Energy Loss and Flavor Dynamics from Single Particle Measurements in PHENIX
aa r X i v : . [ nu c l - e x ] S e p Energy Loss and Flavor Dynamics from Single ParticleMeasurements in PHENIX
R. Belmont for the PHENIX Collaboration
Department of Physics and Astronomy, Station B 1807, Vanderbilt University, Nashville, TN 37235, USA
Abstract
The transverse momentum spectra, yields, and ratios of charged pions, protons, and antipro-tons have been studied up to 5 GeV / c in p T in 5 di ff erent centrality classes in Au + Au collisionsat √ s NN =
200 GeV. These results are compared and contrasted with the observables calcu-lated in recombination models of hadronization. They are also used to examine the color chargedependence of parton energy loss in the medium.
1. Introduction
One of the most intriguing results from the early years of RHIC operation was the apparentnon-suppression of baryon production at intermediate values of transverse momentum p T . Thisresult stimulated new models of hadronization mechanisms in the quark-gluon plasma (QGP),many of them based on some form of parton recombination. These models have shown qualita-tive agreement with the published experimental data, but more detailed results with extended p T reach are needed to further discriminate between them.Additionally, the flavor dependence of parton energy loss is expected to become apparentat su ffi ciently high p T . The di ff erence in energy loss between gluons and quarks propogatinga pure color field is the Casimir factor of 9 /
4, but this e ff ect may be mitigated by parton flavorconversion. By this, we mean elastic scattering processes in which the leading parton of thejet changes flavor or type. The two dominant modes of this are 1) annihilation and its inversereaction gluon fusion, ¯ q + q ↔ g + g , and 2) Compton scattering, q + g ↔ g + q , wherein themomenta between the quark and gluon are exchanged and the particle with the higher momentumhas consequently changed type [1].This study uses the data set collected by PHENIX during RHIC operations in the year 2007,which is over 800 µ b − in integrated luminosity of Au + Au collisions at √ s NN =
200 GeV. Thenew results shown herein make use of the high p T PID capabilities of the time-of-flight andaerogel cherenkov counter detectors in the west arm of the PHENIX spectrometer, which providetrack-by-track identification of protons, antiprotons, and charged pions up to a p T of 5 GeV / c in5 di ff erent centrality classes, 0-10%, 10-20%, 20-40%, 40-60%, and 60-92%. In future studies,higher p T reach and finer centrality binning will be possible. Preprint submitted to Nuclear Physics A November 18, 2018 . Results
The proton to pion ratio, in both charges, has been examined in each of the 5 centrality classesin this study. Similar to prior observations [2, 3], the p /π + and ¯ p /π − ratios in the most peripheralcollisions are similar to those in p + p collisions and the maximum of both ratios increases mono-tonically with centrality. The left panel of Figure 1 shows the new results of the p /π + and ¯ p /π − ratios together as a function of p T for the 10% most central collisions, and the right panel showsa comparison of p /π + only to various hadronization models [4]. The new data, with extended p T reach, improve the discriminative power against the models. The models adequately describe thedata in the low and intermediate p T regions, but they all fall o ff too quickly above the maximum,at p T ≈ / c. This disagreement could be from the fragmentation functions used, or from theassumption of the relative contribution of recombination and fragmentation to the final hadronspectrum. Figure 1: Left Panel: ratios p /π + and ¯ p /π − as a function of p T , 0-10% centrality; data from this study shown in filledcircles, data from previous studies [2, 3] shown in open squares. Right Panel: ratio p /π + from this study in red circles,ratio p /π + [2, 3] in black circles, and ratio p /π [2, 3] in black squares, with model comparisons. To study the energy loss of hard-scattered partons, we examine the nuclear modification bycomparing the hadron yields in central and peripheral collisions scaled by the number of binarynucleon-nucleon collisions, R CP . The large data sample collected allows for the study of p , ¯ p , π + ,and π − separately and with significantly improved statistical precision and higher p T reach com-pared to previous results [2, 3]. We see in the left panel of Figure 2 that both of the charged pionshave have