Soft sector: promises, promises...
•  ~ 10 GeV/fm3 >> C (Bjorken / hydro)
• Early thermalization (v2 / hydro)
• Tchem ~ TH ~ Tc ~ 170 MeV
• b ~ 30-40 MeV -- near baryon free  connection to lattice?
• further evolution (PdV work?) & expansion leading to...
• cooling and bulk collective motion (pT spectra, v2, HBT, K-)
Can we use an external probe to explore further?
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
1
Discovery and Properties: The Ideal Experiment
The first exploration of subatomic structure was undertaken
Can
we do the same at RHIC?
by Rutherford at Manchester in 1909 using Au atoms as
targets and a particles as probes.
scattered
electron
QGP
incoming
electron
(controlled probe)
But we can get close ………
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
2
Fast Partons (Quarks & Gluons) Traversing Matter
p+p sNN=200 GeV
• Jets:
– high-pT parton produced in a hard (high-Q)
scattering process
– Calculable in QCD (at high-pT)
– partons fragment into many correlated white hadrons
• emitted in a cone
– created early in the collision
hadrons
quark
PHENIX PRL 91 241803 (2003)
quark
hadrons
leading particle
(highest-p particle)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
3
Fate of jets in heavy ion collisions?
idea: p+p collisions @ same
sNN = 200 GeV as reference
p
p
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
The tail probes the bulk
?: what happens in Au+Au to jets
which pass through medium?
June 2005
?
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Au+Au
Poland
4
5x fewer high pT particles than
“expected” in AuAu
Required checks:
• saturation (initial-state) effect?
200 GeV 0 RAA
The Ubiquitous RAA…
PHENIX (and all others) PRL2003
Nuclear overlap model to calculate # incoherent NN collisions (no shadowing etc)
RAA = (# seen in nuclear collision) / (# expected)
June 2005
notation: f=centrality cut)5
mike lisa - Cracow School of Theoretical Physics, XLV - (PHENIX
Zakopane Poland
The control experiment: d+Au, 2003
d
Au
?
Au
Au
Au+Au
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
6
Centrality Dependence
Au + Au Experiment
•
•
June 2005
d + Au Control Experiment
Dramatically different and opposite centrality evolution of Au+Au experiment from
d+Au control.
Jet Suppression is clearly a final state effect.
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
7
5x fewer high pT particles than
“expected” in AuAu
Required checks:
• saturation (initial-state) effect?
• d-Au null result
• hard scattering rates understood?
• measure “robust” hard processes
200 GeV 0 RAA
The Ubiquitous RAA…
PHENIX (and all others) PRL2003
Nuclear overlap model to calculate # incoherent NN collisions (no shadowing etc)
RAA = (# seen in nuclear collision) / (# expected)
June 2005
notation: f=centrality cut)8
mike lisa - Cracow School of Theoretical Physics, XLV - (PHENIX
Zakopane Poland
Hard non-mesonic (“direct”) photons
& non-photonic (“charming”) electrons
• Created in earliest, hardest parton
scatterings & conserved thereafter
• binary scaling √
• (probably some Cronin inside errors)
PHENIX QM04
Direct photons visible
because 0 is suppressed!!
PHENIX
June 2005
• probe-once-created and creation itself in
AA are “calibrated” (better than d-Au)
NB: pT-integrated
PHENIX
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane
Poland2004
9
5x fewer high pT particles than
“expected” in AuAu
Required checks:
• saturation (initial-state) effect?
• d-Au null result
• hard scattering rates understood?
• measure “robust” hard processes
• charm / direct  follow Nbinary
200 GeV 0 RAA
The Ubiquitous RAA…
PHENIX (& all others): PRL 2003
Accepted view
• final state effect is partonic energy loss
in color-charge-dense
(not necc. deconfined) medium
•  = 15 GeV/fm3 @  = 0.2 fm
• ~ consistent w/ BJ and hydro…
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
10
5x fewer high pT particles than
“expected” in AuAu
200 GeV 0 RAA
The Ubiquitous RAA…
Required checks:
• saturation (initial-state) effect?
• d-Au null result
• hard scattering rates understood?
• measure “robust” hard processes
is it “jet” quenching? PHENIX (& all others): PRL 2003
• charm / direct  follow
binary
Go Nbeyond
leading particles
Accepted view
• final state effect is partonic energy loss
in color-charge-dense
(not necc. deconfined) medium
•  = 15 GeV/fm3 @  = 0.2 fm
• ~ consistent w/ BJ and hydro…
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
11
Beyond leading particles
STAR p+p event
find this…
June 2005
STAR Au+Au event
…in this
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
12
Jets via azimuthal correlations
STAR p+p event
• trigger: highest pT track, pT>4 GeV/c
• Df distribution for 2<pT<pTtrigger
• normalize to number of triggers
trigger
June 2005
PRL 90, 082302
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
13
Jets via azimuthal correlations
STAR Au+Au event
Try the same in Au-Au
(large combinatorics)…
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
14
Azimuthal distributions in Au+Au
pedestal and flow subtracted
PRL 90, 082302
Peripheral collisions:
• very similar to p+p
June 2005
Central collisions:
• strong suppression
of away-side jet
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
15
Further geometric detail
midcentral (20-60%) collisions
nucl-ex/0407007
Suppression depends on pathlength thru medium
This analysis: cannot be initial state effect
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
16
anything new in-between?
Probes of the system (hard)
• interesting results
• important open questions
System itself (soft)
• interesting/important results
• important open questions
If we bridge the gap (“firm”)?
?
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
17
Connecting the sectors
soft-firm connection: v2
• hydro breaks down above
~ 2 GeV/c (expected)
• mass systematic
meson/baryon
systematic?
• definitely not hydro!
v2 generated
by anisotropic
energy loss?
symmetry, thermal smearing
June
P.
Sorensen,
2005 SQM04
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
18
Connecting the sectors
soft-firm connection: v2
• hydro breaks down above
~ 2 GeV/c (expected)
• mass systematic
meson/baryon
systematic?
• definitely not hydro!
v2 generated
by anisotropic
energy loss?
hard-firm connection:
RPID-differential
AA
meson/baryon systematic!
BUT!…
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
19
meson/baryon (i.e. constituent quark
number) systematics suggest…
coalescence (“recombination”) of
already-flowing (!!) partons (*)
PRL 92
92 (2004)
(2004) 052302
052302
PRL
Non-hydro / non-DE
origin of firm v2?
v 2hadron( pThadron)  nv 2quark( pTquark)
pThadron  npTquark
*June
well,
quarks
2005 constituentmike
lisa - Cracow School
P. Sorensen,
of Theoretical
SQM04Physics, XLV - Zakopane Poland
20
A growth industry...
Many model variants
Lin, Molnar, Pratt,
Fries, Bass, Mueller,
Ko, Das, Levai, Hwa,
Greco, Voloshin, Nonaka…
… for a thermal source
Fragmentation wins
out for a power law
tail
Baryons compete with
mesons
Many schemes, all describe
“anomalous” B/M, RAA, v2
But, data themselves (IMHO)
clearly indicate exciting,
collective partonic behaviour
Hope for a “hadronization
testbed”?
June 2005
mike B.
lisaMueller
- CracowTBS05
School of Theoretical Physics, XLV - Zakopane Poland
21
“calibrated” probes:
final state effect
color-dense medium
new at RHIC (?)
sensible geometrical systematics
 ~ 10 0
baryon / meson systematics
unexpected!
(constituent?) partonic collectivity?!
[possibility to test hadronization
mechanisms?]
bulk, collective system
 ~ 10 0
Tch~ TC ~ 170 MeV; b~30 MeV
Tkin~100 MeV, flow~0.7c
hydro works
June 2005
early thermalization.
sensitivity
to early-stage
EoS
mike lisa - Cracow
School of Theoretical
Physics,
XLV - Zakopane Poland
22
Open
Issues
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
23
soft sector questions
• hydro and v2
– hydro “limits” ?
– EoS - does Lattice really work
• RHIC HBT “puzzle”
– puzzle #1: failure of (otherwise-successful) dynamical models
– puzzle #0: generic “failure” - no sign of P.T.
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
24
hydro “limit”?
we have a bad history in RHI
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
25
Motivation
Formalism
Experiment
Trends
Models
Trends, soft sector, and RHI history
Finally, we
understand it!
Just one
event!
Gyulassy 1995
June 2005
6 decades of E/A
(2 decades of sNN)
Art’s talk. Compiled by
A. Wetzler (2005)
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
26
hydro “limit”?
dN/dy
Kolb, Sollfrank, Heinz, PRC62 054909 (2000)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
27
Disintegration timescale
Relative momentum between pions is a vector
 can extract 3D shape information
q  p1  p2
Rlong – along beam direction
Rout – along “line of sight”  increases with emission timescale

