Workshop on Heavy Ion Collisions at the LHC: Last Call for Predictions

Abstract

This writeup is a compilation of the predictions for the forthcoming Heavy Ion Program at the Large Hadron Collider, as presented at the CERN Theory Institute 'Heavy Ion Collisions at the LHC - Last Call for Predictions', held from May 14th to June 10th 2007.

Full text

arXiv:0711.0974v1 [hep-ph] 6 Nov 2007
Heavy Ion Collisions at the LHC - Last Call for Predictions
N Armesto1, N Borghini2, S Jeon3and U A Wiedemann4(editors)
S Abreu5, V Akkelin6, J Alam7, J L Albacete8, A Andronic9, D
Antonov10, F Arleo4‡, N Armesto1, I C Arsene11, G G Barnaföldi12, J
Barrette3, B Bäuchle13,14, F Becattini15, B Betz13,16, M Bleicher13, M
Bluhm17, D Boer18, F W Bopp19, P Braun-Munzinger9,20, L Bravina11,21,
W Busza22, M Cacciari23, A Capella24, J Casalderrey-Solana25, R
Chatterjee8,7, L-W Chen26,27, J Cleymans28, B A Cole29, Z Conesa Del
Valle30, L P Csernai14,12, L Cunqueiro1, A Dainese31, J Dias de Deus5,
H-T Ding32, M Djordjevic8, H Drescher33, I M Dremin34 A Dumitru13, A
El13, R Engel35, D d’Enterria36, K J Eskola37,38, G Fai39, E G Ferreiro1,
R J Fries40,41, E Frodermann8, H Fujii42, C Gale3, F Gelis4,
V P Gonçalves43, V Greco44, C Greiner13, M Gyulassy45,33, H van Hees40,
U Heinz4§, H Honkanen37,38,46, W A Horowitz45,33, E Iancu47,
G Ingelman48, J Jalilian-Marian49, S Jeon3, A B Kaidalov50, B
Kämpfer17,51, Z-B Kang52, Iu A Karpenko6, G Kestin8, D Kharzeev53,
C M Ko40, B Koch33,13, B Kopeliovich54,55,10, M Kozlov2, I Kraus20,56, I
Kuznetsova57, S H Lee58, R Lednicky55,59, J Letessier23, E Levin60,
B-A Li61, Z-W Lin62, H Liu63, W Liu40, C Loizides22, I P Lokhtin21,
M V T Machado43, L V Malinina21,55, A M Managadze21, M L
Mangano4, M Mannarelli64, C Manuel64, G Martínez30, J G Milhano5,
Á Mócsy41, D Molnár65,41, M Nardi66, J K Nayak7, H Niemi37,38,
H Oeschler20, J-Y Ollitrault47, G Pai´
c67, C Pajares1, V S Pantuev68, G
Papp69, D Peressounko70, P Petreczky53, S V Petrushanko21, F
Piccinini71, T Pierog35, H J Pirner10, S Porteboeuf30, I Potashnikova54,
G Y Qin3, J-W Qiu52,53, J Rafelski57,4, K Rajagopal63, J Ranft19, R
Rapp40, S S Räsänen37, J Rathsman48, P Rau13, K Redlich72,
T Renk37,38, A H Rezaeian54, D Rischke13,33, S Roesler73, J Ruppert3,
P V Ruuskanen37,38, C A Salgado1,74, S Sapeta4,75, I Sarcevic57, S
Sarkar7, L I Sarycheva21, I Schmidt54, A I Shoshi2, B Sinha7, Yu M
Sinyukov6, A M Snigirev21, D K Srivastava7, J Stachel76, A Stasto77, H
Stöcker13,33,9, C Yu Teplov21, R L Thews57, G Torrieri13,33,
V Topor Pop3, D N Triantafyllopoulos78, K L Tuchin52,41, S Turbide3, K
‡On leave from Laboratoire d’Annecy-le-Vieux de Physique Théorique (LAPTH), UMR 5108 du CNRS
associée à l’Université de Savoie, B.P. 110, 74941 Annecy-le-Vieux Cedex, France.
§On leave from 8.

Heavy Ion Collisions at the LHC - Last Call for Predictions 2
Tywoniuk11, A Utermann18, R Venugopalan53, I Vitev79, R Vogt80,81, E
Wang32,25, X N Wang25, K Werner30, E Wessels18, S Wheaton20,28, S
Wicks45,33, U A Wiedemann4, G Wolschin10, B-W Xiao45, Z Xu13,
S Yasui57, E Zabrodin11,21, K Zapp77, B Zhang82, B-W Zhang40k, H
Zhang32 and D Zhou32 (authors)¶
1Departamento de Física de Partículas and IGFAE, Universidade de Santiago de Compostela,
15782 Santiago de Compostela, Spain
2Fakultät für Physik, Universität Bielefeld, D-33501 Bielefeld, Germany
3Department of Physics, McGill University, Montréal, Canada H3A 2T8
4CERN, PH Department, TH Division, 1211 Geneva 23, Switzerland
5Instituto Superior Técnico/CENTRA, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal
6Bogolyubov Institute for Theoretical Physics, Metrolohichna str. 14-b, 03680 Kiev-143,
Ukraine
7Variable Energy Cyclotron Centre, 1/AF Bidhan Nagar, Kolkata 700 064, India
8Department of Physics, The Ohio State University, 191 W. WoodruffAvenue, OH-43210,
Columbus, USA
9Gesellschaft für Schwerionenforschung, GSI, D-64291 Darmstadt, Germany
10 Institut für Theoretische Physik, Universität Heidelberg, Philosophenweg 19, D-69120
Heidelberg, Germany
11 Department of Physics, University of Oslo, N-0316 Oslo, Norway
12 MTA KFKI RMKI, P.O. Box 49, Budapest 1525, Hungary
13 Institut für Theoretische Physik, Universität Frankfurt, Max-von-Laue-Straße 1, D-60438
Frankfurt am Main, Germany
14 Section for Theoretical Physics, Departement of Physics, University of Bergen, Allégaten
55, 5007 Bergen, Norway
15 Università di Firenze and INFN Sezione di Firenze, Via G. Sansone 1, I-50019, Sesto F.no,
Firenze, Italy
16 Helmholtz Research School, GSI, FIAS and Universität Frankfurt, Germany
17 Forschungszentrum Dresden-Rossendorf, PF 510119, 01314 Dresden, Germany
18 Department of Physics and Astronomy, VU University Amsterdam, De Boelelaan 1081,
1081 HV Amsterdam, The Netherlands
19 Siegen University, Siegen, Germany
20 Institut für Kernphysik, Technical University Darmstadt, D-64283 Darmstadt, Germany
21 Skobeltsyn Institute of Nuclear Physics, Moscow State University, RU-119899 Moscow,
Russia
22 Massachusetts Institute of Technology, Cambridge MA, USA
23 LPTHE, Université Pierre et Marie Curie (Paris VI), France
24 Laboratoire de Physique Théorique, Université de Paris XI, Bâtiment 210, 91405 Orsay
Cedex, France
25 Lawrence Berkeley National Laboratory, 1 Cyclotron Road, MS 70R0319, Berkeley, CA
94720, USA
26 Institute of Theoretical Physics, Shanghai Jiao Tong University, Shanghai 200240, China
27 Center of Theoretical Nuclear Physics, National Laboratory of Heavy Ion Accelerator,
Lanzhou 730000, China
28 UCT-CERN Research Centre and Department of Physics, University of Cape Town,
Rondebosch 7701, South Africa
kOn leave from 32.
¶The contributors on this author list have contributed only to those subsections of the report, which they cosign
with their name. Only those have collaborated together, whose names appear together in the header of a given
subsection.

Heavy Ion Collisions at the LHC - Last Call for Predictions 3
29 Nevis Laboratory, Columbia University, New York, USA
30 Subatech (CNRS/IN2P3 - Ecole des Mines - Université de Nantes) Nantes, France
31 INFN, Laboratori Nazionali di Legnaro, Legnaro (Padova), Italy
32 Institute of Particle Physics, Central China Normal University, Wuhan, China
33 Frankfurt Institute for Advanced Studies (FIAS), Johann Wolfgang Goethe-Universität,
Max-von-Laue-Str. 1, 60438 Frankfurt am Main, Germany
34 Lebedev Physical Institute, Leninsky pr. 53, 119991 Moscow, Russia
35 Forschungszentrum Karlsruhe, Karlsruhe, Germany
36 CERN/PH, CH-1211 Geneva 23, Switzerland
37 Department of Physics, P.O. Box 35, FI-40014, University of Jyväskylä, Finland
38 Helsinki Institute of Physics, P.O. Box 64, FI-00014,University of Helsinki, Finland
39 Department of Physics, Kent State University, Kent, OH, USA
40 Cyclotron Institute and Department of Physics, Texas A&M University, College Station
TX 77843, USA
41 RIKEN/BNL Research Center, Brookhaven National Laboratory, Upton NY 11973, USA
42 Institute of Physics, University of Tokyo, Komaba, Tokyo 153-8 902, Japan
43 Universidade Federal de Pelotas, Caixa Postal 354, CEP 96010-090, Pelotas, RS, Brazil
44 Dipartimento di Fisica e Astronomia, Via S. Sofia 64, I-95125 Catania, Italy
45 Physics Department, Columbia University, New York, New York 10027, USA
46 Department of Physics, University of Virginia, Charlottesville, VA, USA
47 Service de Physique Théorique, CEA/DSM/SPhT, CNRS/MPPU/URA2306, CEA Saclay,
F-91191 Gif-sur-Yvette Cedex
48 High Energy Physics, Uppsala University, Box 535, S-75121 Uppsala, Sweden
49 Department of Natural Sciences, Baruch College, New York, NY 10010, USA
50 Institute of Theoretical and Experimental Physics, RU-117259 Moscow, Russia
51 Institut für Theoretische Physik, TU Dresden, 01062 Dresden, Germany
52 Department of Physics and Astronomy, Iowa State University, Ames IA 50011, USA
53 Physics Department, Brookhaven National Laboratory, Upton, NY 11793-5000,USA
54 Departamento de Física y Centro de Estudios Subatómicos, Universidad Técnica Federico
Santa María, Casilla 110-V, Valparaíso, Chile
55Joint Institute for Nuclear Research, Dubna, Moscow Region, 141980, Russia
56 Nikhef, Kruislaan 409, 1098 SJ Amsterdam, The Netherlands
57 Department of Physics, University of Arizona, Tucson, Arizona 85721, USA
58 Institute of Physics and Applied Physics, Yonsei University, Seoul 120-749, Korea
59 Institute of Physics ASCR, Prague, 18221, Czech Republic
60 HEP Department, School of Physics, Raymond and Beverly Sackler Faculty of Exact
Science, Tel Aviv University, Tel Aviv 69978, Israel
61 Department of Physics, Texas A&M University-Commerce, Commerce, Texas
75429-3011, USA
62 Mail Stop VP62, NSSTC, 320 Sparkman Dr., Huntsville, AL 35805 and Department of
Physics, East Carolina University,Greenville, North Carolina 27858-4353, USA
63 Center for Theoretical Physics, MIT, Cambridge, MA 02139, USA
64 Instituto de Ciencias del Espacio (IEEC/CSIC), Campus U.A.B., Fac. de Ciències, Torre
C5, E-08193 Bellaterra (Barcelona), Spain
65 Physics Department, Purdue University, West Lafayette, IN 47907, USA
66 INFN, Sezione di Torino, via Giuria N.1, 10125 Torino, Italy
67 Instituto de Ciencias Nucleares, UNAM, Mexico City, Mexico
68 University at Stony Brook, Stony Brook, New York 11794, USA
69 Department of Theoretical Physics, ELTE, Pázmány P. 1/A, Budapest 1117, Hungary
70 RRC “Kurchatov Institute”, Kurchatov Sq. 1, Moscow 123182, Russia
71 INFN Sezione di Pavia, Pavia, Italy

Heavy Ion Collisions at the LHC - Last Call for Predictions 4
72 Institute of Theoretical Physics, University of Wrocław, PL-50204 Wrocław, Poland
73 CERN/SC, CH-1211 Geneva 23, Switzerland
74 Dipartimento di Fisica, Università di Roma “La Sapienza” and INFN, Roma, Italy
75 M. Smoluchowski Institute of Physics, Jagellonian University, Reymonta 4, 30-059
Cracow, Poland
76 Physikalisches Institut der Universität Heidelberg, D-69120 Heidelberg, Germany
77 Physics Department, Penn State University, PA 16802-6300, USA
78 ECT∗, Villa Tambosi, Strada delle Tabarelle 286, I-38050 Villazzano (TN), Italy
79 Los Alamos National Laboratory, Theoretical Division, Mail Stop B283, Los Alamos, NM
87545, USA
80 Lawrence Livermore National Laboratory, Livermore, CA, USA
81 Physics Department, University of California at Davis, Davis, CA, USA
82 Department of Chemistry and Physics, Arkansas State University, State University,
Arkansas 72467-0419, USA
Abstract. This writeup is a compilation of the predictions for the forthcoming Heavy Ion
Program at the Large Hadron Collider, as presented at the CERN Theory Institute ’Heavy Ion
Collisions at the LHC - Last Call for Predictions’, held from May 14th to June 10th 2007.

CONTENTS 5
Contents
1 Multiplicities and multiplicity distributions 10
1.1 Multiplicity distributions in rapidity for Pb-Pb and p-Pb central collisions
from a simple model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2 Multiplicities in Pb-Pb central collisions at the LHC from running coupling
evolution and RHIC data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3 Identified hadron spectra in Pb-Pb at √sNN =5.5 TeV: hydrodynamics+pQCD
predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Multiplicities at the LHC in a geometric scaling model . . . . . . . . . . . . 14
1.5 Multiplicity and cold-nuclear matter effects from Glauber-Gribov theory . . . 16
1.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.5.2 Particle production at LHC . . . . . . . . . . . . . . . . . . . . . . . 17
1.6 Stopping Power from SPS to LHC energies. . . . . . . . . . . . . . . . . . . 18
1.7 Investigating the extended geometric scaling region at LHC with polarized
and unpolarized final states . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.7.1 DHJ model prediction for charged hadron production . . . . . . . . . 21
1.7.2 DHJ model prediction for Λpolarization . . . . . . . . . . . . . . . 22
1.8 Inclusive distributions at the LHC as predicted from the DPMJET-III model
with chain fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.9 Some “predictions” for PbPb and pp at LHC, based on the extrapolation of
data at lower energies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.10 Multiplicities and J/ψ suppression at LHC energies . . . . . . . . . . . . . . 26
1.10.1 Multiplicities with shadowing corrections . . . . . . . . . . . . . . . 26
1.10.2 J/ψ suppression . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.11 Heavy ion collisions at LHC in a Multiphase Transport Model . . . . . . . . 30
1.12 Multiplicity distributions and percolation of strings . . . . . . . . . . . . . . 33
1.13 Shear Viscosity to Entropy within a Parton Cascade . . . . . . . . . . . . . . 34
1.14 Hadron multiplicities, pTspectra and net-baryon number in central Pb+Pb
collisions at the LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
1.15 Melting the Color Glass Condensate at the LHC . . . . . . . . . . . . . . . . 38
1.15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
1.15.2 Particle multiplicity in central AA collisions . . . . . . . . . . . . . . 38
1.15.3 Heavy quark production in pA collisions . . . . . . . . . . . . . . . 38
1.16 RpA ratio: total shadowing due to running coupling . . . . . . . . . . . . . . 39
1.17 LHC dNch/dηand Nch from Universal Behaviors . . . . . . . . . . . . . . . 41
1.18 Hadron multiplicities at the LHC . . . . . . . . . . . . . . . . . . . . . . . . 42
1.19 CGC at LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
1.20 Fluctuation Effects on RpA at High Energy . . . . . . . . . . . . . . . . . . . 45
1.21 Particle Production at the LHC: Predictions from EPOS . . . . . . . . . . . . 46
1.22 Forward hadron production in high energy pA collisions . . . . . . . . . . . 48
1.23 Rapidity distributions at LHC in the Relativistic Diffusion Model . . . . . . . 51

CONTENTS 6
2 Azimuthal asymmetries 53
2.1 Transverse momentum spectra and elliptic flow: Hydrodynamics with QCD-
based equations of state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.1.2 Predictions for heavy-ion collisions at LHC . . . . . . . . . . . . . . 55
2.2 The centrality dependence of elliptic flow at LHC . . . . . . . . . . . . . . . 55
2.3 Elliptic flow from pQCD+saturation+hydro model . . . . . . . . . . . . . . 56
2.4 From RHIC to LHC: Elliptic and radial flow effects on hadron spectra . . . . 59
2.5 Differential elliptic flow prediction at the LHC from parton transport . . . . . 60
3 Hadronic flavor observables 62
3.1 Thermal model predictions of hadron ratios . . . . . . . . . . . . . . . . . . 62
3.2 (Multi)Strangeness Production in Pb+Pb collisions at LHC. HIJING/B¯
B v2.0
predictions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.3 Antibaryon to Baryon Production Ratios in Pb-Pb and p-p collision at LHC
energies of the DPMJET-III Monte Carlo . . . . . . . . . . . . . . . . . . . 65
3.4 Statistical model predictions for pp and Pb-Pb collisions at LHC . . . . . . . 68
3.5 Universal behavior of baryons and mesons’ transverse momentum distribu-
tions in the framework of percolation of strings . . . . . . . . . . . . . . . . 69
3.6 Bulk hadron(ratio)s at the LHC-ions . . . . . . . . . . . . . . . . . . . . . . 70
4 Correlations at low transverse momentum 75
4.1 Pion spectra and HBT radii at RHIC and LHC . . . . . . . . . . . . . . . . . 75
4.2 Mach Cones at central LHC Collisions via MACE . . . . . . . . . . . . . . . 76
4.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.2.2 Correlation functions . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.2.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
4.3 Study of Mach Cones in (3+1)d Ideal Hydrodynamics at LHC Energies . . . 78
4.4 Forward-Backward (F-B) rapidity correlations in a two step scenario . . . . . 79
4.5 Cherenkov rings of hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.6 Evolution of pion HBT radii from RHIC to LHC – predictions from ideal
hydrodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.7 Correlation radii by FAST HADRON FREEZE-OUT GENERATOR . . . . . 85
4.8 Exciting the quark-gluon plasma with a relativistic jet . . . . . . . . . . . . . 85
5 Fluctuations 87
5.1 Fluctuations and the clustering of color sources . . . . . . . . . . . . . . . . 87
5.2 Fluctuations of particle multiplicities from RHIC to LHC . . . . . . . . . . . 89
6 High transverse momentum observables and jets 91
6.1 Jet quenching parameter ˆqfrom Wilson loops in a thermal environment . . . 91
6.2 Particle Ratios at High pTat LHC Energies . . . . . . . . . . . . . . . . . . 92
6.3 π0fixed p⊥suppression and elliptic flow at LHC . . . . . . . . . . . . . . . . 94

CONTENTS 7
6.3.1 π0fixed p⊥suppression . . . . . . . . . . . . . . . . . . . . . . . . 94
6.3.2 Elliptic flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
6.4 Energy dependence of jet transport parameter . . . . . . . . . . . . . . . . . 96
6.5 PQM prediction of RAA(pT) and RCP(pT) at midrapidity in Pb–Pb collisions
at the LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.6 Effect of dynamical QCD medium on radiative heavy quark energy loss . . . 98
6.7 Charged hadron RAA as a function of pTat LHC . . . . . . . . . . . . . . . . 101
6.8 Nuclear suppression of jets and RAA at the LHC . . . . . . . . . . . . . . . . 102
6.9 Perturbative jet energy loss mechanisms: learning from RHIC, extrapolating
to LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.10 Jet evolution in the Quark Gluon Plasma . . . . . . . . . . . . . . . . . . . . 106
6.11 Pion and Photon Spectra at LHC . . . . . . . . . . . . . . . . . . . . . . . . 108
6.12 Transverse momentum broadening of vector bosons in heavy ion collisions at
the LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
6.13 Nuclear modification factors for high transverse momentum pions and protons
at LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
6.14 Quenching of high-pThadrons: Alternative scenario . . . . . . . . . . . . . 112
6.15 Expectations from AdS/CFT for Heavy Ion Collisions at the LHC . . . . . . 114
6.15.1 Jet quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
6.15.2 Quarkonium suppression . . . . . . . . . . . . . . . . . . . . . . . . 115
6.16 High-pTobservables in PYQUEN model . . . . . . . . . . . . . . . . . . . 116
6.16.1 Nuclear modification factors for jet and high-pThadrons . . . . . . . 116
6.16.2 Medium-modified jet fragmentation function . . . . . . . . . . . . . 116
6.16.3 Azimuthal anisotropy of jet quenching . . . . . . . . . . . . . . . . . 118
6.16.4 PT-imbalance in dimuon tagged jet events . . . . . . . . . . . . . . . 118
6.16.5 High-mass dimuon and secondary J/ψ spectra . . . . . . . . . . . . 119
6.17 Predictions for LHC heavy ion program within finite sQGP formation time . . 119
6.18 Hadrochemistry of jet quenching at the LHC . . . . . . . . . . . . . . . . . . 122
6.19 GLV predictions for light hadron production and suppression at the LHC . . . 123
6.20 NLO Predictions for Single and Dihadron Suppression in Heavy-ion
Collisions at LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
7 Heavy quarks and quarkonium 126
7.1 Statistical hadronization model predictions for charmed hadrons . . . . . . . 126
7.2 Nuclear suppression for heavy flavors in PbPb collisions at the LHC . . . . . 127
7.3 Heavy-quark production from Glauber-Gribov theory at LHC . . . . . . . . . 130
7.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
7.3.2 Heavy-quark production at the LHC . . . . . . . . . . . . . . . . . . 131
7.4 RAA(pt) and RCP(pt) of single muons from heavy quark and vector boson
decays at the LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
7.5 Quarkonium production in coherent pp/AA collisions and small-xphysics . . 134
7.6 Heavy-Quark Kinetics in the QGP at LHC . . . . . . . . . . . . . . . . . . . 135

Heavy Ion Collisions at the LHC - Last Call for Predictions 8
7.7 Ratio of charm to bottom RAA as a test of pQCD vs. AdS/CFT energy loss . . 136
7.8 Thermal charm production at LHC . . . . . . . . . . . . . . . . . . . . . . . 139
7.9 Charm production in nuclear collisions . . . . . . . . . . . . . . . . . . . . . 140
7.9.1 Higher twist shadowing . . . . . . . . . . . . . . . . . . . . . . . . 141
7.9.2 Process dependent leading twist gluon shadowing . . . . . . . . . . . 141
7.10 Charm and Beauty Hadrons from Strangeness-rich QGP at LHC . . . . . . . 142
7.11 Charmonium Suppression in Strangeness-rich QGP . . . . . . . . . . . . . . 143
7.12 J/ψ pTspectra from in-medium recombination . . . . . . . . . . . . . . . . 145
7.13 Predictions for quarkonia dissociation . . . . . . . . . . . . . . . . . . . . . 147
7.14 Heavy flavor production and suppression at the LHC . . . . . . . . . . . . . 148
7.15 Quarkonium shadowing in pPb and Pb+Pb collisions . . . . . . . . . . . . . 149
7.16 Quarkonium suppression as a function of pT. . . . . . . . . . . . . . . . . . 151
8 Leptonic probes and photons 153
8.1 Thermal photons to dileptons ratio at LHC . . . . . . . . . . . . . . . . . . . 153
8.2 Prompt photon in heavy ion collisions at the LHC: A “multi-purpose” observable156
8.3 Direct photon spectra in Pb-Pb at √sNN =5.5 TeV: hydrodynamics+pQCD
predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
8.4 Elliptic flow of thermal photons from RHIC to LHC . . . . . . . . . . . . . . 158
8.5 Asymmetrical in-medium mesons . . . . . . . . . . . . . . . . . . . . . . . 160
8.6 Photons and Dileptons at LHC . . . . . . . . . . . . . . . . . . . . . . . . . 162
8.7 Direct photons at LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
8.7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
8.7.2 Color dipole approach and predictions for LHC . . . . . . . . . . . . 164
8.8 Thermal Dileptons at LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
8.9 Direct γproduction and modification at the LHC . . . . . . . . . . . . . . . 167
9 Others 168
9.1 The effects of angular momentum conservation in relativistic heavy ion
collisions at the LHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
9.2 Black hole predictions for LHC . . . . . . . . . . . . . . . . . . . . . . . . . 170
9.2.1 From the hierarchy-problem to black holes in large extra dimensions . 171
9.2.2 From black hole evaporation to LHC observables . . . . . . . . . . . 171
9.3 Charmed exotics from heavy ion collision . . . . . . . . . . . . . . . . . . . 172
9.4 Alignment as a result from QCD jet production or new still unknown physics
at the LHC? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
Preface
In August 2006, the CERN Theory Unit announced to restructure its visitor program and to
create a "CERN Theory Institute", where 1-3 month long specific programs can take place.
The first such Institute was held from 14 May to 10 June 2007, focussing on "Heavy Ion

Heavy Ion Collisions at the LHC - Last Call for Predictions 9
Collisions at the LHC - Last Call for Predictions". It brought together close to 100 scientists
working on the theory of ultra-relativistic heavy ion collisions. The aim of this workshop was
to review and document the status of expectations and predictions for the heavy ion program
at the Large Hadron Collider LHC before its start. LHC will explore heavy ion collisions at
∼30 times higher center of mass energy than explored previously at the Relativistic Heavy Ion
Collider RHIC. So, on the one hand, the charge of this workshop provided a natural forum for
the exchange of the most recent ideas, and allowed to monitor how the understanding of heavy
ion collisions has evolved in recent years with the data from RHIC, and with the preparation of
the LHC experimental program. On the other hand, the workshop aimed at a documentation
which helps to distinguish pre- from post-dictions. An analogous documentation of the "Last
Call for Predictions" [1] was prepared prior to the start of the heavy-ion program at the
Relativistic Heavy Ion Collider RHIC, and it proved useful in the subsequent discussion and
interpretation of RHIC data. The present write-up is the documentation of predictions for the
LHC heavy ion program, received or presented during the CERN TH Institute. The set-up of
the CERN TH Institute allowed us to aim for the wide-most coverage of predictions. There
were more than 100 presentations and discussions during the workshop. Moreover, those
unable to attend could still participate by submitting predictions in written form during the
workshop. This followed the spirit that everybody interested in making a prediction had the
right to be heard.
To arrive at a concise document, we required that each prediction should be summarized
on at most two pages, and that predictions should be presented, whenever possible, in figures
which display measurable quantities. Full model descriptions were not accepted - the authors
were encouraged to indicate the relevant references for the interested reader. Participants
had the possibility to submit multiple contributions on different topics, but it was part of
the subsequent editing process to ensure that predictions on neighboring topics were merged
wherever possible. The contributions summarized here are organized in several sections,
- though some of them contain material related with more than one section -, roughly by
going from low transverse momentum to high transverse momentum and from abundant to
rare measurements. In the low transverse momentum regime, we start with predictions on
multiplicity distributions, azimuthal asymmetries in particle production and hadronic flavor
observables, followed by correlation and fluctuation measurements. The contributions on
hard probes at the LHC start with predictions for single inclusive high transverse momentum
spectra, and jets, followed by heavy quark and quarkonium measurements, leptonic probes
and photons. A final section "Others" encompasses those predictions which do not fall
naturally within one of the above-mentioned categories, or discuss the more speculative
phenomena that may be explored at the LHC.
We would like to end this Preface by thanking the TH Unit at CERN for its generous
support of this workshop. Special thanks go to Elena Gianolio, Michelle Mazerand, Nanie
Perrin and Jeanne Rostant, whose help and patience was invaluable.
Néstor Armesto
Nicolas Borghini
Sangyong Jeon

Heavy Ion Collisions at the LHC - Last Call for Predictions 10
Urs Achim Wiedemann
1. Multiplicities and multiplicity distributions
1.1. Multiplicity distributions in rapidity for Pb-Pb and p-Pb central collisions from a simple
model
S. Abreu, J. Dias de Deus and J. G. Milhano
The simple model [2] for the distribution of rapidity extended objects (longitudinal
glasma colour fields or coloured strings) created in a heavy ion collision combines the
generation of lower centre-of-mass rapidity objects from higher rapidity ones with asymptotic
saturation in the form of the well known logistic equation for population dynamics
∂ρ
∂(−∆)=1
δ(ρ−Aρ2),(1)
where ρ≡ρ(∆,Y) is the particle density, Yis the beam rapidity, and ∆≡ |y| −Y. The Y-
dependent limiting value of ρis determined by the saturation condition ∂(−∆)ρ=0−→ ρY=
1/A, while the separation between the low density (positive curvature) and high density
(negative curvature) regions is given by ∂2(−∆)ρ|∆0=0−→ ρ0≡ρ(∆0,Y)=ρY/2. Integrating
(1) we get
ρ(∆,Y)=ρY
e∆−∆0
δ+1.(2)
In the String Percolation Model [2] the particle density is proportional, once the colour
reduction factor is taken into account, to the average number of participants ρ∝NA; the
normalized particle density at mid-rapidity is related to the gluon distribution at small
Bjorken-x,ρ∝eλY; and the dense-dilute separation scale decreases, from energy conservation,
linearly with Y,∆0=−αYwith 0 < α < 1. Rewriting (2) in rapidity
ρ≡dN
dy=NA·eλY
e|y|−(1−α)Y
δ+1.(3)
The values λ=0.247, α=0.269 and δ=0.67 for the parameters in the solution (3) are fixed
by an overall fit [3] of Au-Au RHIC data [4].
In Fig. 1 we show the predicted multiplicity distribution for the 10% most central Pb-Pb
collisions at √sNN =5.5 TeV with NA=173.3 taken from the Glauber calculation in [5].
In Fig. 2 we show the predicted multiplicity distribution for the 20% most central p-Pb
collisions at √sNN =8.8 TeV with Npart =13.07 also from [5]. In this case the solution (3)
have been modified to account for the asymmetric geometry and the shift of the centre of
mass of the system relatively to the laboratory centre of mass [2]. The resulting rapidity shift
yc=−2.08 is marked in Fig. 2.
1.2. Multiplicities in Pb-Pb central collisions at the LHC from running coupling evolution
and RHIC data
J. L. Albacete

Heavy Ion Collisions at the LHC - Last Call for Predictions 11
y
-10 -5 0 5 10
dy
dN
0
200
400
600
800
1000
1200
1400
1600
Figure 1: dN
dyfrom (3) for Pb-Pb (0-10% central) collisions at √sNN =5.5 TeV with NA=173.3
y
-10 -5 0 5 10
dy
dN
0
5
10
15
20
25
30
35
=-2.08
c
y
Figure 2: dN
dyfrom asymmetric version of (3) [2] for p-Pb (0-20% central) collisions at
√sNN =8.8 TeV with Npart =13.07.
Predictions for the pseudorapidity density of charged particles produced in Pb-Pb central
collisions at √sNN =5.5 TeV are presented. Particle production in such collisions is computed in the
framework of kt-factorization, using running coupling non-linear evolution to determine the transverse
momentum and rapidity dependence of the nuclear unintegrated gluon distributions.
Predictions for the pseudorapidity density of charged particles produced in Pb-Pb central
collisions at √sNN =5.5 TeV presented in [6] are summarized. Primary gluon production in
such collisions can be computed perturbatively in the framework of kt-factorization. Under
the additional assumption of local parton-hadron duality, the rapidity density of produced
charged particles in nucleus-nucleus collisions at energy √sand impact parameter bis given
by, [7]:

Heavy Ion Collisions at the LHC - Last Call for Predictions 12
dN
dyd2b=C4πNc
N2
c−1Zpkin d2pt
p2
tZptd2ktαs(Q)ϕx1,|kt+pt|
2ϕx2,|kt−pt|
2,(4)
where ptand yare the transverse momentum and rapidity of the produced particle, x1,2=
(pt/√s)e±yand Q=0.5max{|pt±kt|}. The lack of impact parameter integration in this
calculation and the gluon to charged hadron ratio are accounted for by the constant C, which
sets the normalization. The nuclear unintegrated gluon distributions (u.g.d.), ϕ(x,k), entering
Eq. (4) are taken from numerical solutions of the Balitsky-Kovchegov evolution equation
including running coupling corrections, [8]:
∂N(Y,r)
∂Y=R[N(Y,r)] −S[N(Y,r)] (5)
Explicit expressions for the running,R[N], and subtraction,S[N], functionals in the r.h.s.
of Eq. (5) can be found in [8]. The nuclear u.g.d. are given by the Fourier transform of the
dipole scattering amplitude evolved according to Eq. (5), ϕ(Y,k)=Rd2r
2πr2eik·rN(Y,r), with
Y=ln(0.05/x)+ ∆Yev, where ∆Yev is a free parameter. Large-xeffects have been included by
replacing ϕ(x,k)→ϕ(x,k)(1 −x)4. The initial condition for the evolution is taken from the
McLerran-Venugopalan model [9], which is believed to provide a good description of nuclear
distribution functions at moderate energies:
NMV(Y=0,r)=1−exp





−r2Q2
0
4ln 1
rΛ+e!





,(6)
where Q0is the initial saturation scale and Λ = 0.2 GeV. In order to compare Eq. (4)
with experimental data it is necessary to correct the difference between rapidity, y, and the
experimentally measured pseudorapidity, η. This is managed by introducing an effective
hadron mass, me f f . The variable transformation, y(η, pt,me f f ), and its corresponding jacobian
are given by Eqs.(25-26) in [7]. Corrections to the kinematics due to the hadron mass are also
considered by replacing pt→mt=p2
t+m2
e f f 1/2in the evaluation of x1,2. This replacement
affects the predictions for the LHC by less than a 5%, [6].
The results for the pseudorapidity density of charged particles in central Au-Au collisions
at √sNN =130, 200 and 5500 GeV are shown in Fig. 3. A remarkably good description of
RHIC data is obtained with Q0=0.75÷1.25 GeV, ∆Yev.3 and mef f =0.2÷0.3 GeV. Assuming
no difference between Au and Pb nuclei, the extrapolation of the fits to RHIC data yields the
following band: dNPb−Pb
ch
dη(√sNN =5.5TeV)|η=0≈1290÷1480 for central Pb-Pb collisions at the
LHC. The central value of our predictions dNPb−Pb
ch
dη(√sNN =5.5TeV)|η=0≈1390 corresponds
to the best fits to RHIC data.
1.3. Identified hadron spectra in Pb-Pb at √sNN =5.5 TeV: hydrodynamics+pQCD
predictions
F. Arleo, D. d’Enterria and D. Peressounko
The single inclusive charged hadron pTspectra in Pb-Pb collisions at the LHC, predicted by a
combined hydrodynamics+perturbative QCD (pQCD) approach are presented.

Heavy Ion Collisions at the LHC - Last Call for Predictions 13
-4 -2 0 2 4
200
400
600
800
1000
1200
1400
1600
ηdch
dN
η
=200 GeV
NN
sAu-Au 0-6%,
=130 GeV
NN
sAu-Au 0-6%,
=5.5 TeV
NN
sPb-Pb
Figure 3: Multiplicity densities for Au-Au central collisions at RHIC (experimental data taken
from [4]), and prediction for Pb-Pb central collisions at √sNN =5.5 TeV. The best fits to data
(solid lines) are obtained with Q0=1 GeV, ∆Yev =1 and me f f =0.25 GeV. The upper limit of
the error bands correspond to ∆Yev =3 and Q0=0.75 GeV, and the lower limit to ∆Yev =0.5
and Q0=1.25 GeV, with me f f =0.25 GeV in both cases.
We present predictions for the inclusive transverse momentum distributions of pions,
kaons and (anti)protons produced at mid-rapidity in Pb-Pb collisions at √sNN =5.5 TeV
based on hydrodynamics+pQCD calculations. The bulk of the spectra (pT.5 GeV/c) in
central Pb-Pb at the LHC is computed with a hydrodynamical model – successfully tested
at RHIC [10] – using an initial entropy density extrapolated empirically from the hadron
multiplicities measured at RHIC: dNch/dη|η=0/(0.5Npart)≈0.75ln(√sNN /1.5) [11]. Above
pT≈3 GeV/c, additional hadron production from (mini)jet fragmentation is computed from
collinearly factorized pQCD cross sections at next-to-leading-order (NLO) accuracy [12]. We
use recent parton distribution functions (PDF) [13] and fragmentation functions (FF) [14],
modified respectively to account for initial-state shadowing[15] and final-state parton energy
loss [16].
We use cylindrically symmetric boost-invariant 2+1-D relativistic hydrodynamics, fix-
ing the initial conditions for Pb-Pb at b=0 fm and employing a simple Glauber prescription
to obtain the corresponding source profiles at all other centralities [10]. The initial source
is assumed to be formed at a time τ0=1/Qs≈0.1 fm/c, with an initial entropy density of
s0=1120 fm−3(i.e. εo∝s4/3
0≈650 GeV/fm3) so as to reproduce the expected final hadron

Heavy Ion Collisions at the LHC - Last Call for Predictions 14
multiplicity dNch/dη|η=0≈1300 at the LHC [11]. We follow the evolution of the system
by solving the equations of ideal hydrodynamics including the current conservation for net-
baryon number (the system is almost baryon-free, µB≈5 MeV). For temperatures above
(below) Tcrit ≈170 MeV the system is described with a QGP (hadron gas) equation of state
(EoS). The QGP EoS – obtained from a parametrization to recent lattice QCD results – is
Maxwell connected to the hadron resonance gas phase assuming a first-order phase transition.
As done for RHIC energies, we chemically freeze-out the system (i.e. fix the hadron ratios) at
Tcrit. Final state hadron spectra are obtained with the Cooper-Frye prescription at Tfo ≈120
MeV followed by decays of unstable resonances using the known branching ratios. Details
can be found at [10].
Our NLO pQCD predictions are obtained with the code of ref. [12] with all scales set to
µ=pT. Pb-Pb yields are obtained scaling the NLO cross-sections by the number of incoherent
nucleon-nucleon collisions for each centrality class given by a Glauber model (Ncoll =1670,
12.9 for 0-10%-central and 60-90%-peripheral). Nuclear (isospin and shadowing) correc-
tions of the CTEQ6.5M PDFs [13] are introduced using the NLO nDSg parametrization [15].
Final-state energy loss in the hot and dense medium is accounted for by modifying the AKK
FFs [14] with BDMPS quenching weights as described in [16]. The BDMPS medium-induced
gluon spectrum depends on a single scale ωc=hˆqiL2, related to the transport coefficient and
length of the medium. We use ωc≈50 GeV, from the expected energy dependence of the
quenching parameter and the measured ωc≈20 GeV at RHIC [16]. The inclusive hadron
spectra in central Pb-Pb are suppressed by up to a factor ∼10 (2), RPbPb ≈0.1 (0.5), at pT=
10 (100) GeV/c.
Our predictions for the identified hadron spectra in Pb-Pb collisions at 5.5 TeV are shown
in Figure 4. The hydrodynamical contribution dominates over the (quenched) pQCD one up to
pT≈4 (1.5) GeV/cin central (peripheral) Pb-Pb. As expected, the hydro-pQCD pTcrossing
point increases with the hadron mass. In the absence of recombination effects (not included
here), bulk protons may be boosted up to pT≈5 GeV/cin central Pb-Pb at the LHC.
1.4. Multiplicities at the LHC in a geometric scaling model
N. Armesto, C. A. Salgado and U. A. Wiedemann
We present predictions for charged multiplicities at mid-rapidity in PbPb collisions, as well as
transverse momentum distributions at different pseudorapidities in pPb collisions, at LHC energies.
We use geometric scaling as found in lepton-proton and lepton-nucleus scattering, to determine
the evolution of multiplicities with energy, pseudorapidity and centrality. The only additional free
parameter required to obtain the multiplicities is fixed from RHIC data.
Geometric scaling - the phenomenological finding that virtual photon-hadron cross
sections in lepton-proton [17] and lepton-nucleus [18,19] collisions, are functions of a single
variable which encodes all dependences on Bjorken-x, virtuality Q2and nuclear size A-
is usually considered as one of most important evidences in favor of saturation physics

Heavy Ion Collisions at the LHC - Last Call for Predictions 15
(GeV/c)
T
p
0 1 2 3 4 5 6 7 8
-2
dy) (GeV/c)
2
T
dpπN/(
2
d
-10
10
-9
10
-8
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
1
10
2
10
3
10
= 50 GeV)
c
ω (
loss
E
)/2
-
+h
+
total (h )/2
-
+h
+
hydro (h )/2
-
+h
+
NLO pQCD (h
pions
-2
10×kaons
-5
10× p
p,
Pb-Pb @ 5.5 TeV [0-10% central]
(GeV/c)
T
p
0 1 2 3 4 5 6 7 8
-2
dy) (GeV/c)
2
T
dpπN/(
2
d
-11
10
-10
10
-9
10
-8
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
1
10
2
10
)/2
-
+h
+
total (h
)/2
-
+h
+
hydro (h
)/2
-
+h
+
NLO pQCD (h
pions
-2
10×kaons
-5
10× p
p,
Pb-Pb @ 5.5 TeV [60-90% peripheral]
Figure 4: Spectra at y=0 for π±,K±,p,¯pin 0-10% central (left) and 60-70% peripheral (right)
Pb-Pb at √sNN =5.5 TeV, obtained with hydrodynamics +(quenched) pQCD calculations.
at work [19] . In the scaling variable τA=Q2/Q2
s,A(x) the quantity Qs,A, the saturation
momentum, shows a behavior with energy or Bjorken-xdetermined by lepton-proton data,
while the dependence on Ais fixed by lepton-nucleus data [18]:
Q2
s,A(x)∝x−λA1/(3δ), λ =0.288, δ =0.79±0.02.(7)
To compute particle production, we assume that geometric scaling holds for the
distributions assigned to the projectile and target. Without invoking factorization, dimensional
analysis allows us to factor out the geometrical information. Then, the multiplicity at central
pseudorapidity can be written in the form [18] (with Npart ∝A)
2
Npart
dNAA
dηη∼0
=N0s/GeV2λ/2N1−δ
3δ
part .(8)
The only new parameter is N0, a normalization constant which takes into account the parton-
hadron conversion and the change from mid-rapidity to mid-pseudorapidity. Once fixed for
a set of data (N0=0.47), this formula has predictive power and establishes a factorization
of the energy and centrality dependences in agreement with data. In Fig. 5 we show the
results of Eq. (8) compared to RHIC data (including those of intermediate energies [20])
and our prediction for the LHC, where our numbers for dNAA/dη|η∼0are 1550 ÷1760 for
Npart =350 and 1670÷1900 for Npart =375, with the range in the predictions reflecting the
uncertainty coming from δ, see Eq. (7). We note that these values are based on a √s-powerlaw
dependence in Eq. (2) and can be discriminated clearly from a log-extrapolation of RHIC data.
A more model-dependent application of this formalism [18] concerns particle production
in hadron-nucleus collisions at forward rapidities or large energies. Assuming factorization,
geometric scaling and a steeply falling parton distribution in the proton or deuteron, one gets

Heavy Ion Collisions at the LHC - Last Call for Predictions 16
Figure 5: Charged multiplicity at mid-pseudorapidity per participant pair, for four RHIC
energies and for LHC energies, from Eq. (8). The band shows theuncertainty coming from δ,
see Eq. (7).
for one-particle distributions in two centrality classes c1and c2
dNdAu
c1
Ncoll1dηd2pt,dNdAu
c2
Ncoll2dηd2pt≈Ncoll2φA(pt/Qsat1)
Ncoll1φA(pt/Qsat2)≈Ncoll2Φ(τ1)
Ncoll1Φ(τ2),(9)
where Φis the geometric scaling function in lepton-hadron collisions. In this way, particle
ratios in hadron-nucleus collisions provide a check of parton densities in the nucleus through
geometric scaling. In Fig. 6 we show the results compared to RHIC data and our predictions
for the LHC. The definition of the centrality classes is Ncoll1=13.6±0.3 (central), 7.9±0.4
(semicentral) and Ncoll2=3.3±0.4 (peripheral). The suppression for mid-pseudorapidities at
the LHC turns out to be as large as that for forward pseudorapidities at RHIC.
1.5. Multiplicity and cold-nuclear matter effects from Glauber-Gribov theory
I. C. Arsene, L. Bravina, A. B. Kaidalov, K. Tywoniuk and E. Zabrodin
We present predictions for nuclear modification factor in proton-lead collisions at LHC energy 5.5
TeV from Glauber-Gribov theory of nuclear shadowing. We have also made predictions for baseline
cold-matter nuclear effects in lead-lead collisions at the same energy.
1.5.1. Introduction The system formed in nucleus-nucles (AA) collisions at LHC will
provide further insight into the dynamics of the deconfined state of nuclear matter. There
are also interesting effects anticipated for the initial state of the incoming nuclei related to
shadowing of nuclear parton distributions and the space-time picture of the interaction. These
should be studied in the more “clean” environment of a proton-nucleus collision. The initial-
state effects constitute a baseline for calculation of the density of particles at all rapidities
and affect therefore also high-p⊥particle suppression and jet quenching, as well as the total
multiplicity.
Both soft and relatively high-p⊥,p⊥<10 GeV/c, particle production in pA at LHC
energies probe the low-xgluon distribution of the target nucleus at moderate scales, Q2∼p2
⊥,

Heavy Ion Collisions at the LHC - Last Call for Predictions 17
Figure 6: RCP versus pt, Eq. (9), in dAu collisions at RHIC compared to experimental
data (upper and middle plots), and in pPb collisions at the LHC (lower plot), for different
pseudorapidities and centrality classes: central to peripheral (lower curves) and semicentral
to peripheral (upper curves). The bands reflect the uncertainty in the definition of the centrality
class.
and is therefore mainly influenced by nuclear shadowing. In the Glauber-Gribov theory [21],
shadowing at low-xis related to diffractive structure functions of the nucleon, which are
studied at HERA. The space-time picture of the interaction is altered from a longitudinally
ordered rescattering at low energies, to a coherent interaction of the constituents of the
incoming wave-functions at high energy. Shadowing affects both soft and hard processes.
Calculation of gluon shadowing was performed in our recent paper [22], where Gribov
approach for the calculation of nuclear structure functions was used. The Schwimmer model
was used to account for higher-order rescatterings. The gluon diffractive distributions are
taken from the most recent experimental parameterizations [23].
1.5.2. Particle production at LHC Shadowing will lead to a suppression both at mid- and
forward rapidities in p+Pb collisions at √s=5.5 TeV as seen in Fig. 7. We have plotted the
curves for two distinct kinematical scenarios of particle production; one-jet kinematics which
may be well motivated for particle production at p⊥<2 GeV/c and two-jet kinematics that
apply for high-p⊥particle producion. The uncertainty in the curves is due to uncertainty in
the parameterization of gluon diffractive distribution functions. Cronin effect is not included
in the curves of Fig. 7. We estimate it to be a 10% effect at these energies.

Heavy Ion Collisions at the LHC - Last Call for Predictions 18
[GeV/c]
T
p
0 2 4 6 8 10 12 14 16 18 20
pA
R
0.2
0.4
0.6
0.8
1
One-jet kinematics y = 0
y = 2
y = 4
y = 6
Two-jet kinematics
y = 0
y = 2
y = 4
y = 6
Figure 7: Shadowing as a function of transverse for p+Pb collisions at √s=5.5 TeV.
In Fig. 8 we present the suppresion due to cold-nuclear effects in Pb+Pb collisions at
√s=5.5 TeV as a function of centrality (top) and rapidity (bottom). Also here we present the
results for two kinematics.
1.6. Stopping Power from SPS to LHC energies.
V. Topor Pop, J. Barrette, C. Gale, S. Jeon and M. Gyulassy
We investigate the energy dependence of hadron production and of stopping power based on
HIJING/B¯
B v2.0 model calculations. Pseudorapidity spectra and pTdistributions for produced charged
particles as well as net baryons (per pair of partcipants) and their rapidity loss are compared to data at
RHIC and predictions for LHC energies are discussed.
In previous papers [24] we studied the possible role of topological baryon junctions
[25] [26], and the effects of strong color field (SCF) in nucleus-nucleus collisions at RHIC
energies. In the framework of HIJING/B¯
B v2.0 model, the new algorithm for junction anti-
junction J¯
J loops provide a possible explanation for baryon/meson anomaly. The SCF effects
as implemented within our model gives a better description of this anomaly. At LHC energies,
due to higher initial energy density (or temperature) we expect an increase of the mean value
of the string tension (κ) [27].
The day 1 measurements at the LHC will include results on multiplicity distributions with
important consequences for our understanding of matter produced in the collisions [28], [29].
From our model calculations one expects dNch
PbPb/dη≈3500 at η=0 in central (0-5 %) Pb
+Pb collision. This correspond to ≈17.5 produced charged hadrons per participant pair.
These values are higher than those obtained by requiring that both limiting fragmentation and

Heavy Ion Collisions at the LHC - Last Call for Predictions 19
b [fm]
0 2 4 6 8 10 12 14 16 18 20
R (Pb+Pb)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9 = 1.5 GeV/c
T
p
= 5500 GeVs
y* = 0
GG one-jet
GG two-jet
b [fm]
2 4 6 8 10 12 14 16 18 20
y* = 3
Y
-3 -2 -1 0 1 2 3 4 5 6
Pb+Pb
R
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9 GG one-jet
GG two-jet = 1.5 GeV/c
T
p
Figure 8: Shadowing as a function of centrality (top) and rapidity (bottom) for Pb+Pb
collisions at √s=5.5 TeV.
the trapezoidal shape of the pseudo-rapidity distribution persist at the LHC [29]. Our model
predicts a characteristic violations of the apparently universal trend, seen up to maximum
RHIC energy. In contrast saturation models [5] offer a justification for the predicted very
weak √sNN dependence of event multiplicity.
Figure 9 presents predictions for pTspectra at midrapidity and NMF Rch
PbPb of total
inclusive charged hadrons for central (0-5%) Pb+Pb and p+pcollisions at √sNN =5.5 TeV.
The predicted NMF Rπ0
PbPb of neutral pions is also presented. From our model calculations
we conclude that baryon/meson anomaly, will persist at the LHC with a slight increase for
increasing strength of the chromoelectric field (κ=ee f f E). A somewhat higher sensitivity to
κis obtained for NMF of identified particles [27].
The net-baryon rapidity distribution measured at RHIC is both qualitatively and
quantitatively very different from those at lower energies indicating that a significantly
different system is formed near mid rapidity [30]. Fig. 10 (left panel) presents the energy
dependence of net-baryon at mid-rapidity per participant pair. Shown are the results for central
(0-5%) Au+Au collisions, which indicate a net decrease with increasing energy. This picture,
corroborated with an increase of the ratio ¯p/pto ≈1 suggests that the reaction at the LHC
is more transparent in contrast to the situation at lower energy. For central (0-5%) Pb+Pb

Heavy Ion Collisions at the LHC - Last Call for Predictions 20
Figure 9: Left: HIJING/B¯
B v2.0 predictions for pTspectra at mid-rapidity of total inclusive
charged hadrons for central (0-5%) Pb+Pb and p+pcollisions. Right: Predicted nuclear
modification factors for charged hadrons and for neutral pions.
collisions at √sNN =5.5 TeV, our prediction for net-baryon per participant pair is ≈0.065
with Npart =398, assuming κ=5 GeV/fm. Similar values (open squares) are obtained within
pQCD+hydro model [31]. However,this model predicts (Fig. 15 from ref. [31]) muchsteeper
slopes of charged hadron pTspectra.
Figure 10: Left: HIJING/B¯
B v2.0 predictions for net-baryon (per participant pair) at mid-
rapidity as function of √sNN. Right: Average rapidity loss versus beam rapidity. The data
and dashed line extrapolation are from ref. [30] and from BRAHMS [32].
In our model the main mechanisms for baryon production are quark di-quark (q−qq)
strings fragmentation and J¯
J loops in which baryons are produced approximatively in pairs.
The energy dependence is ∝(s/s0)−1/4+∆/2similar with those predicted in ref. [25] (eq. 11)
with the assumption that J¯
J is a dominant mechanisms. This dependence is obtained if we
choose for the parameters: s0=1 GeV2the usual parameter of Regge theory, α(0) =1/2
the reggeon (MJ
0) intercept and αP(0) =1+ ∆ (where ∆≈0.01) for the pomeron intercept. If
confirmed, the measurements at LHC energies will help us to determine better these values.

Heavy Ion Collisions at the LHC - Last Call for Predictions 21
In contrast, results from HIJING/B¯
B v1.10 model [26] (star symbol) give a slow energy
dependence with a higher pomeron intercept αP(0) =1+0.08 and over-estimate the stopping
in the entire energy region.
Baryon conservation in the reactions can be used to predict rapidity loss and the energy
loss per baryon. The results are illustrated in Fig. 10 (right panel) for average rapidity loss
< δy>defined as in ref. [24]. The predicted values for RHIC and LHC energies, clearly depart
from the linear extrapolation for constant relative rapidity loss [30], which seems to be valid
only at lower energies (√sNN ≤20 GeV).
1.7. Investigating the extended geometric scaling region at LHC with polarized and
unpolarized final states
D. Boer, A. Utermann and E. Wessels
We present predictions for charged hadron production and Λpolarization in p-pand p-Pb
collisions at the LHC using the saturation inspired DHJ model for the dipole cross section in the
extended geometric scaling region.
At high energy, scattering of a particle offa nucleus can be described in terms of a
colour dipole scattering offsmall-xpartons, predominantly gluons, in the nucleus. At very
high energy (small x), the dipole amplitude starts to evolve nonlinearly with x, leading to
saturation of the density of these small-xgluons. The scale associated with this nonlinearity,
the saturation scale Qs(x), grows exponentially with log(1/x).
The nonlinear evolution of the dipole amplitude is expected to be characterized by
geometric scaling, which means that the dipole amplitude depends only on the combination
r2
tQ2
s(x), instead of on r2
tand xindependently. Moreover, the scaling behaviour is expected to
hold approximately in the so-called extended geometric scaling (EGS) region between Q2
s(x)
and Q2
gs(x)∼Q4
s(x)/Λ2.
The small-xDIS data from HERA, which show geometric scaling, were successfully
described by the GBW model [33]. To describe the RHIC data on hadron production in d-Au
in the EGS region a modification of the GBW model was proposed by Dumitru, Hayashigaki
and Jalilian-Marian (DHJ), incorporating scaling violations in terms of a function γ+[34].
This DHJ model also describes p-pdata at forward rapidities [35].
1.7.1. DHJ model prediction for charged hadron production Using the DHJ model we can
make a prediction for the pt-spectrum of charged hadron production in both p-Pb and p-p
collisions at the LHC, at respectively √s=8.8 TeV and √s=14 TeV. Figure 11a shows the
minimum bias invariant yield for an observed hadron rapidity of yh=2 in the centre of mass
frame, which for 1 GeV .pt.10 GeV predominantly probes the EGS region. We note that
at this rapidity the result is not sensitive to details of the DHJ model in the saturation region
r2
t>1/Q2
s. Further, from [34] we expect that pt-independent K-factors are needed to fix the
+We note that at central rapidities we cannot reproduce exactly the results of [34] for large pt. Therefore, a
modification of the model may be needed to describe all RHIC data.

Heavy Ion Collisions at the LHC - Last Call for Predictions 22
normalization. We conclude that the LHC data on hadron production in both p-Pb and p-p
collisions will provide valuable data to further study the dipole scattering amplitude near the
onset of saturation, particularly the behaviour of the function γ, which is discussed in e.g. [36].
1.7.2. DHJ model prediction for Λpolarization Another interesting small-xobservable is
the polarization of Λhyperons produced in p-Acollisions, PΛ. This polarization, oriented
transversely to the production plane, was shown to essentially probe the derivative of the
dipole scattering amplitude, hence displaying a peak around Qswhen described in the
McLerran-Venugopalan model [37]. If this feature persists when x-evolution of the dipole
scattering amplitude is taken into account, PΛwould be a valuable probe of saturation effects.
Using the DHJ model for the x-evolution of the scattering amplitude, we find that PΛdisplays
similar behaviour as in the MV model. This is depicted for fixed Λrapidities of 2 and 4
in figure 11b. The position of the peak scales with the average value of the saturation scale
hQs(x)i. In the plotted region, the peak is located roughly at hQs(x)i/2.
The figure also shows that, like in the MV model, in the DHJ model |PΛ|scales
approximately linearly with xF, which means that at the LHC it is very small due to √s
being very large: rapidities around 6 are required for PΛto be on the 1% level, although there
is a considerable model uncertainty in the normalization.
We conclude that the polarization of Λparticles in p-Pb collisions is an interesting probe
of hQs(x)i, but is probably of measurable size only at very forward rapidities.
1 2 3 4 5 6 78 9 10
1×10-6
1×10-5
1×10-4
1×10-3
1×10-2
1×10-1
1×100
p-p, 14 TeV, yh= 2
p-Pb, 8.8 TeV, yh= 2
d3N
dyd2pT[(GeV/c)−2]
pT[GeV/c]
11.5 22.5
-0.005
-0.004
-0.003
-0.002
-0.001
0
p-Pb, 8.8 TeV, yh= 2
p-Pb, 8.8 TeV, yh= 4
PΛ
pT[GeV/c]
Figure 11: a. Charged hadron production. b. Λpolarization. In both plots, Aeff=20, and
parton distributions and fragmentation functions of [34] and [37] were used.
1.8. Inclusive distributions at the LHC as predicted from the DPMJET-III model with chain
fusion
F. Bopp, R. Engel, J. Ranft and S. Roesler
DPMJET-III with chain fusion is used to calculate inclusive distributions of Pb-Pb collisions at
LHC energies. We present rapidity distributions as well as scaled multiplicities at mid-rapidity as
function of the collision energy and the number of participants.

Heavy Ion Collisions at the LHC - Last Call for Predictions 23
1e-04
0.001
0.01
0.1
0 5000 10000 15000 20000 25000 30000
A(N) arbitrary units
N
pbpbmbmul
Pb-Pb 5500 GeV %
Pb-Pb 5500 GeV |etacm| < 2.5 %
Pb-Pb 5500 GeV 0-10 %
Pb-Pb 5500 GeV 0-10 % |etacm| < 2.5 %
Pb-Pb 2000 GeV %
Pb-Pb 2000 GeV |etacm| < 2.5 %
Pb-Pb 700 GeV %
Pb-Pb 700 GeV |etacm| < 2.5 %
Figure 12: Multiplicity distributions in minimum bias and 0-10% central collisions in Pb-Pb
collisions in the full ηcm range and for |ηcm| ≤ 2.5 (from DPMJET-III).
Monte Carlo codes based on the two–component Dual Parton Model (soft hadronic
chains and hard hadronic collisions) are available since 10–20 years: The present codes are
PHOJET for hh and γhcollisions [38,39] and DPMJET-III based on PHOJET for hA and AA
collisions [40]. To apply DPMJET–III to central collisions of heavy nuclei, the percolation
and fusion of the hadronic chains had to be implemented [41].
In figures 12 and 13 we apply this model to minimum bias and central collisions of heavy
nuclei at the LHC and at RHIC. We find an excellent agreement to RHIC data on inclusive
distributions.
The behaviour of the inclusive hadron production becomes particular simple if we plot
it in the form dN
dηcm /Npart
2.Npart is the number of participants in the AA collisions. In figure 14
we plot this quantity as function of Npart and as function of Ecm, in both plots we find a rather
simple behaviour.
The limiting fragmentation hypothesis was proposed in 1969 by Benecke et al [42].
If we apply it to nuclear collisions we have to plot dN
dηcm /Npart
2as function of ηcm −ybeam. In
figure 15 we plot central and less central Au-Au collisions at RHIC and LHC energies in this
form. We find that DPMJET-III shows in the fragmentation region only small deviations from
limiting fragmentation.
1.9. Some “predictions” for PbPb and pp at LHC, based on the extrapolation of data at
lower energies
W. Busza
The global characteristics of multiparticle production in pp, pA, AA and even e+e−

Heavy Ion Collisions at the LHC - Last Call for Predictions 24
Figure 13: (left) Central RHIC and LHC collisions. (right) LHC Pb-Pb collisions from
DPMJET-III.
Figure 14: dN
dηcm /Npart
2(left) over Npart (right) over Ecm, Pb-Pb and Au-Au collisions.

