1. Ultra-relativistic heavy ion collisions and quark-gluon plasma

In ultra-relativistic heavy ion collisions, a high-temperature state of matter is created: the quark and gluon plasma (QGP). This state of matter forms above a temperature of 1.7 10¹⁰ degrees Celsius and was therefore predominant in the universe during the first microseconds after the Big-Bang. Experimental measurements have led to the conclusion that the PQG behaves like a near-perfect fluid ¹. The PQG is today studied in two large particle gas pedals: the "relativistic heavy-ion collider" (RHIC) at Brookhaven National Laboratory in the USA (gold-gold collisions at an energy √ sNN = 200 GeV) and the "large hadron collider" (LHC) at CERN in Geneva (lead-lead collisions at an energy √ sNN = 5000 GeV). However, this picture of the quark and gluon plasma seen as a near perfect fluid living for 10 fm/c is not exclusively based on the results of the current experiments, uses a variety of theoretical models, from different scenarios of the formation of the PQG. However, the "initial state" of heavy ion collisions is precisely characterized thanks to the study of the structure of the proton, which allowed to assert that the energy deposition in the collision zone is largely due to the contribution of gluons (these are the messengers of the strong interaction, one of the four fundamental interactions known in modern physics). An illustration of the modeling of this energy deposition in the initial state is shown in figure 1. However, the way in which these gluons evolve to form a fluid described by hydrodynamics, in an extremely small time span of the order of 1 fm/c = 3.10^-24 seconds, is not understood at all.

¹This fluid has the smallest entropy density normalized viscosity that has ever been experimentally observed. This property implies that the expansion in the hydrodynamic stage is very little affected by friction.

2. An experimental probe of choice: the radiation of matter

How could we experimentally access the transition between this initial state and the hydrody-namic description? To face this challenge, we need to find an observable, i.e. a measurable physical quantity, which allows to characterize this transition stage without being altered after its creation. The electromagnetic radiation seems to be the ideal probe in this respect. Even if its production is low, the probability that this radiation re-interacts with the other products of the collision is low and its transport is almost unperturbed by the nuclear medium through which it passes. This probe has already been used to understand other stages of the collision, but had not yet been used to study the transition phase between the initial state and the hydrodynamic regime. A team of experimental physics researchers from the DPhN of Irfu has thus initiated a collaboration with theorists of heavy ion collision hydrodynamics from the IPhT of CEA Saclay and theorists of the early PQG physics at the University of Bielefeld in the framework of the Gluodynamics collaboration. Among the competing electromagnetic observables, the team identified the study of dilepton production (electron-positron pairs or muon-antimuon² pairs) as a promising probe of the first moments of the collision. Indeed, the production of dileptons is strongly influenced by two key factors that characterize the transition between the initial state and the hydrodynamic regime. First, the dilepton cannot be directly produced by the gluons emerging from the initial state. It requires electric charges, quarks, which are created from the gluons through the strong interaction. Its production is illustrated in Figure 2. Thus, the production of dileptons directly measures the emergence of quarks in the soup of quarks and gluons. The second important factor is the rate at which the initial out-of-equilibrium system "thermalizes" kinematically towards a hydrodynamic description. This equilibration is quantified by the viscosity of the system η normalized by the noted entropy density. Consequently, a measurement of the dilepton production rate directly informs us about these two features of the transition from the initial state to the hydrodynamic stage: the creation of charges and the speed of the transition itself  ³.

^2. Electrons and muons are elementary particles of the standard model of particle physics, positrons and antimuons are the corresponding antiparticles. They are all charged leptons, i.e. particles that participate in the electromagnetic interaction and the weak interaction.

³. This sensitivity to the speed of the transition is possible because we could estimate the final entropy of the collision with the measurement of most of the particles after the collision.

The results for the dilepton production rate shown in Figure 3 show the sensitivity as a function of the viscosity value and the sensitivity to charge creation. The production of dileptons is very different for the viscosity to entropy density ratios η/s=0.16 (red line) and η/s=0.32 (blue line), visible in logarithmic scale. Moreover, the dashed curves show the production under the assumption that quarks are present from the beginning of the collision (at impact). Thus, the difference between the dotted and solid curves shows the effect of the absence of quarks at the first moments of the collision. These simulation results are important because, compared to the experimental results, they should eventually allow to characterize the nature of the creation phase of the PQG. The collaboration has resulted in a first publication in Physics letters B [1] and a second one has been submitted to Nuclear Physics A [2].