  from 1.7 GeV/c2 !!)

11
And after thata lot morefor 12 years
SU 90 low masses
SU 87 low intensity
SU 92 J/?
etc..etc..etc..
12
The muon pair mass spectrum


.
15 years later

• Shapes of various physical contributions from MC
• Combinatorial background (? and K decays) from
  like-sign measured pairs
• ? and ? mass resolution 100 MeV
• Final fit performed for M gt 2.9 GeV/c2

lt
lt
13
Why do we keep using Drell-Yan ?

• Drell-Yan (muon pairs) is a well known
  computable process,
  
  proportional to the of elementary
  nucleon-nucleon collisions, with the following
  priceless advantages

• identical experimental biases
• identical inefficiencies
• identical selection criteria
• identical cuts

as J/?
Therefore the corrections cancel out in the ratio
? (J/?)
?
(DY)
which is insensitive to normalization
factors/uncertainties
14
Advantages and drawbacks of Drell-Yan

• leads to real robust results
• ? (DY) is proportional to the number of
  nucleon-nucleon collisions from pp up to Pb-Pb
  (in our phase space
  domain, at least)
• ideal to compare different colliding nuclei
• needs isospin correction
• DY ltltlt J/? statistics

15
Why do we use a reference curve ?

• The question we have tried to answer
• Is J/? abnormally suppressed in nucleus-nucleus
  collisions and, in particular, in Pb-Pb
  collisions at 158 GeV/n ???
• The standard (our definition of normal
  suppression)
• The apparent suppression (in fact nuclear
  absorption) in pA collisions at 158 GeV
  (normal, by definition)
• Our presently non ideal but only available
  tool
• (while waiting for NA60 direct measurement at 158
  GeV)
• A set of pA measurements at 450 and 400
  GeV/c
• pA and AB measurements at
  200 GeV

16
The year 2000 special vintage data
Going directly to the besttaking advantage of
what we learned from the previous 5 years (4
times same experiment) 1 single (4mm thick)
target in vacuum Use of tracking in MD to
identify primary interaction vertex ? No
Pb-Air contamination in peripheral interactions
? Efficient primary vertex on
target identification ? No reinteractions
in central collisions
The cleanest of all our samples !!! MD tracking
technique was later extended to the 1998 data
1998 and, in particular, peripheral
1998, reanalyzed with only low Pb-Air
contamination
17
The year 2000 analysis
Standard J/? / DY analysis with Affect
only absolute normalization, not pattern shape
itself Special effort on the reference curve
Normal Nuclear Absorption based on all our
recent p-A data at 450 and 400 GeV
and using at 200 GeV

• adapted minimal cuts (allowed by extra clean
  sample)
• use of GRVLO94 (practically same result with
  GRVLO98
• improved J/? line shape
• either, as in the past, both p-A and S-U data
• or, newest development ONLY p-A data

18
The year 2000 results (I)

• As a a function of ET
• used as a centrality estimator
• 
  
• the ratio of cross-sections
• steadily decreases,
• from peripheral to central
• collisions by a factor 2.5
• no saturation is seen
• for the most central collisions
• statistical errors, due to DY,
• are in the range 9 - 7

19
The year 2000 results (II)
20
The year 2000 results (III)

• Normal Nuclear Absorption determined from
  
• new pA data
• at 450 and 400 GeV
• S-U at 200 GeV
• leading to ?abs 4.2 0.4 mb
• with the rescaling factor
• 450/400 ? 200 GeV obtained
• from simultaneous (same ?abs )
• fit at the 3 energies.
  
• the ratio of cross-sections
• is normal for peripheral collisions
• more and more abnormal
• with increasing centrality

21
The year 2000 results (IV)
22
New pure p-A reference (I)
Determine absorption reference at 158 GeV from
pA data only as SU could be already
abnormal i.e. maybe affected by comovers ?
Only our most precise data ? All available 200
GeV data (NA38) plus pp and pPt (NA3) ?
No Drell-Yan at 200 GeV ? absolute J/?
cross-sections ? Separate fits at 450/400/200
show Excellent slope compatibility
? Simultaneous fit leads to ?abs and
rescaling factor 450/200
23
New pure p-A reference (II)

• Glauber fit (excluding NA51) on pA data only
  leads to
• ?abs 4.1 0.5 mb from x-section
• ?abs 4.2 0.5 mb from J/? / DY
• Absolute cross-sections
• experimentally rescaled
• to 200 GeV, from pA only
• OCu, OU and even SU
• are just plotted
• BUT NOT INCLUDED in the fit.
• They show, however,
• within errors,
• a pA like behaviour

24
J/? suppression from x-sections

• Same as previous plot with all data rescaled at
  158 GeV
• Confirm in PbPb,
• with pure pA reference,
• J/? is anomalously suppressed
• For PbPb the ratio
• measured/expected
• amounts to 0.65 0.08
• for J/? / DY normal absorption
• reference (compared to previous
• determination using SU)
• normalization by 0.6 !!
• its uncertainty by a factor 2 !!