Heavy Ion Collisions at the LHC - Last Call for Predictions 25
Figure 15: dN
dηcm /Npart
2in Au-Au collisions over ηcm −ybeam (left) central, (right) less central.
collisions, over the entire energy range studied to date, show remarkably similar trends.
Furthermore it is a fair characterization of the data to say that the data appears simpler than
current explanations of it. These trends allow us to “predict”, with high precision, several
important results which will be seen in pp and PbPb collisions at LHC.
Such predictions are valuable from a practical point of view. More important, if they turn
out to be correct, and the trends seen to date are not some accidental consequence of averaging
over many species and momenta, the observed trends must be telling us something profound
about how QCD (most likely how the vacuum) determines particle production. At a minimum,
if the current belief is correct that the intermediate state between the instant of collision and
final free-streaming of the produced particles is very different in e+e−and AA colisions, or
for that matter, in pp collisions, AA collisions below SPS energies and AA collisions at the
top RHIC energy, the global characteristics of multiparticle productions must be insensitive
to the intermediate state. One consequence is that no successful prediction of any selected set
of global properties can be used as evidence that a particular model correctly describes the
intermediate state.
On the other hand, if these “predictions" turn out to be false, it will be a strong indicator
of the onset of some new physics at LHC.
So what are these universal simple trends?
We find [43, 44], as a first approximation, that
(i) The global distributions of charged particles factorize into an energy dependent part and
a geometry, or incident system, dependent part.
(ii) At a given energy, in pA and AA the distributions do depend in detail on the colliding
systems or geometry (eg. impact parameter). However the total number of produced
particles is simply proportional to the total number of participants Npart (or wounded
nucleons, in the language of Bialas and Czyz). [Note: there is a systematic difference in
the constant of proportionality in pA and AA, that can be attributedto the leading particle

Heavy Ion Collisions at the LHC - Last Call for Predictions 26
effect in pp collisions]
(iii) The total charged particle density dN/dη(where ηis the pseudorapidity), and the directed
and elliptic flow parameters ν1and ν2satisfy extended longitudinal scaling. Furthermore,
over most of its range the “limiting curve" is linear.
(iv) The mid-rapidity (in the cm system) particle density dN
dη|y=0and the elliptic flow
parameter ν2both increase linearly with ln √s. It is not clear if this is the origin or
consequence of item (iii) above. [Note: for elliptic flow, (iii) and (iv) are directly related
only if we postulate that at all energies there is a “pedestal" in the value of ν2, i.e. there
is a part of the source of flow that is independent of energy]
In the figures 16, 17 and 18 we use the above observed trends at lower energies to “predict"
LHC results. A more detailed version of this work will be submitted to Acta Physica Polonica.
1.10. Multiplicities and J/ψ suppression at LHC energies
A. Capella and E. G. Ferreiro
We present our predictions on multiplicities and J/ψ suppression at LHC energies. Our results
take into account shadowing effects in the initial state and final state interactions with the hot medium.
We obtain 1800 charged particles at LHC and the J/ψ suppression increases by a factor 5 to 6 compared
to RHIC.
1.10.1. Multiplicities with shadowing corrections At high energy, different mechanisms
in the initial state -shadowing-, that lower the total multiplicity, have to be taken into
account. The shadowing makes the nuclear structure functions in nuclei different from the
superposition of those of their constituents nucleons. Its effect increases with decreasing x
and decreases with increasing Q2. We have included a dynamical, non linear of shadowing
[49], controlled by triple pomeron diagrams. It is determined in terms of the diffractive
cross sections. Our results for charged particles multiplicities at RHIC and LHC energies
are presented in Fig. 19. In absence of shadowing we obtain a maximal multiplicity,
dNAA/dy =A4/3. With shadowing corrections the multiplicity behaves as dNAA/dy =Aα,
with α=1.13 at RHIC and α=1.1 at LHC.
1.10.2. J/ψ suppression An anomalous J/ψ suppression -that clearly exceeds the one
expected from nuclear absorption- has been found in PbPb collisions at SPS. Such a
phenomenon was predicted by Matsui and Satz as a consequence of deconfinement in a dense
medium. It can also be described as a result of final state interaction of the ccpair with
the dense medium produced in the collision: comovers interaction [50]. Here we present

Heavy Ion Collisions at the LHC - Last Call for Predictions 27
NN
s
1 10 2
10 3
10 4
10
〉/2
part
N〈/η/d
ch
dN
0
2
4
6
8AGS: Au+Au
SPS: Pb+Pb
PHOBOS: Au+Au
PHOBOS: Cu+Cu
ln(s) ×-0.5+0.39
fit PHOBOS data
LHC
Figure 1 Extrapolation of midrapidity particle density
per participant pair, for central PbPb collisions at
√sNN =5.5TeV. The data are from a PHOBOS
compilation [4, 45]. The predicted value for
dNch
dη/(Npart/2) =6.2±0.4, which for Npart =386 (top
3 %) corresponds to dNch/dη=1200 ±100.
η
-5 0 5
ηd
dN
0
500
1000
200 GeV
130 GeV
19.6 GeV
=5.5TeV Extrapolated from Lower Energy Datas for PbPb @
ηd
dN
Figure 2 Extrapolated central (Npart =360, 0-10%
centrality) PbPb pseudorapidity distribution at √sNN
=5.5TeV. PHOBOS AuAu data [4, 45], longitudinal
scaling with a linear “limiting curve", and the
observed ln √sNN energy dependence of the mid
rapidity density were used in the extrapolation.
η
-8 -6 -4 -2 0 2 4 6 8
ηd
dN
0
200
400
600
800
1000
AuAu 200 GeV
AuAu 130 GeV
AuAu 19.6 GeV
CuCu 200 GeV
CuCu 62 GeV
CuCu 22 GeV
=360
part
N
=250
part
N
=100
part
N
=60
part
N
=5.5TeV Extrapolated from Lower Energiess for PbPb @
ηd
dN
Figure 3 Predictions of PbPb LHC pseudorapidity
distributions for different centrality collisions. For
each extrapolated curve, PHOBOS AuAu or CuCu
data [4, 45], with Npart closest to the required value,
were taken and normalized to the quoted Npart value.
The vertical and horizontal scales were then re-scaled
by ln√sNN. This is equivalent to trends (iii) and (iv),
provided that dn/dη→0 as η→beamrapidity. It
must be emphasized that each curve is an independent
extrapolation, however each extrapolation relies on the
same method.
s
2
ln
0 20 40 60 80 100
/2>)
part
> (/<N
ch
<N
0
20
40
60
80
100 AA Data
p) Data (Inelastic)ppp(
p) Data (NSD)ppp(
Data
-
e
+
e
LHC Projection
Figure 4 Total charged particle multiplicity per
participant pair plotted as a function of ln2√s(with
√sin GeV) for various colliding systems. The
data are taken from the compilation in Busza [43].
The extrapolation of the AA data to √sNN=5.5 TeV
predicts 15000 ±1000 charges particles at LHC for
Npart =386 (top 3%). Extrapolation of the non-
single diffractive (NSD) pp data to √sNN =14
TeV predicts 70 ±8 charged particles at LHC. The
ln2√sextrapolation for the total multiplicity is a
consequence of the extrapolation procedure described
in the caption of fig 3.
Figure 16:
our results for the ratio of the J/ψ yield over the average number of binary nucleon-nucleon

Heavy Ion Collisions at the LHC - Last Call for Predictions 28
η
-5 0 5
/2
part
Nηd
dN
0
2
4
60-6% Centrality
35-45% Centrality
PbPb @ 5.5TeV Scaled From AuAu @ 200GeV (Phobos Data)
Figure 1 Predicted PbPb pseudorapidity distributions
of charged particles at √sNN=5.5 TeV for two cen-
tralities, 0-6% and 35-45%. These predicted curves
are obtained by extrapolating the corresponding cen-
trality 200GeV PHOBOS AuAu data [4, 45] using the
same procedure as in fig. 16.3.
part
N
0 100 200 300
/2
part
N|<1η|
ηdN/d
4
5
6
7
Non-Single Diffractivepp+
Inclusivep
p+
@ 5.5TeV Scaled from Lower Energy Datap
p+
PbPb @ 5.5TeV Scaled from AuAu @ 200 GeV (Phobos Data)
part
N
0 100 200 300
/2
part
N|<1η|
ηdN/d
4
5
6
7
Non-Single Diffractivepp+
Inclusivep
p+
@ 5.5TeV Scaled from Lower Energy Datap
p+
PbPb @ 5.5TeV Scaled from AuAu @ 200 GeV (Phobos Data)
part
N
0 100 200 300
/2
part
N|<1η|
ηdN/d
4
5
6
7
Non-Single Diffractivepp+
Inclusivep
p+
@ 5.5TeV Scaled from Lower Energy Datap
p+
PbPb @ 5.5TeV Scaled from AuAu @ 200 GeV (Phobos Data)
part
N
0 100 200 300
/2
part
N|<1η|
ηdN/d
4
5
6
7
Non-Single Diffractivepp+
Inclusivep
p+
@ 5.5TeV Scaled from Lower Energy Datap
p+
PbPb @ 5.5TeV Scaled from AuAu @ 200 GeV (Phobos Data)
part
N
0 100 200 300
/2
part
N|<1η|
ηdN/d
4
5
6
7
Non-Single Diffractivepp+
Inclusivep
p+
@ 5.5TeV Scaled from Lower Energy Datap
p+
PbPb @ 5.5TeV Scaled from AuAu @ 200 GeV (Phobos Data)
Figure 2 Predicted Npart dependence of the mid
rapidity charged particle density per participant pair
for PbPb at √sNN =5.5 TeV. These results are
obtained from the PHOBOS 200 GeV AuAu data
[4, 45] scaled by ln √sNN (see trends (i) and (iv) in
the text).
η
-10 0 10 20
ηd
dN
0
5
10 s
38.8 GeV
27.4 GeV
23.8 GeV
19.4 GeV
11.3 GeV
p-Emulsion Scaled to 5.5 TeV
Figure 3 Predicted pA pseudorapidity distribution
for √sNN =5.5 TeV for Npart =3.4. Each curve
is an independent extrapolation of lower energy p-
emulsion data using the same procedure as in Fig.
16.3. The p-emulsion data are from ref [46], and
covers the energy range √sNN =11.3 GeV to 38.8
GeV (ie. proton beams with momentum 67 GeV/c to
800 GeV/c)
-5 0 5
2
4
6
sData and p
Scaled from p
1800 GeV
900 GeV
630 GeV
630 GeV
546 GeV
200 GeV
Non-Single-Diffractive Scaled to 14000 GeVpp+
Figure 4 Predicted pp Non-Single-Diffractive (NSD)
pseudorapidity distributions at √sNN =14 TeV. Each
curve is an independent extrapolation of lower energy
p¯pdata [47], using the same procedure as in Fig.
16.3. The scatter of points at central rapidities
most likely reflects systematic errors in these difficult
measurements.
Figure 17:
collisions at RHIC and LHC energies:
RJ/ψ
AB (b)=dNJ/ψ
AB (b)/dy
n(b)=dNJ/ψ
pp
dy Rd2sσAB(b)n(b,s)Sabs(b,s)Sco(b,s)
Rd2sσAB(b)n(b,s).(10)
Sabs refers to the survival probability due to nuclear absorption and Sco is the survival
probability due to the medium interactions. The data on dAu collisions at RHIC favorize
a small σabs =0 mb, so Sabs =1 [51]. The interaction of a particle or a parton with the
medium is described by the gain and loss differential equations which govern the final state

Heavy Ion Collisions at the LHC - Last Call for Predictions 29
η
-10 -5 0 5 10
2
v
0.02
0.04
0.06
19.6 GeV
62.4 GeV
130.0 GeV
200.0 GeV
PbPb Extrapolated to 5.5TeV
(40% Most Central)
Figure 1 Predicted pseudorapidity dependance of
the elliptic flow parameter ν2for the 40% most
central collisions of PbPb at √sNN =5.5 TeV. An
extrapolation procedure similar to that in fig. 16.2 was
used, with input data from PHOBOS [47].
s
1 10 2
10 3
10 4
10
2
V
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10 AGS (E877)
NA49 cumul
NA49 std/mod
CERES
Phobos
STAR
Phenix
LHC
Figure 2 Extrapolation of the elliptic flow parameter
ν2for the 40% most central collisions in PbPb at
√sNN =5.5 TeV. The data is a compilation in [48].
Figure 18:
Figure 19: Multiplicities of charged particles
with (solid lines) and without (dashed lines)
shadowing corrections at RHIC and LHC.
Figure 20: J/ψ production at RHIC and LHC.
Dashed: J/ψ shadowing, pointed: comovers
suppression, continuous: total suppression.
interactions:
τdρJ/ψ(b,s,y)
dτ=−σco ρJ/ψ(b,s,y)ρmedium(b,s,y),(11)
where ρJ/ψ and ρco =ρmedium are the densities of J/ψ and comovers. We neglect a gain term
resulting from the recombination of c-cinto J/ψ. Our equations have to be integrated between
initial time τ0and freeze-out time τf. We use the inverse proportionality between proper time
and densities, τf/τ0=ρ(b,s,y)/ρpp(y). Our densities can be either hadrons or partons, so σco
represents an effective cross-section averaged over the interaction time. We obtain the survival

Heavy Ion Collisions at the LHC - Last Call for Predictions 30
probability Sco(b,s) of the J/ψ due to the medium interaction:
Sco(b,s)≡NJ/ψ(final)(b,s,y)
NJ/ψ(initial)(b,s,y)=exp"−σco ρco(b,s,y)ℓn ρco(b,s,y)
ρpp(0) !#.(12)
The shadowing produces a decrease of the medium density. Because of this, the shadowing
corrections on comovers increase the J/ψ survival probability Sco. On the other side, the
shadowing corrections on J/ψ decrease the J/ψ yield. Our results for RHIC and LHC are
presented in Fig. 20. We use the same value of the comovers cross-section, σco =0.65 mb
that we have used at SPS energies. We neglect the nuclear absorption. The shadowing is
introduced in both the comovers and the J/ψ yields.
1.11. Heavy ion collisions at LHC in a Multiphase Transport Model
L.-W. Chen, C. M. Ko, B.-A. Li, Z.-W. Lin and B.-W. Zhang
The AMPT model [52] is a hybrid model that uses the HIJING model [53] to generate the
initial conditions, the ZPC [54] for modeling the partonic scatterings, and the ART model [55]
for treating hadronic scatterings. In the default version [56], the initial conditions are strings
and minijets from the HIJING model and particle production is based on the Lund string
fragmentation, while in the string melting version [57], the initial conditions are valence
quarks and antiquarks from hadrons produced in the HIJING model and hadronization is
described by a coordinate-space coalescence model. Using the AMPT model, we predict in
the following the hadron rapidity and transverse momentum distributions, the elliptic flows
of both light and heavy hadrons, the two-pion and two-kaon correlation functions in Pb+Pb
collisions at center-of-mass energy of √sNN =5.5 TeV at LHC [58].
Shown in the left window of Fig. 21 are the charged hadron pseudorapidity distribution
and the rapidity distributions of identified hadrons obtained with (lines with circles) and
without (solid lines) nuclear shadowing of nucleon parton distribution functions. Compared
to results from the AMPT model for RHIC [59], the distributions at LHC are significantly
wider and higher. For mid-pseudorapidity charged hadrons, the distribution shows a clear
plateau structure with a value of about 4500 and 2500, respectively, without and with nuclear
shadowing. The latter is more than a factor of three higher than that at RHIC. The transverse
momentum spectra of identified midrapidity hadrons are shown in the right window of Fig. 21
by lines with circles. The inverse slope parameters, particularly for kaons and protons with
transverse momenta below 0.5 GeV/cand 1 GeV/c, respectively, are larger than those at RHIC
(solid lines) as a result of stronger final-state rescatterings at LHC than at RHIC. Similar to
that observed at RHIC, the proton spectrum is below that of pions at low transversemomenta,
but they become comparable at about 2 GeV/c.
Hadron elliptic flows based on a parton scattering cross section of 10 mb, which is needed
to describe observed hadron elliptic flows at RHIC [57], are shown in Fig. 22. The left window
gives the elliptic flows of light and heavy quarks as functions of their transverse momenta, and
they display the expected mass ordering at low transverse momenta, i.e., the elliptic flow is
smaller for quarks with larger masses. At larger transverse momenta, the elliptic flows of

Heavy Ion Collisions at the LHC - Last Call for Predictions 31
−6 −4 −2 0 2 4 6
η
0
2000
4000
6000
dNch/dη
no shadowing
AMPT
−6 −4 −2 0 2 4 6
y
0
50
100
150
dN/dy
p
p
−6 −4 −2 0 2 4 6
y
0
1000
2000
dN/dy
π+
π−
−6 −4 −2 0 2 4 6
y
0
200
400
dN/dy
K+
K−
−
0 0.5 1 1.5 2 2.5 3
pT (GeV/c)
10−3
10−2
10−1
100
101
102
103
104
d2N/(2πpTdpTdy) (−0.5<y<0.5) (GeV−2c2)
π+ at LHC
K+
p
at RHIC
Figure 21: Left window: Pseudorapidity distributions of charged hadrons and rapidity
distributions of identified hadrons in central (b≤3 fm) Pb+Pb collisions at √sNN =5.5
TeV from the default AMPT model with (lines with circles) and without (solid lines) nuclear
shadowing. Right window: Transverse momentum spectra of identified midrapidity hadrons
from same collisions as well as central Au+Au collisions at √sNN =200 GeV.
Figure 22: Elliptic flows of quarks (left window), light hadrons (middle window), and heavy
mesons (right window) in Pb+Pb collisions at √sNN =5.5 TeV and b=8 fm from the AMPT
model with string melting and a parton scattering cross section of 10 mb.
heavy quarks become, however, larger than those of light and strange quarks, which peak
at around 1-1.5 GeV/c. The elliptic flows of pions and protons at LHC are shown in the
middle window of Fig. 22. Compared to corresponding ones at RHIC for Au+Au collisions
at √sNN =200 GeV and same impact parameter shown in the figure, the elliptic flow of pions
at LHC is larger while that of protons is smaller. As at RHIC [60], elliptic flows of heavy
mesons are estimated from those of quarks using the quark coalescence or recombination
model [61, 62] and are shown in the right window of Fig. 22. While elliptic flows of heavy

Heavy Ion Collisions at the LHC - Last Call for Predictions 32
mesons are dominated by those of heavy quarks, particularly for bottomed mesons, those of
heavy mesons with hidden charm or bottom, i.e., quarkonia J/ψ and Υconsisting of a heavy
quark and its antiquark, at transverse momentum pTare simply twice those of their constituent
heavy quarks at pT/2.
0 20 40 60 80 100
1.0
1.1
1.2
1.3
1.4
1.0
1.1
1.2
1.3
1.4
AMPT v2.12 String Melting
10mb, b=0 fm
RHIC 200AGeV Au+Au
LHC 5500AGeV Pb+Pb
1.0
1.1
1.2
1.3
1.4
π+π+ & π−π−
−0.5<y<0.5, 0.3<pt1,2<1.5GeV/c
0 20 40 60 80
1.0
1.1
1.2
1.3
1.4
Q (MeV/c)
C2
Qinv Qout
Qside Qlong
cut=35MeV/c for 1−D projections
0 40 80 120
1.0
1.1
1.2
1.3
1.4
1.0
1.1
1.2
1.3
1.4
AMPT v2.12 String Melting
10mb, b=0 fm
RHIC Au+Au
LHC Pb+Pb
1.0
1.1
1.2
1.3
1.4
K+K+ & K−K−
−0.5<y<0.5, 0.3<pt1,2<1.5GeV/c
0 40 80
1.0
1.1
1.2
1.3
1.4
Q (MeV/c)
C2
Qinv Qout
Qside Qlong
cut=40MeV/c for 1−D projections
Figure 23: Correlation functions for midrapidity charged pions (left windows) and kaons
(right window) with 300 <pT<1500 MeV/cfrom the AMPT model with string melting and
a parton cross section of 10 mb for central (b=0 fm) Pb+Pb collisions at √sNN =5.5 TeV
(solid lines) and Au+Au collisions at √sNN =200 GeV (dashed lines).
From the positions and momenta of pions or kaons at freeze out, their correlation
functions in the longitudinally comoving frame can be calculated using the program
Correlation After Burner [63] to take into account their final-state strong and Coulomb
interactions. Shown in the left and right windows of Fig. 23 are, respectively, one-dimensional
projections of the correlation functions of midrapidity (−0.5<y<0.5) charged pions and
kaons with transverse momentum 300 <pT<1500 MeV/cand their comparison with
corresponding ones for central Au+Au collisions at √sNN =200 GeV at RHIC, which have
been shown to reproduce reasonably measured ones for pions [64]. The correlation functions
at LHC are seen to be narrower than at RHIC.
Table 1: Radii from Gaussian fit to correlation functions.
Rout(fm) Rside(fm) Rlong(fm) λRout/Rside
RHIC (π) 3.60 3.52 3.23 0.50 1.02
LHC (π) 4.23 4.70 4.86 0.43 0.90
RHIC (K) 2.95 2.79 2.62 0.94 1.06
LHC (K) 3.56 3.20 3.16 0.89 1.11
Fitting the correlation functions by the Gaussian function C2(Q,K)=1+
λexp(−PiR2
ii(K)Q2
i), where Kis the average momentum of two mesons. Extracted radii of

Heavy Ion Collisions at the LHC - Last Call for Predictions 33
the emission source are shown in Table I. Predicted source radii at LHC are larger than those
at RHIC, consistent with the narrower correlation functions at LHC than at RHIC. In both
collisions, radii of the emission source for pions are larger than those for kaons. The smaller
lambda parameter for pions than for kaons is due to the large halo in the pion emission source
from decays of omega mesons. Also, the emission source is non-Gaussian and shifted in the
direction of pion or kaon transverse momentum.
1.12. Multiplicity distributions and percolation of strings
J. Dias de Deus and C. Pajares
In the framework of percolations of strings the rapidity distributions for central AA collisions are
shown for SPS, RHIC and LHC energies. The obtained value for LHC is lower than the one predicted
for the rest of models but larger than the linear energy extrapolation from SPS and RHIC.
Multiparticle production is currently described in terms of color strings stretched
between the partons of the projectile and the target these color strings may be viewed as
small areas in the transverse space πr2
0,r0≃0.2−0.3 fm, filled with color field created
by the colliding partons. Particles are produced via emision of q¯qpairs in this field. With
growing energy and/or atomic number of colliding nuclei, the number of strings grows, and
they start to overlap, forming clusters, very much similar to disks in the two dimensional
percolation theory. At a certain critical density a macroscopical cluster appears that marks the
percolation phase transition [65]. A cluster behaves as a single string with a higher color field
~
Qncorresponding to the vectorial sum of the color changes of each individual ~
Q1string. The
resulting color field covers the area Snof the cluster. As ~
Qn=P~
Q1, and the individual string
colors may be oriented in an arbitrary manner respective to one another, the average ~
Q1i·~
Q1j
is zero and ~
Q2
n=n~
Q2
1.
In this way, the multiplicity µnand the average p2
Tof particles <p2
T>nproduced by a
cluster of nstrings, are given by
µn=rnSn
S1µ1;hp2
Tin=rnS1
Snhp2
Ti1(13)
where µ1and hp2
Ti1are the mean multiplicity and the mean transverse momentumof particles
produced by a simple string with a transverse area S1=πr2
0.
Equation (13) is the main tool of our calculations. In order to compute the multiplicities
we generate strings according to the quark gluon string model and using a Monte Carlo code.
Each string is produced at an identified impact parameter. From this, knowing the transverse
area of each string, we identify all the clusters formed is each collision and subsequently
compute for each of them the rapidity multiplicity spectrum.
In figure 24 is shown the results, (see reference [3] for details) for central Au-Au
collisions at different energies, including the curve for √sNN =5.5 TeV. The value at
midrapidity 8.5 is similar to other computations in the same framework (7.3 [66], 8.6 [67]).
This strong reduction of the multiplicities relative to simple multicollision models, due to the
interaction of strings, was anticipated 12 years ago [68]. Nowdays models have incorporated

Heavy Ion Collisions at the LHC - Last Call for Predictions 34
Figure 24: LHC prediction, together with RHIC data and results.
effects, like strong shadowing or triple pomeron couplings to suppress their original values.
However, our value is smaller than the one obtained by most of the other existing models.
Only, the extrapolation of the observed geometrical scaling in lN to AA given a close value:
9.5. The linear log of energy extrapolation of the SPS and RHIC values gives a lower value
of around 6.5.
At SPS and RHIC has been observed an aproximated limiting fragmentation scaling,
which is well reproduced in our approach. A clear breaking of this scaling is predicted at
LHC.
1.13. Shear Viscosity to Entropy within a Parton Cascade
A. El, C. Greiner and Z. Xu
The shear viscosity is calculated by means of the perturbative kinetic partonic cascade BAMPS
with CGC initial conditons for various saturation momentum scale Qs.η/s≈0.15 stays approximately
constant when going from RHIC to LHC.
The measured momentum anisotropy parameter v2at RHIC energy can be well
understood if the expanding quark-gluon matter is assumed to be described by ideal
hydrodynamics. This suggests that a strongly interacting and locally thermalized state of
matter has been created which behaves almost like a perfect fluid. Since the initial situation of
the quark-gluon system is far from thermal equilibrium, it is important to understand how and
which microscopic partonic interactions can thermalize the system within a short timescale
and can be responsible as well for its (nearly) ideal hydrodynamical behaviour. Furthermore
one would like to know the transport properties of the QGP, most prominently the shear
viscosity.

Heavy Ion Collisions at the LHC - Last Call for Predictions 35
0
500
1000
1500
2000
2500
3000
3500
0 0.5 1 1.5 2 2.5 3
dN/dη|cenral bin
t (fm/c)
dN/dη |CENTRAL
QS=2 GeV
QS=3 GeV
QS=4 GeV
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.5 1 1.5 2 2.5
T*t1/3 (GeV*fm1/3)
t (fm/c)
T*t1/3
QS=2 GeV
QS=3 GeV
QS=4 GeV 0
0.05
0.1
0.15
0.2
0.25
0.3
0 0.5 1 1.5 2 2.5 3 3.5
<pz2/E2>
t (fm/c)
Momentum Isotropy <pz2/E2>
QS=2 GeV
fit, trel=0.85
QS=3 GeV
fit, trel=0.52
QS=4 GeV
fit, trel=0.42
Figure 25: Time evolution of dN
dη(left), of the effective temperature (middle) and of the
momentum anisotropy (right).
A kinetic parton cascade (BAMPS) [69,70] has been developed with strictly perturbative
QCD inspired processes including for the first time inelastic (”Bremsstrahlung”) collisions
gg ↔ggg. The multiparticle back reation channel is treated fully consistently by respecting
detailed balance within the same algorithm. In [70] it is demonstrated that the inelastic
processes dominate the total transport collision rate and thus contribute much stronger
to momentum isotropization then elastic ones. Within a default setting of minijet initial
conditions, the overall build up of elliptic flow v2can be reasonably described [71] (a more
dedicated study is presently undertaken [72]).
One can thus expect to see thermalization of a QGP on a short time scale less than
1fm/cfor LHC relevant initial conditions as can be seen in the evolution in time of the
temperature and the momentum isotropy depicted in Fig. 25. We apply Bjorken expanding
geometry in one dimension. For the initial condition a simple Color Glass Condensate
(CGC) gluon distribution is assumed: The initial partons are described by the boost-invariant
form of the distribution function f(x,p)|z=0=c
αsNc1
τfδ(pz)Θ(Q2
s−p2
T) at a characteristic time
τ0=c/(αsNcQs).
Due to 3 →2 collisions the particle number first decreases (see Fig. 25) [73]. This is in
contrast to the idealistic ”Bottom-Up” scenario of thermalization, where an ongoing particle
production in the soft sector (pT< αsQS) is predicted with a strong increase in the total
particle number. The present calculation show that the particle number roughly stays constant.
For the above simple CGC parametrization Qs=2 GeV corresponds to RHIC energy whereas
Qs≈3−4 GeV is expected for LHC.
For all energies a nearly ideal hydrodynamical behavior is observed after 0.5fm/c
(middle Fig. 25). The thermalization time lies in the same range when looking at the
momentum isotropy. It is of crucial importance to extract out of these simulations the
transport properties of QCD matter to quantify the dissipative properties of the fluid. Using
standard dissipative hydrodynamics in expanding geometry shear viscosity and ratio η
scan
be calculated [73]: η=τ
4Txx +Tyy −2·Tzzand s=4n−n·ln(λ), where λdenotes the gluon
fugacity. As depicted in Fig. 26, the value η
s≈0.15 proves to be a universal number within
the BAMPS simulations, being nearly independent of QS. This is in line also with full 3-

Heavy Ion Collisions at the LHC - Last Call for Predictions 36
0
0.05
0.1
0.15
0.2
0 0.5 1 1.5 2 2.5 3
η/s
t (fm/c)
QS=2 GeV
QS=3 GeV
QS=4 GeV
Figure 26: Ratio of the shear viscosity to entropy density (αs=0.3).
dim calculations employing minijets and Glauber geometry for the initial condition [72].
η
sbasically only depends on the employed coupling strength αs(taken to be 0.3 as default
setting). Hence, within BAMPS, we do not expect any change in the shear viscosity ratio η/s
when going from RHIC to LHC.
1.14. Hadron multiplicities, pTspectra and net-baryon number in central Pb+Pb collisions
at the LHC
K. J. Eskola, H. Honkanen, H. Niemi, P. V. Ruuskanen and S. S. Räsänen
We summarize here our recent LHC predictions [31], obtained in the framework of
perturbative QCD (pQCD)+saturation+hydrodynamics (EKRT model for brief) [74]. This
model has successfully predicted [74,75] the charged particle multiplicities in central Au+Au
collisions at different √sNN , and it also describes the low-pTspectra of pions and kaons at
RHIC quite well [31,76].
Primary parton production in the EKRT model is computed from collinearly factorized
pQCD cross sections [77] by extending the calculation towards smaller pTuntil the abundant
gluon production vertices overlap and gluon fusions [78] saturate the number of produced
partons (gluons). The saturation scale is determined as p0=psat from a saturation
condition [74] NAA(p0,√s)·π/p2
0=c·πR2
A, where NAA(p0,√s) is the average number of
partons produced at |y| ≤ 0.5 and pT≥p0. With a constant c=1 the framework is closed.
For central Pb+Pb collisions at the LHC psat ≈2 GeV. We obtain the initial conditions for
the cylindrically symmetric boost invariant (2+1)-dimensional hydrodynamical description
by converting the computed transverse energy ET(psat) and net-baryon number NB(psat) into
densities ǫ(r,τ0) and nB(r,τ0) using binary collision profiles and formation time τ0=1/psat.
Assuming a fast thermalization at τ0, and zero initial transverse fluid velocity, we proceed
by solving the standard equations of ideal hydrodynamics including the current conservation
equation for net-baryon number. In the Equation of State we assume an ideal gas of gluons
and massless quarks (Nf=3), the QGP, with a bag constant Bat T>Tc, and a hadron
resonance gas of all states with m<2 GeV at T<Tc. Taking B1/4=239 MeV leads to
first-order transition with Tc=165 MeV. Final state hadron spectra are obtained with Cooper-
Frye procedure on a decoupling surface at Tdec followed by strong and electromagnetic

Heavy Ion Collisions at the LHC - Last Call for Predictions 37
2- and 3-body decays of unstable states using the known branching ratios. Extensive
comparison [31,76] with RHIC data suggests a single decoupling temperature Tdec =150 MeV
which is also used to calculate the predictions for the LHC. For details, see [31].
Our predictions [31] for the LHC multiplicities, transverse energies and net-baryon
number at y=η=0 for 5% most central Pb+Pb collisions at √sNN =5.5 TeV are summarized
in table 2. Note that the predicted charged particle multiplicity dNch/dηis 2570, i.e. only a
third of the initial ALICE design value (see also [75]). Whereas the multiplicity of initially
produced partons and observable hadrons are close to each other, the transverse energy is
reduced by a factor as large as 3.4 in the evolution from initial state to final hadrons. Due to
this reduction the very high initial temperature, T0&1 GeV, possibly observable through the
emission of photons, need not lead to contradiction between predicted and observed ET.
Table 2:
dN
dytot dN
dη
tot dN
dych dN
dη
ch dN
dyBdE
dyTdE
dη
TdN
dyπ±dN
dyπ0dN
dyK±dN
dypdN
dy¯pp/¯p
4730 4240 2850 2570 3.11 4070 3710 1120 1240 214 70.8 69.6 0.98
Our prediction for the charged hadron pTspectrum is the lower limit of the red band
(HYDRO, the width corresponding to Tdec =120...150 MeV) in the l.h.s. of figure 27 [31].
The corresponding pTdistributions of π+and K+are shown in the r.h.s. of the figure (solid
lines). The pQCD reference spectra, obtained by folding the LO pQCD cross sections with the
nuclear PDFs and fragmentation functions (KKP) and accounting for the NLO contributions
with a √sNN -dependent K-factor from [79], are also shown (pQCD) on the r.h.s. The yellow
bands (pQCD+E-loss) show the results with parton energy losses included as in [80]. We thus
predict the applicability region of hydrodynamics at the LHC to be pT.4...5 GeV, i.e. a
wider region than at RHIC.
0 1 2 3 4 5 6 7 8 9 10
pT[GeV]
10-4
10-3
10-2
10-1
100
101
102
103
1/2 pTdN/dpTd [1/GeV2]
5 % most central
s1/2 = 5500 GeV
LHC Pb+Pb
(h-+h+)/2
H YDR O
pQCD
pQCD wE-loss
0 1 2 3 4 5 6 7 8 9 10
pT[GeV]
10-4
10-3
10-2
10-1
100
101
102
103
1/2 pTdN/dpTd [1/GeV2]
5 % most central
s1/2 = 5500 GeV
LHC Pb+Pb
(h-+h+)/2
H YDR O
pQCD
pQCD wE-loss
Figure 27:

Heavy Ion Collisions at the LHC - Last Call for Predictions 38
1.15. Melting the Color Glass Condensate at the LHC
H. Fujii, F. Gelis, A. Stasto and R. Venugopalan
The charged particle multiplicity in central AA collisions and the production of heavy flavors in
pA collisions at the LHC is predicted in the CGC framework.
1.15.1. Introduction In the Color Glass Condensate (CGC) framework, fast (large x) partons
are described as frozen light cone color sources while the soft (small x) partons are described
as gauge fields. The distribution of the fast color sources and their evolution with rapidity is
described by the JIMWLK evolution equation; it is well approximated for large nuclei by the
Balitsky-Kovchegov (BK) equation. When two hadrons collide, a time dependent color field
is produced that eventually decays into gluons [81]. When the projectile is dilute (e.g.,AA
collisions at forward rapidity or pA collisions), k⊥factorization holds for gluon production,
thereby simplifying computations. For quark production, k⊥factorization breaks down and is
recovered only for large invariant masses and momenta.
1.15.2. Particle multiplicity in central AA collisions The k⊥factorized cross-sections are
convolutions over “dipole" scattering amplitudes in the projectile and target. Initial conditions
for the BK evolution of these are specified at an initial x=x0(chosen here to be x0≈10−2).
In this work [82], we consider two initial conditions, based respectively on the McLerran-
Venugopalan (MV) model or on the Golec-Biernat–Wusthoff(GBW) model.We adjust the
free parameters to reproduce the limiting fragmentation curves measured at RHIC from √s=
20 GeV to √s=200 GeV. The value of αsin the
fixed coupling
BK equation is tuned to
obtain the observed rate of growth of the saturation scale.The rapidity distribution dN/dy
is converted into the pseudo-rapidity distribution dN/dηby asuming the produced particles
have m∼200 MeV. A prediction for AA collisions at the LHC is obtained by changing √sto
0
1
2
3
4
5
6
7
8
-8 -6 -4 -2 0 2
η − ηbeam
dN/dη/(0.5 Npart)
Extrapolation to 5500 GeV
Figure 28: Number of charged particles per unit of pseudo-rapidity at the LHC energy.
5.5 GeV. From Fig. 28, we can infer dNch/dη|η=0=1000−1400; the two endpoints correspond
to GBW and MV initial conditions respectively.
1.15.3. Heavy quark production in pA collisions The cross-section for the production of a
pair of heavy quarks [83] is the simplest process for which k⊥-factorization breaks down [84]

Heavy Ion Collisions at the LHC - Last Call for Predictions 39
in pA collisions. This is due to the sensitivity of the cross-section to 3- and 4-point correlations
in the nucleus. Integrating out the antiquark and convoluting with a fragmentation function,
one obtains the cross-section for open heavy flavor production, e.g., Dmesons. Alternatively,
one can use the Color Evaporation Model to obtain the cross-section for quarkonia bound
states. The nuclear modification ratio is displayed in figure 29. The main difference at the
LHC compared to RHIC energy is that this ratio is smaller than unity already at mid rapidity,
and decreases further towards the proton fragmentation region.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 2 4 6 8 10
RpA
pT ( GeV/c )
y = 0, 2
y = 0, 2, 4
RHIC
LHC
0
0.2
0.4
0.6
0.8
1
1.2
-1 0 1 2 3 4
RpA
y
J/ψ
Charm
Figure 29: Left: nuclear modification factor for Dmesons as a function of p⊥. Right: the
same ratio as a function of rapidity, for Dmesons and for J/ψ.
1.16. RpA ratio: total shadowing due to running coupling
E. Iancu and D. N. Triantafyllopoulos
We predict that the RpAratio at the most forward rapidities to be measured at LHC should be
strongly suppressed, close to “total shadowing” (RpA≃A−1/3), as a consequence of running coupling
effects in the nonlinear QCD evolution.
We present predictions for the nuclear modification factor, or “RpAratio”, at forward
pseudorapidities (η > 0) and relatively large transverse momenta (p⊥) for the produced
particles, in the kinematical range to be accessible at LHC. These predictions are based on
a previous, systematic, study of the RpAratio within the Color Glass Condensate formalism
with running coupling [85]. The ratio can be approximated by
RpA≃1
A1/3
ΦA(Y,p⊥)
Φp(Y,p⊥),(14)
where Y=η+ln √s/p⊥and Φ(Y,p⊥) is the unintegrated gluon distribution of the
corresponding target hadron at fixed impact parameter. When the energy increases one expects
more and more momentum modes of this distribution to saturate to a value of order 1/αs, and
the corresponding saturation momentum reads
Q2
s(Y)= Λ2exp sB(Y−Y0)+ln2Q2
s(Y0)
Λ2,(15)
with Λ = 0.2 GeV, B=2.25 and Y0=4. The initial condition for the nucleus and the
proton are taken as Q2
s(A,Y0)=1.5 GeV2and Q2
s(p,Y0)=0.25 GeV2respectively, so that

Heavy Ion Collisions at the LHC - Last Call for Predictions 40
Q2
s(A,Y0)=A1/3Q2
s(p,Y0) for A =208. The functional form of this expression is motivated
by the solution to the nonlinear QCD evolution equations with running coupling [86, 87],
while the actual values of the numbers Band Y0have been chosen in such a way to agree with
the HERA/RHIC phenomenology. As shown in Fig. 30, with increasing Ythe two saturation
momenta approach to each other and clearly for sufficiently large Y, a nucleus will not be
more dense than a proton [87].
Fixed Αs
Running Αs
6
8
10
12
Y
2
3
4
5
6
Qs
2HA,YLQs
2Hp,YL
Qs
2HA,YL
Qs
2Hp,YL
Qg
2HA,YL
Qg
2Hp,YL
6
8
10
12
Y
5
10
15
20
25
30
Q2
Figure 30: Left: The ratio of the saturation momenta. (Y=12 corresponds to a pseudorapidity
η=6 for the produced particles). Right: Geometric scaling windows.
For momenta p⊥larger than Qs, the gluon distribution satisfies geometrical scaling [86,
88], i.e. it is a function of only the combined variable p⊥/Qs(Y):
Φ(p⊥,Y)∝"Q2
s(Y)
p2
⊥#γ ln p2
⊥
Q2
s(Y)+c!,(16)
with γ=0.63 and c=O(1). This holds within the scaling window Qs.p⊥.Qg,
where lnQ2
g(Y)/Q2
s(Y)∼[lnQ2
s(Y)/Λ2]1/3and for large Ythis is proportional to Y1/6. The
geometrical scaling lines for a proton and a nucleus are shown in Fig. 30. Note that, since Qgis
increasing much faster than Qs, a common scaling window exists, at Qs(A,Y).p⊥.Qg(p,Y)
(and for sufficiently large Y), where the gluon distributions for both the nucleus and the proton
are described by Eq. (16).
Within this window, it is straightforward to calculate the RpAratio. This is shown in
Fig. 31 for two values of pseudorapidity. The upper, dotted, line is the asymptotic prediction
of a fixed-coupling scenario, in which the ratio Q2
s(A,Y)/Q2
s(p,Y)=const.=A1/3, while the
lowest, straight, curve is the line of total shadowing RpA=1/A1/3. Our prediction with running
coupling is the line in between and it is very close to total shadowing. This is clearly a
consequence of the fact that the proton and the nuclear saturation momenta approach each
other with increasing energy.
Note finally that in the present analysis we have neglected the effects of particle number
fluctuations (or “Pomeron loops”). This is appropriate since Pomeron loops effects are
suppressed by the running of the coupling [89], and thus can be indeed ignored at all energies
of phenomenological interest (in particular, at LHC).

Heavy Ion Collisions at the LHC - Last Call for Predictions 41
A-13
Η=3
@A lnHA13LD-H1-ΓL3
9
10
11
12
13
14
15
p¦
2
0.1
0.2
0.3
0.4
0.5
RpA
A-13
Η=5
@A lnHA13LD-H1-ΓL3
14
16
18
20
22
24
p¦
2
0.1
0.2
0.3
0.4
0.5
RpA
Figure 31: The ratio RpAas a function of p2
⊥at √s=8.8TeV.
1.17. LHC dNch/dηand Nch from Universal Behaviors
S. Jeon, V. Topor Pop and M. Bleicher
RHIC dNch/dηcontains two universal curves, one for limiting fragmentation and one for the
transition region. By extrapolating, we predict dNch/dηand Nch/Npart at the LHC energy.
Data from RHIC at all energies clearly show limiting fragmentation phenomena [90] for
very forward and very backward rapidities. In reference [91], we have shown that in the RHIC
dN/dη(normalized to the number of colliding nucleon pairs) spectra at various energies, there
are in fact two universal curves. This fact is not readily visible if one compares the dN/dη
from different energies directly. It is, however, clearly visible in the slope d2N/dη2as shown
in the left panel in figure 32 In this panel, we have plotted dN/dηper participant pair from the
-3 -2 -1 0 1 2 3 4 5 6
η-ymax
-2
-1
0
1
2
3
4
-d2N/dη2 dN/dη
Au-Au 200 GeV
Au-Au 130 GeV
Au-Au 62.4 GeV
Au-Au 19.6 GeV
0 1 2 3 4 5 6
η-yp
-d2N/dη2 -(dN/dη)shifted
Figure 32: Evidence of two universal curves in RHIC dN/dηdata. The slope d2N/dη2is
inverted for visibility. In the left panel, ymax ≈ln( √s/mN) are matched whereas in the right
panel, the shoulders of dN/dηare matched.
PHOBOS collaboration for √sNN =19.6,62.4,130,200 GeV as a function of η−ymax with the

Heavy Ion Collisions at the LHC - Last Call for Predictions 42
corresponding d2N/dη2.
Even though the curves all look similar in dN/dη, it is rather obvious in d2N/dη2that
the true universal behavior is maintained only up to about 50% of the maximum height. More
interestingly, there emerges another universal behavior beyond that point as shown in the right
panel in Fig.1. In this panel, we have shifted dN/dηvertically and horizontally to match the
shoulder. The common straight line in d2N/dη2in this region clearly show that the shoulder
region in dN/dηis a quadratic function of η. Moreover the curvature of the quadratic function
is independent of the colliding energy. The universality of these two curves also implies that
(dN/dη)η=0will at most grow like ln2(√sNN /mN) and the total number of produced particle
Nch can at most grow like ln3(√sNN /mN).
Parameterizing the d2N/dη2with simple functions in two slightly different ways (for
details see reference [91], we can easily extrapolate to the LHC energy as shownin figure 33.
Our prediction is slightly higher than purely linear extrapolation carried out by W. Busza in
reference [44].
(dN/dη)0Ntotal
Param I 6.9 87
Param II 6.5 83
K & L 10.7 110
HIJING w/p0=3.5GeV 21.4 160
HIJING w/p0=5.0GeV 13.6 110
−9 −8 −7 −6 −5 −4 −3 −2 −1 0 1 2 3
η − ymax
0
2
4
6
8
10
12
14
16
18
20
22
(dN/dη)/(Npart/2)
LHC: Param I
LHC: Param II
RHIC 200 GeV
K & L
HIJING: p0 = 3.5 GeV
HIJING: p0 = 5.0 GeV
Figure 33: Predictions for 6% central Au-Au collisions at LHC. Curves and rows labeled
Param I & II are our predictions. For comparison, HIJING predictions with two different
minijet parameters and Kharzeev and Levin formula [7] extrapolated to LHC are also shown.
1.18. Hadron multiplicities at the LHC
D. Kharzeev, E. M. Levin and M. Nardi
We present the predictions for hadron multiplicities in pp,pA and AA collisions at the LHC
based on our approach to the Color Glass Condensate.
We expect that at LHC energies, the dynamics of soft and semi-hard interactions
will be dominated by parton saturation. In this short note we summarize our results for
hadron multiplicities basing on the approach that we have developed and tested at RHIC
energies in recent years [7, 92–94]; a detailed description of our predictions for the LHC
energies can be found in reference [5]. In addition, we will briefly discuss the properties

Heavy Ion Collisions at the LHC - Last Call for Predictions 43
of non-linear evolution at high energies, and their implications; details will be presented
elsewhere [95]. Our approach is based on the description of initial wave functions of
colliding hadrons and nuclei as sheets of Color Glass Condensate. We use a corresponding
ansatz for the unintegrated parton distributions, and compute the inclusive cross sections of
parton production using k⊥-factorization. The hadronization is implemented through the local
parton-hadron duality – namely, we assume that the transformation of partons to hadrons is
a soft process which does not change significantly the angular (and thus pseudo-rapidity)
distribution of the produced particles. Because of these assumptions, we do not expect our
results be accurate for the transverse momentum distributions in AA collisions, but hope that
our calculations (see figure 34a) will apply to the total multiplicities.
This work
(2/Npart)dNAA/dη
η=0, Centrality 0-6 % 1
2
W
0
2
4
6
8
10
12
10210 3
(a) (b)
Figure 34: (a) Charged hadron multiplicity in Pb-Pb collisions as a function of pseudo-rapidity
at √sNN =5.5 TeV; also shown are predictions from other approaches (from [5]); (b) Energy
dependence of charged hadron multiplicity per participant pair in central AA collisions for
different approaches to parton evolution (curves 1 and 2); also shown is the logarithmic fit,
dashed curve (from [95]).
While our approach has been extensively tested at RHIC, an extrapolation of our
calculations to the LHC energies requires a good theoretical control over the rapidity
dependence of the saturation momentum Qs(y). The non-linear parton evolution in QCD
is a topic of vigorous theoretical investigations at present. Recently, we have investigated the
role of longitudinal color fields in parton evolution at small x, and found that they lead to the
following dependence of the saturation momentum on rapidity [95]:
Q2
s(Y)=Q2
s(Y=Y0) exp2αS
π(Y−Y0)
1+BQ2
s(Y=Y0)exp2αS
π(Y−Y0)−1,(17)
where B=1/(32π2) (πR2
A/αS); RAis the area of the nucleus, and αSis the strong coupling
constant. At moderate energies, equation (17) describes an exponential growth of the
saturation momentum with rapidity; when extrapolated to the LHC energy this results in the

Heavy Ion Collisions at the LHC - Last Call for Predictions 44
corresponding growth of hadron multiplicity, see curve "1" in figure 34b. At high energies,
equation (17) predicts substantial slowing down of the evolution, which results in the decrease
of hadron multiplicity as shown in figure 34b by the curve "2". In both cases, the growth of
multiplicity is much slower than predicted in the conventional "soft plus hard" models, see
figure 34. We thus expect that the LHC data on hadron multiplicities will greatly advance the
understanding of QCD in the strong color field regime.
1.19. CGC at LHC
B. Kopeliovich and I. Schmidt
Data strongly indicate the localization of glue in hadrons within small spots. This leads to a small
transverse overlap of gluons in nuclei, i.e. to weak CGC effects. We predict a weak Cronin effect for
LHC, not considerably altered by gluon shadowing.
There are many experimental evidences for the localization of the glue in hadrons within
spots of small size, r0≈0.3 fm [96,97]. Correspondingly, the mean transverse momentum
of gluons in the proton should be rather large, about 700 MeV/c. One of the manifestation
of this phenomenon is a weak Cronin enhancement for gluons. Indeed, the Cronin effect is
a result of the interplay between the primordial transverse momentum, hk2
Ti, of the incoming
parton and the additional momentum, ∆hk2
Ti, gained in the nucleus (broadening). The relative
significance of the latter controls the magnitude of the Cronin enhancement. Apparently, the
larger the original hk2
Tiis, the weaker is the Cronin effect. The pT-slope of the cross section
also matters: the steeper it is, the stronger is the nuclear enhancement.
Although a rather strong Cronin effect was observed in fixed target experiments, the
production of high-pThadrons is dominated by scattering of valence quarks [98]. One can
access the gluons only at sufficiently high energies. Relying on the above consideration, a
very weak Cronin enhancement was predicted in [98] at √sNN =200 GeV, as is depicted in
figure 35. A several times stronger effect was predicted in [99]∗, and a suppression, rather
than enhancement, was the expectation of the color glass condensate (CGC) model [100]. The
latest data from the PHENIX experiment at RHIC support the prediction of [98].
At LHC energies one can access quite small values of Bjorken x, such that the lifetime
of gluonic fluctuations, tc≈0.05/(xmN) [101], becomes longer than the nuclear size. Then
one might expect coherence effects, in particular pronounced signatures of CGC. However,
the longitudinal overlap of gluons is not sufficient, since they also have to overlap in impact
parameter, which is something problematic for small gluonic spots. The mean number of
gluons overlapping with a given one in a heavy nucleus is, hni=3π
4r2
0hTAi=πr2
0ρARA=0.3,
and such a small overlap results in a quite weak CGC and gluon shadowing. The latter is
confirmed by the NLO analysis of nuclear structure functions performed in [15]. Missing this
important observation, one could easily overestimateboth the CGC and gluon shadowing.
Thus, we expect that the effects of CGC, both the Cronin enhancement and shadowing
∗The extremely strong gluon shadowing implemented into the HIJING model is ruled out by the recent NLO
analysis [15] of DIS data.

Heavy Ion Collisions at the LHC - Last Call for Predictions 45
0.8
0.9
1
1.1
1.2
0.8
0.9
1
1.1
1.2
0 2 4 6 8 10 12 14 16
pT (GeV/c)
RA(pT)
√s
−=200 GeV
√s
−=5.5 TeV
Figure 35: Nucleus-to-proton ratio for pion production versus pT. Dashed and solid curves
correspond to calculations without or with gluon shadowing [98].
suppression, to be rather weak at the LHC, and nearly compensating each other. Therefore in
this case the nucleus-to proton ratio is expected to approach unity from below at high pT.
1.20. Fluctuation Effects on RpA at High Energy
M. Kozlov, A. I. Shoshi and B.-W. Xiao
We discuss a new physical phenomenon for RpA in the fixed coupling case, the total gluon
shadowing, which arises due to the effect of gluon number fluctuations.
We study the ratio of the unintegrated gluon distribution of a nucleus hA(k⊥,Y) over the
unintegrated gluon distribution of a proton hp(k⊥,Y) scaled up by A1/3
RpA =hA(k⊥,Y)
A1
3hp(k⊥,Y).(18)
This ratio is a measure of the number of particles produced in a proton-nucleus collision
versus the number of particles in proton-proton collisions times the number of collisions. The
transverse momentum of gluons is denoted by k⊥and the rapidity variable by Y.
In the geometric scaling region shown in Fig. 36a the small-xphysics is reasonably
described by the BK-equation which emerges in the mean field approximation. Using the BK-
equation one finds in the geometric scaling regime in the fixed coupling case that the shape of
the unintegrated gluon distribution of the nucleus and proton as a function of k⊥is preserved
with increasing Y, because of the geometric scaling behaviour hp,A(k⊥,Y)=hp,A(k2
⊥/Q2
s(Y)),
and therefore the leading contribution to the ratio RpA is k⊥and Yindependent, scaling with
the atomic number Aas RpA =1/A1/3(1−γ0), where γ0=0.6275 [87]. This means that gluons
inside the nucleus and proton are somewhat shadowed since hA/hp=Aγ0/3lies between total

Heavy Ion Collisions at the LHC - Last Call for Predictions 46
(hA/hp=1) and zero (hA/hp=A1/3) gluon shadowing. The partial gluon shadowing comes
from the anomalous behaviour of the unintegrated gluon distributions which stems from the
BFKL evolution.
We have recently shown [102] that the behaviour of RpA as a function of k⊥and Yin the
fixed coupling case is completely changed because of the effects of gluon number fluctuations
or Pomeron loops at high rapidity. According to [103] the influence of fluctuations on the
unintegrated gluon distribution is as follows: Starting with an intial gluon distribution of the
nucleus/proton at zero rapidity, the stochastic evolution generates an ensamble of distributions
at rapidity Y, where the individual distributions seen by a probe typically have different
saturation momenta and correspond to different events in an experiment. To include gluon
number fluctuations one has to average over all individual events, hfluc.
p,A(k⊥,Y)=hhp,A(k⊥,Y)i,
with hp,A(k⊥,Y) the distribution for a single event. The main consequence of fluctuations is
the replacement of the geometric scaling by a new scaling, the diffusive scaling [103, 104],
hhp,A(k⊥,Y)i=hp,Aln(k2
⊥/hQs(Y)i2))/[DY]. The diffusive scaling, see Fig. 36a, sets in
when the dispersion of the different events is large, σ2=hρs(Y)2i− hρs(Y)i2=DY ≫1,
i.e., Y≫YDS =1/D, where ρs(Y)=ln(Q2
s(Y)/k2
0) and Dis the diffusion coefficient, and is
valid in the region σ≪ln(k2
⊥/hQs(Y)i2)≪γ0σ2. The new scaling means that the shape
of the unintegrated gluon distribution of the nucleus/proton becomes flatter and flatter with
increasing rapidity Y, in contrast to the preserved shape in the geometric scaling regime. This
is the reason why the ratio in the diffusive scaling regime [102]
RpA(k⊥,Y)≃1
A1
31−lnA1/3
2σ2"k2
⊥
hQs(A,Y)i2#lnA1/3
σ2(19)
yields total gluon shadowing,RpA =1/A1/3, at asymptotic rapidity Y(at fixed A). This result
is universal since it does not depend on the initial conditions. Moreover the slope of RpA as a
function of k⊥descreases with increasing Y. The qualitative behaviour of RpA at fixed αsdue
to fluctuation effects is shown in Fig. 36b.
The above effects of fluctuations on RpA are valid in the fixed coupling case and at very
large energy. It isn’t clear yet whether the energy at LHC is high enough for them to become
important. Moreover, in the case where fluctuation effects are neglected but the coupling
is allowed to run, a similar behaviour for RpA is obtained [85], including the total gluon
shadowing. It remains for the future to be clarified how important fluctuation or running
coupling effects are at given energy windows, e.g., at LHC energy.
1.21. Particle Production at the LHC: Predictions from EPOS
S. Porteboeuf, T. Pierog and K. Werner
We present EPOS predictions for proton-proton scattering and for lead-lead collisions at different
centralities at LHC energies. Wefocus on soft physics and show particle spectra of identified particles
and some results on elliptical flow. We claim that collective affects are already quite important in
proton-proton scattering.

Heavy Ion Collisions at the LHC - Last Call for Predictions 47
diffusive
scaling
geometric
scaling
saturation
low density
hρs(A, Y )i
ρ= ln(k2
⊥/k2
0)ln(Λ2
QCD/k 2
0)
Y= ln 1/x
YDS
YF
Qs(Y1)Qs(Y4)
RpA
k⊥
1
A
1
3
1
Y1
Y2
Y3
Y4
Figure 36: (a) Phase diagram of a highly evolved nucleus/proton. (b) Rp,Aversus k⊥at
different rapidities Y4≫Y3≫Y2≫Y1.
EPOS is a consistent quantum mechanical multiple scattering approach based on partons
and strings, where cross sections and the particle production are calculated consistently, taking
into account energy conservation in both cases [105]. A special feature is a careful treatment
of projectile and target remnants.
Nuclear effects related to Cronin transverse momentum broadening, parton saturation,
and screening have been introduced into EPOS [106].
Furthermore, high density effects leading to collective behavior in heavy ion collisions
are also taken into account ("plasma core") [107].
We first show in fig. 37 pseudorapidity and transverse momentum spectra of charged
particles and of different identified hadrons, as well as some particle ratios, in proton-proton
scattering at 14 TeV. As for heavy ions, the default version of EPOS considers also in proton-
proton scattering the formation of a core (dense area), with a hydrodynamical collective
expansion. Whereas such "mini-plasma cores" are negligible in proton-proton scattering at
RHIC, they play an important role at the LHC, which can be seen from the difference between
the full curves (full EPOS, including "mini-plasma") and the dotted curves ("mini-plasma
option turned off"). The effect is even more drastic when we investigate the multiplicity
dependence of particle production, see fig. 37.
In the following, we investigate lead-lead collisions at 5.5 TeV. In fig. 38, we plot the
centrality dependence of particle yields for charged particles and different identified hadrons.
We observe an increase by roughly 2.5 for pions, and a bigger factor for the heavier particles.
In fig. 39, we show pseudorapidity spectra, for different particles, at different centralities.
The pseudorapidity density of charged particles at η=0 is around 2500, for central collisions.
In fig. 40, we show nuclear modification factors RAA (ratios with respect to proton-
proton, divided by Ncoll), considering charged particles and different identified hadrons. The
peak structure of the baryon results is related to the concave form of the baryon spectra from
the radially flowing core in PbPb collisions. All curves are well below one, indicating strong
screening effects.

Heavy Ion Collisions at the LHC - Last Call for Predictions 48
0
2
4
6
8
0 5
η
dn/dη
14TeV chgd
inel
NSD
10 -1
1
10
0 5 10
η
dn/dη
14TeV
π
K
p
p
_
10 -7
10 -5
10 -3
10 -1
0 5
pt (GeV/c)
dn/dηd2pt (c2/GeV2)
14TeV chgd
π
K
p
Λ
Ξ
η=0
0
0.25
0.5
0.75
1
0 5
pt (GeV/c)
K+/π+
14TeV
0
0.25
0.5
0.75
1
0 5
pt (GeV/c)
14TeV
ap/π-
0
0.5
1
1.5
0 5
pt (GeV/c)
Λ/K+
14TeV
0.2
0.4
0.6
0.8
0 20
dNch/dη(0)
mean pt (GeV/c)
14TeV
chgd
0.5
1
1.5
0 20
dNch/dη(0)
mean pt (GeV/c)
14TeV
p
_
π
0.5
1
1.5
2
0 20
dNch/dη(0)
mean pt (GeV/c)
14TeV
Λ
K
Figure 37: Proton-proton scattering at 14 TeV: pseudorapidity distributions of charged
particles (upper row left), and of different identified hadrons (upper row middle), as
well as tranverse momentum spectra of different identified hadrons at η=0 (upper row
right),transverse momentum dependence of particle ratios at η=0 (middle row),the average
transverse momentum of charged particles and of different identified hadrons at η=0
(lower row). The full lines refer to the "mini-plasma option", the dotted ones refer to the
"conventional option (mini-plasma turned off)".
In fig. 41, we finally show the transverse momentum dependence of the elliptical flow.
The full line is the full calculation, the dashed one only the core contribution. The big
difference between the two is due to the fact that high ptjets are allowed to freely leave
the core (no jet quenching).
1.22. Forward hadron production in high energy pA collisions
K. L. Tuchin
We present a calculation of π,Dand Bproduction at RHIC and LHC energies based upon the
KKT model of gluon saturation.
In this proceedings we present a calculation of forward hadron production in pA

Heavy Ion Collisions at the LHC - Last Call for Predictions 49
0
1
2
3
4
5
6
7
0 200 400
dn/dη /Npart
chgd
0
0.5
1
1.5
2
2.5
3
0 200 400
dn/dη /Npart
π+
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0 200 400
dn/dη /Npart
K+
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0 200 400
Npart
p
_
0
0.02
0.04
0.06
0.08
0.1
0.12
0 200 400
Npart
Λ
0
0.02
0.04
0.06
0.08
0.1
0 200 400
Npart
Λ
_
Figure 38: Lead-lead collisions at 5.5 TeV: centrality dependence of particle yields (central
pseudorapidity density per participant), for charged particles and different identified hadrons.
0
500
1000
1500
2000
2500
0 2 4 6
dn/dη
chgd 0-5%...70-80%
0
200
400
600
800
1000
0 2 4 6
dn/dη
π+ 0-5%...70-80%
0
20
40
60
80
100
120
140
160
180
0246
dn/dη
K+ 0-5%...70-80%
0
10
20
30
40
50
60
70
80
0 2 4 6
pseudorapidity η
p 0-5%...70-80%
0
10
20
30
40
50
60
70
0 2 4 6
pseudorapidity η
p
_ 0-5%...70-80%
0
10
20
30
40
50
0246
pseudorapidity η
Λ 0-5%...70-80%
Figure 39: Lead-lead collisions at 5.5 TeV: pseudorapidity distributions of charged particles
and of different identified hadrons, at different centralities. For each plot, from top to bottom:
0-5%, 10-20%, 25-35%, 40-50% 70-80%.

