Summary on high pT probes

Results on high-pT probes shown at the Hard Probes 2008 Conference are summarized, with an appreciation of the improvements in precision of the measurements and experimental techniques since the beginning of RHIC operation. Particular attention is given to the latest measurements of the nuclear modification factor of identified particles, photon-hadron correlation measurements, and full jet reconstruction.


Introduction
The goal of heavy-ion collisions is to study the medium produced and ultimately quantify the properties of the medium. High transverse momentum (p T ) particles provide particularly good probes of the medium created in heavy-ion collisions because they are created early in the collision and thus are sensitive to the transport properties. Probes of heavy-ion collisions having high transverse momentum became accessible at the Relativistic Heavy Ion Collider (RHIC), due to the large center of mass collision energies (the largest of which has been √ s NN = 200 GeV). At these high energies, hard production cross sections are large enough for the probes at high p T to become abundant. The experiments, positioned at intersection points along the RHIC ring, began recording data in 2000.

High-p T spectra
The first measurements of high p T spectra at RHIC came from Au + Au collisions at √ s NN = 130 GeV. Already in these data, a factor of 5 suppression relative to binary scaling was observed for the first time [1]. The suppression is quantified with the nuclear modification factor, defined as R AA = d 2 N AA /dp T dη (N binary )(d 2 σ pp /dp T  where the numerator is the yield of particles measured in A + A collisions and the denominator is the yield measured in p + p collisions scaled by the mean number of binary collisions in the centrality selection of A + A collisions. In the first measurements from RHIC, the spectra extended to p T ∼ 4 GeV/c for identified pions and ∼5 GeV/c for nonidentified charged hadrons [1] and had systematic errors of ∼ 30% at the highest measured p T values. The systematic errors in the measured nuclear modification factor R AA were as large as ∼ 50% due to an additional uncertainty of 35% on the p + p reference spectrum, which was deduced from an interpolation of data taken for other center of mass energies [2]. The most recent measurements of particle production as a function of p T include many different identified particle species, extend to as high as p T ∼ 20 GeV/c, and, in general, have systematic errors of approximately 20%. Figure 2.1 shows the R AA of a compilation of identified particles measured by the PHENIX and STAR experiments [3] for p T up to 10 GeV/c. This figure highlights not only the wealth of data provided by RHIC, but summarizes some of the key features of particle spectra at high p T . There seems to be an ordering to the suppression factors in an intermediate p T range (2 < p T < 6 GeV/c). The η are as suppressed as the pions, the kaons and φ are less suppressed than pions, and protons are even enhanced over some range of p T . Before the measurement of the R AA of the φ, it was assumed that there would be a meson/baryon ordering to the amount of suppression, as is observed in the momentum anisotropy v 2 of the particles [5, 6]. However, the φ R AA seems to lie between the pions and protons, following the trend of the kaon R AA at lower p T . Figure 2.2 shows the nuclear modification factor for mesons compared to that for direct photons (photons not from hadronic decays) up to p T ∼ 20 GeV/c [4,7]. The photons are not suppressed up to p T ∼ 10 GeV/c. However, for p T > 14 GeV/c, the photons are suppressed relative to binary scaling. This can be explained theoretically from different effects (perhaps each contributing to the suppression that  [4] is measured) [8]. The recent high-statistics d +Au RHIC run should provide an important test of the cold nuclear matter effect (shadowing or gluon saturation).
Theoretical calculations have been largely successful at describing high-p T spectra of pions and non-identified charged hadrons [9][10][11]. However, there are several different theoretical approaches that lead to contradicting conclusions about properties of the medium. With different conclusions of the transport coefficientq = k 2 T /L from different models [12,13], it is important to understand the different assumptions that lead to these different conclusions. There are ongoing attempts to understand the differences in the energy loss models [13][14][15].
Despite the successes of these models to describe the suppression observed in the light hadrons, the surprising result of an equally strong suppression of heavy-flavor at high p T was not predicted by these same models, and could not be described by radiative energy loss alone [16]. The strong elliptic flow measured at RHIC also pointed toward a strongly coupled medium, which led to the question of applicability of perturbative calculations to describe the interactions which lead to energy loss in this medium. Calcu-lations of partonic energy loss in a strong coupling regime, using Ads/CFT to describe the heavy-ion collisions, were also discussed at the conference [17][18][19].

