From Glasma to Plasma in Heavy Ion Collisions

From Glasma to Plasma in Heavy Ion Collisions Raju Venugopalan Brookhaven National Laboratory Topical Overview Talk, QM 2008, Jaipur, Feb. 4 th, 2008

What is the Glasma ? Ludlam, Mc. Lerran, Physics Today (2003) Glasma (Glahs-maa): Noun: non-equilibrium matter between Color Glass Condensate (CGC) & Quark Gluon Plasma (QGP) 2

Why is the Glasma relevant ? o Intrinsic interest: Glasma fields are among strongest Electric & Magnetic fields in nature. What are their properties ? o Initial conditions for the QGP: Ø How does bulk matter flow in the Glasma influence transport in the perfect fluid ? Ø How do jets interact with the Glasma ? The Glasma is key to quantitative understanding of matter produced in HI collisions 3

Little Bang Big Bang WMAP data (3 x 105 years) Hot Era Inflation QGP CGC/ Glasma Plot by T. Hatsuda 4

Big Bang vs. Little Bang Decaying Inflaton field with occupation # 1/g 2 Decaying Glasma field with occupation # 1/g 2 Explosive amplification of low mom. small fluctuations (preheating) Explosive amplification of low mom. small fluctuations (Weibel instability ? ) Interaction of fluct. /inflaton - thermalization Interaction of fluct. /Glasma - thermalization ? Other common features: topological defects, turbulence ? 5

Before the Little Bang v Nuclear wavefunction at high energies Bremsstrahlung SY + Recombination = Saturation: v Renormalization Group (JIMWLK/BK) equations sum leading logs and high parton densities v Successful CGC phenomenology of HERA e+p; NMC e+A; RHIC d+A & A+A Review: RV, ar. Xiv: 0707. 1867, DIS 2007 6

Hadron wave-fns: universal features CGC Effective Theory = classic fields + strong stochastic sources >> 1 S(QS 2) << 1 T. Ullrich (see talk) -based on Kowalski, Lappi, RV ; PRL 100, 022303 (2008) v Theory developments: running coupling in BK Balitsky; Albacete, Gardi, Kovchegov, Rummukainen, Weigert v Upcoming test: current RHIC d+Au run - eg. , forward di 7 jets(talk by C. Marquet)

How is Glasma formed in a Little Bang ? v Problem: Compute particle production in field theories with strong time dependent sources 8

Glasma dynamics perturbative vs non-perturbative Non-perturbative for questions of interest in this talk strong coupling vs weak coupling Interesting set of issues…not discussed here (talks by Rajagopal and Iancu) 9

Systematic expansion for multiplicity moments (=O(1/g 2) and all orders in (g )n) In QCD, solve Yang-Mills Eqns. for two nuclei Glasma initial conditions from matching classical CGC wave-fns on light cone Kovner, Mc. Lerran, Weigert 10

Numerical Simulations of classical Glasma fields Krasnitz, Nara, RV Lappi (see talk) LO Glasma fields are boost invariant for from extrapolating DIS data to RHIC energies 11

I) RHIC LO Glasma Multiplicity Au-Au mult. at eta=0 Krasnitz, RV Kharzeev, Levin, Nardi II) LHC Pb+Pb at = 0 ≈ 950 - 1350 for Npart = 350 (See Armesto talk for other LHC predictions) Gelis, Stasto, RV 12

Flow in the Glasma (I) • Large initial ET QS & NCGC Nhad consistent with strong isentropic flow. Initial conditions for hydro Hirano, Nara • v 2 (initial eccentricity) Hirano et al. , ; Drescher, Nara Lappi, RV CGC- type initial conditions leave room for larger dissipation (viscosity) in hydro stage ? 13

Flow in the Glasma (II) Partial thermalization and v 2 fluctuations: Bhalerao, Borghini, Blaizot, Ollitrault Knudsen # K = /R with 1/K = c. S d. N/dy /area Partial thermalization fit suggests CGC gives lower v 2 than Glauber Drescher, Dumitru, Gombeaud, Ollitrault K 0 14

