Recent PHENIX Spin Recent PHENIX Spin ResultsResults
Astrid MorrealeAstrid Morreale
University of California at RiversideUniversity of California at Riverside
On behalf of the collaborationOn behalf of the collaboration
WWND, April 12, 2008
2
OutlineOutline
About About polarized polarized RHIC, PHENIXRHIC, PHENIX The Proton Spin Structure via The Proton Spin Structure via
AsymmetriesAsymmetriesRecent PHENIX results from polarized Recent PHENIX results from polarized
proton running proton running (longitudinal, transverse if time (longitudinal, transverse if time permits)permits)
Summary Summary
3
RHICRHIC
Acceleration of polarized protons between 25 and 250 GeVAcceleration of polarized protons between 25 and 250 GeV Up to 120 Bunches with 2*10Up to 120 Bunches with 2*101111 protons (100ns apart) protons (100ns apart) Polarization up to 70%Polarization up to 70% http://www-nh.scphys.kyoto-u.ac.jp/SPIN2006/SciPro/pres/plenary/MeiBai.ppthttp://www-nh.scphys.kyoto-u.ac.jp/SPIN2006/SciPro/pres/plenary/MeiBai.ppt
4
scatteredproton
recoilCarbon
polarizedbeam
Carbontarget
PolarimetryPolarimetry ““CNI”-Polarimeter measures left-right asymmetries in CNI”-Polarimeter measures left-right asymmetries in
elastic pC collisionselastic pC collisionso Provides fast relative polarization measurementProvides fast relative polarization measuremento Scan polarization over x and y range of the beamScan polarization over x and y range of the beam
270.352002003 *
400.122002004 *
48
55
49
Polarization [%]
0.0862.42006 *
7.52002006 *
3.42002005 *
Luminosity [pb-1] (recorded)[GeV]Year
Forward scattered proton
proton target
proton beam
““Jet” Polarimeter measures left-right asymmetries in elastic pp Jet” Polarimeter measures left-right asymmetries in elastic pp collisions collisions
oUses the known polarization of H atoms from Atomic Beam SourceUses the known polarization of H atoms from Atomic Beam Source
oAbsolute measurementAbsolute measurement
oIntegrating over beam profileIntegrating over beam profile
oRun 6: total syst. polarization uncertainty Run 6: total syst. polarization uncertainty PPBBPPYY/(P/(PBBPPYY)=8.4%)=8.4%
5
The PHENIX Detector for Spin PhysicsThe PHENIX Detector for Spin PhysicsCentral Detector Central Detector Acceptance: (|Acceptance: (|||x x ::• High efficiency High efficiency trigger trigger • detectiondetection
– Electromagnetic Calorimeter:Electromagnetic Calorimeter: PbSc + PbGl, PbSc + PbGl, < |0.35|, < |0.35|, = 2 x 90 = 2 x 90
• ee++/e/e--
– Drift ChamberDrift Chamber– Ring Imaging Ring Imaging ČČerenkov erenkov
Detector(RICH) Detector(RICH)
Muon Arms Muon Arms (forward kinematics (~1.1<| (forward kinematics (~1.1<| |<2.4) |<2.4)::
•JJ–Muon ID/Muon Tracker (Muon ID/Muon Tracker ())
•
–Electromagnetic Calorimeter (MPC)Electromagnetic Calorimeter (MPC)
Global Detectors:Global Detectors:
•Relative LuminosityRelative Luminosity
–Beam-Beam Counter (BBC) Beam-Beam Counter (BBC)
Zero-Degree Calorimeter (ZDC)Zero-Degree Calorimeter (ZDC)
•Local Polarimetry - ZDCLocal Polarimetry - ZDC
Intrinsic Spin Violates our intuition:
How can an elementary particle such as the e¯ be point like and have perpetual angular momentum.?
The Proton also violates our intuition.The Proton is composed of quarks, gluons and anti quarks.