Rside –  “line of sight”
K  p1  p2
K  p1  p 2
p1
q
Rside
Rout

p2
Rout

Rside
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
28
Disintegration timescale - expectation
3D 1-fluid Hydrodynamics
Rischke & Gyulassy, NPA 608, 479 (1996)
with
transition
with
transition
“”
“”
Long-standing favorite signature of QGP:
• increase in , ROUT/RSIDE due to deconfinement  confinement transition
• expected to “turn on” as QGP energy threshold is reached
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
29
Disintegration timescale - observation
• no threshold effect seen
RO (fm)
RS (fm)
• RO/RS ~ 1
8
6
4
8
6
4
RO / RS
1.5
1.25
1.0
increasing collision energy
RHIC
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
30
Disintegration timescale - observation
• no threshold effect seen
• RO/RS ~ 1
• toy model calculations
suggest very short timescales
F. Retiere & MAL, PRC 70 (2004)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
31
Disintegration timescale - observation
• no threshold effect seen
• RO/RS ~ 1
• toy model calculations
suggest very short timescales
• rapid, explosive evolution
• too explosive for “real” models
which explain all other data
An important space-time
“puzzle” at RHIC
- actively under study
Heinz & Kolb, hep-ph/0204061
CYM & LGT
N()
PCM & clust. hadronization
NFD
NFD & hadronic TM
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
32
Measured final source shape
central
collisions

