Probing High Parton Densities at Low- x in d+Au Collisions at PHENIX Using the New Forward and Backward Muon Piston Calorimeters
PProbing High Parton Densities at Low- x in d + Au Collisions atPHENIX Using the New Forward and Backward Muon PistonCalorimeters
Beau Meredith a for the PHENIX collaboration a University of Illinois at Urbana Champaign, 1110 W Green St, Urbana, IL, 61801, USA
Abstract
The new forward Muon Piston Calorimeters allow PHENIX to explore low- x parton distributionsin d + Au collisions with hopes of observing gluon saturation. We present a two-particle azimuthal ∆ φ correlation measurement made between a mid-rapidity particle ( | η | < .
35) and a forward π (3 . < η < .
9) wherein we compare correlation widths in d + Au to p + p and compute I dA .
1. Introduction
Deuteron-gold collisions at RHIC provide a system wherein one can explore nuclear e ff ectson initial-state parton densities in the absence of final-state medium e ff ects in heavy ion colli-sions. RHIC experiments have shown a suppression in nuclear modification factors ( R dA , R cp )for √ s NN = GeV d + Au collisions in the forward (deuteron) direction and an enhancementin the backward (gold) direction [1-3]. Multiple theories exist that can explain the observed sup-pression and enhancement, but a conclusive measurement discriminating between the di ff erentmechanisms has yet to be carried out. Two new forward electromagnetic calorimeters (MuonPiston Calorimeters or MPCs, − . < η < − .
1, 3 . < η < .
9) were recently installed in thePHENIX experiment allowing study of parton densities at low x . The MPCs make it possible tomeasure nuclear modification factors in the forward and backward directions as well as azimuthalcorrelations of di-hadron pairs at di ff erent pseudorapidities. In these proceedings, we present theresults of the first correlation measurements made with the new MPCs. The analysis presentedis based on the ≈ nb − integrated luminosity data sample of d + Au collisions at √ s NN = GeV taken at RHIC in 2008. The correlation measurements are especially interesting because itis expected that they provide a strong test of gluon saturation at low x in the Au nucleus [4], [5].Two-particle correlations have previously been measured at PHENIX in √ s = GeV d + Au and p + p collisions for two charged hadrons ( h ± ) detected by the central arm spectrometers( | η , η | < .
35) [6] and for rapidity separated particles where one particle is a h ± in the centralspectrometer ( | η | < .
35) and the other particle is a punch-through hadron [3] detected in theforward or backward muon spectrometer (1 . < | η | < .
0) [7]. Throughout these proceedings“forward” (“backward”) refers to the direction of the deuteron (gold) beam.The correlation measurement is performed with a h ± or a π in the mid-rapidity detectors( | η | < .
35) and a π detected in the forward MPC. The forward π serves to lower the x atwhich we probe the gold nucleus and to provide a larger rapidity gap ( ∆ η ≈ .
5) than has beenobserved previously ( | ∆ η | ≈ . h ± / punch-through hadron correlations). Preprint submitted to Nuclear Physics A October 26, 2018 a r X i v : . [ nu c l - e x ] S e p . Muon Piston Calorimeters The MPCs are PbWO electromagnetic calorimeters that add to the PHENIX calorimeteracceptance in the interesting forward and backward regions; one MPC is installed in the northmuon piston hole (3 . < η < .
9) and the other in the south ( − . < η < − . π . The towers have lateral dimensions of 2 . cm × . cm . The MPCs areinstalled 220 cm from the nominal interaction point [8].In any electromagnetic calorimeter, if the momentum of a π is su ffi ciently high, the energyfrom both decay photons will be reconstructed as a single cluster. For the MPC, this merginge ff ect dominates at and above p tot = GeV / c . Hence, below 20 GeV / c , π s are identified bythe invariant mass spectrum of all photon pairs; while above 20 GeV / c , π s are identified usingsingle clusters.To identify π s from two photons, we use the following set of cuts: 7 GeV < E γγ < GeV , a cluster separation cut of | ∆ r | > . cm , an energy asymmetry cut on the two clusters of α = E − E E + E < .
6. To identify π s using single clusters, we specify that 20 GeV < E γγ < GeV .Sample invariant mass plots are shown in Fig. 1. (a) (b) (c)Figure 1: North MPC invariant mass distributions of photon pairs (black) and the background distributions obtainedusing an event mixing method (red) for 0 . GeV / c < p T < . GeV / c for (a) p + p, (b) d + Au 60-88% centrality bin, (c) d + Au 0-20% centrality bin.
