Collective flow in event-by-event partonic transport plus hydrodynamics hybrid approach

TL;DR

This paper presents a hybrid model combining partonic transport and viscous hydrodynamics to study collective flow in ultrarelativistic heavy-ion collisions, successfully explaining experimental data and revealing the role of initial fluctuations.

Contribution

It introduces a coupled Boltzmann and hydrodynamics approach with event-by-event fluctuations, providing new insights into the generation of flow harmonics and initial eccentricities.

Findings

Model reproduces v_n(p_T) data from RHIC and LHC.

Eccentricities epsilon_n are generated by initial fluctuations.

Flow coefficients v_n are insensitive to switching time variations.

Abstract

Complete evolution of the strongly interacting matter formed in ultrarelativistic heavy-ion collisions is studied within a coupled Boltzmann and relativistic viscous hydrodynamics approach. For the initial nonequilibrium evolution phase, we employ the AMPT model that explicitly includes event-by-event fluctuations in the number and positions of the participating nucleons as well as of the produced partons with subsequent parton transport. The ensuing near-equilibrium evolution of quark-gluon and hadronic matter is modeled within the (2+1)-dimensional viscous hydrodynamics. We probe the role of parton dynamics in generating and maintaining the spatial anisotropy in the preequilibrium phase. Substantial eccentricities epsilon_n are found to be generated in the event-by-event fluctuations in parton production from initial nucleon-nucleon collisions. For ultracentral heavy-ion collisions, the model is able to explain qualitatively the unexpected hierarchy of the harmonic flow coefficients v_n(p_T)(n=2-6) observed at LHC. We find that the results for v_n(p_T) are rather insensitive to the variation (within a range) of the time of switchover from AMPT parton transport to hydrodynamic evolution. The usual Grad and the recently proposed Chapman-Enskog-like (nonequilibrium) single-particle distribution functions are found to give very similar results for v_n(n=2-4). The model describes well both the RHIC and LHC data for v_n(p_T) at various centralities, with a constant shear viscosity to entropy density ratio of 0.08 and 0.12, respectively. The event-by-event distributions of v_{2,3} are in good agreement with the LHC data for midcentral collisions. The linear response relation v_n = k_n epsilon_n is found to be true for n=2,3, except at large values of epsilon_n, where a larger value of k_n is required, suggesting a small admixture of positive nonlinear response even for n=2,3.