Paul Romatschke

Relativistic viscous hydrodynamics, conformal invariance, and holography

We consider second-order viscous hydrodynamics in conformal field theories at finite temperature. We show that conformal invariance imposes powerful constraints on the form of the second-order corrections. By matching to the AdS/CFT calculations of correlators, and to recent results for Bjorken flow obtained by Heller and Janik, we find three (out of five) second-order transport coefficients in the strongly coupled = 4 supersymmetric Yang-Mills theory. We also discuss how these new coefficents can arise within the kinetic theory of weakly coupled conformal plasmas. We point out that the Muller-Israel-Stewart theory, often used in numerical simulations, does not contain all allowed second-order terms and, frequently, terms required by conformal invariance.

Viscosity Information from Relativistic Nuclear Collisions: How Perfect is the Fluid Observed at RHIC?

Relativistic viscous hydrodynamic fits to Brookhaven Relativistic Heavy Ion Collider data on the centrality dependence of multiplicity, transverse, and elliptic flow for square root s = 200 GeV Au+Au collisions are presented. For standard (Glauber-type) initial conditions, while data on the integrated elliptic flow coefficient v(2) are consistent with a ratio of viscosity over entropy density up to eta/s approximately 0.16, data on minimum bias v(2) seem to favor a much smaller viscosity over entropy ratio, below the bound from the anti-de Sitter conformal field theory conjecture. Some caveats on this result are discussed.

New Developments in Relativistic Viscous Hydrodynamics

Starting with a brief introduction into the basics of relativistic fluid dynamics, I discuss our current knowledge of a relativistic theory of fluid dynamics in the presence of (mostly shear) viscosity. Derivations based on the generalized second law of thermodynamics, kinetic theory, and a complete second-order gradient expansion are reviewed. The resulting fluid dynamic equations are shown to be consistent for all these derivations, when properly accounting for the respective region of applicability, and can be applied to both weakly and strongly coupled systems. In its modern formulation, relativistic viscous hydrodynamics can directly be solved numerically. This has been useful for the problem of ultrarelativistic heavy-ion collisions, and I will review the setup and results of a hydrodynamic description of experimental data for this case.

Collective non-Abelian instabilities in a melting color glass condensate.

We present first results for (3 + 1)D simulations of SU(2) Yang-Mills equations for matter expanding into the vacuum after a heavy ion collision. Violations of boost invariance cause a non-Abelian Weibel instability leading soft modes to grow with proper time tau as exp(gamma square root(g2 mu tau)), where g2 mu is a scale arising from the saturation of gluons in the nuclear wave function. The scale for the growth rate gamma is set by a plasmon mass, defined as omega(pl) = kappa0 square root(g2 mu/tau)), generated dynamically in the collision. We compare the numerical ratio gamma/kappa0 to the corresponding value predicted by the hard thermal loop formalism for anisotropic plasmas.

Relativistic Viscous Fluid Dynamics and Non-Equilibrium Entropy

Fluid dynamics corresponds to the dynamics of a substance in the long wavelength limit. Writing down all terms in a gradient (long wavelength) expansion up to second order for a relativistic system at vanishing charge density, one obtains the most general (causal) equations of motion for a fluid in the presence of shear and bulk viscosity, as well as the structure of the non-equilibrium entropy current. Requiring positivity of the divergence of the non-equilibrium entropy current relates some of its coefficients to those entering the equations of motion. I comment on possible applications of these results for conformal and non-conformal fluids.

On the collision of two shock waves in AdS5

We consider two ultrarelativistic shock waves propagating and colliding in five-dimensional Anti-de-Sitter spacetime. By transforming to Rosen coordinates, we are able to find the form of the metric shortly after the collision. Using holographic renormalization, we calculate the energy-momentum tensor on the boundary of AdS space for early times after the collision. Via the gauge-gravity duality, this gives some insights on bulk dynamics of systems created by high energy scattering in strongly coupled gauge theories. We find that Bjorken boost-invariance is explicitely violated at early times and we obtain an estimate for the thermalization time in this simple system.

The unstable glasma

We discuss results from 3+1-D numerical simulations of SU(2) Yang-Mills equations for an unstable glasma expanding into the vacuum after a high energy heavy-ion collision. We expand on our earlier work on a non-Abelian Weibel instability in such a system and study the behavior of the instability in greater detail on significantly larger lattices than previously. We establish the time scale for the onset of the instability and demonstrate that the growth rate is robust as one approaches the continuum limit. For large violations of boost invariance, non-Abelian effects cause the growth of soft modes to saturate. At late times, we observe significant creation of longitudinal pressure and a systematic trend towards isotropy. These time scales however are significantly larger than those required for early thermalization in heavy-ion collisions. We discuss additional effects in the produced glasma that may speed up thermalization.

Hard-loop dynamics of non-abelian plasma instabilities

I discuss recent advances in the understanding of non-equilibrium gauge field dynamics in plasmas which have particle distributions which are locally anisotropic in momentum space. In contrast to locally isotropic plasmas such anisotropic plasmas have a spectrum of soft unstable modes which are characterized by exponential growth of transverse (chromo)-magnetic fields at short times. The long-time behavior of such instabilities depends on whether or not the gauge group is abelian or non-abelian. Here I will report on recent numerical simulations which attempt to determine the long-time behavior of an anisotropic non-abelian plasma within hard-loop effective theory.

Cold quark matter

We perform an O({alpha}{sub s}{sup 2}) perturbative calculation of the equation of state of cold but dense QCD matter with two massless and one massive quark flavor, finding that perturbation theory converges reasonably well for quark chemical potentials above 1 GeV. Using a running coupling constant and strange quark mass, and allowing for further nonperturbative effects, our results point to a narrow range where absolutely stable strange quark matter may exist. Absent stable strange quark matter, our findings suggest that quark matter in (slowly rotating) compact star cores becomes confined to hadrons only slightly above the density of atomic nuclei. Finally, we show that equations of state including quark matter lead to hybrid star masses up to M{approx}2M{sub {center_dot},} in agreement with current observations. For strange stars, we find maximal masses of M{approx}2.75M{sub {center_dot}}and conclude that confirmed observations of compact stars with M>2M{sub {center_dot}}would strongly favor the existence of stable strange quark matter.

Dissipative hydrodynamics and heavy-ion collisions

Recent discussions of RHIC data emphasized the exciting possibility that the matter produced in nucleus-nucleus collisions shows properties of a near-perfect fluid. Here, we aim at delineating the applicability of fluid dynamics, which is needed to quantify the size of corresponding dissipative effects. We start from the equations for dissipative fluid dynamics, which we derive from kinetic theory up to second order (Israel-Stewart theory) in a systematic gradient expansion. In model studies, we then establish that for too early initialization of the hydrodynamic evolution ({tau}{sub 0} < or approx. 1 fm/c) or for too high transverse momentum (p{sub T} < or approx. 1 GeV) in the final state, the expected dissipative corrections are too large for a fluid description to be reliable. Moreover, viscosity-induced modifications of hadronic transverse momentum spectra can be accommodated to a significant degree in an ideal fluid description by modifications of the decoupling stage. We argue that these conclusions, drawn from model studies, can also be expected to arise in significantly more complex, realistic fluid dynamics simulations of heavy ion collisions.