3. Data Analysis
In the correlation analysis, the mid-rapidity particle is the trigger particle, and the forwardparticle is the associate particle [9]. A peak at ∆ φ = ∆ φ distributionsbecause the particles are separated in rapidity by approximately 3.5 units, which is wider thanthe width of the near side jet structure. Hence only an away side peak is expected at ∆ φ = π .The correlation function, CF ( ∆ φ ), is the ∆ φ distribution of the two particles corrected for thenonuniform detector acceptance ( acc ( ∆ φ ) is the two-particle ∆ φ distribution where the particlesare from di ff erent events), or CF ( ∆ φ ) = acc ( ∆ φ ) × dN measured ( ∆ φ ) d ( ∆ φ ) [9]. Example acceptance-corrected ∆ φ correlation functions from this analysis are shown in Fig. 2. The correlation functions arefit as a Gaussian di-jet signal on top of a constant background. Two interesting quantities tocompare d + Au with p + p are the width of the Gaussian peak and the conditional (or per-trigger)yield, CY . The conditional yield is the e ffi ciency-corrected pair-yield of particles produced pertrigger particle detected, or 2 a) (b) (c)Figure 2: Example ∆ φ correlation functions for trigger π ( | η | < .
35) 2
GeV / c < p T < GeV / c , associate π (3 . <η < .
9) 0 . GeV / c < p T < . GeV / c for (a) p + p, (b) d + Au 60-88% centrality bin, (c) d + Au 0-20% centrality bin. CY = (cid:82) π d ( ∆ φ )( CF ( ∆ φ ) − bg ( ∆ φ )) N trig × (cid:15) (1)where N trig is the number of trigger particles, (cid:15) is the detection e ffi ciency of the associate parti-cle, and bg ( ∆ φ ) is the constant combinatorial background determined by fitting the correlationfunction. We then form a ratio of the CY s for d + Au and p + p which is the nuclear modificationfactor I dA : I dA = CY dA CY pp (2) (a) (b)Figure 3: ∆ φ width vs. p T for mid-rapidity π / forward MPC π correlations. (a) mid-rapidity π p T = − GeV / c . (b) mid-rapidity π p T = − GeV / c . . Discussion The correlation widths are shown in Fig. 3. To extend the p T range of the associate particleboth the MPC π (two-photon identification) and MPC cluster widths are shown. The widthsdecrease with increasing p T as expected from jet fragmentation. Within the precision of statisti-cal and systematic errors, little variation can be seen when comparing the correlation widths ford + Au (central or peripheral) and p + p. Figure 4: (a) I dA vs. N coll for mid-rapidity hadron / forwardMPC π correlations. On the other hand, I dAu (see Fig. 4,formed only for MPC π s and not clusters)shows a suppression with increasing collisioncentrality for both species of trigger particles( π s, h ± , | η | < .
35) that is significant.Multiple theories are being used to ex-plain these results (especially I dA ); the mostnotable are the Color Glass Condensate(CGC) model for gluon saturation [4, 5] andpQCD based pictures such as non-leadingtwist shadowing used by Vitev [10]. It will beinteresting to see if the d + Au forward RHICcorrelation results will be able to distinguishbetween the above models.This interesting result is an exciting startto a set of PHENIX di-hadron correlationsthat will span di ff erent total rapidities (utiliz-ing all PHENIX detectors) and rapidity gaps,allowing studies of the e ff ects over varyingranges of x . Acknowledgments
This work is supported by NSF PHY 0601067 and by the Department of Energy whichoperates RHIC and PHENIX.
References [1] J. Adams, et al. (STAR), Phys. Rev. Lett. 97, 152302 (2006).[2] I. Arsene, et al. (BRAHMS), Phys. Rev. Lett. 93, 242303 (2004).[3] S. S. Adler, et al. (PHENIX), Phys. Rev. Lett. 94, 082302 (2005).[4] D. Kharzeev, E. Levin and L. McLerran, Nucl. Phys. A748, 627-640 (2005).[5] L. McLerran and R. Venugopalan, Phys. Rev. D 49, 2233 (1994); Phys. Rev. D 49, 3352 (1994).[6] S. S. Adler et al. (PHENIX), Phys. Rev. C 73, 054903 (2006).[7] S. S. Adler et al. (PHENIX), Phys. Rev. Lett. 96 222301 (2006).[8] M. Chiu. nucl-ex / / et al. (PHENIX), Phys. Rev. C 73, 054903 (2006).[10] J. Qiu and I. Vitev, Phys.Lett. B632, 507 (2006) [hep-ph /0405068].