David L. Winter for the PHENIX Collaboration High-p T Particle Production with Respect to the Reaction Plane Winter Workshop on Nuclear Dynamics La Jolla, CA 12 March – 18 March 2006

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane2 Outline Physics Motivation Measurement Method PHENIX Results Models Summary

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane3 R AA and v 2 at high p T PHENIX preliminary  0 R AA appears flat for p T >3.0 GeV/c Large v 2 at high p T Examples of theoretical studies: Gyulassy, Vitev, Wang, PRL 86: 2537, 2001 Shuryak, Phys. Rev. C: (2002) Drees, Feng, Jia, Phys. Rev. C: (2005)

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane4 Physics Motivation How well do we understand the origin of azimuthal anisotropy (v 2 ) at high p T ? The “usual” explanations –Arises from azimuthal variation in energy loss –Which is in turn due to geometry: spatial anisotropy of parton density in non-central collisions.  0 s provide ideal “laboratory” to probe this physics at high p T : –Expected to be less subject to effects of recombination –High p T acceptance Studying anisotropy out to high p T provides powerful tool for studying transition from soft to hard physics at p T >~ 3 GeV/c

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane5 The PHENIX Detector Pioneering High-Energy Nuclear Interaction eXperiment Two central arms for measuring hadrons, photons and electrons Two forward arms for measuring muons Event characterization detectors in center PHENIX (image ca. Jan 1999)

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane6 Measuring  0 s in PHENIX In Run 4, PHENIX recorded 1.5 Billion AuAu Collisions –Data presented here represents ~ 1B of those events For measuring ,  0 s, we have 8 EmCal sectors Two technologies –PbSc: Sampling –PbGl: Cerenkov 00 In a given p T, centrality, and reaction plane bin, we: –Form pairs of clusters –Subtract mixed event background –Integrate counts in mass window –(Determined by fit to Gaussian) 

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane7 Measuring the R.P. in PHENIX Use the Beam-Beam 3<|  |<4, azimuthally symmetric Measure charged particle multiplicity as function of  Event-by-event determination 2 independent measurements from north and south counters – estimates resolution x y z Reaction Plane Reaction plane

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane8 Reaction Plane Biases? –Can hard scattering bias the reaction plane measurement ? –Evaluate using Pythia: Calculate  between pions in (central arm) And charged particles in (BBC) For different pion p T bins.  dn/d  d  3<  <4 Au-Au dn/d  *2v % 13  % 12  % 9  % 5.8  1.3  Much larger than hard scattering correlation. Black:p T > 2 Red:p T > 4 Blue:p T > 10

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane9 Relative Yields wrt Reaction Plane Measure  0 dN/dp T in 6  bins over [0,  /2]. –Correct yields for reaction plane resolution Multiply the ratio With previously measured R AA  R AA (  ) x y z Reaction Plane

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane10 But first… accounting for the detector Reaction Plane as measured has resolution  Fit raw yield(  )  raw v 2 Correct raw v 2 for resolution Correct raw yield(  ) with Measure RP with Beam-Beam Counters 20-30% 2<pt<2.5

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane11 From Relative Yields to R AA Multiply By inclusive R AA Red: Sys. Due to resolution correction Blue: Error on R AA 0  /2

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane12 Centrality dependence of  0 R AA

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane13 R AA ( ,p T ) vs. p T (Cent. Dependence) Grey bands: Error in R AA In-plane Out-of-plane

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane14 R AA ( ,p T ) vs. N part In-plane Out-of-plane Grey Bands: Inclusive R AA w/ Error

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane15  0 v 2 Red: Sys. error (abs) Large v 2 at high p T !

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane16 Compare with charged hadrons

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane17 Energy Loss and Path Length Suppose energy loss is dominant mechanism at high p T These two (coupled) parameters both give handles on the parton’s path length through medium: –Centrality –Angle with respect to Reaction Plane Can we find an equivalent single parameter? –Have to depend on density of partons,  –Need to include time-dependence/formation time in 

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane18 Geometry and “Canonical” Energy Loss Initial parton (areal) density Intrinsic energy loss: Assume: Calculate: Further refine with Glauber MC sampling of path origin to take into account fluctuations in hard-scattering center This quantity should contain all geometric effects, and therefore  E should be proportional it. Participant Density Y X

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane19 R AA vs. “  L dL” % Centrality % Centrality % Centrality % Centrality % Centrality Angular and centrality dependence described by single curve!!

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane20 Calculations based on Arnold, Moore, Yaffe (AMY) formalism –JHEP 0305: Energy loss only (BDMS++) High-p T –v 2 appears to decrease to energy loss calculation Low(er)-p T –Something additional going on… While the data appear to approach the energy loss limit at high p T, there is something extra going on in 3-6 GeV/c region  0 v 2 Theory Comparison: AMY (Turbide et al.)

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane21  0 v 2 Theory Comparison: D.Molnar Molnar Parton Cascade (MPC) –nucl-th/ Contains: –Corona effects –Energy loss due to interactions –p T boost due to interactions Consistency would suggest: –QGP? –sQGP? Model shown here is for one set of parameters –Can larger opacity reproduce the v 2 ? High-p T “slopes” consistent

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane22 Summary While R AA (p T ) appears to be flat out to high p T, R AA ( ,p T ) reveals both p T - and angle- dependent substructure For the first time we see a clear decrease in the  0 v 2 at high p T – to a non-zero value! Non-zero high-p T v 2 is consistent with energy loss calculations Comparison of high-p T (>7.0 GeV/c) behavior of v 2 with models points to pQCD + energy loss as dominant sources –What’s responsible for larger v 2 at intermediate p T ? Partons pushed to higher p T (à la Molnar)? Larger energy loss crossing the flow field (Wiedemann et al)? Collisional energy loss? Flow + recombination?

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane23 Backups

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane24 “Zooming in” on MinBias h ± Minimum-Bias  s=200 Au+Au

13 March 2006D.Winter: High-Pt Production wrt Reaction Plane25 Goal: Combine centrality and angle dependence into one geometric picture First approach: calculate a simple length (L) assuming an elliptical shape Next level: include variation in density using a Glauber model and plot R AA (  ) vs.  0   L/  dL Even Better: include fluctuations in L within Glauber Model and plot R AA (  ) vs.  0   L eff /  dL (with the effective length L eff ) Geometric pictures and path length