Studying Hot Many Body QCD
W. A. Horowitz
University of Cape Town
June 22, 2012
With many thanks to Razieh Morad,
Miklos Gyulassy, and Yuri Kovchegov
6/30/2016
Wits Seminar
1
What Are We Interested In?
• Measure manybody physics of
strong force
• Test & understand
theory of manybody non-Abelian
fields
Long Range Plan, 2008
6/30/2016
Wits Seminar
2
Experiments
• RHIC
• LHC
–
–
–
–
–
–
–
–
BRAHMS
PHENIX
PHOBOS
STAR
ATLAS
PHENIX
6/30/2016
ALICE
ATLAS
CMS
LHCb
Wits Seminar
3
Four Known Forces
Electromagnetism
Gravity
starchild.gsfc.nasa.gov
John Maarschalk, travelblog.portfoliocollection.com
Weak
Strong
lhs.lps.org/staff/sputnam/chem_notes/tritium_decay.gif
6/30/2016
Wits Seminar
4
Many Body Electromagnetism
•
=>
Hanneke, Fogwell, and Gabrielse, PRL100 (2008)
• “Simple” Hydrogen
Phase Diagram
Calculated, Burkhard Militzer,
Diploma Thesis, Berlin, 2000
6/30/2016
Wits Seminar
5
Many Body QCD
•
=>
PDG
6/30/2016
Wits Seminar
de Florian, Sassot, Stratmann,
PRD75 (2007)
6
Big Bang vs. Little Bang
ALICE Collaboration
t=-
t=0
Initial State
Initial Overlap
6/30/2016
t = 1 fm/c
Thermalization
t = 3 fm/c
QGP
Wits Seminar
t = 4 fm/c
Hadronization
t=+
Hadron Gas
7
Viscous Hydrodynamics
• Viscosity reduces elliptic flow
– Naive pQCD => h/s ~ 1
– Naive AdS/CFT => h/s ~ 1/4p
Shear Viscosity, Wikipedia
Luzum and Romatschke,
Phys.Rev.C78:034915,2008
6/30/2016
Wits Seminar
8
QGP Energy Loss
• Learn about E-loss mechanism
– Most direct probe of DOF
pQCD Picture
AdS/CFT
Picture
6/30/2016
Wits Seminar
9
Hot Nuclear Matter: AdS
6/30/2016
Wits Seminar
10
Strong Coupling Calculation
• The supergravity double conjecture:
QCD  SYM  IIB
– IF super Yang-Mills (SYM) is not too
different from QCD, &
– IF we believe Maldacena conjecture
– Then a tool exists to calculate stronglycoupled QCD in SUGRA
6/30/2016
Wits Seminar
11
Heavy Quark E-Loss in AdS/CFT
• Model heavy quark jet energy loss by
embedding string in AdS space
dpT/dt = - m pT
m = pl1/2 T2/2Mq
– Similar to Bethe-Heitler
dpT/dt ~ -(T3/Mq2) pT
J Friess, S Gubser, G Michalogiorgakis, S Pufu, Phys Rev D75 (2007)
– Very different from usual pQCD and LPM
dpT/dt ~ -LT3 log(pT/Mq)
6/30/2016
Wits Seminar
12
Qual. Expectations: pQCD vs. AdS/CFT
• For e = DpT/pT
– Asymptotic pQCD
erad ~ as
L2 log(pT/Mq)/pT
– Asymptotic AdS
eST ~ 1 - Exp(-m L),
m = pl1/2 T2/2Mq
– Independent of pT and strongly dependent on Mq!
– T2 dependence in exponent makes for a very sensitive
probe
– Expect: epQCD
0 vs. eAdS indep of pT!!
