Heavy Ion Physics
Lecture 2
Thomas K Hemmick
Stony Brook University
Outline of Lectures

What have we done?





Is There a There There?







Azimuthally Anisotropic Flow
Hydrodynamic Limit
Heavy Flavor Modification
Recombination Scaling
Lecture 2
Is the matter exotic?


The Medium & The Probe
High Pt Suppression
Control Experiments: gdirect, W, Z
What is It Like?


Energy Density
Initial Temperature
Chemical & Kinetic Equilibrium
System Size
Quarkonia, Jet Asymmetry,
Color Glass Condensate 2
What does the Future Hold?
Stony Brook University
Thomas K Hemmick
LHC Experiments
ALICE
CMS
ATLAS
Stony Brook University
3
Thomas K Hemmick
Could Suppression be Merely from the PDFs?
probe rest frame
ggg r/g



The lower in x one
measures, the more
gluons you find.
At some low enough x,
phase space saturates
and gluons swallow one
another.
Another novel phase:
Color Glass
Condensate
Control Experiment
4
d+Au Control Experiment
Medium?
No Medium!
Proton/deuteron
nucleus
collision
Nucleusnucleus
collision


Collisions of small with large nuclei quantify all cold nuclear effects.
Small + Large distinguishes all initial and final state effects.
Stony Brook University
5
Thomas K Hemmick
Terminology
Peripheral Collision
Semi-Central Collision
100%


Centrality and
Reaction Plane
determined on an
Event-by-Event
basis.
Npart= Number of
Participants


2  394
Ncoll = # Collisions

11000
Stony Brook University
Centrality
0%
f

Central Collision
Reaction Plane
Fourier decompose azimuthal yield:
d 3N
 1  2v1 cosf   2v2 cos2f   ...
dfdpT6dy
Thomas K Hemmick
What is it Like? “elliptic flow”
Origin: spatial anisotropy of the system when created, followed by
multiple scattering of particles in the evolving system
spatial anisotropy  momentum anisotropy
v2: 2nd harmonic Fourier
coefficient in azimuthal
distribution of particles with
respect to the reaction
plane
Almond shape
overlap region
in coordinate
2
2
y

x

space
 2
2
v2  cos2f
y  x 
Stony Brook University
7
f  atan
py
px
Thomas K Hemmick
Anisotropic Flow
Liquid Li Explodes
into Vacuum


Process is SELF-LIMITING
Sensitive to the initial time
Position Space anisotropy
(eccentricity) is transferred to
a momentum space anisotropy
visible to experiment



Stony Brook University
Gases explode into
vacuum uniformly in
all directions.
Liquids flow violently
along the short axis
and gently along the
long axis.
We can observe the
RHIC medium and
decide if it is more
liquid-like or gas-like
8

Delays in the initiation of anisotropic
flow not only change the magnitude of
the flow but also the centrality
dependence increasing the sensitivity
of the results to the initial time.
Thomas K Hemmick
Fourier Expansion

Most general expression for ANY invariant cross section
uses explicit Fourier-Series for explicit f dependence:
1 d 3N
1 d 2N
1  2v1 ( pT , y) cosf   2v2 ( pT , y) cos2f   ...

pT dpT dfdy 2pT dpT dy

here the sin terms are skipped by symmetry agruments.
For a symmetric system (AuAu, CuCu) at y=0, vodd vanishes
1 d 3N
1 d 2N
1  2v2 ( pT ) cos2f   2v4 ( pT ) cos4f   ...

pT dpT dfdy 2pT dpT dy

v4 and higher terms are non-zero and measured but will be
neglected for this discussion.
1 d 3N
1 d 2N
1  2v2 ( pT ) cos2f 

pT dpT dfdy 2pT dpT dy
Stony Brook University
9
Thomas K Hemmick
Huge v2!




