Basic phenomenology for relativistic heavy-ion collisions ∗

Wojciech Florkowski
arXiv:1410.7904v1 [nucl-th] 29 Oct 2014

The H. Niewodniczański Institute of Nuclear Physics, Polish Academy of

Sciences, PL-31342 Kraków, Poland, and
Institute of Physics, Jan Kochanowski University, PL-25406 Kielce, Poland

Basic concepts used to interpret the soft-hadronic data collected in

relativistic heavy-ion collisions are reviewed at the elementary level.

1. Introduction
Physics of relativistic heavy-ion collisions is a very broad, interdisci-
plinary field of physics. It is impossible to cover its all important aspects
in a short text. Therefore, one has to decide which topics are selected for
presentation. The idea behind these lectures is to give a possibly general
overview which concentrates on soft-hadronic observables and may serve as
the background for other advanced lectures presented during the School.
The omitted discussions are partly compensated by references given to
many original papers. They will guide the reader in further studies. We
also note that at present there are several textbooks available which discuss
heavy-ion collisions and physics of the quark-gluon plasma [1, 2, 3, 4, 5].
We refer to them and to the collected review articles [6] for additional in-
formation.
In the remaining part of Introduction we give a historical outline of the
development of the heavy-ion physics, present the main theoretical tools,
and discuss the concepts of the quark-gluon plasma. In Sec. 2 we discuss
the main physics terminology used in our field. Section 3 is devoted to the
concept of the limiting Hagedorn temperature. This concept has turned out
to be very inspiring in the physics of strong interactions and led us directly
to the idea of the phase transition from ordinary hadronic matter to quark
matter (later called the quark-gluon plasma [7, 8]). In Sec. 4 we discuss
shortly the coefficients of the Fourier expansion of the momentum distribu-
tion in the azimuthal angle. The first three coefficients are known as the
∗
email address: wojciech.florkowski@ifj.edu.pl

(1)
2 wf printed on October 30, 2014

directed, elliptic and triangular flows. The large values of the elliptic flow
have been reproduced within the hydrodynamic calculations and suggest
the small viscosity to entropy density ratio of the quark-gluon plasma. In
Sec. 5 we present the basics of the Glauber model, which is commonly used
to determine initial distributions of the entropy or energy density in the
transverse plane. The Glauber model serves also as a tool in comparisons
between heavy-ion and more elementary proton–nucleus and proton–proton
collisions. The hydrodynamic approaches are discussed in Sec. 6, where we
first recall the famous Fermi, Landau and Bjorken models and then switch
to the characteristics of the perfect-fluid and viscous-fluid dynamics. The
final stage of the space-time evolution of matter is shortly called a freeze-
out. Several approaches to deal with this rather complicated process are
presented in Sec. 7. The space-time dimensions of the produced system at
freeze-out can be inferred by the study of the identical particle correlations,
which we shortly discuss in Sec. 8. The lectures are closed with short con-
clusions. Throughout the text we use natural units where c = h̄ = kB = 1.

1.1. Historical background

The name “heavy-ions” is used for heavy atomic nuclei, and the term
“relativistic energy” denotes the energy regime where the kinetic energy is
much larger than the rest energy. The first experiments with the relativistic
heavy ions (energies larger than 10 GeV per nucleon in the projectile beam)
took place at the Brookhaven National Laboratory (BNL) and at the Euro-
pean Organisation for Nuclear Research (CERN) in 1986. The Alternating
Gradient Synchrotron (AGS) at BNL accelerated 28 Si at 14 GeV per nu-
cleon. At CERN, the Super Proton Synchrotron (SPS) accelerated 16 O at
60 and 200 GeV per nucleon in 1986, and 32 S at 200 GeV per nucleon in
1987. In 1990 the next project on heavy-ion physics was organised at CERN
with 32 S beams. In 1992 the experiments with 197 Au beams at 11 GeV per
nucleon were initiated at BNL. In 1995 the new experiments took place at
CERN with 208 Pb beams at 158 GeV per nucleon.
In 2000 the first data from the Relativistic Heavy Ion Collider (RHIC)
at BNL were taken. During the first run, the maximum energy of 130 GeV
per nucleon pair was achieved. In the next years new runs took place with
the maximum energy of 200 GeV per nucleon pair. One of those runs was
devoted to study deuteron–gold collisions which were analysed in order to
get a reference point for more complicated gold–gold collisions.
At present, the main activity in the field is connected with Large Hadron
√
Collider (LHC) at CERN (Pb on Pb reactions at sNN = 2.76 TeV, first run
in Nov.-Dec., 2010). However, new experiments at lower energies (NA61 at
CERN, STAR at BNL) are also very important, since this allows us to study
 wf printed on October 30, 2014 3

the energy dependence of many characteristics of the particle production

and analyse the systems at finite baryon chemical potential (see Marek
Gaździcki’s lecture).

1.2. Theoretical tools

In the relativistic heavy-ion collisions very large numbers (multiplicities)
of particles are produced. For instance, in the central Au–Au collisions at
√
RHIC, at the beam energy sNN = 200 GeV, the total charged particle
multiplicity is about 5000. Hence, the number of produced particles ex-
ceeds the number of initial nucleons by a factor of 10. In this situation,
different theoretical methods are used, which are appropriate for descrip-
tion of large macroscopic systems, e.g., thermodynamics, hydrodynamics,
kinetic (transport) theory, field theory at finite temperature and density,
non-equilibrium field theory, Monte-Carlo simulations. More importantly,
we may also apply the fundamental theory of strong interactions.
In high-energy nuclear collisions a many-body system of strongly in-
teracting particles is produced. The fundamental theory of strong inter-
actions is Quantum Chromodynamics (QCD), the theory of quarks and
gluons which are confined in hadrons, i.e., baryons and mesons. The dis-
covery of asymptotic freedom in the strong interactions in 1973 by Gross,
Politzer, and Wilczek [9, 10] allowed for making precise predictions of the
results of many high-energy experiments in the framework of the perturba-
tive quantum field theory — the asymptotic freedom is the property that
the interaction between particles becomes weaker at shorter distances (see
the School lectures by George Sterman).
Probably the most striking feature of QCD is color confinement, which
is the other side of the asymptotic freedom. This is the phenomenon that
color charged particles (such as quarks and gluons) cannot be isolated as
separate objects. In other words, quarks and gluons cannot be directly
observed. The physical concept of confinement may be illustrated by a
string which is spanned between the quarks when we try to separate them.
If the quarks are pulled apart too far, large energy is deposited in the string
and it breaks into smaller pieces.

1.3. Quark-gluon plasma

The main aim of the relativistic heavy-ion collisions is the observation
of the two phase transitions predicted by QCD, i.e., the deconfinement and
chiral phase transitions. At Earth conditions (i.e., at low energy densities)
quarks and gluons are confined in hadrons. However, with increasing tem-
perature (heating) and/or increasing baryon density (compression), a phase
transition may occur to the state where ordinary hadrons do not exist any
4 wf printed on October 30, 2014

longer; quarks and gluons become the right degrees of freedom, and their
motion is not confined to hadrons.
This popular picture is based on the asymptotic freedom — QGP is con-
sidered as an asymptotic state available at extremely high energies. Most
likely, such a state has not been reached in the present experiments and,
more importantly, it is very difficult to find an experimental evidence for
its formation. On the other hand, we can accept a more pragmatic point of
view and consider QGP as a new state of strongly interacting matter, whose
properties can be inferred from experimental and theoretical investigations
carried out at the currently available energies (with direct connections to
QCD wherever it is possible). The present evidence suggests that the matter
produced in heavy-ion collisions consists of quarks and gluons (due to the
strong coupling these might not be elementary excitations in the system),
it is locally well equilibrated, and characterised by the small (shear) vis-
cosity to entropy density ratio. These striking experimental and theoretical
findings suggest that QGP behaves more like a fluid than a gas [11, 12].
In the limit of vanishing masses, the left– and right–handed quarks be-
come decoupled from each other and QCD becomes invariant under their
interchange — left- and right-handed quark currents are separately con-
served, each state of the theory should have a degenerate partner of the
opposite parity. On the other hand, we know that hadrons have well de-
fined parity, and no such parity partners are observed. This paradox is
resolved by the phenomenon of the spontaneous breakdown of chiral sym-
metry [13, 14]: the chiral symmetry of the interaction is broken by the true
ground state of the theory. One expects that this symmetry is restored at
high energies where quarks and gluons become the correct degrees of free-
dom. This is a very exciting subject but we are not going to follow it any
longer in these lectures.

