aa r X i v : . [ nu c l - e x ] M a y SPIN RESULTS FROM the PHENIX DETECTOR AT RHIC
ASTRID MORREALE for the PHENIX Collaboration
The polarized proton beams at the Relativistic Heavy Ion Collider at Brookhaven NationalLaboratory provide a unique environment to observe hard scattering between gluons andquarks. The PHENIX experiment has recorded collisions at p ( s NN ) =200 GeV and 62.4GeV to yield data which are complementary to those measured by deep inelastic scatteringexperiments. Polarized proton-proton collisions can directly probe the polarized gluon andanti-quark distributions as the collisions couple the color charges of the participants. ThePHENIX detector is well suited to measure many final-state particles sensitive to the proton’sspin structure. We will give a brief overview of the PHENIX Spin Program and we will reportresults, status and outlook of the many probes accessible to the PHENIX experiment whichwill be incorporated into future global analyses of world data on polarized hard scattering. The proton is a composite particle made of more fundamental subatomic entities called quarksand gluons which generate the observed quantum mechanical spin 1/2 properties. spin 1/2, isone of the most important cases to explain nuclear interacting matter and thus probes underlyingtheoretical structures deeply. A surprising result was found in the late 1980’s by the EuropeanMuon Collaboration (EMC) at CERN that the spin of the quarks contributed a very smallfraction to the proton’s spin. The original EMC publication1 that triggered the spin crisis hassince resulted in a large theoretical and experimental effort to find the pieces to the proton’sspin puzzle, mainly, what role gluons, sea quarks and orbital angular momentum (OAM) play.This can be summarized in the helicity sum rule (Eq. 1) where ∆(u, d, s, G) are the probabilitiesof finding a q, ¯ q or gluon with spin parallel or anti-parallel to the spin of the nucleon and Lis the OAM of the parton. Questions which spin physics experiments such as PHENIX aimto investigate, include the role played by the strong force mediators (gluons), the role of thevirtual quark, antiquark pairs originating from the gluons’ strong interaction field (sea quarks)nd lastly the role of angular momentum to the nucleon spin. S z = 12 = 12 (∆ u + ∆¯ u, +∆ d + ∆ ¯ d + ∆ s + ∆¯ s ) + L q + ∆ G + L G (1)The RHIC spin program complements the work done in DIS by making use of strongly-interacting polarized quark and gluon probes which are sensitive to the gluon polarization ∆G, amajor emphasis of RHIC-Spin2. The transverse spin structure of the proton is also being exploredwith measurements sensitive to the sivers effect, transversity and the collins effect. Currentand future running at √ s = 500 GeV will focus at disentagling quark and anti quark spin-flavor separation in W-Boson production thus measuring the flavor asymmetry of the polarizedantiquark sea.2Polarizing protons is not a trivial task; maintaining proton polarization is a challenge dueto the proton’s large anomalous magnetic moment. The novel use of RHIC’s siberian snakes,cancells out major depolarizing spin resonances in the beams and a stable spin direction can beobtained perpendicular to the direction of the beam.3The PHENIX detector located at the 8 o’clock position at RHIC has two central armswith pseudorapidity acceptance of | η | < .
35. These are equipped with fine-grained calorimetry100 times finer than previous collider detectors, making particle identification excellent, thegranularity of the electromagnetic calorimeter (EMCal) is ∆ η × ∆ φ = 0 . × . .