We should expect the proton's spin to be predominatelycarried by its 3 valence quarks
6
That nucleon has a large anomalous magnetic moment proves that this is not a fundamental spin1/2 Dirac particle.
Within the nucleon:Quarks, gluons and their angular momentum caused by their high speed motion within the nucleon are contributors to the Nucleon's spin.
8
Contributions to the Proton's Contributions to the Proton's Spin Spin
Quarks and Gluons carry about 50%(each) of the Quarks and Gluons carry about 50%(each) of the longitudinal momentumlongitudinal momentum
What about Spin?What about Spin?
Valence Quarks (QPM) Valence Quarks (QPM) ~30%~30%
((QQCCDD) Gluons, Sea Quarks: ) Gluons, Sea Quarks: ~>, =, ~>, =, <0?<0?
Orbital Angular Momentum Orbital Angular Momentum ~?~?
9
The Spin Structure of the ProtonThe Spin Structure of the ProtonLongitudinal Spin Sum Rule:
Transverse Spin Sum Rule?
12=S z=
12
ΔΣ+ΔG+L z
12=S x=
12δΣ+L x
Double Spin Asymmetries (pp,SIDIS)
W-production (pp)
Exclusive processes (DVCS,etc)
Chiral-odd Fragmentation functions (Collins,IFF,L)
Sivers effect??
ΔG, Δq=are the probabilities of finding a parton with spin parallel or anti parallel to the spin of the nucleon.
10
Accessing Accessing g with Asymmetriesg with Asymmetries
I Hard subprocess asymmetries (LO)
Increasing x, pT
1 2LL LL
1 2
g(x ) q(x )ˆA a (qg q )
g(x ) q(x )
ˆ ˆ
11
AsymmetriesAsymmetries For our For our g program the tools are measurements of g program the tools are measurements of
helicity cross section asymmetries Ahelicity cross section asymmetries ALLLL
Bunch spin configuration alternates every 106 ns
Data for all bunch spin configurations are collected at the same time
Possibility for false asymmetries are greatly reduced
(N) Yield (R) Relative Luminosity
BBC vs ZDC(P) Polarization
RHIC Polarimeter (at 12 o’clock)Local Polarimeters (SMD&ZDC)
12
AsymmetriesAsymmetriesAccessing Accessing g: g: Inclusive channels AInclusive channels ALLLL(pp(ppAX) AX)
measurable at Phenixmeasurable at Phenix 00: wide p: wide pTT range, mixture with gg range, mixture with ggX dominant at low X dominant at low
ppTT , similar to , similar to 0 , 0 , different FF’s. different FF’s. , mixture sensitive to qg, mixture sensitive to qgqX at high pqX at high pTT
o Multiparticle clusters (parts of jets), correlated with Multiparticle clusters (parts of jets), correlated with 0,0,
o Direct photons: pDirect photons: pTT range 6-20+ GeV/c, dominated by range 6-20+ GeV/c, dominated by qgqgqq
o J/J/, , , e, e (gg (ggcc)cc)
LLzz : Current k : Current kT,T, , D , DLL, A, AN,UT, TN,UT, T
measurements at Phenixmeasurements at Phenixo AANN π π00/π/π//hh, J/, J/forward neutrons forward neutrons o DLL Anti-Λ Spin Transfero kkTT azimuthal di-h azimuthal di-h correlations correlations
13
00 cross section cross section measurementmeasurement
Agreement between Agreement between data and pQCD data and pQCD theorytheory
Shows that pQCD Shows that pQCD and and unpolarizedunpolarized PDFs determined in PDFs determined in DIS can describe pp DIS can describe pp datadata
Choice of Choice of fragmentation fragmentation function crucial function crucial (dominated by gluon (dominated by gluon fragmentation)fragmentation)
Scale uncertainty Scale uncertainty still large at lower still large at lower ppTT<5 GeV<5 GeV
PHENIX: 0 mid-rapidity, 200GeVhep-ex-0704.3599
14
Measured asymmetries for pp0X from Run 5 ,Run 6
Run3,4,5: PRL 93, 202002; PRD 73, 091102; hep-ex-0704.3599
00 Asymmetries Asymmetries