2. Basic physics terms

2.1. Participants, spectators, and wounded nucleons
At high energies, simple geometric concepts are often used, for exam-
ple, one distinguishes participants from spectators — if we assume that all
nucleons propagate initially (i.e., before a collision) along parallel, straight
line trajectories, then the nucleons which do not strike any other nucleons
on their way are called spectators. Other nucleons which interact with each
other are called participants. The participants which suffered at least one
inelastic collision are called the wounded nucleons. A two-dimensional vec-
tor connecting centres of the colliding nuclei is called the impact vector (its
length is the impact parameter). In particle and nuclear physics one intro-
duces a coordinate system, where the spatial z-axis is parallel to the beam,
 wf printed on October 30, 2014 5

and the impact vector b points in x-direction. The axes x and z span the
reaction plane of a given collision.
The simple picture outlined above is most convenient for theoretical in-
vestigations where we control the (initial) geometry of the collision process.
On the other hand, the quantities such as the impact parameter or the reac-
tion plane are not directly measured observables and we have to introduce
them in a more sophisticated way. Before we do it, let us define the popular
ways to parametrise the four-momenta of the produced particles.

2.2. Transverse mass and rapidity

The component of a three-vector A parallel to z-axis is usually denoted
by Ak , and the transverse component is A⊥ = A − Ak . The transverse
q
mass of a particle is defined as m⊥ = m2 + p2⊥ , where m and p are the
particle’s mass and three-momentum (the “transverse” quantities are also
denoted by T , e.g., mT or pT , the “longitudinal” quantities are then denoted
by L, e.g., pL ).
Since we deal with relativistic energies, it is useful to use the rapidity
instead of the standard velocity

1 (E + pk ) p 
k

y= ln = arctanh = arctanh vk . (1)
2 (E − pk ) E
p
Here E is the energy of a particle, E = m2 + p2 , and vk = pk /E is the
longitudinal component of the velocity.
Rapidity is additive under Lorentz boosts along the z-axis. This means
that the difference dy as well as the rapidity density dN/dy do not change
under Lorentz boosts along the collision axis. The invariance under this type
of transformation (corresponding to a constant dN/dy) is shortly called the
boost-invariance.
Using the rapidity and the transverse mass, we can calculate the en-
ergy and the longitudinal momentum of a particle from the two equations:
E = p0 = m⊥ cosh y and pk = m⊥ sinh y. Experimentalists distinguish be-
tween rapidity and pseudorapidity. The latter is defined by the formula

1 (|p| + pk )
   
θ θ
η = ln = ln cot = − ln tan , (2)
2 (|p| − pk ) 2 2

where θ is the scattering angle. Pseudorapidity is easier to measure than

rapidity (it is just a measure of the angle at which a particle has been
emitted). To measure rapidity one has to identify the particle. Since at large
energies E ≈ |p| one is often tempted to assume that dN/dy ≈ dN/dη. In
6 wf printed on October 30, 2014

practice, this approximation is poor, especially in the region where rapidity

is close to zero.
In theoretical calculations one usually uses the space-time rapidity

1 t+z
ηk = ln . (3)
2 t−z

The space-time rapidity ηk together with the (longitudinal) proper time

√
τ = t2 − z 2 are used instead of the coordinates t and z to parametrise
the interior of the light cone t2 = z 2 (t > 0). One has to be careful to
distinguish between rapidity, pseudorapidity, and space-time rapidity. In
realistic calculations they are usually quite different. Only in very simple
boost invariant models one may assume that these three quantities are equal.

3. Hagedorn limiting temperature

In 1960s the statistical bootstrap model (SBM) was introduced [15, 16]
that was based on the observation that hadrons form bound and resonance
states. This led to the concept of a possibly unlimited sequence of heavy
resonance states, each being a constituent of a still heavier resonance. The
number of such states in the mass interval (m, m + dm) is denoted by
ρ(m)dm, and the function ρ(m) is known as the SBM mass spectrum. The
requirement that resonances are formed from other resonances in the self-
consistent manner leads to the bootstrap condition for the mass spectrum
ρ(m). The solution of the bootstrap equation shows that the mass spec-
trum for large masses m grows exponentially, as found by Hagedorn al-
ready in 1965 [15]. As a consequence, any thermodynamics employing this
mass spectrum has a singular temperature TH generated by the asymptotics
ρ(m) ∼ exp(m/TH ). At present TH (Hagedorn temperature) is interpreted
as the temperature where the phase transition from the hadron gas to the
quark-gluon plasma occurs [17].
The subject of the limiting temperature is still an intriguing issue. More
recent studies of the hadron mass spectrum have revealed that the Hagedorn
temperatures of mesons and baryons are different [18, 19]. The concepts that
we are still missing some resonance states and they may be responsible for
(fast) thermalization of the produced matter are widely discussed [20].

4. Harmonic flows
At present, the extraction of the reaction plane is one aspect of the
very advanced flow analysis of the collisions [21, 22]. In this type of the
investigations one represents the momentum distribution of the produced
 wf printed on October 30, 2014 7

particles in the form

∞
" #
dN dN X
= 1+ 2vk cos (k(φp − Ψk )) , (4)
dyd2 p⊥ 2πp⊥ dp⊥ dy
k=1

where Ψk is the reference angle defined by the condition hsin(kΨk )i = 0,

where the averaging is done over all particles in one event. Until very
recently it has been common to assume Ψk = ΨRP for all k’s and to identify
the angle ΨRP with the angle which specifies the position of the reaction
plane.
Averaging of (4) over the azimuthal angle gives the transverse-momentum
distribution. The coefficients vk characterize the momentum anisotropy.
The coefficient v1 is called the directed flow, whereas the coefficient v2 is
called the elliptic flow. In general, the coefficients vk are functions of rapid-
ity and transverse momentum, vk = vk (y, p⊥ ), and in this form often called
the kth harmonic differential flow.

4.1. Directed flow

At low energies, the directed flow is manifested by the reflection of in-
coming particles by the first produced regions of highly compressed nuclear
matter. At the SPS energies the situation is already quite complex [23].
At positive rapidities the proton directed flow is positive, while the pion
directed flow is negative. This suggests a different origin of v1 of protons
and pions. At the RHIC energies the directed flow of charged particles is
negative, whereas the v1 of the spectator neutrons is positive. This trend
in the data suggests different behavior of the matter created in the central
region and in the target/projectile fragmentation regions [24, 25].