4. Triggeringin the central arms allow us to select high p T γ , e ± and π ± . The PHENIX muon arms cover1 . < η < .
4, they surround the beams and include µ ± identifiers, tracking stations and ironsheets with detectors in the gaps in each sheet. ∆ G Measurements sensitive to ∆G have been part of the main goals of the PHENIX spin program.With the use of factorization, a differential cross section can be written as the convolutionof a parton density function (pdf) and a hard scattering process. Factorization along withuniversality of pdf’s and fragmentation functions (FF, D h ) allows for separation of long andshort distance contributions in the cross sections. These assumptions, allow predictions whichdepend both in experimental measurements and theoretical calculations. Experimentally, whatcan be measured are asymmetries: the ratio of the polarized to unpolarized cross sections (Eq.2). Asymmetries give an elegant way of accessing parton information in a factorized framwork,by counting observed particle yields in different helicity states of incident protons (++ , −− vs+ − , − +) normalized by the polarization in each beam ( P B,Y ) A LL = X a,b,c =q , ¯q , g ∆ f a ⊗ ∆ f b ⊗ ∆ˆ σ ⊗ D h/c X a,b,c =q , ¯q , g f a ⊗ f b ⊗ ˆ σ ⊗ D h/c = σ ++ − σ + − σ ++ + σ + − ,A LL = 1 P B P Y N ++ − RN + − N ++ + RN + − , R = L ++ L + − . (2) A LL results Masuring A LL in certain final states is a valuable tool to measure polarized gluon distributionfunctions in the proton. The most accurately way to do so is to study those processes whichcan be calculated in the framework of pQCD. PHENIX unpolarized π γ production cross sections at mid rapidity have shown that the next to leading order (NLO)perturbative calculations describe the data well at RHIC energies. A variety of hadron, lepton -0.04-0.0200.020.040.060.080.1 T x Run2006 62.4GeVRun2005 200GeV A LL G R S V - m a x ( G e V ) G R S V - s t d ( G e V ) G R S V - s t d ( G e V ) G R S V - m a x ( G e V ) (62.4GeV) T p (200GeV) T p Figure 1: Left: π A LL ( x T ) at √ s =62.4 GeV and 200 GeV. Right: π ( p T ) cross section (GeV/c) T p2 4 6 8 10 12 π LL A −0.0200.020.040.06 (a) (GeV/c) T p1 1.5 2 2.5 3 π LL A −0.00500.005 G=0.24) ∆ GRSV std (G=0 ∆ GRSV G=−1.05 ∆ GRSV Run−5Run−6 ) =4 GeV µ ( GRSVx=[0.02,0.3] G ∆ −1.5 −1 −0.5 0 0.5 1 1.5 χ χ∆ =9 χ∆ G=−1.05" ∆ " G=0" ∆ " "std" Stat OnlyR ∆ + R ∆ − P ∆ + P ∆ − Figure 2: Left: π A LL ( p T ) Right: χ derived by comparing the measurement to a range of solutions provided byGRSV, for which the value of the integral ∆G was constrained and the lepton scattering data refit. and photon probes have been measured at PHENIX. π
10 asymmetries (Fig. 2) with measure-ments at √ s = 200 GeV, and 62.4 GeV (Fig. 1) have now been included for the first time ina NLO global analysis. 5 While π ’s have significantly constrained ∆G in a limited x range,large uncertainties remain and sign information of ∆G is still unknown. Measurements of π ± are an independent probe which are sensitive to the sign, and magnitude of ∆G: quark-gluon(qg) scattering dominates mid-rapidity pion production at RHIC at transverse momenta above5 GeV/c. Preferential fragmentation of up quarks (u) to π + , and down quarks(d) to π − , leads tothe dominance of u-g, and d-g contributions. This dominance of u or d combined with the differ-ent signs of their polarized distributions translates into asymmetry differences for the differentspecies π + , π , and π − that depend on the sign of ∆G. For example, a positive ∆G could beindicated by an order of π asymmetries, i.e: A LL ( π + ) > A LL ( π ) > A LL ( π − ), and viceversa fora negative contribution. Particle cluster asymmetries, as well as A LL of η have been calculated.The recent preliminary extraction of η ’s D η , has allowed for A LL theory comparisons. D η shows Figure 3: Left: π + A LL ( P T ), Right: π − A LL ( P T ). √ s =200GeV.igure 4: (left) η A LL ( P T ) at sqrts γ A LL ( p T ) at √ s =200GeV a slight enhanced sensitivity to qg when compared to π . Observation of difference in asymme-tries could help disentangle the contributions from the different quarks and gluons.(Fig. 4) Rarechannels measured at PHENIX include: A LL of µ ± coming from J/ψ production, e ± A LL comingfrom heavy quarks, and prompt γ (Fig. 4.) Prompt γ are a clean elegant probe dominated byqg compton and thus can give a better access to ∆G, however as is the case for rare probes,these measurements require high luminosities, not yet achieved at RHIC. PHENIX has measured small single spin asymmetries (SSA) A N of π and h ± at small p T for | η | < .
35 and has helped constrain the magnitude of the gluon Sivers function[8]. Incontrast, measured A N at forward X F show large asymmetries in the the positive but not thenegative X F region. These interesting measurements may provide quantitative tests for theoriesinvolving valence quark effects. Other measurements include the A N of J/ Ψ which may alsobe sensitive to g-Sivers via D Meson production as its produced from g-g fusion, neutron’s A N ,di-hadron interference fragmentation function (IFF) asymmetries and k T asymmetries whichaim at probing orbital angular momentum[9] PHENIX is well suited to the study of spin structure of the proton with a wide variety of probes.A variety of new results aiming to disentangle spin partonic contributions are emerging. In thenext coming years, statistics needed to explore different channels for different gluon kinematicsand different mixtures of subprocesses will become available and allow a more accurate pictureof the spin of the proton.
1. Ashman J, et al
Phys. Lett.
B206 (1988) 364.2. G. Bunce et al.,
Annu. Rev. Nucl. Part. Sci. (2000) 525.3. Alekseev I, et al. Polarized RHIC (1998.)4. K. Adcox et al.,
Nucl. Inst. Meth. A (2003) 469.5. D. de Florian et al, arXiv:0804.0422 (2008)6. M. Gl¨ uu