25
The Y2K results with new p-A reference
26
and with traditional pA/SU reference
27
From ppto PbPb
28
The Y2K results vs. energy density
29
J/? suppression in InIn collisions (NA60)

• ONGOING study by experiment NA60 in InIn and
  also pA
• Same energy as NA50 158 GeV/nucleon
• Same muon spectrometer as NA50
• Highly upgraded detector, in the target region
  (pixel telescope)
• Centrality estimated from zero degree
  calorimeter
• Study of a la NA50
• Study of making
  no use of DY events
• ONLY PRELIMINARY results, as shown in QM2005,
• are reminded here.

J/? / DY
J/? / expected
30
Standard analysis of J/y production in In-In
collisions
phase space window -0.5 lt cos ?CS lt 0.5 0.0 lt
ycms lt 1.0
73 000 J/y events in total 520 events for M
gt 4.2 GeV
The ratio J/y / DY in 3 centrality bins, as
limited by DY statistics.
31
Direct study of the J/y production as a function
of centrality
Normal Nuclear absorption
Directly compare the measured J/? centrality
distribution with pure nuclear absorption as
determined by NA50. Only matched dimuons used.
No fits used to extract the J/y yield, just
counting events in appropriate mass range.
32
J/y production in InIn vs. other colliding
systems
NA60 P R E L I M I N A R Y !!!
L is not a good scaling variable between p-A
absorption and In-In suppression Number of
participants provides a better overlap between
In-In and Pb-Pb data
33
J/y suppression by comovers in InIn collisions
The model takes into account nuclear
absorptionand comovers interaction with sco
0.65 mb
C O M O V E R S F I N A L !!!
This prediction clearly fails to describe the
In-In data BUT NA60 PRELIMINARY !!!
Capella and Ferreiro, hep-ph/0505032
34
For the ? too, life is difficult
.. in NN collisions, as compared with pA
(rescaled to 158 GeV) Significant
suppression as a function of centrality,
in line with observed in SU interactions at
200 GeV.
35
even more than for J/?
as a function of energy density
J/? and ? suppressions and normal nuclear
absorption
36
Conclusion
What I learned from experiment after 20 years in
J/? suppression 1/
Many years of hard work, repeating the same
measurement under improved conditions are
needed to reach stable, coherent and
comparable experimental results (just
progress) and also 2/ It is very risky to
build models, to try and reproduce still
UNPUBLISHED, and therefore, PRELIMINARY,
results. For PUBLISHED results, .. go to
1/ and be extremely cautious...
37
Conclusion
What I learned from experiment after 20 years in
J/? suppression 1/
The correct paper is, systematically, the
NEXT one, in preparation for publication
(this is just called progress) and also 2/
It is risky to build models, to try and
reproduce still UNPUBLISHED, and therefore,
PRELIMINARY, results. For PUBLISHED
results, .. go to 1/ and be extremely
cautious...
38
Acknowledgements

• Thank you so much, Helmut,
• in my own personal name but also,
• in the name of the 200 colleagues of the
• NA38, NA51, NA50 and NA60 Collaborations.
• Thanks to the original Matsui and Satz article,
• we all made a living for 20 years,
• built and successfully operated new detectors,
• put some salt and pepper in the field,
• and had, together, a lot of fun and excitement.
  

39
Thank you
40
From somewhere on the web.

• Satz hatte 1986 zusammen mit einem japanischen
  Wissenschaftler, Tetsuo Matsui (der heute in
  Kyoto lehrt) theoretisch hergeleitet, daß erst
  dann, wenn der Plasmazustand wirklich erreicht
  ist, ein bestimmtes seltenes Elementarteilchen
  (das J/psi-Meson) zu schmelzen beginnt. Wenn die
  Ausbeute dieser Mesonen drastisch zurückgeht, muß
  der Plasmazustand erreicht worden sein. Und genau
  das ist in dem CERN-Experiment passiert Die
  Energie, mit der im Experiment die Atomkerne
  aufeinanderprallten, war hoch genug, um das
  Plasma entstehen zu lassen erkennbar daran, daß
  das J/psi-Meson nur noch in deutlich geringerer
  Menge nachweisbar war. Ohne die Arbeiten von
  Professor Satz wäre der Nachweis des
  "Urknall-Plasmas" nicht möglich gewesen. Satz war
  übrigens lange Jahre sowohl am CERN als auch am
  Brookhaven National Laboratory in Long Island,
  USA, tätig, den beiden größten Elementarteilchenbe
  schleunigern der Welt.