Hydrodynamic limit
exhausted at RHIC for
low pT particles.
Can microscopic
models work as well?
Flow is sensitive to
thermalization time
since expanding
system loses spatial
asymmetry over time.
Hydro models require
thermalization in less
than t=1 fm/c
Adler et al., nucl-ex/0206006
WTF!
10
What is needed, partonically for v2?
Huge cross sections!!
if (r3==45 mb) {r=1.2 fm};
11
Comparison to Hydro Limit



Hydro limit drops with energy.
RHIC “exhausts” hydro limit.
Does the data flatten
12 to LHC or rise?
LHC Flow results match RHIC


Magnitude of flow as a FUNCTION of pT is nearly
exactly the same as at RHIC.
LHC data reach to very high moments (v6).
What else we can get from Hydro?
So far we have tracked the hydrodynamic evolution of the system back in
time to the initial state. Let now Hydro do something good for us.
Approximately: ∂nTmn =0   P dV = DEK  mT – m0  DKET = √pT2+m02
Baryons
Mesons
v2 for different m0 shows good agreement with “ideal fluid” hydrodynamics
An “ideal fluid” which knows about quarks!
14
Recombination Concept
1
dN
dz E dN a
E 3h  2
D h ( z )
3
d P 0 z z d P / z
Fragmentation:
• for exponential parton
spectrum, recombination is
more effective than
fragmentation
• baryons are shifted to
higher pt than mesons, for
same quark distribution
 understand behavior of
protons!
Stony Brook University
fragmenting parton:
ph = z p, z<1
recombining partons:
p1+p2=ph
15
Thomas K Hemmick
Baryon Anomaly


Recombination models
assume particles are
formed by the coalescence
of “constituent” quarks.
Explain baryon excess by
simple counting of valence
quark content.
16
Where does the Energy: LHC


Outside of large cone (R=0.8)
Carried by soft particles
Away Jet cannot “Disappear”
1 < pT (assoc) < 2.5 GeV/c
PHENIX
STAR


Energy conservation says “lost” jet must be found.
“Loss” was seen for partner momenta just below the
trigger particle…Search low in momentum for the
18
remnants.
Correlation of soft ~1-2 GeV/c jet partners
Emergence of a Volcano Shape
“split” of away side jet!
PHENIX
Mach Cone??
(nuclex/0507004)
Gluon Cherenkov??
120o…is
it just v3???
19
Stay Tuned…
Strings: Duality of Theories that Look Different



Tool in string theory for 10 years
Strong coupling in one theory corresponds to weak
coupling in other theory
AdS/CFT duality
(Anti deSitter Space/ Conformal field theory)
(in QCD)
(N=4 SYM)
Calculated from AdS/CFT Duality
Another Exotic Structure: Ridge
“The Ridge”
Is this bulk response to stimulus…long range flux tubes…v3?
1. pT spectra similar to bulk (or slightly harder)
2. baryon/meson enhancement similar to bulk
3. Scales per trigger like Npart similar to bulk
Stony Brook University
21
Thomas K Hemmick
Rise and fall of “ridge/cone”—Centrality evolution
0-1%
Pay attention to
1.02
how long-range
1.01
1
structures
0.99
disappear and clear -1
0
1
jet-related peaks
2
Df 3
4
emerge on the
away-side
Strength of soft
component
increase and then
decrease
0-10%
1.02
1
12
-10
-3-2 D h
-5-4
34
-1
0
1
Df
2
3
4
20-30%
12
-10
-3-2 D h
-5-4
34
10-20%
1.04
1.1
1.02
1.05
1
1
0.98
-1
0
1
Df
2
3
4
30-40%
12
-10
-3-2 D h
-5-4
34
0.95
-1
1.1
1.1
1.1
1
1
1
1
0
1
2
3
4
23
1
0
-1
-3-2
Dh
-5-4
0.9
-1
0
1
Df
2
3
4
60-70%
23
1
0
-1
-3-2
Dh
-5-4
4
-1
0
1
Df
2
3
12
3
4
-10
-3-2 D h
-4
-5
4
0
1
2
3
4
70-80%
4
23
1
0
-1
-3-2
Dh
-5-4
4
0.9
-1
0
1
Df
2
3
4
23
1
0
-1
-3-2
Dh
-5-4
80-90%
1.1
1.2
1
1
0.9
0.8
-1
3
34
1.4
1.2
0.9
0.9
-1
Df
1.1
1
2
12
-10
-3-2 D h
-5-4
50-60%
1.1
4
1
40-50%
1.2
-1
0
Df
1.2
Df
Near-side jet peak is
truncated from top to
better reveal long
range structure
0-5%
0
1
Df
2
3
12
3
4
-10
-3-2 D h
-4
-5
4
-1
ATLAS Preliminary
ò Ldt = 8 mb
0
1
Df
2
3
12
3
4
-10
-3-2 D h
-4
-5
4
a
-1
2 < pT , pbT < 3 GeV
4
Ridge and Cone = v3???



