Starting from the Maxwell-Jüttner equilibrium distribution, we develop a relativistic lattice Boltzmann (LB) algorithm capable of handling ultrarelativistic systems with flat, but expanding, spacetimes. The algorithm is validated through simulations of quark-gluon plasma, yielding excellent agreement with hydrodynamic simulations. The present scheme opens the possibility of transferring the recognized computational advantages of lattice kinetic theory to the context of both weakly and ultra-relativistic systems.PACS numbers: 47.75.+f, 47.11.-j Keywords: Relativistic fluid dynamics, Quark-gluon plasmas, Lattice Boltzmann
I. MOTIVATIONThe great success of the Relativistic Heavy-Ion Collider (RHIC) experimental program [1][2][3][4] has provided the motivation to come up with realistic and quantitative simulations of heavy-ion collisions.Since the bulk of particles produced in relativistic heavy-ion collisions is described by fluid dynamics [5], the center-piece of any complete simulation attempt will involve a viscous fluid dynamics algorithm. The majority of presently available fluid dynamics codes is able to handle smooth initial conditions in 2+1 dimensions in the presence of shear viscosity [6][7][8][9][10]. However, it has by now been understood that the presence of event-by-event fluctuations in the initial state can lead to significantly different quantitative results with respect to smooth initial conditions [11], and may in some cases even explain qualitatively new phenomena. To be more specific, the presence of event-by-event fluctuations is the source of the non-vanishing elliptic flow found in RHIC experiments at central collisions, the source of hydrodynamic flow-fluctuations, and may (through the presence of socalled triangular flow v 3 ) naturally explain the presence of the 'ridge phenomenon' found in experiments [12][13][14][15]. Thus, it is probably fair to say that without including the effect of event-by-event fluctuations, a description of the medium created in heavy-ion collisions cannot be regarded as realistic. This provides the motivation to develop a fully relativistic and computationally efficient viscous fluid dynamics algorithm that can handle initial state fluctuations. Also, such an algorithm can be used Further motivation is provided by other systems where relativistic viscous fluid flows are of interest, such as astrophysical systems and condensed matter systems such as graphene [17]. One particular question that arises in all this different systems is when relativistic fluid flow becomes turbulent, which involves a determination of the critical Reynolds number and the turbulent spectrum [18,19].
II. LATTICE KINETIC APPROACH TO HYDRODYNAMICSFluid turbulence, both classical and relativistic, sets one of the most compelling challenges in modern computational physics. This motivates a relentless search for new and ever more efficient methods for solving the hydrodynamic equations of motion in the high-Reynolds turbulent regimes. In the last two decades, a new computational paradigm has...