Fluid Dynamics In Relativistic Heavy Ion Collisions

Quantum Chromodynamics (QCD) is the fundamental theory for strong interac-tion, one of the four forces in nature. Different from electromagnetic interaction, due to its non-Abelian color symmetry, the QCD has the property of asymptotic freedom at large momentum transfer, while remains strongly coupled at low energies. This prop-erty leads to color confinement, i.e. there are no free quarks and gluons carrying color degree of freedom, and quarks and gluons are confined inside hadrons. In 1974-75, Lee and Collins-Perry suggested that the deconfinement can be reached through the ultra-relativistic heavy ion collisions, where the vacuum can be excited to a new state of matter or a quark gluon plasma. In the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Lab (BNL), the gold nuclei are accelerated and collide head-to-head at the center-of-mass energy of 200 GeV per nucleon. After the collisions, the huge amount of energy is deposited in the central rapidity region to excite vacuum and produce many quarks and gluons. In very short time, these quarks and gluons collide to each other and the fireball reaches the local thermal equilibrium, and a quark-gluon-plasma (QGP) is then formed. After expansion and cooling, the temperature drops sub-stantially and the quarks recombine into hadrons which are finally observed in detectors. The collective flows such as radial and elliptic flows are observed at RHIC and can be well described by ideal fluid dynamics. Further study of the elliptic flow using dissi-pative fluid dynamics gives very low values of the ratio of shear viscosity to entropy density, close to the lower bound 1/4Ï€by AdS/CFT correspondence. This is one of the most surprising observation at RHIC:the QGP is strongly coupled or a nearly perfect liquid instead of a gas-like weakly coupled system by convention. The relativistic fluid dynamics provides a useful tool to bridge the gap between the initial state of quarks and gluons and the experimental data. Pushed by heavy ion collision experiments, the theory of relativistic fluid dynamics also has a lot of new development in recent years. This thesis is about the study of three important issues in the theory of relativistic fluid dynamics:the stability of dissipative fluid dynamics, the AdS/CFT application to shear viscosity in a Bjorken expanding fluid, and a consistent description of kinetic equation with triangle anomaly.Hydrodynamics is a long-wavelength effective theory for many particle system-s. The basic hydrodynamic equation consists of conservation equations of energy- momentum and charge numbers. One can carry out gradient expansion for energy-momentum tensor and charge currents, or equivalently expansion in powers of the Knudsen number. The Knudsen number is defined as the ratio of the macroscopic length scale (hydrodynamic wave-length) to the microscopic one (mean free path). When the macroscopic length scale is much larger than the microscopic one, the fluid dynamics is a good effective theory. In expansion, the zeroth order corresponds to the ideal fluid. The first order gives the Navier-Stokes equations with dissipative terms like shear and bulk viscous terms. There are a lot of candidates for the second order theory. We fo-cus on the widely used Israel-Stewart (IS) theory including the simplified and complete version. The simplified version can be obtained by correspondence to the macroscopic phenomena. To derive the complete version the kinetic theory or the Boltzmann equa-tion is necessary. In this thesis, we will give the details as how to obtain the complete version. The connection between the transport coefficients in the first and the second order theories will be demonstrated.In the first order theory, causality is violated since it takes no time for a system in a non-equilibrium state to reach equilibrium, i.e. the propagating speed of the signal is arbitrarily large. In the second order theory, due to finite relaxation time introduced, it takes finite time for the system to reach equilibrium. The propagating speed of the signal is limited. Therefore the second order theory is necessary for causality. However the causality cannot be guaranteed for all parameters. The constraints for parameters are then given. We also point out that the causality and the stability are inter-correlated. Relativity requires that the signal propagate in light-cones. An acausal propagating modes must be forbidden by instability or singularity. The connection between causality and stability is also discussed. For convenience we work in a general boost frame. It is found that a causal system must be stable, but an acausal system in the boost frame at high speed must be unstable.The transport coefficients can be determined in kinetic theory. There are two main techniques to compute the transport coefficients. One is to employ the Boltzmann equa-tion, the other is to use the Kubo formula in field theory. We will firstly discuss about derivation of the shear viscosity via variational method in the Boltzmann equation. Sec-ondly, after a short review of Kubo formula, we will compute the shear viscosity via AdS/CFT duality.Different from the work of Policastro, Son and Starinets (PSS), we focus on strong-ly coupled QGP (sQGP) with the radial expansion and Bjorken boost invariance. The information of the sQGP is encoded on the boundary of AdS space via the holographic renormalization. Solving the Einstein equation with the boundary condition given by the stress tensor yields the metric of the AdS space. In this case, the metric is Bjorken boost invariant and has the radial flow. The evolution of the shear viscosity as a function of proper time can be obtained via the Kubo formula for the retarded Green function given by the AdS/CFT duality. It is found that the ratio of the shear viscosity to entropy density is consistent with PSS.As another application of AdS/CFT duality, we investigate the property of baryon-s in sQGP in a Wilson-loops-like model. The quarks located at the boundary of AdS space are connected to a probe D5 brane by superstrings. By studying the configura-tions of baryons with different spins, the screening length of baryons can be obtained as a function of spin and temperature. We also study the relationship between the angu-lar momentum and energy for different kinds of baryons, which shows the Regge-like behavior, i.e. the total angular momentum is proportional to the energy squared.As the last topic, we investigate the fluid dynamics with quantum triangle anoma-lies. Generally the relativistic fluid dynamics does not allow the vorticity due to parity conservation. Recently it is pointed out that the vorticity has to be introduced to rel-ativistic fluid dynamics with anomalies to satisfy the second law of thermodynamics. These new terms are also relevant to the Chiral Magnetic Effect (CME) or Chiral Vor-ticity Effect (CVE). Such terms can be derived from the kinetic approach. The coef-ficients of the vorticity in the case of right-handed quarks (or left-handed anti-quarks) and quarks-antiquarks of mixed chirality are evaluated.