Heavy quarks may be produced in the earliest stage of ultra-relativistic heavy-ion collisions, and may or may not bind into quarkonia. They probe the full evolution of the strongly-coupled medium created in these collisions.
The kinetic thermalization of unbound heavy quarks is understood in terms of the heavy-quark diffusion coefficient. Recently, first results with a non-trivial quark sea were made possible by the use of gradient flow. These results covered temperatures of 195 MeV < T < 352 MeV and included the dependence on the heavy-quark mass. As the light sea quarks become more important near the crossover, we extend these studies with almost physical quark sea to lower temperatures and with the same unphysical quark sea to higher temperatures, thus mapping out the quark sea dependence in the temperature window from around 𝑇 ≃ 150 MeV to the GeV level. At such high temperatures appropriately resummed weak-coupling methods have become predictive, whereas the constraining power of the lattice approach has ceased.
In-medium quarkonia are subject to a dynamical melting process, which can be understood in terms of the static potential or of low-lying quarkonia levels. EFT calculations predict a complex potential, whose imaginary part corresponds to the thermal width of these levels. The real part is Debye screened only in certain hierarchies between thermal and non-relativistic scales. Yet these hierarchies apply in the large-time limit and thus are irrelevant for the dynamical melting. A recent calculation in (2+1)-flavor QCD at T < 352 MeV reveals a large imaginary part and provides no evidence for Debye screening. While another recent study in the quenched approximation is consistent with these results. Direct lattice calculations of bottomonia correlators support these conclusions, either of temporal correlators via lattice non-relativistic QCD or of spatial correlators via highly-improved staggered quarks.