Asymmetry of combinatorial Asymmetry of combinatorial background estimated from background estimated from sidebands and subtractedsidebands and subtracted
Initial state parton configurations contributing to unpolarized cross
section (Fractions)
W. Vogelsang et al. Dominated by gg for pT<3, qg
for 3<pT<10 GeV/c
15
Information from Information from 00 AsymmetriesAsymmetries
vs. xT=2pT/S
Maximum scenarios ruled out -> Constrain of Maximum scenarios ruled out -> Constrain of g(x) g(x) Inclusive Inclusive 00 A ALLLL cannot access cannot access g(x) directlyg(x) directly
o Only sensitive to an average over a wide x range Only sensitive to an average over a wide x range o No conclusions about moment of No conclusions about moment of g(x) possible without a g(x) possible without a
model for its shapemodel for its shape
More (indirect) information from varying cms energiesMore (indirect) information from varying cms energiesoHigher (500 GeV) Higher (500 GeV) lower x lower x
oSmaller (62 GeV) Smaller (62 GeV) higher x (and larger scale higher x (and larger scale uncertainty)uncertainty)
16
+,-+,- Asymmetries Asymmetries
New set from M. Stratmann et al. is the first to use charged separated data from SIDIS for fragmentation functions
Charged pions above 5 GeV/c Charged pions above 5 GeV/c identified with RICH. identified with RICH.
At higher pAt higher pTT, qg interactions , qg interactions become dominant: become dominant: qqg term.g term.
AALLLL becomes significant allowing becomes significant allowing access to the sign of access to the sign of GG
Tp ~ 5GeV
+ 0 -
+ 0 -u u u
LL LL LL
qg starts to dominate for and D D >DExpect sensitivity to sign of G, e.g., positive A A A
Fraction of pion production
17
Information from Information from +,-+,- AsymmetriesAsymmetries
Inclusive Inclusive +,-,0+,-,0 A ALLLL has access to sign has access to sign g(x) directlyg(x) directly “Model independent” conclusion possible once enough
data is available.
18
Recent preliminary extraction of fragmentation functions, Eta has (slightly) enhanced sensitivity to gg (when compared to the 0 ) Observation of difference in asymmetries could help disentangle the contributions from the different quarks and the gluons.
19
AsymmetriesAsymmetries
Dominated by qg Compton:- Small uncertainty from FFs- Better access to sign of G (qG)-Clean “Golden Channel” :-)-Luminosity Hungry :-(
20
Information from Information from ee+,-+,- Asymmetries Asymmetries
Provide access different x Provide access different x rangerange
o Thresholds Thresholds o J/J/range (forward arms)range (forward arms)o PromptPrompt: no fragmentation z=1: no fragmentation z=1o Rare channels with large backgroundRare channels with large backgroundo Need more luminosityNeed more luminosity
prompt photon
cceX
bbeXJ/
xG
(x)
21
G(x) Global Analysis
Results from various channels combined into single results for G(x) Correlations with other PDFs for each channel properly accounted Every single channel result is usually smeared over x global
analysis can do deconvolution (map of G vs x) based on various channel results
NLO pQCD framework can be used Global analysis framework already exist for pol. DIS data and being
developed to include RHIC pp data, by different groups
Now Run5-Final and Run6-Preliminary 0 data are available and has joined this effort
22
G(x) Global AnalysisLatest Results
Global Analysis of Helicity Parton Densities and Their Uncertainties
(de Florian, Sassot, Stratmann and Wogelsang) ArXiv:0804.0422 (April 2008)
-Flavor dependence of the sea -SU3 symmetry breaking?.