Heavy Ion Collisions at the LHC - Last Call for Predictions 50
0
0.25
0.5
0.75
1
024
RAA
chgd 0-5% ... 70-80%
0
0.25
0.5
0.75
1
024
RAA
π+ 0-5% ... 70-80%
0
0.25
0.5
0.75
1
0 2 4
RAA
K+ 0-5% ... 70-80%
0
0.25
0.5
0.75
1
024
pt
p 0-5% ... 70-80%
0
0.25
0.5
0.75
1
024
pt
p
_ 0-5% ... 70-80%
0
0.5
1
0 2 4
pt
Λ 0-5% ... 70-80%
Figure 40: Lead-lead collisions at 5.5 TeV: the nuclear modification factor RAA at η=0 of
charged particles and of different identified hadrons, at different centralities. For each plot,
from top to bottom: 70-80%, 40-50%, 25-35%, 10-20%, 0-5%.
0
0.1
0.2
0.3
024
pt
v2
C
0
0.1
0.2
0.3
024
pt
v2
π
0
0.1
0.2
0.3
0 2 4
pt
v2
K
0
0.1
0.2
0.3
024
pt
v2
p
0
0.1
0.2
0.3
024
pt
v2
Λ
0
0.1
0.2
0.3
0 2 4
pt
v2
Ξ
Figure 41: Lead-lead collisions at 5.5 TeV: the transverse momentum dependence of the
elliptical flow at η=0 of charged particles and of different identified hadrons, for minimum
bias collisions. The full line is the full calculation, the dashed one only the core contribution.

Heavy Ion Collisions at the LHC - Last Call for Predictions 51
collisions at RHIC and LHC. The theoretical framework for inclusive gluon production
including the effect of gluon saturation was set up in Ref. [108]. It has been successfully
applied to study the inclusive light hadron production at RHIC [109]. Since the KKT model
of Ref. [109] works so well at RHIC we decided to extend it to the LHC kinematical region.
Doing so we explicitly neglect a possible effect of gluon saturation in a proton which is
perhaps a good approximation for the nuclear modification factor. The results of calculation
of inclusive pion production are shown in Fig. 42.
pT (GeV)
RpA
ηRHIC = 3.2
ηLHC = 0.
ηLHC = 3.
ηLHC = 6.
(min. bias)Pions
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7 8
pT (GeV)
RpA
ηRHIC = 3.2
ηLHC = 0.
ηLHC = 3.
ηLHC = 6.
(0-20%)Pions
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7 8
Figure 42: Nuclear modification factor for pion production at RHIC and LHC.
Production of heavy quarks at small xis also affected by gluon saturation in a way similar
to that of gluons [110]. The main difference, however, is that the effect of gluon saturation
is postponed to higher energies/rapidities for heavier quarks as compared to lighter quarks
and gluons. This is because the relevant xis proportional to m⊥=(m2+k2
⊥)1/2and hence is
higher for heavier quarks at the same values of √s,y,k⊥. In Fig. 43 and Fig. 44 the nuclear
modification factors for open charm and beauty are shown. The calculations are based upon
the theoretical result of Ref. [111] and the KKT model [108].
If the nuclear modification factor is measured to as high transverse mass as possible,
we can observe transition from the geometric scaling (described by the KKT model) to the
collinear factorization regime. This is shown in Fig. 45. Had the geometric scaling held for
all m⊥and x<0.01, the nuclear modification factor would have been described by the solid
line. However, one expect the breakdown of the geometric scaling as illustrated by the dotted
lines.
A more detailed description of the theoretical approach to the heavy quark production as
well as discussion of the obtained results will be provided in a forthcoming publication.
1.23. Rapidity distributions at LHC in the Relativistic Diffusion Model
G. Wolschin
Stopping and particle production in heavy-ion collisions at LHC energies are investigated in a

Heavy Ion Collisions at the LHC - Last Call for Predictions 52
pT (GeV)
RpA
yRHIC = 0
yRHIC = 2
yLHC = 0
yLHC = 2
yLHC = 4
(Min. bias)Open charm
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7
pT (GeV)
RpA
yRHIC = 0
yRHIC = 2
yLHC = 0
yLHC = 2
yLHC = 4
(0-20%)Open charm
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7
Figure 43: Nuclear modification factor for open charm production at RHIC and LHC.
pT (GeV)
RpA
yRHIC = 0
yRHIC = 2
yLHC = 0
yLHC = 2
yLHC = 4
(Min. bias)
Open beauty
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7 8 9
pT (GeV)
RpA
yRHIC = 0
yRHIC = 2
yLHC = 0
yLHC = 2
yLHC = 4
(0-20%)
Open beauty
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 1 2 3 4 5 6 7 8 9
Figure 44: Nuclear modification factor for open beauty production at RHIC and LHC. Note,
that the calculations of [111] break down at y=0 at RHIC (xis not small enough); the
corresponding result (solid line) is shown for completeness.
Relativistic Diffusion Model (RDM). Using three sources for particle production, the energy- and
centrality dependence of rapidity distributions of net protons, and pseudorapidity spectra of charged
hadrons in heavy systems are studied from SPS to LHC energies. The transport coefficients are
extrapolated from Au +Au at RHIC energies (√sNN=19.6 - 200 GeV) to Pb +Pb at LHC energies
of √sNN =5.52 TeV. Rapidity distributions for net protons, and pseudorapidity spectra for produced
charged particles are calculated at LHC energies.
Net-proton and charged-hadron distributions in collisions of heavy systems have
been calculated in a three-sources Relativistic Diffusion Model (RDM) for multiparticle
interactions from SPS to LHC energies. Analytical results for the rapidity distribution of net
protons in central collisions, and produced charged hadrons are found to be in good agreement

Heavy Ion Collisions at the LHC - Last Call for Predictions 53
m(GeV)
RpA
s
cb
t
√s=5500 GeV, y=2, k⊥=0.5 GeV
0
0.2
0.4
0.6
0.8
1
10 -1 1 10 102
Figure 45: Dependence of the nuclear modification factor on quark mass. Solid line is RpA for
quarks. Geometric scaling is expected to break down at m⊥∼Qgeom ≃Q2
s/Λ, and therefore
RpA is anticipated to deviate from the solid line towards unity. Dotted lines illustrate a possible
behavior of RpA.
with the available data (Figs. 46, 47) at RHIC.
An extrapolation of the transport coefficients for net protons, and produced hadrons to
Pb +Pb at LHC energies of √sNN =5.52 TeV has been performed in [112,113], and the
corresponding rapidity distributions have been calculated as shown in Figs. 46, 47.
The net-proton result for LHC is shown for particle contents of 7 % and 14 % in the
central source, respectively [112]. Kinematical constraints will modify the result at large
values of the rapidity y. For produced particles, the curves (A) - (D) in Fig. 47 are discussed
in [113]. The essential parameters relaxation time, diffusion coefficients or widths of the
distribution functions of the three sources, and number of particles in the local equilibrium
source will have to be adjusted to the ALICE data.
2. Azimuthal asymmetries
2.1. Transverse momentum spectra and elliptic flow: Hydrodynamics with QCD-based
equations of state
M. Bluhm, B. Kämpfer and U. Heinz
We present a family of equations of state within a quasiparticle model adjusted to lattice
QCD and study the impact on azimuthal flow anisotropies and transverse momentum spectra within
hydrodynamic simulations for heavy-ion collisions at energies relevant for LHC.
2.1.1. Introduction The equation of state (EoS) represents the heart of hydrodynamic
simulations for ultra-relativistic heavy-ion collisions. Here, we present a realistic EoS for
QCD matter delivered by our quasiparticle model (QPM) faithfully reproducing lattice QCD
results. The approach is based on [114–118] adjusted to the pressure pand energy density

Heavy Ion Collisions at the LHC - Last Call for Predictions 54
Figure 46: Net-proton rapidity spectra [112] in
the Relativistic Diffusion Model (RDM), solid
curves: Transition from the double-humped
shape at SPS energies of √sNN =17.3 GeV to
a broad midrapidity valley at RHIC (200 GeV)
and LHC (5.52 TeV).
-10 -8 -6 -4 -2 0 2 46 8 10
Η
00
300
600
900
1200
1500
1800
0
200
400
600
800
1000
dN

dΗ
@AD
@BD
@CD
@DD
@ADLHC
RHIC
Figure 47: Produced charged hadrons
for central Au +Au collisions at RHIC
compared with 200 A GeV PHOBOS data,
and diffusion-model extrapolation to Pb +Pb
at LHC energies of 5520 GeV. See [113] for
curves [A] to [D] at LHC energies.
eof Nf=2+1 quark flavors [119, 120]. As the QPM EoS does not automatically fit to
the hadron resonance gas EoS in the confinement region, we construct a family of EoS’s by
an interpolation between the hadron resonance gas at e1=0.45 GeV/fm3and the QPM at
flexible em(cf. [121] for details). In this way, the influence of details in the transition region
on hydrodynamic flow can be studied, since for e<e1and e>emthe EoS is uniquely given
by the resonance gas and the QCD-based QPM, respectively. In Figure 48, we exhibit the
EoS family in the form p=p(e) and the corresponding speed of sound v2
s=∂p/∂e. For LHC,
baryon density effects are negligible.
Figure 48: Left panel: Family of EoS’s p(e) labelled in the following as QPM(em) with em=
4.0, 2.0, 1.25, 1.0 GeV/fm3(solid curves) combining QPM adjusted to lattice data [119,120]
and hadron resonance gas at matching point em. For comparison the bag model EoS (dashed
line) is shown. Right panel: corresponding v2
s.

Heavy Ion Collisions at the LHC - Last Call for Predictions 55
2.1.2. Predictions for heavy-ion collisions at LHC We concentrate on two extreme EoS’s,
QPM(4.0) and the bag model EoS being similar to QPM(1.0). We calculate transverse
momentum spectra and elliptic flow v2(pT) using the relativistic hydrodynamic program
package [122, 123] with initial conditions for Pb+Pb collisions at impact parameter b=5.2
fm. For the further initial parameters required by the program we conservatively guess s0=
330 fm−3,n0=0.4 fm−3and τ0=0.6 fm/c for initial entropy density, baryon density and
time. Within the QPM these translate into e0=127 GeV/fm3,p0=42 GeV/fm3and T0=
515 MeV. The freeze-out temperature is set Tf.o.=100 MeV. In Figure 49, we exhibit our
results at midrapidity for various primordial hadron species. Striking is the strong radial flow
as evident from the flat pT-spectra and a noticeably smaller v2(pT) than at RHIC in particular
at low pT[121]. Details of the Eos in the transition region as mapped out by our family are
still visible.
Figure 49: Transverse momentum spectra (left panels) and azimuthal anisotropy (right panels)
for directly emitted pions, kaons and protons (upper row) and strange baryons (lower row).
Solid and dashed curves are for EoS QPM(4.0) and the bag model EoS, respectively.
2.2. The centrality dependence of elliptic flow at LHC
H.-J. Drescher, A. Dumitru and J.-Y. Ollitrault
We present predictions for the centrality dependence of elliptic flow at mid-rapidity in Pb-Pb
collisions at the LHC.
The centrality and system-size dependence of elliptic flow (v2) provides direct
information on the thermalization of the matter created in the collision. Ideal (non-viscous)

Heavy Ion Collisions at the LHC - Last Call for Predictions 56
hydrodynamics predicts that v2scales like the eccentricity, ε, of the initial distribution of
matter in the transverse plane. Our predictions are based on this eccentricity scaling, together
with a simple parameterization of deviations from hydrodynamics [124]:
v2=hε
1+K/0.7,(20)
where the scale factor his independent of system size and centrality, but may depend on the
collision energy. The Knudsen number Kcan be expressed as
1
K=σ
SdN
dy1
√3.
It vanishes in the hydrodynamic limit. dN/dyis the total (charged +neutral) multiplicity per
unit rapidity, Sis the transverse overlap area between the two nuclei, and σis an effective
(transport) partonic cross section.
The model has two free parameters, the “hydrodynamic limit” h, and the partonic cross
section σ. The other quantities, ε,S, dN/dy, must be obtained from a model for the initial
condition. Here, we choose the Color Glass Condensate (CGC) approach, including the
effect of fluctuations in the positions of participant nucleons, which increase ε[125]. The
model provides a perfect fit to RHIC data for Au-Au and Cu-Cu collisions with h=0.22 and
σ=5.5 mb [124].
We now briefly discuss the extrapolation to LHC. The hydrodynamic limit his likely to
increase from RHIC to LHC, as the QGP phase will last longer; however, we do not have a
quantitative prediction for h. We predict only the centrality dependence of v2, not its absolute
value. Figure 50 is drawn with h=0.22.
The second parameter is σ, which parameterizes deviations from ideal hydrodynamics,
i.e., viscous effects. We consider two possibilities: 1) σ=5.5 mb at LHC, as at RHIC. 2)
σ∼1/T2(on dimensional grounds, assuming that no non-perturbative scales arise), where
the temperature T∼(dN/dy)1/3. This gives the value 3.3 mb in figure 50.
The remaining quantities (S, dN/dyand ε) are obtained by extrapolating the CGC
from RHIC to LHC, either with fixed-coupling (fc) or running-coupling (rc) evolution of
the saturation scale Qs. The multiplicity per participant increases by a factor of 3 (resp. 2.4)
with fc (resp. rc). The eccentricity εis 10% larger with fc (solid curve in figure 50) than
with rc (dash-dotted curve) evolution. Deviations from hydrodynamics (the K-dependent
factor in equation (20)) are somewhat smaller than at RHIC: v2is 90% (resp. 80%) of the
hydrodynamic limit for central collisions if σ=5.5 mb (resp. 3.3 mb). Our predictions lie
between the dashed and dotted curves, up to an overall normalization factor. The maximum
value of v2occurs for Npart between 60 (σ≈const.) and 80 (σ∼1/T2).
Elliptic flow will be a first-day observable at LHC. Both its absolute magnitude and its
centrality dependence are sensitive probes of initial conditions, and will help to improve our
understanding of high-density QCD.
2.3. Elliptic flow from pQCD+saturation+hydro model
K. J. Eskola, H. Niemi and P. V. Ruuskanen

Heavy Ion Collisions at the LHC - Last Call for Predictions 57
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
50 100 150 200 250 300 350 400
v2
Npart
hydro fixed coupling
hydro running coupling
σ=5.5 mb fixed coupling
σ=3.3 mb running coupling
Phobos
Figure 50: v2as a function of Npart at mid-rapidity for Pb-Pb collisions at LHC ( √sNN =
5.5 TeV). solid- and dash-dotted lines: εscaling (K=0 in (20)); dashed- and dotted lines: incl.
incomplete thermalization, with two values of the partonic cross section. Squares: PHOBOS
data for Au-Au collisions at RHIC [126]. The vertical scale is arbitrary (see text).
We have previously predicted multiplicities and transverse momentum spectra for the
most central LHC Pb+Pb collisions at √sNN =5.5 TeV using pQCD +saturation +hydro
(EKRT model) [31, 74]. We now extend these calculations for non-central collisions and
predict low-pTelliptic flow. Our model is in good agreement with RHIC data for central
collisions, and we show that our extension of the model is also in good agreement with
minimum bias v2data from RHIC Au+Au collisions at √sNN =200 GeV.
We obtain the primary partonic transverse energy production and the formation time
in central AAcollisions from the EKRT model [74]. With the assumption of immediate
thermalization we can use these to estimate the initial state for hydrodynamic evolution. For
centrality dependence we consider here two limits which correspond to models eWN and eBC
in [127], where the profile and normalization are obtained from optical Glauber model, once
the parameters in central collisions are fixed. In the eWN (eBC) model the energy density
profile and normalization are proportional to the density and the number of wounded nucleons
(binary collisions), respectively. These energy density profiles are used to initialize boost
invariant hydro code with transverse expansion. We use the bag model equation of state with
massless gluons and quarks (Nf=3), and hadronic phase with all hadronic states up to a mass
2 GeV included. Phase transition temperature is fixed to 165 MeV. Decoupling is calculated
using standard Cooper-Frye formula, and all decays of unstable hadronic states are included.
Freeze-out temperature is fixed from RHIC pTspectra for the most central collisions and
is 150 MeV for binary collision profile [31] and 140 MeV for wounded nucleon profile. The
same freeze-out temperatures are used at the LHC. Both initializations give a good description
of the low-pTspectra for different centralities at RHIC.

Heavy Ion Collisions at the LHC - Last Call for Predictions 58
0
0.05
0.1
0.15
0.2
0.25
0 0.5 1 1.5 2 2.5 3
v2
pT [GeV]
π+
LHC eBC
LHC eWN
RHIC eBC
RHIC eWN
STAR π+ + π−
PHENIX π+
0
0.05
0.1
0.15
0.2
0.25
0 0.5 1 1.5 2 2.5 3
v2
pT [GeV]
p
LHC eBC
LHC eWN
RHIC eBC
RHIC eWN
STAR anti-p
PHENIX
Figure 51:
The left panel of figure 51 shows our calculations for pTdependence of minimum bias
v2for positive pions. RHIC results are compared with STAR [128] and PHENIX [129]
data. Our minimum bias centrality selection (0−80%) corresponds to the one used by STAR
collaboration. Solid lines are calculations with the eBC model and dashed lines are from the
eWN model. Thin lines are our results for RHIC Au+Au collisions at √sNN =200 GeV and
thick lines show our predictions for the LHC Pb+Pb collisions at √sNN =5.5 TeV. Largest
uncertainty in v2calculations for pions comes here from the initial transverse profile of the
energy density. Sensitivity to initial time and freeze-out temperature is much weaker. In
general the eWN profile leads to weaker elliptic flow than the eBC profile. At the LHC the
lifetime of the QGP phase is longer, which results in stronger flow asymmetry than at RHIC.
On the other hand the magnitude of transverse flow is also larger, which decreases the v2
value at fixed pT. The net effect is that, for a given profile, v2of low-pTpions is larger at the
LHC than at RHIC. Since jet production at the LHC starts to dominate over the hydrodynamic
spectra at larger pTthan at RHIC [31], we expect that the hydrodynamic calculations should
cover a larger pTrange at the LHC. Thus we predict that the minimum bias v2of pions at
fixed pTis larger at the LHC than at RHIC, and can reach values as high as 0.2.
Our model clearly overshoots the proton v2data from STAR [128] and PHENIX [129].
A more detailed treatment of the hadron gas dynamics and freeze-out is needed to describe
both the proton spectra and elliptic flow simultaneously. However, we can still predict the
change in the behaviour of v2of protons when going from RHIC to the LHC. This is shown in
the r.h.s. of figure 51. Although the flow asymmetry increases at the LHC, for more massive
particles like protons the overall increase in the magnitude of radial flow is more important
than for light pions. This results in smaller v2at the LHC than at RHIC in the whole pTrange
for protons. Even if v2at fixed pTis smaller at the LHC, pT-integrated v2is always larger at
the LHC for all particles, due to the increase in the relative weight at larger pT’s.

Heavy Ion Collisions at the LHC - Last Call for Predictions 59
2.4. From RHIC to LHC: Elliptic and radial flow effects on hadron spectra
G. Kestin and U. Heinz
Using (2+1)-d ideal hydrodynamics [130], we computed the evolution from AGS to LHC
energies of the pT-spectra and elliptic flow at midrapidity for several hadrons [131]. While
ideal fluid dynamics begins to break down below RHIC energies, due to viscous effects in
the late hadronic stage which persist even at RHIC [132], its validity is expected to improve
at the LHC where the elliptic flow saturates in the quark-gluon plasma (QGP) stage, and
effects from late hadronic viscosity become negligible [133]. Early QGP viscous effects seem
small at RHIC [132,134], and recent results from Lattice QCD indicate little change of its
specific shear viscosity η/sfrom RHIC to LHC [135]. The following ideal fluid dynamical
predictions for soft (pT.2−3GeV/c) hadron production in (A≈200)+(A≈200) collisions at
the LHC should thus be robust.
For Au+Au at RHIC we use standard initial (s0=117/fm3,nB0=0.44/fm3at
τ0=0.6fm/c, corresponding to dNch/dy(y=b=0)=680) and final conditions (ef=75 MeV/fm3,
Tf=100MeV) [130,134]. For the LHC we assume dNch/dy(y=b=0)=1200 (the lower end
of the predicted range), using s0=271/fm3and nB0=0 at τ0=0.45 fm/c, keeping the product
T0τ0and Tfunchanged. Predictions for other multiplicities, for interpolation to the actually
measured LHC value, can be found in [131].
1. Elliptic flow of pions and protons: Figure 52 shows the pion and proton elliptic flow at
RHIC and LHC. While the total (pT-integrated) pion elliptic flow increases from RHIC to
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 0.5 1 1.5 2 2.5 3 3.5 4
v2
pT (GeV)
π+
p
π+ (S0=117 fm-3)
π+ (S0=271 fm-3)
p (S0=117 fm-3)
p (S0=271 fm-3)
Figure 52: (Color online) Pion and proton elliptic flow as function of pTfor b=7 fm Au+Au
collisions at RHIC (s0=117fm−3) and LHC (s0=271fm−3).
LHC by about 25% [133], very little of this increase (∼5%) is of ideal fluid dynamical origin,
most of it stemming from the disappearance of late hadronic viscous effects between RHIC
and LHC. At fixed pT, Figure 52 shows a decrease of v2, reflecting a shift of the momentum
anisotropy to larger pTby increased radial flow, which flattens the LHC pT-spectra, affecting

Heavy Ion Collisions at the LHC - Last Call for Predictions 60
0
1
2
3
p/π+
0
2
4
1
3
Λ/Κ+
pT(GeV)
Ω/ϕ
0 1 2 3
0
0.2
0.4
0.3
0.1
1
10-2
102
103
0.1
10
dN/(pTdpTdy)
1
10-2
102
10-3
103
0.1
10
0 1 2 3 4
pT(GeV)
10-3
10-2
0.1
1
10
π+
p
K+
Λ
φ
Ω
S0 = 117 fm-3, nB,0 = 0.44 fm-3 S0 = 271 fm-3, nB,0 = 0 fm-3
RHIC
LHC
RHIC
RHIC
LHC
LHC
Figure 53: (Color online) Normalized pT-spectra (right) and pT-dependent particle ratios
(left) for (¯p, π+), (Λ,K+), and (Ω, φ) in central Au+Au collisions at RHIC and LHC. Hadron
yields are assumed to freeze out at Tc=164 MeV.
the heavier protons more than the lighter pions (Figure 53, right column). These radial flow
effects on v2(pT) are very small for pions but clearly visible for protons.
2. pT-dependence of hadron ratios: Hydrodynamic flow, which leads to flatter pT-spectra
for heavy than light particles, is a key contributor to the observed strong rise of the ¯p/π and
Λ/Kratios at low pTat RHIC [134]. Figure 53 shows that this rise is slower at LHC than
at RHIC (left column) since all spectra are flatter at LHC due to increased radial flow (right
column) while their asymptotic ratios at pT→∞ (given by their fugacity and spin degeneracy
ratios [134]) remain similar.
2.5. Differential elliptic flow prediction at the LHC from parton transport
D. Molnár
Introduction. General physics arguments and calculations for a class of conformal

Heavy Ion Collisions at the LHC - Last Call for Predictions 61
field theories suggest [136, 137] that quantum effects impose a lower bound on transport
coefficients. For example, the shear viscosity to entropy density ratio is above a small value
η/s&0.1 (“most perfect fluid” limit). Dissipative effects can therefore never vanish in a finite,
expanding system. On the other hand, ideal (nondissipative) hydrodynamic modelling of
Au+Au collisions at RHIC ( √sNN ∼100 GeV) is rather successful, leading many to postulate
that the hot and dense QCD matter created is in fact such a “most perfect fluid” (at least during
the early stages of the RHIC evolution). We predict here how differential elliptic flow v2(pT)
changes from RHIC to LHC collision energies (Pb+Pb at √sNN =5.5 TeV), if the quark-gluon
system created at both RHIC and the LHC has a “minimal” shear viscosity η/s=1/(4π).
Covariant transport theory is a nonequilibrium framework with two main advantages: i)
it has a hydrodynamic limit (i.e., capable of thermalization); and ii) it is always causal and
stable. In contrast, hydrodynamics (whether ideal, Navier-Stokes, or second-order Israel-
Stewart theory [138]) shows instabilities and acausal behavior in certain, potentially large,
regions of the hydrodynamic “phase space”.
We consider here Lorentz-covariant, on-shell Boltzmann transport theory, with a 2 →2
rate [54,139]
pµ
1∂µf1=S(x, ~p1)+1
πZ2Z3Z4(f3f4−f1f2)W12→34 δ4(p1+p2−p3−p4)
The integrals are shorthands for Ri≡Rd3pi/(2Ei). For dilute systems, fis the phase space
distribution of quasi-particles, while the transition probability W=s(s−4m2)dσ/dtis given
by the scattering matrix element. Our interest here, on the other hand, is to study the theory
near its hydrodynamic (local equilibrium) limit.
Near local equilibrium, the transport evolution can be characterized via transport
coefficients of shear and bulk viscosities (η, ζ) and heat conductivity (λ) that are determined by
the differential cross section. For the massless dynamics (ǫ=3pequation of state) considered
here η≈0.8T/σtr,ζ=0, and λ≈1.3/σtr,τπ≈1.2λtr [138,140] (σtr and λtr are the transport
cross section and mean free path, respectively).
Minimal viscosity and elliptic flow. Finite cross sections lead to dissipative effects that
reduce elliptic flow [141, 142]. For a system near thermal and chemical equilibrium
undergoing longitudinal Bjorken expansion, T∼τ−1/3,s≈4n∼T3, and thus η/s=const
requires a growing σtr ∼τ2/3. With 2 →2 processes chemical equilibrium is broken, therefore
σtr also depends on the density through µ/T∼lnn(because s=4(n−µ/T)). We ignore this
weak logarithm and take σtr(τ)=σ0,tr(τ/0.1 fm)2/3with σ0,tr large enough to ensure that most
of the system is at, or below, the viscosity bound (thus we somewhat underestimate viscous
effects, i.e., overestimate v2(pT)).
For AA at b=8 fm impact parameter we use the class of initial conditions in [139] that
has three parameters: parton density dN/dη, formation time τ0, and effective temperature T0
that sets the momentum scale. Because of scalings of the transport solutions [139], v2(pT/T0)
only depends on two combinations σtr dN/dη∼A⊥τ0/λtr and τ0. This may look worrisome
because dN/dηat the LHC is uncertain by at least a factor of two. However, the “minimal
viscosity” requirement Tλtr ≈0.5fixes σtrdN/dη(e.g., with dN/dη(b=0) =1000 at RHIC,
σ0,tr ≈2.7 mb), while on dimensional grounds τ0∼1/T0.

Heavy Ion Collisions at the LHC - Last Call for Predictions 62
This means that the main difference between LHC and RHIC is in the typical momentum
scale T0(gold and lead nuclei are basically identical), and therefore to good approximation
one expects the simple scaling vLHC
2(pT)≈vRHIC
2(pTTLHC
0/TRHIC
0). From gluon saturation
physics we estimate r≡TLHC
0/TRHIC
0≈1.3−1.5 at b=8 fm via Gribov-Levin-Ryshkin
formula as applied in [143] (we take Teff∼Qs∼qhp2
Ti).
As depicted in figure 54, at a given pTthe scaling predicts a striking reduction of v2(pT)
at the LHC relative to RHIC. This is the opposite of both ideal hydrodynamic expectations
and what was seen going from SPS to RHIC (where v2(pT) increased slightly with energy).
Experimental determination of the scaling factor r≡QLHC
s/QRHIC
swould provide a further
test of gluon saturation models.
Pb+Pb, min. visc
Au+Au, min. visc
τ0= 0.6 fm, b= 8 fm
QLHC
s/QRHI C
s= 1.5
pT[GeV]
v2
3210
0.3
0.2
0.1
0
Figure 54: Differential elliptic flow at RHIC and the LHC, assuming a “minimally viscous”
quark-gluon system η/s=1/(4π) at both energies.
We note that higher momenta at the LHC would also imply somewhat earlier
thermalization τ0∼1/T0. This is expected to prolong longitudinal Bjorken cooling at the
LHC, changing the scale factor in v2(pT) from rtowards r1−1/3=r2/3≈1.2−1.3.
3. Hadronic flavor observables
3.1. Thermal model predictions of hadron ratios
A. Andronic, P. Braun-Munzinger and J. Stachel
We present predictions of the thermal model for hadron ratios in central Pb+Pb collisions at LHC.
Based on the latest analysis within the thermal model of the hadron yields in central
nucleus-nucleus collisions [144], the expected values at LHC for the chemical freeze-
out temperature and baryochemical potential are T=161±4 MeV and µb=0.8+1.2
−0.6MeV,
respectively. For these values, the thermal model predictions for hadron yield ratios in central
Pb+Pb collisions at LHC are shown in Table 3. We have assumed no contribution of weak
decays to the yield of pions, kaons and protons.
The antiparticle/particle ratios are all very close to unity, with the exception of the ratio
¯p/p, reflecting the expected small, but nonzero, µbvalue. The errors are determined by the

Heavy Ion Collisions at the LHC - Last Call for Predictions 63
Table 3: Predictions of the thermal model for hadron ratios in central Pb+Pb collisions at
LHC. The numbers in parantheses represent the error in the last digit(s) of the calculated
ratios.
π−/π+K−/K+¯p/p¯
Λ/Λ¯
Ξ/Ξ¯
Ω/Ω
1.001(0) 0.993(4) 0.948−0.013
+0.008 0.997−0.011
+0.004 1.005−0.007
+0.001 1.013(4)
p/π+K+/π+K−/π−Λ/π−Ξ−/π−Ω−/π−
0.074(6) 0.180(0) 0.179(1) 0.040(4) 0.0058(6) 0.00101(15)
errors of µbin case of antiparticle/particle ratios and by the errors of Tfor all other ratios.
Table 4: Predictions for the relative abundance of resonances at chemical freeze-out.
φ/K−K∗0/K0
S∆++/pΣ(1385)+/Λ Λ∗/Λ Ξ(1530)0/Ξ−
0.137(5) 0.318(9) 0.216(2) 0.140(2) 0.075(3) 0.396(7)
Assuming that the yield of resonances is fixed at chemical freeze-out, we show in Table 4
predictions for the relative yield of various resonance species. We emphasize that the above
hypothesis needs to be checked at LHC, in viewof the data at RHIC [145], which may indicate
rescattering and regeneration of resonances after chemical freeze-out.
3.2. (Multi)Strangeness Production in Pb+Pb collisions at LHC. HIJING/B¯
B v2.0
predictions.
V. Topor Pop, J. Barrette, C. Gale, S. Jeon and M. Gyulassy
Strangeness and multi-strangeness particles production can be used to explore the initial transient
field fluctuations in heavy ion collisions. We emphasize the role played by Junction anti-Junction (J¯
J)
loops and strong color electric fields (SCF) in these collisions. Transient field fluctuations of SCF on
the baryon production in central (0-5 %) Pb+Pb collisions at √sNN =5.5 TeV will be discussed in
the framework of HIJING/B¯
B v2.0 model, looking in particular to the predicted evolution of nuclear
modification factors (RAA) from RHICto LHC energies. Our results indicate the importance of a good
description of the baseline elementary p+pcollisions at this energy.
In previous publications [146] we studied the possible role of topological baryon
junctions [25], and the effects of strong color field (SCF) in nucleus-nucleus collisions at
RHIC energies. We have shown that the dynamics of the production process can deviate
considerably from that based on Schwinger-like estimates for homogeneous and constant
color fields. An increase of the string tension from κ0=1 GeV/fm, to in medium mean

Heavy Ion Collisions at the LHC - Last Call for Predictions 64
values of 1.5-2.0 GeV/fm and 2.0-3.0 GeV/fm, for d+Au and Au+Au respectively, results
in a consistent description of the observed nuclear modification factors (NMF) RAA in both
reactions and point to the relevance of fluctuations on transient color fields. The model
provides also an explanation of the baryon/meson anomaly, and is an alternative dynamical
description of the data to recombination models [147].
Strangeness enhancement [148], strong baryon transport, and increase of intrinsic
transverse momenta kT[149] are all expected consequences of SCF. These are modeled in
our microscopic models as an increase of the effective string tension that controls the quark-
anti-quark (q¯q) and diquark - anti-diquark (qqqq) pair creation rates and the strangeness
suppression factors. A reduction of the strange (s) quark mass from Ms=350 MeV to the
current quark mass of approximately ms=150 MeV, gives a strangeness suppression factor
γ1
s≈0.70. A similar value of γ1
s(0.69) is obtained by increasing the string tension from
κ0=1.0 GeV/fm to κ=3.0 GeV/fm [146]. Howeover, if we consider that Schwinger tunneling
could explain the thermal character of hadron spectra we can define an apparent temperature
as function of the average value of string tension (< κ >), T=√3< κ > /4π[150]. The
predictions at LHC for initial energy density and temperature are ǫLHC ≈200 GeV/fm3and
TLHC ≈500 MeV, respectively [151]. Both values would lead in the framework of our model
to an estimated increase of the average value of string tension to κ≈5.0 GeV/fm at LHC
energy.
Figure 55: HIJING/B¯
B v2.0 predictions including SCF effects for NMF of identified particles.
The results for proton and lambda particles are for inclusive measurements.
The p+pcross sections serve as a baseline reference to calculate NMF for A+Acollisions
(RAA). In p+pcollisions high baryon/mesons ratio (i.e. close to unity) at intermediate pT
were reported at √sNN=1.8 TeV [152]. These data could be fitted assuming a string tension
κ=2.0 GeV/fm. This value is used in our calculations at √sNN=5.5 TeV. This stresses the
need for a reference p+pmeasurements at LHC energies.
The predictions for NMF RPbPb of identified particles at the LHC energy are presented in
Fig. 55 for two values of the string tension. From our model we conclude that baryon/meson
anomaly, will persist at the LHC with a slight increase for increasing strength of the

Heavy Ion Collisions at the LHC - Last Call for Predictions 65
chromoelectric field. The NMF RPbPb also exhibit an ordering with strangeness content at
low and intermediate pT. The increase of the yield being higher for multi-strange hyperons
than for (non)strange hyperons (RPbPb(Ω)>RPbPb(Ξ)>RPbPb(Λ)>RPbPb(p) ). At high
pT>4GeV/cfor κ=3.0 GeV/fm, a suppression independent of flavours is predicted due to
quench effects. In contrast, this independence seems to happen at pT>8 GeV/c for κ=5.0
GeV/fm.
As expected, a higher sensitivity to SCF effects on the pTdependence of multi-strange
particle yield ratio is predicted. As an example, Fig. 56 presents our results for the ratio
(Ω−+ Ω+)/Φ in central (0-5%) Pb+Pb collisions and p+pcollisions. The results and data at
RHIC top energy are also included (left panel).
Figure 56: Predictions of HIJING/B¯
B v2.0 for the (Ω++ Ω−)/Φratio as function of pT
for RHIC (left panel) and LHC (right panel) energies. The experimental data are from
STAR [153].
The mechanisms of (multi)strange particles production is very sensitive to the early phase
of nuclear collisions, when fluctuation in the color field strength are highest. Their mid-
rapidity yield favors a large value of the average string tension as shown at RHIC and we
expect similar dynamical effects at LHC energy. The precision of these predictions depens on
our knowledge of initial conditions, parton distribution functions at low Bjorken-x, the values
of the scale parameter p0, constituent and current (di)quark masses, energy loss for gluon and
quark jets.
3.3. Antibaryon to Baryon Production Ratios in Pb-Pb and p-p collision at LHC energies of
the DPMJET-III Monte Carlo
F. Bopp, R. Engel, J. Ranft and S. Roesler
A sizable component of stopped baryons is predicted for pp and PbPb collisions at LHC. Based
on an analysis of RHIC data within framework of our multichain Monte Carlo DPMJET-III the LHC
predictions are presented.
This addendum to Ranft’s talk about the main DPMJET III prediction addresses

Heavy Ion Collisions at the LHC - Last Call for Predictions 66
baryon stopping. The interest is a component without leading quarks. Where the flavor
decomposition is not determined by final state interactions the valence-quarkless component
can be enhanced by considering net strange baryons.
In models, in which soft gluons can arbitrarily arrange colors, a configuration can appear
in which the baryonic charge ends up moved to the center. The actual transport is just an
effect of the orientation of the color-compensation during the soft hadronisation. Various
other ideas about fast baryon stopping exist but to have it caused by such an “initial” process
is an attractive option.
The “Dual-Topological” phenomenology of such baryon transport processes was
developed 30 years ago [154]. Critical are various baryonium-exchange intercepts which
were estimated at that time. Some ambiguity remains until today for the quarkless compo-
nent (also called “string junction" exchange denoted as {S J}) and a confirmation of the flat
net-baryon distribution indicated by RHIC data at LHC would be helpful.
Nowadays it is postulated that at very high energy hadronic scattering can be understood
as extrapolation of BFKL Pomeron exchanges [155] and their condensates in the minimum
bias region. BFKL Pomerons are described by ladders of dispersion graphs, in which soft
effects are included using effective gluons. In principle these soft effects include the color
compensating mechanism usually modelled as two strings neutralizing triplet colors. A
necessary ingredient in this approach are Odderons exchanges with Pomeron-like intercepts
and with presumably much smaller couplings. As these Odderons can produce a baryon
exchange of the type discussed above, a small rather flat net baryon component is expected.
Experimentally, the first indication for a flat component came from never finalized
preliminary ZEUS data at HERA. As RHIC runs pp or heavy ions instead of p¯pthis question
could be addressed much better than before and the data seem torequire a flat contribution. In
a factorizing Quark-Gluon-String model calculation [156] the best fit to RHIC BRAHMS pp
data at √s=200 GeV required diquarks with a probability of ǫ=0.024 to involve a quarkless
baryonium-exchange with an intercept α{S J}=0.9.
To obtain such a quarkless baryonium-exchange in the microscopic generator DPMJET
III [157] a new string interaction reshuffling the initial strings was introduced. It introduces
an exchange with a conservative intercept of α{S J}=0.5. With this baryonium addition
good fits were obtained for various baryon ratios in p−pand d−Au RHIC and π−p
FERMILAB processes [158]. There are of course a number of more conventional baryon
transport mechanisms implemented in the model. As the string interaction requires multiple
Pomeron exchanges the new mechanism is actually only a 10% effect at pp RHIC. It is,
however, important for heavy ion scattering or at LHC energies.

Heavy Ion Collisions at the LHC - Last Call for Predictions 67
For pp LHC the DPMJET III
prediction for the pseudo rapidity
of p, ¯p, and p−¯pis shown in the
Figure on the right. The new
baryon stopping is now a 40%
effect.Of course, with the effective
intercept of 0.5 the present
implementation of the baryon
stopping is a rather conservative
estimate. For an intercept of 1.0 the
value at η=0 would roughly
correspond to the present value of
η=4
We now turn to DPM-
JET III prediction for
central PbPb LHC. For
the most central 10%
of the heavy ion events
the pseudorapidity pro-
ton and Λdistributions
are given in figures be-
low. The PbPb results
are preliminary, as the
model is not well tested
in this region.

Heavy Ion Collisions at the LHC - Last Call for Predictions 68
3.4. Statistical model predictions for pp and Pb-Pb collisions at LHC
I. Kraus, J. Cleymans, H. Oeschler, K. Redlich and S. Wheaton
Predictions for particle production at LHC are discussed in the context of the statistical model.
Moreover, the capability of particle ratios to determine the freeze-out point experimentally is studied,
and the best suited ratios are specified. Finally, canonical suppression in p-p collisions at LHC energies
is discussed in a cluster framework. Measurements with pp collisions will allow us to estimate the
strangeness correlation volume and to study its evolution over a large range of incident energies.
Particle production in heavy-ion collisions is, over a wide energy range, consistent with
the assumption that hadrons originate from a thermal source with a given temperature Tand
a given baryon chemical potential µB. In the framework of the statistical model, we exploit
the feature that the freeze-out points appear on a common curve in the T−µBplane. The
parameterization of this curve, taken from reference [159], is used to extrapolate to the LHC
energy of √sNN =5.5 TeV: T≈170 MeV, µB≈1 MeV.
For the given thermal conditions, particle ratios in central Pb-Pb collisions were
calculated; numerical values are given in reference [160]. As soon as experimental results
become available, the extrapolation can be cross-checked with particle ratios that exhibit
a large sensitivity to the thermal parameters. The ratios shown in figure 57 (left) hardly
vary over a broad range of Tand µB. This feature can be used to investigate the validity
of the statistical model at LHC: Especially the prediction for the K/π ratio is limited to a
narrow range. It would be hard to reconcile experimental results outside of this band with the
statistical model.
Antiparticle over particle ratios, on the other hand, strongly depend on µB(figure 57,
middle panel). Most of all, the ¯p/pratio almost directly translates to the baryon chemical
potential, since the Tdependence is very weak. Better suited for the temperature
determination are ratios with large mass differences, i.e. Ω/π and Ω/K, which increase in
the studied range by 25% per 10 MeV change in T. The astonishing similarity between Kand
πin this respect is caused by the huge contribution of 75% from resonance decays to pions
for the given thermal conditions [161].
In collisions of smaller systems, the strange-particle phase-space exhibits a suppression
beyond the expected canonical suppression. A modification of the statistical model is
proposed in references [162, 163], which is based on the assumption that strangeness
conservation is maintained in correlated sub-volumes of the fireball. The size of these clusters,
which could be smaller than the volume defined by all hadrons, was estimated from relative
strangeness production in collisions of small systems at top SPS and RHIC energy. The
radius RCof a spherical cluster is of the order of 1 - 2 fm and shows only a weak energy
dependence. Additionally it is not clear at which stage of the interaction the strangeness
abundance is formed. Possibly the early, dense phase is crucial, so the cluster size should be
the same at RHIC and LHC, or, on the contrary, the total number of particles at the late stage
of hadronisation is relevant; thus RCshould increase as the multiplicity will increase with
colliding energy.

Heavy Ion Collisions at the LHC - Last Call for Predictions 69
(MeV)
B
µ
0 20
R
0
0.2
T = 170 MeV
S = 1∆
T (MeV)
160 180
+
π /
+
K -
π /
-
K -
Ξ /
-
Ω Λ /
-
Ξ 0.3∗ / p Λ
= 1 MeV
B
µ
(MeV)
B
µ
0 20
R
0.8
1
+
/ K
-
K -
Ω /
+
Ω
-
Ξ /
+
Ξ
Λ / Λ / pp
T = 170 MeV a
T (MeV)
160 180 200
R
0
0.02
-
/ K
-
Ω 5
-
π /
-
Ω 0.5
-
π /
-
Ξ 0.2 / p
-
Ω
= 1 MeV
B
µb
-
π/
+
π-
/K
+
K / pp Λ/Λ-
Ξ/
+
Ξ-
Ω/
+
Ω-
πp/ +
π/
+
K-
π/
-
K / pΛΛ/
-
Ξ-
Ξ/
-
Ω-
/K
-
Ω-
π/
-
Ω
-
π/
+
π-
/K
+
K / pp Λ/Λ-
Ξ/
+
Ξ-
Ω/
+
Ω-
πp/ +
π/
+
K-
π/
-
K / pΛΛ/
-
Ξ-
Ξ/
-
Ω-
/K
-
Ω-
π/
-
Ω
-6
10
-4
10
-2
10
1
PbPb
Grand Canonical
=2 fm
C
, R
pp
Canonical
=1.5 fm
C
, R
pp
Canonical
=1 fm
C
, R
pp
Canonical
=0.75 fm
C
, R
pp
Canonical
Figure 57: Left: Ratios Rof particles with unequal strangeness content as a function of µBfor
T=170 MeV (left) and as a function of Tfor µB=1 MeV (right).
Middle: Antiparticle/particle ratios Ras a function of µBfor T=170 MeV (left) (the
horizontal line at 1 is meant to guide the eye). Particle ratios Rinvolving hyperons as a
function of Tfor µB=1 MeV (right).
Right: Ratios Rof particles in the grand-canonical ensemble and with suppressed strange-
particle phase-space in different canonical volumes indicated by the spherical radius RC,
calculated at µB=1 MeV and T=170 MeV.
Instead of precise predictions as shown for Pb-Pb collisions, the correlation volume will
be extracted from measurement. As displayed in figure 57 (right), especially the Ω/π ratio
varies over orders of magnitude in a reasonable range of the correlation length. This allows
for a good estimate of the cluster size which will give us more insight into the mechanism of
strangeness production.
3.5. Universal behavior of baryons and mesons’ transverse momentum distributions in the
framework of percolation of strings
L. Cunqueiro, J. Dias de Deus, E. G. Ferreiro and C. Pajares
The clustering of color sources [65] reduces the average multiplicity and enhances
the average hpTiof an event in a factor F(η) with respect to those resulting from pure
superposition of strings:
hµi=NsF(η)hµi1,hp2
Ti=hp2
Ti1/F(η) (21)
where Nsis the number of strings and F(η)=p(1−e−η)/η is a function of the density
of strings η[164]. The invariant cross section can be written as a superposition of the
transverse momentum distributions of each cluster, f(x,pT) (Schwinger formula for the decay
of a cluster), weighted with the distribution of the different tension of the clusters, W(x)
(W(x) is the gamma function whose width is proportional to 1/kwhere kis a determined
function of ηrelated to the measured dynamical transverse momentum and multiplicity

Heavy Ion Collisions at the LHC - Last Call for Predictions 70
pt
0 1 2 3 4 5
CP
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
pt
0 1 2 3 4 5 6
0
π/p/
0
0.2
0.4
0.6
0.8
1
Figure 58: Left: RCP for neutral pions (solid) and antiprotons (dashed). Right: ¯pto π0ratio
for the centrality bins 0-10% (solid) and 60-92% (dashed). RHIC results in black and LHC
predictions in blue.
fluctuations) [67,165–167]:
dN
dp2
Tdy=Z∞
0dxW(x)f(pT,x)=dN
dyk−1
k1
hp2
Ti1iF(η)1
(1+F(η)p2
T
khp2
Ti1i)k.(22)
For (anti)baryons, equation (21) must be changed to hµ¯
Bi=N1+α
sF(η¯
B)hµ1¯
Bito take
into account that baryons are enhanced over mesons in the fragmentation of a high density
cluster. The parameter α=0.09 is fixed from the experimental dependence of ¯p/π on Npart.
The (anti)baryons probe higher densities than mesons, ηB=Nα
sη. On the other hand, from
the constituent counting rules applied to the high pTbehavior we deduce that for baryons
kB=k(ηB)+1. In figure 58, we show the ratios RCP and ¯p/π0defined as usual, compared
to RHIC experimental data for pions and antiprotons together with the LHC predictions. In
figure 59 left we show the nuclear modification factor RAA for pions and protons for central
collisions at RHIC. LHC predictions are also shown. We note that pp collisions at LHC
energies will reach enough string density for nuclear-like effects to occur. In this respect, in
figure 59 right, we show the ratio RCP for pp →πXas a function of pT, where the denominator
is given by the minimum bias inclusive cross section and the numerator is the inclusive cross
section corresponding to events with twice multiplicity than minimum bias. According to our
formula (22) a suppression at large pToccurs.
3.6. Bulk hadron(ratio)s at the LHC-ions
J. Rafelski and J. Letessier
The expected LHC-heavy ion yields of strange and non-strange hadrons, mesons and baryons,
are evaluated within the statistical hadronization model.
This summary of our recent work on bulk hadronization in LHC-ion interactions is based
on methods and ideas presented in [168], with the present update using the results obtained

Heavy Ion Collisions at the LHC - Last Call for Predictions 71
qt
0 2 4 6 8 10 12 14 16 18 20 22
AA(qt)
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0-10% CENTRAL RHIC
0-10% CENTRAL LHC
solid/dashed=pions/protons
pt
0 2 4 6 8 10
in p-p collisions
CP
R
0.6
0.7
0.8
0.9
1
1.1
1.2
Figure 59: Left: Nuclear modification factor for π0(solid) and ¯p(dashed) for 0-10% central
events, RHIC results in black and LHC predictions in blue. Right: RCP for pions in pp
collisions at LHC.
for strangeness production in [169]. This presentation is more specific regarding the yields
in order to allow “first-day" understanding of the mechanisms of hadronization dynamics of
the deconfined quark–gluon plasma phase formed in most central √sNN =5520 GeV Pb–Pb
reactions at the LHC.
Our detailed results rely on SHARE-2.2 suite of programs [170], which have been
extensively tested, with several typos, and errors corrected compared to earlier releases
SHARE-1.x [171], and SHARE-2.1 . An important feature of the SHARE suite of
programs is that one can obtain the particle multiplicities for any consistent mixed set of
extensive/intensive bulk matter parameters and/or particle yields. What ‘consistency’ means
can be understood considering the variables in the Gibbs-Duham relation:
PV +E=TS +X
a
µaNa, µa=X
i∈aTlnγi+biµB+siµS,(23)
where the extensive V(volume), E≡Vǫ(energy), S≡Vσ(entropy), Ni≡Vρi(particle
number) appears along with intensive P(pressure), T(temperature), and µi(particle chemical
potential). Aside of the above strict constraint, other qualitative constraints arise and thus, in
our approach, we allow for a deviation from prescribed parameter values within a margin of
a few percent, to be chosen in a quasi-fit procedure in order to alleviate inconsistencies in the
choices made.
Considering the limited central rapidity experimental coverage, we refer instead of
the total volume Vto the range associated with the central rapidity dV/dy, thus dS/dy =
(dS/dV)(dV/dy) is the entropy (multiplicity) yield per unit of rapidity. One can show that
dS/dy is conserved in the hydrodynamic expansion of the bulk matter, thus the final observed
entropy (multiplicity) content per unit of rapidity is the outcome of the initial state entropy
production.
The soft hadron production at LHC-Ion relies on the following input:
IThe entropy content: dS/dy ≡hadron multiplicity — this is normalizer of all particle

Heavy Ion Collisions at the LHC - Last Call for Predictions 72
yields for which the predictions most widely vary. The straight line extrapolation as function
of ln √sNN implies an increase of dS/dy by only a factor 1.65 from RHIC top energy reach
√sNN =200 GeV to the LHC-ion top energy of √sNN =5520 GeV. The charged particle
yield per unit rapidity is expected, in this case, at about hch =1150. Since this extrapolation
is done based on PHOBOS multiplicity, only partial Ksweak interaction decay is allowed
for. We will also state the corresponding hvis
ch which is computed assuming acceptance of
weak decays akin to the STAR detector. The entropy content determines up to about 15% the
energy content dE/dy ≃ThdS/dy which thus increases, in essence, by the same factor. We
note that model differences in hadronization temperature Thwhich are in the range of up to
20% impact accordingly the thermal energy content.
However, one can wonder if the factor 1.65 correctly accounts for the 28-fold reaction
energy increase between RHIC and LHC-ion. The widening of the particle production rapidity
window accounts for much of the collision energy increase. How this widening occurs i.e the
strength of stopping, determines the central rapidity energy deposition. We thus consider in
the second example the case with 3.4-fold increase in entropy/multiplicity content per unit
of rapidity. This value is fine-tuned such that the visible charged hadron yield is identical to
the TPC-visible charged hadron multiplicity yield in the chemical equilibrium model, where
the hadronization volume was set to be V=6200fm3(our 3rd table entry). This allows to
compare the yields of both models normalized to same hadron yield.
II The strangeness content ds/dy =d¯s/dy and/or (ds/dy)/(dS/dy)=s/S. The production
of strangeness has been evaluated within pQCD, for a given entropy content. The final
strangeness yield does not depend in a significant way on how the parton entropy content
is implemented in the early reaction times where thermal distributions are reached (e.g., high
T, low chemical abundances, low T, high chemical abundances). This is so, since strangeness,
being a relatively strongly interacting probe, does not convey a detailed information about the
early τ < 2fm/c times of the heavy ion collision. For the case of a greater (3.4-times increased)
entropy/multiplicity content, the pQCD computation suggests s/S≃0.037 yield, which is
10–15% above QGP chemical equilibrium, the lower entropy variant (extrapolated factor
1.65 increase in multiplicity) implies for QGP-strangeness a small excess above chemical
equilibrium, we will use s/S≃0.034. For the third case, the hadron chemical equilibrium,
the ratio s/S=0.025 results. Thus, strangeness enhancement, where it is not washed out by
a lower hadronization temperature, is the salient feature of the non-equilibrium hadronization
picture we have developed and present here.
III The net baryon stopping d(b−¯
b)/dy is unknown, and will be difficult to measure. An
extrapolation of the energy per baryon retained per unit of rapidity yields E/b≃412±20 GeV
at LHC. This value is consistent with the here considered two cases, when, as an example,
we fix λq=1.0056 which determines the baryon and hyperon chemical potentials µBand µS.
We note, in passing, that in all the cases considered here, we find for the baryon asymmetry
at LHC (b−¯
b)/(b+¯
b)≃0.015, which is 6–7 orders of magnitude larger compared to the
conditions prevailing in the early universe.
There are constrains which we use to fully determine the system properties:
1) For the chemical non-equilibrium hadronization we will use Th=140 MeV while for

Heavy Ion Collisions at the LHC - Last Call for Predictions 73
chemical equilibrium we adopt Th=162 MeV. Both values are taken from the study of
highest RHIC energies. The lower Tarises due to supercooling expansion, leading to sudden
hadronization [172], and thus, we also impose a bias for E/TS >1.
2) Strangeness balance hsi=h¯siin the central unit of rapidity.
3) Net charge per net baryon ratio Q/b=0.4 (value in colliding nuclei) is implemented. Since
the net baryon number is rather small, the charge asymmetry is for all purposes invisible, the
purpose of this exercise is to assure physical consistency and to fix the isospin asymmetry
statistical parameter λ3.
Our results are presented in detail in the table. We note that the total charged hadron
multiplicity will be a first-day observable at LHC and hence much of the uncertainty we
have in discussing the absolute hadron yields will disappear. When comparing hadronization
models at fixed total hadron yield one sees clear differences in yield pattern:
a) Multi-strange hadron yields are, in general, greatly enhanced in our non-equilibrium
approach as compared to yields assuming chemical equilibrium hadronization, yet single
strange yields are often similar, since the differences in hadronization (temperature) conditions
compensate for the strangeness yield enhancement;
b) The yields of non-strange resonances are, in general, significantly greater in the chemical
equilibrium model, due to the higher hadronization temperature.
c) This suppression is compensated in resonances with single and partial multi-strange content
(η,η′).
The above differences, already seen at RHIC, are much more striking at LHC, since the
specific strangeness per entropy yield enhancement is by factor 1.5. Even the visible K+/π+
vis
ratio is increased from the RHIC level, to K+/π+
vis ≃0.17 –0.18, however this enhancement
effect is much better visible once weak decays have been vetoed in the pion yield, in which
case, we predict K+/π+≃0.21. While the yield of nucleons may be difficult to determine,
the measurement of baryon resonances such as ∆(1230) could help considerably in the
characterization of the baryon yield.

Heavy Ion Collisions at the LHC - Last Call for Predictions 74
T[MeV] 140∗140∗162∗
dV/dy[ fm3] 2036 4187 6200∗
dS/dy 7517 15262 18021
dhch/dy 1150∗2351 2430
dhvis
ch /dy 1351 2797∗2797
1000·(λq,s−1) 5.6∗,2.1 5.6∗,2.1 5.6∗,2.0
µB,S[MeV] 2.4,0.5 2.3,0.5 2.7,0.6
γq,s1.62,2.42 1.6∗,2.6 1∗,1∗
s/S0.034∗0.037∗0.025
E/b420∗428 408
E/TS 1.02 1.05 0.86
P/E0.165 0.164 0.162
E/V[MeV/fm3] 530 538 400
P[MeV] 87 88 65
p25/45 49/95 66/104
b−¯
b2.6 5.3 6.1
(b+¯
b)/h−0.335 0.345 0.363
0.1·π±49/67 99/126 103/126
K±94 207 175
φ14 33 23
Λ19/28 41/62 37/50
Ξ−4 9.5 5.8
Ω−0.82 2.08 0.98
∆0,∆++ 4.7 9.3 13.7
K∗
0(892) 22 48 52
η62 136 127
η′5.2 11.8 11.5
ρ36 73 113
ω32 64 104
f02.7 5.5 9.7
K+/π+
vis 0.165 0.176 0.148
Ξ−/Λvis 0.145 0.153 0.116
Λ(1520)/Λvis 0.043 0.042 0.060
Ξ(1530)0/Ξ−0.33 0.33 0.36
φ/K+0.15 0.16 0.13
K∗
0(892)/K−0.236 0.234 0.301
Table: LHC predictions, two variants
of our non-equilibrium hadronization
model are shown on left, the chemi-
cal equilibrium model results are stated
for comparison in the right column.
To obtain results n the first column,
we considered an overall hadron yield
chosen to increase at central rapidity
by factor 1.65 compared to PHOBOS
results (star ‘*’ indicates a fixed in-
put value). The chemical equilibrium
model shown on right is matched in
the middle column by assuming a TPC-
visible charged hadron yield to be the
same, 2797. These characteristic prop-
erties along with the entropy content,
and chemical conditions at hadroniza-
tion, are stated in the two top sections
of the table. In the third section, we
show bulk properties at hadronization,
with specific strangeness content pre-
scribed as arising in pQCD computa-
tion [169], except for the equilibrium
model in which case the specific yield
s/Sis a consequence of the equilib-
rium assumption. One notes for the
equilibrium model that the energy den-
sity and pressure at hadronization is
smaller, which agrees with the greater
volume of hadronization required to
obtain the same hadron yield. This is
due to particle density scaling roughly
with γ2
qT3, the change in γqoutweighs
that in T. When we present the hadron
yields, we give (separated by slash)
the ranges with/without weak decays
for protons p,πand Λ. Clearly the
properties of the detector will impact
the uncorrected yields. We also note
that, while baryon density in rapidity
can vary depending on dynamics of
the reaction, the specific total baryon
yield, compared to that of mesons, re-
mains nearly constant and model inde-
pendent. The difference to the equilib-
rium model is most pronounced in the
multi-strange hadron Ξ, ω and φyields.
The ratios or resonances with the stable
decay product are shown in the bottom
section of the table.

