J Suppression in Heavy Ion Collisions at RHIC

J/ψ Suppression in Heavy Ion Collisions at RHIC and LHC Matthew Wysocki, University of Colorado October 18, 2011 Seminar at Oak Ridge National Laboratory

What happens to matter at very high temperature? • Construct a phase diagram for QCD matter • “Normal” matter is at low temperature and net baryon density. • Deconfinement at high temperatures: Quark-Gluon Plasma! Early Universe LHC • Access high temperatures at heavy ion colliders. RHIC • Other phases at high baryon density, such as in neutron stars. T (Me. V) 200 nuclei QGP deconfined quarks, gluons Color superconductor Neutron Stars m (Me. V) 10/18/2011 2

Phase Transition on the Lattice Transition to the QGP studied with Lattice QCD calculations. • e ~ (n. d. f. )*T 4 • Evidence of partonic degrees of freedom for the system! • 80% of SB limit != weakly-interacting plasma. 10/18/2011 3

QGP Evolution The QGP will go through several phases as it expands: initial state hadronic phase and freeze-out QGP and hydrodynamic expansion Tc pre-equilibrium Tfo hadronization Probes of the QGP must be created in the initial collision, and non-QGP phases could have important effects. 10/18/2011 4

Charmonium The J/ψ meson is one of many charmonium (cc ) bound states. SLAC: e+e- J/y BNL: p + Be J/y + X - First discovered in 1974 at both SLAC and BNL - First evidence of heavier quarks than u, d, s 10/18/2011 5

Debye Screening � Debye screening of color charges reduces the binding potential energy of quark-antiquark states. � In a QGP this could result in the melting of hadronic bound states with a size larger than the Debye screening radius of the plasma. � Matsui and Satz showed that the J/ψ should melt just above TC of the QGP. o Based on this, J/y suppression was expected to be a “smoking-gun” signature of QGP formation. 10/18/2011 6

J/y In-Medium Modification • Potential well becomes very shallow. • Can be seen in a simple potential model • …and on the lattice. In Medium • Some lattice results showed J/ψs surviving to 2 Tc, but that seems to be disfavored now. T=0 T=Tc P. Petreczky, 1001. 5284 10/18/2011 7

NA 50 Pb+Pb Measurment NA 50 at CERN-SPS measured J/y suppression in Pb+Pb collisions at sqrt(s) = 17 Ge. V • “Anomalous suppression” beyond expectation 10/18/2011 8

Relativistic Heavy Ion Collider RHIC is a 3. 8 km-long hadron collider located at Brookhaven National Lab on Long Island, NY. • It is able to collide Au+Au at up to √s. NN=200 Ge. V, or p+p at up to √s=500 Ge. V, as well as other heavy ion combinations. • The PHENIX experiment is located at one of the six beam intersection points. 10/18/2011 9

PHENIX records J/ys via: • the e+e- channel in the central arms at mid-rapidity (|y|<0. 35) • the m+m- channel in the muon arms at forward rapidity (1. 2<|y|<2. 2) Muon Tracker measures momentum with 16 tracking layers in 3 groups. Muon Identifier separates muons by their penetration through many layers of steel. 10/18/2011 10

PHENIX records J/ys via: • the e+e- channel in the central arms at mid-rapidity (|y|<0. 35) • the m+m- channel in the muon arms at forward rapidity (1. 2<|y|<2. 2) The Beam-Beam Counters measure the collision z-position and impact parameter, and are used as the minimum bias trigger for Au+Au collisions. Peripheral 10/18/2011 Central 11

New PHENIX Results! PHENIX analyzed new higher-statistics data: p+p in 2006 and 2008 Au+Au in 2007 d+Au in 2008 Smaller uncertainties and finer binning provide better constraints. The d+Au data in particular represents ~30 x the J/ψ sample that was recorded during 2003 and used in previous PHENIX d+Au analyses. d+Au 0 -20% 1. 7<y<1. 95 d+Au - Phys. Rev. Lett. 107 (2011) 142301, ar. Xiv: 1010. 1246 (longer paper in preparation) Au+Au - In Press in Phys. Rev. C, ar. Xiv: 1103. 6269 10/18/2011 12

