The rapidity dependence of the proton-to-pion ratio in Au+Au and p+p collisions at sqrt(sNN) = 62.4 and 200 GeV
aa r X i v : . [ nu c l - e x ] S e p The rapidity dependence of the proton-to-pion ratio in Au + Au andp + p collisions at √ s NN = . P. Staszel a for the BRAHMS collaboration a Smoluchowski Inst. of Physics, Jagiellonian University, ul. Reymonta 4, 30-059 Krak´ow, Poland
Abstract
The BRAHMS measured proton-to-pion ratios in Au + Au and p + p collisions at √ s NN = . √ s NN =
200 GeV are presented as a function of transverse momentum and colli-sion centrality within the pseudo-rapidity range 0 ≤ η ≤ + Au at √ s NN =
200 GeV are compared with predictions from models which incorporate hydro-dynamics, hadronrescattering and jet production, in the η interval covered. In Au + Au collisions at √ s NN =
200 GeV, η ≈ .
2, and at √ s NN = . η =
0, the bulk medium can be characterizedby the common value of µ B ≈
65 MeV. The p /π + ( p T ) ratios measured for these two selectionsdisplay a striking agreement in the p T range covered (up to 2 . / c). At a collision energy of62.4 GeV and at forward pseudo-rapidity we found a crossing point of p /π + ratios measured incentral and semi-peripheral Au + Au and in p + p reactions. The crossing occurs in the narrow η bin around value of 3.2, simultaneously in the whole covered p T range (0.3 GeV / c < p T < / c).The measured p T dependence of the baryon-to-meson ratio appears to be related to modifica-tions in the hadronization mechanisms as it happens in a partonic medium. It was pointed out thatthe baryon-to-meson ratio p T dependence should be sensitive to the hadronization scenario dueto the di ff erent quark content of baryons and mesons [1] and / or to radial flow of the bulk mediumbecause of significant di ff erences in baryon and meson masses [2]. Both flow and medium quarkcoalescence are expected to enhance protons over pions at intermediate p T .The PHENIX ¯ p /π − data at mid-rapidity is well described by the Greco, Ko, and Levai quarkcoalescence model where the introduced coalescence involves partons from the medium (ther-mal) and partons from mini-jets [3]. The Hwa and Yang quark recombination model is alsosuccessful in describing BRAHMS and PHENIX mid-rapidity data for p /π + [4].On the other hand, the comparison with the hydrodynamical model shows that hydro-flowcannot itself account for the large observed ratio above ≈ / c and that the model overpre-dicts the data at low p T [5]. These results support the view of a hadronization process driven byparton recombination with negligible final state interactions between produced hadrons. Never-theless, at large µ B , a significant gap between the temperature of the transition from the partonicto the hadronic phase, T c , and the temperature of chemical freeze-out is predicted by QCD latticecalculations [6]. Thus at large µ B , the picture, suggested by mid-rapidity measurements, mightbe contaminated by final state hadron interactions leading to a transition from the parton recom-bination scheme to a hydrodynamical description that has a common velocity field for baryonsand mesons [7, 8].The setup of the BRAHMS experiment is described in details in [9]. Here we just point Preprint submitted to Nuclear Physics A May 29, 2018 ut that the arrangement of BRAHMS spectrometers, namely of the Mid-Rapidity Spectrome-ter (MRS) and the Forward Spectrometer (FS), makes it possible to measure identified particlespectra over a pseudo-rapidity interval from η = η = .