Heavy Ion Collisions at the LHC - Last Call for Predictions 75
4. Correlations at low transverse momentum
4.1. Pion spectra and HBT radii at RHIC and LHC
Yu. M. Sinyukov, S. V. Akkelin and Iu. A. Karpenko
We describe RHIC pion data in central A+A collisions and make predictions for LHC based on
hydro-kinetic model, describing continuous 4D particle emission, and initial conditions taken from
Color Glass Condensate (CGC) model.
Hydro-kinetic approach to heavy ion collisions proposed in Ref. [173] accounts for
continuous particle emission from 4D volume of hydrodynamically expanding fireball as
well as back reaction of the emission on the fluid dynamics. The approach is based on the
generalized relaxation time approximation for relativistic finite expanding systems,
pµ
p0
∂f(x,p)
∂xµ=−f(x,p)−fl.eq.(x,p)
τrel(x,p),(24)
where f(x,p) is phase-space distribution function (DF), f(l.eq.)(x,p) is local equilibrium
distribution and τrel(x,p) is relaxation time, τrel(x,p) as well as fl.eq are functional of
hydrodynamic variables. Complete algorithm described in detail in Ref. [174] includes:
solution of equations of ideal hydro; calculation of a non local equilibrium DF and emission
function in the first approximation; solution of equations for ideal hydro with non-zero right-
hand-side that accounts for conservation laws at the particle emission during expansion;
calculation of "improved" DF and emission function; evaluation of spectra and Bose-Einstein
correlations. Here we present our results for the pion momentum spectra and interferometry
radii calculated for RHIC and LHC energies in the first approximation of the hydro-kinetic
approach.
For simulations we utilize ideal fluid model [175–177] and realistic equation of state
(EoS) that combines high temperature EoS with crossover transition [178] adjusted to the
QCD lattice data and EoS of hadron resonance gas with partial chemical equilibrium [175–
177]. The gradual disappearance of pions during the crossover transition to deconfinement
and different intensity of interactions of pions in pure hadronic and "mixed" phases are taken
into account in the hydro-kinetic model (HKM), but resonance contribution to pion spectra
and interferometry radii is not taken into account in the present version of the HKM. We
assume the following initial conditions at proper time τ0=1 fm/c for HKM calculations:
boost-invariance of a system in longitudinal direction and cylindrical symmetry with Gaussian
energy density profile in transverse plane. The maximal energy densities at RHIC, ǫ0=30
GeV/fm3and at LHC, ǫ0=70 GeV/fm3, were calculated from Ref. [179] in approximation of
Bjorken expansion of free ultrarelativistic partons till τ0and adjusted for transverse Gaussian
density profile. The (pre-equilibrium) initial transverse flows at τ0were estimated assuming
again a free-streaming of partons, with transverse modes distributed according to CGC
picture, from proper time ≈0.1 fm/c till τ0=1 fm/c. Finally, we approximate the transverse
velocity profile by vT=tanh(α·rT
RT) where α=0.2 both for RHIC and LHC energies and we
suppose the fitting Gaussian radius for RHIC top energy, RT=4.3, to be the same for LHC

Heavy Ion Collisions at the LHC - Last Call for Predictions 76
[GeV]
T
p
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
]
-2
dy) [GeV
T
dp
T
pπN/(2
2
d
1
10
2
10
PHENIX (RHIC)
3
=30 GeV/fm
0
∈HKM, (LHC)
3
=70 GeV/fm
0
∈HKM, (LHC)
3
=110 GeV/fm
0
∈HKM,
spectraπ
[GeV]
T
p
0.2 0.4 0.6 0.8 1 1.2
R [fm]
3
4
5
6
7
PHENIX (RHIC)
3
=30 GeV/fm
0
∈HKM, (LHC)
3
=70 GeV/fm
0
∈HKM, (LHC)
3
=110 GeV/fm
0
∈HKM,
out
R
[GeV]
T
p
0.2 0.4 0.6 0.8 1 1.2
R [fm]
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5 PHENIX (RHIC)
3
=30 GeV/fm
0
∈HKM, (LHC)
3
=70 GeV/fm
0
∈HKM, (LHC)
3
=110 GeV/fm
0
∈HKM,
side
R
[GeV]
T
p
0.2 0.4 0.6 0.8 1 1.2
R [fm]
2
3
4
5
6
7
8
9
10 PHENIX (RHIC)
3
=30 GeV/fm
0
∈HKM, (LHC)
3
=70 GeV/fm
0
∈HKM, (LHC)
3
=110 GeV/fm
0
∈HKM,
long
R
Figure 60: Comparison of the single-particle momentum spectra of pions and pion Rout,Rside,Rlong radii
measured by the PHENIX Collaboration for Au+Au central collisions (HBT radii data were recalculated for
0−5% centrality) at RHIC with the HKM calculations, and HKM predictions for Pb+Pb central collisions at
LHC. For the sake of convenience the calculated one-particle spectra are enhanced in 1.4 times.
energy. Our results for RHIC and predictions for LHC are presented in Fig. 60. The relatively
small increase of the interferometry radii with energy in HKM calculations is determined by
early (as compare to sharp freeze-out prescription) emission of hadrons, and also by increase
of transverse flow at LHC caused by longer time of expansion. It is noteworthy that in the
case of EoS related to first order phase transition, the satisfactory fitting of the RHIC HBT
data requires non-realistic high initial transverse flows at τ0=1 fm/c: α=0.3.
4.2. Mach Cones at central LHC Collisions via MACE
B. Bäuchle, H. Stöcker and L. P. Csernai
The shape of Mach Cones in central lead on lead collisions at √sNN =5.5 TeV are calculated and
discussed using MACE.
4.2.1. Introduction After the discovery of “non-trivial parts” in three-particle correlations at
RHIC [180], which are compatible with the existence of Mach cones [181], it is interesting to
see how the signal for Mach cones will look like under the influence of a medium created at
the LHC in PbPb-Collisions.
Mach cones caused by ultrarelativistic jets going in midrapidity will create a double-
peaked two-particle correlation function dN/d(∆ϕ). Those peaks are located at ∆ϕ=π±
cos−1cS, where cSis the speed of sound as obtained by the equation of state. The model
MACE (“Mach Cones Evolution”) has been introduced to simulate the propagation of sound

Heavy Ion Collisions at the LHC - Last Call for Predictions 77
0 0 1 1 2 2 3 3 4 4 5 5
center
right
left
(a) (b)
∆ϕ), a.u.dN/d(
∆ϕ
∆ϕ), a.u.dN/d(
∆ϕ
Figure 61: Two-particle-correlation function (away-side-part) for central PbPb-Collisions at
√sNN =5.5 TeV. The peak created by the forward jet is not calculated. (a): minimum jet bias
(see text) with peaks at ∆ϕ≈π±1.2. (b): Midrapidity jets starting from a position 70 % on
the way outside left and right of as well as in the middle.
waves through a medium and recognize and evaluate mach cones [182].
The medium is calculated without influence of a jet using the hydrodynamical Particle-
in-Cell-method (PIC) [183]. For the equation of state, a massless ideal gas is assumed, so
that cS=1/√3 and cos−1cS=0.96. The sound waves are propagated independently of the
propagation of the medium and without solving hydrodynamical equations. Only the velocity
field created by PIC is used. To recognize collective phenomena, the shape of the region
affected by sound waves is evaluated.
4.2.2. Correlation functions The correlation functions from the backward peak show a clear
double-peaked structure. The data for arbitrary jet origin and jet direction (minimum jet bias)
is shown in figure 61 (a). Here, the peaks are visible at ∆ϕ≈π±1.2. This corresponds to a
speed of sound of cS≈0.36. Note that the contributions from the forward jet are not shown.
Deeper insight into different jet directions do not show a qualitatively different picture.
Triggers on the origin of the jet, though, show the dependence ofthe correlation function
∆ϕ1
∆ϕ2
dN2
/ d(∆ϕ1) d(∆ϕ2), a.u.
0 1 2 3 4 5
0
1
2
3
4
5
Figure 62: Three-particle-correlation function for the same data as in figure 61.

Heavy Ion Collisions at the LHC - Last Call for Predictions 78
on the position where the jet was created (see figure 61 (b)). It shows that only the jet coming
from the middle of the reaction results in a symmetric correlation function with peaks at the
mach angle ∆ϕ=π±0.96. All other jets result in correlations that have peaks at different
angles, with the deviation getting bigger when going away from the middle. Therefore, the
speed of sound will always appear to be smaller than it actually is.
4.2.3. Conclusions If sound waves are produced from jet quenching in LHC-Collisions,
the two-particle correlation function will show the expected double-humped structure in the
backward region. The peaks will, though, be further apart than δ(∆ϕ)=2cos−1cS, thus
alluding to a speed of sound smaller than is actually present in the medium.
The only case in which the true speed of sound can be measured is a midrapidity jet that
creates a symmetric correlation function.
4.3. Study of Mach Cones in (3+1)d Ideal Hydrodynamics at LHC Energies
B. Betz, P. Rau, G. Torrieri, D. Rischkeand H. Stöcker
The energy loss of jets created in heavy–ion collisions shows an anomalous behaviour of the
angular distribution of particles created by the away-side jet due to the interaction of the jet with the
medium [184, 185]. Recent three–particle correlations [180, 186,187] confirm that a Mach cone is
created. Ideal (3+1)d hydrodynamics [188] is used to study the creation and propagation of such Mach
cones under LHC conditions.
Jets are one possible probe to study the medium created in a heavy–ion collision. They
are assumed to be formed in an early stage of the collision and to interact with the hot and
dense nuclear matter.
Experimental results from the Relativistic Heavy Ion Collider (RHIC) show a
suppression of the away–side jet in Au+Au collisions for high-p⊥particles as compared to
the away–side jet in p+p collisions. This effect is commonly interpreted as jet energy loss or
jet quenching [184,185]. However, studies including low-p⊥particles [180, 186,187] exhibit
a double peaked away–side jet. Recent three–particle correlations confirm that this pattern is
due to a creation of a Mach cone [180,186,187].
The interaction of a jet with the medium is theoretically not well enough understood.
Therefore, we compare two models of energy loss under LHC conditions. We consider a
medium with an initial radius of 3.5 fm and an initial energy density of e0=1.7 GeV/fm3that
undergoes a Bjorken–like expansion according to a bag–model equation of state (EoS) with a
first–order phase transition from a hadron gas to the quark–gluon plasma (QGP) with a mixed
phase between eH=0.1 GeV/fm3and eQ=1.69 GeV/fm3.
In the first scenario, we implement a jet that completely deposits its energy and
momentum during a very short time in a 0.25 fm3spatial volume. Initially, the jet is located
between −3.5fm <x<−2.5fm,|y|<0.25fm,|z|<0.25fm, has a velocity of vx=0.99 c and
traverses the medium along the x-axis. Totally, it deposits an energy of 15 GeV, no rapidity
cut is applied.

Heavy Ion Collisions at the LHC - Last Call for Predictions 79
In a second step, we study a 15 GeV jet that gradually deposits its energy and momentum
in equal time steps of ∆t=0.8 fm/c. As in the first scenario, the jet traverses the medium with
a velocity of vx=0.99 c along the x–axis.
The hydrodynamic evolution is stopped after a time of 7.2 fm/c. Using a Monte Carlo
simulation based on the SHARE program [171], an isochronous freezeout according to
the Cooper–Frye formula is performed, considering a gas of rhos, pions and etas in the
pseudorapidity interval of [-2.3,2.3].
Figure 63 shows the angular distribution of particles for the first (left panel) and second
(right panel) scenario, without any background subtraction. The omitted near–side jet would
appear at φ=0.
In case of a short–time energy and momentum deposition, a broad away–side distribution
(left panel) occurs, due to the deposition and dissipation of kinetic energy caused by the
jet. However, if the jet gradually dispenses its energy and momentum (right panel), two
maxima appear. This Mach cone–like structure agrees with the recent STAR and PHENIX
data [180,186,187].
4.4. Forward-Backward (F-B) rapidity correlations in a two step scenario
J. Dias de Deus and J. G. Milhano
We argue that in models where particles are produced in two steps, formation first of longitudinal
sources (glasma and string models), followed by local emission, the Forward-Backward correlation
parameter bmust have the structure b=(hnBi/hnFi)/(1+K/hnFi) where hnBi(hnFi) is the multiplicity in
the backward (forward) rapidity window and 1/Kis the (centrality and energy dependent) normalized
variance of the number of sources.
Two-step scenario models for particle production are based on:
30
32
34
36
38
40
42
44
46
48
50
2π3π/2ππ/20
dN/dφ
φ [rad]
34
35
36
37
38
39
40
41
42
2π3π/2ππ/20
dN/dφ
φ [rad]
Figure 63: Angular distribution of particles after isochronous freeze–out of a (3+1)d ideal
hydrodynamical evolution for a jet that deposits its energy and momentum a) completely
within a very short time (left panel) b) in equal timesteps (right panel) in a medium that
undergoes a Bjorken–like expansion according to a bag–model eos.

Heavy Ion Collisions at the LHC - Last Call for Predictions 80
(i) creation of extended objects in rapidity (glasma longitudinal colour fields or coloured
strings); followed by
(ii) local emission of particles.
The first step guarantees the presence of F-B correlations due to fluctuations in the
colour/number of sources, while the second step accounts for local effects such as resonances.
The F-B correlation parameter bis defined via
hnBiF=a+bnF,b≡D2
FB/D2
FF ,(25)
where D2is the variance. In general, correlations are measured in two rapidity windows
separated by a rapidity gap so that F-B short range correlations are eliminated. In the two-
step scenario models we write [189–192],
D2
FB ≡ hnFnBi−hnFihnBi=hnFihnBi
K,(26)
D2
FF ≡ hn2
Fi−hnFi2=hnFi2
K+hnFi,(27)
where 1/Kis the normalized — e.g., in the number of elementary collisions — long range
fluctuation and depends on centrality, energy and rapidity length of the windows. We have
assumed, for simplicity, that local emission is of Poisson type.
From equations (25, 26, 27) we obtain
b=hnBi/hnFi
1+K/hnFi.(28)
It should be noticed that bmay be larger than 1, and that a Colour Glass Condensate (CGC)
model calculation [193] shows a structure similar to (28): b=A[1+B]−1(for a discussion on
general properties of (28) and on the CGC model, see [192]).
A simple way of testing (28) is by fixing the backward rapidity window, or hnBi, in the
region of high particle density and move the forward window along the rapidity axis. We can
rewrite equation (28) in the form
b=x
1+K′x,(29)
where K′≡K/hnBiis a constant and x≡ hnBi/hnFi. In (29), one has 1 <x<∞with the
limiting behaviour:
x→1,b→1
1+K′;x→ ∞,b→1
K′.(30)
The behaviour of (29) is shown in figure 64 (drawn for K′=1).
A similar curve is obtained for B-F correlations in the backward region of rapidity. Note
that in aA collisions, a≤A, the centrality and energy dependence of K′is given by [2, 191],
K′∼a1/2A−1/6eλY,
where Yis the beam rapidity and λa positive parameter. In the symmetric situation, a=Aand
K′increases with centrality (and the curve of the figure moves down) while in the asymmetric
situation, a=1,2≪Aand K′decreases with centrality (and the curve in the figure moves
up). As the energy increases K′increases (and the curve moves down).

Heavy Ion Collisions at the LHC - Last Call for Predictions 81
x
b
0.5
1
=1
K’
1
2
1
=
1+K’
1
Figure 64: F-B correlation parameter b(29) with K′=1.
4.5. Cherenkov rings of hadrons
I. M. Dremin
The ring-like structure of inelastic events in heavy ion collisions becomes pronounced when the
condition for the emission of Cherenkov gluons is fulfilled.
In heavy ion collisions any parton can emit a gluon. On its way through the nuclear
medium the gluon collides with some internal modes. Therefore it affects the medium as
an “effective” wave which accounts also for the waves emitted by other scattering centers.
Beside incoherent scattering, there are processes which can be described as the refraction of
the initial wave along the path of the coherent wave. The Cherenkov effect is the induced
coherent radiation by a set of scattering centers placed on the way of propagation of the
gluon. Considered first for events at very high energies [194,195], the idea about Cherenkov
gluons was extended to resonance production [196,197]. The refractive index and the forward
scattering amplitude F(E,0o) at energy E=√sare related as
∆n=Ren−1=8πNsReF(E,0o)
E2.(31)
Nsis the density of the scattering centers in the medium.
The necessary condition for Cherenkov radiation is
∆n>0 or ReF(E,0o)>0.(32)
If these inequalities are satisfied, Cherenkov gluons are emitted along the conewith half-angle
θcin the rest system of the medium determined by n:
cosθc=1
n(33)
Prediction The rings of hadrons similar to usual Cherenkov rings of photons can be observed
in the plane perpendicular to the cone (jet) axis if n>1.

Heavy Ion Collisions at the LHC - Last Call for Predictions 82
Proposal Plot the one-dimensional pseudorapidity (η=−lntanθ/2) hadron distribution with
trigger jet momentum as z-axis. It should have maximum at (33).
This is the best possible one-dimensional projection of the ring. To define the refractive
index in the absence of the theory of nuclear media (for a simplified approach see [198]) I
prefer to rely on our knowledge of hadronic reactions. From experiments at comparatively
low energies we learn that the resonances are abundantly produced. They are described by
the Breit-Wigner amplitudes which have a common feature of the positive real part in the
low-mass wing (e.g., see Feynman lectures). Therefore the hadronic refractive index exceeds
1 in these energy regions.
At high energies the experiment and dispersion relations indicate on positive real parts
of amplitudes for all hadronic reactions above a very high threshold. Considering gluons as
carriers of strong forces one can assumethat the similar features are typical for their amplitude
as well. Then one should await for two energy regions in which Cherenkov gluons play a role.
Those are either gluons with energies which fit the left wings of resonances produced in their
collisions with internal modes of the medium or with very high energies over some threshold.
The indications on “low” energy effects come from RHIC [187] where the two-bump
structure of the angular distribution of hadrons belonging to the so-called companion (away-
side) jet in central heavy-ion collisions has been observed. It arises as the projection of a
ring on its diameter and provides important information on the properties of the nuclear
medium [196, 197]. From the distance between peaks the cone half-angle is found to be
about 60o−70oin the c.m.s. which is equivalent to the target rest system for the trigger at
central rapidities. Derived from it and Eq.(33) are the large refractive index (n≈3) and parton
density (ν≈20 within a nucleon volume) that favor the state of a liquid. The energy loss
(dE/dx ≈1GeV/fm) is moderate and the free path length is of a nuclear size. The three-
particle correlations also favor the ring-like structure.
The indications on high energy effects came from the cosmic ray event [199] at energy
about 1016eV (LHC!) with two ring-likeregions. They are formed at such angles in the target
rest system which are equivalent to 60o−70oand 110o−120oin c.m.s. It corresponds to the
refractive index close to 1 that well fits results of dispersion relations and experiment at these
energies. Such dependence on parton energy shows that the same medium could be seen as a
liquid by rather slow partons and as a gas by very fast ones.
It is crucial for applicability of Eq.(33) to define properly the target rest system. In RHIC
experiments the parton-trigger moves in the transverse direction to the collision axis and, on
the average, “sees” the target (the primary fireball) at rest in c.m.s. dealing with rather low
xand Q2. In the cosmic event the narrow forward ring is produced by fast forward moving
partons (large x) which “see” the target at rest in the lab. system. At LHC one can await for
both types of Cherenkov gluons produced. Thus, the hadronic Cherenkov effect can be used
as a tool to scan (1/x,Q2)-plane and plot on it the parton densities (see Eq.(31)) corresponding
to its different regions.
To conclude, the ring-like structure of inelastic processes must be observed if the gluonic
Cherenkov effects are strong enough. The ring parameters reveal the properties of the nuclear
medium and their energy dependence.

Heavy Ion Collisions at the LHC - Last Call for Predictions 83
4.6. Evolution of pion HBT radii from RHIC to LHC – predictions from ideal hydrodynamics
E. Frodermann, R. Chatterjee and U. Heinz
We use the longitudinally boost-invariant relativistic ideal hydrodynamic code
AZHYDRO [130] to predict the expected trends for the evolution from RHIC to LHC of
the HBT radii at mid-rapidity in central (A≈200)+(A≈200) collisions, as well as that of their
normalized oscillation amplitudes in non-central collisions. We believe that these trends may
be trustworthy, in spite of the model’s failure to correctly predict the HBT radii at RHIC [200].
The results shown here are selected from Ref. [201].
Hydrodynamics can not predict the √s-dependence of its own initial conditions, but it
relates uniquely the initial entropy density to the final hadron multiplicity. We compute hadron
spectra and HBT radii as functions of final multiplicity, parametrized by the initial peak
entropy density s0at thermalization time τ0in b=0 collisions. We hold T0τ0constant (where
2
4
6
8
Rs(fm)
KT = 0.0 GeV
KT = 0.2 GeV
KT = 0.4 GeV 2
4
6
8
Ro(fm)
KT = 0.6 GeV
KT = 0.8 GeV
KT = 1.0 GeV
200 300 400 500 600
s0(fm-3)
0
5
10
15
20
Rl(fm)
680 980 1280 1580 1880
dNch/dy
0
2
4
6
8
Rs(fm)
s0=117 fm-3
s0=215 fm-3
0
2
4
6
Ro (fm)
s0=326 fm-3
s0=458 fm-3
0 0.2 0.4 0.6 0.8 1
KT (GeV)
0
5
10
15
Rl(fm)
s0=602 fm-3
Figure 65: (Color online) Pion HBT radii for central (b=0) Au+Au collisions as a function
of transverse pair momentum KT(left) and of initial entropy density s0or final charged
multiplicity dNch
dy (right). For details see [201].
T0∼s1/3
0is the initial peak temperature). Our results cover a range from dNch
dy =680 (“RHIC

Heavy Ion Collisions at the LHC - Last Call for Predictions 84
initial conditions”: s0=117fm−3at τ0=0.6 fm/c) to dNch
dy =2040 (“LHC initial conditions”:
s0=602fm−3at τ0=0.35 fm/c).
1. Central collisions: Figure 65 shows the pion HBT radii for central Au+Au (Pb+Pb) col-
lisions in the (osl) coordinate system [200]. Since we computed the HBT radii from the
space-time variances of the emission function instead of doing a Gaussian fit to the two-pion
correlation function, all Rlvalues should be corrected downward by about 20% [202]. We
see no dramatic changes, neither in magnitude nor in KT-dependence, of the HBT radii as we
increase the multiplicity by up to a factor 3. The largest increase (by ∼30% at low KT) is
seen for Rs, while Roeven slightly decreases at large KT.Rlchanges hardly at all. The main
deficiency of hydrodynamic predictions for the HBT radii at RHIC (too weak KT-dependence
of Rsand Roand a ratio Ro/Rsmuch larger than 1) is not likely to be resolved at the LHC
unless future LHC data completely break with the systematic tendencies observed so far [200].
2. Non-central collisions: Figure 66 shows the normalized azimuthal oscillation amplitudes
[203] of the HBT radii for b=7fm Au+Au collisions. The dashed line in the lower left panel
0 0.2 0.4 0.6 0.8 1
KT
-0.4
-0.2
0
0.2
0.4
Normalized Oscillation
Amplitude
RHIC (dNch/dy=680)
0 0.2 0.4 0.6 0.8 1
KT
-0.4
-0.2
0
0.2
0.4
Normalized Oscillation
Amplitude
LHC (dNch/dy=2040)
200 300 400 500 600
s0
-0.1
-0.05
0
0.05
0.1
Normalized Oscillation
Amplitude
KT=0.0 GeV
200 300 400 500 600
s0
-0.4
-0.2
0
0.2
0.4
Normalized Oscillation
Amplitude
R2{o,2}/R2{s,0}
R2{s,2}/R2{s,0}
R2{os,2}/R2{s,0}
R2{l,2}/R2{l,0}
KT=1.0 GeV
Figure 66: (Color online) Normalized HBT oscillation amplitudes as a function of KTat
RHIC and LHC (top) and as function of s0for two values of KT(bottom).
gives the spatial eccentricity of the source at freeze-out [203]: ǫf.o.
x≈2limKT→0R2
s,2/R2
s,0.
The freeze-out eccentricity is seen to flip sign between RHIC and LHC: at the LHC the
freeze-out source is elongated in the reaction plane direction by almost as much as it was
still out-of-plane elongated at RHIC.

Heavy Ion Collisions at the LHC - Last Call for Predictions 85
4.7. Correlation radii by FAST HADRON FREEZE-OUT GENERATOR
Iu. A. Karpenko, R. Lednicky, I. P. Lokhtin, L. V. Malinina, Yu. M. Sinyukov and A. M. Snigirev
The predictions for correlation radii in the central Pb+Pb collisions for LHC √sNN =5500 GeV
are given in the frame of FAST HADRON FREEZE-OUT GENERATOR (FASTMC).
One of the most spectacular features of the RHIC data, refereed as “RHIC puzzle”, is
the impossibility to describe simultaneously momentum-space measurements and the freeze-
out coordinate-space ones (femtoscopy) by the existing hydrodynamic and cascade models
or their hybrids. However, a good description of SPS and RHIC data have been obtained in
various models based on hydro-inspired parametrizations of freeze-out hypersurface. Thus,
we have achievedthis goal within our fast hadron freeze-out MC generator (FASTMC) [204].
In FASTMC, particle multiplicities are determined based on the concept of chemical freeze-
out. Particles and hadronic resonances are generated on the thermal freeze-out hypersurface,
the hadronic composition at this stage is defined by the parameters of the system at chemical
freeze-out [204]. The input parameters which control the execution of our MC hadron
generator in the case of Bjorken-like parameterization of the thermal freeze-out hypersurface
(similar to the well known “Blast-Wave” parametrization with the transverse flow) for central
collisions are the following: temperature Tch and chemical potentials per a unit charge
eµB,eµS,eµQat chemical freeze-out, temperature Tth at thermal freeze-out, the fireball transverse
radius R, the mean freeze-out proper time τand its standard deviation ∆τ(emission duration),
the maximal transverse flow rapidity ρmax
u. We considered here the naive “scaling” of
the existing physical picture of heavy ion interactions over two order of magnitude in √s
to the maximal LHC energy √sNN =5500 GeV. The model parameters obtained by the
fitting within FASTMC generator of the existing experimental data on mt-spectra, particle
ratios, rapidity density dN/dy,kt-dependence of the correlation radii Rout,Rside,Rlong from
SPS (√sNN =8.7−17.3 GeV) to RHIC ( √sNN =200 GeV) are shown in Fig. 67. For
LHC energies we have fixed the thermodynamic parameters at chemical freeze-out as the
asymptotic ones: Tch =170 MeV, eµB=eµS=eµQ=0 MeV. The linear extrapolation of the
model parameters in log( √s) to LHC ( √sNN =5500 GeV) is shown in Fig. 67 by open
symbols. The extrapolated values are the following: R∼11 fm, τ∼10 fm/c, ∆τ∼3.0 fm/c,
ρmax
u∼1.0, Tth ∼130 MeV. The density of charged particles at mid-rapidity obtained with
these parameters is dN/dy =1400, i.e. twice larger than at RHIC √sNN =200 GeV in
coincidence with the naive extrapolation of dN/dy. These parameters yield only a small
increase of the correlation radii Rout,Rside,Rlong (Fig. 68).
4.8. Exciting the quark-gluon plasma with a relativistic jet
M. Mannarelli and C. Manuel
We discuss the properties of a system composed by a static plasma traversed by a jet of particles.
Assuming that both the jet and the plasma can be described using a hydrodynamical approach, and
in the conformal limit, we find that unstable modes arise when the velocity of the jet is larger than
the speed of the sound of the plasma and only modes with momenta smaller than a certain values

Heavy Ion Collisions at the LHC - Last Call for Predictions 86
10 2
10 3
10 4
10
(MeV)
ch
, T
th
, T
ch
B
µ
0
50
100
150
200 a
10 2
10 3
10 4
10
max
u
ρ
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2 b
(GeV)s
10 2
10 3
10 4
10
(fm/c), R (fm)τ, τ ∆
0
2
4
6
8
10
12 c
Figure 67: FASTMC parameters versus
log(√s) for SPS √s=8.7−17.3 GeV (black
squares), RHIC √s=200 GeV (black triangles)
and LHC √s=5500 GeV(open circles): (a)Tch,
Tth,µB, (b)ρmax
u, (c)τ,Rand ∆τ.
(GeV/c)
t
k
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
(fm)
out
R
2
3
4
5
6
7
8
9
10
(GeV/c)
t
k
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
(fm)
side
R
2
3
4
5
6
7
8
9
10
(GeV/c)
t
k
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
(fm)
long
R
2
3
4
5
6
7
8
9
10
Figure 68: The π+π+correlation radii in
longitudinally comoving system at mid-rapidity
in central Au+Au collisions at √sNN =200 GeV
from the STAR experiment [205] (open circles)
and the FASTMC calculations for LHC √s=
5500 GeV (black squares).
are unstable. Moreover, for ultrarelativistic velocities of the jet the most unstable modes correspond
to relative angles between the velocity of the jet and momentum of the collective mode ∼π/4. Our
results suggest an alternative mechanism for the description of the jet quenching phenomenon, where
the jet crossing the plasma loses energy exciting colored unstable modes. In LHC this effect should be
seen with an enhanced production of hadrons for some specific values of their momenta and in certain
directions of momenta space.
It has been suggested that a high pTjet crossing the medium produced after a relativistic
heavy ion collision, and travelling at a velocity higher than the speed of sound should form
shock waves with a Mach cone structure [206,207]. Such shock waves should be detectable
in the low pTparton distributions at angles π±1.2 with respect to the direction of the trigger
particle. A preliminary analysis of the azimuthal dihadron correlation performed by the
PHENIX Collaboration [208] seems to suggest the formation of such a conical flow.
We propose a novel possible collective process to describe the jet quenching
phenomenon. In our approach a neutral beam of colored particles crossing an equilibrated
quark-gluon plasma induces plasma instabilities [209]. Such instabilities represent a very
efficient mechanism for converting the energy and momenta stored in the total system
(composed by the plasma and the jet) into (growing) energy and momenta of gauge fields,
which are initially absent. To the best of our knowledge, only reference [210] considers the

Heavy Ion Collisions at the LHC - Last Call for Predictions 87
possibility of the appearance of filamentation instabilities produced by hard jets in heavy-ion
collisions.
We have studied this phenomenon using the chromohydrodynamical approach developed
in [211], assuming the conformal limit for the plasma. Since we are describing the system
employing ideal fluid-like equations, our results are valid at time scales shorter than the
average time for collisions. A similar analysis using kinetic theory, and reaching to similar
results, will soon be reported.
We have studied the dispersion laws of the gauge collective modes and their dependence
on the velocity of the jet v, the magnitude of the momentum of the collective mode k, the angle
θbetween these quantities, and of the plasma frequencies of both the plasma ωpand the jet
ωjet. We find that there is always one unstable mode if the velocity of the jet is larger than the
speed of sound cs=1/√3, and if the momentum of the collective mode is in modulus smaller
than a threshold value. Quite interestingly we find that the unstable modes with momentum
parallel to the velocity of the jet is the dominant one for velocity of the jet v.0.8. For larger
values of the jet velocity only the modes with angles larger than ∼π/8 are significant and the
dominant unstable modes correspond to angles ∼π/4 (see figure 69).
00.05 0.1
b
0
0.08
0.16
Γmax/ωt
θ = 0
θ = π/8
θ = π/4
θ = 3π/8
θ = π/2
00.05 0.1
b
0
0.08
0.16
Γmax/ωt
θ = 0
θ = π/8
θ = π/4
θ = 3π/8
θ = π/2
Figure 69: Largest value of the imaginary part of the dispersion law for the unstable mode as
a function of b=ω2
jet/ω2
pfor two different values of the velocity of the jet vand five different
angles between kand v. The left/right panels correspond to v=0.8/0.9, respectively.
Our numerical results imply that both in RHIC and in the LHC these instabilities develop
very fast, faster in the case of the LHC as there one assumes that ωpwill attain larger values.
Further, the soft gauge fields will eventually decay into soft hadrons, and may affect the
hydrodynamical simulations of shock waves mentioned in reference [206,207].
5. Fluctuations
5.1. Fluctuations and the clustering of color sources
L. Cunqueiro, E. G. Ferreiro and C. Pajares
We present our results on multiplicity and pTfluctuations at LHC energies in the framework of
the clustering of color sources. In this approach, elementary color sources -strings- overlap forming

Heavy Ion Collisions at the LHC - Last Call for Predictions 88
clusters, so the number of effective sources is modified. We find that the fluctuations are proportional
to the number of those clusters.
Non-statistical event-by-event fluctuations in relativistic heavy ion collisions have been
proposed as a probe of phase instabilities near de QCD phase transition. The transverse
momentum and the multiplicity fluctuations have been measured at SPS and RHIC energies.
These fluctuations show a non-monotonic behavior with the centrality of the collision: they
grow as the centrality increases, showing a maximum at mid centralities, followed by a
decrease at larger centralities. Different mechanisms have been proposed in order to explain
those data. Here, we will apply the clustering of color sources. In this approach, color
strings are stretched between the colliding partons. Those strings act as color sources of
particles which are successively broken by creation of q¯qpairs from the sea. The color strings
correspond to small areas in the transverse space filled with color field created by the colliding
partons. If the density of strings increases, they overlap in the transverse space, giving rise
to a phenomenon of string fusion and percolation [65]. Percolation indicates that the cluster
size diverges, reaching the size of the system. Thus, variations of the initial state can lead to
a transition from disconnected to connected color clusters. The percolation point signals the
onset of color deconfinement.
These clusters decay into particles with mean transverse momentum and mean
multiplicity that depend on the number of elementary sources that conform each cluster,
and the area occupied by the cluster. In this approach, the behavior of the pT[166] and
multiplicity [167] fluctuations can be understood as follows: at low density, most of the
particles are produced by individual strings with the same transverse momentum <pT>1
and the same multiplicity < µ1>, so fluctuations are small. At large density, above the critical
point of percolation, we have only one cluster, so fluctuations are not expected either. Just
below the percolation critical density, we have a large number of clusters formed by different
number of strings n, with different size and thus different <pT>nand different < µ >nso the
fluctuations are maximal.
The variables to measure event-by-event pTfluctuations are φand FpT, that quantify the
deviation of the observed fluctuations from statistically independent particle emission:
φ=s<Z2>
< µ > −p<z2> , (34)
where zi=pTi−<pT>is defined for each particle and Zi=PNi
j=1zjis defined for each event,
and
FpT=ωdata −ωrandom
ωrandom , ω =q<p2
T>−<pT>2
<pT>.(35)
Moreover, in order to measure the multiplicity fluctuations, the variance of the multiplicity
distribution scaled to the mean value of the multiplicity has been used. Its behavior is similar
to the one obtained for Φ(pT), used to quantify the pT-fluctuations, suggesting that they are
related to each other. The Φ-measure is independent of the distribution of number of particle

Heavy Ion Collisions at the LHC - Last Call for Predictions 89
par
N
0 50 100 150 200 250 300 350 400
%
T
p
F
0
1
2
3
4
5
6
Figure 70: FpTat LHC.
par
N
0 50 100 150 200 250 300 350 400
>
-
)/<n
-
(n
2
Σ
0.5
1
1.5
2
2.5
3
3.5
4
LHC
RHIC
SPS
=1.5
y
δ
Figure 71: Scaled variance on negatively
charged particles at, from up to down, LHC,
RHIC and SPS.
sources if the sources are identical and independent from each other. That is, Φshould be
independent of the impact parameter if the nucleus-nucleus collision is a simple superposition
of nucleon-nucleon interactions.
In Fig. 70 we present our results on pTfluctuations at LHC. Note that the increase of
the energy essentially shifts the maximum position to a lower number of participants [166].
In Fig. 71 we show our values for the scaled variance of negatively charged particles at SPS,
RHIC and LHC energies.
Summarizing: the pTand multiplicityfluctuations are due in our approach to the different
mean <pT>and mean multiplicities of the clusters, and they depend essentially on the
number of clusters. In other words, a decrease in the number of effective sources leads to
a decrease of the fluctuations.
5.2. Fluctuations of particle multiplicities from RHIC to LHC
G. Torrieri
We define an observable capable of determining which statistical model, if any, governs freeze-
out in very high energy heavy ion collisions such as RHIC and LHC. We calculate this observable for
K/π fluctuations, and show that it should be the same for RHIC and LHC, as well as independent of
centrality, if the Grand-Canonical statistical model is appropriate and chemical equilibrium applies.
We describe variations of this scaling for deviations from this scenario, such as light quark chemical
non-equilibrium, strange quark over-saturation and local (canonical) equilibrium for strange quarks.
Particle yield fluctuations are a promising observable to falsify the statistical model and
to constrain its parameters (choice of ensemble, strangeness/light quark chemical equilibrium)
[212]. The uncertainities associated with fluctuations, however, warrant that care be taken to
choose a fluctuation observable.

Heavy Ion Collisions at the LHC - Last Call for Predictions 90
For instance, volume fluctuations could be originating from both initial state effects and
dynamical processes, and are not well understood. Their effect has to be factored out from
multiplicity fluctuations data. One way to do this is to concentrate on fluctuations of particle
ratios, where volume factors out event by event [213]
σ2
N1/N2=D(∆N1)2E
hN1i2+D(∆N2)2E
hN2i2−2h∆N1∆N2i
hN1ihN2i.(36)
This, however, introduces an average hadronization volume dependence through the N1,2
terms (two in the denominator, one in the numerator of Eq. 36).
This feature allows us to perform an invaluable consistency chech for the statistical
model, since the volume going into the ratio fluctuations must, for consistency, be the same as
the volume going into the yields. Thus, observables such as dhN1i
dy σ2
N1/N2should be strictly
independent of multiplicity and centrality, as long as the statistical model holds and the
physically appropriate ensemble is Grand Canonical.
We propose doing this test, at both RHIC and LHC, using the corrected variance
ΨN1
N1/N2=dN1
dy νdyn
N1/N2(37)
where νdyn
N1/N2is theoretically equal to the corrected mixed variance [214]
νdyn
N1/N2=(σdyn
N1/N2)2=σ2
N1/N2−(σPoisson
N1/N2)2=
=hN1(N1−1)i
hN1i2+hN2(N2−1)i
hN2i2−2hN1N2i
hN1ihN2i(38)
SHAREv2.X [170] provides the possibility of calculating all ingredients of ΨN1
N1/N2for
any hadrons, incorporating the effect of all resonance decays, as well as chemical
(non)equilibrium. The calculation for Ψπ−
K−/π−, as well as Ψπ−
K−/π−is shown in Fig. 72. These
species were chosen because their correlations (from resonance decays, N∗→N1N2), which
would need corrections for limited experimental acceptance, are small.
Equilibrium thermal and chemical parameters are very similar at RHIC and the LHC(the
baryo-chemical potential will be lower at the LHC, but it is so low at RHIC that the difference
is not experimentally detectable). Thus, ΨN1
N1/N2should be identical, to within experimental
error, for boththe LHC and RHIC, overall multiplicities were the statistical model is thought
to apply.
According to [168], chemical conditions at freeze-out deviate from equilibrium, and
reflect the higher entropy contect and strangeness per entropy content of the early deconfined
phase through an over-saturated phase space occupancy for the light and strange quarks
(γs> γq>1). If this is true, than ΨN1
N1/N2should still be independent of centrality for a given
energy range, but should go markedly up for the LHC from RHIC, because of the increase in
γqand γs. Fig. 72 shows what effect three different sets of γq,sinferred in [168] would have
on Ψπ−
K−/π−and Ψπ−
K−/K+
If non-statistical processes (minijets, string breaking etc.) dominate event-by-event
physics, the flat ΨN1
N1/N2scaling on centrality/multiplicity should be broken, and ΨN1
N1/N2would
exhibit a non-trivial dependence on Npart or dN/dy.

Heavy Ion Collisions at the LHC - Last Call for Predictions 91
This is also true if global correlations persist, such as is the case in Canonical and
micro-canonical models [215] If global correlations persist for particle N2and/or N1, than
ΨN1
N1/N2becomes reduced, and starts strongly varying with centrality in lower multiplicity
events. Thus, if strangeness at RHIC/the LHC is created and maintained locally, ΨN1
N1/N2
should develop a “wiggle” at low centrality, and be considerably lower than Grand Canonical
expectation. For Ψπ−
K+/K−it should be lower by a factor of two.
In conclusion, measuring Ψπ−
K−/π−and Ψπ−
K+/K−, at comparing the results between the LHC
and RHIC can provide an invaluable falsification of the statistical model, as well as constraints
as to which statistical model applies in these regimes.
0500 1000 1500 2000 2500 3000
dNπ−/dy
0
0.5
1
1.5
2
(dNπ−/dy)νK-/π−
Equilibrium (T=156,γq=γs=1)
Non-eq. 1 (T=145,γs=γq=1.62)
Non-eq. 2 (T=134,γq=1.67,γs=3)
Non-eq 3 (T=125,γq=1.73,γs=5)
Higher γq,s
RHIC (Matches
T=140 MeV,γq=1.5,γs=2)
0500 1000 1500 2000 2500 3000
dNπ−/dy
102
103
104
(dNπ−/dy)νK-/K+
Equilibrium (T=156,γq=γs=1)
Non-eq. 1 (T=145,γs=γq=1.62)
Non-eq. 2 (T=134,γq=1.67,γs=3)
Non-eq 3 (T=125,γq=1.73,γs=5)
Equilibrium Canonical
~dNK/dy=10
Figure 72:
6. High transverse momentum observables and jets
6.1. Jet quenching parameter ˆq from Wilson loops in a thermal environment
D. Antonov and H. J. Pirner
The gluon jet quenching parameter is calculated in SU(3) quenched QCD within the stochastic
vacuum model. At the LHC-relevant temperatures, it is defined by the gluon condensate and the
vacuum correlation length. Numerically, when the temperature varies from Tc=270MeV to the inverse
vacuum correlation length µ=894MeV, the jet quenching parameter rises from zero to 1.1GeV2/fm.
At LHC energies, radiative energy loss is the dominant mechanism of jet energy loss in
the quark-gluon plasma. The expectation value of a light-like adjoint Wilson loop provides an
estimate for the radiative energy loss of a gluon [216]:
Wadj.
Lk×L⊥=exp −ˆq
4√2LkL2
⊥!.(39)
The contour of the loop at zero temperature is depicted in Fig. 73. We have calculated
the jet quenching parameter ˆqin the SU(3) quenched theory through the evaluation of the

Heavy Ion Collisions at the LHC - Last Call for Predictions 92
L
L
L
z
t
x
Figure 73: The contour of the Wilson loop of a gluon.
Wilson loop (39). To this end, we have used the stochastic vacuum model [217] at T>Tc,
where Tc=270MeV is the deconfinement temperature. This model incorporates the gluon
condensate which, together with the vacuum correlation length, defines the jet quenching
parameter. This is different from the results obtained within perturbative QCD [218] and
conformal field theories [216], where ˆq∝T3.
The hierarchy of scales in our problem is µ−1≪L⊥≪β≪Lk, where βis the inverse
temperature, and µ=894MeV is the inverse vacuum correlation length. Due to the x4-
periodicity at finite temperature, the contour depicted in Fig. 73 effectively splits into
segments whose extensions along the 3rd and the 4th axes are β. Furthermore, due to the
short-rangeness of gluonic correlations, which fall offat the vacuum correlation length, the
dominant contribution to ˆqstems from self-interactions of individual segments. We have also
calculated the contribution stemming from the correlations of neighboring segments, which
turns out to be parametrically (and numerically) suppressed by the factor e−µ/T. For this
reason, the even smaller contributions from the next-to-nearest neighboring segments on are
disregarded. The contributions of individual and neighboring segments read
ˆq=g2D(Fa
µν)2ET=0
16µ"√2−T
µ1−e−√2µ/T#"cothµ
2T−coth µ
2Tc!#and
∆ˆq=g2D(Fa
µν)2ET=0
16µe−µ/T"1−T
µ1−e−µ/T#"cothµ
2T−coth µ
2Tc!#,
respectively. The right most brackets in these equations define the temperature dependence
of the gluon condensate, corresponding to the exponential fall-offof its nonlocal
counterpart [219]. As for the zero-temperature value of the gluon condensate, it can be
expressed through the vacuum correlation length and the string tension in the fundamental
representation of SU(3), σ=(440MeV)2, and reads [220] g2D(Fa
µν)2ET=0=(72/π)σµ2=
3.55GeV4. The above contributions together with their sum are plotted in Fig. 74. Note
finally that, in the large-Nclimit, our full result for the jet quenching parameter behaves as
N0
c, i.e. it does not scale with Nc. This behavior is similar to those of other models [216,218].
6.2. Particle Ratios at High pTat LHC Energies
G. G. Barnaföldi, P. Lévai, B. A. Cole, G. Fai and G. Papp

Heavy Ion Collisions at the LHC - Last Call for Predictions 93
0
0.2
0.4
0.6
0.8
1
1.2
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
T [GeV]
full jet quenching parameter
individual segments
neighboring segments
ˆq[GeV2/fm]
Figure 74: The full jet quenching parameter and relativecontributions to it.
Hadron production has been calculated in a pQCD improved parton model for pp,dA and heavy
ion collisions. We applied KKPand AKK fragmentation functions. Our jet fragmentation study shows,
that hadron ratios at high pTdepend on quark contribution mostly and less on the gluonic one. This
finding can be seen in jet-energy loss calculations, also. We display the suppression pattern on different
hadron ratios in PbPb collisions at LHC energies.
The precision of pQCD based parton model calculations was enhanced during the last
decade. The calculated spectra allow to make predictions not only for the hadron yields, but
for sensitive particle ratios and nuclear modifications. For the calculation of particle ratios
new fragmentation functions are needed not only for the most produced light mesons, but for
protons also. From the experimental point of view one requires identified particle spectra by
RHIC and LHC. Especially the ALICE detector has a unique capability to measure identified
particles at highest transverse momenta via ˇ
Cherenkov detectors. The π±/K±and K±/p( ¯p)
ratios can be measured up to 3 GeV/c and 5 GeV/c respectively.
Here we calculate hadron ratios in our next-to-leading order pQCD improved parton
model based on Ref. [221] with intrinsic transverse momenta, determined by the expected
c.m. energy evolution along the lines of Ref. [221]. The presented ratios are based on π,K
and pspectra which were calculated by AKK fragmentation functions [14]. First we compare
calculated particle ratios to the data of the STAR collaboration measured in AuAu collisions
at √s=200 AGeV RHIC energy [222,223]. Predictions for high-pThadron ratios at RHIC
and at LHC energies in most central (0−10%) PbPb collisions are also shown in Fig. 75.
On the
left panel
of Fig. 75, particle ratios are compared to AuAu collisions at √s=200
AGeV STAR K/π (
dots
) and p/π (
triangles
) data. The agreement between the RHIC data
and the calculations at RHIC energy can be considered acceptable at pT&5 GeV/c, with an
opacity of L/λ =4. However, at lower momenta, where pQCD is no longer reliable, the ratios
differ from the calculated curves.
The
right panel
shows calculations for PbPb collisions for √s=5.5ATeV energy.
Using a simple dN/dy∼1500 −3000 estimation, we expect a L/λ ≈8 opacity in most
central PbPb collisions. For comparison, we plotted the L/λ =0 and 4 values also. The
lower- and intermediate-pTvariation of the hadron ratios arise from the different strengths

Heavy Ion Collisions at the LHC - Last Call for Predictions 94
Figure 75: Calculated charge-averaged K/π and p/π ratios in AA collisions at RHIC and LHC
energies. RHIC curves are compared to STAR [222,223] data at √s=200 AGeV.
of the jet quenching for quark and gluon contributions [224]. Due to the quark dominated
fragmentation, the difference disappears at high-pTin the ratios.
6.3. π0fixed p⊥suppression and elliptic flow at LHC
A. Capella, E. G. Ferreiro, A. Kaidalov and K. Tywoniuk
Using a final state interaction model which describes the data onthese two observables, at RHIC,
we make predictions at the LHC – using the same cross-section and p⊥-shift. The increase in the
medium density between these two energies (by a factor close to three) produces an increase of the
fixed p⊥π0suppression by a factor 2 at large p⊥and of v2by a factor 1.5.
6.3.1. π0fixed p⊥suppression Final state interaction (FSI) effects have been observed in
AA collisions. They are responsible of strangeness enhancement, J/ψ supression, fixed p⊥
supression, azimuthal asymmetry, ... Is it the manifestation of the formation of a new state
of matter or can it be described in a FSI model with no reference to an equation of state,
thermalization, hydrodynamics, ... ? We take the latter view and try to describe all these
obseervables within a unique formalism : the well known gain and loss differential equations.
We assume [225] that, at least for particles with p⊥larger than <p⊥>, the interaction with
the hot medium produces a p⊥-shift δp⊥towards lower values and thus the yield at a given p⊥
is reduced. There is also a gain term due to particles produced at p⊥+δp⊥. Due to the strong
decrease of the p⊥-distributions with increasing p⊥, the loss is much larger than the gain.
Asuming boost invariance and dilution of the densities in 1/τ due to longitudinal expansion,
we obtain
τdNπ0(b,s,p⊥)
dτ=−σN(b,s)Nπ0(b,s,p⊥)−Nπ0(b,s,p⊥+δp⊥)(40)

Heavy Ion Collisions at the LHC - Last Call for Predictions 95
[GeV/c]
T
p
0 10 20 30 40 50
AA
R
-1
10
1
0
π
PHENIX (0-10%)
0
πCentral
Suppression @ RHIC
One-jet @ LHC
Two-jet @ LHC
Figure 76: From up to down: RHIC initial,
2 LHC initial, RHIC final, LHC FSI, LHC
FSI+shadowing.
[GeV/c]
T
p
0 1 2 3 4 5
2
v
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
PHENIX (13%-26%)
0
πMid-central
Figure 77: v2for π0at RHIC (lower curve) and
LHC (upper curve).
Here N≡dN/dyd2sis the transverse density of the medium and Nπ0the corresponding
one of the π0[226]. This has to be integrated between initial time τ0and freeze-out time τf.
The solution deepnds only on τf/τ0. We use σ=1.4 mb at both energies and δp⊥=p1.5
⊥/20
for p⊥<2.9 GeV and δp⊥=p0.8
⊥/9.5 for p⊥>2.9 GeV [227]. Eq. (40) at small τdescribes
an interaction at the partonic level. Indeed, here the densities are very large and the hadrons
not yet formed. At later times the interaction is hadronic. Most of the effect takes place in
the partonic phase. We use a single (effective) value of σfor all values of the proper time τ.
The results at RHIC and LHC are given in Fig. 76. At LHC only shadowing [226] has been
included in the initial state. The suppression is given by the dashed line. It coincides with RAA
for p⊥large enough – when shadowing and Cronin efffects are no longer present. The LHC
suppression is thus a factor of two larger than at RHIC.
6.3.2. Elliptic flow Final state interaction in our approach gives rise to a positive
v2[227] (no need for an equation of state or hydro). Indeed, when the π0is
emitted at θR=90◦its path length is maximal (maximal absorption). In order to
compute it we assume that the density of the hot medium is proportional to the path
length RθR(b,s) of the π0inside the interaction region determined by its transverse
position sand its azimuthal angle θR. Hence, we replace N(b,s) by N(b,s)RθR(b,s)/
<RθR(b,s)>where RθRis the π0path length and <> denotes its average over θR. (In this
way the averaged transverse density N(b,s) is unchanged). The suppression Sπ0(b,s) depends

Heavy Ion Collisions at the LHC - Last Call for Predictions 96
now on θRand v2is given by
v2(b,p⊥)=ZdθRSπ0(b,p⊥,θR)cos2θR
ZdθRSπ0(b,p⊥,θR)(41)
The results at RHIC and LHC are presented in Fig. 77.
6.4. Energy dependence of jet transport parameter
J. Casalderrey-Solana and X. N. Wang
We study the evolution and saturation of the gluon distribution function in the quark-gluon plasma
as probed by a propagating parton and its effect on the computation of the jet quenching or transport
parameter ˆq. For hard probes, this evolution at small x=Q2
s/6ET leads to a jet energy dependence of
ˆq
Within the picture of multiple parton scattering in QCD, the energy loss for an energetic
parton propagating in a dense medium is dominated by induced gluon bremsstrahlung.
Taking into account of the non-Abelian Landau-Pomeranchuck-Midgal (LPM) interference,
the radiative parton energy loss [228],
∆E=αsNc
4ˆqRL2,(42)
is found to depend on the jet transport or energy loss parameter ˆqwhich describes the averaged
transverse momentum transfer squared per unit distance (or mean-free-path). Here Ris the
color representation of the propagating parton in SU(3).
The transport parameter ˆqRexperienced by a propagating parton can be defined in terms
of the unintegrated gluon distributions φk(x,q2
T) of the color sources in the quark-gluon
plasma,
ˆqR=4π2CR
N2
c−1ρZµ2
0
d2qT
(2π)2Zdxδ(x−q2
T
2p−hk+i)αs(q2
T)φ(x,q2
T),(43)
where hk+iis the average energy of the color sources and φ(x,q2
T) is the corresponding average
unintegrated gluon distribution function per color source. The integrated gluon distribution is
xG(x,µ2)=Zµ2
0
d2qT
(2π)2φ(x,qT).(44)
Since we are interested in the determination of ˆqRat large jet energies, we need to know
the unintegrated parton distribution φ(x,q2
T) in Eq. (43) at small x∼Dq2
TE/6ET. For a large
path length, the typical total momentum transfer, ˆqL, which will set the scale of the process,
is also large. These scales lead to the evolution of the gluon distribution function. In the
medium, this evolution may be modified due to the interaction of the radiated gluons with
thermal partons. However, since the medium effects are of the order of µD<< T, we neglect
those at hard scales. Given that both the scale and the rapidity are large, we describe the
the (linear) vacuum evolution in the double logarithmic approximation (DLA) [229]. The

Heavy Ion Collisions at the LHC - Last Call for Predictions 97
L (fm)
0 1 2 3 4 5 6 7 8
/fm)
2
(GeVq
0
5
10
15
20
E=200 GeV T=0.6 GeV
E=200 GeV T=0.4 GeV
E=20 GeV T=0.6 GeV
E=20 GeV T=0.4 GeV
E (GeV)
0 50 100 150 200 250 300
/fm)
2
(GeVq
0
5
10
15
20 T=0.6 GeV
c
L>L
T=0.4 GeV
c
L>L
L=1 fm T=0.6 GeV
L=1 fm T=0.4 GeV
Figure 78: Jet quenching parameter ˆqas a function of the path length (left) and jet energy
(right). The square (triangle) marks the value of ˆqfor thermal particle at T=0.4 GeV (T=0.6
GeV). Significant corrections to the energy dependence are expected at low energy which
should approach their thermal value at E=3T.
thermal gluon distribution function at a scale µ2=T2is determined via the hard thermal loop
approximation and it is used as an initial condition for the evolution. As in vacuum, the
growth of the gluon distribution function leads to saturation which tame this growth for scales
µ2<Q2
s. The saturation scale is estimated from the linearly evolved distribution. The details
of the computation can be found in [230].
The evolution leads to a jet energy dependence of the transport parameter that is stronger
than any power of logarithmic dependence. The saturation effect also gives rise to a non-
trivial length dependence of the jet transport parameter. These two features are shown in
Figure 78, where we compute the transport parameter for T=0.4 GeV (RHIC) and T=0.6
GeV (LHC). In both cases, the energy dependence of ˆqis significant, leading to a factor of 2
difference between jets of 20 and 200 GeV. This difference is larger for small jet path lengths.
The computation also shows that ˆqgrows as the path length decreases. Both dependences
translate into different amount of radiative energy loss Eq. (42). Let us note, however, that the
derivation of Eq. (42) assumes a constant ˆq; thus, the relation between the radiative energy
loss and the transport parameter should be revisited for an energy/length dependent ˆq.
6.5. PQM prediction of RAA(pT)and RCP(pT)at midrapidity in Pb–Pb collisions at the
LHC
A. Dainese, C. Loizides and G. Pai´c
The Parton Quenching Model (PQM) couples the BDMPS-SW quenching weights for radiative
energy loss with a realistic description of the nucleus–nucleus collision geometry, based on the
Glauber model. We present the predictions for the nuclear modification factors, in Pb–Pb relative
to pp collisions (RAA) and in central relative to peripheral Pb–Pb collisions (RCP), of the transverse
momentum distributions of light-flavour hadrons at midrapidity.
The Parton Quenching Model (PQM) [231], which combines the pQCD BDMPS-SW

Heavy Ion Collisions at the LHC - Last Call for Predictions 98
framework for the probabilistic calculation of parton energy loss in extended partonic matter
of given size and density [232] with a realistic description of the collision overlap geometry
(Glauber model) in a static medium, was shown to describe the transverse momentum and
centrality dependence of the leading particle suppression in Au–Au collisions at top RHIC
energy. The model has one single parameter that sets the scale of the BDMPS transport
coefficient ˆq, hence of the medium density. The parameter has been tuned [231] on the basis
of the RAA data at √sNN =200 GeV, that indicate a transport coefficient in the range 4–
14 GeV2/fm. We scale the model parameter to LHC energy assuming its proportionality to
the expected volume-density of gluons ng. Using the value of ngpredicted for the LHC by the
EKRT saturation model [74] (which gives dNch/dy≃3000), we obtain ˆq≃25–100 GeV2/fm.
In PQM we obtain the leading-particle suppression in nucleus–nucleus collisions by
calculating the hadron-level transverse momentum distributions in a Monte Carlo approach.
The ‘event loop’ that we iterate is the following: 1) Generation of a parton, quark or gluon,
with pT>5 GeV, using the PYTHIA event generator in pp mode with CTEQ 4L parton
distribution functions; nuclear shadowing is neglected, since it’s effect is expected to be
small above 5–10 GeV in pT; the pT-dependence of the quarks-to-gluons ratio is taken
from PYTHIA. 2) Sampling of a parton production point and propagation direction in the
transverse plane, according to the density of binary collisions, and determination of the in-
medium path length and of the path-averaged ˆq, the inputs for the calculation of the quenching
weights, i.e. the energy-loss probability distribution P(∆E). 3) Sampling of an energy
loss ∆Eaccording to P(∆E) (non-reweighted case [231]) and definition of the new parton
transverse momentum, pT−∆E; 4) Fragmentation of the parton to a hadron using the leading-
order Kniehl-Kramer-Pötter (KKP) fragmentation functions. Quenched and unquenched pT
distributions are obtained including or excluding the third step of the chain. The nuclear
modification factor RAA(pT) is given by their ratio.
The left-hand panel of Fig. 79 shows the pT-dependence of the RAA nuclear modification
factor in 0–10% central Pb–Pb at √sNN =5.5 TeV relative to pp. The RAA for central Au–
Au collisions at top RHIC energy is also shown and compared to π0data from the PHENIX
experiment [233]. PQM predicts for central Pb–Pb at the LHC a very slow increase of RAA
with pT, from about 0.1 at 10 GeV to about 0.2 at 100 GeV. The right-hand panel of the
figure shows the RCP central-to-peripheral nuclear modification factor for different centrality
classes relative to the peripheral class 70–80%.
6.6. Effect of dynamical QCD medium on radiative heavy quark energy loss
M. Djordjevic and U. Heinz
The computation of radiative energy loss in a dynamically screened QCD medium is a key
ingredient for obtaining reliable predictions for jet quenching in ultra-relativistic heavy ion collisions.
We calculate, to first order in the number of scattering centers, the energy loss of a heavy quark
traveling through an infinite and time-independent QCD medium consisting of dynamical constituents.
We show that the result for a dynamical medium is almost twice that obtained previously for a
medium consisting of randomly distributed static scattering centers. A quantitative description of

Heavy Ion Collisions at the LHC - Last Call for Predictions 99
[GeV]
t
p
1 10 2
10
AA
R
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
[GeV]
t
p
1 10 2
10
AA
R
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
/fm
2
= 25-100 GeVq = 5.5 TeV,
NN
sPb-Pb,
/fm
2
= 4-14 GeVq = 200 GeV,
NN
sAu-Au,
0
π = 200 GeV, PHENIX
NN
s
Au-Au,
(nucl-ex/0611007)
PQM, centrality 0-10%
[GeV]
t
p
10 2
10
CP
R
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
/fm
2
= 25-100 GeVq = 5.5 TeV,
NN
sPb-Pb,
40-50% / 70-80%
20-30% / 70-80%
0-10% / 70-80%
PQM
Figure 79: Left: RAA(pT) for central Pb–Pb collisions at √sNN =5.5 TeV and central Au–Au
collisions at √sNN =200 GeV. The PHENIX π0data are shown with statistical errors only and
they have a 10% normalization systematic error [233]. Right: RCP(pT) for Pb–Pb collisions
at √sNN =5.5 TeV.
jet suppression in RHIC and LHC experiments thus must correctly account for the dynamics of the
medium’s constituents.
Heavy flavor suppression is considered to be a powerful tool to study the properties of a
QCD medium created in ultra-relativistic heavy ion collisions [234]. The suppression results
from the energy loss of high energy partons moving through the plasma [235]. Therefore, the
reliable computations of heavy quark (collisional and radiative) energy loss mechanisms are
essential for the reliable predictions of jet suppression.
However, currently available heavy quark radiative energy loss studies suffer from one
crucial drawback: The medium induced radiative energy loss is computed in a QCD medium
consisting of randomly distributed but static scattering centers (“static QCD medium”).
Within such approximation, the collisional energy loss is exactly zero, which is contrary to
the recent calculations [236] that showed that the collisional contribution is important and
comparable to the radiative energy loss. Due to this, it became necessary to obtain the heavy
quark radiative energy loss in a dynamical QCD medium, and to test how good is the static
approximation in these calculations.
In this proceeding, we report on a first important step, the calculation of heavy
quark radiative energy loss in an infinite and time-independent QCD medium consisting of
dynamical constituents. By comparing with the static medium calculation this permits us to
qualitatively assess the importance of dynamical effects on radiative energy loss.