A Recipe for Suppression The Ingredients Cold Nuclear Matter (CNM) effects: shadowing, gluon saturation, nuclear absorption, initialstate parton energy loss Hot Nuclear Matter (HNM) effects: dissociation, regeneration But what are the proportions? ? ? Start by looking at CNM using d+Au collisions. 10/18/2011 13

d+Au modification of J/ψs Nuclear Modification Factor: Significant suppression at mid and forward rapidities. arxiv: 1010. 1246 Compare these data to model calculations… d Au Bars = point-to-point uncorrelated uncertainties Boxes = point-to-point correlated uncertainties 10/18/2011 14

Nuclear modification of PDFs Nuclear PDFs are known to be modified in various xranges. Shadowing, anti-shadowing, EMC effect, etc. Fermi Motion anti-shadowing Saturation? shadowing EMC effect x PHENIX probes three ranges of x in the gold nucleus, in both the shadowing and anti-shadowing regions, using detectors at: forward y, x~0. 005 mid y, x~0. 03 backward y, x~0. 1 x 2 10/18/2011 15

Calculation I � R. Vogt calculated the J/ψ production from: o EPS 09 n. PDF with shadowing effects � We compare to both the “best-fit” and maximum-variation EPS 09 curves Eskola, Paukkunen, Salgado, JHEP 04 (2009) 065 o Include sbreakup to account for break-up of the cc pair while passing through the nucleus. EPS 09 10/18/2011 16

Rd. Au for minimum bias collisions Reasonable agreement with EPS 09 nuclear PDF + sbr = 4 mb (red curves). sbr is the only free parameter. Dashed lines are the maximum variation included in EPS 09. arxiv: 1010. 1246 EPS 09, as published, is averaged over all b, so we would expect decent agreement with Rd. Au(0 -100%). 10/18/2011 17

Rd. Au for minimum bias collisions Gluon saturation model of Kharzeev and Tuchin Good agreement at forward rapidity Deviates quickly as y<1 arxiv: 1010. 1246 Kharzeev and Tuchin, Nucl. Phys. A 770 (2006) 40 We can break the data down further by dividing events into small and large impact parameter. 10/18/2011 18

Centrality Dependence How does the modification vary with (longitudinal) nuclear thickness? Peripheral Central For EPS 09 shadowing we assume a linear dependence on the nuclear thickness. Nobody really knows! 10/18/2011 19

Rd. Au central and peripheral (new) EPS 09 + linear thickness dependence + sbreakup Unable to reproduce the rapidity dependence of Rd. Au in both central and peripheral Gluon saturation still only matches at forward rapidity We can further reduce systematics by taking the ratio… 10/18/2011 arxiv: 1010. 1246 20

RCP vs. rapidity RCP cancels most of the systematic uncertainties. EPS 09 + sbreakup still doesn’t get the rapidity dependence Gluon saturation… well… arxiv: 1010. 1246 10/18/2011 21

To further examine the centrality and rapidity dependence of Rd. Au: 1. Start from a Glauber MC of the nucleon-nucleon hit positions. 2. Add a simple parameterization based on the longitudinal thickness of the gold nucleus. 10/18/2011 22

p+Au Geometry b 10/18/2011 p+Au impact parameter tells us exactly what we want to know, ie. the transverse radius of the N-N collision(s). 23

d+Au Geometry r. T b d+Au impact parameter is not as useful since we really want the radial positions of all of the struck nucleons. Call it r. T to differentiate. 0 -20% Central For this event, there are three r. T values (transverse radial positions for the struck gold nucleus nucleons). These are the values in the histograms to the right. 20 -40% Central 40 -60% Central 60 -88% Central r. T (fm) 10/18/2011 24

d+Au Geometry r. T b 0 -20% Central 20 -40% Central Now figure out the modification M(r. T)! Convolve with the r. T dists to get Rd. Au! 40 -60% Central 60 -88% Central r. T (fm) 10/18/2011 25