4.2. Elliptic flow and thermalisation

In a non-central collision (characterised by a finite impact parameter b),
the particles are produced in an almond-like region in the transverse plane.
This region can be characterised by the spatial anisotropy parameter ǫ. The
initial longitudinal momenta of the produced particles are very large (due
to the initial impact). On the other hand, the typical initial transverse mo-
menta are small and distributed isotropically. If the produced particles do
not interact, their final transverse momenta should be also isotropic. Con-
trary, if the particles do interact, the spatial anisotropy is being transferred
into the momentum anisotropy [21]. This leads to non-zero values of the
coefficient v2 in Eq. (4), which becomes proportional to ǫ.
At RHIC energies, the measured values of the elliptic flow were ex-
plained first by the approaches based on the perfect-fluid hydrodynamics
8 wf printed on October 30, 2014

[26]. This led to the conclusion that the system produced in such collisions
is characterised by the very small shear viscosity η [11, 12]. Further studies
done within dissipative hydrodynamics set an upper limit on the ratio of
the shear viscosity η to the entropy density s, which is still very low [27, 28],
1 ≤ 4πη/s ≤ 3 (smaller than for Helium at its critical temperature). One
has to clarify that the shear viscosity itself is not a good measure of the
viscous effects as it is a dimensional quantity. A better measure is the ra-
tio η/s as it is a dimensionless observable (in the natural system of units).
Therefore, the recent hydrodynamic studies aim at fixing η/s.
Originally, the large values of v2 were treated as the evidence of fast ther-
malisation of the produced matter. A frequently repeated statement was
that v2 could be generated only in the very early times, smaller than 1 fm/c.
Model calculations show, however, that the elliptic flow can be generated
during the whole time evolution of the system [29, 30] (although the initial
growth is the strongest). The realistic values of v2 can be obtained with
scenarios assuming initial free streaming of patrons or large anisotropy of
the initial pressure [31, 32]. This allows for delayed thermalisation taking
place at the times 1–2 fm/c, which is compatible with the results of differ-
ent microscopic calculations (see the School lectures by Francois Gelis and
Michael Strickland).

4.3. Triangular flow

Recent hydrodynamic calculations include event-by-event fluctuations

in the distribution of the initial energy or entropy density [33, 34] (this ef-
fect follows from the fluctuations of the nucleon-nucleon collisions which are
described below in the framework of the Glauber model). The fluctuations
make this distribution less symmetric and this gives rise to both cosine and
sine terms in the Fourier decomposition of the particle momenta in the az-
imuthal angle. Equivalently, different reference angles Ψk should be used
for each value of k in the decomposition (4). In event-by-event hydrody-
namics, one performs the hydrodynamic calculation for a single event first,
extracts vk ’s, and then the averaging over the events is done. This leads to
non-trivial results for the odd harmonics such as, for example, v3 which is
called the triangular flow [35]. The combined measurements of the elliptic
and triangular flow enhance our ability to determine the value of the shear
viscosity.
 wf printed on October 30, 2014 9

5. Glauber model
Glauber model is used to determine the initial energy (or entropy) den-
sity of matter produced in heavy-ion collisions. 1 With an additional as-
sumption that the matter is thermalised, one can use the results obtained
with the Glauber model as the input for hydrodynamic calculations which
determine the subsequent space-time evolution of matter until the freeze-out
point.
Originally, the Glauber model was applied only to elastic collisions. In
this case a nucleon does not change its properties in the individual collisions,
so all nucleon interactions can be well described by the same (elastic) cross
section. Applying the Glauber model to inelastic collisions, we assume that
after a single inelastic collision an excited nucleon-like object is created that
interacts basically with the same inelastic cross section with other nucleons.
An alternative to the Glaber model calculations are theoretical studies
related more directly to QCD. The most common approach of this type
is the color–glass–condensate (CGC) theory [36, 37] which is based on the
concept of gluon saturation [38, 39] (see the lectures by Francois Gelis, Yuri
Kovchegov and Larry McLerran).

5.1. Nucleon-nucleon collisions

At high energies, the inelastic nucleon-nucleon cross section σin gives
the main contribution to the total cross section. A certain subclass of the
inelastic processes is the diffractive dissociation process. In this process a
nucleon is only slightly excited and a small number of particles is produced.
The diffractive processes contribute to about 10% of all inelastic collisions.
In non-diffractive inelastic nucleon-nucleon collisions a certain number of
charged particles is produced. The average charged particle multiplicity
N NN is described by√ phenomenological formulas which give N NN as a func-
tion of the energy s . Such formulas can be used in the comparisons done
between heavy-ion and hadronic collisions. √
Let us consider a nucleon-nucleon collision at a given energy s and at
an impact parameter b. In the eikonal approximation, we may introduce
the probability of having a nucleon-nucleon inelastic collision
 
2
iχ(b)
p (b) = 1 − e ≡ t (b) σin , (5)

where χ(b) is the phase shift (times a factor of two). The function t (b),
defined by (5), is called the nucleon-nucleon thickness function. The integral
1
For a recent review of the applications of the Glauber model to describe initial stages
of relativistic heavy-ion collisions see, for example, Sec. III in Ref. [5].
10 wf printed on October 30, 2014

of p (b) over the whole range of the impact parameter should be normalized
to σin . Thus, the thickness function is normalized to unity.

5.2. Nucleon-nucleus collisions

The probability of finding a nucleon in the nucleus with the atomic mass
number A is the usual baryon density divided by the number of baryons in
the nucleus. For large nuclei, one commonly uses the Woods-Saxon function
(our definition of ρA (r) includes A in the denominator, because we want to
interpret ρA (r) as the probability distribution)
  −1
ρ0 r − r0
ρA (r) = 1 + exp , (6)
A a

with the parameters: r0 = (1.12A1/3 − 0.86A−1/3 ) fm, a = 0.54 fm, ρ0 =

0.17 fm−3 . The parameter ρ0 is the nuclear saturation density.
The nucleon-nucleus thickness function for the nucleus A is obtained
from a simple geometric consideration and the assumption that the nucleon
positions in the nucleus A are not changed during the collision process,
Z Z
TA (b) = dzA d 2 sA ρA (sA , zA ) t(sA − b). (7)

Here the transverse coordinates in the nucleus A are denoted by the vector
sA and q 
ρA (sA , zA ) = ρA 2 2
sA + zA . (8)

Our definition of the Woods-Saxon distribution implies the normalization

condition Z
d 2 b TA (b) = 1. (9)

The quantity TA (b) σin is the probability that a single nucleon-nucleon

collision takes place in a nucleon-nucleus collision at the impact parameter
b. Treating all possible nucleon-nucleon collisions in the nucleon-nucleus
collision as completely independent and characterized by the same cross
section, we easily find the probability of having n such collisions
 
A
P (n; A; b) = [1 − TA (b) σin ]A−n [TA (b) σin ]n . (10)
n

The average number of binary nucleon-nucleon collisions may be calculated

from the expression n (A; b) = A TA (b) σin .
 wf printed on October 30, 2014 11

5.3. Nucleus-nucleus collisions

The thickness function for the nucleus-nucleus collision follows also from
the geometric considerations which give in this case
Z Z Z Z
TAB (b) = dzA d sA ρA (sA , zA ) dzB d 2 sB ρB (sB , zB ) t(b+sB −sA ),
2

(11)
with the corresponding normalization condition d 2 b TAB (b) = 1.
R
The quantity TAB (b) σin is the averaged probability that a nucleon-
nucleon collision takes place in a nucleus-nucleus collision characterized by
the impact parameter b. In the limit t(b) → δ(2) (b) we may write
Z
TAB (b) = d 2 sA TA (sA ) TB (sA − b).
(12)
The nucleus-nucleus thickness function TAB (b) can be used to calculate
the probability of having n inelastic binary nucleon-nucleon collisions in a
nucleus-nucleus collision at the impact parameter b
 
AB
P (n; AB; b) = [1 − TAB (b) σin ]AB−n [TAB (b) σin ]n . (13)
n
The average number of the collisions is n (AB; b) = AB TAB (b) σin .
The total probability of an inelastic nuclear collision is the sum over n
from n = 1 to n = AB
AB
P (n; AB; b) = 1 − [1 − TAB (b) σin ]AB .
X
Pin (AB; b) = (14)
n=1
In more realistic calculations, the positions of nucleons in the target and
projectile nucleus are fixed, and the averaging should be done later. The
probability of an inelastic collision for a fixed nucleon configuration equals
A Y
Y B
1 − t b + sB A


1− i − sj σin . (15)
j=1 i=1

The probability of an inelastic nuclear collision at the impact parameter b

is then
Z Z
Pin (AB; b) = d2 sA A 2 A
d2 sB A B 2 B B
1 TB (s1 ) · · · d sB TB (sB )
1 TA (s1 ) · · · d sA TA (sA )
 
 A Y
Y B 
B A


× 1− 1 − t b + si − sj σin . (16)
 
j=1 i=1
12 wf printed on October 30, 2014

AB . Equations (14) and (16) differ from

The integration of (16) over b gives σin
each other! The more accurate formula (16) is much more complicated to
handle and cannot be simply reduced to (14). Only for nucleon-nucleus
collisions the two methods are equivalent. Since there is no good analytic
method to evaluate (16) for large values of A and B, one is most often
satisfied with Eq. (14) only. It is called the optical limit of the Glauber
model. In order to have a more reliable distributions one does the Monte-
Carlo calculations [40, 41].