Heavy Ion Collisions at the LHC - Last Call for Predictions 100
We compute the medium induced radiative energy loss for a heavy quark to first (lowest)
order in number of scattering centers. To compute this process, we consider the radiation
of one gluon induced by one collisional interaction with the medium. In distinction to the
static case, we take into account that the collisional interactions are exhibited with dynamical
(moving) medium partons. To simplify the calculations, we consider an infinite QCD medium
and assume that the on-shell heavy quark is produced at time x0=−∞, i.e. we consider
the Bethe-Heitler limit. The calculations were performed by using two Hard-Thermal Loop
approach, and are presented in [237]. As the end result, we obtained a closed expression for
the radiative energy loss in dynamical QCD medium. This result allows us to compare the
radiative energy loss in dynamical and static QCD medium, from which we can observe two
main differences. First, there is an O(15%) decrease in the mean free path which increases
the energy loss rate in the dynamical medium by O(20%). Second, there is a change in the
shape and normalization of the emitted gluon spectrum. This second difference leads to an
additional significant increase of the heavy quark energy loss rate and of the emitted gluon
radiation spectrum by about 50% for the dynamical QCD medium. The numerical results are
briefly discussed below.
0 0.5 1 1.5
E@TeVD
1.6
1.8
2
DErad
DYN

DErad
STAT
0 1 2 3 4 5
M@GeVD
1.65
1.7
1.75
1.8
DErad
DYN

DErad
STAT
u,d
c
b
Figure 80:
Left panel:
Ratio of the fractional radiative energy loss in dynamical and static
media for charm quarks as a function of initial quark energy E.
Right panel:
Asymptotic
value of the radiative energy loss ratio for high energy quarks as a function of their mass, with
marks indicating the light, charm and bottom quarks. For the parameter values, see [237].
Left panel of the Fig. 80 shows the energy loss ratio between dynamical and static media
for charm quark under the LHC conditions. We see that the ratio is almost independent of
the momentum pof the fast charm quark, saturating at ≃1.75 above p&100 GeV and being
even somewhat larger at smaller momenta. The dynamical enhancement persists at constant
level to the largest possible charm quark energies. Therefore, we can conclude that there is
no quark energy domain where the assumption of static scatterers in the medium becomes a
valid approximation. Further, the mass of the fast quark plays only a minor role for its energy
loss. The right panel in Fig. 80 shows the asymptotic energy loss ratio for very high energy
quarks as a function of the quark mass. While the dynamical enhancement is largest for light
quarks, the difference between light and bottom quarks is only about 15%, and bquarks still
suffer about 70% more energy loss in a dynamical medium than in one with static scattering
centers.
In summary, we obtained an important qualitative conclusion that the constituents

Heavy Ion Collisions at the LHC - Last Call for Predictions 101
of QCD medium can not be approximated as static scattering centers in the energy loss
computations. Therefore, the dynamical effects have to be included for the reliable prediction
of radiative energy loss and heavy flavor suppression in the upcoming high luminosity RHIC
and LHC experiments.
6.7. Charged hadron RAA as a function of pTat LHC
T. Renk and K. J. Eskola
We compute the nuclear suppression factor RAA for charged hadrons within a radiative energy
loss picture using a hydrodynamical evolution to describe the soft medium inducing energy loss. A
minijet +saturation picture provides initial conditions for LHC energies and leading order perturbative
QCD (LO pQCD) is used to compute the parton spectrum before distortion by energy loss.
We calculate the suppression of hard hadrons induced by the presence of a soft medium
produced in central Pb-Pb collisions at √sNN =5.5 TeV at the LHC. Note that this prediction
depends on knowledge of the medium. In the present calculation, the medium evolution is
likewise predicted and has to be confirmed before the suppression can be tested. Note further
that the calculation is only valid where hadron production is dominated by fragmentation and
that it cannot be generalized to the suppression of jets since the requirement of observing a
hard hadron leads to showers in which the momentum flow is predominantly through a single
parton. This is not so for jets in which the momentum flow is shared on average among several
partons (which requires a different framework).
We describe the soft medium evolution by the boost-invariant hydrodynamical model
discussed in [31] where the initial conditions for LHC are computed from perturbative
QCD+saturation [74]. Our calculation for the propagation of partons through the medium
follows the BDMPS formalism for radiative energy loss using quenching weights [232].
Details of the implementation can be found in [238].
The probability density P(x0,y0) for finding a hard vertex at the transverse position
r0=(x0,y0) and impact parameter bis given by the normalized product of the nuclear profile
functions. We compute the energy loss probability P(∆E)path for any given path from a vertex
through the medium by evaluating the line integrals
ωc(r0,φ)=Z∞
0dξ ξ ˆq(ξ) and hˆqLi(r0,φ)=Z∞
0dξˆq(ξ).
Along the path where we assume the relation
ˆq(ξ)=K·2·ǫ3/4(ξ)(coshρ−sinhρcosα)
between the local transport coefficient ˆq(ξ), the energy density ǫand the local flow rapidity ρ
as given in the hydrodynamical model. The angle αis between flow and parton trajectory. We
view the constant Kas a tool to account for the uncertainty in the selection of αsand possible
non-perturbative effects increasing the quenching power of the medium (see [238]) and adjust
it such that pionic RAA for central Au-Au collisions at RHIC is described. The result for LHC
is then an extrapolation with Kfixed.

Heavy Ion Collisions at the LHC - Last Call for Predictions 102
Using the numerical results of [232], we obtain P(∆E;ωc,R)path for ωcand R=2ω2
c/hˆqLi.
From this distribution given a single path, we can define the averaged energy loss probability
distribution P(∆E)iTAA by averaging over all possible paths, weighted with the probability
density P(x0,y0) for finding a hard vertex in the transverse plane.
We consider all partons as absorbed whose energy loss is formally larger than their
initial energy. The momentum spectrum of produced partons is calculated in LO pQCD.
The medium-modified perturbative production of hadrons is obtained from the convolution
dσAA→h+X
med =X
f
dσAA→f+X
vac ⊗hP(∆E)iTAA ⊗Dvac
f→h(z,µ2
F)
with Dvac
f→h(z,µ2
F) the fragmentation function. From this we compute the nuclear modification
factor RAA as
RAA(pT,y)=dNh
AA/dpTdy
TAA(b)dσpp/dpTdy.
0 100 200 300 400
pT [GeV]
0
0.2
0.4
0.6
0.8
1
RAA
RAA for LHC
Figure 81: Expectation for the pTdependence of the nuclear suppression factor RAA for
charged hadrons in central Pb-Pb collisions at midrapidity at the LHC.
Figure 81 shows the expected behaviour of RAA with hadronic transverse momentum pT
at midrapidity. On quite general grounds, we expect a rise of RAA with pTfor any energy
loss model in which the energy loss probability does not strongly depend on the initial parton
energy as more of the shift in energy becomes accessible (see [238]). The detailed form of
the rise is then sensitive to the form of P(∆E)iTAA .
6.8. Nuclear suppression of jets and RAA at the LHC
G. Y. Qin, J. Ruppert, S. Turbide, C. Gale and S. Jeon
The nuclear modification factor RAA for charged hadron production at the LHC is predicted from
jet energy loss induced by gluon bremsstrahlung. The Arnold, Moore, and Yaffe [239–241] formalism
is used, together with an ideal hydrodynamical model [31].
We present a calculation of the nuclear modification factor RAA for charged hadron
production as a function of pTin Pb+Pb collisions at √sNN =5.5 TeV in central collisions at

Heavy Ion Collisions at the LHC - Last Call for Predictions 103
mid-rapidity at the LHC. The net-energy loss of the partonic jets is calculated by applying the
Arnold, Moore, and Yaffe (AMY) formalism to calculate gluon bremsstrahlung [239–241].
The details of jet suppression relies on an understanding of the nuclear medium, namely
the temperatures and flow profiles that are experienced by partonic jets while they interact
with partonic matter at T≥Tc. Our predictions use a boost-invariant ideal hydrodynamic
model with initial conditions calculated from perturbative QCD +saturation [31, 74]. It
is emphasized that the reliability of this work hinges on the validity of hydrodynamics at
the LHC. It has been verified that RAA for π0production as a function of pTas obtained
in the same boost-invariant ideal hydrodynamical model adjusted to Au+Au collisions at
√sNN =0.2 TeV [31] is in agreement with preliminary data from PHENIX in central collisions
at RHIC (and the result is very close to the one obtained in 3D hydrodynamics presented
in [242]). In AMY the strong coupling constant αsis a direct measure of the interaction
strength between the jet and the thermalized soft medium and is the only quantity not
uniquely determined in the model, once the temperature and flow evolution is fixed by the
initial conditions and subsequent hydrodynamical expansion. We found that assuming a
constant αs=0.33 describes the experimental data in most central collisions at RHIC. It is
conjectured that αsshould not be changed very much at the LHC since the initial temperature
is about twice larger than the one at RHIC whereas αsis only logarithmically dependent on
temperature. We present results for αs=0.33 and 0.25.
For details of the calculation of nuclear suppression, we refer the reader to [242]. The
extension to the LHC once the medium evolution and αsare fixed is straightforward. The
initial jets are produced with an initial momentum distribution of jets computed from pQCD
in the factorization formalism including nuclear shadowing effects. The probability density
PAA(~r⊥) of finding a hard jet at the transverse position ~r⊥in central A+A collisions is given
by the normalized product of the nuclear thickness functions, PAA(~r⊥)=TA(~r⊥)TA(~r⊥)/TAA
and is calculated for Pb+Pb collisions. The evolution of the jet momentum distribution
Pj(p,t)=dNj(p,t)/dpdyin the medium is calculated by solving a set of coupled rate equations
with the following generic form,
dPj(p,t)
dt=X
ab ZdkPa(p+k,t)dΓa
jb(p+k,p)
dkdt−Pj(p,t)dΓj
ab(p,k)
dkdt,
where dΓj
ab(p,k)/ddtis the transition rate for the partonic process j→a+bwhich depends on
the temperature and flow profiles experienced by the jets traversing the medium. The hadron
spectrum dNh
AA/d2pTdyis obtained by the fragmentation of jets after their passing through
the medium. The nuclear modification factor RAA is computed as
Rh
AA(~pT,y)=1
Ncoll
dNh
AA/d2pTdy
dNh
pp/d2pTdy.
In figure 82 we present a prediction for charged hadron RAA as a function of pTat mid-
rapidity for central collisions at the LHC. We consider that these two values of αsdefine a
sensible band of physical parameters.

Heavy Ion Collisions at the LHC - Last Call for Predictions 104
0 10 20 30 40 50 60
pT (GeV/c)
0
0.2
0.4
0.6
0.8
1
charged hadron RAA
Hydro + AMY, αs = 0.33
Hydro + AMY, αs = 0.25
Figure 82: The pTdependence of the nuclear modification factor RAA for charged hadrons in
central Pb+Pb collisions at mid-rapidity at the LHC.
6.9. Perturbative jet energy loss mechanisms: learning from RHIC, extrapolating to LHC
S. Wicks and M. Gyulassy
In many recent papers, collisional energy loss has been found to be of the same order as
radiative energy loss for parameters applicable to the QGP at RHIC. As the temperature and jet
energy dependence of collisional energy loss differs from that of radiative loss, the interpretation of the
results at RHIC affects our extrapolation to predictions for the LHC. We present results from a hybrid
collisional plus radiative model, combining DGLV radiative loss with HTL-modified collisional loss,
including the fluctuation spectrum for small numbers of collisions and gluons emitted.
Collisional energy loss is an essential component of the physics of high momentum
partonic jets traversing the quark-gluon plasma [243, 244]. If we do not properly understand
the energy loss mechanisms that are important at RHIC, then we cannot accurately extrapolate
in medium density and jet energy to make predictions for the LHC.
WHDG [244] made a first attempt at including both collisional and DGLV radiative
energy loss processes. A simple model of the collisional energy loss was used: leading
log average loss with a Gaussian distribution around this average, the width given by
the fluctuation-dissipation theorem. For the short lengths of interest in the QGP fireball
(≈0−6fm), we expect a jet to undergo only a small number of significant collisions. But
the fluctuation spectrum for this will be different than that implemented in the WHDG
model: instead, the distinctly non-Gaussian fluctuation in energy loss in 0,1,2,3 collisions
is necessary. We present here results and predictions from an improved hybrid radiative plus
collisional energy loss model which include a full evaluation of these fluctuations.
A significant uncertainty in the model is the use of a fixed strong coupling constant. In
WHDG, a canonical value αs=0.3 was used, validated by the fitting of the pion RAA(pT) at
RHIC. Here, for a fixed density dNg/dy =1000, an increased coupling αs=0.4 is necessary to
do the same. In fact, if the collisional component of the energy loss is neglected completely,
a further increased coupling of αs=0.5 would be necessary, as shown in the left-hand side of

Heavy Ion Collisions at the LHC - Last Call for Predictions 105
Fig. 83. Both values, while large, are still in a possible perturbative kinematical region, and
are evaluated with medium densities constrained by the total entropy and multiplicity of the
collision.
Is it possible to differentiate between these two scenarios: one including collisional loss,
the other neglecting it but increasing the coupling to compensate? Staying with the most
simple observables, single particle inclusives in central collisions, we have three dependences
to test: the dependence on medium density, jet energy and jet mass. The first is tested by the
predicted increased density of the medium to be produced at the LHC (consistency between
the left plot and either the central or right plot in Fig. 83). The very high momentum reach
available for measurements involving gluon and light quark jets is valuable for the second
(radiative versus radiative plus collisional in the central and right hand plots of Fig. 83), and
the separate detection of D and B mesons gives us the third (as in Fig. 84). All these together
will provide very strong constraints on the energy loss models, even before considering
observables beyond the single-particle inclusives.
Figure 83: RAA for pions for RHIC (left) and two possible densities at LHC (central and
right). The main result, the hybrid radiative plus collisional energy loss model for αs=0.4, is
compared to a radiative energy loss alone model for an increase value of the strong coupling.
The increased range in momentum available at the LHC enables the different slopes of the
two models to be seen.
There are still significant uncertainties in the energy loss model. The most important
kinematic region for evaluation of both the collisional and radiative energy losses are for
energy and momentum transfers from the medium greater than µD, the Debye mass. This is
the region in which we know the least about the physics of the QGP: beyond the HTL region,
but before a region of vacuum gluon exchange, especially if processes close to the light-cone
of the exchanged gluon are important (as it is for collisional energy loss). This can produce
an uncertainty of ≈50% for the average collisional loss, which may not be correlated with an
uncertainty in the radiative loss. Such large uncertainties affect both the explanation of RHIC
data and the extrapolation to the LHC.

Heavy Ion Collisions at the LHC - Last Call for Predictions 106
Figure 84: RAA for observable products of heavy quark jets at RHIC (electrons - left) and two
possible densities at the LHC (D and B mesons - right). There is considerable uncertainty
in the perturbative production of c and b jets. This shows up in the results for electrons
at RHIC in the large uncertainty band, ±0.1 or greater - as the ratio of c to b jets is very
uncertain. However, the uncertainty in D and B meson RAAs is small (approximately ±0.02) -
the different slopes on the individual spectra have very little effect on the meson RAA results.
6.10. Jet evolution in the Quark Gluon Plasma
H. J. Pirner, K. Zapp, J. Stachel, G. Ingelman and J. Rathsman
Jet evolution is calculated in the leading log approximation. We solve the evolution equation
for the branching of gluons in vacuum, using a triple differential fragmentation function D(x,Q2,p2
⊥).
Adding an extra scattering term for evolution in the quark gluon plasma we investigate the influence
of the temperature of the plasma on the differential cross section of partons dN/dln(1/x) in a jet of
virtuality Q2=(90 GeV)2. Due to scattering on the gluons in the plasma the multiplicity increases, the
centroid of the distribution shifts to smaller xvalues and the width narrows.
The evolution equation for the transition of a parton iwith virtuality Q2and momentum
(1,k⊥) into a parton jwith momentum (z,p⊥) can be constructed in leading logarithmic
approximation [245]. In a dense medium they are modified due to the possibility that the
parton is scattered. The scatterings change the transverse momentum of the leading fast parton
by giving it ~q⊥kicks, but they do not change the mass scale or virtuality of the fast parton.
The lifetime of a virtual parton can be estimated as dτ=E/Q2
0(dQ2/Q2) using the uncertainty
principle (Eis the parton energy and Q2
0is the infrared scale). Evolving along a straight line
path in a homogeneous plasma with a density of gluons ngwe obtain a modified evolution
equation
Q2∂Dj
i(z,Q2, ~p⊥)
∂Q2=
αs(Q2)
2πZ1
z
du
uPr
i(u,αs(Q2))d2~q⊥
πδu(1−u)Q2−Q2
0
4−q2
⊥Dj
rz
u,Q2, ~p⊥−z
u~q⊥
+S(z,Q2, ~p⊥)

Heavy Ion Collisions at the LHC - Last Call for Predictions 107
with the scattering term S(z,Q2,p⊥)
S(Q2, ~p⊥)=
zEng
Q2
0Z1
zdwZd2~q⊥
dσr
i
d2~q⊥hDj
r(w,Q2, ~p⊥−w~q⊥)−Dj
r(z,Q2, ~p⊥)iδ w−z−q2
⊥
2mgE!.
The scattering term includes the probability for scattering into and out of the p⊥bin as well as
the energy loss of the parton. The gluon mass in the plasma mgis related [246] to the Debye
mass mg=1/2mD.
There is an analytic solution for the p⊥integrated equation restricted to gluons, which
give the dominant contribution to the multiplicity. The solution can be found via Mellin
transformation in a similar fashion as in vacuum [247], the running of the coupling is taken
into account.
00
2000 1
4000 2
6000 3
8000 4
10000 5
Q2ln(1/x)
0
5
10
15
0
20
2
25
4
6
8
10
multiplicity
dN/dln(1/x)
Figure 85: LHS: Multiplicity of two jets with invariant mass Q2in vacuum (dashed line), at
T=0.8 GeV (dotted line) and at T=1.0 GeV (full line)
RHS: Differential multiplicity dN/dln(1/x) of jet particles inside a jet with invariant mass
Q2=(90 GeV)2in vacuum (dashed line) and at T=1.0 GeV (full line)
One finds an increase of the multiplicity with temperature and a shift of the centroid of the
ln(1/x) distribution towards smaller x, see figure 85. The width of the distribution, however,
becomes smaller. It remains to be studied in vacuum how the choice of the parameters can be
optimized to the LEP data, for simplicity the above curves are calculated for ΛQCD =250 MeV.
It is well known that in the evolution equation the QCD scale parameter may well be adjusted.
Concerning the effects of the plasma, the form of the cross section and its dependence on
αs(Q2) has to be further investigated. The results look encouraging and serve as an analytical
model with which numerical Monte Carlo calculations can be compared.
There has been a calculation of jet evolution in the modified leading log
approximation [248] which has produced similar shapes for the differential multiplicity
distribution. The advantage of our calculation is that it takes into account the scattering term
explicitly and therefore gives results which depend on the plasma properties. The equation
can also be used to investigatethe p⊥broadening of the parton in the medium, since our input
function contains the transverse momentum as an extra variable explicitly.
Note added in proof: The calculation described in the text has been undergoing several
changes during the last months. Therefore we refer to a forthcoming publication where these

Heavy Ion Collisions at the LHC - Last Call for Predictions 108
improvements are included.
6.11. Pion and Photon Spectra at LHC
S. Jeon, I. Sarcevic and J. Jalilian-Marian
Using simple modification of jet fragmentation function that is tuned to reproduce the RHIC π0
data, we had previously predicted photon production at RHIC which is confirmed by recent PHENIX
data. Using the same parameter set, we predict hight pTpion and prompt photon spectra in Pb-Pb
collisions at LHC.
In perturbative QCD, the inclusive cross section for pion production in a hadronic
collision is given by:
Eπd3σ
d3pπ(√s,pπ)=ZdxadxbdzX
i,jFi(xa,Q2)Fj(xb,Q2)Dc/π(z,Q2
f)Eπd3ˆσi j→cX
d3pπ
where iand jlabel hadrons or nuclei and a,b,clabel partons.
In heavy-ion collisions, one needs to include nuclear effects. In our model, we take the
parton distribution function for a nucleus to be
Fa/A(x,Q2,bt)=TA(bt)Sa/A(x,Q2)Fa/N(x,Q2)
where TAis the nuclear thickness function and Sa/Ais the shadowing function (we use EKS98
parametrization).
Unfortunately, the interaction of parton-medium cannot be calculated within perturbative
QCD, but need to be modeled. The purpose of our model [249–251] is to be as simplistic as
possible so that the essential nature of the energy loss process can manifest. To achieve this
goal, we modify the fragmentation function in the following way [252]
zDc/π(z,∆L,Q2)=
N
X
n=0
Pa(n)za
nD0
c/π(za
n,Q2)+hnaiz′
aD0
g/π(z′
a,Q2
0),
where za
n=z/(1−nǫa/ET), z′
a=zET/ǫa,Nis the maximum number of collisions for which
za
n≤1 and D0
c/π is the hadronic fragmentation function. The second term comes from the
emitted gluons each having energy ǫaon the average. The average number of scatterings
within a distance ∆Lis hnai= ∆L/λa. We take λa=1 fm and ∆L=RA.Pa(n) is the Poisson
distribution function with hni=(∆L/λa).
The three energy loss models we use are ∆E=1.0 GeV (Const) ∆E=√ELPME(LPM)
and ∆E=κE(BH) per collision. For RHIC, BH (Bethe-Heitler) gives best description of π0
data, and predictions for direct photons using the same energy loss is recently found to be in
agreement with PHENIX data [253]. Within the same framework we present our predictions
for the LHC.
Photons can be either produced during the primary collision or via fragmentation. The
reason that the photon RAA behaves qualitatively differently than that of π0is because in this
energy range, the direct photons that come out of the primary collisions dominate over the
fragmentation photons. Therefore the effect of energy loss is substantially reduced compared
to the pion case.

Heavy Ion Collisions at the LHC - Last Call for Predictions 109
5 10 15 20 25 30 35 40
pT (GeV)
100
101
102
103
104
105
106
107
(1/Ncoll)Ed3σ/d3p (pb/GeV2)
PP
Const : ε=1.0 GeV
LPM : ELPM=0.351 GeV
BH : κ=6.1 %
s1/2 = 5500 GeV
EKS98 Shadowing
5 10 15 20 25 30 35 40
pT (GeV)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
(Ed3σAA/d3p)/(NcollEd3σPP/d3p)
Const : ε=1.0 GeV
LPM : ELPM=0.351 GeV
BH : κ=6.1 %
s1/2 = 5500 GeV
EKS98 Shadowing
Figure 86: Neutral pion spectrum and RAA at LHC. The energy loss parameter κis fixed by
fitting the RHIC data.
5 10 15 20 25 30 35 40
pT (GeV)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
(Ed3σAA/d3p)/(NcollEd3σPP/d3p)
π0 : κ=6.1 %
γ : κ=6.1 %
s1/2 = 5500 GeV
EKS98 Shadowing
5 10 15 20 25 30 35 40
pT (GeV)
0
0.1
0.2
0.3
γ/π0
PP
LPM : ELPM=0.351 GeV
BH : κ=6.1 %
s1/2 = 5500 GeV
EKS98 shadowing
Figure 87: Direct photon RAA and γ/π0ratio at LHC.
6.12. Transverse momentum broadening of vector bosons in heavy ion collisions at the LHC
Z.-B. Kang and J.-W. Qiu
We calculate in perturbative QCD the transverse momentum broadening of vector bosons in heavy
ion collisions at the Large Hadron Collider (LHC). We predict transverse momentum broadening of
W/Zbosons constructed from their leptonic decay channels, which should be a clean probe of initial-
state medium effect. We also predict the upper limit of transverse momentum broadening of J/ψand Υ
production as a function of Npart at the LHC energy.
Nuclear transverse momentum broadening of heavy vector bosons (γ∗,W/Z, and heavy
quarkonia) is defined as a difference between the averaged transverse momentum square
measured in nuclear collisions and that measured in collisions of free nucleons,
∆hq2
TiAB ≡ hq2
TiAB −hq2
TiNN ≈Zdq2
Tq2
Tdσ(D)
AB
dq2
T,Zdq2
TdσNN
dq2
T
.(45)
Since single scattering is localized in space, the broadening is a result of multiple parton
scattering, and is a good probe for nuclear medium properties. Because the mass scale of the
vector bosons is much larger than the characteristic momentum scale of the hot medium, the

Heavy Ion Collisions at the LHC - Last Call for Predictions 110
broadening is likely dominated by double partonic scattering as indicated in equation (45).
The broadening caused by the double scattering can be systematically calculated in terms of
high twist formalism in QCD factorization [254,255].
At the LHC energies, a lot Wand Z, and J/ψand Υwill be produced. Most reconstructed
W/Zbosons will come from their leptonic decays. Their transverse momentum broadening is
a result of purely initial-state multiple scattering. By calculating the double scattering effect,
we obtain [255,256]
∆hq2
TiW
pA =4π2αs(MW)
3λ2
WA1/3,∆hq2
TiZ
pA =4π2αs(MZ)
3λ2
ZA1/3(46)
for hadron-nucleus collisions. The λ2A1/3in equation (46) was introduced in [254] as a
ratio of nuclear four parton correlation function over normal parton distribution. The λ
is proportional to the virtuality or transverse momentum of soft gluons participating in the
coherent double scattering. For collisions with a large momentum transfer, Q, the λ2should
be proportional to ln(Q2) [256] and the saturation scale Q2
sif the active parton xis small. By
fitting Fermilab E772 Drell-Yan data, it was found that λ2
DY ≈0.01 GeV2at √s=38.8 GeV
[255]. From the λ2
DY, we estimate the value of λ2for production of a vector boson of mass
MVat the LHC energy as
λ2
V(LHC) ≈λ2
DY ln(M2
V)
ln(Q2
DY) MV/5500
QDY/38.8!−0.3
,(47)
where we used Q2
s∝1/xδwith δ≈0.3 [33] and √sNN =5500 GeV for the LHC heavy ion
collisions. For an averaged QDY ∼6 GeV, we obtain λ2
W/Z≈0.05 at the LHC energy. We
can also apply our formula in equation (46) to the broadening in nucleus-nucleus collisions
by replacing A1/3by an effective medium length Leff. We calculate Leffin Glauber model
with inelastic nucleon-nucleon cross section σin
NN =70 mb at the LHC energy. We plot our
predictions (lower set curves) for the broadening of W/Zbosons in figure 88.
0
0.5
1
1.5
2
2.5
3
3.5
1 10 102103
W
Z
Y
J/ψ
Atomic Mass
∆<qT
2> (GeV2)
0
0.5
1
1.5
2
2.5
3
3.5
0 50 100 150 200 250 300 350 400
W
Z
Y
J/ψ
Npart
∆<qT
2> (GeV2)
Figure 88: Predicted broadening (maximum broadening) for Wand Z(J/ψand Υ) production
in pA (left) and Pb-Pb (right) collisions at √sNN =5500 GeV.
Heavy quark pairs are produced at a distance scale much less than the physical size of
heavy quarkonia in high energy collisions. The pairs produced in heavy ion collisions can

Heavy Ion Collisions at the LHC - Last Call for Predictions 111
have final-state interactions before bound quarkonia could be formed. We found [256] that
with both initial- and final-state double scattering, the broadening of heavy quarkonia is close
to 2CA/CFtimes the Drell-Yan broadening in proton-nucleus collision, which is consistent
with existing data [257]. If all soft gluons of heavy ion beams are stopped to form the
hot dense medium in nucleus-nucleus collisions, final-state interaction between the almost
stationary medium and the fast moving heavy quarks (or quarkonia) of transverse momentum
qTis unlikely to broaden the qTspectrum, instead, it is likely to slow down the heavy quarks
(or quarkonia)[256]. From equation (47). weobtain λ2
J/ψ ≈0.035, and λ2
Υ≈0.049 at the LHC
energy; and we predict the maximum broadening for J/ψand Υproduction (upper set curves)
in figure 88.
6.13. Nuclear modification factors for high transverse momentum pions and protons at LHC
W. Liu, B.-W. Zhang and C. M. Ko
The inclusion of conversions between quark and gluon jets in a quark-gluon plasma
(QGP) via both elastic qg ↔gq and inelastic q¯q↔gg reactions [258] has recently been
shown to give a plausible explanation for the observed similar p/π+and ¯p/π−ratios at large
transverse momenta in both central Au+Au and pp collisions at √sNN =200 GeV [223].
Extending this study to LHC, we predict the nuclear modification factor for both protons
and pions as well as their ratios at large transverse momenta in central Pb+Pb collisions at
√sNN =5.5 TeV.
For the dynamics of formed QGP at LHC, we assume that it evolves boost invariantly
in the longitudinal direction but with an accelerated transverse expansion. Specifically, its
volume expands in the proper time τaccording to V(τ)=πR2(τ)τc, where R(τ)=R0+
a(τ−τ0)2/2 is the transverse radius with an initial value R0=7 fm, the QGP formation
time τ0=0.5 fm/c, and the transverse acceleration a=0.1c2/fm. Starting with an initial
temperature T0=700 MeV, the time dependence of the temperature is obtained from entropy
conservation, leading to the critical temperature TC=170 MeV at proper time τC=8.4 fm/c.
For a quark or gluon jet moving through the QGP, the rate for the change in its mean transverse
momentum hpTiis given by dhpTi/dτ≈γ(hpTi,T)hpTi. The drag coefficient γ(hpTi,T) is
calculated from two-body scattering with thermal quark and gluon masses and the strong
QCD coupling αs(T)=g2(T)/4π≈2.1αpert(T) from lattice calculations [259]. To take into
account the contribution from two-body radiative scattering, we multiply the calculated drag
coefficient by a factor KE∼2, which is determined from fitting the light meson nuclear
modification factor at RHIC. Because of conversion scatterings, the quark or gluon jet can
also be convertedto a gluon or quark jet with a rate given by corresponding collisional widths,
which are also calculated by using the strong QCD coupling constant and multiplying with
KC=KE∼2.
Using initial transverse momentum spectra of minijet gluons, quarks, and anti-quarks
obtained by multiplying those from the PYTHIA for pp collisions at same energy with the
number of binary collisions, we simulate the propagation of jets in the QGP using the Monte

Heavy Ion Collisions at the LHC - Last Call for Predictions 112
Figure 89: (Color online) Left window: Nuclear modification factor RAA for π+(solid line)
and proton (dashed line) in central Pb+Pb collisions at √sNN =5.5 TeV. Right window: p/π+
ratio without (dotted lines) or with jet conversions (solid lines). Dashed lines correspond to
p+p collisions at same energy.
Carlo method with test particles [258]. Resulting charged pion and proton spectra from freeze-
out quark and gluon jets are obtained via the AKK fragmentation functions [14]. In the left
window of figure 89, we show predicted nuclear modification factor RAA for π+and pat
large transverse momenta in central Pb+Pb collisions at √sNN =5.5 TeV at LHC. It is seen
that the RAA of pions increases from 0.18 at pT=5 GeV/cto 0.4 at pT=40 GeV/cdue to
a smaller drag coefficient at large transverse momenta. The RAA of protons has a similar
behavior, but its value is smaller because of stronger suppression of gluon than quark jets.
The resulting p/π+ratio, shown by the solid line in the right window of figure 89, approaches
that in pp collisions at same energy when the transverse momenta become very large. At
lower transverse momenta, the p/π+ratio in Pb+Pb collisions remains, however, smaller than
that in pp collisions, which is different from that in heavy ion collisions at RHIC as a result of
the larger ratio of gluon to quark jets at LHC. Without conversions between quark and gluon
jets, the p/π+ratio decreases by a factor of two as shown by the dotted line.
6.14. Quenching of high-pThadrons: Alternative scenario
B .Z. Kopeliovich, I. K. Potashnikova and I. Schmidt
A new scenario, alternative to energy loss, for the observed suppression of high-pThadrons
observed at RHIC is proposed. In the limit of a very dense medium crated in nuclear collisions
the mean free-path of the produced (pre)hadron vanishes, and and the nuclear suppression, RAA is
completely controlled by the production length. The RHIC data are well explained in a parameter free
way, and predictions for LHC are provided.
The key assumption of the energy loss scenario for the observed suppression of high-pT
hadrons in nuclear collisions is a long length of the quark hadronization which ends up in the

Heavy Ion Collisions at the LHC - Last Call for Predictions 113
medium. This has got no justification so far and was challenged in [260].
The quark fragmentation function (FF) was calculated in Born approximation in [261]:
∂DBorn
π/q(z)
∂k2∝1
k4(1−z)2,(48)
where kand zare the transverse and fractional longitudinal momenta of the pion. One can
rewrite this in terms of the coherence length lc=z(1−z)E/k2, where Eis the jet energy. Then,
∂DBorn
π/q(z)/∂lc∝(1 −z), is lcindependent. Inclusion of gluon radiation leads to the jet lag
effect [262] which brings lcdependence,
∂Dπ/q(z)
∂lc∝(1−˜z)S(lc,z).(49)
Here ˜z=z[1 + ∆E(lc)/E] accounts for the higher Fock components of the quark, which are
incorporated via the vacuum energy loss ∆E(lc) calculated perturbatively with a running
coupling. The induced energy loss playing a minor role is added as well. S(lc,z) is the
Sudakov suppression caused by energy conservation. Fig. 90 shows an example for the lc-
distributions calculated for z=0.7 and different jet energies at √s=200GeV.
0
0.1
0.2
0.3
0 1 2 3 4 5 6 7 8 9 10
lc (fm)
dD/dlc
E=6 GeV
E=20 GeV
0
0.1
0.2
0.3
0.4
0.5
0 2 4 6 8 10 12 14 16 18 20
PT (GeV)
RAA(PT)
Figure 90: Left: ∂D(z)/∂lc(in arbitrary units) at jet energies 6,10,16,20GeV and z=0.7.
Right: Pion suppression in central AA collisions (A∼200) at √s=200GeV (solid) and
√s=5500GeV (dashed). Data are from the PHENIX experiment.
The pre-hadron, a ¯qq dipole, may be produced with a rather large initial separation
hr2
0i ≈ 2lc/E+1/E2and it keeps expanding.
To keep calculations analytic we consider a central, b=0, collision of identical heavy
nuclei with nuclear density ρA(r)=ρAΘ(RA−r). Then we find,
RAA =hl2
ci
R2
A"1−AL
hlci+BL2
hl2
ci#,(50)
where the effective absorption length has the form, L3=3pT/(8ρ2
ARAX), and Xincludes the
unknown density of the medium and is to be fitted to data on RAA. However. if the medium
is very dense, i.e. Xis large, the last two terms in (50) can be neglected, and we can predict
RAA,
Rh
AA =hl2
ci
R2
A
.(51)

Heavy Ion Collisions at the LHC - Last Call for Predictions 114
With this expression we calculated RAA at the energies of RHIC and LHC and in fig. 90
(right). This parameter free result well agrees with the data supporting the assumption that
the medium is very dense. Summarizing:
•The A-dependence, eq. (51), predicts RAA ≈0.42 for Cu−Cu confirmed by data.
•Vacuum radiation which depends only on the current trajectory should be flavor
independent. This fact and the above consideration explains the strong suppression for
heavy flavors observed at RHIC.
•Since the strength of absorption does not affect RAA, eq. (51), a single hadron and a pair
of hadrons should be suppressed equally.
•The observed suppression RAA may not contain much information about the medium
properties, except it is very dense.
6.15. Expectations from AdS/CFT for Heavy Ion Collisions at the LHC
H. Liu, K. Rajagopal and U. A. Wiedemann
We summarize results obtained by use of the AdS/CFT correspondence for jet quenching and
quarkonium dissociation, and we discuss the resulting expectations for heavy ion collisions at the
LHC.
The AdS/CFT correspondence maps nonperturbative problems in a large class of strongly
coupled non-abelian gauge theories onto calculable problems in dual gravity theories. The
gravity dual of Quantum Chromodynamics is not known. However, one finds many
commonalities amongst the quark-gluon plasmas in large classes of strongly coupled non-
abelian thermal gauge theories, independent of their significantly differing microscopic
degrees of freedom and interactions. Since these results are generic and do not seem to depend
on microscopic features of the theory such as its particle content at weak coupling, one may
expect that they are shared by QCD. Where this can be tested against QCD lattice results,
the qualitative agreement is fair (see Ref. [216]). However, many measurements in heavy ion
collisions involve strong coupling and real-time dynamics, where lattice QCD results are not
available or in their infancy. The practitioner faces the uncomfortable choice of calculating
either with inappropriate (e.g. perturbative) techniques in QCD, or using appropriate strong
coupling techniques but working in a class of gauge theories that may not include QCD itself
and seeking universal commonalities. We report on two results from the latter approach.
6.15.1. Jet quenching In QCD itself, the jet quenching parameter ˆqhas not been calculated
in the strong coupling regime. For the N=4 SYM theory, it has been calculated for large
t’Hooft coupling λ=g2Ncby use of the AdS/CFT correspondence [263]:
ˆqS Y M =π3/2Γ(3/4)
Γ(5/4) √λT3.(52)
If one relates this to QCD by fixing Nc=3 and αS Y M =.5, then ˆqS Y M =32.7T3=4.5 GeV2/fm
at T=300 MeV. This shows that a medium characterized by a momentum scale Tcan give

Heavy Ion Collisions at the LHC - Last Call for Predictions 115
rise to an apparently large quenching parameter, significantly larger than T3. For a certain
infinite class of theories with gravity dual, one finds that the quenching parameter scales with
the square root of the entropy density [263]. Assuming that QCD follows this systematic, one
finds
ˆqQCD =rsQCD
sN=4ˆqN=4=r47.5
120 ˆqN=4≃0.63 ˆqN=4.(53)
In extrapolating from RHIC to LHC, we assume that the change in ˆqis dominated by
the change in T3, see eq. (52). In the presence of expansion, the relevant temperature
Tat RHIC and at the LHC must be compared at the same time τ. This can be seen,
e.g., in a Bjorken expansion scenario in which T(τ)=T0(τ0/τ)1/3. The time-averaged
ˆq=(2/L2)RL
0dττ ˆq(τ), which determines parton energy loss and which is the quantity that
has been extracted by comparison with RHIC data, is then ˆq∝(2τ0/L)T3
0=(2τ/L)T(τ)3,
independent of the reference time τ0. Since the volume of the collision region at early
times depends only on the nuclear overlap and is energy independent, we can assume that
at any particular τ,T3
LHC/T3
RHIC =(dNLHC
ch /dη)/(dNRHIC
ch /dη) and hence make the prediction
ˆqLHC =ˆqRHIC(dNLHC
ch /dη)/(dNRHIC
ch /dη).
6.15.2. Quarkonium suppression In lattice QCD, the temperature dependent potential
between a heavy quark and anti-quark has been calculated as a function of their separation
L. At finite temperature, this potential is screened above a length Ls∼0.5/T(see references
in [264]). These studies indicate that the J/Ψdissociates at a temperature between 1.5Tcand
2.5Tc. For N=4 SYM theory, one finds Ls∼0.277/T. In contrast to QCD, the calculation in
theories with gravity dual can be done also for heavy quark-antiquark pairs which are moving
with a velocity vthrough the heat bath. One finds that the screening length decreases with
increasing γ=p1/(1−v2) [264]:
Ls(v,T)≃Ls(0,T)/√γ−→ Tdiss(v)≃Tdiss(0)/√γ . (54)
So, bound states with a dissociation temperature Tdiss(v) will survive if at rest in a medium at
temperature Tif Tdiss(0) >T. Yet, they will dissociate if they move sufficiently fast through
the medium, such that Tdiss(v)<T. LHC data may test this prediction, depending on the
quarkonium formation mechanism. Let us consider three possibilities for the latter: i) A parent
quark (cor b) propagates through the medium but the quarkonium forms later, outside the
medium. ii) As in (i) but with a parent gluon. iii) A quarkonium bound state forms (from either
a parent quark or gluon) and propagates through the medium. These three scenarios can be
discriminated as follows: i) The nuclear modification factor of quarkonium is the same as that
of open charm or beauty, which are known to be dominated by quark parents. It is the same for
all quarkonium bound states. ii) The nuclear modification factor of quarkonium is the same
as that of light hadrons, which at the LHC are dominated by gluon parents. Again, all bound
states are equally suppressed. iii) The nuclear modification factor will differ for different
bound states, since they will dissociate for different values of the transverse momentum pT.
The hierarchy in the pT-dependence of the quarkonium suppression pattern would test (54).
For example, Υ(or J/Ψ) suppression could set in only above some pTwhile Υ′(or Ψ′) are

Heavy Ion Collisions at the LHC - Last Call for Predictions 116
suppressed even at low pT. Details of the formation mechanism cancel in ratios like Υ/Υ′,
making the pT-dependent pattern predicted by (54) visible as long as the quarkonia form in
the medium.
6.16. High-pTobservables in PYQUEN model
I. P. Lokhtin, A. M. Snigirev and C. Yu. Teplov
Predictions of PYQUEN energy loss model for high-pTobservables at the LHC are discussed.
Nuclear modification factors and elliptic flow for hard jets and high-pThadrons, medium-modified jet
fragmentation function, pT-imbalance for dimuon tagged jets, high-mass dimuon and secondary J/ψ
spectra are calculated for PbPb collisions.
In this paper, the various high-pTobservables in PbPb collisions at √sNN =5.5ATeV
are analyzed in the frame of PYQUEN partonic energy loss model [265]. The pseudorapidity
cuts for jets |ηjet|<3, charged hadrons |ηh±|<2.5 and muons |ηµ|<2.5 were applied. The
jet energy was determined here as the total transverse energy of the final particles around the
direction of a leading particle inside a cone R=p∆η2+ ∆ϕ2=0.5 (ϕis the azimuthal angle).
6.16.1. Nuclear modification factors for jet and high-pThadrons The nuclear modification
factor is defined as a ratio of particle yields in AA and pp collisions normalized on the number
of binary nucleon-nucleon collisions. Figures 91 and 92 show pT-dependences of nuclear
modification factors for inclusive charged hadrons (in central PbPb events triggered on jets
with Ejet
T>100 GeV) and for jets respectively. The number of entries and the statistical errors
correspond to the estimated event rate for one month of LHC run and a nominal integrated
luminosity of 0.5 nb−1[266]. The predicted hadron suppression factor slightly increases with
pT(>20 GeV), from ∼0.25 at pT∼20 GeV to ∼0.4 at pT∼200 GeV. This behaviour
manifests the specific implementation of partonic energy loss in the model, rather weak
energy dependence of loss and the shape of initial parton spectra. Without event triggering
on high-ETjet(s), the suppression factor is stronger (∼0.15 at 20 GeV and slightly increasing
with pTup to ∼0.3 at 200 GeV). The predicted jet suppression factor (due to partial gluon
bremsstrahlung out of jet cone and collisional loss) is about 2 and almost independent on jet
energy. It is clear that the measured jet nuclear modification factor will be very sensitive to
the fraction of partonic energy loss carried out of the jet cone.
6.16.2. Medium-modified jet fragmentation function The “jet fragmentation function” (JFF),
D(z), is defined as the probability for a given product of the jet fragmentation to carry a
fraction zof the jet transverse energy. Figure 93 shows JFF’s in central PbPb collisions with
and without partonic energy loss. The number of entries and the statistical errors correspond
again to the estimated event rate for one month of LHC run. Significant softening of the JFF
(by a factor of ∼4 and slightly increasing with z) is predicted.
The medium-modified JFF is sensitive to a fraction εof partonic energy loss carried out
of the jet cone. Figure 94 shows the ε-dependences of jet nuclear modification factor Rjet
AA and

Heavy Ion Collisions at the LHC - Last Call for Predictions 117
, GeV/c
h
T
p
0 20 40 60 80 100120140160180200
h
AA
R
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Figure 91: The nuclear modification factor
for charged hadrons in central PbPb colli-
sions triggered on jets with Ejet
T>100 GeV.
, GeV
jet
T
E
100 150 200 250 300 350 400 450 500
jet
AA
R
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Figure 92: The nuclear modification factor
for jets in central PbPb collisions.
jet
T
/p
h
T
z=p
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
D(z)
1
10
2
10
3
10
4
10
5
10 PbPb, no energy loss
PbPb, PYQUEN energy loss
Jet fragmentation function
Figure 93: Jet fragmentation function for
leading hadrons in central PbPb collisions
triggered on jets with Ejet
T>100 GeV without
(squares) and with (circles) partonic energy
loss.
Figure 94: Jet nuclear modification factor
(solid curve) and ratio of JFF with loss to JFF
without loss (dashed, dash-dotted curves) as
a function of ε(see text).

Heavy Ion Collisions at the LHC - Last Call for Predictions 118
ratio of JFF with energy loss to JFF without loss, DAA(z>z0)/Dpp(z>z0), for z0=0.5 and 0.7.
If εclose to 0, then Rjet
AA ∼1 (there is no jet rate suppression), and JFF softening is maximal.
Increasing εresults in stronger jet rate suppression, but effect on JFF softening becomes
smaller. Indeed, final jet transverse momentum (which is the denominator in definition of
z) decreases in this case without an influence on the numerator of zand, as a consequence,
the effect on JFF softening reduces, while the integral jet suppression factor becomes larger.
Thus a novel study of the softening of the JFF and suppression of the absolute jet rates can be
carried out in order to differentiate between various energy loss mechanisms (“small-angular”
radiative loss versus “wide angular” and collisional loss) [267].
Other correlation measurements which also can be useful extracting information about
medium-modified jets are jet shape broadening and jet quenching versus rapidity [268] and
monojet-to-dijet ratio versus dijet acoplanarity [269].
6.16.3. Azimuthal anisotropy of jet quenching The azimuthal anisotropy of particle spectrum
is characterized by the second coefficient of the Fourier expansion of particle azimuthal
distribution, elliptic flow coefficient, v2. The non-uniform dependence of medium-induced
partonic energy loss in non-central heavy ion collisions on the parton azimuthal angle ϕ(with
respect to the reaction plane) is mapped onto the final hadron spectra [270, 271]. Figure 95
shows the calculated impact parameter dependence of v2coefficient for jets with Ejet
T>100
GeV and for inclusivecharged hadrons with pT>20 GeV/cin PbPb events triggered on jets.
The absolute values of v2for high-pThadrons is larger that one’s for jets by a factor of ∼2−3.
However, the shape of b-dependence of vh
2and vjet
2is similar: it increases almost linearly with
the growth of band becomes a maximum at b∼1.6RA(where RAis the nucleus radius). After
that, the v2coefficients drop rapidly with increasing b.
6.16.4. PT-imbalance in dimuon tagged jet events An important probe of medium-induced
partonic energy loss in ultrarelativistic heavy ion collisions is production of a single jet
opposite to a gauge boson such as γ⋆/Z0decaying into dileptons. The advantage of such
processes is that the mean initial transverse momentum of the hard jet equal to the mean
initial/final transverse momentum of boson, and the energy lost by the parton can be estimated
from the observed pT-imbalance between the leading particle in a jet and the lepton pair.
Figure 96 shows the difference between the transverse momentum of a µ+µ−pair from γ⋆/Z0
decay, pµ+µ−
T, and five times the transverse energy of the leading particle in a jet (since the
average fraction of the parent parton energy carried by a leading hadron at these energies
is z≈0.2) for minimum bias PbPb collisions [272]. The cuts, pµ+µ−
T,Ejet
T>50 GeV/cwere
applied. Despite the fact that the initial distribution is smeared and asymmetric due to initial-
state gluon radiation, hadronization effects, etc., one can clearly see the additional smearing
and the displaced mean and maximum values of the pT-imbalance due to partonic energy loss.
The pT-imbalance between the µ+µ−pair and a leading particle in a jet is directly related to
the absolute value of partonic energy loss, and almost insensitive to the form of the angular
spectrum of the emitted gluons and to the experimental jet energy resolution [272].

Heavy Ion Collisions at the LHC - Last Call for Predictions 119
b, fm
0 2 4 6 8 10 12 14
2
v
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
jets
charged hadrons
Figure 95: The impact parameter dependence
of elliptic flow coefficients vjet
2for jets
with Ejet
T>100 GeV (black circles) and vh
2
for inclusive charged hadrons with pT>
20 GeV/c(open circles) in PbPb events
triggered on jets.
, GeV
lead
T
) - 5E
-
µ
+
µ(
T
=p
T
p∆
-100-80 -60 -40 -20 0 20 40 60 80 100
)
T
p∆dN/d(
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01 PbPb, no energy loss
PbPb, PYQUEN
energy loss
) + jet
-
µ
+
µ→(
0
/Z
*
γ
Figure 96: The distribution of the difference
between the transverse momentum of a µ+µ−
pair and five times the transverse energy of
the leading particle in a jet in PbPb collisions
with (dashed histogram) and without (solid
histogram) energy loss.
6.16.5. High-mass dimuon and secondary J/ψ spectra While the study of inclusive high-pT
jet production in heavy ion collisions provides information on the response of created medium
to gluons and light quarks, the study of open heavy flavour production gives corresponding
information on massive colour charges. The open charm and bottom semileptonic decays are
the main sources of muon pairs in the resonance-free high invariant mass region, 10 <Mµ+µ−<
70 GeV/c2[266]. Other processes which also carry information about medium-induced
bottom rescattering are secondary J/ψ production from the B meson decay [273, 274] and
muon tagged b-jets [275]. Figures 97 and 98 show the spectra of high-mass µ+µ−pairs and the
pT-distributions of the secondary J/ψ’s respectively, for minimum bias PbPb collisions with
and without energy loss of bottom quarks (pµ
T>5 GeV/c). A factor of around 2.5 suppression
for b¯
b→µ+µ−and 2 for secondary J/ψ would be clearly observed over the initial state nuclear
shadowing expected in this kinematic region [273, 274].
6.17. Predictions for LHC heavy ion program within finite sQGP formation time
V. S. Pantuev
Predictions for some experimental physical observables in nucleus-nucleus collisions at LHC
energies are presented. I extend the previous suggestion that the retarded jet absorption, at RHIC by
time about 2.3 fm/c, in opaque core is a natural explanation of many experimental data. At LHC
this time should be inversely proportional to the square root of parton hard scattering density, thus
about 2 times shorter than at RHIC, or 1.2 fm/c. Predictions were done for hadrons, including charm

Heavy Ion Collisions at the LHC - Last Call for Predictions 120
2
), GeV/c
-
µ
+
µM(
20 25 30 35 40 45 50
-1
)
2
dN/dM, (GeV/c
2
10
3
10
PYTHIA
PYQUEN
-
µ
+
µ → B
B
Figure 97: Invariant mass distribution of
µ+µ−pairs from b¯
bdecays in minimum
bias PbPb collisions, with (dashed histogram)
and without (solid histogram) bottom quark
energy loss.
, GeV/c
T
p
10 12 14 16 18 20 22 24
-1
, (GeV/c)
T
dN/dp
3
10
4
10 PYTHIA
PYQUEN
)
-
µ
+
µ→ (Ψ J/→B
Figure 98: Transverse momentum spectrum
of secondary J/ψ in minimum bias PbPb col-
lisions, with (dashed histogram) and without
(solid histogram) bottom quark energy loss.
hadrons, with transverse momentum above 5 GeV/c. I calculate nuclear modification factor RAA,
azimuthal anisotropy parameter v2, jet suppression IAA for the away side jet and its dependence versus
the reaction plane orientation. The system under consideration is Au+Au at central rapidities.
In previous paper [276] I propose a simple model, driven by experimental data, to
explain the angular dependence of the nuclear modification factor RAA at high transverse
momentum in and out the reaction plane. I introduce one free parameter L≃2.3 fm to
describe the the thickness of the corona area with no absorption wich was adjusted to fit
the experimental data of Au-Au collisions at centrality 50-60%. The model uses realistic
Woods-Saxon nuclear density distribution and nicely describes the RAA dependence for all
centrality classes. I extract the second Fourier component amplitude, v2, for high pTparticle
azimuthal distribution and found v2should be at the level of 11-12% purely from the geometry
of the collision with particle absorption in the core. At that time I made a prediction for
RAA in Cu+Cu collisions at 200 GeV which, as later was found, is in very good agreement
with experimental data. Physical interpretation of the parameter Lcould be that it is actually
retarded jet absorption caused by the plasma formation time T=L/c≃2.3 fm/cat RHIC,
or at least non-trivial response of strongly interacting plasma to fast moving color charge.
From experimental data at 62 GeV center-of-mass beam energy I found that this time
should be about 3.5 fm/c. This follows the expectation on the significance of mean distance
between the centers with mini jet production (hard scatterings) at particular beam energy. At
LHC energy of about 5 TeV we expect ≃1.2 fm/cformation time [277].
In figure 99 I show predictions for RAA and IAA at central rapidities. As usual, nuclear

Heavy Ion Collisions at the LHC - Last Call for Predictions 121
modification factor RAA is defined as:
RAA(pT)=(1/Nevt) d2NAA/dpTdη
(hNbinaryi/σNN
inel) d2σNN/dpTdη,
where hNbinaryiis a number of binary nucleon-nucleon collisions at particular centrality class.
Npart
0 50 100 150 200 250 300 350 400
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
62
200
5500
Raa vs. Npart for AuAu at 62, 200 and 5500 GeV
Npart
0 50 100 150 200 250 300 350 400
, di-jets
AA
I
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
di-jets in plane
di-jets out of plane
Figure 99: RAA, nuclear modification factor, (left) and IAA, suppression of away-side
jet compared to pp data versus number of participants, right. Hadrons are at transverse
momentum 5 to 20 GeV/c. Width of the away side jet was assumed to be σ=0.22 radians.
In all cases I consider hadrons (mesons and baryons, including charm) at pTabove 4-
5 GeV/c.RAA(pT) at a such momentum should be independent on pT, flat distribution at least
to 20 GeV/c.
IAA is defined as a ratio of away-side yield per trigger high pTparticle to the similar
value from pp collisions. The major feature of this model is the dominant tangential back to
back di-jet production from the surface region. Because of that we may expect significantly
larger di-jet production out of the reaction plane, figure 99, in contrast to punch through
jet scenario. Predictions for azimuthal asymmetry parameter v2are shown in figure 100.
Npart
0 50 100 150 200 250 300 350 400
v2 (%)
0
1
2
3
4
5
6
7
8
> 5 GeV/c
T
LHC, p
Figure 100: Azimuthal assimetry parameter v2for mesons and baryons at transverse
momentum between 5 to 20 GeV/c.

Heavy Ion Collisions at the LHC - Last Call for Predictions 122
6.18. Hadrochemistry of jet quenching at the LHC
S. Sapeta and U. A. Wiedemann
We point out that jet quenching can leave signatures not only in the longitudinal and transverse
multiplicity distributions, but also in the hadrochemical composition of the jet fragments. As a
theoretical framework, we use the MLLA+LPHD formalism, supplemented by medium-modified
splitting functions.
In heavy ion collisions at the LHC, the higher energies of produced jets will facilitate
their separation from the soft background. The interactions of jets with the matter produced
in these collisions is expected to modify both the longitudinal and transverse jet distributions.
In addition, we expect that these interactions affect also the hadrochemical composition of
jets. Within current models of jet quenching, this may be expected, since color is transfered
between the projectile and the medium - and a changed color flow in the parton shower can
be expected to change the hadronization. More generally, one may imagine that partonic
fragments of the jet participate in hadronization mechanisms not available in the vacuum (such
as a recombination mechanism, which depends on the density of recombination partners), or
that recoil effects kick components of the medium into the jet cone. Also, any exchange
of quantum numbers between medium and jet (e.g. baryon number or strangeness) may be
reflected in the hadrochemical composition. In the following, we consider a model which
does not implement such mechanisms, but considers solely the enhanced parton splitting due
to medium effects [278].
To calculate multiplicities of the identified hadrons we use the framework of
Modified Leading Logarithmic Approximation (MLLA) [279]. This perturbative approach
supplemented by the hypothesis of Local Parton-Hadron Duality (LPHD) was shown to
reproduce correctly the single inclusive hadron spectra in jets both in e+e−and pp/p¯p
collisions. It provides good description not only for the distributions of all charged particles
but also for the spectra of identified hadrons such as pions, kaons and protons [280,281].
Moreover, the dependence on jet opening angle can be implemented. The general form of the
multiplicity of hadrons of mass Mhin the jet of energy Ejet and opening angle θcis given by
dNh
dξ=KLPHD D(ξ, Ejet,θc,Mh,Λ),(55)
where ξ=ln1/xand x=p/Ejet is the fraction of the jet energy carried by the hadron h. The
regularization scale Λis a parameter of the model.
The medium-modification of jets is formulated within the MLLA formalism [282] by
enhancing the singular parts of the LO splitting functions by a factor 1 +fmed. This accounts
for the nuclear modification factor at RHIC when fmed is of the order of 1, and it provides a
model for the distribution of subleading jet fragments.
One result of our studies is shown in Figure 101. We observe a significant difference
of the K±/π±and p±/π±ratios of medium-modified (with fmed =1) and ’standard’ vacuum
fragmenting jets. We have also shown, that Figure 101 remains largely unchanged if the soft

Heavy Ion Collisions at the LHC - Last Call for Predictions 123
Ratios of dNh/dp
Ejet = 50 GeV
K±/π±
Ejet = 100 GeV
K±/π±
Ejet = 200 GeV
K±/π±
vacuum jet
medium jet
p±/π±p±/π±p±/π±
p (GeV)
0
0.1
0.2
0.3
0.4
0.5
0.6
0
0.1
0.2
0.3
0.4
0.5
0.6
5 10 15 20 5 10 15 20 5 10 15 20
Figure 101: Ratios of kaons and protons to pions in the jets with or without medium
modification for jet opening angle θc=0.28 rad. The θc-dependence is weak. The kaon
multiplicity was in adjusted by a strangeness suppression factor 0.73 as in [281].
background is included in forming the ratio [278].
The precise numerical change of the hadrochemical composition, shown in Figure 101,
is model-depement of course. We emphasize, however, that in our model, medium effects are
implemented on the partonic level only, and the hadronization mechanism remains unchanged.
Nevertheless, the observed change is significant. Thus, our model provides a first example for
our expectation, that the hadrochemical composition of jets may be very fragile to medium
effects, and provides additional information about the microscopic mechanism underlying jet
quenching.
6.19. GLV predictions for light hadron production and suppression at the LHC
I. Vitev
Simulations of neutral pion quenching in Pb+Pb reactions at s1/2=5.5 A.TeV at the LHC are
presented to high transverse momentum pT. At low and moderate pT, we study the contribution
of medium-induced gluon bremsstrahlung to single inclusive hadron production. At the LHC, the
redistribution of the lost energy is shown to play a critical role in yielding nuclear suppression that
does not violate the participant scaling limit. Energy loss in cold nuclear matter prior to the formation
of the QGP is also investigated and shown to have effect on particle suppression as large as doubling
the parton rapidity density.
Pb+Pb collisions at the LHC represent thefuture energy frontier of QGP studies in heavy
ion reactions. Energy loss of jets in the final state is calculated in the GLV formalism [184].
Numerical simulations follow the technique outlined in [283] and incorporate the Cronin
effect [284]. We have explored the sensitivity of RAA(pT) to the parton rapidity density in

Heavy Ion Collisions at the LHC - Last Call for Predictions 124
central nuclear reactions with dNg/dy ≃2000,3000 and 4000. In this work we adhere to a
more modest two- to four-fold increase of the soft hadron rapidity density and emphasize that
future measurements of jet quenching must be correlated to dNg/dy ≈(3/2)dNch/dy [283,
284] to verify the consistency of the phenomenological results. See left panel of Fig. 102.
The contribution of the bremsstrahlung gluons to low- and moderate-pTinclusive particle
production at the LHC is shown to be significant. See right panel of Fig. 102.
Energy loss of jets in cold nuclear matter has not been considered before. Recent
calculations in the GLV approach show that, in contract to final-state energy loss, the
cancellation of the bremsstrahlung in the initial-state is finite [285]. With ∆E/E∼few %, the
observable effect of the bremsstrahlung associated with the multiple soft scattering in nuclei
is non-negligible even for very energetic partons in the nuclear rest frame. See left panel of
Fig. 103. At the LHC, in central Pb+Pb collisions, the effect of cold nuclear matter energy
loss can be as large as doubling the parton rapidity density dNg/dy mainly due to reduced
sensitivity in the final state. See right panel of Fig. 103.
6.20. NLO Predictions for Single and Dihadron Suppression in Heavy-ion Collisions at LHC
E. Wang, X.-N. Wang and H. Zhang
Suppresions of high transverse momentum single and dihadron spectra at LHC are calculated
within a next-to-leading order perturbative QCD model with energy parton energy loss.
The predictions presented here are calculated within a NLO pQCD Monte Carlo based
program [286]. For the study of large pTsingle and dihadron production in A+Acollisions,
0 50 100 150 200
pT [GeV]
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
RAA(pT)
π0 in central Pb+Pb, GLV E-loss dNg/dy = 2000
π0 in central Pb+Pb, GLV E-loss dNg/dy = 3000
π0 in central Pb+Pb, GLV E-loss dNg/dy = 4000
0 50 100 150
pT [GeV]
10-12
10-10
10-8
10-6
10-4
10-2
dσ/dyd2pT [mb.GeV-2]
π0 in p+p at s1/2 = 5.5 TeV
π0 in p+p at s1/2 = 200 GeV
s1/2 = 5.5 TeV
-
0 1 2 3 4 5
pT [GeV]
0
0.1
0.2
0.3
0.4
0.5
RAA(pT)
dNg/dy = 1175 with gluon feedback
dNg/dy = 1175 without gluon feedback
PHENIX prelim. π0 in 0-10% Au+Au
0 5 10 15 20
pT [GeV]
0
0.1
0.2
0.3
0.4
RAA(pT)
dNg/dy = 2000 with(out) gluon feedback
dNg/dy = 3000 with(out) gluon feedback
dNg/dy = 4000 with(out) gluon feedback
s1/2 = 200 GeV
s1/2 = 5.5 TeV Below participant scaling
Below participant scaling
Figure 102: Left panel: Suppression of π0production in central Pb+Pb collisions at the LHC
as a function of the parton rapidity density. Insert shows the baseline p+pπ0cross section at
√s=200 GeV and √s=5.5 TeV [283, 284]. Right panel: nuclear modification factor RAA
in central Au+Au collisions at moderate pTwith (solid line) and without (dashed line) gluon
feedback, dNg/dy =1175. Central Pb+Pb collisions with (solid line) and without (dashed
line) gluon feedback are shown, dNg/dy ≃2000,3000,4000 [284].

