Using the non-hydrodynamic mode to study the onset of hydrodynamic behavior in ultraperipheral symmetric nuclear collisions

Abstract

With the attempts of extending the hydrodynamic framework of heavy-ion collision to proton-proton and other small and low energy systems, we are confronted with the question of how small the system can get and still be safely modelled as a fluid. One of the transport coefficients required in the $2^{nd}$ order relativistic viscous hydrodynamics is the shear relaxation time, inclusion of which solves the causality violation problem in the Navier-Stokes equation. In phenomenological studies this coefficient has been taken as a constant and much attention has gone into finding and fixing the shear viscosity to entropy density ratio, $\eta/s$. This transport coefficient also happens to control the non-hydrodynamic mode of the out-of-equilibrium hydrodynamics theory. It has been predicted that for decreasing system size, observables become sensitive to variation in shear relaxation time as a result of increasing dominance of non-hydrodynamic mode, which could potentially indicate breakdown of hydrodynamics. In this study, we try to test this prediction in the peripheral Pb-Pb collisions at $2.76$ TeV and Au-Au collisions at $200$ GeV, with IPGlasma initial condition and $(2+1)-$Dimensional viscous hydrodynamics. We find that elliptic flow does show adequate sensitivity to variation in relaxation time for decreasing system size. The multiplicity rapidity density limit for applicability of hydrodynamics is found to be around $dN/dy\approx10$, with the possibility of refinement in this value given a way to improve the centrality resolution in experimental data for referencing in peripheral collisions.