Table Of ContentSingle Spin Transverse Asymmetries of Neutral
Pions at Forward Rapidities in √s = 62.4 GeV
Polarized Proton Collisions in PHENIX
Mickey Chiu for the PHENIX Collaboration
7
0 Dept.ofPhysics,UniversityofIllinois,Urbana,IL61801USA1
0
2
Abstract. In the RHIC run of 2006, PHENIX recorded data from about 20 nb 1 of transversely
n −
polarized p+p collisions at √s = 62.4 GeV and polarization of about 50%. Also in this last run
a
J PHENIX successfullycommissioneda newPbWO4 basedelectromagneticcalorimeter,theMuon
PistonCalorimeter(MPC),covering2p inazimuthand3.1<h <3.7.Theforwardcoverageallows
2
2 PHENIX to measure high xF p 0 production.We presentthe currentstatus of the analysis for the
singleinclusivep 0transverseasymmetryAN atforwardrapiditiesinPHENIX.
1
Keywords: Protonspin,Singletransversespinasymmetry
v
PACS: 13.85.Ni,13.88.+e,14.20.Dh
1
3
0
1 INTRODUCTION
0
7
Transverse single spin asymmetries in hadronic collisions have had a long history of
0
/ surprises, such as the observation by the E704 collaboration of very large asymmetries
x
e in inclusive pion production at high xF in √s = 19.4 GeV transversely polarized p↑p
- collisions[1, 2]. Naively in leading twist perturbative QCD one expects these asymme-
l
c tries to be powersuppressed such that A a m /p [3]. It was thought that at higher
u N S q T
∼
n energies these asymmetries would be strongly suppressed. Quite surprisingly, however,
: largeasymmetrieswerediscoveredtopersistevenatcolliderenergiesbyfirstSTARand
v
i thentheBrahmscollaboration[4, 5].
X
These asymmetries are interesting because they point toward some new understand-
r
a ingofinternalprotonstructure,suchastheexistenceofanewnon-perturbativefunction
liketheSiversfunction[6],orperhapsanon-zerotransversitydistributioninconjunction
with a transversely dependent fragmentation function[7] (Collins effect). Of particular
interest is that these effects occur in a regime where NLO pQCD calculations correctly
predict the unpolarized cross-sections. Therefore the hope is that one can understand
theseasymmetriesunambiguously,suchaswithahighertwistexpansionapproach[8,9],
orbytakingintoaccountthetransversemotionofpartonsinaLOcollinearlyfactorized
approach with non-perturbative, transverse-momentum dependent functions[10]. Fur-
therprogressintheoreticalunderstandingrequiresmoredifferentialmeasurementssuch
as the p and x dependenceoftheasymmetry,as wellas di-hadron correlations[11]to
T F
distinguishbetween theaboveapproaches.
1 Presentaddress:BrookhavenNationalLaboratory,Upton,NY11375
MPC DATA ANALYSIS AND CURRENT STATUS
Before the 2006 run, PHENIX initiated a program to install a calorimeter inside the
hole at the front of the muon magnet piston2 to extend the kinematic reach of PHENIX
calorimetry to more forward rapidities.
The muon piston holes are cylindri-
cal with a depth of 43.1 cm and a di-
ameter of 42 cm. The small size of
the area, proximity to the interaction
point, and sizable magnetic fields en-
force tight constraints on the calorime-
ter’s design, requiring a compact ma-
terial with short radiation length and
small moliere radius and a readout
that is insensitive to magnetic fields.
PHENIX chose a design, based on that
1:ThePHENIXSouthMPCLayout.
of the ALICE PHOS detector[12], of a
highlysegmentedlead-tungstatecrystal
arraywithAvalanchePhotodiode(APD)readout.Lead-tungstateisoneofthebestcandi-
datematerialsforacompactcalorimetersinceithasoneofthesmallestradiationlength
(0.89cm)and moliereradius (2.0 cm)ofany knownscintillator.PHENIX installed192
crystalsofsize2.2 2.2 18cm3 inthesouthpistonholeintimeforthe2006RHICrun.