5.4. Wounded nucleons

The Glauber model can be used also to calculate the number of the
participants. To be more precise we distinguish between the participants
which may interact elastically and the participants which interact only in-
elastically. The latter are called Rthe wounded nucleons [42]. The number
of nucleons in the nucleus A is A d2 s TA (s). Probability, that the nucleon
from A at the position s collides one or more times with the nucleons in B
(in an AB collision at the impact parameter b) is

B
P (n; B; b − s) = 1 − [1 − σin TB (b − s)]B .
X

n=1
(17)

The number of wounded nucleons in A can be obtained from the expression

Z  
w A (A; B; b) = A d2 s TA (s) 1 − [1 − σin TB (b − s)]B . (18)

An analogous expression holds for the number of wounded nucleons in B.

Since the number of wounded nucleons in the collision of A and B is the
sum of the wounded nucleons in the nucleus A and B, we obtain [42] (after
making the appropriate shifts in the integration over positions s)
Z  
w (A; B; b) = A d2 s TA (b − s) 1 − [1 − σin TB (s)]B
Z  
+ B d2 s TB (b − s) 1 − [1 − σin TA (s)]A . (19)

5.5. Nuclear modification factor

A simple way to quantify the differences between the nucleus-nucleus
collisions and the nucleon-nucleon collisions is to calculate the nuclear mod-
 wf printed on October 30, 2014 13

ification factor,

pp
1 d2 N AB 1 dσincl
RAB (p⊥ ) = / pp . (20)
nAB dp⊥ dη σtot dp⊥ dη

N AB – average number of particles produced in the collisions of the nuclei

A and B, nAB – number of the binary nucleon-nucleon collisions obtained
in the framework of the Glauber model.
The denominator of (20) is the inclusive cross section for pp collisions
divided by the total cross section. This quantity is equal to the average
number of particles produced in pp collisions in the appropriate phase-space
interval,

pp
dNpp 1 dσincl
= pp . (21)
dp⊥ dη σtot dp⊥ dη

If the collisions of the nuclei A and B are simple superpositions of the

elementary pp collisions, the scaling with the number of binary collisions
should hold, and the nuclear modification factor is expected to be equal to
unity. This is the reason why the nuclear modification factor is so frequently
used in direct searches for non-trivial dynamics.

6. Hydrodynamic description of heavy-ion collisions

6.1. Early statistical and hydrodynamic models
The use of relativistic hydrodynamics to describe particle production in
hadronic collisions has a long history which starts with the famous work of
Landau in the early 1950s [43], see also [44, 45]. Landau’s considerations
were preceded, however, by a few approaches that used pure statistical and
thermodynamic concepts in the analysis of hadronic collisions [46, 47, 48,
49]. Such approaches may be regarded as pre-hydrodynamic models and it
is important to present them before we turn to the discussion of the genuine
hydrodynamic models.

6.1.1. Fermi model

Fermi assumed that when two relativistic nucleons collide, the deposited
energy is released in a very small volume V , whose magnitude corresponds
to the Lorentz contracted characteristic pion field volume V0 , i.e., V =
14 wf printed on October 30, 2014

√ √
2mN V0 / s, where V0 = (4/3) πRπ3 with Rπ = 1/mπ , and where s is the
center-of-mass energy. Subsequently, such a dense system decays into one
of many accessible multiparticle states. The decay probability is calculated
in the framework of the standard statistical physics.
The main reason for the introduction of the statistical concepts was the
breakdown of the perturbation theory in description of strongly interacting
systems. Clearly, the large values of the coupling constant prohibit the ap-
plication of the perturbation theory. On the other hand, the large coupling
is responsible for the phenomenon of multiparticle production, which is a
very characteristic feature of strong interactions.
The probability of the transition into a given state is proportional to the
square of the matrix element and to the density of states. In the statistical
description, the matrix elements are treated as constants and the main
effect comes from the phase space. Thus, the statistical approach represents
a simple theoretical modeling of collisions which may be regarded as the
complementary approach to the perturbation schemes which typically break
down at a certain scale. The main heuristic argument for the justification
of the use of the statistical approach is that the role of the phase space
naturally grows with the increasing energy of the collisions.
According to Fermi, the probability for the formation of the state with
n particles is proportional to the factor
 (n−1)
V dQ(W )
S(n) = 3
. (22)
(2π) dW
Here W is the total energy of the colliding system, dQ/dW is the number
of states per unit energy, and V is the interaction volume. The power n − 1
arises from the fact that the momenta of only n−1 particles are independent.
In the last chapter of his seminal paper from 1950, Fermi considers the
collisions at extremely high energies. He argues that in this case a detailed
statistical considerations may be replaced by the simple thermodynamic
arguments. Assuming that the matter is thermalized one can calculate
the temperature of the produced hadronic system from the thermodynamic
relations valid for massless particles. Fermi took into account only the
production of pions, nucleons and antinucleons, and used the formula

π2 V T 4
επ + εN +N̄ V = = W. (23)
3
Using the expression for the Lorentz contracted volume we rewrite (23) in
the following form
3 W2 9 W 2 m3π
T4 = = . (24)
2π 2 V0 mN 8π 3 mN
 wf printed on October 30, 2014 15

This equation may be used to calculate the abundances of the produced pi-
ons, nucleons and antinucleons from the thermodynamic relations giving the
particle densities in terms of the temperature. It is interesting to note that
Fermi’s idea forms the ground for present thermal model analyses discussed
below in Sec. 7.2.

6.1.2. Landau model

In the Landau model [43] an expansion of matter before the hadron de-
coupling is included. The idea of modification of the Fermi approach was
indicated by Pomeranchuk [49]. He argued that the particles in the system
should interact until the average distance between them becomes larger
than the typical interaction distance. Landau proposed his hydrodynamic
approach to describe proton-proton collisions. Following Fermi, he assumed
that the two colliding protons released their energy in the volume corre-
sponding to the Lorentz-contracted size of a proton. Under the influence of
the longitudinal gradient the system starts expanding. The transverse gra-
dient is also present but initially the gradient in the longitudinal direction is
much larger and the early expansion may be regarded as one-dimensional.