Heavy Ion Collisions at the LHC - Last Call for Predictions 125
we assume that the initial hard scattering cross sections are factorized as in p+pcollisions.
We further assume that the effect of final-state interaction between produced parton and the
bulk medium can be described by the effective medium-modified FF’s. The total parton energy
loss in a finite and expanding medium is approximated as a path integral,
∆E≈ hdE
dL i1dZ∞
τ0
dττ−τ0
τ0ρ0ρg(τ,b,r+nτ),(56)
for a parton produced at a transverse position rand traveling along the direction n.hdE/dLi1d
is the average parton energy loss per unit length in a 1-d expanding medium with an initial
uniform gluon density ρ0at a formation time τ0for the medium gluons. The energy
dependence of the energy loss is parameterized as
hdE
dL i1d=ǫ0(E/µ0−1.6)1.2/(7.5+E/µ0),(57)
from the numerical results in Ref. [287,288]. The parameter ǫ0should be proportional to the
initial gluon density ρ0. The gluon density distribution in a 1-d expanding medium in A+A
collisions at impact-parameter bis assumed to be proportional to the transverse profile of
participant nucleons ,
ρg(τ,b,r)=τ0ρ0
τ
πR2
A
2A[tA(r)+tA(|b−r|)].(58)
In fitting the RHIC data [289] we have chosen the parameters as µ0=1.5 GeV, τ0=0.2 fm/c
and ǫ0=1.68 GeV/fm. We assume ǫ0is proportional to the final multiplicity density and
ǫ0=5.6 GeV/fm in the central Pb+Pb collisions √s=5.5 TeV.
0.01
0.1
1
∆p+ / p+
µ = 0.35 GeV, λg = 1 fm
µ = 0.70 GeV, λg = 4 fm
0.01
0.1
1
∆p+ / p+
100101102103104105106
p+/2 ~ Ejet [GeV]
0.001
0.01
0.1
1
∆p+ / p+
Bertsch-Gunion E-loss
Initial state E-loss
Final state E-loss L = 5 fm
Q0 = mN = 0.94 GeV
Quark jets
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
RAA(pT)
dNg/dy = 2000 + Cold E-loss
dNg/dy = 4000 + Cold E-loss
050 100 150 200 250 300
pT [GeV]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
RAA(pT)
dNg/dy = 2000, no Cold E-loss
dNg/dy = 4000, no Cold E-loss
Figure 103: Left panel: Comparison of Bertsch-Gunion, initial-state and final-state quark
energy loss in a large nucleus, such as Au or Pb. The cancellation of initial-state energy
loss is finite and cannot be neglected even at high parton energies [285]. Right panel:
Effects of could nuclear matter energy loss on suppressed π0production in central Pb+Pb
collisions at the LHC. Two parton rapidity densities dNg/dy ≃2000,4000. are shown; cold
nuclear matter energy loss effects can be as large as the effect of doubling the parton rapidity
density [284,285].

Heavy Ion Collisions at the LHC - Last Call for Predictions 126
We use the factorization scale µ=1.2pTin both p+pand A+Acollisions in our
calculation. Shown in Fig. 104(a) are the single π0spectra in both p+pand Pb +Pb
collisions at √s=5.5 TeV and the corresponding nuclear modification factor, RAA =
dσAA/dp2
Tdy[Rd2bTAA(b)dσNN/dp2
Tdy]−1.
Figure 104: (a) π0spectra and suppression factor in Pb +Pb(0 −5%) collisions at √s=5.5
TeV. (b)Hadron-triggered fragmentation functions DAA(zT) and the medium modification
factors IAA(zT) in NLO pQCD in central Pb +Pb collisions at √s=5.5 TeV.
The hadron-triggered fragmentation function,
DAA(zT,ptrig
T)≡ptrig
Tdσh1h2
AA /dytrigdptrig
Tdyassodpasso
T[dσh1
AA/dytrigdptrig
T]−1as a function of zT=
passo
T/ptrig
Tis essentially the away-side hadron spectrum associated with a triggered hadron
within |ytrig,asso|<0.5 and the azimuthal angle relative to the triggered hadron is integrated
over |∆φ|>2.5.
The factorization scale in the NLO calculation of dihadron spectra is chosen to be
µ=1.2M, where Mis the invariant mass of the dihadron M2=(p1+p2)2. The associated
hadron spectra Dpp(zT,ptrig
T) in p+pand central Pb +Pb collisions at √s=5.5 TeV and the
suppression factor IAA =DAA(zT,ptrig
T)[Dpp(zT,ptrig
T)]−1for central Pb+Pb collision at LHC
are shown in Fig. 104(b).
7. Heavy quarks and quarkonium
7.1. Statistical hadronization model predictions for charmed hadrons
A. Andronic, P. Braun-Munzinger, K. Redlich and J. Stachel
We present predictions of the statistical hadronization model for charmed hadrons production in
Pb+Pb collisions at LHC.

Heavy Ion Collisions at the LHC - Last Call for Predictions 127
The results presented below are discussed in detail in our recent publication [290].
We summarize here the values of the model parameters: i) characteristics at chemical
freeze-out: temperature, T=161±4 MeV; baryochemical potential, µb=0.8+1.2
−0.6MeV; volume
corresponding to one unit of rapidity V=6200 fm3; ii) charm production cross section:
σpp
c¯c/y=0.64+0.64
−0.32 mb.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-4 -3 -2 -1 0 1 2 3 4
y
dNJ/ψ /dy
√sNN =5.5 TeV, Npart=350
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
50 100 150 200 250 300 350
Npart
100 x (dNJ/ψ /dy) / (dNcc /dy)
dσcc/dy (mb)
0.32
0.43
0.64
0.85
1.28
Figure 105: Predictions for J/ψ yield: rapidity distribution for central collisions (left panel)
and centrality dependence of the yield relative to the charm production yield for different
values of the charm cross section indicated on the curves (right panel).
In Fig. 105 we present predictions for the yield of J/ψ. The left panel shows the rapidity
distribution with the band reflecting the uncertainty in the charm production cross section.
The right panel shows the centrality dependence of the yield relative to the charm production
yield for five values of the input charm cross section.
The statistical hadronization model predictions for charmed hadron yield ratios in central
Pb+Pb collisions at LHC are shown in Table 5. We expect that these ratios are independent
of centrality down to values of Npart ≃100.
Following from our model assumption of charm quark thermalization and assuming
decoupling of charm at hadronization, the transverse momentum spectra of charmed hadrons
can be calculated [290]. As seen in Fig. 106, a precision measurement of the spectrum of J/ψ
meson will allow the determination of the expansion velocity in QGP.
7.2. Nuclear suppression for heavy flavors in PbPb collisions at the LHC
N. Armesto, M. Cacciari, A. Dainese, C. A. Salgado and U. A. Wiedemann
We predict the nuclear suppression factors for D and B mesons, and for electrons from their semi-
leptonic decays, in PbPb collisions at the LHC. The results are obtained supplementing a perturbative
next-to-leading order +next-to-leading log (FONLL) calculation with appropriate non-perturbative
fragmentation functions and radiative energy loss.

Heavy Ion Collisions at the LHC - Last Call for Predictions 128
Table 5: Predictions of the statistical hadronization model for charmed hadron ratios for
Pb+Pb collisions at LHC. The numbers in parantheses represent the error in the last digit(s)
due to the uncertainty of T.
D−/D+¯
D0/D0D∗−/D∗+D−
s/D+
s¯
Λc/ΛcD+/D0D∗+/D0
1.00(0) 1.01(0) 1.01( 0) 1.00(1) 1.00(1) 0.425(18) 0.387(15)
D+
s/D0Λc/D0ψ′/ψ ηc/ψ χc1/ψ χc2/ψ
0.349(14) 0.163(16) 0.031(3) 0.617(14) 0.086(5) 0.110(8)
10 -5
10 -4
10 -3
10 -2
10 -1
0 1 2 3 4 5 6 7 8
pt (GeV/c)
dNJ/ψ /dpt (a.u.)
T=160 MeV
β0.3
0.4
0.5
pp
–, √s =1.96 GeV (CDF data, prel.)
Figure 106: Predictions for momentum
spectrum of J/ψ meson for different val-
ues of the average expansion velocity, β,
for central Pb+Pb collisions (Npart=350).
Also included is the measured spectrum in
p¯p collisions at Tevatron [291], which is
used to calculate the contribution from the
corona (see ref. [290]).
Medium-induced gluon radiation is usually identified as the dominant source of energy
loss of high-pTparticles traversing a hot medium. Different models which use
different approximations to this physical mechanism of energy loss, provide a successful
phenomenological description of available experimental data for light hadron suppression.
Most of these calculations assume independent multiple gluon emission to model
the exclusive distributions essential to compute the suppression which convolutes the
fragmentation functions with a steeply falling perturbative spectrum. This convolution biases
the observed particle yields to small in-medium energy losses and surface emission which,
on the other hand, leads to a lack of precision in the determination of the medium parameters
[80, 231]. The value of the transport coefficient obtained in these approaches, by using the

Heavy Ion Collisions at the LHC - Last Call for Predictions 129
multiple soft scattering approximation is [80,231]
ˆq=5÷15GeV2/fm.(59)
One proposal to increase the sensitivity to the value of the transport coefficient is to measure
the corresponding effects on heavy mesons as the formalism predicts a calculable hierarchy
of energy losses ∆Eg>∆Eq
mq=0>∆EQ
mQ,0, due color factors for the first and mass terms
for the second inequality [292]. The implementation of mass effects does not add any new
parameter to the calculation once the transport coefficient ˆqis fixed by e.g. light hadrons
(59). The description of non-photonic electrons data from RHIC given by this formalism
is reasonable [293] although the uncertainties in the benchmark relative contribution from
beauty and charm quarks are still large.
0
0.2
0.4
0.6
0.8
1
0 10 20 30 40 50
RAA
pT (GeV)
mesons from charm
q = 10
q = 25
q = 100
0
0.2
0.4
0.6
0.8
1
0 10 20 30 40 50
RAA
pT (GeV)
electrons from bottom
q = 10
q = 25
q = 100
Figure 107: RAA for D’s (left) and for electrons coming from bottom decays (right) at y=0
for 10% PbPb collisions at √s=5.5 TeV/A, for different ˆq(in GeV2/fm).
At the LHC, where charm and beauty suppression will be measured separately, the
situation will be improved and a definite check on the influence of mass terms in the medium-
induced gluon radiation and the corresponding energy loss will be done. We here present
predictions based on the formalism developed in references [80, 231, 292–294], for 0 −10%
and 30−60% PbPb collisions at LHC energy. This also updates the calculations in [294]
by taking into account the FONLL baseline for the perturbative calculation (see [293] and
references therein).
In Figs. 107 and 108 we present our predictions for RAA, double ratios and v2, for mesons
and/or decay electrons at y=0. While the mass effects in charm are very modest, they are
clearly visible for bottom quarks at pT.20 GeV. At larger pTthey tend to disappear and the
typical suppression is that of massless particles [80, 231]. We have used several values of ˆq
ranging from 10 GeV2/fm, which is the lowest one still compatible with RHIC non-photonic
electrons data, to 100 GeV2/fm, which is our estimated upper limit, from the most extreme
extrapolation of the multiplicities at the LHC.

Heavy Ion Collisions at the LHC - Last Call for Predictions 130
0.5
1
1.5
2
2.5
3
3.5
0 10 20 30 40 50
RAA(bottom)/RAA(charm)
pT (GeV)
mesons
m = 0
q = 10
q = 25
q = 100
[GeV]
T
meson p
0 5 10 15 20 25 30 35 40 45 50
2
v
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
[GeV]
T
meson p
0 5 10 15 20 25 30 35 40 45 50
2
v
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
/fm
2
=10GeVq /fm
2
=25GeVq /fm
2
=100GeVq
Pb-Pb, 30-60%, charm mesons
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
2.8
0 10 20 30 40 50
RAA(bottom)/RAA(charm)
pT (GeV)
electrons
m = 0
q = 10
q = 25
q = 100
[GeV]
T
electron p
0 5 10 15 20 25 30 35 40 45 50
2
v
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
[GeV]
T
electron p
0 5 10 15 20 25 30 35 40 45 50
2
v
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
/fm
2
=10GeVq /fm
2
=25GeVq /fm
2
=100GeVq
Pb-Pb, 30-60%, beauty-decay electrons
Figure 108: Upper plots: double ratio for mesons (left) and decay electrons (right) for 10%
PbPb collisions at √s=5.5 TeV/A for y=0, for different ˆq(in GeV2/fm). Lower plots: v2for
D’s (left) and from electrons coming from bottom decays (right) at y=0 for 30−60% PbPb
collisions at √s=5.5 TeV/A, for different ˆq.
7.3. Heavy-quark production from Glauber-Gribov theory at LHC
I. C. Arsene, L. Bravina, A. B. Kaidalov, K. Tywoniuk and E. Zabrodin
We present predictions for heavy-quark production for proton-lead collisions at LHC energy 5.5
TeV from Glauber-Gribov theory of nuclear shadowing. We have also made predictions for baseline
cold-matter (in other words inital-state) nuclear effects in lead-lead collisions at the same energy that
has to be taken into account to understand properly final-state effects.
7.3.1. Introduction In the Glauber-Gribov theory [21] nuclear shadowing at low-xis related
to diffractive structure functions of the nucleon, which are studied experimentally at HERA.
The space-time picture of the interaction for production of a heavy-quark state on nuclei
changes from longitudinally ordered rescatterings at energies below the critical energy,
corresponding to x2of an active parton from a nucleus becoming smaller than 1/mNRA, to the
coherent interaction of constituents of the projectile with a target nucleus at energies higher
thant the critical one [295]. For production of J/ψ and Υin the central rapidity region the
transition happens at RHIC energies. In this kinematical region the contribution of Glauber-
type diagrams is small and it is necessary to calculate diagrams with interactions between
pomerons, which, in our approach, are accomodated in the gluon shadowing. A similar model

Heavy Ion Collisions at the LHC - Last Call for Predictions 131
y
-4 -3 -2 -1 0 1 2 3 4
pA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4 PHENIX (min. bias)
-
e
+
e
PHENIX (min. bias)
-
µ
+
µ
= 200 GeVsGGB @
= 5.5 TeVsGGB @
ψJ/
y
-4 -3 -2 -1 0 1 2 3 4
= 200 GeVsGGB @
= 5.5 TeVsGGB @
Υ
coll
N
0 2 4 6 8 10 12 14 16
pA
R
0.2
0.4
0.6
0.8
1
1.2
ψJ/
y = -3.5
y = 0
y = 3.5
coll
N
2 4 6 8 10 12 14 16
Υ
y = -3.5
y = 0
y = 3.5
Figure 109: Rapidity (top) and centrality (bottom) dependence of the nuclear modification
factor for J/ψ (left) and Υ(right) production in p+Pb (d+Au) collisions at √s=5500 (200)
GeV. Experimental data are from [296].
for J/ψ-suppression in d+Au collisions at RHIC has been considered in Ref. [51].
Calculation of gluon shadowing was performed in our recent paper [22], where Gribov
approach for the calculation of nuclear structure functions was used. The gluon diffractive
distributions were taken from the most recent experimental parameterizations of HERA data
[23]. The Schwimmer model was used to account for higher-order rescatterings.
7.3.2. Heavy-quark production at the LHC We present predictions for the rapidity and
centrality dependence of the nuclear modification factor in proton-lead (p+Pb) collisions
for both J/ψ and Υin Fig. 109 (the data on J/ψ suppression at √s=200 GeV is taken
from [296], where also a definition of the nuclear modification factor can be found). We
predict a similar suppression for open charm, c¯c, and bottom, b¯
b, as for the hidden-flavour
particles. The observed xFscaling at low energies of the parameter α(from σpA =σppAα) for
J/ψ production, which is broken already at RHIC, will go to a scaling in x2at higher energies.
This will also be the case for Υand open charm and bottom.
In Fig. 110 we present predictions for cold-nuclear matter effects due to gluon shadowing
in lead-lead (Pb+Pb) collisions at LHC energy √s=5.5 TeV for the production of J/ψ and
Υ. The suppression is given as a function of rapidity and centrality. .

Heavy Ion Collisions at the LHC - Last Call for Predictions 132
part
N
0 50 100 150 200 250 300 350 400
Pb+Pb
R
0
0.2
0.4
0.6
0.8
1
y = 0
y = 2.2
y = 3.5
y = 0
y = 2.2
y = 3.5
ψJ/
Υ
Figure 110: Baseline cold-nuclear matter effects in Pb+Pb collisions at 5.5 TeV for J/ψ and
Υproduction.
7.4. RAA(pt)and RCP(pt)of single muons from heavy quark and vector boson decays at the
LHC
Z. Conesa Del Valle, A. Dainese, H.-T. Ding, G. Martínez and D. Zhou
We study the effect of heavy-quark energy loss on the nuclear modification factors RAA and RCP
of the high-ptdistribution of single muons in Pb–Pb collisions at √sNN=5.5 TeV. The energy loss
of heavy quarks is calculated using the mass-dependent BDMPS quenching weights and taking into
account the decrease of medium density at large rapidity. Muons from W and Z decays, that dominate
the yield at high pt, can be used as a medium-blind reference that scales with the number of binary
collisions.
The PHENIX and STAR experiments at RHIC have measured a suppression, in central
Au–Au relative to pp collisions, of the high-ptyield of non-photonic electrons, which are
assumed to come from semi-electronic decays of charm and beauty particles. This suppression
is interpreted as an indication of a strong energy loss of c and b quarks in the medium formed
in Au–Au collisions. At the LHC, the nuclear modification factors RAA and RCP of the single-
muon inclusive ptdistribution will be among the first measurements sensitive to heavy-quark
energy loss. Moreover, the very high ptdomain (pt>30 GeV/c) of the muon spectrum will
be dominated by muonic decays of electroweak boson W (mainly) and Z, that should be
medium-insensitive and follow binary scaling, thus making of the nuclear modification factor
a self-normalized observable.
We obtain the charm and beauty contributions to the muon spectrum from the NLO
pQCD calculation (MNR [297]) supplemented with the mass-dependent BDMPS quenching
weights for radiative energy loss [294], quark fragmentation à la Peterson and semi-muonic

Heavy Ion Collisions at the LHC - Last Call for Predictions 133
decay with the spectator model. We account for the medium density decrease at large rapidity
by assuming the transport coefficient to scale as ˆq(η)∝dNch/dη. We use PYTHIA to calculate
the W and Z decay contribution [298]. More details can be found in Ref. [299].
[GeV/c]
t
µ
p
0 10 20 30 40 50 60
c]
-1
[pbGeV
t
/dp
µ
σd
-2
10
-1
10
1
10
2
10
3
10
4
10
5
10
6
10
7
10
8
10
9
10
from W, Z, c, bµ from W, Zµ from cµ from bµ
Pb-Pb, E loss, no cuts
=0q
solid: /fm
2
=25GeVq
short-dash: /fm
2
=100GeVq
long-dash:
[GeV/c]
t
µ
p
0 10 20 30 40 50 60
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
Figure 111: ptdistribution (left) and RAA (right) of single muons in central (0–10%) Pb–Pb
collisions at √sNN =5.5 TeV.
[GeV/c]
t
µ
p
0 10 20 30 40 50 60
(0-10%)/(40-70%)
CP
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
from W, Z, c, bµ from W, Zµ from cµ from bµ
Pb-Pb, E loss, no cuts
/fm
2
=25GeVq
short-dash: /fm
2
=100GeVq
long-dash:
[GeV/c]
t
µ
p
0 10 20 30 40 50 60
(0-10%)/(40-70%)
CP
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
/fm
2
= 25 GeVq
/fm
2
= 100 GeVq =0
c,b
/fm, m
2
= 25 GeVq =0
c,b
/fm, m
2
= 100 GeVq
Pb-Pb, E loss, no cuts
Figure 112: RCP (0–10%)/(40–70%) of single muons in Pb–Pb collisions at √sNN =5.5 TeV.
Fig. 111 shows the ptspectrum and RAA(pt) of the single muons from heavy quark and
W/Z bosons in the central Pb–Pb collisions at √sNN =5.5 TeV, with the transport coefficient
values ˆq=0,25,100 GeV2/fm. The crossing point of b and W decay muons shifts down by
5–7 GeV/c.RAA rapidly increases from 0.3 to 0.8 between 20 (b-dominated) and 40 GeV/c
(W-dominated), as does RCP (0–10%)/(40–70%), shown in Fig. 112. The effect of the heavy-
quark mass on the medium-induced suppression of RCP is shown in the left-hand panel of

Heavy Ion Collisions at the LHC - Last Call for Predictions 134
Fig. 112.
7.5. Quarkonium production in coherent pp/AA collisions and small-x physics
V. P. Gonçalves and M. V. T. Machado
We study the photoproduction of quarkonium in coherent proton-proton and nucleus-nucleus
interactions at the LHC. The integrated cross sections and rapidity distributions are estimated using
the Color Glass Condensate (CGC) formalism, which takes into account the parton saturation effects
at high energies. Nuclear shadowing effects are also taken into account.
In this contribution we study the photoproduction of vector mesons in the coherent
pp/AA interactions at the LHC energies. The main advantage of using colliding hadrons
and nuclear beams for studying photon induced interactions is the high equivalent photon
energies and luminosities that can be achieved at existing and future accelerators (for a review
see reference [300]). Consequently, studies of γpinteractions at LHC could provide valuable
information on the QCD dynamics at high energies. The basic idea in coherent hadron
collisions is that the total cross section for a given process can be factorized in terms of
the equivalent flux of photons of the hadron projectile and the photon-photon or photon-target
production cross section. In exclusive processes, a certain particle is produced while the
target remains in the ground state (or is only internally excited). The typical examples of
these processes are the exclusive vector meson production, described by the process γh→Vh
(V=ρ,J/Ψ,Υ). In the last years we have discussed this process in detail considering pp [301],
pA [302] and AA [301] collisions as an alternative to investigate the QCD dynamics at high
energies. Here, we revised these results and present for the first time our predictions for the Υ
production.
The cross section for the photoproduction of a vector meson Xin an ultra-peripheral
hadron-hadron collision is given by
σ(h1h2→h1h2X)=Z∞
ωmin
dωdNγ(ω)
dωσγh→Xh(W2
γh),
where ωis the photon energy and dNγ(ω)/dωis the equivalent flux of photons from a charged
hadron. The total cross section for vector meson photoproduction is calculated considering
the color dipole approach, which is directly related with the dipole-target forward amplitude
N. In the Color Glass Condensate (CGC) formalism (see e.g. [303]), Nencodes all the
information about the hadronic scattering, and thus about the non-linear and quantum effects
in the hadron wave function. In our analyzes we have used the phenomenological saturation
model proposed in references [33,304]. Nuclear effects are also properly taken into account.
Our predictions for the rapidity distributions are presented in figure 113 and for the total
cross section in table 6. The main uncertainties are the photon flux, the quark mass and the
size of nuclear effects for the photonuclear case. In addition, specific predictions for ρand
J/Ψphoproduction in pA collisions can be found in reference [302]. The rates are very high,
mostly for light mesons. Although the rates are lower than hadroproduction, the coherent

Heavy Ion Collisions at the LHC - Last Call for Predictions 135
−12 −9 −6 −3 0 3 6 9 12
y
10−1
100
101
102
103
dσ/dy [nb]
LHC: pp −−> pp + (VM = ρ,ω,φ,J/Ψ)
J/Ψ(3097)
φ(1019)
ω(782)
ρ(770)
−9 −6 −3 0 3 6 9
y
10−1
100
101
102
103
dσ/dy [mb]
LHC: PbPb −−> PbPb + (VM = ρ,ω,φ,J/Ψ)
J/Ψ(3097)
φ(1019)
ω(782)
ρ(770)
Figure 113: The rapidity distribution for nuclear vector meson photoproduction on coherent
pp (left panel) and AA (right panel) reactions at the LHC.
Table 6: The integrated cross section for nuclear vector mesons photoproduction in coherent
pp and AA collisions at the LHC.
Υ(9460) J/Ψ(3097) φ(1019) ω(782) ρ(770)
pp 0.8 nb 132 nb 980 nb 1.24 µb 9.75 µb
Ca-Ca 9.7 µb 436 µb 12 mb 14 mb 128 mb
Pb-Pb 96 µb 41.5 mb 998 mb 1131 mb 10069 mb
photoproduction signal would be clearly separated by applying a transverse momentum cut
pT<1 GeV and two rapidity gaps in the final state.
7.6. Heavy-Quark Kinetics in the QGP at LHC
H. van Hees, V. Greco and R. Rapp
We present predictions for the nuclear modification factor and elliptic flow of Dand Bmesons,
as well as of their decay electrons, in semicentral Pb-Pb collisions at the LHC. Heavy quarks are
propagated in a Quark-Gluon Plasma using a relativistic Langevin simulation with drag and diffusion
coefficients from elastic interactions with light anti-/quarks and gluons, including non-perturbative
resonance scattering. Hadronization at Tcis performed within a combined coalescence-fragmentation
scheme.
In Au-Au collisions at the Relativistic Heavy Ion Collider (RHIC) a surprisingly
large suppression and elliptic flow of “non-photonic” single electrons (e±, originating from
semileptonic decays of Dand Bmesons) has been found, indicating a strong coupling of
charm (c) and bottom (b) quarks in the Quark-Gluon Plasma (QGP).

Heavy Ion Collisions at the LHC - Last Call for Predictions 136
We employ a Fokker-Planck approach to evaluate drag and diffusion coefficients for
cand bquarks in the QGP based on elastic scattering with light quarks and antiquarks
via D- and B-meson resonances (supplemented by perturbative interactions in color non-
singlet channels) [305]. This picture is motivated by lattice QCD computations which
suggest a survival of mesonic states above the critical temperature, Tc. Heavy-quark (HQ)
kinetics in the QGP is simulated with a relativistic Langevin process [306]. Since the initial
temperatures at the LHC are expected to exceed the resonance dissociation temperatures,
we implement a “melting” of D- and B-mesons above Tdiss=2Tc=360 MeV by a factor
(1+exp[(T−Tdiss)/∆])−1(∆=50 MeV) in the transport coefficients.
The medium in a heavy-ion reaction is modeled by a spatially homogeneous elliptic
thermal fireball which expands isentropically. The temperature is inferred from an ideal
gas QGP equation of state with Nf=2.5 massless quark flavors, with the total entropy
fixed by the number of charged hadrons which we extrapolate to dNch/dy≃1400 for central
√sNN =5.5 TeV Pb-Pb collisions. The expansion parameters are adjusted to hydrodynamic
simulations, resulting in a total lifetime of τfb≃6 fm/c at the end of a hadron-gas QGP mixed
phase and an inclusive light-quark elliptic flow of hv2i=7.5%. The QGP formation time, τ0, is
estimated using τ0T0=const (T0: initial temperature), which for semicentral collisions (impact
parameter b≃7 fm) yields T0≃520 MeV.
Initial HQ pTspectra are computed using PYTHIA with parameters as used by the
ALICE Collaboration. cand bquarks are hadronized into Dand Bmesons at Tcby
coalescence with light quarks [62]; “left over” heavy quarks are hadronized with δ-function
fragmentation. For semileptonic electron decays we assume 3-body kinematics [306].
Fig. 114 summarizes our results for HQ diffusion in a QGP in terms of RAA(pT)
and v2(pT) at the quark, meson and e±level for b=7 fm Pb-Pb collisions at the LHC
(approximately representing minimum-bias conditions). Our most important findings are:
(a) resonance interactions substantially increase (decrease) v2(RAA) compared to perturbative
interactions; (b) bquarks are much less affected than cquarks, reducing the effects in the
e±spectra; (c) there is a strong correlation between a large v2and a small RAA at the quark
level, which, however, is partially reversed by coalescence contributions which increase both
v2and RAA at the meson (and e±) level. This feature turned out to be important in the
prediction of e±spectra at RHIC; (d) the predictions for LHC are quantitatively rather similar
to our RHIC results [306,307], due to a combination of harder initial HQ-pTspectra with a
moderate increase in interaction strength in the early phases where non-perturbative resonance
scattering is inoperative.
7.7. Ratio of charm to bottom RAA as a test of pQCD vs. AdS/CFT energy loss
W. A. Horowitz
The theoretical framework of a weakly-coupled QGP used in pQCD models that
quantitatively describe the high-pTπ0,ηsuppression at RHIC is challenged by several
experimental observables, not limited to high-pTonly, suggesting the possibility that a

Heavy Ion Collisions at the LHC - Last Call for Predictions 137
0 1 2 3 4 5 6 78
pT [GeV]
0
0.5
1
1.5
2
2.5
RAA
c, reso (Γ=0.4-0.75 GeV)
c, pQCD+rad
b, reso (Γ=0.4-0.75 GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
0 1 2 3 4 5
pT [GeV]
0
5
10
15
20
v2 [%]
c, reso (Γ=0.4-0.75 GeV)
c, pQCD
b reso (Γ=0.4−0.75 GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 1 2 3 4 5 6 7 8
RAA (D/B)
pT (GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
reso: Γ=0.4-0.75 GeV
D reso
D pQCD
B reso
B pQCD
-5
0
5
10
15
20
25
0 1 2 3 4 5 6 7 8
v2(D/B) (%)
pT (GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
reso: Γ=0.4-0.75 GeV
D reso
D pQCD
B reso
B pQCD
0
0.5
1
1.5
2
0 1 2 3 4 5
RAA(e±)
pT (GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
c+b reso
c+b pQCD
c reso
-5
0
5
10
15
20
25
0 1 2 3 4 5
v2(e±) (%)
pT (GeV)
Pb-Pb √s=5.5 TeV (b=7 fm)
reso: Γ=0.4-0.75 GeV
c+b reso
c+b pQCD
c reso
Figure 114: (Color online) Predictions of relativistic Langevin simulations for heavy quarks in
a sQGP for b=7 fm √sNN=5.5 TeV Pb-Pb collisions: RAA (left column) and v2(right column)
for heavy quarks (1st row), Dand Bmesons (2nd row) and decay-e±(3rd row).
strongly-coupled picture might be more accurate. One seeks a measurement that may clearly
falsify one or both approaches; heavy quark jet suppression is one possibility. Strongly-
coupled calculations, utilizing the AdS/CFT correspondence, have been applied to high-
pTjets in three ways [263, 308, 309]. We will focus on predictions from the AdS/CFT
heavy quark drag model and compare them to pQCD predictions from the full radiative and
elastic loss WHDG model and radiative alone WHDG model [244]. Comparisons between
AdS/CFT models and data are difficult. First, one must accept the double conjecture of
QCD↔SYM↔AdS/CFT. Second, to make contact with experiment, one must make further
assumptions to map quantities such as the coupling and temperature in QCD into the SUGRA
calculation. For example, the AdS/CFT prediction for the heavy quark diffusion coefficient
is D=4/√λ(/2πT) [308], where λ=g2
SYMNcis the ’t Hooft coupling. The “obvious” first
such mapping [310] simply equates constant couplings, gs=gS Y M, and temperatures, TS Y M =
TQCD. Using this prescription with the canonical Nc=3 and αs=.3 yields D≈1.2(/2πT).
It was claimed in [308] that D=3(/2πT) agrees better with data; this requires αs≈.05. An
“alternative” mapping [310] equates the quark-antiquark force found on the lattice to that
computed using AdS/CFT, giving λ≈5.5, and the QCD and SYM energy densities, yielding

Heavy Ion Collisions at the LHC - Last Call for Predictions 138
TSYM =TQCD/31/4.The medium density to be created at LHC is unknown; we will take the
PHOBOS extrapolation of dNg/dy =1750 and the KLN model of the CGC, dNg/dy=2900, as
two sample values.We will search for general trends associated with AdS/CFT drag (denoted
hereafter simply as AdS/CFT) or pQCD as these uncertainties mean little constrains the
possible normalizations of AdS/CFT RQ
AA predictions for LHC.
0 50 100 150
pT (GeV)
0.2
0.4
0.6
0.8
1
RAA(pT)
pQCD Rad+El, PHOBOS
pQCD Rad+El, KLN
AdS/CFT D = 3, PHOBOS
AdS/CFT D = 3, KLN
50 100 150
pT (GeV)
BottomCharm
O
O
(a) (b)
Figure 115: (a) Charm and bottom RAA(pT) predictions with representative input parameters
for LHC. The generic trend of pQCD curves increasing with pTwhile AdS/CFT curves
decrease is seen for representative input parameters; similar trends occurred for the other
input possibilities considered. (b) Ratio of charm to bottom RAA(pT) bunches the two models
for a wide range of input parameters; the LHC should easily distinguish between the two
trends.
AdS/CFT calculations of the drag on a heavy quark yield dpT/dt =−µQpT=
−(π√λT2
SYM/2mQ)pT[309], giving an average fractional energy loss of ¯ǫ=1−exp(−RdtµQ).
Asymptotic pQCD energy loss for heavy quarks in a static medium goes as ¯ǫ≈
κL2ˆqlog(pT/mQ)/pT, where κis a proportionality constant and Lis the pathlength traversed
by the heavy quark. Note that AdS/CFT fractional momentum loss is independent of
momentum while pQCD loss decreases with jet energy. The heavy quark production
spectrum may be approximated by a slowly varying power law of index nQ(pT)+1, then
RQ
AA ≈(1 −¯ǫ)nQ(pT). Since nQ(pT) is a slowly increasing function of momentum, we expect
RQ
AA,S Y M(pT) to decrease while RQ
AA,pQCD(pT) to increase as momentum increases. This
behavior is reflected in the full numerical calculations shown in Fig. 115 (a); details of the
model can be found in [311].
For high suppression pQCD predicts nearly flat RQ
AA, masking the difference between
AdS/CFT and pQCD. One can see in Fig. 115 (b) that the separation of AdS/CFT
and pQCD predictions is enhanced when the double ratio of charm to bottom nuclear
modification, Rcb(pT)=Rc
AA(pT)/Rb
AA(pT), is considered. Asymptotic pQCD energy loss
goes as log(mQ/pT)/pT, becoming insensitive to quark mass for pT≫mQ; hence Rcb
pQCD →
1. Expanding the RAA formula for small ǫyields Rcb
pQCD(pT)≈1−pcb/pT, where pcb =
κn(pT)L2log(mb/mc)ˆqand nc≈nb=n. Therefore the ratio approaches unity more slowly for
larger suppression. This behavior is reflected in the full numerical results for the moderately
quenched pQCD curves, but is violated by the highly oversuppressed ˆq=100 curve. The
AdS/CFT drag, however, is independent of pT. A back of the envelope approximation gives

Heavy Ion Collisions at the LHC - Last Call for Predictions 139
RQ
AA ≈RL
0dℓexp(−nQµqℓ)≈1/nQµqwhich yields Rcb(pT)≈nb(pT)mc/nc(pT)mb≈mc/mb≈
.27. This behavior is also reflected in the full numerical results shown in Fig. 115 (b), and so,
remarkably, the pQCD and AdS/CFT curves fall into easily distinguishable bunches, robust
to changes in input parameters. An estimate for the momentum after which corrections to
the above AdS/CFT drag formula are needed, γ > γc, found in the static string geometry is
γc=1/1+(2mQ/T√λ) [312]. Since temperature is not constant we show the smallest speed
limit, using T(τ0, ~x=~
0), and largest, from Tc, represented by “O” and “|,” respectively. A
deviation of Rcb away from unity at LHC in year 1 would pose a serious challenge to the usual
pQCD paradigm. An observation of a significant increase in Rcb with jet momenta would
imply that the current AdS/CFT picture is only applicable at low momenta, if at all. For a
definitive statement to be made a p+Pb control run will be crucial.
7.8. Thermal charm production at LHC
B.-W. Zhang, C. M. Ko and W. Liu
Charm production from an equilibrated quark-gluon plasma (QGP) produced in heavy
ion collisions at LHC is studied to the next-to-leading order in perturbative QCD [313].
Specifically, we consider the process q(g)+¯q(g)→c+¯cand its virtual correction as well
as the processes q(g)+¯q(g)→c+¯c+g, and g+q(¯q)→c+¯c+q(¯q). The amplitudes for
these processes are taken from Refs. [314–317] using massless quarks and gluons, the QCD
coupling constant αs(mc)≈0.37, and a charm quark mass mc=1.3 GeV. The charm quark
production rate in the QGP is then evaluated by integrating over the thermal quark and gluon
distributions in the QGP. Both thermal quarks and gluons are taken to have thermal masses
given by mq=mg=gT/√6, where Tis the temperature of the QGP and gis related to the
thermal QCD coupling constant αs(2πT)=g2/4π, which has values ranging from ∼0.23 for
T=700 MeV to ∼0.42 for T=170 MeV.
For the dynamics of formed QGP in central Pb+Pb collisions at √sNN =5.5 TeV at
LHC, we assume that it evolves boost invariantly in the longitudinal direction but with an
accelerated transverse expansion. Specifically, its volume expands in the proper time τ
according to V(τ)=πR2(τ)τc, where R(τ)=R0+a(τ−τ0)2/2 is the transverse radius with an
initial value R0=7 fm, the QGP formation time τ0=0.2 fm/c, and the transverse acceleration
a=0.1c2/fm. Starting with an initial temperature T0=700 MeV, which gives an initial
energy density of about 50% higher than that predicted by the AMPT model [52] or the Color
Glass Condensate [179], the time dependence of the temperature is obtained from entropy
conservation, leading to the critical temperature TC=170 MeV at proper time τC=6.4 fm/c.
The initial number of charm pairs is taken to be dNc¯c/dy =20 at midrapidity, which is of
similar magnitude as that estimated from initial hard nucleon-nucleon collisions based on the
next-to-leading order pQCD calculations.
In the left window of Fig. 116, we show the temperature dependence of the charm quark
pair production rates from the leading order (dashed line) and the next-to-leading order (solid
line) with their ratio shown in the inset. The contributions from the leading order and next-

Heavy Ion Collisions at the LHC - Last Call for Predictions 140
0.2 0.4 0.6 0.8
T (GeV)
1.0
2.0
3.0
4.0
5.0
Ratio
0.2 0.3 0.4 0.5 0.6 0.7 0.8
T (GeV)
10-6
10-5
10-4
10-3
10-2
0.1
1.0
R (c/fm4)
Next-to-leading order
Leading order
0 1 2 3 4 5 6
τ (fm/c)
10
20
30
Ncc
T0=700 MeV, τ0=0.2 fm/c
mc=1.3 GeV, massive partons
Next-to-leading order
Leading order
Figure 116: Time evolution of charm pair production rate (left window) and number (right
window) in central Pb+Pb collisions at √sNN =5.5 TeV for an initial QGP temperature of
700 MeV. Dashed and solid lines are results from the leading order and next-to-leading order
calculations, respectively. The inset in left window gives the ratio of charm production rate in
the next-to-leading order to that in the leading order.
leading order are of similar magnitude and both are appreciable at high temperatures. The
total number of charm pairs as a function of the proper time τin an expanding QGP produced
at LHC is shown in the right window of Fig. 116. As shown by the dashed line, including only
the leading-order contribution from two-body processes increases the number of charm pairs
by about 10% during the evolution of the QGP. Adding the next-leading-order contribution
through virtual corrections to two-body precesses as well as the 2 →3 processes further
increases the charm quark pair number by about 25% as shown by the solid line. The charm
quark pair number reaches its peak value at τ∼2 fm/cand then deceases with the proper
time as a result of larger charm annihilation than creation rates when the temperature of the
QGP drops. At the end of the QGP phase, it remains greater than both its initial value and the
chemically equilibrium value of about 5 at TC=170 MeV. The number of charm quark pairs
produced from the QGP would be reduced by a factor of about 3 if a larger charm quark mass
of 1.5 GeV or a lower initial temperature of T0=630 MeV is used. It is, however, not much
affected by using massless gluons due to increase in the gluon density. On the other hand,
increasing the initial temperature to 750 MeV would enhance the thermally produced charm
quark pairs by about a factor of 2.
7.9. Charm production in nuclear collisions
B. Z. Kopeliovich and I. Schmidt
Nuclear suppression of heavy flavor inclusive production in hard partonic collisions has a leading
twist component related to gluon shadowing, as well as a higher twist contribution related to the
nonzero separation of the produced ¯
QQ pair. Both terms are evaluated and suppression for charm
production in heavy ion collisions at LHC is predicted.

Heavy Ion Collisions at the LHC - Last Call for Predictions 141
7.9.1. Higher twist shadowing Heavy flavors are produced via gluon fusion, therefore they
serve as a good probe for the gluon distribution function in nuclei. The light-cone dipole
approach is an effective tool for the calculation of nuclear effects in these processes, since the
phenomenological dipole cross section includes by default all higher order and higher twist
terms.
The production of heavy flavors can be treated as freeing of a ¯
QQ fluctuation in the
incoming hadron, in which the interaction with such a small dipole (actually, with a three-body
¯
QQg dipole) results in nuclear shadowing, which is a higher twist, 1/m2
Q, effect. Although
very small, it steeply rises with energy and reaches sizable magnitude at the energy of LHC.
The effect of this higher twist shadowing on charm production in minimal bias and central
collisions of heavy ions at the energies of RHIC and LHC is shown in figure 117 as the
difference between solid and dashed curves.
Figure 117: Shadowing for DY reaction in pA (upper curves) and dA (lower curves) collisions
at the energies of RHIC ( √sNN =200 GeV) and LHC ( √sNN =5500 GeV), as function of xF
and dilepton mass M2. The left and right figures are calculated at M=4.5 GeV and xF=0.5
respectively.
7.9.2. Process dependent leading twist gluon shadowing The projectile fluctuations
containing, besides the ¯
QQ, also gluons, are responsible for gluon shadowing, which is a
leading twist effect. Indeed, the aligned jet configurations, i.e. the fluctuation in which the
¯
QQ pair carries the main fraction of the momentum, have a large and scale independent,
transverse size. Gluon shadowing is expected to be a rather weak effect [318] due to the
localization of the glue inside small spots in the proton [96]. This is confirmed by the latest
NLO analysis [15] of data on DIS on nuclei.
Unlike for the DIS case, where the produced ¯qq is predominantly in a color octet state,
in the case of hadroproduction the ¯
QQ may be either colorless or a color octet. Moreover,
in the latter case it may have different symmetries [319,320]. Nonperturbative effects, which

Heavy Ion Collisions at the LHC - Last Call for Predictions 142
cause a contraction of the gluon cloud, may be absent for a colorless ¯
QQ, leading to a much
stronger shadowing compared to DIS. This possibility was taken into account predicting the
rather strong nuclear effects depicted in figure 117. This part of the prediction should be taken
with precaution, since it has never been tested by data.
Figure 117 shows our results for RAA/NN as function of rapidity for minimal bias and
central collisions. These calculations do not include the suppression caused by energy
conservation at the ends of the rapidity interval [321].
The rather strong suppression of charm production that we found should be taken into
account as part of the strong suppression of high pTcharm production observed in central
nuclear collisions at RHIC. At high pTthis effect should fade away because of the rise of x2,
although at the LHC this may be a considerable correction.
7.10. Charm and Beauty Hadrons from Strangeness-rich QGP at LHC
I. Kuznetsova and J. Rafelski
The yields of heavy flavored hadrons emitted by strangeness rich QGP are evaluated within
chemical non-equilibrium statistical hadronization model, conserving strangeness, heavy flavor, and
entropy yields at hadronization.
A relatively large number of hadrons containing charm (dNc/dy ≃10) and bottom
(dNb/dy ≃1) quarks are expected to be produced at central rapidity in heavy ion (Pb–Pb)
collisions at the Large Hadrons Collider (LHC). This report summarized results of our more
extensive recent report [322], and amplifies its findings with reference to the ‘first day’ LHC-
ion results. Differing from other recent studies which assume that the hadron yields after
hadronization are in chemical equilibrium [323], we form the charm hadron yields in the
statistical hadronization approach based on an abundance of u,d,squark pairs fixed by the
bulk properties of a practically chemically equilibrated QGP phase.
In proceeding in this fashion we are respecting the constraints of the recombinant
dynamic model [324]. The absolute yields (absolute chemical equilibrium) depend in addition
to recombination on absolute heavy quark yield dNb,c/dy. We are fully implementing the
relative chemical equilibrium, that is the formation of heavy (charmed) hadrons according
the the relative phase space, thus ratios of yields presented here are a complete and reliable
prediction characterized by QGP entropy and strangeness content.
It is energetically more effective for strange quarks to emerge bound to heavy quarks.
Said differently, the reaction K+D→π+Dsis strongly exothermic, with ∆Q≃240MeV,
and similarly for the bottom quark. Considering that the phase space for hadronization is
characterized by a domain temperature T=160 ±20 MeV, in presence of strangeness the
yield tilts in favor Dsover D, and Bsover B.The variability in the light and strange quark
content at given hadronization temperature Tis accomplished introducing the phase space
occupancy γH
s>1, γH
q>1 of strange, and, respectively, light constituent quarks in the hadron
phase. In chemical equilibrium γH
s=γH
q=1.
A phase space evaluation of the relative yields leads to the results presented in figure 118,

Heavy Ion Collisions at the LHC - Last Call for Predictions 143
0.0 1 2 3 4 5
0.0
2
4
6
8
10
γs/γq
D/Ds
T=140 MeV
T=160 MeV
T=180 MeV
1 2 3 4
0
2
4
6
8
10
12
14
γs/γq
cqq/css, T=140 MeV
cqq/css, T=170 MeV
cqq/css, T=200 MeV
cqq/cqs, T=140 MeV
cqq/cqs, T=170 MeV
cqq/cqs, T=200 MeV
Figure 118: As a function of γH
s/γH
qon left: D/Dsratio and on right: cqq/css =(Λc+ Σc)/Ωc(upper
lines) and cqq/cqs =(Λc+ Σc)/Ξc(lower lines) ratios.
where we show ratio of open charm strange meson and baryons with the corresponding ‘less’
strange open charmed (strange) meson and baryons, as a function of γH
s/γH
q, which is the
controlling variable for three values T=200 MeV, T=180–160 MeV and T=140 MeV. The
corresponding chemical reference results are indicated by the crossing vertical and horizontal
lines. For B,Bsmesons the results are the same as for D,Dsmesons, see [322] for details.
The challenge is to understand what values of γH
s/γH
qa fast hadronizing QGP implies.
We obtain these by requiring that the hadronization of QGP proceeds conserving the entropy
dS/dy and strangeness ds/dy =d¯s/dy content of QGP. For LHC the expected ratio s/S=
0.038 [169] at T=140–180 MeV which implies in the hadron phase γs/γq=1.8–2 [168]. This
entails a considerable shift of open charm hadrons away from hadron chemical equilibrium
yield towards states containing strangeness in all cases considered in figure 118 (and similarly
for the bottom flavor). The hadronization process, as expected, favors formation of strange
charmed meson and baryons, once the actual QGP strangeness yield near/above-chemical
equilibrium is allowed for.
7.11. Charmonium Suppression in Strangeness-rich QGP
I. Kuznetsova and J. Rafelski
The yields of c¯cmesons formed in presence of entropy and strangeness rich QGP are evaluated
within chemical non-equilibrium statistical hadronization model, conserving strangeness and entropy
yields at hadronization. We find that for a given dNc/dy charm yield, the abundant presence of light an
strange quarks favors formation of D,Dsmesons and to suppression of charmonium.
There is considerable energetic advantage for a charm quark to bind with a strange quark
– most, if not all, charmonium–strange meson/baryon reactions of the type c¯c+sX →cX+¯cs,

Heavy Ion Collisions at the LHC - Last Call for Predictions 144
140 160 180 200 220 240 260
10−3
10−2
T [MeV]
dNhid/dy/Nc
2
γq=γs=1
s/S=0.03, SH=SQ
140 160 180 200 220 240 260
10−3
10−2
T [MeV]
γq=γs=1
s/S=0.04, SH=SQ
0 1 2 3 4
0
1
2
J/Ψ/J/Ψeq=γ2
c/γ2
c eq
γs/γq
T=170 MeV, γs=γq=1
T=170 MeV, SH=SQ, γq=1.12
T=170 MeV, γq=γcr
q=1.51
T=140 MeV, SH=SQ, γq=1.6
Figure 119: Left two panels: c¯c/N2
crelative yields as a function of hadronization temperature
T, right panel ratio J/Ψ/J/Ψeq as a function of γH
s/γH
q, see tex.
where X≡¯q=¯u,¯
dor X≡qq,qs,ss are strongly exothermic. In statistical hadronization
this phase space effect favors formation of Dsover c¯c. Seen from the kinetic model
perspective [324], this observation shows a strong channel of charmonium destruction. Thus
presence of strangeness facilitates a novel charmonium suppression mechanism [322, 325].
To implement this effect hadronization of QGP must conserve strangeness and entropy and
thus cannot be ad-hoc associated with chemical equilibrium.
In the non-equilibrium statistical hadronization model we balance total yield of charmed
particles within a given volume dV/dy to the level available in the QGP phase dNc/dy ∝
dV/dy(γH
cγH
i+...), where a few percent of the yield is in multi-charm baryons and
charmonium involving higher powers of γH
c. This constraint determines a value of γH
c, which
for the case of LHC can be considerably above unity. Therefore, the hadronization yields we
compute for hidden charm mesons: dNc¯c/dy ∝dV/dyγH2
c∝(dNc/dy)2/(γH2
idV/dy). depends
on the inverse of the model dependent reaction volume, and scales with the square of the total
charm yields [324]. We also show above that for the case that γH
i>1 a hereto unexpected
suppression of ’onium yield is expected.
In figure 119 the yield of all hidden charm c¯c(sum over all c¯cmesons) is shown,
normalized by the square of dNc/dy=10 (middle panel for LHC environment) and dNc/dy=3
(left panel, RHIC environment), as a function of hadronization temperature T. We show
result for s/S=0.03 with dV/dy =600 fm3,T=200 MeV (solid line, left panel) and for
s/S=0.04 with dV/dy =800 fm3,T=200 MeV (solid line, middle panel). Results shown
for chemical equilibrium case (dashed lines) are for the values γs=γq=1. For the chemical
non-equilibrium hadronization (solid lines γH
i>1,i=q,s), the QGP and hadron phase space
is evaluated conserving entropy SQ=SHand strangeness sQ=sHbetween phases.
We see, comparing the left and middle panel that the yield of c¯cmesons decreases with

Heavy Ion Collisions at the LHC - Last Call for Predictions 145
increasing specific strangeness content (note logarithmic scale). The chemical suppression
effect is further quantified in third, right panel in figure 119, where we show the ratio
J/Ψ/J/Ψeq =γ2
c/γ2
ceq as a function of γH
s/γH
qat fixed value of γH
qand, as required, entropy
conservation for T=140,170 MeV. For T=140 MeV we show result with γq=1.6 (solid
dotted line) which corresponds entropy conservation between QGP and hadronic phase for this
hadronization temperature. For T=170 MeV we show results with γq≈γcr
q=1.51 ≡emπ/2T
(dash-dot line), γq=1.12 (solid line) and γq=1 (dashed line). For γq=1.12 entropy is
conserved in hadronization at T=170 MeV .
The formation of the Bc(¯
bc) proposed as another QGP signature [326] has not been
evaluated in the present work, since this particle yield suffers from additional (canonical)
suppression. Kinetic formation models suggest significant enhancement of this double exotic
meson, as compared to a cascade of NN reactions.
7.12. J/ψ pTspectra from in-medium recombination
R. L. Thews and M. L. Mangano
We consider production of J/ψ by recombination of c, ¯cquarks produced in separate N-N
interactions during Pb-Pb collisions. Inputs for the calculation include the NLO pQCD spectra of
charm quarks, plus a range of nuclear parameters taken from extrapolation of results at RHIC energy.
The possibility that J/ψ could be formed in AA collisions by recombination in a region
of color deconfinement was first developed in Ref. [324]. It was motivated by the realization
that the total formation probability would be proportional to the square of the total number
of c¯cpairs, which at RHIC and especially LHC provide a large enhancement factor. One can
calculate the pTand yspectra of J/ψ formed either through recombination or direct initial
production, using the corresponding quark spectra from a pQCD NLO calculation [297] in
individual nucleon-nucleon interactions. The method involves generating a sample of these
initially-produced c¯cpairs, smearing the transverse momentum with a gaussian distribution
of width hkT2ito simulate nuclear broadening and confinement effects, and weighting each
pair with a formation cross section. This procedure naturally divides the total pair sample
into two categories: the so-called “diagonal" sample, which pairs the c and ¯cfrom the same
nucleon-nucleon interaction and the “off-diagonal" sample, where c and ¯ccome from different
nucleon-nucleon interactions. The spectra of the resulting J/ψ will retain some memory of
the charm quark spectra and provide signatures of the two different origins. For example, one
expects the pTspectrum of non-diagonal pairs to be softer, since it is less likely for high-pT
cand ¯cquarks from independent scatterings to be close enough in phase-space to coalesce
into a J/ψ. Results for RHIC were presented in Ref. [327], where the primary signal was
found to be a narrowing of the non-diagonal yand pTspectra, relative to the diagonal ones.
We show in Fig. 120 the calculated J/ψ width hpT2ias a function of hkT2i, for central and
forward production in ALICE. hpT2igrows with hkT2ifor both the direct initial production
and the in-medium formation, but the latter widths are always smaller than the former. Widths
at small yare also greater than at large y, reflecting the underlying pQCD distributions. To fix

Heavy Ion Collisions at the LHC - Last Call for Predictions 146
0 1 2 3 4 5 6 78 9 10
<kT2> (GeV2)
0
10
20
30
40
50
60
<pT2> (GeV2)
Initial Production, central y
Initial Production, backward y
In-Medium Formation, central y
In-Medium Formation, backward y
10 100 1000
Ncoll
0
10
20
30
40
50
60
70
<pT2> (GeV2)
Initial production, λ2 = RHIC value
Initial production, λ2 = 10 x RHIC value
In-medium formation, λ2 = RHIC value
In-medium formation, λ2 = 10 x RHIC value
10 100 1000
Ncoll
0
10
20
30
40
50
60
70
<pT2> (GeV2)
Initial production, λ2 = RHIC value
Initial production, λ2 = 10 x RHIC value
In-medium formation, λ2 = RHIC value
In-medium formation, λ2 = 10 x RHIC value
Figure 120: Upper: Variation of J/ψ hpT2iwith the nuclear smearing parameter. Lower:
dependence on the intrinsic hkT2ipp, with hkT2ipp =0.0 (left) and 5.0 (right).
the nuclear smearing parameter values, we use a relation between measurable hpT2iin pp and
pA interactions,
hpT2ipA −hpT2ipp =λ2[¯nA−1],(60)
where ¯nAis the impact-averaged number of inelastic interactions of the proton projectile in
nucleus A, and λ2is proportional to the square of the transverse momentum transfer per
initial state collision. We use a Glauber model to calculate the centrality dependence of
the ¯nA, and parameterize the centrality by the total number of collisions, Ncoll. Thus with
measurements of hpT2ipp and hpT2ipA one can extract λ2and calculate the corresponding
nuclear broadening for AA interactions. The lower plots of Fig. 120 show the results for
Pb-Pb at 5.5 TeV, with hkT2ipp=0 and 5 GeV2. For both cases the J/ψ widths will provide a
clear discrimination between direct initial production and in-medium formation. In general,
one would expect some combination of initial production and in-medium formation, so the
prediction is bounded from above and below. There is almost no change in the pT(c) spectra
between 5.5 and 14 TeV. Thus we can use the 14 TeV pp data to determine hkT2ipp at 5.5 TeV.
One can then expect that the absence of energy dependence will also hold for p-Pb results,
allowing us to also determine λ2at 5.5 TeV from a measurement at any LHC energy, thus
fixing the prediction for curves such as those in Fig. 120.