× ×
Thecalorimetersitsaroundthebeam-pipe223cmfromtheinteractionpointandcovers
3.1<h <3.7. The layout of the south MPC is shown in figure 1. Another220 crystals
willbeinstalledin thenorth pistonholefor2007. Further detailson theMPC hardware
can befoundinthecontributionofA. Kazantsevintheseproceedings[13].
Thedatashownhereweretakenwith
the PHENIX detector at RHIC during
twodaysofthe2006 p prun,atabeam 1400
↑
energy of √s=62.4 GeV. A total inte- 1200
grated luminosityof 20 nb 1 of data
− 1000
∼
were collected. The data are triggered
800
with the MPC using 4x4 tower energy
600
sums at a threshold of 5 GeV, cor-
∼ 400
responding to good trigger efficiency
for p 0 at 10 GeV. An additional 80 200
nb 1 ofda∼tawere takenwithlongitudi- 00 0.2 0.4 0.6 0.8 1
− inv. mass mg g (GeV)
nallypolarizedbeamsatthesamebeam
energy[13].
2: Theinvariantmassdistributionofphotonpairs
The MPC clustering algorithm
in the South MPC for p+p collisions at √s =
is based on that developed for the NN
62.4 GeV. The mixed eventdistribution is shown
PHENIX central arm calorimeter. In
inthereddashed-dottedline.
this algorithm the known shower shape
from test-beam measurements is used to reconstruct the hit location and energy of
photonswhen thetwophotonsareseparated enough toseetwo peaks. Alternativealgo-
rithmsarebeinginvestigatedwhichwouldhavegreatereffectivenessathigherenergies,
2 This idea was first suggested by K. Imai and Y. Goto in 1996, and rediscovered recently. See
http://www.phenix.bnl.gov/WWW/publish/goto/Polarimetry/pol1.html
where the probability for the two decay photons to merge within the same or adjacent
towersislarge, and itbecomesdifficultto separateoneversustwophotonshowers.The
algorithmsareevaluatedbysimulation.Infigure2theinvariantmassdistributionmgg of
clusterpairsisshown,alongwithpairsgeneratedforclustersfrommixedevents.Mixing
clusters from different events provides an estimate of the combinatorial background.
Notethat a region in the lowerright quadrant ofthe MPC was excluded in this plot due
to issues with electronics noise. Currently, the energy scale has been checked with the
MIPpeak andp 0 peak andbothhavebeen foundto beconsistenttobetterthan 10%.
Thesingletransversespinrawasymmetryasafunctionoff wascalculatedusingthe
square-rootformula[14, 15]
A(f )= 1 e (f )= 1 ds ↑,f −ds ↓,f = 1 qNL↑,f NR↓,f −qNL↓,f NR↑,f (1)
PB N PBds ↑,f +ds ↓,f PBqNL↑,f NR↓,f +qNL↓,f NR↑,f
which largely cancels out differences in detector and beam asymmetries. f is the az-
imuthalanglerelativetothebeampolarizationdirection,andds ,f representsthecross-
↑
section into the f direction for a vertically polarized beam with polarization P . The
B
asymmetryA is thentheamplitudeofthisasymmetrymodulation,A(f )=A sin(f ).
N N
In figure 3 we have plotted on the left (right) the raw asymmetry e (f ) for photon
N
pairs within the p 0 mass peak for the cases where the yellow (blue) beam is polarized.
InPHENIXtheyellowbeamistheprotonbeamheadingtowardstheMPC,andtheblue
beamis theoneheadingaway.
0.04 p 0 ˛ , forward x 0.04 p 0 ˛ , backward x
N F N F
0.03 0.03
0.02 0.02
0.01 0.01
0 0
-0.01 -0.01
-0.02 -0.02
-0.03 -0.03
-0.04 -0.04
-0.05 -0.05
-1.5 -1 -0.5 0 0.5 1 1.5 -1.5 -1 -0.5 0 0.5 1 1.5
f f
3: Raw asymmetries from reconstructed photon pairs within 0.05 < mgg < 0.25 GeV and
9 < Egg < 12 GeV. The asymmetry has not been corrected for combinatoric background or
beampolarization.