a similar suppression pattern at all p T and that at higher momentum they agree quitenicely with the neutral pions as well. Additionally, and perhaps more interesting, the p and ¯ p also agree within errors at all p T , indicating that the jets that form them have experienced similarsuppression. It is expected that antibaryons will have a larger gluon fraction than baryons which,owing to the stronger energy loss expected for gluons, should result in a stronger suppressionpattern for antibaryons than baryons.In Reference [1], a novel measurement, the so-called “double ratio,” is proposed as an ob-servable to test for jet flavor conversion. The fragmentation functions of protons, indeed allbaryons, have larger gluon contributions than those of pions, or mesons in general. Considering2he expected stronger suppression of gluons, protons should exhibit stronger suppression thanpions, meaning the R AA or R CP for protons should be lower than that of pions, and so the ratio r ( p /π ) = R AA ( p ) / R AA ( π ) should be less than unity. As seen in the right panel of Figure 2, theratio in the available p T regime is much greater than one, so there is clearly still some interplaybetween parton coalescence and jet fragmentation. Note that, at this time, p + p reference data arenot available with the appropriate p T reach, so we need to use R CP as a proxy to R AA . In the p T regions already studied, the p /π ratio is comparable but slightly greater than that in p + p, so theconclusion would not change. Figure 2: R CP vs p T of pions and protons (left panel) and double ratio R CP ( p ) / R CP ( π ) vs p T (right panel). Azimuthal anisotropy measurements of identified hadrons can also help us to further under-stand the hadronization mechanisms. Anisotropy flow parameters have been studied using thesame experimental apparatus as used in this study [5]. So-called “constituent quark scaling” hasbeen known for some time and is a feature of quark coalescence models [6]. Previous measure-ments have focused on low to intermediate p T where we believe that elliptic flow is the relevantmechanism for the generation of v . Elliptic flow is of course a soft physics phenomenon, so atsu ffi ciently high p T we can expect jet properties to dominate v . The left panel of Figure 3 showsanisotropy parameter v as a function of transverse kinetic energy KE T for pions, kaons, andprotons, with each axis scaled by the number of constituent quarks. In non-central events, thereis less material in-plane and more material out-of-plane, so making the very reasonable assump-tion that there is some path dependence of the energy loss one naturally comes to the conclusionthat v should be positive even at high p T , which is seen in the data. Additionally, the colorcharge dependence of partonic energy loss should cause larger v for gluon jets than for quarkjets, and this should manifest as protons having higher v than pions. The right panel of Figure 3shows v as a function of p T for pions, kaons, and protons, and the axes are not scaled. In theintermediate p T region where recombination dominates, the proton v is higher than the pion v ,as expected. However, as one goes to higher p T , where jet fragmentation begins to dominateparticle production, we see that the proton v starts to decrease and appears to merge with thepion v . 3 igure 3: Left panel: anisotropy parameter v of identified hadrons as a function of KE T , each axis scaled by the numberof constituent quarks [5]. Right panel: anisotropy parameter v of identified hadrons as a function of p T [5].
3. Conclusion
We have herein reported the salient features of our recent study of p T spectra of chargedpions, protons, and antiprotons in Au + Au collisions at √ s NN =
200 GeV. The p /π ratio as afunction of p T shows qualitative agreement with the various recombination models, but a moredetailed theoretical study with updated fragmentation functions is needed. On the experimentalside, higher p T reach and reduced systematic uncertainty should be achievable. The R CP and v suggests a strong amount of interplay between recombination and jet fragmentation and flavordependent e ff ects. References [1] W. Liu and R. Fries,
Phys. Rev. C , 054902 (2008).[2] S. S. Adler et al. , Phys. Rev. Lett. , 172301 (2003).[3] S. S. Adler et al. , Phys. Rev. C , 034909 (2004).[4] K. Adcox et al. , Nucl. Phys. A , 184 (2003).[5] S. Huang,
J. Phys. G. , 064061 (2009).[6] S. Afanasiev et al. , Phys. Rev. Lett. , 052301 (2007)., 052301 (2007).