“We also find that the SIDIS data give rise to a Robust pattern for the sea polarizations, clearly deviating from SU(3) symmetry, which awaits further clarification from the upcoming W boson Program at RHIC”
23
G(x) Global AnalysisLatest Results
ArXiv:0804.0422 (April 2008)
-A first demonstration that p-p data can be included in a consistent way in a NLO pQCD calculation.
-RHIC data set significantly constraints on the gluon helicity distribution
-”Inclusion of theoretical uncertainties and the treatment of experimental ones should and will be improved”
hep-ex/0507073(hep-ex/0507073)
Transverse SpinTransverse Spin
(Collins effect)spin-dependentfragmentation
functions
(Sivers effect)transversely asymmetric
kt quark distributions
(Twist-3)quark gluon field
Interference
^ q1Tq,f ,L
• Mid-rapidity data at small pT sensitive to gluons, constrains magnitude of gluon Sivers function (Anselmino et al., PRD 74, 2006)•What happens if qq sets in (valence quarks) at high pT?
process contribution to 0, =0, s=200 GeV
PLB 603,173 (2004)p+p0+X at s=200 GeV/c2
PRL 95, 202001 (2005)
Transverse Spin Transverse Spin Mid-rapidity AMid-rapidity ANN of of 00 and h and h for y~0 for y~0 at at s=200GeVs=200GeV
AN is 0 within 1% interesting contrast with forward
AN : h+/h-
26
Transverse Spin Transverse Spin 00 A ANN at large x at large xFF
3.0<<4.0
p+p0+X at s=62.4 GeV
p+p0+X at s=62.4 GeV
Asymmetry seen in yellow beam (positive xF), but not in blue (negative xF)Large asymmetries at forward xF Valence quark effect?xF, pT, s, and dependence provide quantitative tests for theories
process contribution to 0, =3.3, s=200 GeV
PLB 603,173 (2004)
27
Transverse Spin Transverse Spin AANN of of J/J/ at at s=200GeVs=200GeV
Sensitive to gluon Sivers as Sensitive to gluon Sivers as produced through g-g produced through g-g fusionfusion
Charm theory prediction is Charm theory prediction is availableavailableo How does J/How does J/production production
affect prediction?affect prediction?
28
Transverse Spin Transverse Spin Other asymmetries at Other asymmetries at s=200GeVs=200GeV
neutronchargedparticles
Run 5
Strange quark Components via Spin Strange quark Components via Spin TransferTransferIn PHENIX the Self-analyzing decay channel (anti-Λ) has been found to be sensitive to the polarization of the anti-strange sea of the nucleon (See: hep-ph/0511061)
Probing Orbital angular Momentum with kProbing Orbital angular Momentum with kTT Asymmetries Asymmetries (See:(See: Phys. Rev. D 74, 072002 (2006)
Neutron asymmetries. Neutron asymmetries. (See:(See:AIP Conf.Proc.915:689-AIP Conf.Proc.915:689-
692,2007692,2007 ))
Results for <jT2> Results for <kT2>
29
SummarySummary PHENIX is well suited to the study of spin physics with a wide PHENIX is well suited to the study of spin physics with a wide
variety of probes.variety of probes.o Inclusive Inclusive 00 data for A data for ALLLL has reached statistical significance to has reached statistical significance to
constrain ΔG in a limited x-range (~0.02-0.3).constrain ΔG in a limited x-range (~0.02-0.3).
NeedNeed more statistics (RHIC running time) to explore more statistics (RHIC running time) to explore different (rare) channels fordifferent (rare) channels foro Different gluon kinematicsDifferent gluon kinematicso Different mixtures of subprocessesDifferent mixtures of subprocesses
Global Analysis of many channels together with DIS, SIDIS Global Analysis of many channels together with DIS, SIDIS data will give us a more accurate picture of data will give us a more accurate picture of g(x)g(x)
Upcoming Upcoming W programW program will give more information about will give more information about quarksquarks
PHENIX has an upgrade program that will give us the triggers PHENIX has an upgrade program that will give us the triggers and vertex information that we need for precise future and vertex information that we need for precise future measurements measurements
of of G, G, q and new physics at higher luminosity and energyq and new physics at higher luminosity and energy
We Think That We Understand the Concept of “Rotational Motion“
...but how does it workout at scales of about one fermi? (Marco Stratman. Lectures on the Longitudinal Spin Structure of the Nucleon, Wako, Japan)
We Should Measure it and find out!!