= longitudinal density-weighted thickness Use Rd. Au(0 -100%) for the xaxis. This is the overall level of modification averaged across impact parameters. Use RCP as y-axis. This is relative modification between central & peripheral. a=0 a=a 1 For any value of a, we can put a point in the RCP(a) - Rd. Au(a) plane. As we vary a, we map out one curve for each of our three modification functions M(r. T). a=a 2 Any model using a particular M(r. T) must follow that curve. 10/18/2011 26

Now add the d+Au data points. Backward and mid-rapidity points agree within uncertainties for the three cases presented here. Ellipses = the point-to-point correlated systematics on Rd. Au and RCP arxiv: 1010. 1246 10/18/2011 27

However, the forward rapidity points require the suppression to be stronger than exponential or linear with the thickness. This is reflected by the inability of an EPS 09(linear) + sbr(exponential) to reproduce the Rd. Au in both central and peripheral. The only extra model dependence is the PHENIX centrality calculation, which is included in the systematics on the data. 10/18/2011 arxiv: 1010. 1246 28

Take-home Message(s) a) Gluon shadowing plays an important role in d+Au. b) Does gluon saturation? o Would help explain the large suppression at forward rapidity and the geometric dependence. Additional treatment in: Nagle, Frawley, Linden Levy, and Wysocki, ar. Xiv: 1011. 4534 (Phys Rev C, In Press) 10/18/2011 29

Hot Nuclear Matter � Now that we have some idea of the CNM effects, let’s turn to the HNM. � New Au+Au RAA at forward rapidity using the 2007 dataset. ~3 x increase in J/ψ statistics. 10/18/2011 30

RAA w/ NEW forward data Strong suppression in central Au+Au events. NA 50 Pb+Pb Forward rapidity is more suppressed than midrapidity. Midrapidity comparable to SPS energies. arxiv: 1103. 6269 10/18/2011 31

Questions How can we explain the Au+Au results? First look at CNM effects in Au+Au collisions. Project the calculation using EPS 09 n. PDFs and sbr to Au+Au. 10/18/2011 32

Projection of CNM Effects Project EPS 09 shadowing and sbr to Au+Au Doesn’t reproduce RAA or the ratio between rapidities. Forward rapidity J/ψs largely come from a high-x gluon and a low-x gluon, so shadowing effects largely cancel. arxiv: 1103. 6269 10/18/2011 33

Gluon Saturation Gluon saturation calculation (extension of method from d+Au) Matches the ratio of New calculation from forward/midrapidity very well. M. Nardi et al. indicate Arbitrarily normalizedeffect. to RAA much smaller data points. At Hard Probes 2010, she declared that there are definitely final state effects. arxiv: 1103. 6269 10/18/2011 D. Kharzeev, et al, Nucl. Phys. A 826, 230 (2009), 0809. 2933 34

HNM Effects - Zhao & Rapp Model a QGP phase followed by a hadron gas phase. Central midrapidity suppression matched, not forward rapidity. * Note how similar regeneration is between the two rapidities. Forward/midrapidity ratio not small enough • HNM pushes in the wrong direction X. Zhao and R. Rapp, Phys. Lett. B 664, 253 (2008), 0712. 2407 10/18/2011 arxiv: 1103. 6269 35

Regeneration of J/ψs helps reduce the overall amount of suppression. It does not appear to explain the larger suppression at forward rapidity. Idea was that regeneration would go as Nccbar 2, but this is only true if it is due to off-diagonal pairs (case 3). Most calculations seem to be dominated by diagonal pairs recombining (case 2), which goes as Nccbar. 10/18/2011 36

Take-home Message(s) a) Gluon shadowing plays an important role in d+Au. b) Does gluon saturation? o Would help explain the large suppression at forward rapidity and the geometric dependence. c) HNM effects are confirmed o Cannot divide them out without precise CNM description. d) Larger suppression at forward rapidity still not explained o 10/18/2011 Does not seem to be from recombination as previously proposed. 37