Heavy Ion Collisions at the LHC - Last Call for Predictions 147
0
0.1
0.2
0.3
0.4
0.5
0.6
1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8
Ebinding [GeV]
T/Tc
Υ
J/ψ
χb
Υ’
0
0.2
0.4
0.6
0.8
1
0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
Γ(T) [GeV]
T/Tc
Υ
Υ’
χb
J/ψ
Figure 121: Upper limit of the binding energy (left) and the width (right) of quarkonium
states. For better visibility, in the limit of small binding, the open squares show the width of
the 1S bottomonium state multiplied by six.
7.13. Predictions for quarkonia dissociation
Á. Mócsy and P. Petreczky
We predict the upper bound on the dissociation temperatures of different quarkonium states.
In a recent paper [328] we analyzed in detail the quarkonium spectral functions. This
analysis has shown that spectral functions calculated using potential model for the non-
relativistic Green’s function combined with perturbative QCD can describe the available
lattice data on quarkonium correlators both at zero and finite temperature in QCD with no light
quarks [328]. Charmonia, however, were found to be dissolved at temperatures significantly
lower than quoted in lattice QCD studies, and in contradiction with other claims made in
recent years from different potential model studies. In [329] we extended the analysis to
real QCD with one strange quark and two light quarks using new lattice QCD data on quark
anti-quark free energy obtained with small quark masses [330].
Here we briefly outline the main results of the analysis of [329], in particular the estimate
for the upper limit on the dissociation temperatures. There is an uncertainty in choosing
the quark-antiquark potential at finite temperature. In [329] we considered two choices of
the potential, both consistent with the lattice data [330]. The more extreme choice, still
compatible with lattice data, leads to the largest possible binding energy. In this most binding
potential some of the quarkonium states survive above deconfinement, but their strongly
temperature-dependent binding energy is significantly reduced. This is shown in figure 121.
Due to the reduced binding energy thermal activation can lead to the dissociation of quarkonia,
even when the corresponding peak is present in the spectral function. Knowing the binding
energy we estimate the thermal width using the analysis of [331]. The expression of the rate
of thermal excitation has particularly simple form in the two limiting cases:
Γ(T)=(LT)2
3πMe−Ebin/T,Ebin ≫TΓ(T)=4
LrT
2πM,Ebin ≪T.
Here Mis the quarkonium mass, Lis the size of the spatial region of the potential, given by the
distance from the average quarkonium radius to the top of the potential, i.e. L=rmed −hr2i1/2,

Heavy Ion Collisions at the LHC - Last Call for Predictions 148
rmed being the effective range of the potential [329]. Using the above formulas we estimate the
thermal width of charmonium and bottomonium states. Since in the deconfined phase Ebin <T
the 1Scharmonium and 2Sand 1Pbottomonium states are in the regime of weak binding,
and their width is large, as shown in figure 121. The 1Sbottomonium is strongly bound for
T<1.6Tcand its thermal width is smaller than 40 MeV. For T>1.6Tc, however, even the 1S
bottomonium states is in the weak binding regime resulting in the large increase of the width,
see figure 121. When the thermal width is significantly larger than the binding energy no peak
structure will be present in the spectral functions, even though the simple potential model
calculation predicts a peak. Therefore, we define a conservative dissociation temperature by
the condition Γ>2Ebin. The obtained dissociation temperatures are summarized in table 7.13.
Table 7: Upper bound on quarkonium dissociation temperatures.
state χcψ′J/ψ Υ′χbΥ
Tdis ≤Tc≤Tc1.2Tc1.2Tc1.3Tc2Tc
From the table it is clear that all quarkonium states, except the 1Sbottomonium, will melt
at temperatures considerably smaller than previous estimates, and will for certain be dissolved
in the matter produced in heavy ion collision at LHC. Furthermore, it is likely that at energy
densities reached at the LHC a large fraction of the 1Sbottomonium states will also dissolve.
It has to be seen to what extent these findings will result in large RAA suppression at LHC. For
this more information about initial state effects is needed. Moreover, the spectral functions are
strongly enhanced over the free case even when quarkonium states are dissolved [328,329]
indicating significant correlations between the heavy quark and antiquark. Therefore, one
should take into account also the possibility of quarkonium regeneration from correlated initial
quark-antiquark pairs.
7.14. Heavy flavor production and suppression at the LHC
I. Vitev
Predictions for the baseline D- and B-mesons production cross sections at s1/2=5.5 TeV at
the LHC in p+p collisions are given for pT>Mc,b, respectively. New measurements that allow to
identify the underlying hard partonic processes in heavy flavor production are discussed. Based on
the short D- and B-mesons formation times, medium-induced dissociation is proposed as a mechanism
of heavy flavor suppression in the QGP at intermediate pT. In contrast to previous results on heavy
quark modification, this approach predicts suppression of B-mesons comparable to that of D-mesons at
transverse momenta as low as pT∼10 GeV. Suppression of non-photonic electrons form the primary
semi-leptonic decays of charm and beauty hadrons is calculated in the pTregion where collisional
dissociation is expected to be relevant.
Predictions for the baseline D0,D+,B0,B+cross sections in p+p collisions at the LHC
at s1/2=5.5 TeV are given in the left panel of Fig. 122 [332]. At lowest order we also include

Heavy Ion Collisions at the LHC - Last Call for Predictions 149
0 10 20 30 40 50
pT [GeV]
10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
10-2
10-1
dσ/dyd2pT [mb.GeV-2]
D0, r = 0.2
D+, r = 0.2
B0, r = 0.07
B+, r = 0.07
510 15 20 25
pT [GeV]
B+ at s1/2 = 1.8 TeV
B+ at s1/2 = 1.96 TeV
510 15 20
pT [GeV]
100
101
102
103
104
105
dσ/dpT [nb.GeV-1]
D0 at s1/2 = 1.96 TeV
D+ at s1/2 = 1.96 TeV
Tevatron (I+II) |y| < 1
mc = 1.3 GeV mb = 4.5 GeV
s1/2 = 1.96 TeV |y| < 1
PQCD, K = 1 - 1.5 PQCD, K = 1 - 1.5
y = 0
s1/2 = 5500 GeV K = 1.5
LHC
0 2 4 68 10 12 14
pT2 [GeV]
0.5
1
1.5
2
2.5
dσD meson/dσtotal, dσlight h/dσtotal
Anti-D mesons
Light hadrons
0 0.2 0.4 0.6 0.8 1
z
0.001
1
1000
D(z)
h++h- (q,g)
D0+D+ (c)
pT1 = 10 GeV
dσtotal
y1 = y2 = 0
r = 0.1, 0.2, 0.4
s1/2 = 5500 GeV
D trigger
Figure 122: Left panel: D- and B-meson production cross sections st s1/2=5.5 TeV [332].
Comparison to available data at Tevatron is also shown. Away-side hadron composition of
pT=10 GeV D-meson triggered jet [332]. Right panel: Hadron composition of the away-
side D-meson triggered jet at LHC energies as a function of the hardness of the heavy quark
fragmentation function.
Q+g→Q+g,Q+q(¯q)→Q+q(¯q) and processes that give a dominant contribution to the
single inclusive D- and B-mesons [332]. The right panel of Fig. 122 illustrates a method to
determine the underlying heavy flavor production mechanism through the away-side hadron
composition of D−and B−meson triggered jets [332].
The GLV approach is to multiple parton scattering [333] can be easily generalized to
various compelling high energy nuclear physics problems, such as meson dissociation in
dense nuclear matter [334]. RAA(pT) results for charm and beauty from this novel suppression
mechanism at RHIC and LHC are shown in the left panel of Fig. 123. Attenuation rate similar
to the light hadron quenching from radiative energy loss [333] is achieved. The right panel
of Fig. 123 shows the suppression of the single non-photonic 0.5(e++e−) in central Au+Au
and Pb+Pb collisions at RHIC and LHC respectively [334]. The separate measurement of
intermediate pTD−and B−meson quenching will allow to experimentally determine the
correct physics mechanism of heavy flavor suppression [335].
7.15. Quarkonium shadowing in pPb and Pb+Pb collisions
R. Vogt
The d+Au data from RHIC, including the pA results from the fixed-target CERN SPS
pA data, suggest increased importance of initial-state shadowing and decreasing nuclear
absorption with increasing energy [336]. This is not surprising since smaller xis probed

Heavy Ion Collisions at the LHC - Last Call for Predictions 150
0
0.25
0.5
0.75
1
RAA(pT)
B+D mesons+baryons
B mesons+baryons
D mesons+baryons
0 10 20
pT [GeV]
0
0.25
0.5
0.75
RAA(pT)
0 10 20 30
pT [GeV]
Central Au+Au at RHIC (dNg/dy = 1175) Central Cu+Cu at RHIC (dNg/dy = 350)
Central Pb+Pb at LHC (dNg/dy = 2000) Central Pb+Pb at LHC (dNg/dy = 3500)
s1/2NN = 5500 GeV
s1/2NN = 200 GeVs1/2NN = 200 GeV
s1/2NN = 5500 GeV
ξ = 2 - 3
B
D
0 2 4 68 10 12
pT [GeV]
0
0.2
0.4
0.6
0.8
1
RAA(pT)
0.5(e++e-), QGP dissociation ξ = 2-3
PHENIX 0.5(e++e-), 0-10% Au+Au
STAR 0.5(e++e-), 0-5% Au+Au
STAR 0.5(e++e-), 0-12% Au+Au
0 10 20 30
pT [GeV]
0
0.2
0.4
0.6
0.8
1
RAA(pT)
0/5(e++e-), QGP dissociation ξ = 2-3
No nuclear effect
Central Au+Au
No nuclear effect
LHC central Pb+Pb, dNg/dy = 3500
Figure 123: Left panel: Suppression of D- and B-meson production via collisional
dissociation in the QGP. Results on RAA(pT) in central Pb+Pb collisions at the LHC are
compared to central Au+Au and Cu+Cu collisions at RHIC [334, 335]. Right panel:
Suppression of inclusive non-photonic electrons from D- and B-meson spectra softened by
collisional dissociation in central Au+Au collisions at RHIC compared to data and Pb+Pb
collisions at the LHC.
at higher energy while absorption due to multiple scattering is predicted to decrease with
energy [337]. The CERN SPS data suggest a J/ψ absorption cross section of about 4 mb
without shadowing, and a larger absorption cross section if it is included since the SPS xrange
is in the antishadowing region. The d+Au RHIC data support smaller absorption, σJ/ψ
abs ∼0−2
mb. Thus our predictions for J/ψ and Υproduction in pPb and Pb+Pb interactions at the LHC
are shown for initial-state shadowing alone with no absorption or dense matter effects. We
note that including absorption would only move the calculated ratios down in proportion to the
absorption survival probability since, at LHC energies, any rapidity dependence of absorption
is at very large |y|[338], outside the detector acceptance.
We present RpPb(y)=pPb/pp and RPbPb(y)=PbPb/pp for J/ψ and Υ. Since the pp,
pPb and Pb+Pb data are likely to be taken at different energies (14 TeV, 8.8 TeV and 5.5
TeV respectively), to make the calculations as realistic as possible we show several different
scenarios for RpPb(y) and RPbPb(y). The lead nucleus is assumed to come from the right in
pPb. All the pA calculations employ the EKS98 shadowing parameterization [339, 340]. The
difference in the J/ψ and Υresults is primarily due to the larger Υmass which increases the
xvalues by about a factor of three. In addition, the higher Q2reduces the overall shadowing
effect.
The top of Fig. 124 shows RpPb(y) for pPb/pp with both systems at √SNN =8.8 and 5.5
TeV (dashed and dot-dashed curves respectively), ignoring the ∆y=0.46 rapidity shift at 8.8
TeV. For the J/ψ, these ratios are relatively flat at forward rapidity where the xin the lead

Heavy Ion Collisions at the LHC - Last Call for Predictions 151
is small. The larger xand greater Q2for the Υbrings the onset of antishadowing closer to
midrapidity, within the range of the ALICE dimuon spectrometer. At far backward rapidity, a
rise due to the antishadowing region is seen. The lower energy moves the antishadowing peak
to the right for both quarkonia states. We show RpPb(y) with pPb at 8.8 TeV and pp at 14 TeV
with ∆y=0 in the dotted curves. The effect on the J/ψ is an apparent lowering of the dashed
curve. Since the Υrapidity distribution is narrower at 8.8 TeV than at 14 TeV in the rapidity
range shown here, the Υcurve turns over at large |y|. (This effect occurs at |y|>6 for the J/ψ.)
The solid curves show RpPb(y) for 8.8 TeV pPb and 14 TeV pp with the rapidity shift. Both
the J/ψ and Υratios are essentially constant for y>−2.5. Thus relying on ratios of pA to pp
collisions at different energies to study shadowing (or other small xeffects) may be difficult
because the shadowing function is hard to unfold when accounting for the pA ∆yas well as
the difference in x. If d+Pb collisions were used, ∆ywould be significantly reduced [341].
The lower part of Fig. 124 shows RPbPb(y) for the J/ψ and Υat 5.5 TeV for both
systems. No additional dense matter effects such as QQcoalescence or plasma screening
are included. The EKS98 (dashed) and nDSg [15] (dot-dashed) shadowing parameterizations
are compared. The results are very similar over the entire rapidity range. (Other shadowing
parameterizations,which do not agree with the RHIC d+Au data, give different RPbPb(y).)
There are antishadowing peaks at far forward and backward rapidity. As at RHIC, including
shadowing on both nuclei lowers the overall ratio relative to RpPb(y) as well as making
RPbPb(y>2) similar to or larger than RPbPb(y=0) because, without any other effects,
RPbPb(y)∼RpPb(y)RpPb(−y) when all systems are compared at the same √SNN. The solid
curves show the ratios for Pb+Pb at 5.5 TeV relative to pp at 14 TeV with the EKS98
parameterization. The trends are similar but the magnitude is lower.
Since these calculations reflect what should be seen if nothing else occurs, RPbPb(y) is
expected to differ significantly due to dense matter effects. If the initial J/ψ production is
strongly suppressed by plasma screening, then the only observed J/ψ’s would be from cc
coalesence [327] or Bmeson decays. It should be possible to experimentally distinguish
secondary production from the primordial distributions by displaced vertex cuts. Secondary
J/ψ production should havea narrower rapidity distribution and a loweraverage pT. Both are
indicated in central Au+Au collisions at √SNN =200 GeV at RHIC [342]. If J/ψ production
in central collisions is dominated by secondary J/ψ’s, peripheral collisions should still reflect
initial-state effects. Predictions of the centrality dependence of shadowing on J/ψ production
at RHIC agree with the most peripheral Au+Au data.
Finally, the J/ψ and Υrapidity distributions are likely to be inclusive, including feed
down from higher quarkonium states. Initial-state effects should be the same for all members
of a quarkonium family so that these ratios would be the same for direct and inclusive
production.
7.16. Quarkonium suppression as a function of pT
R. Vogt

Heavy Ion Collisions at the LHC - Last Call for Predictions 152
Figure 124: The J/ψ (left) and Υ(right) pPb/pp (top) and PbPb/pp (bottom) ratios as a
function of rapidity. The pPb/pp ratios are given for 8.8 (dashed) and 5.5 (dot-dashed) TeV
collisions in both cases and 8.8 TeV pPb to 14 TeV pp without (dotted) and with (solid) the
beam rapidity shift taken into account. The Pb beam comes from the right. The PbPb/pp ratios
are shown for 5.5 TeV in both cases with EKS98 (dashed) and nDSg (dot-dashed) shadowing
and also for 5.5 TeV Pb+Pb and 14 TeV pp (solid).
We present a revised look at the predictions of Ref. [343], taking into account
newer calculations of the screening mass with temperature and the quarkonia dissociation
temperature based on both potential models and calculations of quarkonium spectral
functions. The estimates of Digal et al. [344] predict lower quarkonium dissociation
temperatures, 1.1Tcfor the J/ψ and 2.3Tcfor the Υ, with µ=1.15T. A later review by
Satz [345], predicts higher values, more in line with the recent calculations of quarkonium
spectral functions, 2.1Tcfor the J/ψ and 4.1Tcfor the Υ, as well as µ∼1.45Tfor T>1.1Tc.
We assume 700 <T0<850 MeV and τ0=0.2 fm [346]. The pTdependence of the screening
is calculated as first discussed in Ref. [347]. Since it may be unlikely for feed down
contributions to be separated from the inclusive ψand Υyields in AA collisions, we present
the indirect ψ′/ψ and Υ′/Υratios, with feed down included, in Fig. 125. While the individual
suppression factors are smooth as a function of pT, as shown in Fig. 126 for all four sets of
initial conditions and dissociation temperatures, due to their different predicted dissociation
temperatures and formation times, they contribute differently to the ratios in Fig. 125.
We have assumed that the ψ′/ψ and Υ′/Υratios are independent of pT, as predicted in
the color evaporation model [348]. However, if this is not the case, any slope of the pTratios
in pp collisions can be calculated and/or evalulated experimentally and deconvoluted from
the data. Quarkonium regeneration by coalescence [327] has not been included here. While
it is unknown how coalescence production populates the quarkonium levels, since the pTof
quarkonium states produced by coalescence is lower than those produced in the initial NN
collisions, higher pTquarkonia should have a smaller coalescence contribution. The lower

Heavy Ion Collisions at the LHC - Last Call for Predictions 153
Figure 125: The indirect ψ′/ψ (left) and Υ′/Υ(right) ratios as a function of pTin Pb+Pb
collisions at 5.5 TeV for T0=700 MeV (solid and dashed) and 850 MeV (dot-dashed and
dotted). The ψ(Υ) results are shown for assumed dissociation temperatures of 1.1Tc(2.3Tc)
(solid and dot-dashed) and 2.1Tc(4.1Tc) (dashed and dotted) respectively.
BBrates should reduce the coalescence probability of Υproduction. By taking the ψ′/ψ and
Υ′/Υratios, we reduce systematics and initial-state effects.
In the case where TD=1.1Tcfor the J/ψ, its shorter formation time leads to suppression
over a larger pTrange than that for the χcand ψ′, leading to a larger ψ′/ψ ratio than the pp
value over all pT. On the other hand, for the higher dissociation temperature, the pTrange of
J/ψ suppression is shorter than for the other charmonium states, giving a smaller ratio than in
pp. The low pTbehavior of the dashed and dotted curves in the left-hand side of Fig. 125 is
due to the disappearance of χcsuppression since the χcis suppressed over a shorter pTrange
than the ψ′.
Since there are more states below the BBthreshold for the Υfamily, the suppression is
more complicated, in part because there are also feed down contributions to the Υ′, leading
to more structure in the Υ′/Υratios on the right-hand side of Fig. 125. For µ=1.15T, the
Υitself is suppressed, albeit over a short pTrange. The dips in the solid and dashed curves
occur at the pTwhere direct Υsuppression ceases. In the case where TD=4.1Tcfor the Υ,
the initial temperature is not large enough to suppress direct Υproduction so that Υ′/Υ<1
for all pT. The χbcontributions are responsible for the slopes of the ratios at pT>12 GeV.
8. Leptonic probes and photons
8.1. Thermal photons to dileptons ratio at LHC
J. K. Nayak, J. Alam, S. Sarkar and B. Sinha
Photons and dileptons are considered to be efficient probes of quark gluon plasma

Heavy Ion Collisions at the LHC - Last Call for Predictions 154
Figure 126: The survival probabilities as a function of pTfor the charmonium (left-hand side)
and bottomonium (right-hand side) states for initial conditions at the LHC. The charmonium
survival probabilities are J/ψ (solid), χc(dot-dashed) and ψ′(dashed) respectively. The
bottomonium survival probabilities are given for Υ(solid), χ1b(dot-dashed), Υ′(dashed),
χ2b(dot-dot-dash-dashed) and Υ′′ (dotted) respectively. The top plots are for T0=700 MeV
while the bottom are for T0=850 MeV. The left-hand sides of the plots for each state are for
the lower dissociation temperatures, 1.1Tcfor the J/ψ and 2.3Tcfor the Υwhile the right-
hand sides show the results for the higher dissociation temperatures, 2.1Tcfor the J/ψ and
4.1Tcfor the Υ.
(QGP) expected to be created in heavy ion collisions at ultra-relativistic energies.
However, the theoretical calculations of the transverse momentum (pT) spectra of photons
(d2Nγ/d2pTdyy=0) and dileptons (d2Nγ∗/d2pTdyy=0) depend on several parameters which are
model dependent (see [349,350] and references therein). In the present work it is shown that
the model dependences involved in individual photon and dilepton spectra are canceled out in
the ratio, Rem defined as: Rem =(d2Nγ/d2pTdy)y=0/(d2Nγ∗/d2pTdy)y=0.
The invariant yield of thermal photons can be written as d2Nγ/d2pTdy =
Pi=Q,M,HRid2Rγ/d2pTdyid4x, where Q,Mand Hrepresent QGP, mixed (coexisting phase
of QGP and hadrons) and hadronic phases respectively. (d2R/d2pTdy)iis the static rate of
photon production from the phase i, which is convoluted over the expansion dynamics through
the integration over d4x. The thermal photon rate from QGP up to O(ααs) have been consid-
ered. For photons from hadronic matter an exhaustive set of reactions (including those involv-
ing strange mesons) and radiative decays of higher resonance states have been considered in
which form factor effects have been included.
Similar to photons, the pTdistribution of thermal dileptons is given by, d2Nγ∗/d2pTdy =
Pi=Q,M,HRid2Rγ∗/d2pTdydM2idM2d4x.The limits for the integration over Mare fixed from
experimental measurements. Here we consider 2mπ<M<1.05 GeV. Thermal dilepton
rate from QGP up to O(α2αs) has been considered. For the hadronic phase we include the

Heavy Ion Collisions at the LHC - Last Call for Predictions 155
024
101
102
103
104
HM
Total (QM+HM)
101
102
103
104
Rem
HM
Total (QM+HM)
0 2 4
pT (GeV)
HM
Total (QM+HM)
HM
Total (QM+HM)
SPS RHIC
LHC (dN/dy=4000) LHC (dN/dy=4730)
200 400 600 800 1000
Ti (MeV)
100
150
200
250
300
Rem (pT = 2 GeV)
Figure 127: Left panel: Variation of Rem with pT, right panel: variation of Rem(pT=2GeV)
with Ti.
Table 8: The values of various parameters - thermalization time (τi), initial temperature (Ti),
freeze-out temperature (Tf) and hadronic multiplicity dN/dy - used in the present calculations.
Accelerator dN
dy τi(fm)Ti(GeV)Tf(MeV)
SPS 700 1 0.2 120
RHIC 1100 0.2 0.4 120
LHC 4000 0.08 0.85 120
LHC 4730 0.08 0.905 120
dileptons from the decays of light vector mesons [349]. The space time evolution of the
system has been studied using (2+1) dimensional relativistic hydrodynamics with longitudinal
boost invariance and cylindrical symmetry. The calculations have been performed for the
initial conditions mentioned in table 8 (see also [349]). The values of parameters shown
in table 8 reproduce the various experimental data from SPS and RHIC. For LHC we have
chosen two values of Ticorresponding to two values of dN/dy. We use the Bag model EOS
for the QGP phase. For EOS of the hadronic matter all resonances with mass ≤2.5 GeV have
been considered
The variation of Rem with pTfor different initial conditions are depicted in Fig. 127 (left
panel). At SPS, the contributions from hadronic matter (HM) coincides with the total and
hence it becomes difficult to make any conclusion about the formation of QGP. However,
for RHIC and LHC the contributions from HM are less than the total indicating large
contributions from quark matter. The quantity, Rem, reaches a plateau beyond pT=1 GeV
for all the three cases i.e. for SPS, RHIC and LHC. However, it is very important to note that
the values of Rem at the plateau region are different, e.g. RLHC
em >RRHIC
em >RS PS
em . Now for all
the three cases, SPS, RHIC and LHC, except Tiall other quantities e.g. Tc,v0,Tfand EOS
are same, indicating that the difference in the value of Rem in the plateau region originates
only due to different values of Tifor the three cases (Fig. 127, right panel). This, hence can
be used as a measure of Ti.

Heavy Ion Collisions at the LHC - Last Call for Predictions 156
We have observed that although the individual pTdistribution of photons and lepton pairs
are sensitive to different EOS (lattice QCD, for example) the ratio Rem is not. It is also noticed
that Rem in the plateau region is not sensitive to the medium effects on hadrons, radial flow,
Tc,Tfand other parameters.
It is interesting to note that the nature of variation of the quantity, RpQCD
em , which is the
corresponding ratio of photons and lepton pairs from hard processes only is quite different
from Rthermal
em for pTup to ∼3 GeV indicating that the observed saturation is a thermal effect.
8.2. Prompt photon in heavy ion collisions at the LHC: A “multi-purpose” observable
F. Arleo
I emphasize in this contribution how prompt photons can be used to probe nuclear parton densities
as well as medium-modified fragmentation functions in heavy ion collisions. Various predictions in
p–A and A–A collisions at LHC energies are given.
Prompt photon production in hadronic collisions has been extensively studied, both
experimentally and theoretically, over the past 25 years (see [351] and references therein).
As indicated in Ref. [351], it is remarkable that almost all existing data from fixed-target
to collider energies can be very well understood within perturbative QCD at NLO. In
these proceedings, I briefly discuss how prompt photons in nuclear collisions (p–A and
A–A) may allow for a better understanding of interesting aspects discussed in heavy-ion
collisions, namely the physics of nuclear parton distribution functions and medium-modified
fragmentation functions. Parton distribution functions in nuclei are so far poorly constrained,
especially in contrast with the high degree of accuracy currently reached in the proton
channel, over a wide xand Q2domain. In particular, only high-x(x&10−2) and low Q2
(Q2.100 GeV2) have been probed in fixed-target experiments. In order to predict hard
processes in nuclear collisions at the LHC, a more accurate knowledge on a wider kinematic
range is necessary. As stressed in [352], the nuclear production ratio of isolated photons in
p-A collisions,
RpA(xT,y)=1
Ad3σ
dyd2p⊥
(p+A→γ+X).d3σ
dyd2p⊥
(p+p→γ+X)
can be related to a good accuracy (say, less than 5%) to the parton density ratios
Rapprox(xT,y=0) ≃0.5RA
F2(xT)+0.5RA
G(xT) ; Rapprox (xT,y=3) ≃RA
G(xTe−y),
with xT=2p⊥/√sNN. To illustrate this, the ratio RpAis computed for isolated photons
produced at mid-rapidity in p–Pb collisions at √sNN =8.8 TeV in Fig. 128 (solid line),
assuming the de Florian and Sassot (nDSg) nuclear parton distributions [15]. The
above analytic approximation Rapprox
y=0(dotted line) demonstrates how well this observable is
connected to the nuclear modifications of the gluon density and structure function; see also the
agreement (RpA−Rapprox
y=0)/RpAas a dash-dotted line in Fig. 128. In nucleus-nucleus scattering,
the energy loss of hard quarks and gluons in the dense medium presumably produced at
LHC may lead to the suppression of prompt photons coming from the collinear fragmentation

Heavy Ion Collisions at the LHC - Last Call for Predictions 157
process. In Fig. 129, the expected photon quenching in Pb–Pb collisions at √sNN =5.5 TeV
is plotted. A significant suppression due to energy loss (taking ωc=50 GeV, see [16] for
details) is observed, unlike what is expected when only nuclear effects in the parton densities
are assumed in the calculation (dash-dotted line).
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
10 -3 10 -2
x⊥
RpPb (x⊥)
isolated γ NLO w/ nDSg
0.5 ( RF2 + RG )
√s = 8.8 TeV y = 0
Figure 128: RpPb of y=0 isolated photons
in p–Pb collisions at √sNN =8.8 TeV.
0
0.2
0.4
0.6
0.8
1
1.2
10 102
w/o nDSg
w/ nDSg
w/ nDSg ωc = 50 GeV
incl. γ √s=5.5 TeV y=0
p⊥ (GeV)
RPbPb (p⊥)
Figure 129: RPbPb of y=0 inclusive
photons in Pb–Pb collisions at √sNN =
5.5 TeV.
Finally performing momentum correlations between a prompt photon and a leading
hadron in p–pand A–A collisions, yet experimentally challenging, appears to be an
interesting probe of vacuum and medium-modified fragmentation function, as discussed in
detail in Refs. [353,354]. We refer in particular the interested reader to Fig. 10 of [353] for
the predictions of γ–π0momentum-imbalance distributions at the LHC.
8.3. Direct photon spectra in Pb-Pb at √sNN =5.5 TeV: hydrodynamics+pQCD predictions
F. Arleo, D. d’Enterria and D. Peressounko
The pT-differential spectra for direct photons produced in Pb-Pb collisions at the LHC, including
thermal (hydrodynamics) and prompt (pQCD) emissions are presented.
We present predictions for the transverse momentum distributions of direct-γ(i.e.
photons not coming from hadron decays) produced at mid-rapidity in Pb-Pb collisions at
√sNN =5.5 TeV based on a combined hydrodynamics+pQCD approach. Thermal photon
emission in Pb-Pb at the LHC is computed with a hydrodynamical model successfully used in
nucleus-nucleus collisions at RHIC energies [10]. The initial entropy density of the produced
system at LHC is obtained by extrapolating empirically the hadron multiplicities measured
at RHIC [355]. Above pT≈3 GeV/c, additional prompt-γproduction from parton-parton
scatterings is computed perturbatively at next-to-leading-order (NLO) accuracy [356]. We use
recent parton distribution functions (PDF) [13] and parton-to-photon fragmentation functions

Heavy Ion Collisions at the LHC - Last Call for Predictions 158
(FF) [357], modified resp. to account for initial-state shadowing+isospin effects [15] and
final-state parton energy loss [358].
We follow the evolution of the hot and dense system produced in central Pb-Pb at LHC
by solving the equations of (ideal) relativistic 2D+1 hydrodynamics [10, 355] starting at
a time τ0=1/Qs≈0.1 fm/c. The system is assumed to have an initial entropy density
of s0=1120 fm−3, which corresponds to a maximum temperature at the center of T0≈
650 MeV (hT0i ≈ 470 MeV). We use a quark gluon plasma (QGP) and hadron resonance
gas (HRG) equation of state above and below Tcrit ≈170 MeV resp., connected by a
standard Maxwell construction assuming a first-order phase transition at Tcrit. Thermal
photon emission is computed using the most recent parametrizations of the QGP and HRG
γrates. For the QGP phase we use the AMY complete leading-log emission rates including
LPM suppression [240]. For the HRG phase, we employ the improved parametrization from
Turbide et al. [359].
Our NLO pQCD predictions are obtained with the code of ref. [356] with all scales
set to µ=pT. Pb-Pb yields are obtained scaling the NLO cross-sections by the number of
incoherent nucleon-nucleon collisions: Ncoll =1670, 12.9 for 0-10% central (hbi=3.2 fm)
and 60-90% peripheral (hbi=13 fm). Nuclear (isospin and shadowing) corrections of the
CTEQ6.5M PDFs [13] are introduced using the NLO nDSg parametrization [15]. At relatively
low pT, prompt photon yields have a large contribution from jet fragmentation processes.
As a result, final-state parton energy loss in central Pb-Pb affects also the expected prompt
γyields. We account for medium-effects on the γ-fragmentation component by modifying
the BFG parton-to-photon FFs [357] with BDMPS quenching weights. The effects of the
energy loss are encoded in a single parameter, ωc=hˆqiL2≈50 GeV, extrapolated from RHIC.
The combination of initial-state (shadowing) and final-state (energy loss) effects results in a
quenching factor for prompt photons of RPbPb ≈0.2 (0.8) at pT=10 (100) GeV/c[358].
Our predictions for the direct photon spectra at y=0 in Pb-Pb at 5.5 TeV are shown in
Fig. 130. The thermal contribution dominates over the (quenched) pQCD one up to pT≈
4 (1.5) GeV/cin central (peripheral) Pb-Pb. Two differences are worth noting compared to
RHIC results [10]: (i) the thermal-prompt crossing point moves up from pT≈2.5 GeV/c
to pT≈4.5 GeV/c, and (ii) most of the thermal production in this transition region comes
solely from the QGP phase. Both characteristics make of semi-hard direct photons at LHC, a
valuable probe of the thermodynamical properties of the system.
8.4. Elliptic flow of thermal photons from RHIC to LHC
R. Chatterjee, E. Frodermann, U. Heinz and D. K. Srivastava
We use the longitudinally boost-invariant relativistic ideal hydrodynamic code
AZHYDRO [130] to predict the evolution from RHIC to LHC of the transverse
momentum spectra and elliptic flow of thermal photons and dileptons at mid-rapidity in
(A≈200)+(A≈200) collisions. Here we discuss only photons for Au+Au collisions at b=7 fm;
for other results and more details see Refs. [360].
The hydrodynamic initial conditions for RHIC collisions are described in [360]. For the

Heavy Ion Collisions at the LHC - Last Call for Predictions 159
(GeV/c)
T
p
0 1 2 3 4 5 6 7 8
-2
dy) (GeV/c)
2
T
dpπN/(
2
d
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
1
10
2
10
3
10 : prompt + thermalγTotal = 50 GeV
c
ω
loss
, E
coll
N×Prompt: NLO
Thermal: QGP
Thermal: HRG
+X, 5.5 TeV [0-10% central]γ →Pb-Pb
=0.1 fm/c) = 650 MeV
0
τ(
0
T
(GeV/c)
T
p
0 1 2 3 4 5 6 7 8
-2
dy) (GeV/c)
2
T
dpπN/(
2
d
-8
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
1
10 +X, 5.5 TeV [60-90% periph]γ →Pb-Pb
: prompt + thermalγTotal
[60-90%]
coll
N×Prompt: NLO pQCD
Thermal: QGP
Thermal: HRG
Figure 130: Direct-γspectra in 0-10% central (left) and 60-90% peripheral (right) Pb-Pb
at √sNN =5.5 TeV, with the thermal (QGP and HRG) and prompt (pQCD) contributions
differentiated.
LHC simulations shown in comparison we assume a final charged hadron multiplicity near
the upper end of the predicted range: dNch
dy (b=y=0)=2350 (680 at RHIC). Correspondingly
we increase the initial peak entropy density in central Au+Au collisions from s0=351 fm−3
at τ0=0.2fm/cfor RHIC to s0=2438 fm−3at τ0=0.1 fm/cfor LHC.
1. Thermal photon spectra: Figure 131 shows the thermal photon pT-spectra (angle-
integrated) for RHIC and LHC. At both collision energies the total spectrum is dominated
by quark matter once pTexceeds a few hundred MeV. Its inverse slope (“effective tempera-
ture”) in the range 1.5<pT<3 GeV/cincreases by almost 50%, from 303MeV at RHIC to
442MeV at LHC, reflecting the higher initial temperature and significantly increased radial
flow (visible in the HM contribution) at LHC.
2. Thermal photon elliptic flow: Figure 132 shows the differential elliptic flow of thermal
photons at RHIC and LHC, with quark matter (QM) and hadronic matter (HM) radiation
shown separately for comparison. The decrease at high pTof the QM and total photon
v2reflects the dominance of QM radiation at high pT(emission from the early, hot stage
when radial and elliptic flow are still small). At fixed pT, the photon elliptic flow from QM
radiation is larger at LHC than at RHIC since the LHC fireballs start hotter and fluid cells
with a given temperature thus flow more rapidly. At low pT, hadronic radiation dominates,
and since it flows more rapidly at LHC than at RHIC the corresponding photon elliptic is
significantly larger at LHC than RHIC. This is different from hadrons whose elliptic flow at
low pTdecreases from RHIC to LHC, reflecting a redistribution of the momentum anisotropy
to higher pTby increased radial flow [361]. For photons, the elliptic flow is not yet saturated
at RHIC, and at low pTit keeps increasing towards LHC at a rate that overwhelms the loss of
momentum anisotropy to the high-pTdomain via radial flow. Contrary to pion v2[361], the

Heavy Ion Collisions at the LHC - Last Call for Predictions 160
pT-integrated photon elliptic flow roughly doubles (!) from RHIC to LHC.
8.5. Asymmetrical in-medium mesons
I. M. Dremin
Cherenkov gluons may be in charge of mass asymmetry of in-medium mesons which reveals
itself in the asymmetry of dilepton spectra.
The hypothesis about the nuclear analogue of the well known Cherenkov effect [194–
198] is widely discussed now. The necessary condition for Cherenkov effect in usual or
hadronic media is the excess of the corresponding refractive index nover 1. There exists the
general linear relation between this excess ∆n=n−1 and the real part of the forward scattering
amplitude F(E,0o). In electrodynamics, it is the dipole excitation of atoms in the medium by
light which results in the Breit-Wigner shape of the photon amplitude. In a nuclear medium,
this should be the amplitude of gluon scattering on some internal modes of the medium. In
absence of the theory of such media I prefer to rely on our knowledge of hadronic reactions.
From experiments at comparatively lowenergies we learn that the resonances are abundantly
produced. They are described by the Breit-Wigner amplitudes which have a common feature
00.5 11.5 22.5 3
pT (GeV)
1e-06
0.0001
0.01
1
100
dN/d2pTdy (GeV-2)
(QM+HM)@LHC
(QM+HM)@RHIC
QM@LHC
QM@RHIC
HM@LHC
HM@RHIC
Thermal Photons
b= 7 fm
Figure 131: (Color online) Thermal photon spectra Au+Au collisions at RHIC and Pb+Pb
collisions at LHC, both at b=7fm.

Heavy Ion Collisions at the LHC - Last Call for Predictions 161
0 1 2 3 4 5 6
pT (GeV)
0
0.025
0.05
0.075
0.1
v2(pT)
v2(QM+HM)@LHC
v2(QM+HM)@RHIC
v2(QM)@LHC
v2(QM)@RHIC
0.2*v2(HM)@LHC
0.2*v2(HM)@RHIC
Thermal Photons
b= 7 fm
Figure 132: (Color online) Thermal photon elliptic flow for Au+Au collisions at RHIC
(dashed) and Pb+Pb at LHC (solid lines), both at b=7 fm.
of the positive real part in the low-mass wing (for the electrodynamic analogy see, e.g.,
Feynman lectures). Therefore the hadronic refractive index exceeds 1 in these energy regions.
Prediction Masses of Cherenkov states are less than in-vacuum meson masses. This leads to
the asymmetry of decay spectra of resonances with increased role of low masses.
Proposal Plot the mass distribution of π+π−,µ+µ−,e+e−-pairs near resonance peaks. Thus,
apart from the ordinary Breit-Wigner shape of the cross section for resonance production, the
dilepton mass spectrum would acquire the additional term proportional to ∆n(that is typical
for Cherenkov effects) at masses below the resonance peak [362]. Therefore its excess (e.g.,
near the ρ-meson) can be described by the following formula♯
dNll
dM =A
(m2
ρ−M2)2+M2Γ21+wm2
ρ−M2
M2θ(mρ−M)(61)
Here Mis the total c.m.s. energy of two colliding objects (the dilepton mass), mρ=775 MeV
is the in-vacuum ρ-meson mass. The first term corresponds to the Breit-Wigner cross section.
According to the optical theorem it is proportional to the imaginary part of the forward
scattering amplitude. The second term is proportional to ∆nwhere the well known ratio of real
to imaginary parts of Breit-Wigner amplitudes has been used. It vanishes for M>mρbecause
♯We consider only ρ-mesons here because the most precise experimental data are available for them. To include
other mesons, one should evaluate the corresponding sum of similar expressions.

Heavy Ion Collisions at the LHC - Last Call for Predictions 162
only positive ∆nlead to the Cherenkov effect. Namely it describes the distribution of masses
of Cherenkov states. In Eq.(61) one should take into account the in-medium modification of
the height of the peak and its width. We just fit the parameters Aand Γby describing the
shape of the mass spectrum at 0.75 <M<0.9 GeV measured in [363–365]. Let us note that
wis not used in this procedure. The values A=104GeV3and Γ = 0.354 GeV were obtained.
The width of the in-medium peak is larger than the in-vacuum ρ-meson width equal to 150
MeV.
Thus the low mass spectrum at M<mρdepends only on a single parameter wwhich is
determined by the relative role of Cherenkov effects and ordinary mechanism of resonance
production. It is clearly seen from Eq.(61) that the role of the second term in the brackets
increases for smaller masses M. The excess spectrum [363–365]. in the mass region from 0.4
GeV to 0.75 GeV has been fitted by w=0.19. The slight downward shift about 40 MeV of
the peak of the distribution compared with mρmay be estimated from Eq.(61) at these values
of the parameters.
Whether the in-medium Cherenkov gluonic effect is strong can be verified by measuring
the angular distribution of the lepton pairs with different masses. The trigger-jet experiments
similar to that at RHIC are necessary to check this prediction. One should measure the angles
between the companion jet axis and the total momentum of the lepton pair. The Cherenkov
pairs with masses between 0.4 GeV and 0.7 GeV should tend to fill in the rings around the jet
axis. The angular radius θof the ring is determined by the usual condition
cosθ=1
n(62)
Another way to demonstrate it is to measure the average mass of lepton pairs as a function of
their polar emission angle (pseudorapidity) with the companion jet direction chosen as z-axis.
Some excess of low-mass pairs may be observed at the angle (62).
The prediction of asymmetric in-medium widening of any resonance at its low-mass side
due to Cherenkov gluons is universal. This universality is definitely supported by experiment.
Very clear signals of the excess on the low-mass sides of ρ,ω and φmesons have been seen
in KEK. This effect for ω-meson is also studied by CBELSA/TAPScollaboration. There are
some indications at RHIC on this effect for J/ψ-meson.
To conclude, the universal asymmetry of in-medium mesons with an excess over the
usual Breit-Wigner form at low masses is predicted as a signature of Cherenkov gluons
produced with energies which fit the left wings of resonances where nexceeds 1.
8.6. Photons and Dileptons at LHC
R. J. Fries, S. Turbide, C. Gale and D. K. Srivastava
We discuss real and virtual photon sources in heavy ion collisions and present results for dilepton
yields in Pb+Pb collisions at the LHC at intermediate and large transverse momentum pT.
Electromagnetic radiation provides a valuable tool to understand the dynamics of heavy
ion collisions. Due to their long mean free path real and virtual photons carry information

Heavy Ion Collisions at the LHC - Last Call for Predictions 163
from very early times and from deep inside the fireball. We discuss the sources of photons
which will be important for the upcoming heavy ion experiments at LHC. We focus on
intermediate and large transverse momenta pTand masses M. We also present our numerical
results for dilepton yields.
At asymptotically large pTthe most important source of real and virtual photons is the
direct hard production in primary parton-parton collisions between the nuclei, via Compton
scattering, annihilation, and the Drell-Yan processes. These photons do not carry any
signature of the fireball. They are augmented by photons fragmenting from hard jets also
created in primary parton-parton collisions. The emission of this vacuum bremsstrahlung
is described by real and virtual photon fragmentation functions. Vacuum fragmentation is
assumed to happen outside the fireball, so the jets are subject to the full energy loss in the
medium. This contribution to the photon and dilepton yield is therefore depleted in heavy ion
collisions analogous to the high-pThadron yield.
At intermediate scales jet-induced photons from the medium become important. It has
been shown that high-pTjets interacting with the medium can produce real and virtual photons
by one of two processes: (i) by Compton scattering or annihilation with a thermal parton,
leading to an effective conversion of the jet into a photon [366]; (ii) by medium induced
Bremsstrahlung [367]. Jet-medium photons have a steeper spectrum than primary photons
and carry information about the temperature of the medium. They are also sensitive to the
partial energy loss that a jet suffers from its creation to the point of emission of the photon. At
even lower pTand Mthermal radiation from the quark gluon plasma (and also the hadronic
phase not considered here) has to be taken into account.
Figure 133 shows numerical evaluations of the different contributions discussed above to
the e+e−transverse momentum and mass spectrum for central Pb+Pb collisions at LHC. We
use next-to-leading order pQCD calculations for Drell Yan and a leading order calculation
for jet production. Energy loss of jets is computed with the AMY formalism [368]. Jet-
medium emission and thermal emission have been evaluated in the Hard Thermal Loop (HTL)
resummation scheme. For the mass spectrum we also show the expected background from
correlated heavy quark decays. The full calculation for dileptons with a more extended
discussion is presented in [369]. Predictions for direct photon yields including jet-medium
photons can be found in [368].
Dileptons from jet-medium interactions will be more important at LHC than at previous
lower energy experiments. They will be as important or even exceeding the Drell-Yan yields
at intermediate masses up to about 8 GeV. They offer a new way to access information about
the temperature and the partonic nature of the fireball.
8.7. Direct photons at LHC
A. H. Rezaeian, B. Z. Kopeliovich, H. J. Pirner and I. Schmidt
The DGLAP improved color dipole approach provides a good description of data for inclusive
direct photon spectra at the energies of RHIC and Tevatron. Within the same framework we predict the
transverse momentum distribution of direct photons at the CERN LHC energies.

Heavy Ion Collisions at the LHC - Last Call for Predictions 164
012 3 4 5678 9 10
M (GeV)
10-11
10-10
10-9
10-8
10-7
10-6
10-5
10-4
dNe+e-/dydM 2 (GeV-2)
heavy quarks decay
Drell-Yan (direct)
jet-therm, HTL
jet-frag (E-loss)
therm-therm, HTL
0-10% Central
Ti= 845 MeV
yd =0, |ye| < 0.5
PT > 8 GeV
LHC
8 10 12 14 16
pT (GeV)
10-11
10-10
10-9
10-8
10-7
10-6
10-5
dNe+e-/dyd2pT (GeV-2)
Drell-Yan (direct)
jet-frag. (E-loss)
jet-therm, HTL
therm-therm, HTL
Sum
0.5 GeV < M < 1 GeV
LHC 0-10% Central
|ye| < 0.5
yd=0
Figure 133: The yield of e+e−pairs in central Pb+Pb collisions at √sNN =5.5 TeV. Left: Mass
spectrum dN/(dyddM2) integrated over the transverse momentum pTof the pair for pT>8
GeV/c.Right: Transverse momentum spectrum dN/(dydd2pT) integrated over a mass range
0.5 GeV <M<1 GeV. Both panels show the case yd=0 for the pair rapidity ydand a cut
|ye|<0.5 for the single electron rapidity.
8.7.1. Introduction Direct photons, i.e. photons not from hadronic decay, provide a powerful
probe for the initial state of matter created in heavy ion collisions, since they interact with the
medium only electromagnetically and therefore provide a baseline for the interpretation of jet-
quenching models. The primary motivationfor studying the direct photons has been to extract
information about the gluon density inside proton in conjunction with DIS data. However, this
task has yet to be fulfilled due to difference between the measurement and perturbative QCD
calculation which is difficult to explain by altering the gluon density function (see Ref. [370]
and references therein). We have recently shown that the color dipole formalism coupled to
DGLAP evolution is an viable alternativeto the parton model and provided a good description
of inclusive photon and dilepton pair production in hadron-hadron collisions [370]. Here we
predict the transverse momentum spectra of direct photons at the LHC energies √s=5.5 TeV
and 14 TeV within the same framework.
8.7.2. Color dipole approach and predictions for LHC Although in the process of
electromagnetic bremsstrahlung by a quark no dipole participates, the cross section can be
expressed via the more elementary cross section σq¯qof interaction of a ¯qq dipole. For
the dipole cross section, we employ the saturation model of Golec-Biernat and Wüsthoff
coupled to DGLAP evolution (GBW-DGLAP) [371] which is better suited at large transverse
momenta. Without inclusion of DGLAP evolution, the direct photon cross section is
overestimated [370]. In Fig. 134, we show the GBW-DGLAP dipole model predictions for
inclusive direct photon production at midrapidities for RHIC, CDF and LHC energies. We
stress that the theoretical curves in Fig. 134, are the results of a parameter free calculation.
Notice also that in contrast to the parton model, neither K-factor (NLO corrections), nor higher
twist corrections are to be added. No quark-to-photon fragmentation function is needed either.

Heavy Ion Collisions at the LHC - Last Call for Predictions 165
0 2 4 68 10 12 14 16 18 20 22 24
pT(GeV)
10-2
10-1
100
101
102
103
104
Ed3σ/d3P [pb/GeV2]
PHENIX, √
s = 200 GeV
√
s = 200 GeV
20 40 60 80 100 120 140 160 180 200
pT(GeV)
10-2
10-1
100
101
102
103
104
105
d2σ/dpTdη [pb/GeV]
CDF, √
s = 1.8 TeV
√
s = 1.8 TeV
√
s = 5.5 TeV
√
s = 14 TeV
Figure 134: Direct photon spectra obtained from GBW-DGLAP dipole model at midrapidity
for RHIC, CDF and LHC energies. Experimental data (right) are for inclusive isolated photon
from CDF experiment for |η|<0.9 at √s=1.8 TeV [372] and (left) for direct photon at η=0
for RHIC energy √s=200 GeV [373]. The error bars are the linear sum of the statistical and
systematic uncertainties.
Indeed, the phenomenological dipole cross section is fitted to DIS data and incorporates all
perturbative and non-perturbative radiation contributions. For the same reason, in contrast
to the parton model, in the dipole approach there is no ambiguity in defining the primordial
transverse momentum of partons. Such a small purely non-perturbative primordial momentum
does not play a significant role for direct photon production at the given range of pTin
Fig. 134. Notice that the color dipole picture accounts only for Pomeron exchange from
the target, while ignoring its valence content. Therefore, Reggeons are not taken into account,
and as a consequence, the dipole is well suited mainly for high-energy processes. As our
result for RHIC and CDF energies indicate, we expect that dipole prescription to be at work
for the LHC energies. At the Tevatron, in order to reject the overwhelming background of
secondary photons isolation cuts are imposed [372]. Isolation conditions are not imposed
in our calculation. However, the cross section does not vary by more than 10% under CDF
isolation conditions [370]. One should also notice that the parametrizations of the dipole cross
section and proton structure function employed in our computation have been fitted to data at
considerably lower pTvalues [370].
8.8. Thermal Dileptons at LHC
H. van Hees and R. Rapp
We predict dilepton invariant-mass spectra for central 5.5 ATeV Pb-Pb collisions at LHC.
Hadronic emission in the low-mass region is calculated using in-medium spectral functions of light
vector mesons within hadronic many-body theory. In the intermediate-mass region thermal radiation
from the Quark-Gluon Plasma, evaluated perturbatively with hard-thermal loop corrections, takes over.
An important source over the entire mass range are decays of correlated open-charm hadrons, rendering
the nuclear modification of charm and bottom spectra a critical ingredient.

Heavy Ion Collisions at the LHC - Last Call for Predictions 166
Due to their penetrating nature, electromagnetic probes (dileptons and photons) are an
invaluable tool to investigate direct radiation from the hot/dense matter created in heavy-ion
collisions. At low invariant mass, M≤1 GeV, the main source of dileptons is the decay of the
light vector mesons, ρ,ωand φ, giving unique access to their in-medium spectral properties,
most prominently for the short-lived ρmeson. If the chiral properties of the ρ-meson can be
understood theoretically, dilepton spectra can serve as a signal for the restoration of chiral
symmetry at high temperatures and densities.
We employ medium-modified vector-meson spectral functions in hot/dense matter
following from hadronic many-body theory, phenomenologically constrained by vacuum ππ
scattering, decay branching ratios for baryonic and mesonic resonances, photo-absorption
cross sections on nucleons and nuclei, etc. [374]. The resulting spectral functions, especially
for the ρmeson, exhibit large broadening with little mass shift, with baryonic interactions
as the prevalent agent, especially in the mass region below the resonance peaks. Note that
CP invariance of strong interactions implies equal interactions with baryons and antibaryons.
Thus, even in a net-baryon free environment, the ρresonance essentially “melts” around the
expected phase transition temperature, Tc≃180 MeV. Other sources of thermal dileptons taken
into account are (i) four-pion type annihilation in the hadronic phase (augmented by chiral
vector-axialvector mixing) [375], which takes over the resonance contributions at intermediate
mass, and (ii) radiation from the Quark-Gluon Plasma (QGP), computed within hard-thermal
loop improved perturbation theory for in-medium q-¯qannihilation.
Thermal dilepton spectra are computed by evolving pertinent emission rates over the
time evolution of the medium in central 5.5 ATeV Pb-Pb collisions. To this end, we employ a
cylindrical homogeneous thermal fireball with isentropic expansion and a total entropy fixed
by the number of charged particles, which we estimate from a phenomenological extrapolation
to be dNch/dy≃1400. We use an ideal-gas equation of state (EoS) with massless gluons and
Nf=2.5 quark flavors for the QGP, and a resonance gas for the hadronic EoS with chemical
freezeout at (µc
B,Tc)=(2,180) MeV (finite meson and anti-/baryon chemical potentials are
implemented to conserve the particle ratios until thermal freezeout at Tfo≃100 MeV, with a
mass-action law for short-lived resonances). We start the evolution in the QGP phase at initial
time τ0=0.17 fm/c, translating into T0≃560 MeV. The volume expansion parameters are taken
to resemble hydrodynamic simulations. A standard mixed-phase construction connects QGP
and hadronic phase at Tc, and the total fireball lifetime is τfb≃18 fm/c.
As for non-thermal sources, we include primordial Drell-Yan annihilation and decays
of correlated charm pairs. The latter are estimated by scaling the spectrum at RHIC with a
charm-cross section anticipated at LHC, which implies somewhat softer charm spectra than
expected for primordial N-N collisions (and thus softer invariant-mass spectra). We neglect
contributions from jet-plasma interactions.
Our predictions are summarized in Fig. 135. At low mass thermal dileptons are
dominated by hadronic radiation, with large modifications due to in-medium vector-meson
spectral functions. The QGP contribution takes over at around M&1.1 GeV. The yield
from correlated open-charm decays is comparable to hadronic emission already at low mass,
and dominant at intermediate mass. However, this result will have to be scrutinized by

Heavy Ion Collisions at the LHC - Last Call for Predictions 167
0.2 0.4 0.6 0.8 1.0 1.2
Mee [GeV]
10-3
10-2
10-1
dNee/(dM dy) [GeV-1]
free HG
in-med HG
QGP
DD pte>0.2GeV
Central Pb+Pb √s=5.5 ATeV
|ye|<0.35
dNch/dy=1400
2.0 3.0 4.0 5.0
Mee [GeV]
10-6
10-5
10-4
10-3
10-2
dNee/(dM dy) [GeV-1]
Hadron Gas
QGP
Drell-Yan
sum
corr. charm
Central Pb+Pb √s=5.5 ATeV
|ye|<0.35
dNch/dy=1400
Figure 135: (Color online) Predictions for dilepton spectra in central 5.5 ATeV Pb-Pb
collisions at LHC in the low- (left panel) and intermediate-mass region (right panel).
including the nuclear modification of heavy-quark spectra in the QGP (as well as analogous
contributions from correlated bottom decays). Also, larger values of dNch/dywould help to
outshine correlated open-charm decays, at least at low mass.
8.9. Direct γproduction and modification at the LHC
I. Vitev
Baseline direct photon production cross sections arestudied in √s=5.5 TeV p+p collisions at the
LHC. The fraction of fragmentation photons, which suffer QGP effects, is shown to be non-negligible
even at very high pT∼200 GeV. We first examine important cold nuclear matter effects for direct
photon production, related to dynamical shadowing, isospin and initial state energy loss, in comparison
to neutral pion production at √s=200 GeV. Simulations of direct γsuppression in Pb+Pb reactions at
s1/2=5.5 A.TeV at the LHC are also presented to high transverse momentum. Results are given in for
central nuclear collisions and energy loss in the QGP calculated in the GLV approach. Direct photon
quenching is shown to strongly depend on the ratio γprompt/γfragmentation At high pT>100 GeV cold
nuclear matter attenuation can be as large as the QGP effects for the net suppression of direct photons.
It has been argued that direct photon production and direct photon tagged jets provide
error-free gauge for the quenching of quarks and gluons and for fixing their initial energy.
We show that quantitatively large nuclear corrections must be taken into account for direct
γto become precision probes of the QGP. The left panel of Fig. 136 shows the direct
photon production cross section in p+p collisions at √s=5.5 TeV the LHC compared to
the corresponding cross section at RHIC √s=200 GeV to LO in perturbative QCD [376].
Insert shows the fraction of fragmentation to prompt photons versus pT. The right panel of
Fig. 136 shows cold nuclear effects, the Cronin [283], dynamical shadowing [377] and cold
nuclear matter energy loss [285], in d+A reactions at LHC energies. Comparison to data in
0-20% central d+Au collisions at RHIC is also presented.

Heavy Ion Collisions at the LHC - Last Call for Predictions 168
The left panel of Fig. 137 shows the QGP effect (final-state interactions) in central Pb+Pb
collisions at √s=5.5 TeV. Parton rapidity densities dNg/dy ∼2000−4000 [283], as for π0
quenching and heavy meson dissociation, are used. Direct photon quenching closely follows
the ratio γprompt/γfragmentation [376]. At low pTattenuation is QGP-dominated with significant
and measurable suppression RAA(pT)∼0.5. Nevertheless, such quenching is smaller than
the one for π0’s and reflects the CF/CAaverage squared color charge difference for quark
and gluon jets. The right panel of Fig. 137 includes the effect of initial-state cold nuclear
matter energy loss. At high pTthese can be comparable to the final-state quenching in the
QGP [285,376,377].
9. Others
9.1. The effects of angular momentum conservation in relativistic heavy ion collisions at the
LHC
F. Becattini and F. Piccinini
We argue that in peripheral heavy ion collisions at the LHC there might be the formation of
a spinning plasma with large intrinsic angular momentum. If the angular momentum is sufficiently
large, there could be striking observable effects: a decrease of chemical freeze-out temperature and an
increase of transverse momentum spectra broadening (enhanced radial flow) as a function of centrality;
0510 15 20
pT [GeV]
10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
dσγ /dyd2pT [mb.GeV-2]
s1/2 = 200 GeV
050 100 150 200
pT [GeV]
0
0.5
1
1.5
2
Frag. / Pr.
050 100 150 200
pT [GeV]
10-14
10-13
10-12
10-11
10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
dσγ /dyd2pT [mb.GeV-2]
s1/2 = 200 GeV
s1/2 = 5500 GeV
0 10 20 30
pT [GeV]
0
0.5
1
1.5
2
Frag. / Pr.
K = 1.9
p+p collisions
s1/2 = 200 GeV
<kT2> = 0.9 GeV2
RHIC
LHC
p+p collisions
050 100 150 200
pT [GeV]
0.5
0.75
1
1.25
1.5
Rd+Pb (pT)
Cronin + HTS + EMC
Cronin + HTS + EMC + E-loss
0 2 4 68 10 12 14 16
pT [GeV]
0.5
0.75
1
1.25
1.5
1.75
2
Rd+Au (pT)
Cronin effect only
Cronin + HTS + EMC effects
Cronin + HTS + EMC + E-loss
PHENIX π0, 0-20% central
Direct γ
0-20% central
0-20% central d+Pb
Figure 136: Left panel: Direct photon production cross section in p+p collisions at the LHC
√s=5.5 TeV. Comparison to the same cross section calculation at RHIC at √s=200 GeV
and to current high pTdata is also shown. Insert illustrates the ratio of fragmentation to
prompt photons vs pTat LO. Right panel: Nuclear modification factor RdA in central d+Au
collisions at RHIC and central d+Pb at the LHC, 0-20%. The high pTbehavior indicates
the isospin (charge) effect and initial-state energy loss in cold nuclear matter. Comparison to
similar effects on neutral pion production in d+Au collisions at RHIC, indicative for the first
time for cold nuclear matter −∆Erad effects at high pTis also shown.

Heavy Ion Collisions at the LHC - Last Call for Predictions 169
a large enhancement of elliptic flow; a polarization of emitted particles along the direction of angular
momentum. The latter would be the cleanest signature of such effect.
In peripheral relativistic heavy ion collisions colliding ions have a large relative orbital
angular momentum. While the fragments keep flying away from the interaction region
essentially unaffected, a fraction of the initial angular momentum is transferred to the
interaction region. Much of it is probably spent into relative orbital angular momentum of
the newly formed fireballs at large rapidity, but it may happen that another significant fraction
is given to the midrapidity region giving rise to a spinning plasma with an intrinsic angular
momentum J. If Jis sufficiently large, one has remarkable observable effects. It has been
suggested that such a phenomenon can produce an azimuthal anisotropy in the transverse
plane very similar to the well known elliptic flow [150]. Also, a large Jmay result in a
polarization of emitted particles [378]. We make a quantitative determination of observable
effects by assuming that the spinning system is at statistical equilibrium, taking advantage of
a recent calculation of the microcanonical partition function of a relativistic quantum gas with
fixed angular momentum [379, 380] which allowed us to provide the expression of particle
spin density matrix and polarization in a rotating thermodynamical system. Here, a possible
scenario for the LHC energy is just sketched; a more detailed paper will appear [380].
Under reasonable assumptions, the main observables which signal the presence of
an equilibrated spinning system are (see figure 138): a decrease of chemical freeze-out
temperature and an increase of transverse momentum spectra broadening (enhanced radial
flow) as a function of centrality; a large enhancement of elliptic flow and a polarization of
emitted particles along the direction of angular momentum. The latter is the cleanest signature
of a spinning system. These observables scale with the parameter J/T4
cR4,Tcbeing the critical
temperature and Rthe maximal transverse radius of the system. They are shown in figures
below, as a function of the impact parameter or transverse momentum, for the upper bound
050 100 150 200
pT [GeV]
0
0.25
0.5
0.75
1
1.25
1.5
Rd+Pb (pT), RPb+Pb (pT)
Cronin + HTS + EMC
Cronon + HTS + EMC + QGP E-loss
Direct γ
Central Pb+Pb, s1/2 = 5.5 TeV
dNg/dy ~ 2000 - 4000
050 100 150 200
pT [GeV]
0
0.25
0.5
0.75
1
1.25
1.5
Rd+Pb (pT), RPb+Pb (pT)
Cronin + HTS + EMC + E-loss
Cronon + HTS + E-loss + QGP E-loss
Direct γ
Central Pb+Pb, s1/2 = 5.5 TeV dNg/dy ~ 2000 - 4000
Figure 137: Left panel: Comparison of cold nuclear matter effects to QGP effects on
direct photon production at the LHC. Central d+Pb and central Pb+Pb at √s=5.5 TeV
are shown. Calculations do not include initial-state energy loss. QGP suppression trend
with dNg/dy ∼2000−4000 follows the fragmentation/prompt ratio for direct γ. Right panel:
Similar calculations including initial-state cold nuclear matter energy loss effects. Note that
these can yield 50% larger suppression at high pT.