There is a non-zero asymmetry for the case where the yellow beam is polarized,
corresponding to positive x with respect to the polarized beam. This asymmetry has
F
been seen to increase from low x to higher x , and is well described by a sin(f )
F F
functionalform.Forthecaseof thepolarized bluebeam (correspondingtonegativex )
F
the asymmetry is consistent with zero. As an additional cross-check, the asymmetries
were found to be consistent with zero for the longitudinallypolarized runs, as expected
sincethen theresidual transversepolarizationissmall.
CONCLUSIONS AND FUTURE PLANS
PHENIXhassuccessfullycommissionedanewforwardelectromagneticcalorimeter,the
south MPC, during the 2006 RHIC run, and plans to install the north MPC in time for
2007.Withanearlyversionoftheclusteringcode,p 0’shavebeenidentifiedintheMPC
andanon-zerotransverseasymmetryhasbeen seenin p p collisionsat√s=62.4GeV.
↑
Muchworkremainsinimprovingthereconstructioncodeandinunderstandinganyother
artifactsofthedata,suchasnoiseissues,backgrounds,andimprovingthedetermination
oftheenergyscale,beforePHENIXcnafinalizethecross-sectionandasymmetryresults
from the MPC. The validity of a perturbative theoretical interpretation of this data-set
is currently under study. It is known that NLO pQCD fails to correctly describe the
inclusivecross-sectioncorrectlyattheseenergiesintheforwardrapidities[16].However,
includingthresholdresummationshouldimprovetheagreement[17].
Besides thetransversely polarized data at 62.4 GeV, PHENIX also collected 80 nb 1
−
oflongitudinallypolarizeddataat√s=62.4GeVandanother7.5 pb 1oflongitudinally
−
polarized data at √s = 200 GeV during the 2006 run. The installation of the MPC
allows PHENIX to confirm and contribute to previous measurements at high x in
F
RHIC collisions. Beyond mapping out the x and p dependence of A , the high rate
F T N
capabilitiesofthePHENIXDAQwillallowexcellenttriggeringcapabilityfordi-hadron
correlation measurements between forward p 0 and mid-rapidity hadrons. Di-hadron
measurementsare particularly interesting sincethey shouldhelp to distinguishbetween
an asymmetry from an initial state effect (Sivers) or a final state effect (Collins plus
transversity).Additionally,di-hadron correlations constrain thekinematicsof theinitial
scattered partons and therefore provide greater sensitivity to the sampled momentum
fractionx ofthosepartons.
REFERENCES
1. D.L.Adams,etal.,Phys.Lett.B261,201–206(1991).
2. D.L.Adams,etal.,Phys.Lett.B264,462–466(1991).
3. G.L.Kane,J.Pumplin,andW.Repko,Phys.Rev.Lett.41,1689(1978).
4. J.Adams,etal.,Phys.Rev.Lett.92,171801(2004),hep-ex/0310058.
5. F.Videbaek,AIPConf.Proc.792,993–996(2005),nucl-ex/0508015.
6. D.W.Sivers,Phys.Rev.D41,83(1990).
7. J.C.Collins,Nucl.Phys.B396,161–182(1993),hep-ph/9208213.
8. J.-w.Qiu,andG.Sterman,Phys.Rev.D59,014004(1999),hep-ph/9806356.
9. Y.Koike,AIPConf.Proc.675,449–453(2003),hep-ph/0210396.
10. M.Anselmino,etal.,Phys.Rev.D73,014020(2006),hep-ph/0509035.
11. D.Boer,andW.Vogelsang,Phys.Rev.D69,094025(2004),hep-ph/0312320.
12. CERN/LHCC99-4,ALICETDR2(1999).
13. A.Kazantsev,theseproceedings(2006).
14. G.G.Ohlsen,andP.W.Keaton,Nucl.Instrum.Meth.109,41–59(1973).
15. H.Spinka,ArgonneNationalLaboratoryReportANL-HEP-TR-99-113(1999).
16. C.Bourrely,andJ.Soffer,Eur.Phys.J.C36,371–374(2004),hep-ph/0311110.
17. D.deFlorian,andW.Vogelsang,Phys.Rev.D71,114004(2005),hep-ph/0501258.