8. Particle identification in the FSis provided by TOF measurements for low and medium particle momenta. High momentumparticles are identified using a Ring Imaging Cherenkov detector (RICH) [10].The data analysis utilizes the feature of the same pion and proton acceptance in the η versus p T space in the same real time measurement. For a given η - p T bin the p /π ratios are calculated ona setting by setting basis. In order to avoid mixing di ff erent PID techniques, which usually leadto di ff erent systematic uncertainties, the ratios are calculated separately for the TOF PID and theRICH PID. In this way all factors such as acceptance corrections, tracking e ffi ciencies, triggernormalization and bias related to the centrality cut cancel out in the ratio. The only remainingcorrections are those that are species related which are:(i) decay in flight, interaction with the beam pipe, and the detector material budget,(ii) the PID e ffi ciency correction.The corrections for (i) are determined from the single particle response of pions and protons ina realistic GEANT [11] model description of the BRAHMS experimental setup. We estimatethat the overall systematic uncertainty related to this correction is at the level of 2%. The TOFPID is done separately for small momentum bins by fitting a multi-Gaussian function to theexperimental squared mass M distribution and applying a ± σ cut to select a given particletype. For measurements done with the FS spectrometer in the momentum range where pionsoverlap with kaons, (usually above 3 . / c) the RICH detector can be used in veto modeto select kaons with momentum smaller than the kaon Cherenkov threshold which is about 9GeV / c. Above the proton threshold momentum, which is about 15 GeV / c, the RICH providesa direct proton identification. In this momentum range the RICH PID is based on the particleseparation in the M versus momentum space. The described PID procedure leads to a relativelyclean sample of pions with some contamination by kaons having spurious rings associated in theRICH counter. Together with the kaon - proton overlap at larger momenta, this contaminatione ff ect is a source of systematic errors which have been estimated.Figure 1 (left) shows p /π + ratios obtained for Au + Au reacting at √ s NN =
200 GeV for twocentrality sets of events, namely, for centrality 0 −
10% (solid dots) and 40 −
80% (open squares)The shaded boxes plotted for the most central events only, represent the systematic uncertaintiesdiscussed in the previous section. The ratios extracted from p + p data at the same energy areplotted for comparison (solid stars). The p T coverage depends on the pseudo-rapidity bins andextends up to p T = / c for η = . .
1. At low p T ( < . / c) the p /π + ratios exhibita rising trend with a weak dependence on centrality. The dependency on centrality begins above1 . / c. The ratios appear to reach a maximum value at p T around 2.5 GeV / c (whenever thereis enough p T coverage). The maxima of the ratios increase with the level of centrality and at η = .
1, are equal to about 2.5 and 1.5 for the 0 −
10% and 40 −
80% centrality bins, respectively.The p + p ratios are consistent with Au + Au data at low p T and begin to deviate significantly above p T = / c. At η = . + p collisionswhich is a factor of 4.5 smaller than that observed for central Au + Au reactions.The values of the ¯ p /π − ratios plotted in Fig. 1 (right) are significantly lower than the p /π + ratios (note the di ff erence in the vertical scale), however, the centrality dependence shows thesame features as those observed in the p /π + ratios, namely, that the ratios for di ff erent centralityclasses are consistent with each other up to p T ≈ . / c and a strong dependence on cen-trality appears at larger transverse momenta reaching a maxmimum at similar p T as the positive2 .5 1 1.5 2 2.5 3 3.5 4 = 2.25 η BRAHMS + π p / + π p / = 3.1 η [GeV/c] T p [GeV/c] T p = 3.5 η = 2.6 η preliminary = 3.3 η [GeV/c] T p [GeV/c] T p = 3.8 η = 2.25 η BRAHMS - π / p - π / p = 3.1 η [GeV/c] T p [GeV/c] T p = 3.5 η = 2.6 η preliminary = 3.3 η [GeV/c] T p [GeV/c] T p = 3.8 η Figure 1: Centrality dependent p /π + (left) and ¯ p /π − (right) ratio for Au + Au system colliding at √ s NN =
200 GeV forcentral (0 − − + p collision data at the same energy. Thevertical bars represent the statistical errors and the shaded bands (plotted only for central Au + Au) show the systematicuncertainties. particles. Looking at the p + p data alone, one can note the di ff erence in shape between the p /π + and ¯ p /π − ratios: a clear shift of the ¯ p /π − peaks towards lower p T , as well as a much broader p /π + peaks. These large di ff erence between the Au + Au and p + p both in shape and overall magnitudemay reflect significant medium e ff ects in Au + Au at √ s NN =
200 GeV in the pseudo-rapidityintervals covered.Fig. 2 shows the p /π + ratios as function of p T in the pseudo-rapidity range 2 . < η < . + p reactions at √ s =