• dRAA(pT)/dpT > 0 => pQCD; dRAA(pT)/dpT < 0 => ST
6/30/2016
Wits Seminar
13
AdS/CFT and HQ at RHIC
• String drag: qualitative agreement in
heavy flavor sector
WAH, PhD Thesis
Akamatsu, Hatsuda, and Hirano, PRC79, 2009
6/30/2016
Wits Seminar
14
AdS/CFT and HQ at LHC
• D Predictions
• B Predictions
ALICE 0-20%
D
CMS B→J/y
WAH, PANIC11 (arXiv:1108.5876)
ALICE, arXiv:1203.2160
CMS, JHEP 1205 (2012) 063
• AdS HQ Drag appears to oversuppress D
• Roughly correct description of B→J/y
6/30/2016
Wits Seminar
15
Light Quark E-Loss in AdS
Chesler et al., PRD79 (2009)
• Complications:
– string endpoints fall
=> painful numerics
– relation to HI meas.
• less obvious than HQ
• In principle, compute Tmn
from graviton emission
– Extremely hard
6/30/2016
Wits Seminar
16
AdS/CFT Lights E-Loss Prescription
• Make jet def., hope not too diff from Tmn
– Orig: define jet as string Dx ~ 1/T from end
Dx ~ 1/T
 k (t 1 )
 k (t 2 )
 k (t 3 )
Chesler et al., PRD79 (2009)
– E-Loss:
6/30/2016
Wits Seminar
17
dE/dt for Lights from AdS/CFT
• Original definition + original energy loss
calculation => generic Bragg peak
Chesler et al., PRD79 (2009)
See also: Gubser et al., JHEP 0810 (2008)
Arnold and Vaman, JHEP 1010 (2010)
– Intermediate-t dE/dt depends strongly on IC
6/30/2016
Wits Seminar
18
AdS/CFT Light Parton Energy Loss
• Simple Bragg Peak Model
– Ef ~ q(ttherm – t)
0.2 TeV
2.76 TeV
WAH, JPhysG38 (2011)
• Large uncertainty due to T(t)
6/30/2016
Wits Seminar
19
Recent AdS Lights Developments
• Corrected E-loss formula
A Ficnar, arXiv:1201.1780
• Bragg peak
disappears
– But now DE ≠ E
at ttherm!
6/30/2016
Wits Seminar
20
New Prescription
• Define jet by sep. hard and soft scales
0
uc
 0
 0
u
uh
 0
u0
1
 k (t 1 )
 k (t 2 )
 k (t 3 )
Tx
u h  u (t ,  k )
 0.2
uh
– DE = E for t = ttherm
6/30/2016
Wits Seminar
21
Results
• Use T(t) ~ t-1/3 Janik
& Peschanski metric
-(dE/dt)/(T E0)
• In usual AdS/Sch.
Bragg reappears!
Tt
– ttherm very short
6/30/2016
Tt
R Morad and WAH, in prep.
– Bragg peak gone!
Wits Seminar
22
AdS/CFT Light q E-Loss
0.2 TeV
• Static thermal medium =>
very short therm. time
– tth ~ 2.7 fm
• AdS likely oversuppresses
compared to data
• Examine T ~ 1/t1/3 geom
– tth ~ 4.1 fm
R Morad
2.76 TeV
WAH, JPhysG38 (2011)
Simple Bragg peak model
6/30/2016
Wits Seminar
23
Hot Nuclear Matter: pQCD
6/30/2016
Wits Seminar
24
pQCD Rad Picture
• Bremsstrahlung Radiation
– Weakly-coupled plasma
• Medium organizes into Debye-screened centers
– T ~ 350 (450) MeV, g ~ 1.9 (1.8)
• m ~ gT ~ 0.7 (0.8) GeV
• lmfp ~ 1/g2T ~ 0.8 (0.7) fm
• RAu,Pb ~ 6 fm
– 1/m << lmfp << L
• multiple coherent emission
– Bethe-Heitler
– LPM
dpT/dt ~ -LT3 log(pT/Mq)
6/30/2016
Gyulassy, Levai, and Vitev, NPB571 (2000)
Wits Seminar
dpT/dt ~ -(T3/Mq2) pT
25
What About Elastic Loss?