Event Plane method yields <vn> (vodd=0).
2-particle yields SQRT(<vn2>) (vodd>0).
How to disentangle:



PHENIX = EP method + factorization.
ATLAS = Rapidity OUTSIDE other Jet.
Everyone else = Factorization.
v3 explains double-hump
24
correction
correction

v2 correction only

double-hump

v2, v3, v4 correction
double-hump
disappeared
 Peak still broadened

Higher order moments



Higher order moments can
be measured WRT their
own “reaction plane”.
Determines how initial
state fluctuations are
carried by fluid to final
state.
Higher order moments will
serve to provide strong
constraint on viscosity.
25
Thomas Hemmick
How can charm (bottom) be measured?
 ideal (but challenging)
 direct reconstruction of charm decays (e.g. D0  K- +)
 much easier if displaced vertex is measured
(PHENIX upgrade)

e
K
 alternative (but indirect)
ne
 contribution of semi leptonic
charm decays to
– single lepton spectra
– lepton-pair spectra
e
D0
c
c
e
K
e+

D
ne
Stony Brook University
0
D K
0
 
D0  K 
26

ne
n
D0 D0  e  e  K  K  n e n e
D0 D0  e  m  K  K  n e n m
D0 D0  m  m  K  K  n m n m
Thomas K Hemmick
Inferred Heavy Flavor



Measurement inclusive e±.
Measure 0, h0
Construct “Cocktail” of electron sources other than c/b
 light
hadron decays
 photon conversions

Subtract e± “cocktail” leaves e from c/b.
Stony Brook University
27
Thomas K Hemmick
Hard Probes: Open Heavy Flavor
Electrons from c/b hadron decays
Calibrated probe?
pQCD now predicts cross section well
Total charm follows binary scaling
Strong medium effects
Significant suppression
Upper bound on viscosity!
Little room for bottom production
Limited agreement with energy loss
calculations
28
Single Muons from ATLAS




High Momentum muons dominantly from heavy flavor.
Eliminate unwanted background by statistical method.
At these high momenta, the muons are likely dominated
by bottom.
Is there a limit to the power
of the river?...Stay tuned.
29
Stony Brook University
Thomas K Hemmick
Heavy Flavor Quarks are Flowing!
We can imagine that the flowing QGP is a river that sweeps quarks.
A “perfect fluid” is like a school of fish…all change direction at once.
Our QGP river carries off heavy stones (not BOTTOM???)
Requiring a model to SIMULTANEOUSLY fit RAA and v2 “measures”
the h/s of the QGP fluid.
30
How Perfect is “Perfect” ?
h


( Entropy Density ) 
s
4
4
RHIC “fluid” is at ~1-3 on this scale (!)
 The Quark-Gluon Plasma is, within preset error, the
most perfect fluid possible in nature.
31
 High order vn measurements
to yield superb precision!