200 GeV. A very clear di ff erence is found as the pseudo-rapidity changes from η = . η = .
8. But at high p T all these ratios tend towards a commonvalue of about 0.4 consistent with pQCD predictions [12]. Fig. 3 compares the p /π + ratio fromp + p and Au + Au collisions at √ s NN = η ≈ . µ B ≈
250 MeV, [13]). Thereis remarkable little di ff erence in the p /π + ratios over a very wide range of the colliding systemvolume, namely, from p + p reactions up to central Au + Au collisions. It should be noted that(what is now shown) at η = p /π + ratio from central Au + Au reactions is enhanced inrespect to p + p collisions by a factor of 1.6, wheres at η = p /π + ratio in p + p exceeds that measured in central Au + Au by a factor of about 1.4. This indicatesthat the consistency observed at η ≈ . ff ernt systems (from p + p up to central Au + Au) at this particular pseudorapidity bin.We presented the p T dependence of the p /π ratios measured in Au + Au and p + p collisionsat energies 62.4 and 200 GeV as a function of pseudo-rapidity and collision centrality (Au + Au).The data provide the opportunity for studying baryon-to-meson production over a wide range ofthe baryo-chemical potential, µ B . For Au + Au and p + p reactions at √ s NN =
200 GeV the p /π + and ¯ p /π − ratios show noticeable dependency on centrality at intermediate p T with a rising trendfrom p + p to central Au + Au collisions. We have shown that p /π + ratios are remarkably similarfor central Au + Au at √ s NN =
200 GeV, η ≈ . + Au at √ s NN = . η ≈ p / p [14]. This observation, together3 [GeV/c] T p + π p / p+p at 200 GeV BRAHMS preliminary = 2.6 η = 3.1 η = 3.3 η = 3.5 η = 3.8 η Figure 2: The p /π ratios at forward rapidities (2 . < η < .
8) in p + p at √ s =
200 GeV. [GeV/c] T p + π p / -1 ≈ η = 62.4 GeVsAu+Au at centrality: 0-10 %centrality: 10-20 %centrality: 20-40 %centrality: 40-80 %p + p BRAHMS preliminary
Figure 3: p /π + ratio from p + p and Au + Au collisions at √ s NN = η ≈ . with the observed centrality dependence suggests, that at these energies and pseudo-rapidityintervals, particle production at intermediate p T is governed by the size and chemical propertiesof the created medium. Finally, the Au + Au and p + p measurements at √ s NN = . p /π + ratios for p + p and for all analysed Au + Au centralities cross simultaneously atthe same η value ( ≈ p T range e.g. from0.3 GeV / c up to 1.8 GeV / c. References [1] R. J. Fries et al., Phys. Rev. Lett. , 202303 (2003); S. A. Bass, J. Phys. Conf. Ser. , 279 (2006).[2] T. Peitzmann, Nucl. Phys. A 727 , 179 (2003).[3] V. Greco, C.M. Ko, and P. Levai, Phys. Rev. Lett. C 67
A 774
G 28 , 1971 (2002).[7] T. Hirano, and Y. Nara, Nucl. Phys.
A743
305 (2004).[8] W. Broniowski, and W. Florkowski, Phys. Rev. Lett. A 499
437 (2003).[10] R. Debbe et. al., Nucl. Instr. Meth.
A 570 / asdoc / geant html3 / [12] T. Hirano, and Y. Nara, Phys. Rev. C 69 , 034908 (2004).[13] I.C. Arsene et. al., for BRAHMS Collaboration, Int. Jour. of Mod. Phys.
E 162035 (2007).[14] F. Videbaek, Quark Matter 2009, this proceedings.