• Appreciable!
• Finite time effects small
Mustafa, PRC72 (2005)
Adil, Gyulassy, WAH,
Wicks, PRC75 (2007)
– For pQCD comparisons with data, use WHDG
Rad+El+Geom model; formalism valid for g/lq & hq
6/30/2016
Wits Seminar
26
Qualitative Expectations for LHC
– For approx. power law production and energy
loss probability P(e), e = (Ei - Ef)/Ei
– Asymptotically, pQCD =>
DE/E ~ log(E/m)/E
• ~ flat RAA(pT) at RHIC
• Rising RAA(pT) at LHC
– True for glue, lights, & heavies
– NB: LHC is a glue machine
6/30/2016
Wits Seminar
27
pQCD Picture Inadequate at RHIC?
PHENIX, PRL 105 (2010)
PHENIX, PRL 98 (2007)
6/30/2016
Wits Seminar
Wicks et al., NPA784, 2007
28
Pert. at LHC energies?
Constrain to RHIC
• Best fit WHDG to
PHENIX p0 RAA
+200
– dNg/dy = 1400-375
PHENIX, PRC77 (2008)
• Extremely conservative zero parameter
extrapolation to LHC
– Assume rmedium ~ dNch/dh
– Keep as = 0.3 fixed
6/30/2016
Wits Seminar
29
LHC Predictions vs. Data
CMS 0-5%
h±
CMS 40-50%
h±
CMS, arXiv:1202.2554
CMS, arXiv:1204.1850
ALICE 0-20%
D
All data
preliminary
√s = 2.76 ATeV
ALICE, arXiv:1203.2160
6/30/2016
Wits Seminar
30
pQCD pp Predictions vs. Data
PHENIX, PRC84 (2011)
CMS, arXiv:1202.2554
6/30/2016
Wits Seminar
31
Quant. (Qual?) Conclusions Require...
• Further experimental results
• Theoretically, investigation of the effects of
– higher orders in
•
•
•
•
as
kT/xE
MQ/E
opacity
(large)
(large)
(large?)
(large?)
– geometry
•
•
•
•
•
•
uncertainty in IC
coupling to flow
Eloss geom. approx.
t < t0
dyn. vs. static centers
hydro background
(small)
(large?)
(?)
(large: see Buzzatti and Gyulassy)
(see Djordjevic)
(see Renk, Majumder)
– better treatment of
• Coh. vs. decoh. multigluons
• elastic E-loss
• E-loss in confined matter
6/30/2016
Wits Seminar
(see Mehtar-Tani)
32
(Data – pQCD)/Data
CMS 40-50%
h±
CMS 0-5%
h±
LO
Calculation
6/30/2016
ALICE 0-20%
D
Wits Seminar
33
Attempt at NLO
• Running coupling ansatz
A Buzzatti, HP2012
6/30/2016
Wits Seminar
34
Asymptotic pQCD vs. AdS/CFT
• But what about the interplay between
mass and momentum?
– Take ratio of c to b RAA(pT)
• pQCD: Mass effects die out with increasing pT
RcbpQCD(pT) ~ 1 - as n(pT) L2 log(Mb/Mc) ( /pT)
– Ratio starts below 1, asymptotically approaches 1.
Approach is slower for higher quenching
• ST: drag independent of pT, inversely
proportional to mass. Simple analytic approx.
of uniform medium gives
RcbpQCD(pT) ~ nbMc/ncMb ~ Mc/Mb ~ .27
– Ratio starts below 1; independent of pT
6/30/2016
Wits Seminar
35
Does pQCD or AdS Yield Correct Mass &
Momentum Dependecies at LHC?
WAH, PANIC11 (arXiv:1108.5876)
– T(t0): “(”, corrections likely small for smaller momenta
– Tc: “]”, corrections likely large for higher momenta
6/30/2016
Wits Seminar
See also:
WAH, M. Gyulassy, PLB666 (2008)
36
Not So Fast!