R 2y
R 2y
 R 2x
 R 2x
STAR PRL93 (2004) 012301
STAR, PRL93 012301 (2004)
• estimate INIT from Glauber
mid-central
R S2, 2
• from asHBT:  FO  2 2
collisions
R
S, 0
peripheral
collisions
Shape evolution: never goes in-plane?
Expected evolution:
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
33
firm/hard sector questions
• firm
– entropy issue (I really don’t care - theorists will work it out)
– what is a “dynamical constituent quark”?
• hard
– no consistent energy loss scenario describing geometrical systematics
of energy loss (Miodosewski/Tannenbaum)
– is charm quenched? What about dead cone?-- can it be gluon
radiation?
– both issues do we really understand the energy loss mechanism?
– if not, how can we say we probe color density
– and, if so, what can we say besides color density (e.g. deconfined or
not)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
34
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
35
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
36
5x fewer high pT particles than
“expected” in AuAu
200 GeV 0 RAA
The Ubiquitous RAA…
Common wisdom:
• d-Au null effect  final state effect
• so obvious that initial
Important
dAu
= AuAu?result!
• look
at other
hard probes (&
fromdiff) ways to getPHENIX
(& all others): PRL 2003
Several
complicated
suppression?
same initial processes, but with
Several
canceling
no
final state
effects…effects?
Common
wisdom
II: in a more natural way…
Look
closer
• final state effect is energy loss in
color-charge-dense
(not necc. deconfined) medium
•  = 15 GeV/fm3 @  = 0.2 fm
• ~ consistent w/ BJ and hydro…
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
37
If we are looking at energy loss…
…why quantify downward shift
of spectrum (as in RAA)?
It seems more natural to
quantify leftward shift.
[Tannenbaum/Mioduzewski]
NB: ignores
• Cronin (see jets(f) coming up…)
• mods in FF (logical to look at shift of
partons not their fragments)
June 2005
PHENIX PRL 2003
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
38
…then let’s look at energy loss
Empirical energy loss
Tannenbaum/Mioduzewski
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
39
Initial pathlength dependence seems to work
PHENIX prelim
Au+Au √sNN=200 GeV
2/3
–Gyulassy, Vitev, Wang: DE  N part
• 1D expansion, simple geometric scaling
 Well reproduced by experimental data.
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
40
Vary pathlength another way…
qualitatively OK, now see about consistency…
PHENIX prelim
Au+Au √sNN=200 GeV
Df (deg)
PHENIX B. Cole, HardProbes 04
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
41
No universal loss versus density path?
Overlap
integral:
dn color dn part

dA
dA
PHENIX
preliminary
• Quantitative consistency required to claim “jet tomography”
1D