Flow in the Glasma (III) o What’s the “pre-thermal” flow generated in the Glasma ? Classical field / Particle f<1 Glasma flow important for quantifying viscosity of s. QGP Glasma v 2 Krasnitz, Nara, RV: PLB 554, 21 (2003) 15

The unstable Glasma (I) Kharzeev, Krasnitz, RV Lappi, Mc. Lerran LO boost invariant E & B fields: - purely longitudinal for = 0+ - generate small amounts of topological charge p. X, p. Y p. Z Such configurations may lead to very anisotropic mom. dists. Weibel instability (see C. Greiner’s talk) 16

The unstable Glasma (II) • Small rapidity dependent quantum fluctuations of the LO Yang-Mills fields grow rapidly as • E and B fields as large as EL and BL at time Romatschke, RV: PRL, PRD(2006) increasing seed size 2500 17

The unstable Glasma (III) Romatschke, RV Frequency of maximally unstable k mode grows rapidly Large angle deflections of colored particles in strong fields (Numerical studies by Frankfurt group - C. Greiner talk) 18

Turbulent isotropization on short time scales ? Arnold, Moore; Mueller, Shoshi, Wong; Bödeker, Rummukainen I) Anomalously low viscosity II) Large energy loss of jets in strong fields ? (talks by Majumder and Müller) III) Explosive generation of P and CP odd transitions via sphalerons (see Warringa’s talk) Small fluctuation spectrum ab initio in the Glasma: multiplicity moments to NLO 19

Another example of a small fluctuation spectrum… 20

Multiplicity to NLO (=O(1) in g and all orders in (g )n ) Gelis, RV + Gluon pair production One loop contribution to classical field Initial value problem with retarded boundary conditions - can be solved on a lattice in real time (a la Gelis, Kajantie, Lappi for Fermion pair production) 21

NLO and QCD Factorization Gelis, Lappi, RV What small fluctuations go into wave fn. and what go into particle production ? Small x (JIMWLK) evolution of nucleus A -- sum ( SY)n terms O( S) but may grow as Small x (JIMWLK) evolution of nucleus B ---sum ( SY)n terms 22

From Glasma to Plasma v NLO factorization formula: “Holy Grail” spectrum of small fluctuations. First computations and numerical simulations underway Gelis, Fukushima, Mc. Lerran Gelis, Lappi, RV v With spectrum, can compute T - and match to hydro/kinetic theory 23

Ridgeology* * Rudy Hwa (see talk) + parallel session Near side peak+ ridge (from talk by J. Putschke, STAR collaboration) Jet spectra STAR preliminary Ridge spectra STAR preliminary inclusive 24 pt, assoc, cut

Two particle correlations in the Glasma: variance at LO Gelis, RV: NPA 779 (2006), 177 Glasma sensitive to long range rapidity correlations: Fourier modes of small fluctuation field: O(1) Fourier modes of classical field: O(1/g) (talk by Gelis) 25

Our take on the Ridge Gelis, Lappi, RV i) Long range rapidity correlations built in at early times because Glasma background field is boost invariant. (These are the “beam” jets. ) ii) Rapidity correlations are preserved because matter density dilutes rapidly along the beam direction iii) Opacity effect in : Strong E & B fields destroy azimuthal correlations because survival probability of larger path lengths in radial direction is small (collimation a la Voloshin/Shuryak) iv) May explain why features of the ridge persist for both soft and semi-hard associated particles Need detailed models with realistic geometry effects 26

Conclusions I. Ab initio (NLO) calculations of the initial Glasma in HI collisions are becoming available II. Quantifying how the Glasma thermalizes strongly constrains parameters of the (near) perfect fluid III. Deep connections between QCD factorization and turbulent thermalization IV. Possible explanation of interesting structures from jet+medium interactions 27