Correlation measurements
The single-particle measurements at high p T revealed the sizable effect that the medium has on hard-scattered partons. They also revealed new and interesting phenomena relevant at intermediate p T that could be explained by a recombination production mechanism [20-22], although not all aspects of this idea have yet been fully developed. However, it is difficult to learn about the details of the energy loss mechanism in the medium because the single particles that are measured are those that emerge from the dense medium with significant energy and thus have a strong bias for having been produced close to the surface of the medium. If one wants to go beyond the measure of an overall suppression factor (via R AA ), then both sides of a di-jet need to be measured simultaneously. The "near-side" of the di-jet is the side that is triggered on, and is thus again somewhat surface-biased due to the trigger condition of the presence of a high-p T particle, and the "away-side" jet particles can then reveal the effect of the medium relative to vacuum expectation. Full jet reconstruction has been considered a difficult problem in heavy-ion collisions, where the multiplicities are large (thousands of produced particles). To overcome this technical challenge, RHIC experiments measured azimuthal correlations between two particles to determine jet yields on the near-side and the away-side. Many interesting effects, modifications in the shapes and yields jet-associated correlation function, have been reported. On the near-side, there is a long-range correlation in η [23,24] (referred to as the "ridge"), while the away-side shows a modification to the Gaussian-like jet shape that has two maxima away from φ = 180 • [25,26]. Possible explanations for both the ridge on the near-side [27][28][29][30][31][32] and the shape modification of particles correlated with the trigger on the away-side for low to intermediate p T [33][34][35][36][37][38][39] have been widely discussed in the literature. With many theoretical ideas that explain the origin of the ridge, what is needed is predictions from these theoretical ideas that can be verified or falsified.
As the luminosity of RHIC increased, the experiments were able to take more data at high p T , triggering on events with a high-energy cluster in the electromagnetic calorimeter. With larger statistics, correlation measurements extended to higher p T (both for the trigger particle and the associated particles), where these shape modifications are no longer dominant effects. The ultimate goal for high-p T probes is to obtain a measure of the modification of the fragmentation function due to the dense medium. In order to make this measurement, it is necessary to have access to the parton's original energy. This is possible via a photonhadron correlation measurement (γ -jet) [40], where the trigger particle is a high-p T direct photon, which is not affected Fig. 3.1 The yields of charged particles associated with a direct-photon trigger with 8 < p T < 16 GeV/c, as a function of z T = p T ,assoc /p T ,trig for central (0-10%) collisions and peripheral (40-80%) collisions, compared to the yields associated with a π 0 trigger [41] by the medium to lowest order, and the hadrons on the awayside measure the medium-modified jet. Figure 3.1 shows the first result of a modified fragmentation function from the jet yields associated with a direct-photon trigger [41]. The yields associated with a direct photon trigger are compared to those associated with a π 0 trigger. The yields for direct photon triggers are smaller, even in peripheral collisions, because the π 0 triggers come from a fragmentation process where the parton possesses a larger p T than the direct photon that does not come from a fragmentation. In central collisions, there can be additional differences due to a combination of possible effects: energy loss of the π 0 trigger, quark vs. gluon fragmentation, and the amount of the medium that is probed when triggering on an unbiased direct photon. In order to disentangle these effects, the systematic uncertainties on the measurement need to be improved, in addition to including theoretical input for interpretation of the data.

Full jet reconstruction
At this conference, the first ever results from full jet reconstruction in heavy-ion collisions were shown. Theoretical calculations have predicted a modification in the particlespecies content in jet fragmentation [42].  Fig. 4.1 (right panel) for a cone radius of 0.4 [43]. The mass ordering in the peak values of the fragmentation function measurement of kaons and Λ, seems to differ from the calculation of kaons and protons predicted in p + p collisions, emphasizing the importance of having the baseline vacuum measurement before measuring any medium modifications. The effect of the cone radius imposed on the measurement has been studied [43]. Improved precision of these results and the comparison to the theory will serve as important input to constrain the parameters of the calculation.
There has been recent progress on jet reconstruction algorithms to handle the large pile-up expected at the LHC [44]. The improvements have focused on background subtraction, and this has also been relevant for heavy-ion collisions. Figure 4.2 shows the result of applying such algorithms to Au + Au collisions at RHIC [47]. The measurement in central Au + Au collisions agrees with the binary-scaled measurement in p + p collisions for the lowest p T threshold cut, indicating no modification to the fragmentation function when the full jet is reconstructed. With increasing p T thresholds, the agreement is not so good; which may be due   [45,46]. The p T threshold is increased from left to right [47] to not fully correcting for the increasing bias. Corrections for the bias from the p T threshold cut, as well as the jet energy resolution, have been determined by embedding PYTHIA jets into real Au+Au collisions [47,48]. Further studies of the systematics of this difficult measurement need to be performed before drawing strong conclusions. φ distribution for reconstructed jets with p T > 20 GeV/c measured in central Au + Au collisions compared to p + p collisions [48] The progress in jet reconstruction in heavy-ion collisions opens up new possibilities in studying jet energy redistribution in the medium and thus understanding properties of the medium such asq. Figure 4.3 shows the ability of measuring the difference in energy measured on opposite sides of a dijet [48].

Conclusions and summary
Many questions remain about our understanding of the medium created at RHIC. In particular, the value ofq is yet to be determined from the data. Different theoretical calculations have extracted values ofq from the high-p T data, but they vary widely. The applicability of perturbative calculations to account for the parton energy loss in the dense, strongly coupled medium has also been questioned. The measurement of the medium-modified fragmentation function, through γ -jet and/or full jet reconstruction, holds the promise of providing answers to these unresolved questions.