init  0

 color ( xT )   color
expansion

• Next level of precision needed
to clarify underlying
quenching mechanism
pT > 3 GeV/c
• also -tagging, charm/dead-cone, q/g jet selection,
correlated quenching signal versus length
part(x,y) vs centrality
50-60%
10-20%
• (Role of dynamics in differential analysis may be significant)
• cannot just put by-hand some energy loss & find consistency w/BJ
June 2005 B. Cole, HardProbes
PHENIX
mike lisa
04 - Cracow School of Theoretical Physics, XLV - Zakopane Poland
42
HINT of charm
suppression??
Too early to tell (May05)
Converter method
Nuclear modification factor
PHENIX Preliminary
Cocktail subtraction method
RAA (2.5<pT<5.0 GeV/c)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
pT (GeV/c)
43
Backup
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
44
Parton Number Scaling of Elliptic Flow
In the recombination regime, meson and baryon v2 can be obtained
from the parton v2 (using xi = 1/n):
 pt 
2v  
 2
v 2M  pt  
2
 p  pt  
1  2  v2   
  2 
p
2
 p    p 
3v 2p  t   3  v 2p  t  
 3    3 
B
and v 2  pt  
2
 p  pt  
1  6  v2   
  3 
3
Neglecting quadratic and cubic terms, a simple scaling law holds:
 pt 
B
p  pt 
v  pt   2v   and v 2  pt   3v 2  
 2
 3
M
2
p
2
Originally poposed by S. Voloshin
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
45
Suppression vs Collision Energy
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
46
NO suppression in d+Au!
PHENIX
BRAHMS
STAR
June 2005
Phobos
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
47
Hadron dependence of high-pt suppression
Mueller & Fries [Mueller TBS05]
• R+F model describes different RAA behavior of protons and pions
• Jet-quenching becomes universal in the fragmentation region
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
48
Centrality dependence of open charm in Au+Au
central
Compare to
(PYTHIA)
an event
generator
tuned for pp
collisions…
no large
suppressionunlike light
quarks!
peripheral
Spectra of electrons from c e + anything
photonic sources are subtracted
June 2005
Comparison to pp
should be available
this week.
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
49
Flowing Charm? [F. Laue TBS05]
nucl-ex/0502009
inclusive ee v2
charmed ee v2
If charm quarks also flow,
charm quarks are
thermalized and / or
suppressed (interact with
matter) similarly to the
light quarks.
June 2005
•charmed electron v2 > 0
•rule out an extreme case,
charm does not interact with
QGP and fragment in vacuum
into D meson
•get charm quark v2 with quark
coalescence assumption
STAR SQM04
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
50
Direct photons
are enhanced w.r.t. 0 production, but it’s
just because 0 is suppressed, and the
direct photons are in fact binary scaled.
Inclusive Single 
Run2 AuAu 200 GeV
PHENIX Preliminary
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
51
STAR : nucl-ex/0409033
non-flow reduced
v2{RP}
v2{2}
PHENIX Preliminary
Different ways to reduce non-flow correlation, v2 w.r.t. SMD R.P.
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
52
Sonic boom from quenched jets
Casalderrey,ES,Teaney, hep-ph/0410067; H.Stocker…
•
•
•
June 2005
the energy deposited by jets into liquidlike strongly coupled QGP must go into
conical shock waves, similar to the well
known sonic boom from supersonic
planes.
We solved relativistic hydrodynamics
and got the flow picture
If there are start and end points, there
are two spheres and a cone tangent to
both
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
53
away <pT> dependence on angle (STAR,preliminary)
Preliminary
June 2005
<pT> (phi) has a dip structure in central AA.
Mach shock wave?
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
54
Viscosity of QGP
QGP at RHIC seem to be the most ideal
fluid known, viscosity/entropy =.1 or so
water would not flow if only a drop with 1000 molecules be made
• viscous corrections
1st order correction to dist. fn.:
:Sound attenuation length
Velocity gradients
Nearly ideal hydro !?
June 2005
D.Teaney(’03)
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
55
Comparison Au+Au/p+p at Lower
Energies
RAA
• Pb+Pb collisions at s ~ 20 GeV at CERN/SPS
 parton energy
loss (if any) is
overwhelmed by
initial state soft
multiple scattering
(Cronin effect)
June 2005
Crossing at
~ 1.5 GeV/c
Transverse Momentum (GeV/c)
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
56
Many measurements measure at high pT(!)
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
57
What data can tell us / what (we think) it has told us /
open issues
• Hard Sector:
–incoherence of initial state
• calibration of number of initial hard scatterings (binary), which
themselves have calibrated spectrum [charm, gamma, d-Au]
–early system color charge density (not deconfinement!)
• if gluonic bremstrahlung: dNg/d ~ 1000 ~BJ
• BUT… energy loss mechanism unclear (consistent scaling?)
• relevant upcoming data:
–charm quench (early results limited) / dead cone effect
–-tagged jets
–q/g jet selection
–higher stats correlated quenching signal versus length
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
58
What data can tell us / what (we think) it has told us /
open issues
• Soft sector:
–“threshold behaviour”
• no evidence: “evolution, not revolution”: too bad but not a killer
–degree to which it is system - real collectivity
• solid evidence from several observations - very important
[perfect fluid?]
–global/thermodynamic properties: T, , P, 
• consistent with deconfinement - [but! may be P.S. dominance]
– dynamics - system response evolution and timescales
• 2x expansion, collective shape evolution
• timescales- correct dynamic signatures, wrong (x,t) evolution??
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
59
What data can tell us / what (we think) it has told us /
open issues
• Firm Sector:
–“transition bulk to probe” (hadn’t really been considered…)
• clearly something new instead!
• ?is there a transition? does all fit smoothly?
–first “calibration” of QGP p.s. structure
• hope for well-defined ordered (x-p correlated) system of partons
• can we get beyond simple models? entropy issues? connection
between constituent quarks and partons?…
My feeling: solution of open issues in hard & soft sectors will
be well worth it, if we can exploit apparent promise of firm.
Launchpad of next generation program: understanding hadronization
June 2005
mike lisa - Cracow School of Theoretical Physics, XLV - Zakopane Poland
60