Recent J/ψ results from STAR J/ψ RAA from Au+Au p. T = 2 -5 Ge. V/c roughly agrees with PHENIX, which was p. T>0 but narrower rapidity. p. T > 5 Ge. V/c less suppressed. RAA rises with p. T even in central events 10/18/2011 38

Recent J/ψ results from STAR also measured J/ψ v 2 in Au+Au to much better precision than the previous PHENIX preliminary result. Many expect J/ψs from recombination to have a large v 2. After all, we know that open charm mesons have a large v 2. However, the STAR result agrees with zero within the uncertainties. Does this mean no J/ψs from recombination? !? 10/18/2011 39

J/ψ v 2 in a 2 -component model From Zhao & Rapp (ar. Xiv: 0806. 1239): Combine Directly-produced J/ψs Coalescence J/ψs Direct J/ψs dominate above ~2 Ge. V/c, and have small v 2. Coalescence J/ψs dominate at low p. T, where their flow is small. “It is somewhat sobering to find that the resulting v 2(pt) is small, around 2 -3%, over the entire pt range. ” 10/18/2011 40

What can the LHC tell us? There are currently J/ψ results from each of the LHC heavy ion experiments: ATLAS was first out of the gate with a published result. No p+p reference at the time, so only RCP, no RAA. 10/18/2011 41

What can the LHC tell us? There are currently J/ψ results from each of the LHC heavy ion experiments: ALICE RAA vs. Npart • Smaller suppression than either of the PHENIX rapidity ranges. • p. T range (p. T>0) comparable to PHENIX, but rapidity range is 2. 5<y<4. 0! (more forward than the PHENIX muon arms) • Left-most point averages over a region where RAA changes quickly in PHENIX. 10/18/2011 42

What can the LHC tell us? There are currently J/ψ results from each of the LHC heavy ion experiments: CMS RAA vs. Npart • Higher p. T range: 6. 5 < p. T < 30 Ge. V/c • Range is closer to STAR’s high-p. T result, but covers a larger rapidity range, |y|<2. 4. • CMS suppression is much larger than STAR • Central is comparable to PHENIX low p. T and forward rapidity. 10/18/2011 43

What can the LHC tell us? Current answer: not much Too many caveats on the results to come to any whole-picture conclusions. • Need differential measurements (d/dp. T, d/dy, etc. ) over a wide range. • No CNM data yet (p+Pb) • Crucial at CERN-SPS and RHIC With the beautiful p+p charmonia results showing us what the detectors are capable of, though, it’s only a matter of time. 10/18/2011 44

The Future � RHIC J/ψ results have helped make the interpretation of SPS results more robust. I think the same will happen with the LHC results and RHIC J/ψ physics. � PHENIX is adding upsilon measurements and additional charmonium states with our next-generation upgrades. o Varying the state probes the screening length/temperature. o No off-diagonal regeneration for upsilons. o CNM effects should be similar across states. 10/18/2011 45

� PHENIX (and RHIC) currently planning upgrades for the next decade, until EIC starts up. � RHIC II luminosities alone will allow new measurements to higher p. T and new states (Upsilon!). Full PHENIX decadal plan is available online: http: //www. phenix. bnl. gov/phenix/WWW/docs/decadal/2010/phenix_decadal 10_full_refs. pdf 10/18/2011 46

� To take full advantage of RHIC II, PHENIX is looking to replace the Central Arms with new detectors including full 2 pi coverage in phi and ± 1 in eta. � For J/ψs and Upsilons this will give a >10 x increase in acceptance. � s. PHENIX studies are currently ongoing… 10/18/2011 47

Backup 10/18/2011 48

Initial-State Parton Energy Loss (d+Au) Add in another CNM effect: a simple form of initial-state parton energy loss. DE/E ~ 0. 05/fm 2 * L Fit RCP w/ to get sbr, using central EPS 09 n. PDF. Red lines are other EPS 09 n. PDFs. Matches RCP pretty well, but not the separate Rd. Au at forward or backward rapidity. 4/27/11 Nagle, Frawley, Linden Levy, and Wysocki, ar. Xiv: 1011. 4534 M. Wysocki - GHP 2011, Anaheim, CA 49