6.1.3. Bjorken model

In the Landau model the initial conditions are specified for a given lab-
oratory time in the center-of-mass frame, when the matter is highly com-
pressed and at rest. Landau’s description does not include one aspect of
high-energy processes – fast particles are produced later and further away
from the collision center than the slow particles. It is possible to include
this effect by imposing special initial conditions. This idea was proposed
and developed by Bjorken [50].
The Bjorken hydrodynamic model [50] is based on the assumption that
the rapidity distribution dNch /dy is constant in the mid-rapidity region.
This fact means that the central region is boost invariant. In this case the
longitudinal flow has the form vz = z/t and all thermodynamic quantities
characterizing
√ the central region depend only on the longitudinal proper
time τ = t2 − z 2 and the transverse coordinates x and y.
The main success of the Bjorken model is that it allows for simple and
realistic estimates of the initial energy density available at the early stages
of heavy-ion collisions. Such estimates always indicate that the produced
matter has the energy density much larger than the typical energy density
characterising the phase transition from the hadron gas to the quark-gluon
plasma. Thus, using Bjorken’s simple estimates we expect a formation of a
new state of matter in heavy-ion collisions at the relativistic energies.
16 wf printed on October 30, 2014

6.1.4. Towards modern hydro approaches

Modern hydrodynamic calculations follow general concepts introduced

in the Landau and Bjorken models. However, they differ from the Landau
original description in the way how they treat the initial conditions. They
also go beyond the simple Bjorken approach by including transverse expan-
sion [51] and, eventually, by breaking the boost-invariance. Additionally,
the recent hydrodynamic codes use modern equations of state inspired by
the lattice simulations of QCD and advanced hadron-gas calculations. In ad-
dition, the hydrodynamic simulations are performed on the event-by-event
basis [33, 34] .
In the Landau and Bjorken models, the thermalized matter was a gas of
ultra-relativistic particles, mainly pions, satisfying the extreme relativistic
equation of state P = ε/3. At present, more accurate equations of state for
hot and dense matter are known (for the perturbative QCD calculation see
[52, 53] and for the lattice QCD calculations see [54, 55] and Christian Hoel-
bling’s lecture). Expecting the phase transition to the quark-gluon plasma,
one can use the plasma equation of state including the phase transition back
to ordinary hadronic matter, see for example Refs. [56, 57]. In this way the
phase transition is incorporated in the hydrodynamic frameworks in a very
elegant and thermodynamically consistent way.
It is interesting to stress that the use of the appropriate QCD equa-
tion of state was crucial for the correct description of the HBT radii (see
Adam Kisiel’s lecture). Moreover, determination of the real character of the
phase transition may be helpful to perform realistic calculations within the
cosmological models based on the Friedman equation [58].

6.2. Initial conditions

For boost-invariant systems with vanishing baryon chemical potential
one usually assumes that either the initial entropy density, σi (x⊥ ) = σ(τi , x⊥ ),
or the initial energy density, εi (x⊥ ) = ε(τi , x⊥ ), are directly related to the
density of sources of particle production, ρsr (x⊥ ).

The sources considered in this context are wounded nucleons or binary

collisions. The symmetry with respect to the Lorentz boosts along the
collision axis means that it is sufficient to consider all these quantities in
the plane z = 0. In general, a mixed model is used, with a linear combination
of the wounded-nucleon density w (x⊥ ) and the density of binary collisions
n (x⊥ ). This leads to the following expression for the initial entropy [59, 60]:

1−κ
σi (x⊥ ) ∝ ρsr (x⊥ ) = w (x⊥ ) + κ n (x⊥ ) , (25)
2
 wf printed on October 30, 2014 17

where κ is a fit parameter. The initial longitudinal profiles are less known.
One usually uses gaussian parametrisation of the entropy density in space-
time rapidity (with a possible flat insert in the middle). The width of this
distribution is chosen in such a way as to reproduce the measured rapidity
distribution.

6.3. Hydrodynamics of perfect fluid

The perfect fluid is defined formally by the form of its energy-momentum
tensor, namely
µν
Teq = (ε + P )uµ uν − P g µν , (26)

where gµν is the metric tensor with g 00 = 1, ε is the energy density, P is

the pressure, and uµ is the four-velocity of the fluid element.
The form (26) follows from the assumption of local thermal equilibrium.
Equations of motion of the perfect fluid are then obtained from the conser-
vation laws
µν
∂µ Teq = 0. (27)
In (27) we have four equations. On the other hand, we search for five
unknown functions: three independent components of the fluid four-velocity,
the energy density, and pressure. The system of equations becomes closed
if (27) is supplemented with the equation of state, for example, in the form
P = P (ε). Using (27) one may check that the entropy is conserved, hence
the flow is adiabatic. If the fluid has non-zero baryon density, one should
add the baryon number conservation law to (27) as an extra equation.

6.4. Hydrodynamics of viscous fluid

In viscous hydrodynamics, the energy-momentum tensor has the form

T µν = Teq
µν
+ Πµν , (28)
µν
where Teq is the perfect-fluid part given by (26) and Πµν describes dissipa-
tion

Πµν = π µν + Π∆µν , ∆µν = g µν − uµ uν . (29)

Here π µν is the shear tensor and Π describes the viscous bulk pressure. The
equations of hydrodynamics follow from the conservation laws for energy
and momentum, and from the requirement that the entropy production is
positive. These conditions determine the form of equations to be satisfied
by the dissipative terms π µν and Π.
18 wf printed on October 30, 2014

From the formal point of view, the inclusion of the dissipative terms in
(28) follows from the gradient expansion around the local equilibrium. In
the first order in gradients one finds the Navier-Stokes expressions

π µν = η∇<µ uν> , Π = ζ∂α uα , (30)

where the angle brackets project out the traceless symmetric part (the sym-
metric part is denoted by round brackets)

2
∇<µ uν> = 2∇(µ uν) − ∆µν ∇α uα , ∇α = ∆αβ ∂β . (31)
3
The quantities η and ζ in (30) are the shear and bulk viscosity, respectively.
Unfortunately, the relativistic fluid dynamics based on the Navier-Stokes
prescription suffers from problems connected with the acausal transmission
of signals. This is why the second-order theory had been developed by
Israel and Stewart [61]. Within the second-order theory, the shear tensor
π µν and the bulk pressure Π satisfy non-trivial dynamic equations. They
are not any longer expressed by simple formulas such as (30). Moreover, the
second-order theory requires that higher-order kinetic coefficients should be
introduced.
At the moment, the formalism developed by Israel and Stewart is the
most popular version of the dissipative hydrodynamics used to describe
heavy-ion collisions. Usually only the shear viscosity is included in such
calculations. There are, however, suggestions that the bulk viscosity may
also play an important role [62]. More importantly, the second-order for-
malism may lead to unphysical behaviour at the early stages of the collisions
or at the edges of the produced system. Such issues are discussed in the
lectures by Michael Strickland [63] in the context of a new formulation of
dissipative fluid dynamics (anisotropic hydrodynamics [63, 64, 65]).

7. Freeze-out
7.1. Kinetic freeze-out
The thermal or kinetic freeze-out is the stage in the evolution of matter
when the hadrons practically stop to interact. In other words, the thermal
freeze-out is a transition from a strongly coupled system (very likely evolv-
ing from one local equilibrium state to another) to a weakly coupled one
(consisting of essentially free-streaming particles).
It is triggered by the expansion of matter, which causes a rapid growth
of the mean free path, λmfp , of particles. The thermal freeze-out happens
when the timescale connected with the collisions, τcoll ∼ λmfp , becomes
larger than the expansion timescale, τexp . In this case the particles depart
 wf printed on October 30, 2014 19

from each other so fast that the collision processes become ineffective. We
may formulate this condition as the inequality [66]

τcoll ≥ τexp . (32)

The magnitude of the collision time is determined by the product of the

average cross section and the particle density,
1
τcoll ∼ , (33)
σn
whereas the magnitude of the expansion time is characterized by the di-
vergence of the four-velocity field, uµ , describing the hydrodynamic flow of
matter,
1
τexp ∼ . (34)
∂µ uµ

Very often a simplified criterion is assumed which says that the thermal
freeze-out happens at the time when the mean free path of hadrons is of the
same order as the size of the system.

