1. Introduction
The investigation of matter under extreme conditions in temperature and density as prevailing in the aftermath of the
Big Bang is one of the central aims of present theoretical as well as experimental research efforts in high-energy particle
physics. Nowadays, such an extreme state of matter may occur naturally in the core of compact stellar objects like
neutron stars or during neutron star merger events. In a laboratory environment, extreme temperatures and densities can
be created in relativistic collisions of heavy particles, see for example [1,2,3,4,5] for reviews. Such heavy-ion collisions are
furthermore the only means by which bulk properties of a non-Abelian gauge theory, such as Quantum Chromodynamics
(QCD), can be assessed experimentally. Heavy-ion collision experiments are currently performed at the Large Hadron
Collider (LHC) at CERN, the Relativistic Heavy Ion Collider (RHIC) at BNL, the Schwer-Ionen-Synchrotron (SIS) at
GSI and planned at future facilities such as the Facility for Antiproton and Ion Research (FAIR), the Nuclotron-based
Ion Collider fAcility (NICA), the High Intensity heavy ion Accelerator Facility (HIAF), and the heavy-ion program at
the Japan Proton Accelerator Complex (J-PARC).
Electromagnetic (EM) probes, i.e. photons and dileptons, have proven to be exceptionally versatile and useful probes
to study the properties of the hot and dense medium created in such collisions, see for example [6,7,8,9,10] for
reviews. This is due to the fact that they don’t (directly) interact ‘strongly’ with the surrounding medium, i.e. not via the
strong interaction as described by QCD, but predominantly via the electromagnetic interaction as described by Quantum
Electrodynamics (QED). Since the electromagnetic interaction is considerably weaker than the strong interaction, as for
example evident by comparing the EM coupling strength αEM ≈1/137 with the strong coupling αswhich is of the order
O(10−1)−O(1), photons and dileptons have a mean-free path that is larger than the extent of the created fireball. They
can thus traverse the medium almost undisturbed and carry information from their production point to the detector.
The smallness of the EM coupling, however, also entails that photons and dileptons are produced very rarely compared
to strongly-interacting particles such as pions. For example, the decay of the ρ(770) vector meson into dileptons, i.e. into
an electron-positron pair or into a muon-antimuon pair, is suppressed by a factor of ∼5·10−5as compared to the decay
into pions, see for example the corresponding experimental branching ratios [5].
Another important feature of photons and dileptons is that they are produced at all stages of the collision process. In
principle, they can thus be used to obtain information on all phases of the fireball evolution, from initial hard scattering
processes over the pre-equilibrium phase and the Quark-Gluon Plasma (QGP) phase to the hadron gas phase. This
information is, however, convoluted with the space-time evolution of the medium which makes extracting information on
a particular phase, such as the thermally-equilibrated QGP or the hadron-gas phase, very challenging. A good theoretical
understanding of the underlying dilepton production rates within the various phases as well as of the space-time evolution
of the collision process is therefore imperative for a robust interpretation of photon and dilepton spectra.
In this review, we will focus in particular on the theoretical description and experimental results concerning the
soft thermal radiation from the QGP and the hadron gas phase. Those regimes are of particular interest since they
correspond to the extreme state of matter that filled our Universe shortly after the Big Bang and since they allow to
study fundamental properties of QCD such as confinement and chiral symmetry breaking. Color confinement, which
describes the fact that no color-neutral objects have been observed in an isolated state, is expected to disappear at high
enough temperatures and/or densities. Chiral symmetry, on the other hand, is a symmetry of the QCD Lagrangian for
massless quarks that is spontaneously broken in the vacuum, i.e. at zero temperature and density, but eventually gets
restored at high temperatures and/or densities. Mapping out the corresponding QCD phase diagram is one of the central
3