D7 Probe Brane
– Speed limit estimate for
applicability of AdS drag
• g < gcrit = (1 + 2Mq/l1/2 T)2
~ 4Mq2/(l T2)
– Limited by Mcharm ~ 1.2 GeV
• Similar to BH
LPM
– gcrit ~ Mq/(lT)
– Importance of longitudinal
momentum fluctuations
• <(DpL)2> = p l1/2 T3 g5/2
Q
Worldsheet boundary
Spacelike if g > gcrit
x5
Trailing
String
“Brachistochrone”
“z”
D3 Black Brane
– (<(DpL)2>)1/2 ~ pL => gcrit ~ Mq/(4 T)
• Decorrelation of HF?
6/30/2016
Wits Seminar
37
AdS HQ E-Loss in Cold Nuclear Matter
Constant T Thermal Black Brane
Shock Geometries
P Chesler,
Quark Matter 2009
Nucleus as Shock
DIS
Embedded String in Shock
Albacete, Kovchegov, Taliotis,
JHEP 0807, 074 (2008)
Before
After
vshock
Q
z
x
Q
z
vshock
x
WAH and Kovchegov, PLB680 (2009)
6/30/2016
Wits Seminar
38
Putting It All Together
• This leads to
–Recall for BH:
–Shock gives exactly the same drag as BH for L = p T
• L is typical mom. scale of nucleus
• Can test AdS HQ E-Loss in both hot and
cold nuclear matter!
6/30/2016
Wits Seminar
39
Intermediate-pT Muck at RHIC and LHC
• Messy (nonperturbative)
physics below
~ 10 GeV/c
6/30/2016
P Jacobs, HP2012
Wits Seminar
40
Clean High-pT Probes
v2
CMS 40-50%
h±
pT (GeV/c)
– RHIC
• HF sep.
• sPHENIX
– LHC
• High-pT HF
– RAA, v2, corr.
• Control p+A
6/30/2016
B Cole, HP2012
Wits Seminar
– CNM AdS
41
Conclusions
• Exciting times in HI physics!
• LHC & RHIC converging on picture of pQCD E-loss in sQGP
– LO thermal pQCD qualitatively describes LHC single particle
observables
• Constrained by conservative assumptions & RHIC
• Corrections likely large; drive towards data?
• Difficult to reconcile AdS/CFT with new LHC data
– D mesons oversuppressed
– Light flavor hard to compare, but likely oversuppressed
• Much interesting research to do:
– HF sep. at RHIC, and esp. LHC; RAA and v2
• Does bottom flow??
– Properly include, e.g. NLO effects in pQCD E-Loss
– Understand AdS E-Loss better
• Importance of fluctuations
• Light parton jets
• p+A
6/30/2016
Wits Seminar
42
Backup Slides
6/30/2016
Wits Seminar
43
Measuring the IC
• eRHIC could give experimental handle on
initial geometry
– Recall e + A diffraction exps. on A at rest
Hahn, Ravenhall, and Hofstadter, Phys Rev 101 (1956)
6/30/2016
Wits Seminar
44
Gluon Distribution of A at x ~ 10-3
e
e
g*
J/y
• Coherent vector meson
production in e + A
A
A
Must reject incoherent
collisions at ~100%
WAH, arXiv:1102.5058
6/30/2016
Wits Seminar
45
Rise in RAA a Final State Effect?
– Is rise really due to
pQCD?
– Or other quench (flat?)
+ initial state CNM
effects a la CGC?
Y-J Lee, QM11
Albacete and Marquet, PLB687 (2010)
Require p + A and/or direct g
6/30/2016
PHENIX PRL98, 2007
Wits Seminar
46
Null IS at LHC
• Direct photon RAA ~ 1
– Soon add’l p + A null
control exp.
• CNM E-Loss?
C Roland, HP2012
D Perepelitsa, HP2012
6/30/2016
Wits Seminar
47