7.2. Chemical freeze-out and thermal models

Chemical freeze-out defines the stage where the hadron abundances are
fixed — it should precede the kinetic freeze-out. The concept of the chemical
freeze-out is used in thermal models of particle production. In such models
one assumes that a gas of stable hadrons and resonances is formed (at the
chemical freeze-out). The final (measured) multiplicities of hadrons consist
of primary particles, present in the hot fireball, and of secondary particles
coming from the decays of resonances.
There exist several versions of the thermal approach in the literature,
for example, see [67, 68, 69, 70, 71, 72, 73, 74, 75]. The most popular is
the grand canonical version, where one assumes that all hadron species are
formed in local thermal and chemical equilibrium. Other versions of the
thermal approach assume chemical non-equilibrium in the strange sector of
particles, or chemical non-equilibrium in both the strange and non-strange
sectors 2 .
There is also a possibility that the chemical and thermal freeze-out co-
incide. Such a single freeze-out model was constructed in Ref. [77] and
was successfully used to describe the RHIC soft hadronic observables. Its
Monte-Carlo version is defined in Refs. [78, 79].
2
The results of different versions of the thermal model applied at the LHC energy have
been recently reviewed in [76]
20 wf printed on October 30, 2014

The recent LHC data on heavy-ion collisions show that the predictions
of two popular versions of the statistical model (the chemical equilibrium
model and the strangeness non-equilibrium model) give too large values for
the kaon to pion ratio, (K + + K − )/(π + + π − ), and, especially, for the ratio
of protons to pions (p+ p̄)/(π + +π − ) [80, 81]. The recent fit [75] gives almost
three standard deviations higher values for protons and anti-protons com-
pared to the LHC data. Besides the problems with thermal interpretation
of the hadron abundances at the LHC, one encounters also the problems
with the hydrodynamic interpretation of the transverse-momentum spectra
of pions, kaons and protons. It turns out that one can connect the pro-
ton puzzle with the anomalous behavior of the pion pT spectra and solve
the two problems within the chemical non-equilibrium version of the single
freeze-out model [82, 83].
In the hydrodynamic calculations the freeze-out process is modelled in
two alternative ways: either one uses the concept of a single freeze-out and
assumes that the hadrons are completely decoupled on a specific freeze-
out hypersurface or one switches from the hydrodynamic description to the
hadronic cascade model which relies on the kinetic theory [84].

8. Hanbury–Brown-Twiss interferometry
The fundamental object in the HBT interferometry is the two-particle
correlation function C(p1 , p2 ), measured for pairs of identical particles such
as π + π + , π − π − , or K + K + . In general, it is defined by the expression

P2 (p1 , p2 )
C(p1 , p2 ) = , (35)
P1 (p1 )P1 (p2 )

where P1 (p) is the invariant inclusive one-particle distribution function in

the space of rapidity and transverse-momentum,

dN dN
P1 (p) = Ep 3
= , (36)
d p dyd2 p⊥

and P2 (p1 , p2 ) is the analogous two-particle distribution

dN dN
P2 (p1 , p2 ) = Ep1 Ep2 = . (37)
d3 p 3
1 d p2 dy1 d p1⊥ dy2 d2 p2⊥
2

Equations (36) and (37) imply that the correlation function (35) transforms
like a Lorentz scalar. In (35) we may use the average momentum

1
k= (p1 + p2 ) , (38)
2
 wf printed on October 30, 2014 21

and the difference of the two momenta

q = p1 − p2 . (39)

In the analyses of the correlation functions one uses commonly the out-
side-long coordinate system. First, by making the Lorentz boost along the
collision axis we may set kk = 0. In this way we change to the special frame
that is called the longitudinally comoving system (LCMS). In this frame,
the direction of the beam is called the long direction. The direction of the
three-momentum of the pair is the out direction, and the third orthogonal
direction is called the side direction. The measured correlation functions
are usually fitted with the gaussians of the following form 3
2 2 2 2 2 2
 
C(k⊥ , q) = 1 + λ exp −Rlong (k⊥ )qlong − Rout (k⊥ )qout − Rside (k⊥ )qside .
(40)
The parameters Rside , Rout , and Rlong are called the HBT radii. The mea-
surements of the HBT radii as functions of the mean transverse momentum
of the pion pairs gives us information about the space-time sizes of the
system at the kinetic freeze-out (for more details see the lecture by Adam
Kisiel).

9. Conclusions
Successful applications of relativistic hydrodynamics in description of
ultra-relativistic heavy-ion collisions allowed us to establish a uniform pic-
ture of these complicated processes. In some sense we are in a fortunate
situation that such complex systems and processes may be described within
a concise and well-defined framework.
The hydrodynamic approach, combined with the modeling of the initial
state by the Glauber model or the color glass condensate on one side, and
supplemented by the kinetic simulations of the freeze-out process on the
other side, forms the foundation of an approach that may be regarded as
the standard model of ultra-relativistic heavy-ion collisions. Nevertheless,
many details of this picture should be improved and surprises may wait for
us just around the corner.
This work was supported in part by the Polish National Science Center
with Decision No. DEC-2012/06/A/ST2/00390.

3
Strictly speaking, the parametrisation (40) is suitable for boost-invariant and az-
imuthally symmetric systems. In more general cases one should use more complex
formulas.
22 wf printed on October 30, 2014

REFERENCES

[1] L. Csernai, “Introduction to relativistic heavy ion collisions,” (John Wiley &
Sons, Chichester, 1994) .
[2] C. Wong, “Introduction to high-energy heavy ion collisions,” (World
Scientific, Singapore, 2004) .
[3] J. Letessier and J. Rafelski, “Hadrons and quark - gluon plasma,”
Camb.Monogr.Part.Phys.Nucl.Phys.Cosmol. 18 (2002) 1–397.
[4] K. Yagi, T. Hatsuda, and Y. Miake, “Quark-gluon plasma: From big bang to
little bang,” Camb.Monogr.Part.Phys.Nucl.Phys.Cosmol. 23 (2005) 1–446.
[5] W. Florkowski, “Phenomenology of Ultra-Relativistic Heavy-Ion Collisions,”
(World Scientific, Singapore, 2010) .
[6] B. Friman, C. Hohne, J. Knoll, S. Leupold, J. Randrup, et al., “The CBM
physics book: Compressed baryonic matter in laboratory experiments,”
Lect.Notes Phys. 814 (2011) 1–980.
[7] E. V. Shuryak, “Quark-gluon plasma and hadronic production of leptons,
photons and pions,” Phys. Lett. B78 (1978) 150.
[8] E. V. Shuryak, “Quantum chromodynamics and the theory of superdense
matter,” Phys. Rept. 61 (1980) 71–158.
[9] D. J. Gross and F. Wilczek, “Ultraviolet behavior of non-abelian gauge
theories,” Phys. Rev. Lett. 30 (1973) 1343–1346.
[10] H. D. Politzer, “Reliable perturbative results for strong interactions?,”
Phys. Rev. Lett. 30 (1973) 1346–1349.
[11] M. Gyulassy and L. McLerran, “New forms of QCD matter discovered at
RHIC,” Nucl.Phys. A750 (2005) 30–63,
arXiv:nucl-th/0405013 [nucl-th].
[12] E. V. Shuryak, “What RHIC experiments and theory tell us about properties
of quark-gluon plasma?,” Nucl. Phys. A750 (2005) 64–83.
[13] Y. Nambu, “Axial vector current conservation in weak interactions,”
Phys. Rev. Lett. 4 (1960) 380–382.
[14] Y. Nambu and G. Jona-Lasinio, “Dynamical model of elementary particles
based on an analogy with superconductivity. I,”
Phys. Rev. 122 (1961) 345–358.
[15] R. Hagedorn, “Statistical thermodynamics of strong interactions at high-
energies,” Nuovo Cim. Suppl. 3 (1965) 147–186.
[16] R. Hagedorn and J. Ranft, “Statistical thermodynamics of strong
interactions at high-energies. 2. Momentum spectra of particles produced in
pp collisions,” Nuovo Cim. Suppl. 6 (1968) 169–354.
[17] N. Cabibbo and G. Parisi, “Exponential hadronic spectrum and quark
liberation,” Phys. Lett. B59 (1975) 67.
[18] W. Broniowski and W. Florkowski, “Different Hagedorn temperatures for
mesons and baryons,” Phys. Lett. B490 (2000) 223–227.
 wf printed on October 30, 2014 23

