CERN Accelerating science

Above temperatures of 150 MeV, nuclear matter transitions into the quark-gluon plasma (QGP): a phase of unbound quarks and gluons. These conditions are reached in heavy-ion collisions at center-of-mass energies per nucleon-nucleon pair ($\sqrt{s_\mathrm{NN}}$) in the TeV scale that can generate energy densities larger than $10\ \mathrm{GeV/fm^3}$. The spatial distribution of this energy originates from the fluctuating shape of the overlap of nuclei in the initial state. On a timescale of $10\ \mathrm{fm}/c$, the QGP, a near-perfect fluid, transforms the spatial anisotropy into a momentum anisotropy of the emitted particles called anisotropic flow. The comparison of such observations to hydrodynamic model calculations allows extracting the QGP viscosity. The spectator nucleons — remnants of the colliding nuclei which rapidly decouple ($\ll 1\ \mathrm{fm}/c$) before the anisotropic flow emerges — are sensitive to initial-state fluctuations. This thesis presents novel measurements of anisotropic flow and its fluctuations relative to the spectator deflection in lead-lead and xenon-xenon collisions at $\sqrt{s_\mathrm{NN}}$ of 2.76 TeV and 5.44 TeV, respectively, with ALICE at the Large Hadron Collider. These observations show an approximate universal scaling with the shape of the initial energy density. Differences between the flow measurements using spectators and those only using produced particles constrain the initial-state fluctuations. Comparisons to current initial state models without spectator dynamics indicate these dynamics are needed for improving the precision of the QGP viscosity extraction.