Initial-State Parton Energy Loss (cont) Now try with L 2 dependence. DE/E ~ 0. 005/fm 2 * L 2 Again, matches RCP pretty well, but not the separate Rd. Au at forward or backward rapidity. EPS 09 + sbr + initial-state energy loss cannot reproduce the Rd. Au rapidity/centrality distribution. 4/27/11 Nagle, Frawley, Linden Levy, and Wysocki, ar. Xiv: 1011. 4534 M. Wysocki - GHP 2011, Anaheim, CA 50

Initial-State Parton Energy Loss (Au+Au) Redo the projection to Au+Au now including the energy loss: DE/E ~ 0. 005/fm 2 * L 2 Largest effect is at forward rapidity, but even there it is not huge. Still takes a very large sbr to match the RAA, and the ratio is still > 80%. arxiv: 1103. 6269 4/27/11 M. Wysocki - GHP 2011, Anaheim, CA 51

Longitudinal Thickness Modification Take CNM effects to depend on the nuclear geometry via density-weighted longitudinal thickness, L, of the nucleus, as in Klein and Vogt, nucl-th/0305046. Use Woods-Saxon for r(z, r. T) Several simple modification functions using L: 10/18/2011 Exponential is usually used for cc break-up, while the linear case has been used for parameterizing n. PDFs as a function of impact parameter (eg. EPS 09). 52

Rd. Au from geometric modification Given the r. T-distribution of NN collisions and the r. T-dependent modification, it is simple to calculate Rd. Au for any centrality bin: r. T dist. of nucleon-nucleon collisions from PHENIX centrality MC Modification as function of nuclear thickness and a free parameter, a At a fundamental level, we want to study how the modification turns on with centrality. 10/18/2011 53

We discovered a - Rapidity dependence! � Extract best fit to RCP at a given rapidity versus centrality. � Based on predictions from R. Vogt. T. Frawley ETC, Trento � Parameterizes all the effect that shadowing is missing. � Same shape at lower energy (initial state energy loss). 10/18/2011 54

Centrality @ PHENIX b Use BBC charge divided into percentile bins of centrality to classify events. Then use simple Monte Carlo to map this to Ncoll or impact parameter. PHENIX currently uses four centrality bins for d+Au. PHENIX d+Au Centrality Classes Includes Glauber Au geometry, deuteron Hulthen wavefunction, event-to-event fluctuations, modeling of PHENIX BBC response and trigger bias, and final event selection. 10/18/2011 55

Au+Au Reconstruction Efficiency To calculate Au+Au invariant yields, we need to correct by detector acceptance and efficiency. The two spectrometers are not identical. The North Arm has greater acceptance, but worse occupancy effects. 10/18/2011 56

Linear Example The Rd. Au(a) for any centrality bin is generated simply by folding M(r. T) with the r. T distribution for a given centrality bin. RCP is simply the ratio of two centrality bins. 10/18/2011 57

d-Au collision Sometimes only one nucleon from the deuteron hits the Au nucleus. 62% Peripheral 60 -88% 37% Mid-Peripheral 40 -60% 20% Mid-Central 20 -40% 7% Central 0 -20% b = event impact parameter r. T(1) We can actually measure this if the missed nucleon is a neutron (in the PHENIX ZDCs) 9/12/2021 10/18/2011 58 58

What about event-to-event fluctuations in L(r. T)? r. T(1) For each binary collision at r. T, count the number of other nucleons in the nucleus inside the tube defined by r. T ± 2 x 0. 877 fm In this example, the Ntube = 6. Perhaps the nuclear modification is related not to the average thickness L(r. T), but instead the fluctuating quantity related to Ntube defined above. 10/18/2011 59

In the Linear Modification Case, these fluctuations around the average will not matter. Not exactly true for the non-linear cases. Also, the difference of the blue solid and dashed raises the question of how localized in r. T is the effect (blurred over the size of a nucleon? ). Black – TProfile(“RMS”) Blue Solid = L(r. T) scaled Blue Dashed = L(r. T) blurred over ± 2 x 0. 877 fm around r. T 10/18/2011 60