[19] W. Broniowski, W. Florkowski, and L. Y. Glozman, “Update of the

Hagedorn mass spectrum,” Phys. Rev. D70 (2004) 117503.
[20] J. Noronha-Hostler, C. Greiner, and I. A. Shovkovy, “Fast equilibration of
hadrons in an expanding fireball,” Phys. Rev. Lett. 100 (2008) 252301.
[21] J.-Y. Ollitrault, “Anisotropy as a signature of transverse collective flow,”
Phys.Rev. D46 (1992) 229–245.
[22] U. Heinz and R. Snellings, “Collective flow and viscosity in relativistic
heavy-ion collisions,” Ann.Rev.Nucl.Part.Sci. 63 (2013) 123–151,
arXiv:1301.2826 [nucl-th].
[23] H. Stoecker, “Collective flow signals the quark gluon plasma,”
Nucl.Phys. A750 (2005) 121–147, arXiv:nucl-th/0406018 [nucl-th].
[24] A. Bialas and W. Czyz, “Wounded nucleon model and Deuteron-Gold
collisions at RHIC,” Acta Phys.Polon. B36 (2005) 905–918,
arXiv:hep-ph/0410265 [hep-ph].
[25] P. Bozek and I. Wyskiel, “Directed flow in ultrarelativistic heavy-ion
collisions,” Phys.Rev. C81 (2010) 054902, arXiv:1002.4999 [nucl-th].
[26] P. Huovinen, P. Kolb, U. W. Heinz, P. Ruuskanen, and S. Voloshin, “Radial
and elliptic flow at RHIC: Further predictions,”
Phys.Lett. B503 (2001) 58–64, arXiv:hep-ph/0101136 [hep-ph].
[27] P. Romatschke and U. Romatschke, “Viscosity Information from Relativistic
Nuclear Collisions: How Perfect is the Fluid Observed at RHIC?,”
Phys.Rev.Lett. 99 (2007) 172301, arXiv:0706.1522 [nucl-th].
[28] U. Heinz, C. Shen, and H. Song, “The viscosity of quark-gluon plasma at
RHIC and the LHC,” AIP Conf.Proc. 1441 (2012) 766–770,
arXiv:1108.5323 [nucl-th].
[29] A. Bialas, M. Chojnacki, and W. Florkowski, “Early evolution of
transversally thermalized partons,” Phys.Lett. B661 (2008) 325–329,
arXiv:0708.1076 [nucl-th].
[30] R. Ryblewski and W. Florkowski, “Early anisotropic hydrodynamics and the
RHIC early-thermalization and HBT puzzles,”
Phys.Rev. C82 (2010) 024903, arXiv:1004.1594 [nucl-th].
[31] W. Broniowski, W. Florkowski, M. Chojnacki, and A. Kisiel,
“Free-streaming approximation in early dynamics of relativistic heavy-ion
collisions,” Phys.Rev. C80 (2009) 034902, arXiv:0812.3393 [nucl-th].
[32] R. Ryblewski and W. Florkowski, “Highly-anisotropic hydrodynamics in 3+1
space-time dimensions,” Phys.Rev. C85 (2012) 064901,
arXiv:1204.2624 [nucl-th].
[33] C. Gale, S. Jeon, B. Schenke, P. Tribedy, and R. Venugopalan,
“Event-by-event anisotropic flow in heavy-ion collisions from combined
Yang-Mills and viscous fluid dynamics,” Phys.Rev.Lett. 110 (2013) 012302,
arXiv:1209.6330 [nucl-th].
[34] C. Gale, S. Jeon, B. Schenke, P. Tribedy, and R. Venugopalan, “Initial state
fluctuations and higher harmonic flow in heavy-ion collisions,”
Nucl.Phys. A904-905 (2013) 409c–412c, arXiv:1210.5144 [hep-ph].
24 wf printed on October 30, 2014

[35] B. Alver and G. Roland, “Collision geometry fluctuations and triangular flow
in heavy-ion collisions,” Phys.Rev. C81 (2010) 054905,
arXiv:1003.0194 [nucl-th].
[36] L. D. McLerran and R. Venugopalan, “Computing quark and gluon
distribution functions for very large nuclei,” Phys. Rev. D49 (1994)
2233–2241.
[37] L. D. McLerran and R. Venugopalan, “Gluon distribution functions for very
large nuclei at small transverse momentum,” Phys. Rev. D49 (1994)
3352–3355.
[38] K. J. Golec-Biernat and M. Wusthoff, “Saturation effects in deep inelastic
scattering at low Q2 and its implications on diffraction,” Phys. Rev. D59
(1999) 014017.
[39] K. J. Golec-Biernat and M. Wusthoff, “Saturation in diffractive deep
inelastic scattering,” Phys. Rev. D60 (1999) 114023.
[40] W. Broniowski, M. Rybczynski, and P. Bozek, “GLISSANDO: Glauber
initial-state simulation and more..,”
Comput.Phys.Commun. 180 (2009) 69–83, arXiv:0710.5731 [nucl-th].
[41] M. Rybczynski, G. Stefanek, W. Broniowski, and P. Bozek, “GLISSANDO 2:
GLauber Initial-State Simulation AND mOre..., ver. 2,”
Comput.Phys.Commun. 185 (2014) 1759–1772,
arXiv:1310.5475 [nucl-th].
[42] A. Bialas, M. Bleszynski, and W. Czyz, “Multiplicity Distributions in
Nucleus-Nucleus Collisions at High-Energies,” Nucl.Phys. B111 (1976) 461.
[43] L. D. Landau, “On the multiparticle production in high-energy collisions,”
Izv. Akad. Nauk SSSR Ser. Fiz. 17 (1953) 51–64.
[44] I. M. Khalatnikov, “Some questions of the relativistic hydrodynamics,”
Zh.Eksp. Teor. Fiz. . 26 (1954) 529.
[45] S. Z. Belenkij and L. D. Landau, “Hydrodynamic theory of multiple
production of particles,” Nuovo Cim. Suppl. 3S10 (1956) 309.
[46] H. Koppe Z. f. Naturforsch. 3A (1948) 251.
[47] E. Fermi, “High-energy nuclear events,” Prog. Theor. Phys. 5 (1950)
570–583.
[48] E. Fermi, “Angular Distribution of the Pions Produced in High Energy
Nuclear Collisions,” Phys. Rev. 81 (1951) 683–687.
[49] I. Y. Pomeranchuk, “On the theory of multiple particle production in a
single collision,” Dokl. Akad. Nauk Ser. Fiz. 78 (1951) 889–891.
[50] J. Bjorken, “Highly Relativistic Nucleus-Nucleus Collisions: The Central
Rapidity Region,” Phys.Rev. D27 (1983) 140–151.
[51] G. Baym, B. L. Friman, J. P. Blaizot, M. Soyeur, and W. Czyz,
“Hydrodynamics of ultrarelativistic heavy ion collisions,” Nucl. Phys. A407
(1983) 541–570.
[52] N. Haque, J. O. Andersen, M. G. Mustafa, M. Strickland, and N. Su,
“Three-loop HTLpt Pressure and Susceptibilities at Finite Temperature and
Density,” Phys.Rev. D89 (2014) 061701, arXiv:1309.3968 [hep-ph].
 wf printed on October 30, 2014 25