J/psi Suppression by Quark-Gluon Plasma Formation,
T. Matsui and H. Satz, Phys.Lett.B178:416,1986.
If cc dissolved, unlikely to pair with each other.
Suppression of J/Y and Y.
Suppression driven by size of the meson as compared
to the Debye Radius (radius of color conductivity)
Stony Brook University
32
Thomas K Hemmick
How is J/y formed in pp?
Stony Brook University
33
Thomas K Hemmick
J/y is suppressed (everywhere)
Stony Brook University
34
Thomas K Hemmick
RAA
LHC/RHIC comparison
1.4
ALICE (Pb-Pb sNN = 2.76 TeV), 2.5<y<4, p >0 (preliminary)
T
PHENIX (Au-Au sNN = 0.2 TeV), 1.2<|y|<2.2, p >0 (arXiv:1103.6269)
T
1.2
PHENIX (Au-Au sNN = 0.2 TeV), |y|<0.35, p >0 (nucl-ex/0611020)
T
1
0.8
0.6
0.4
0.2
(*) ALICE <N > is weighted by N
coll
part
0
0
STAR (pT>5 GeV) versus
CMS (6.5<pT<30 GeV)
50
100
150
200
250
300
350 400
<Npart*>
PHENIX (pT>0 GeV) versus
ALICE (pT>0 GeV)
Caveat: Different beam energy and rapidity coverage;
35/~182.1 x dN /dh(N )RHIC
dNch/dh(Npart)LHC
ch
part
CMS: all the Y states separately.



The data show that the 2s/3s are reduced compared to the 1s.
This is first strong indication of sequential melting in QGP.
Should yield screening length of our color conductor!
Upsilon Suppression



Upsilon system is “cleaner” than the J/Psi.
1s state suffers from feed-down (~50%).
Consistent with melting all Y except feeddown.
Stony Brook University
37
Thomas K Hemmick
J/Psi as Bottom Suppression?



CMS can separate out J/Psi which are
daughter states of decays from B mesons.
These are a surrogate for a bottom quark.
Suppression same or less than /charm?
Stony Brook University
38
Thomas K Hemmick
Parton Distribution Functions


PDFs are measured by
e-p scattering.
Calculations (PYTHIA)
use theoretically
inspired forms guided
by the data:




CTEQ 5M
others…
Unitarity requires that
the integral under the
PDF adds up to the full
proton momentum.
Dirty Little Secret:
The sum of the parts
exceeds the whole!
F2
39
Crisis in Parton Distributions!
What happens if you pack too
many gluons inside a box?


Parton Distributions
explode at low x.
The rise must be capped.
40
ANSWER: They eat each other.
Glass at the Bottom of the Sea?
probe rest frame

This implies that
nature has a maximal
gluon density.
ggg r/g


Note that the gluon
fusion reaction, g+gg,
“eats gluons”.
Its kind of like a fish tank:

When the fish eat their
young, the tank never
overfills with fish.



Material exhibiting
nature’s ultimate gluon
density is called
Color Glass Condensate.
The existence of this
material would cap the
gluon growth at low x,
restoring unitarity
The Bottom of the Sea
Fuses Into Color Glass.
Nuclear Oomph…



A nucleus compresses more matter and makes
the CGC easily accessible.
Shadowing competes with CGC.
Many believe that shadowing is simply
“parameterized” CGC.
42
J/y complicated by CNM effects




Stony Brook University
43
Electron-nucleus
collisions are the most
promising way to find
CGC.
Proton (deuteron)
collisions are the best
we have for now.
A depletion in the low-x
wave function of a Au
nucleus decreases the
number of scatterings
in the deuteron
direction.
EPS09 shadowing fails.
Thomas K Hemmick
Jets distinguish CGC from shadowing.



The fundamental difference between the CGC model of cold
nuclear matter and the shadowing model is the number of
partons that scatter.
Shadowing changes the PDF, but still does all physics as 1-on-1
parton scatterings.
CGC allows one (from deuteron) against many (from glass), and
thereby splits away-side jet into many small pieces.
HUGE suppression in low X.



The suppression factor from cold nuclear
matter is a factor of ~10!
The away-side jet “decorrelates”.
Jury still out:



Nearly all measurements follow CGC predictions.
Predictions are often qualitative.
Electron-ion collisions will find the truth.
Summary






Nuclear Collisions provide access
to the collective color interaction.
These provide a glimpse at aspects
of the color force inaccessible
through elementary collisions.
Partonic matter just beyond the phase transition is a
strongly-coupled plasma exhibiting explosive flow into
the vacuum.
String-theory has provided “Nature’s lower bound” on
h/s…a limit realized within error by sQGP.
Nuclear collisions can provide access to dense color
fields in cold nuclear matter that may exhibit CGC.
Short time scales for thermalization challenge theory.
Stony Brook University
46
Thomas K Hemmick
Length dependence of J/y
Stony Brook University
47
Thomas K Hemmick
SURPRISE!