Heavy Ion Collisions at the LHC - Last Call for Predictions 170
of this parameter set by the RHIC Λpolarization measurement (=0.2, blue line) and at LHC
(=1.0, black line) under the assumption of a scaling of J/T4
cR4by √s/ln √s5/3.
Caveat: the calculations shown in the plots concern only primary hadrons emitted from an
equilibrated source. Dilution effects such as resonance decays, perturbative production at
large pTand partial equilibration are not taken into account.
b (fm)
Tchem(MeV)
Jmax/Tc4 Re4 = 0.2
Jmax/Tc4 Re4 = 1.0
140
145
150
155
160
165
170
0 2 4 6 8 10 12
b(fm)
Tslope (MeV)
Jmax/Tc4 Re4 = 0.2
Jmax/Tc4 Re4 = 1.0
160
170
180
190
200
210
220
0 2 4 6 8 10 12
Jmax/Tc4Re4=0.2
Centrality 20-70%
pT (GeV)
v2(J)
Jmax/Tc4Re4=1.0
Centrality 15-45%
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Jmax/T5Re5=0.02
Centrality 20-70%
Jmax/T5Re5=0.1
Centrality 15-45%
pT (GeV)
<PΛ>y
0
0.05
0.1
0.15
0.2
0.25
0.3
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Figure 138:
9.2. Black hole predictions for LHC
H. Stöcker and B. Koch
The speculative prediction of the production of microscopical black holes, which would be
possible at the large hadron collider due to large extra dimensions, is discussed. We review observables
for such black holes and for the their possibly stable final state.

Heavy Ion Collisions at the LHC - Last Call for Predictions 171
9.2.1. From the hierarchy-problem to black holes in large extra dimensions One of the
problems in the search for a unified description of gravity and the forces of the standard model
(SM), is the fact that the Planck-scale mPl ∼1019 GeV (derived from Newtons constant GN) is
much bigger than the energy scales like the Z-mass mZ∼90 GeV. This huge difference is the
so-called hierarchy problem. Several theories can explain this hierarchy by the assumption
of extra spatial dimensions [381–383]. These theories assume a true fundamental scale Mf
which is of the order of just a few TeV and they interpret the Planck scale mPl as an effective
magnitude which comes into the game due to unobservable and compactified extra spatial
dimensions. In the model suggested by Arkani-Hamed, Dimopoulos and Dvali [381, 382]
the dextra space-like dimensions are compactified on tori with radii R. In this model the
SM particles are confined to our 3+1-dimensional sub-manifold (brane) and the gravitons
are allowed to propagate freely in the (3+d)+1-dimensional bulk. Planck mass mPl and the
fundamental mass Mfare related by
m2
Pl =Md+2
fRd.(63)
One exciting consequence of such models is that up to 109black holes (BH) might be
produced at the Large Hadron Collider (LHC) [384]. The intuitive approximation of the cross
section for such events can be made by using the Hoop conjecture and taking the classical
area of the (to be produced) BH with radius RH
σ(M)≈πR2
H,(64)
where Mis the BH mass. The Scharzschild radius is given at distances smaller than the size
of the extra dimensions by
Rd+1
H=2
d+1 1
Mf!d+1M
Mf.(65)
This radius is much larger than the Schwarzschild radius corresponding to the same BH mass
in 3+1 dimensions, which translates directly into a much larger cross section (64). This esti-
mate seems to keep its validity also in more elaborated picture .
9.2.2. From black hole evaporation to LHC observables Once a BH is produced it is
assumed to undergo a rapid evaporation process. This happens first in the so called
bolding phase where angular momentum and internal degrees of freedom are assumed to
be radiated off. For a BH mass much bigger than the fundamental mass scale (M≫Mf) the
following phase is the Hawking phase, where particles are thermally radiated offaccording
to the Hawking temperature †† [385] TH≈Mf(Mf/M)1/(d+1). As soon as the BH mass
becomes comparable to the fundamental mass scale, the underlying physics of the BH is
not understood and exact predictions are hardly possible at the current state of knowledge.
Discussed scenarios reach from a sudden final explosion over a slowed down evaporation to
the formation of stable black hole remnant (BHR) As most BHs would be produced close to
††The process of Hawking radiation would in principle allow to transform the BH mass into thermal energy and
was therefore subject to further speculations

Heavy Ion Collisions at the LHC - Last Call for Predictions 172
the production threshold the experimental outcome will be influenced strongly by this final
phase of BH evolution.
We analyzed the predictions for different scenarios. It turned out that the, suppression of
hard (TeV) di-jets above the BH formation threshold would bethe most scenario independent
observable for the LHC. Other observables such as event multiplicities or pTdistributions
should be definitely studied although they are more model dependent. Speculations about the
formation of BHRs can be tested experimentally at the LHC: Charged stable BHRs would
leave single stifftracks in the LHC detectors. Uncharged BHRs with their very small reaction
cross sections could be observed by searching for events with ∼1 TeV missing energy and
quenching of the high pThadron spectra. For further references on BHs, BHRs and their
observables please see [386].
We conclude that BHs at the LHC could provide a unique experimental window to the
understanding of quantum gravity. As many principles of BH production and decay are not
fully understood, a large variability of experimental observables is absolutely essential to pin
down the underlying physics.
9.3. Charmed exotics from heavy ion collision
S. H. Lee, S. Yasui, W. Liu and C. M. Ko
We discuss why charmed multiquark hadrons are likely to exist and explore the possibility of
observing such states in heavy ion reactions at the LHC.
Multiquark hadronic states are usually unstable as their quark configurations are
energetically above those of combined meson and/or baryon states. However, constituent
quark model calculations suggest that multiquark states might become stable when some of
the light quarks are replaced by heavy quarks. Two possible states that could be realistically
observed in heavy ion collisions at LHC are the tetraquark Tcc(ud¯c¯c) [387] and the pentaquark
Θcs(udus¯c) [388]. The driving mechanism for the stability of these states can be traced to
the quark color-spin interaction, which can be effectively parameterized as CHPi>j~si·~sj1
mimj.
Baryon mass splittings between states sensitive to the color-spin interaction are well explained
with a single constant coefficient CB
m2
u=193 MeV [389]. Similarly, corresponding meson mass
splittings are well reproduced with CM
m2
u=635 MeV [389]. Hence, the correlation energy
in a quark-antiquark pair is about a factor 3 larger than that in a quark-quark pair that is
in the color antitriplet channel. For heavy quarks, the size of the relative wave function
decreases substantially, and the parameter CHextracted from the mass difference between
J/ψ and ηcis Cc¯c
m2
c=117 MeV. As in the case of light quarks, we choose Ccc
m2
c=1
3Cc¯c
m2
c=39 MeV.
These numbers suggest that two quarks and two antiquarks would rather become two mesons
than form a single tetraquark state. However, when one or both of the antiquarks become
heavy, the attractions to form mesons are relatively suppressed compared to the strong diquark
correlation among light quarks, making multiquark states possibly stable. Using the constants
CHdiscussed above, we find that the mass of Tcc (Θcs) is -79 MeV below (8 MeV above) its
hadronic decay threshold. These results are well reproduced by full constituent quark model

Heavy Ion Collisions at the LHC - Last Call for Predictions 173
calculations. Although the binding becomes larger when the cquark is replaced by a bquark,
the expected number of bquarks produced in a heavy ion collision at the LHC is small for
a realistic observation of such states. Therefore, we only give predictions for the multiquark
states containing cquarks.
Employing the coalescence model [390], we have studied Tcc and Θcs production in
central Au+Au collisions at RHIC and Pb+Pb collisions at LHC. Using the u(or d) quark
numbers 245 and 662, the anti-strange quark numbers 150 and 405, and the charm quark
numbers 3 and 20 based on initial hard collisions at RHIC and LHC, respectively all in one
unit of midrapidity, we find that the numbers of Tcc produced at RHIC and LHC are about
5.4×10−6and 8.9×10−5, respectively, while those of Θcs are about 1.2×10−4and 8.3×10−4,
respectively. Since these numbers are significantly smaller than 7.5×10−4and 8.6×10−3for
Tcc, and 4.5×10−3and 2.7×10−2for Θcs from the statistical hadronization model for RHIC
and LHC, respectively, we expect additional production of these exotic charmed hadrons from
the hadronic stage of the collisions. We note that these charmed hadrons would be more
abundantly produced, particularly the Tcc, if charm quarks are produced from the QGP formed
in these collisions.
Table 9: Possible decay modes of Tcc. Additional (π+π−)’s are possible in the bracket.
threshold decay mode life time
MTcc >MD∗+MDD∗− ¯
D0hadronic decay
2MD+Mπ<MTcc <MD∗+MD¯
D0¯
D0π−hadronic decay
MTcc <2MD+MπD∗−(K+π−) 0.41 ×10−12 s
¯
D0(π−K+π−) weak decay
Table 10: Possible decay modes of Θcs.
threshold decay mode life time
MΘcs >MN+MDspD−
shadronic decay
MΛ+MD<MΘcs <MN+MDsΛ¯
D0hadronic decay
ΛD−hadronic decay
MΘcs <MΛ+MDΛK+π−,ΛK+π+π−π−0.41×10−12 s
ΛK+π−π−1.05×10−12 s
To observe Tcc and Θcs in experiments, we need to know their decay modes. While our
analysis suggests that Tcc is bound and Θcs is slightly unbound with respect to their hadronic
decays, we give predictions in tables 9 and 10 for all possible Tcc and Θcs masses. These exotic
hadrons can then be observed through reconstructed final states if they decay hadronically or
reconstructed final-state vertices if they decay weakly.

Heavy Ion Collisions at the LHC - Last Call for Predictions 174
9.4. Alignment as a result from QCD jet production or new still unknown physics at the
LHC?
I. P. Lokhtin, A. M. Managadze, L. I. Sarycheva and A. M. Snigirev
We would like to draw attention of the high-energy physics community to very important
experimental results indicating our lack of understanding of features of hadron interactions at super-
high energies and the necessity of improving recent theories.
The intriguing phenomenon of the strong collinearity of cores in emulsion experiments,
closely related to coplanar scattering of secondary particles in the interaction, has been
observed a long time ago. So far there is no simple satisfactory explanation of these cosmic
ray observations in spite of numerous attempts to find it (see, for instance, [391,392] and
references therein). Among them, the jet-like mechanism [393] looks very attractive and
gives a natural explanation of alignment of three spots along a straight line which results from
momentum conservation in a simple parton picture of scattering.
In the Pamir experiment [391] the families with the total energy of the γ-quanta larger
than a certain threshold and at least one hadron present were selected and analyzed. The
alignment becomes apparent considerably at PEγ>0.5 PeV (that corresponds to interaction
energies √s≥4 TeV). The families are produced, mostly,by a proton with energy ≥104TeV
interacting at a height hof several hundred meters to several kilometers in the atmosphere
above the chamber [391]. The collision products are observed within a radial distance rmax up
to several centimeters in the emulsion where the spot separation rmin is of the order of 1 mm.
Our analysis [394,395] shows that the jet-like mechanism can, in principle, attempt to
explain the results of emulsion experiments. For such an explanation it is necessary that
particles from both hard jets (with rapidities close to zero in the center-of-mass system) hit the
observation region due to the large Lorentz factor under the transformation from the center-
of-mass system to the laboratory one. This is possible when the combination of h,√sand
rmax meets the following condition:
2hmp/√s≤krmax,(66)
where mpis the proton mass. k∼1/2 is needed in order to have particles with adjoint positive
and negative rapidities in the center-of-mass system that hit the detection region. At the height
h=1000 m (mostly used in emulsion experiment estimations) and rmax =15 mm the condition
(66) is fulfilled at the energy √s≥270 TeV that is much higher than the LHC energies
√s≃5.5÷14 TeV and the threshold efficient interaction energies √seff≃4 TeV [391,392],
corresponding to the alignment phenomenon. Eq. (66) can be fulfilled and at the LHC energy
(14 TeV) also, but at the considerably lesser height h≤50 m which is in some contradiction
with emulsion experiment vague estimations.
On the other hand if particles from the central rapidity region and the jet-like mechanism
are insufficient to describe the observed alignment, and there is another still unknown
mechanism of its appearance at the energy √s∼5.5÷14 TeV and the accepted height
h∼1000 m, then in any case some sort of alignment should arise at the LHC too in the

Heavy Ion Collisions at the LHC - Last Call for Predictions 175
mid-forward rapidity region (following from the laboratory acceptance criterion for, e.g., pp
collisions) [394,395]:
rmin <ri=⇒ηi< ηmax =ln(r0/rmin)≃4.95,(67)
ri<rmax =⇒ηi> ηmin =ln(r0/rmax)≃2.25,(68)
where r0=2h/eηo,η0=9.55 is the rapidity of center-of-mass system in the laboratory
reference frame, ηiis the particle rapidity in the center-of-mass system, riis the radial particle
spacing in the x-ray film. Namely, at the LHC the strong azimuthal anisotropy of energy
flux (almost all main energy deposition along a radial direction) will be observed for all
events with the total energy deposition in the rapidity interval (67, 68) larger than some
threshold ∼1 TeV. Stress once more that at present there are no models or theories giving such
azimuthal anisotropy following from the experimentally observed alignment phenomenon at
√s≥√seff≃4TeV and h∼1000 m [391,392].
This mid-forward rapidity region must be investigated more carefully on the purpose to
study the azimuthal anisotropy of energy flux in accordance with the procedure applied in the
emulsion and other experiments, i.e. one should analyze the energy deposition in the cells
of η×φ-space in the rapidity interval (67, 68). Note that the absolute rapidity interval can
be shifted in correspondence with the variation of the height: it is necessary only that the
difference (ηmax −ηmin) is equal to ≃2.7 in accordance with the variation of radial distance by
a factor of ∼15 (rmax/rmin =15 independently of r0(h)) due to the relationship ri≃r0/eηi.
Such an investigation both in pp and in heavy ion collisions (to differentiate between
hadronic and nuclear interaction effects) at the LHC can clarify the origin of the alignment
phenomenon, give the new restrictions on the values of height and energy, and possibly
discover new still unknown physics.
Acknowledgements
The research of J. L. Albacete is sponsored in part by the U.S. Department of Energy under
Grant No. DE-FG02-05ER41377.
The work of D. Antonov has been supported by the Marie-Curie fellowship through the
contract MEIF-CT-2005-024196.
Part of the work of F. Arleo has been done in collaboration with T. Gousset [352] and
P. Aurenche, Z. Belghobsi, and J.-P. Guillet [353].
N. Armesto acknowledges financial support by MEC of Spain under a contract Ramón
y Cajal. J. G. Milhano acknowledges the financial support of the Fundação para a Ciência e
a Tecnologia of Portugal (contract SFRH/BPD/12112/2003). C. A. Salgado is supported by
the 6th Framework Programme of the European Community under the Marie Curie contract
MEIF-CT-2005-024624. They and L. Cunqueiro, J. Dias de Deus, E. G. Ferreiro and C.
Pajares acknowledge financial support by MEC under grant FPA2005-01963, and by Xunta
de Galicia (Consellería de Educación).
The work of G. G. Barnaföldi, P. Levai, B. A. Cole, G. Fai and G. Papp was supported
in part by Hungarian OTKA T047050, NK62044 and IN71374, by the U.S. Department of

Heavy Ion Collisions at the LHC - Last Call for Predictions 176
Energy under grant U.S. DE-FG02-86ER40251, and jointly by the U.S. and Hungary under
MTA-NSF-OTKA OISE-0435701. Special thanks to Prof. John J. Portman for computer time
at Kent State University.
The work of V. Topor Pop, J. Barrette, C. Gale and S. Jeon was partly supported by the
Natural Sciences and Engineering Research Council of Canada and by the U. S. DOE under
Contract No. DE-AC03-76SF00098 and DE-FG02-93ER-40764. M. Guylassy gratefully
acknowledges partial support also from FIAS and GSI, Germany.
D. Boer, A. Utermann and E. Wessels thank Adrian Dumitru and Jamal Jalilian-Marian
for helpful discussions.
W. Busza wishes to acknowledge Alex Mott, Yen-Jie Lee and Andre Yoon for help with
many of the plots, and Yetkin Yilmaz for Npart calculations.
The work of C. M. Ko was supported by the US National Science Foundation under
Grant PHY-0457265 and the Welch Foundation under Grant No. A-1358, that of B. Zhang by
NSF under Grant PHY-0554930, that of B.-A. Li by NSF under Grant PHY-0652548 and the
Research Corporation, that of B.-W. Zhang by the NNSF of China under Grant No. 10405011
and MOE of China under project IRT0624, that of L.-W. Chen by the SRF for ROCS, SEM
of China, and their joint work by the NNSF of China under Grants Nos. 10575071 and
10675082, MOE of China under project NCET-05-0392 and Shanghai Rising-Star Program
under Grant No. 06QA14024.
The work of G. Kestin and U. Heinz was supported by NSF grant PHY-0354916, and
theirs and the one of M. Djordjevic by U.S. DOE grant DE-FG02-01ER41190. The work of
E. Frodermann was supported by an Ohio State UniversityPresidential Fellowship.
D. d’Enterria and D. Peressounko acknowledge respectively support from 6th EU FP
contract MEIF-CT-2005-025073 and MPN Russian Federation grant NS-1885.2003.2.
K. J. Eskola, H. Niemi, P. V. Ruuskanen and S. S. Räsänen thank the Academy of Finland,
Projects 206024, 115262, and GRASPANP for financial support.
The work of R. Fries, S. Turbide, C. Gale and D. K. Srivastava was supported in parts by
DOE grants DE-FG02-87ER40328, DE-AC02-98CH10886, RIKEN/BNL, the Texas A&M
College of Science, and the Natural Sciences and Engineering Research Council of Canada.
G. Y. Qin, J. Ruppert, S. Turbide, C. Gale and S. Jeon thank the authors of [31] for
providing their hydrodynamical evolution calculation at RHIC and LHC energies, T. Renk
for discussions, and the Natural Sciences and Engineering Research Council of Canada for
support.
The work of H. van Hees and R. Rapp is supported by a U.S. NSF CAREER Award,
grant no. PHY-0449489.
The work of Z.-B. Kang and J.-W. Qiu is supported in part by the US Department of
Energy under Grant No. DE-FG02-87ER40371 and contract number DE-AC02-98CH10886.
The work of D. Kharzeev was supported by the U.S. Department of Energy under
Contract No. DE-AC02-98CH10886. The work of E. Levin was supported in part by the
grant of Israeli Science Foundation founded by Israeli Academy of Science and Humanity.
The work of H. Stöcker and B. Koch was supported by GSI and BMBF.
The work of A. H. Rezaeian, B. Z. Kopeliovich, H. J. Pirner, I. K. Potashnikova and I.

Heavy Ion Collisions at the LHC - Last Call for Predictions 177
Schmidt was supported in part by Fondecyt (Chile) grants 1070517and 1050519, and by DFG
(Germany) grant PI182/3-1.
The work of I. Kuznetsova, J. Letessier and J. Rafelski has been supported by a grant
from the U.S. Department of Energy DE-FG02-04ER4131. LPTHE, Univ.Paris 6 et 7 is:
Unité mixte de Recherche du CNRS, UMR7589.
The work of M. Mannarelli and C. Manuel has been supported by the Ministerio de
Educación y Ciencia (MEC) under grant AYA 2005-08013-C03-02.
D. Molnár thanks RIKEN, Brookhaven National Laboratory and the US Department of
Energy [DE-AC02-98CH10886] for providing facilities essential for the completion of his
work.
G. Torrieri thanks the Alexander Von Humboldt foundation, the Frankfurt Institute for
Theoretical Physics and FIAS for continued support, and CERN theory division for providing
local support necessary for attending the workshop where this work is presented. He would
also like to thank Sangyong Jeon, Marek Gazdzicki, Mike Hauer, Johann Rafelski and Mark
Gorenstein for useful and productive discussions.
K. Tuchin is grateful to Javier Albacete for showing me the results of his calculations
of open heavy quark production; his results are in a qualitative agreement with Fig. 43 and
Fig. 44. He would like to thank RIKEN, BNL and the U.S. Department of Energy (Contract
No. DE-AC02-98CH10886) for providing the facilities essential for the completion of this
work.
The work of R. Vogt was performed under the auspices of the U.S. Department of Energy
by University of California, Lawrence Livermore National Laboratory under Contract W-
7405-Eng-48 and supported in part by the National Science Foundation Grant NSF PHY-
0555660.
The work of E. Wang, X. N. Wang and H. Zhang was supported by DOE under contracts
No. DE-AC02-05CH11231, by NSFC under Project No. 10440420018, No. 10475031 and
No. 10635020, and by MOE of China under projects No. NCET-04-0744, No. SRFDP-
20040511005 and No. IRT0624.
References
[1] Bass S A et al. 1999 Nucl. Phys. A661 205
[2] Dias de Deus J and Milhano J G 2007 Nucl. Phys. A795 98
[3] Brogueira P, Dias de Deus J and Pajares C 2007 Phys. Rev. C75 054908
[4] Back B B et al. [PHOBOS Collaboration] 2005 Nucl. Phys. A757 28
[5] Kharzeev D, Levin E and Nardi M 2005 Nucl. Phys. A747 609
[6] Albacete J L 2007 Preprint arXiv:0707.2545 [hep-ph]
[7] Kharzeev D and Levin E 2001 Phys. Lett. B523 79
[8] Albacete J L and KovchegovY V 2007 Phys. Rev. D75 125021
[9] McLerran L D and Venugopalan R 1994 Phys. Rev. D49 2233
[10] d’Enterria D and Peressounko D 2006 Eur. Phys. J. C46 451, and references therein
[11] Adler S S et al. [PHENIX Collaboration] 2005 Phys. Rev. C71 034908 [Erratum-ibid. 049901]
[12] Aurenche P et al 2000 Eur. Phys. J. C13 347
[13] Pumplin J et al. 2002 JHEP 0207 012

Heavy Ion Collisions at the LHC - Last Call for Predictions 178
[14] Albino S, Kniehl B A and Kramer G 2005 Nucl. Phys. B725 181; 2006
ibid.
B734 50
[15] de Florian D and Sassot R 2004 Phys. Rev. D69 074028
[16] Arleo F 2007 JHEP 0707 032
[17] Stasto A M, Golec-Biernat K J and Kwiecinski J 2001 Phys. Rev. Lett. 86 596
[18] Armesto N, Salgado C A and Wiedemann U A 2005 Phys. Rev. Lett. 94 022002
[19] Albacete J L et al. 2005 Eur. Phys. J. C43 353
[20] Back B B et al. [PHOBOS Collaboration] 2006 Phys. Rev. C74 021901
[21] Gribov V N 1969 Sov. Phys. JETP 29 483
[22] Tywoniuk K et al. 2007 Preprint arXiv:0705.1596 [hep-ph]
[23] Aktas A et al. [H1 Collaboration] 2006 Eur. Phys. J. C48 715; ibid. 48 749
[24] Topor Pop V et al. 2005 Phys. Rev. C72 054901
[25] Kharzeev D 1996 Phys. Lett. B378 238
[26] Vance S E, Gyulassy M and Wang X N 1998 Phys. Lett. B443 45
[27] Topor Pop V, Barrette J, Gale C, Jeon S and Gyulassy M, these proceedings
[28] Armesto N and Pajares C 2000 Int. J. Mod. Phys. A15 2019
[29] Wiedemann U A 2007 J. Phys. G: Nucl. Part. Phys. 34 S503
[30] Videbaek F 2006 J. Phys. Conf. Ser. 50 134
[31] Eskola K J et al. 2005 Phys. Rev. C72 044904
[32] Bearden I G et al. [BRAHMS Collaboration] 2004 Phys. Rev. Lett. 93 102301
[33] Golec-Biernat K and WusthoffM 1999 Phys. Rev. D59 014017
[34] Dumitru A, Hayashigaki A and Jalilian-Marian J 2006 Nucl. Phys. A770 57
[35] Boer D, Dumitru A and Hayashigaki A 2006 Phys. Rev. D74 074018
[36] Boer D, Utermann A and Wessels E 2007 Phys. Rev. D75 094022
[37] Boer D and Dumitru A 2003 Phys. Lett. B556 33
[38] Engel R 1995 Z. Phys. C66 203
[39] Engel R and Ranft J 1996 Phys. Rev. D54 4244
[40] Roesler S, Engel R and Ranft J 2001 Advanced Monte Carlo 2000 (Springer) p. 1033
[41] Roesler S, Engel R and Ranft J 2003 Nucl. Phys. B (Proc. Suppl.) 122 392
[42] Benecke J, Chou T T, Yang C N and Yen E 1969 Phys. Rev. 188 2159
[43] Earlier discussions of the trends seen and references can be found in Busza W et al. [E178 Collaboration]
1975 Phys. Rev. Lett. 34 836; Elias J et al. [E178 Collaboration] 1980 Phys. Rev. D22 13; Barton D et
al. [E451 Collaboration] 1983 Phys. Rev. D27 2580; Busza W 1977 Acta. Phys. Pol. B8333
[44] Busza W 2004 Acta Phys. Polon. B 35 2873
[45] Roland G [PHOBOS Collaboration], talk at QM2005; Nouicer R [PHOBOS Collaboration], talk at
PANIC 2005
[46] Otterlund I et al. 1978 Nucl. Phys. B142 445; Abduzhamilov A 1987 Phys. Rev. D35 3536
[47] Back B B et al. [PHOBOS Collaboration] 2005 Phys. Rev. Lett. 94 122302; Roland G, talk at PANIC
2005
[48] Alt C et al. 2003 Phys. Rev. C68 (2003) 034903
[49] Capella A and Ferreiro E G 2005 Eur. Phys. J. C42 419
[50] Capella A, Ferreiro E G and Kaidalov A B 2000 Phys. Rev. Lett. 85 2080
[51] Capella A and Ferreiro E G 2006 Preprint arXiv:hep-ph/0610313
[52] Lin Z W et al. 2005 Phys. Rev. C72 064901
[53] Wang X N and Gyulassy M 1991 Phys. Rev. D44 3501
[54] Zhang B 1998 Comput. Phys. Commun. 109 193
[55] Li B A and Ko C M 1995 Phys. Rev. C52 2037
[56] Zhang B, Ko C M, Li B A and Lin Z W 2000 Phys. Rev. C61 067901
[57] Lin Z W and Ko C M 2002 Phys. Rev. C65 034904
[58] Ko C M, Chen L W and Zhang B W 2007 Braz. J. Phys. 37 969
[59] Lin Z W et al. Phys. Rev. C64 011902
[60] Zhang B, Chen L W and Ko C M 2005 Phys. Rev. C72 024906

Heavy Ion Collisions at the LHC - Last Call for Predictions 179
[61] Lin Z W and Molnár D 2003 Phys. Rev. C68 044901
[62] Greco V, Ko C M and Rapp R 2004 Phys. Lett. B595 202
[63] Pratt S et al. 1994 Nucl. Phys. A566 103c
[64] Lin Z-W, Ko C M and Pal S 2002 Phys. Rev. Lett. 89 152301
[65] Armesto N, Braun M A, Ferreiro E G and Pajares C 1996 Phys. Rev. Lett. 77 3736
[66] Dias de Deus J and Ugoccioni R 2000 Phys. Lett. B494 53
[67] Pajares C 2005 Eur. Phys. J. C43 9
[68] van Eijndhoven N et al. 1995 Preprint ALICE/GEN 95-32
[69] Xu Z and Greiner C 2005 Phys. Rev. C71 064901
[70] Xu Z and Greiner C 2007 Preprint arXiv:hep-ph/0703233
[71] Xu Z and Greiner C 2006 Nucl. Phys. A774 787
[72] Xu Z and Greiner C, work in progress
[73] El A, Greiner C and Xu Z 2007 Nucl. Phys. A785 132; manuscript in preparation
[74] Eskola K J, Kajantie K, Ruuskanen P V and Tuominen K 2000 Nucl. Phys. B570 379
[75] Eskola K J, Ruuskanen P V, Räsänen S S and Tuominen K 2001 Nucl. Phys. A696 715
[76] Eskola K J, Niemi H, Ruuskanen P V and Räsänen S S 2003 Phys. Lett. B566 187
[77] Eskola K J, Kajantie K and Lindfors J 1989 Nucl. Phys. B323 37
[78] Gribov L V, Levin L M and Ryskin M G 1983 Phys. Rept. 100 1
[79] Eskola K J and Honkanen H 2003 Nucl. Phys. A713 167
[80] Eskola K J, Honkanen H, Salgado C A and Wiedemann U A 2005 Nucl. Phys. A747 511
[81] Gelis F and Venugopalan R 2006 Acta Phys. Polon. B37 3253
[82] Gelis F, Stasto A M and Venugopalan R 2006 Eur. Phys. J. C48 489
[83] Gelis F and Venugopalan R 2006 Nucl. Phys. A779 177
[84] Fujii H, Gelis F and Venugopalan R 2005 Phys. Rev. Lett. 95 162002
[85] Iancu E, Itakura K and Triantafyllopoulos D N 2004 Nucl. Phys. A742 182
[86] Mueller A H and Triantafyllopoulos D N 2002 Nucl. Phys. B640 331
[87] Mueller A H 2003 Nucl. Phys. A724 223
[88] Iancu E, Itakura K and McLerran L 2002 Nucl. Phys. A708 327
[89] Dumitru A et al. 2007 Preprint arXiv:0706.2540
[90] Back B B et al. [PHOBOS Collaboration] 2003 Phys. Rev. Lett. 91 052303
[91] Jeon S, Topor Pop V and Bleicher M 2004 Phys. Rev. C69 044904
[92] Kharzeev D and Nardi M 2001 Phys. Lett. B507 121
[93] Kharzeev D, Levin E and Nardi M 2005 Phys. Rev. C71 054903
[94] Kharzeev D, Levin E and Nardi M 2004 Nucl. Phys. A730 448
[95] Kharzeev D and Levin E, to appear
[96] Kopeliovich B Z, Povh B and Schmidt I 2007 Nucl. Phys. A782 24
[97] Kopeliovich B Z and Povh B 2004 J. Phys. G: Nucl. Part. Phys. 30 S999
[98] Kopeliovich B Z, Nemchik J, Schäfer A andTarasov A V 2002 Phys. Rev. Lett. 88 232303
[99] Wang X N 2000 Phys. Rev. C61 064910
[100] Kharzeev D, Levin E and McLerran L 2003 Phys. Lett. B561 93
[101] Kopeliovich B Z, Raufeisen J and Tarasov A V 2000 Phys. Rev. C62 035204
[102] Kozlov M, Shoshi A I and Xiao B W 2007 Nucl. Phys. A792 170
[103] Iancu E, Mueller A H and Munier S 2005 Phys. Lett. B606 342
[104] Mueller A H and Shoshi A I 2004 Nucl. Phys. B692 175
[105] Drescher H J et al. 2001 Phys. Rept. 350 93
[106] Werner K, Liu F M and Pierog T 2006 Phys. Rev. C74 044902
[107] Werner K 2007 Phys. Rev. Lett. 98 152301
[108] Kharzeev D, KovchegovY V and Tuchin K 2003 Phys. Rev. D68 094013
[109] Kharzeev D, KovchegovY V and Tuchin K 2004 Phys. Lett. B599 23
[110] Kharzeev D and Tuchin K 2004 Nucl. Phys. A735 248
[111] Tuchin K 2004 Phys. Lett. B593 66

Heavy Ion Collisions at the LHC - Last Call for Predictions 180
[112] Wolschin G 2007 Nucl. Phys. A787 68
[113] Kuiper R and Wolschin G 2007 Annalen Phys. 16 67
[114] Peshier A et al. 1994 Phys. Lett. B337 235
[115] Peshier A et al. 1996 Phys. Rev. D54 2399
[116] Peshier A et al. 2000 Phys. Rev. C61 045203
[117] Peshier A et al. 2002 Phys. Rev. D66 094003
[118] Bluhm M et al. 2007 Eur. J. Phys. C49 205
[119] Karsch F et al. 2003 Eur. J. Phys. C29 549
[120] Karsch F et al. 2003 Phys. Lett. B571 67
[121] Bluhm M et al. 2007 Phys. Rev. C76 034901
[122] Kolb P F et al. 1999 Phys. Lett. B459 667
[123] Kolb P F et al. 2000 Phys. Rev. C62 054909
[124] Drescher H J, Dumitru A, Gombeaud C and Ollitrault J-Y 2007 Phys. Rev. C76 024905
[125] Drescher H J and Nara Y 2007 Phys. Rev. C75 034905
[126] Back B B et al. [PHOBOS Collaboration] 2005 Phys. Rev. C72 051901
[127] Kolb P F et al. 2001 Nucl. Phys. A696 197
[128] Adams J et al. [STAR Collaboration] 2005 Phys. Rev. C72 014904
[129] Adler S S et al. [PHENIX Collaboration] 2003 Phys. Rev. Lett. 91 182301
[130] http://nt3.phys.columbia.edu/people/molnard/OSCAR/and references listed there
[131] Kestin G and Heinz U 2007, to be published
[132] Hirano T et al. 2006 Phys. Lett. B636 299
[133] Hirano T et al. 2007 J. Phys. G: Nucl. Part. Phys. 34 S879
[134] Kolb P F and Heinz U 2003 Preprint arXiv:nucl-th/0305084
[135] Meyer H B 2007 Preprint arXiv:0704.1801 [hep-lat], and to be published
[136] Gyulassy M and Danielewicz P 1985 Phys. Rev. D31 53
[137] Kovtun P, Son D T and Starinets A O 2005 Phys. Rev. Lett. 94 111601
[138] Israel W and Stuart J M 1979 Ann. Phys. 118 349
[139] Molnar D and Gyulassy M 2002 Nucl. Phys. A697 495
[140] de Groot S R, van Leeuwen W A and vanWeert Ch G 1980 Relativistic Kinetic Theory (North Holland)
[141] Molnár D and Huovinen P 2005 Phys. Rev. Lett. 94 012302
[142] Zhang B, Gyulassy M and Ko C M 1999 Phys. Lett. B455 45
[143] Adil A, Gyulassy M and Hirano T 2006 Phys. Rev. D73 074006
[144] Andronic A, Braun-Munzinger P and Stachel J 2006 Nucl. Phys. A772 167
[145] Adams J et al. [STAR Collaboration] 2006 Phys. Rev. Lett. 97 132301
[146] Topor Pop V et al. 2007 Phys. Rev. C75 014904
[147] Fries R J, Muller B, Nonaka C and Bass S A 2003 Phys. Rev. C68 044902
[148] Rafelski J and Muller B 1982 Phys. Rev. Lett. 48 1066 [Erratum-ibid. 1986 56 2334]
[149] SoffSet al. 2004 J. Phys. G: Nucl. Part. Phys. 30 L35
[150] Castorina P, Kharzeev D and Satz S 2007 Eur. Phys. J. C52, 187
[151] Muller B 2007 Nucl. Phys. A783 403
[152] Acosta D et al. [CDF Collaboration] 2005 Phys. Rev. D72 052001
[153] Ma Y G 2006 J. Phys. G32 S373
[154] Veneziano G 1976 Nucl. Phys. B117 519; Rossi G C and Veneziano G 1977 Nucl. Phys. B123 507; 1980
Phys. Rept. 63 153
[155] Bartels J, Lipatov L N and Vacca G P 2000 Phys. Lett. B477 178
[156] Bopp F and Shabelski Y M 2006 Eur. Phys. J. A28 237
[157] Roesler S, Engel R and Ranft J 2000 Preprint hep-ph/0012252, Proc. of Monte Carlo 2000, Lisboa,
October 2000, Springer, p.1033
[158] Bopp F W, Ranft J, Engel R and Roesler S 2005 Preprint arXiv:hep-ph/0505035v2
[159] Cleymans J, Oeschler H, Redlich K and Wheaton S 2006 Phys. Rev. C73 034905
[160] Cleymans J et al. 2006 Phys. Rev. C74 034903

Heavy Ion Collisions at the LHC - Last Call for Predictions 181
[161] Kraus I et al. 2006 J. Phys. G: Nucl. Part. Phys. 32 S495
[162] Kraus I, Oeschler H and Redlich K 2005 PoS (HEP2005) 140
[163] Kraus I et al. 2007 Preprint arXiv:0707.3879 [hep-ph]
[164] Braun M A and Pajares C 2000 Phys. Rev. Lett. 85 4864
[165] Dias de Deus J, Ferreiro E G, Pajares C and Ugoccioni R 2005 Eur. Phys. J. C40 229
[166] Ferreiro E G, del Moral F and Pajares C 2004 Phys. Rev. C69 034901
[167] Cunqueiro L, Ferreiro E G, del Moral F and Pajares C 2005 Phys. Rev. C72 024907
[168] Rafelski J and Letessier J 2006 Eur. Phys. J. C45 61
[169] Letessier J and Rafelski J 2007 Phys. Rev. C75 014905
[170] Torrieri G, Jeon S, Letessier J and Rafelski J 2006 Comput. Phys. Commun. 175 635
[171] Torrieri G et al. 2005 Comput. Phys. Commun. 167 229
[172] Rafelski J and Letessier J 2000 Phys. Rev. Lett. 85 4695
[173] Sinyukov Yu M, Akkelin S V and Hama Y 2002 Phys. Rev. Lett. 89 052301
[174] Akkelin S V, Hama Y, Karpenko Iu A and SinyukovYu M, in preparation
[175] Hirano T 2002 Phys. Rev. C65 011901
[176] Hirano T and Tsuda K 2002 Phys. Rev. C66 054905
[177] Hirano T, private communication
[178] Laine M and Schroder Y 2006 Phys. Rev. D73 085009
[179] Lappi T 2006 Phys. Lett. B643 11
[180] Ulery J G [STAR Collaboration] 2007 Preprint arXiv:0704.0224 [nucl-ex]
[181] Satarov L M, Stoecker H and Mishustin I N 2005 Phys. Lett. B627 64
[182] Baeuchle B, Csernai L P and Stoecker H 2007 Preprint arXiv:0710.1476 [nucl-th]
[183] Clare R B and Strottman D 1986 Phys. Rept. 141 177
[184] Gyulassy M, Levai P and Vitev I 2001 Nucl. Phys. B594 371
[185] Stöcker H 2005 Nucl. Phys. A750 121; Antinori F and Shuryak E V 2005 J. Phys. G31 L19; Adams J et
al. [STAR Collaboration] 2003 Phys. Rev. Lett. 91 072304
[186] Adler S S et al. [PHENIX Collaboration] 2006 Phys. Rev. C73 054903; Grau N [PHENIX Collaboration]
2006 Preprint arXiv:nucl-ex/0612019; Ajitanand N N [PHENIX Collaboration] 2007 Nucl. Phys. A
783 519
[187] Adams J et al. [STAR Collaboration] 2005 Phys. Rev. Lett. 95 152301
[188] Rischke D H et al. 1995 Heavy Ion Phys. 1309
[189] Dias de Deus J, Pajares C and Salgado C A 1997 Phys. Lett. B407 335
[190] Braun M A, Pajares C and Vechernin V V 2000 Phys. Lett. B493 54
[191] Brogueira P and Dias de Deus J 2007 Phys. Lett. B653 202
[192] Brogueira P, Dias de Deus J and Milhano J G 2007 Preprint arXiv:0709.3913 [hep-ph]
[193] Armesto N, Braun M A and Pajares C 2007 Phys. Rev. C75 054902
[194] Dremin I M 1979 JETP Lett. 30 140
[195] Dremin I M 1981 Sov. J. Nucl. Phys. 33 726
[196] Dremin I M 2006 Nucl. Phys. A 767 233
[197] Dremin I M 2007 Nucl. Phys. A 785 365
[198] Koch V, Majumder A and Wang X N 2006 Phys. Rev. Lett. 96 172302
[199] Apanasenko A V, Dobrotin N A, Dremin I M and Kotelnikov K A 1979 JETP Lett. 30 145
[200] Lisa M A, Pratt S, Soltz R and Wiedemann U 2005 Ann. Rev. Nucl. Part. Sci. 55 357
[201] Frodermann E, Chatterjee R and Heinz U 2007 J. Phys. G: Nucl. Part. Phys. 34 2249
[202] Frodermann E, Heinz U and Lisa M A 2006 Phys. Rev. C73 044908
[203] Retiere F and Lisa M A 2004 Phys. Rev. C70 044907
[204] Amelin N S et al. 2006 Phys. Rev. C 74 064901
[205] Adams J et al. [STAR Collaboration] 2005 Phys. Rev. C 71 044906
[206] Casalderrey-Solana J, Shuryak E V and Teaney D 2005 J. Phys. Conf. Ser. 27 22
[207] Casalderrey-Solana J, Shuryak E V and Teaney D 2006 Preprint hep-ph/0602183
[208] Adler S S et al. (PHENIX Collaboration) 2006 Phys. Rev. Lett. 97 052301

Heavy Ion Collisions at the LHC - Last Call for Predictions 182
[209] Mannarelli M and Manuel C 2007 Preprint arXiv:0705.1047 [hep-ph]
[210] Pavlenko O P 1992 Sov. J. Nucl. Phys. 55 1243
[211] Manuel C and Mrowczynski S 2006 Phys. Rev. D74 105003
[212] Torrieri G, Jeon S and Rafelski J 2006 Phys. Rev. C74 024901
[213] Jeon S and Koch V1999 Phys. Rev. Lett. 83 5435
[214] Pruneau C, Gavin S and Voloshin S 2002 Phys. Rev. C66 044904
[215] Begun V V, Gazdzicki M, Gorenstein M I and Zozulya O S 2004 Phys. Rev. C70 034901
[216] Liu H, Rajagopal K and Wiedemann U A 2007 JHEP 0703 066
[217] For a review see: Di Giacomo A et al. 2002 Phys. Rept. 372 319
[218] Baier R et al. 1997 Nucl. Phys. B483 291
[219] Agasian N O 2003 Phys. Lett. B562 257
[220] Shoshi A I et al. 2003 Phys. Rev. D68 074004
[221] Barnaföldi G G et al. 2007 Preprint arXiv:hep-ph/0703059
[222] STAR preliminary data, presented by Ming Y at SQM’07 conference
[223] Abelev B I et al. [STAR Collaboration] 2006 Phys. Rev. Lett. 97 152301
[224] Lévai P, Papp G, Fai G and Gyulassy M 2006 Acta. Phys. Hung. A27 459
[225] Capella A, Ferreiro E G, Kaidalov A B and Sousa D 2005 Eur. Phys. J. C40 129
[226] Capella A and Ferreiro E G 2007 Preprint arXiv:0706.1880 [hep-ph]
[227] Capella A and Ferreiro E G 2007 Phys. Rev. C75 024905
[228] Baier R et al. 1997 Nucl. Phys. B484 265
[229] Gribov L V, Levin E M and Ryskin M G 1981 Nucl. Phys. B188 555
[230] Casalderrey-Solana J and Wang X N 2007 Preprint arXiv:0705.1352 [hep-ph]
[231] Dainese A, Loizides C and Pai´
c G 2005 Eur. Phys. J. C38 461
[232] Salgado C A and Wiedemann U A 2003 Phys. Rev. D68 014008
[233] Adler S S et al. [PHENIX Collaboration] 2007 Phys. Rev. C76 034904
[234] Brambilla N et al. 2004 Preprint hep-ph/0412158.
[235] Guylassy M, Vitev I, Wang X N and Zhang B W, in Quark Gluon Plasma 3, edited by R. C. Hwa and X.
N. Wang, p. 123 (World Scientific, Singapore, 2003)
[236] Djordjevic M 2006 Phys. Rev. C74 064907
[237] Djordjevic M and Heinz U 2007 Preprint arXiv:0705.3439 [nucl-th]
[238] Renk R and Eskola K J 2007 Phys. Rev. C75 054910
[239] Arnold P, Moore G D and Yaffe L G 2001 JHEP 0111 057
[240] Arnold P, Moore G D and Yaffe L G 2001 JHEP 0112 009
[241] Arnold P, Moore G D and Yaffe L G 2001 JHEP 0206 030
[242] Qin G Y et al. 2007 Preprint arXiv:0705.2575 [hep-ph]
[243] Mustafa M G 2005 Phys. Rev. C72 014905
[244] Wicks S, Horowitz W, DjordjevicM and Gyulassy M 2007 Nucl. Phys. A784 426
[245] Bassetto A, Ciafaloni M and Marchesini G 1980 Nucl. Phys. B163 477
[246] Braun J and Pirner H J 2007 Phys. Rev. D75 054031
[247] Ellis R K, Stirling W J and Webber B R 1996 QCD and Collider Physics (Cambridge University Press)
[248] Borghini N and Wiedemann U A 2006 Nucl. Phys. A774 549
[249] Jeon S, Jalilian-Marian J and Sarcevic I 2003 Nucl. Phys. A715 795c
[250] Jeon S, Jalilian-Marian J and Sarcevic I 2003 Phys. Lett. B562 45
[251] Jeon S, Jalilian-Marian J and Sarcevic I 2003 Nucl. Phys. A723 467
[252] Wang X N, Huang Z and Sarcevic I 1996 Phys. Rev. Lett. 77 231
[253] Frantz J 2007 J. Phys. G: Nucl. Part. Phys. 34 S389
[254] Luo M, Qiu J-W and Sterman G 1994 Phys. Rev. D49 4493; D 50 1951
[255] X F Guo 1998 Phys. Rev. D58 114033
[256] Kang Z B and Qiu J-W, in preparation
[257] Peng J C 1999 AIP Conf. Proc. 494 503
[258] Liu W, Ko C M and Zhang B W 2007 Phys. Rev. C75 051901

Heavy Ion Collisions at the LHC - Last Call for Predictions 183
[259] Kaczmarek O, Karsch F, Zantow F and Petreczky P 2004 Phys. Rev. D70 074505 [Erratum-ibid 2005 D
72 059903]
[260] Kopeliovich B Z et al. 2004 Nucl. Phys. A740 211
[261] Berger E L 1979 Phys. Lett. B 89 241
[262] Kopeliovich B Z et al. 2007 Preprint arXiv:0706.3059
[263] Liu H, Rajagopal K and Wiedemann U A 2006 Phys. Rev. Lett. 97 182301
[264] Liu H, Rajagopal K and Wiedemann U A 2007 Phys. Rev. Lett. 98 182301
[265] Lokhtin I P and Snigirev A M 2006 Eur. Phys. J. C 45 211
[266] d’Enterria D et al. 2007 J. Phys. G: Nucl. Part. Phys. 34 2307
[267] Lokhtin I P and Snigirev A M 2003 Phys. Lett. B 567 39
[268] Lokhtin I P, Petrushanko S V, Sarycheva L I and Snigirev A M 2006 Phys. Rev. C 73 064905
[269] Lokhtin I P, Sarycheva L I and Snigirev A M 1999 Phys. Atom. Nucl. 62 1258
[270] Lokhtin I P, Petrushanko S V, Sarycheva L I and Snigirev A M 2003 Pramana 60 1045
[271] Lokhtin I P, Petrushanko S V, Sarycheva L I and Snigirev A M 2002 Phys. At. Nucl. 65 943
[272] Lokhtin I P, Sherstnev A V and Snigirev A M 2004 Phys. Lett. B 599 260
[273] Lokhtin I P and Snigirev A M 2001 J. Phys. G: Nucl. Part. Phys. 27 2365
[274] Lokhtin I P and Snigirev A M 2002 Nucl. Phys. A 702 346
[275] Lokhtin I P, Sarycheva L I, Snigirev A M and Teplov K Yu 2004 Eur. Phys. J. C 37 465
[276] Pantuev V S 2007 JETP Lett. 85 104
[277] Pantuev V S 2006 Preprint hep-ph/0604268
[278] Sapeta S and Wiedemann U A 2007 Preprint arXiv:0707.3494 [hep-ph]
[279] Dokshitzer Y L , Khoze V A and Troian S I 1988 Adv. Ser. Direct. High Energy Phys. 5241
[280] Azimov Y I, Dokshitzer Y L, Khoze V A and Troian S I 1985 Z. Phys. C27 65
[281] Azimov Y I, Dokshitzer Y L, Khoze V A and Troian S I 1986 Z. Phys. C31 213
[282] Borghini N and Wiedemann U A 2005 Preprint arXiv:hep-ph/0506218
[283] Vitev I and Gyulassy M 2002 Phys. Rev. Lett. 89 252301
[284] Vitev I 2003 Phys. Lett. B562 36
[285] Vitev I 2007 Phys. Rev. C75 064906
[286] Owens J F 2002 Phys. Rev. D65 034011
[287] Wang E and Wang X N 2001 Phys. Rev. Lett. 87 142301
[288] Wang E and Wang X N 2002 Phys. Rev. Lett. 89 162301
[289] Zhang H, Owens J F, Wang E and Wang X N 2007 Phys. Rev. Lett. 98 212301
[290] Andronic A, Braun-Munzinger P, Redlich K and Stachel J 2007 Nucl. Phys. A789 334
[291] Pauletta G [CDF and D0 Collaborations] 2005 J. Phys. G: Nucl. Part. Phys. 31 S817
[292] Armesto N, Salgado C A and Wiedemann U A 2004 Phys. Rev. D69 114003
[293] Armesto N et al. 2006 Phys. Lett. B637 362
[294] Armesto N, Dainese A, Salgado C A and Wiedemann U A 2005 Phys. Rev. D71 054027
[295] Boreskov K, Capella A, Kaidalov A and Tran Thanh Van J 1993 Phys. Rev. D47 919
[296] Adler S S et al. [PHENIX Collaboration] 2003 Phys. Rev. Lett. 96 012304
[297] Mangano M, Nason P and Ridolfi G 1992 Nucl. Phys. B373 295
[298] Conesa del Valle Z, Martínez G, Aphecetche L and Finck C 2006 Preprint ALICE-INT-2006-021
[299] Ding H-T et al. 2007 Preprint hep-ph/0702118
[300] Gonçalves V P and Machado M V T 2006 J. Phys. G: Nucl. Part. Phys. 32 295
[301] Gonçalves V P and Machado M V T 2005 Eur. Phys. J. C40 519
[302] Gonçalves V P and Machado M V T 2006 Phys. Rev. C73 044902
[303] Jalilian-Marian J and Kovchegov Yu V 2006 Prog. Part. Nucl. Phys. 56 104
[304] Iancu E, Itakura K and Munier S 2004 Phys. Lett. B590 199
[305] van Hees H and Rapp R 2005 Phys. Rev. C71 034907
[306] van Hees H, Greco V and Rapp R 2006 Phys. Rev. C73 034913
[307] van Hees H and Rapp R 2006 J. Phys. G: Nucl. Part. Phys. G32 S351
[308] Casalderrey-Solana J and Teaney D 2006 Phys. Rev. D74 085012

Heavy Ion Collisions at the LHC - Last Call for Predictions 184
[309] Herzog C P et al. 2006 JHEP 0607 013; Gubser S S 2006 Phys. Rev. D74 126005
[310] Gubser S S 2006 Preprint arXiv:hep-th/0611272
[311] Horowitz W A and Gyulassy M 2007 Preprint arXiv:0706.2336 [nucl-th]
[312] Gubser S S 2006 Preprint arXiv:hep-th/0612143
[313] Zhang B W, Ko C M and Liu W 2007 Preprint arXiv:0709.1684 [nucl-th]
[314] Beenakker W, Kuijf H, van Neerven W L and Smith J 1989 Phys. Rev. D40 54
[315] Beenakker W et al. 1991 Nucl. Phys. B351 507
[316] Nason P, Dawson S and Ellis R K 1988 Nucl. Phys. B303 607
[317] Nason P, Dawson S and Ellis R K 1989 Nucl. Phys. B327 49
[318] Kopeliovich B Z, Schäfer A and Tarasov A V 2000 Phys. Rev. D62 054022
[319] Kopeliovich B Z and Tarasov A V 2002 Nucl. Phys. A710 180
[320] Kopeliovich B Z, Tarasov A V and Hüfner J 2001 Nucl. Phys. A696 669
[321] Kopeliovich B Z et al. 2005 Phys. Rev. C72 054606
[322] Kuznetsova I and Rafelski J 2007 Eur. Phys. J. C51 113
[323] Becattini F 2005 Phys. Rev. Lett. 95 022301
[324] Thews R L, Schroedter M and Rafelski J 2001 Phys. Rev. C63 054905
[325] Kuznetsova I and Rafelski J 2006 J. Phys. G: Nucl. Part. Phys. 32 S499
[326] Schroedter M, Thews R L and Rafelski J 2000 Phys. Rev. C62 024905
[327] Thews R L andMangano M L 2006 Phys. Rev. C73 014904
[328] Mócsy Á and P. Petreczky 2007 Preprint arXiv:0705.2559 [hep-ph]
[329] Mócsy Á and P. Petreczky 2007 Preprint arXiv:0706.2183 [hep-ph]
[330] Petrov K [RBC-Bielefeld Collaboration] 2006 Preprint hep-lat/0610041
[331] Kharzeev D, McLerran L D and Satz H 1995 Phys. Lett. B356 349
[332] Vitev I, Goldman J T, Johnson M B and Qiu J W 2006 Phys. Rev. D74 054010
[333] Gyulassy M, Levai P and Vitev I 2002 Phys. Rev. D66 014005
[334] Adil A and Vitev I 2007 Phys. Lett. B649 139
[335] Vitev I, Adil A and van Hees H 2007 J. Phys. G: Nucl. Part. Phys. 34 S769
[336] Leitch M, Lourenco C and Vogt R, presented at Quark Matter’06, Shanghai, China
[337] Braun M A et al. 1998 Nucl. Phys. B509 357
[338] Vogt R 2005 Phys. Rev. C71 054902
[339] Eskola K J, Kolhinen V J and Ruuskanen P V 1998 Nucl. Phys. B535 351
[340] Eskola K J, Kolhinen V J and Salgado C A 1999 Eur. Phys. J. C961
[341] Klein S R and Vogt R 2003 Phys. Rev. Lett. 91 142301
[342] Adare A et al. [PHENIX Collaboration] 2007 Phys. Rev. Lett. 98 232301
[343] Gunion J F and Vogt R 1997 Nucl. Phys. B492 301
[344] Digal S, Petreczky P and Satz H 2001 Phys. Rev. D64 094015
[345] Satz H 2006 J. Phys. G: Nucl. Part. Phys. 32 R25
[346] Vitev I 2004 J. Phys. G: Nucl. Part. Phys. 30 S791
[347] Karsch F and Petronzio R 1988 Z. Phys. C37 627
[348] Gavai R et al. 1995 Int. J. Mod. Phys. A10 3043
[349] Nayak J K, Alam J, Sarkar S and Sinha B 2007 Preprint arXiv:0705.1591 [nucl-th]
[350] Alam J e et al. 2007 J. Phys. G: Nucl. Part. Phys. 34 871
[351] Aurenche P et al. 2006 Phys. Rev. D73 094007
[352] Arleo F and Gousset T, work in preparation
[353] Arleo F, Aurenche P, Belghobsi Z and Guillet J-P 2004 JHEP 0411 009
[354] Arleo F 2006 JHEP 0609 015
[355] Arleo F, d’Enterria D and Peressounko D 2007 Preprint arXiv:0707.2356 [nucl-th]
[356] Aurenche P et al. 1999 Eur. Phys. J. C9107
[357] Bourhis L, Fontannaz M, Guillet J P and Werlen M 2001 Eur. Phys. J C19 89
[358] Arleo F 2007 Preprint arXiv:0707.2320 [hep-ph]
[359] Turbide S, Rapp R and Gale C 2004 Phys. Rev. C69 014903

Heavy Ion Collisions at the LHC - Last Call for Predictions 185
[360] Chatterjee R, Frodermann E, Heinz U and Srivastava D K 2006 Phys. Rev. Lett. 96 202302; Chatterjee
R, Srivastava D K, Heinz U and Gale C 2007 Phys. Rev. C75 054909; Chatterjee R, Frodermann E,
Heinz U, and Srivastava D K 2007, to be published
[361] Kestin G and Heinz U, these proceedings, Sec. 2.4
[362] Dremin I M and Nechitailo V A 2007 Preprint arXiv:0704.1081 [hep-ph].
[363] Arnaldi R et al. [NA60 Collaboration] 2006 Phys. Rev. Lett. 96 162302
[364] Damjanovic S 2007 Eur. Phys. J. C 49 235
[365] Damjanovic S et al. [NA60 Collaboration] 2007 Nucl. Phys. A 783 327
[366] Fries R J, Müller B and SrivastavaD K 2003 Phys. Rev. Lett. 90 132301
[367] Zakharov B G 2004 JETP Lett. 80 1
[368] Turbide S, Gale C, Jeon S and Moore G D 2005 Phys. Rev. C72 014906
[369] Turbide S, Gale C, Srivastava D K and Fries R J 2006 Phys. Rev. C74 014903
[370] Kopeliovich B Z, Pirner H J, Potashnikova I K and Schmidt I 2007 Preprint arXiv:0706.3059 [hep-ph]
[371] Bartels J, Golec-Biernat K J and Kowalski H 2002 Phys. Rev. D66 014001
[372] F. Abe et al. [CDF Collaboration] 1994 Phys. Rev. Lett. 73 2662 [Erratum 1995 Phys. Rev. Lett. 74 1891]
[373] Adler S S et al. [PHENIX Collaboration] 2007 Phys. Rev. Lett. 98 012002
[374] Rapp R and Wambach J 1999 Eur. Phys. J. A6415
[375] van Hees H and Rapp R 2006 Phys. Rev. Lett. 97 102301
[376] Vitev I, in preparation
[377] Qiu J w and Vitev I 2006 Phys. Lett. B632 507; 2004 Phys. Rev. Lett. 93 262301
[378] Liang Z T and Wang X N 2005 Phys. Rev. Lett. 94 102301
[379] Becattini F and Ferroni L 2007 Preprint arXiv:0707.0793.
[380] Becattini F and Piccinini F, in preparation
[381] Arkani-Hamed N, Dimopoulos S and Dvali G R 1998 Phys. Lett. B429 263
[382] Antoniadis I, Arkani-Hamed N, Dimopoulos S and Dvali G R 1998 Phys. Lett. B436 25 (1998)
[383] Randall L and Sundrum R 1999 Phys. Rev. Lett. 83 4690
[384] Banks T and Fischler W 1999 Preprint arXiv:hep-th/9906038
[385] Stoecker H 2006 J. Phys. G: Nucl. Part. Phys. 32 S429
[386] Koch B, Bleicher M and Stoecker H 2007 J. Phys. G: Nucl. Part. Phys. 34 S535
[387] Navarra F S, Nielsen M and Lee S H 2007 Phys. Lett. B649 166
[388] Sarac Y, Kim H and Lee S H 2006 Phys. Rev. D73 014009
[389] Lee S H, Yasui S, Liu W and Ko C M 2007 Preprint arXiv:0707.1747 [hep-ph]
[390] Chen L W, Ko C M, Liu W and Nielsen M 2007 Phys. Rev. C76 014906
[391] Kopenkin V V, Managadze A K, Rakobolskaya I V and Roganova TM 1995 Phys. Rev. D 52 2766
[392] Mukhamedshin R A 2005 JHEP 0505 049
[393] Halzen F and Morris D A 1990 Phys. Rev. D 42 1435
[394] Lokhtin I P, Managadze A K, Sarycheva L I and Snigirev A M 2005 Eur. Phys. J C 44 51
[395] Lokhtin I P, Managadze A K, Sarycheva L I and Snigirev A M 2006 Phys. Atom. Nucl. 69 113