[53] N. Haque, A. Bandyopadhyay, J. O. Andersen, M. G. Mustafa,

M. Strickland, et al., “Three-loop HTLpt thermodynamics at finite
temperature and chemical potential,” JHEP 1405 (2014) 027,
arXiv:1402.6907 [hep-ph].
[54] S. Borsanyi, G. Endrodi, Z. Fodor, A. Jakovac, S. D. Katz, et al., “The QCD
equation of state with dynamical quarks,” JHEP 1011 (2010) 077,
arXiv:1007.2580 [hep-lat].
[55] T. Bhattacharya, M. I. Buchoff, N. H. Christ, H. T. Ding, R. Gupta, et al.,
“The QCD phase transition with physical-mass, chiral quarks,”
Phys.Rev.Lett. 113 (2014) 082001, arXiv:1402.5175 [hep-lat].
[56] M. Chojnacki and W. Florkowski, “Temperature dependence of sound
velocity and hydrodynamics of ultra-relativistic heavy-ion collisions,” Acta
Phys.Polon. B38 (2007) 3249–3262, arXiv:nucl-th/0702030 [NUCL-TH].
[57] W. Broniowski, M. Chojnacki, W. Florkowski, and A. Kisiel, “Uniform
Description of Soft Observables in Heavy-Ion Collisions at s(NN)**(1/2) =
200 GeV**2,” Phys.Rev.Lett. 101 (2008) 022301,
arXiv:0801.4361 [nucl-th].
[58] W. Florkowski, “The realistic QCD equation of state in relativistic heavy-ion
collisions and the early Universe,” Nucl.Phys. A853 (2011) 173–188,
arXiv:1008.5225 [nucl-th].
[59] X.-N. Wang and M. Gyulassy, “Energy and centrality dependence of rapidity
densities at RHIC,” Phys.Rev.Lett. 86 (2001) 3496–3499,
arXiv:nucl-th/0008014 [nucl-th].
[60] D. Kharzeev and M. Nardi, “Hadron production in nuclear collisions at
RHIC and high density QCD,” Phys.Lett. B507 (2001) 121–128,
arXiv:nucl-th/0012025 [nucl-th].
[61] W. Israel and J. Stewart, “Transient relativistic thermodynamics and kinetic
theory,” Annals Phys. 118 (1979) 341–372.
[62] J. Noronha-Hostler, G. S. Denicol, J. Noronha, R. P. G. Andrade, and
F. Grassi, “Bulk Viscosity Effects in Event-by-Event Relativistic
Hydrodynamics,” Phys.Rev. C88 (2013) 044916,
arXiv:1305.1981 [nucl-th].
[63] M. Strickland, “Anisotropic Hydrodynamics: Three lectures,”
arXiv:1410.5786 [nucl-th].
[64] W. Florkowski and R. Ryblewski, “Highly-anisotropic and
strongly-dissipative hydrodynamics for early stages of relativistic heavy-ion
collisions,” Phys.Rev. C83 (2011) 034907, arXiv:1007.0130 [nucl-th].
[65] M. Martinez and M. Strickland, “Dissipative Dynamics of Highly Anisotropic
Systems,” Nucl.Phys. A848 (2010) 183–197, arXiv:1007.0889 [nucl-th].
[66] C. Hung and E. V. Shuryak, “Equation of state, radial flow and freezeout in
high-energy heavy ion collisions,” Phys.Rev. C57 (1998) 1891–1906,
arXiv:hep-ph/9709264 [hep-ph].
[67] P. Koch and J. Rafelski, “Why the Hadronic Gas Description of Hadronic
Reactions Works: The Example of Strange Hadrons,” South Afr.J.Phys. 9
(1986) 8.
26 wf printed on October 30, 2014

[68] J. Cleymans and H. Satz, “Thermal hadron production in high-energy heavy

ion collisions,” Z.Phys. C57 (1993) 135–148,
arXiv:hep-ph/9207204 [hep-ph].
[69] M. Gazdzicki and M. I. Gorenstein, “On the early stage of nucleus-nucleus
collisions,” Acta Phys.Polon. B30 (1999) 2705,
arXiv:hep-ph/9803462 [hep-ph].
[70] F. Becattini, J. Cleymans, A. Keranen, E. Suhonen, and K. Redlich,
“Features of particle multiplicities and strangeness production in central
heavy ion collisions between 1.7A-GeV/c and 158A-GeV/c,”
Phys.Rev. C64 (2001) 024901, arXiv:hep-ph/0002267 [hep-ph].
[71] P. Braun-Munzinger, D. Magestro, K. Redlich, and J. Stachel, “Hadron
production in Au - Au collisions at RHIC,” Phys.Lett. B518 (2001) 41–46,
arXiv:hep-ph/0105229 [hep-ph].
[72] W. Florkowski, W. Broniowski, and M. Michalec, “Thermal analysis of
particle ratios and p(t) spectra at RHIC,” Acta Phys.Polon. B33 (2002)
761–769, arXiv:nucl-th/0106009 [nucl-th].
[73] G. Torrieri, S. Steinke, W. Broniowski, W. Florkowski, J. Letessier, et al.,
“SHARE: Statistical hadronization with resonances,”
Comput.Phys.Commun. 167 (2005) 229–251,
arXiv:nucl-th/0404083 [nucl-th].
[74] M. Petran, J. Letessier, V. Petracek, and J. Rafelski, “Hadron√production
and quark-gluon plasma hadronization in Pb-Pb collisions at sN N = 2.76
TeV,” Phys.Rev. C88 no. 3, (2013) 034907, arXiv:1303.2098 [hep-ph].
[75] J. Stachel, A. Andronic, P. Braun-Munzinger, and K. Redlich, “Confronting
LHC data with the statistical hadronization model,”
J.Phys.Conf.Ser. 509 (2014) 012019, arXiv:1311.4662 [nucl-th].
[76] M. Floris, “Hadron yields and the phase diagram of strongly interacting
matter,” arXiv:1408.6403 [nucl-ex].
[77] W. Broniowski and W. Florkowski, “Explanation of the RHIC p(T) spectra
in a thermal model with expansion,” Phys.Rev.Lett. 87 (2001) 272302,
arXiv:nucl-th/0106050 [nucl-th].
[78] A. Kisiel, T. Taluc, W. Broniowski, and W. Florkowski, “THERMINATOR:
THERMal heavy-IoN generATOR,”
Comput.Phys.Commun. 174 (2006) 669–687,
arXiv:nucl-th/0504047 [nucl-th].
[79] M. Chojnacki, A. Kisiel, W. Florkowski, and W. Broniowski,
“THERMINATOR 2: THERMal heavy IoN generATOR 2,”
Comput.Phys.Commun. 183 (2012) 746–773, arXiv:1102.0273 [nucl-th].
[80] ALICE Collaboration Collaboration, B. Abelev √ et al., “Pion, Kaon, and
Proton Production in Central Pb–Pb Collisions at sN N = 2.76 TeV,”
Phys.Rev.Lett. 109 (2012) 252301, arXiv:1208.1974 [hep-ex].
[81] ALICE Collaboration Collaboration, B. Abelev et al.,√“Centrality
dependence of π, K, p production in Pb-Pb collisions at sN N = 2.76 TeV,”
Phys.Rev. C88 no. 4, (2013) 044910, arXiv:1303.0737 [hep-ex].
 wf printed on October 30, 2014 27

[82] V. Begun, W. Florkowski, and M. Rybczynski, “Explanation of hadron

transverse-momentum spectra in heavy-ion collisions at sqrt(sNN) = 2.76
TeV within chemical non-equilibrium statistical hadronization model,”
Phys.Rev. C90 (2014) 014906, arXiv:1312.1487 [nucl-th].
[83] V. Begun, W. Florkowski, and M. Rybczynski, “Transverse-momentum
√
spectra of strange particles produced in Pb+Pb collisions at sNN = 2.76
TeV in the chemical non-equilibrium model,” arXiv:1405.7252 [hep-ph].
[84] H. Petersen, J. Steinheimer, G. Burau, M. Bleicher, and H. Stocker, “A Fully
Integrated Transport Approach to Heavy Ion Reactions with an Intermediate
Hydrodynamic Stage,” Phys.Rev. C78 (2008) 044901,
arXiv:0806.1695 [nucl-th].