The direct virtual photons
measured by PHENIX
have been associated
with early stage thermal
radiation.
If true, they should show
little flow.
Surprise…they flow.
We must take care in
interpreting these
photons…
Chiral Magnetic Effect
Dima Kharzeev. QM2011
Stony Brook University
49
Thomas K Hemmick
Chiral Magnetic Effect ('strong parity violation')
cos    
cos     2YRP 
B
RHIC
+
B
?
-
RHIC
Same charge correlations positive
Opposite charge correlations negative RHIC : (++), (+-) different sign and magnitude
RHIC ≈ LHC
LHC: (++),(+-) same sign, similar magnitude
Localunexpected
Parity Violation
somewhat
17 Gauss
in 10
magnetic
should
decrease
with NchField ?
+
may decrease with √s
QM2011 J. Schukraft
50
Backup Slides
Stony Brook University
51
Thomas K Hemmick
Dynamical Charge Correlations
Possible interpretations:
(A) If linked to LPV effect - de-confinement and
chiral symmetry restoration. Absence of difference
in correlations means absence of phase transition.
K. Fukushima et al, PRD 78, 074033 (2008)
Alternate Observables
(B) Charge asymmetry
STAR Preliminary
STAR Preliminary
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
LPV: ‹A+A-›UD < ‹A+A-›LR
(C ) Conservation effects:
momentum & Local charge
and flow.
Reaction plane dependence
balance function ~ difference
between opposite and same
charge correlations.
A. Bzdak, et al., PRC 83 (2011) 014905
S. Schlichting et al., PRC 83 (2011) 014913
Y. Burnier et al., arXiv:1103.1307
How to reconcile (A) with the fact
v2( +) < v2 (-) at 7.7 GeV
QM2011
Bedanga Mohanty
52
Suppression.



1s state should be too
large to melt in the
plasma.
2s/3s could be melted.
Data are above bluedashed which would be
consistent with only 1s
survival and removal of
nearly all 2s/3s.
Freeze-out Conditions
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
11.5 GeV
STAR Preliminary
39 GeV
STAR Preliminary
STAR Preliminary
Chemical freeze-out:
Particle ratios
Andronic et al.,
NPA 834 (2010) 237
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
7.7 GeV
STAR Preliminary
39 GeV
11.5 GeV
STAR Preliminary
STAR Preliminary
Kinetic freeze-out : Momentum distributions
QM2011
L. Kumar, Energy scan, 27th May
Bedanga Mohanty
54
Particle Ratio Fluctuations
Observations:
Fluctuations in particle ratios
-- Sensitive to particle numbers
at chemical FO not kinetic FO
-- Volume effects may cancel
S. Jeon, V. Koch, PRL 83, 5435 (1999)
TPC
PID
TOF
PID
K
p
pT (GeV/c)

Rapidity
Differences could be due to
Constant or monotonic trends observed
difference in acceptance and/or
Apparent differences (results with Kaons) with SPS PID selections --- under discussion
QM2011
Bedanga Mohanty
55
fluct.
Data are still “horny”

Can be naturally explained by change of strangeness
production from LK to KK…
Fragmentation Function at LHC
 Not modified!
 Need to be more
quantitative to
really understand
differences from
RHIC.



Color Glass Condensate
probe rest frame
Gluon fusion reduces number of scattering
centers in initial state.
Theoretically attractive; limits DGLAP
evolution/restores unitarity
ggg r/g
Stony Brook University
59
Thomas K Hemmick
Stony Brook University
60
Thomas K Hemmick
Stony Brook University
61
Thomas K Hemmick
Stony Brook University
62
Thomas K Hemmick
Stony Brook University
63
Thomas K Hemmick
Stony Brook University
64
Thomas K Hemmick
Stony Brook University
65
Thomas K Hemmick
Stony Brook University
66
Thomas K Hemmick
Stony Brook University
67
Thomas K Hemmick
Stony Brook University
68
Thomas K Hemmick
Stony Brook University
69
Thomas K Hemmick
Stony Brook University
70
Thomas K Hemmick