THANK YOU For Listening
THANK YOU For Listening
31
ExtrasExtras
1p
Phys Rev D41 (1990) 83; Phys Rev D43 (1991) 261
Top view
^ q1q kpxp
1S
Quark transverse momentum in transversely polarized proton.
Front view
zyx
y
xz
Sivers effect: Initial stateInitial state of the polarized nucleon
x is longitudinal momentum fraction.
1p
Collins Heppelmann effect:Collins Heppelmann effect:Final stateFinal state of fragmentation hadrons of fragmentation hadrons
Example:: Xhhpp 21
Polarization of struck quark which fragments to hadrons.
Nucl Phys B396 (1993) 161, Nucl Phys B420 (1994) 565
'qp
1h
2h
21 hh
1S
Sivers effectSivers effect and/or and/or Collins- HeppelmannCollins- Heppelmann effect? effect?
Theoretical approaches to explain huge SSAs:
Sivers effect ( is connected to quark orbital angular momentum).
Collins effect (Analyzer of transversity q).
Twist3 effect which is related to both initial and final states. Relation of Twist3 to Sivers effect is introduced.
Relevance of Twist3Twist3 and Sivers effect Sivers effect is studied. PRL97, 082002 (2006)PRD73, 094017 (2006)
Available Probes at RHIC
1
pp+p+ph+Xh+X Both mix
2
pp+p+pdi-jet+Xdi-jet+X Sivers?Sivers?
3 p+ph+h+X (far side) Separate?
4
p+ph+h+X (near side)
p+pjet+X
Collins
Sivers
5
p+pdirect +X Sivers
6
p+pl+l (Drell-Yan) Sivers
Collins function: analyzer of “Transversity”
Transversity: ,xq,xq,xq
: Probability to observe parton whose pol. vector is “with” or “against” the proton pol. vector with the renormalization scale .
,xq
q(x,) has not been measured experimentally.
Lattice QCD calculates the first moments of q(x,) for u,d, s quarks and the sum at 2=2 GeV2.
,xq,xqdxq1
0
sdu
“with”
“against”
36
Relative LuminosityRelative Luminosity
Use BBCs at Use BBCs at 1.5 m from the interaction point 1.5 m from the interaction point to measure bunch-by-bunch luminosityto measure bunch-by-bunch luminosityo LLii=N=Nii/(/(Efficiency) , Efficiency) , Eff.=const.=22.9mbEff.=const.=22.9mb9.7%9.7%
Use independent measurement from ZDCs Use independent measurement from ZDCs ((18m) to check for intrinsic luminosity 18m) to check for intrinsic luminosity asymmetryasymmetry
For Run 6 (200 GeV):For Run 6 (200 GeV): AALLLL(BBC-ZDC)=3.8(BBC-ZDC)=3.81.6*101.6*10-4-4 (>2 standard (>2 standard
deviations) deviations) AALLLL=5.4*10=5.4*10-4 -4
o Sizeable asymmetry compared to stat. error of low pSizeable asymmetry compared to stat. error of low pTT datadata
o Can be instrumental or physics effect Can be instrumental or physics effect need to find need to find outout
37
PHENIX: 0 mid-rapidity, 200GeVhep-ex-0704.3599PHENIX: hep-ex-0704.3599
00 cross section and soft cross section and soft PhysicsPhysics
Exponent (eExponent (e--pTpT) ) describes pion cross describes pion cross section at psection at pTT<<1 1 GeV/c (dominated by GeV/c (dominated by soft physics): soft physics): =5.56=5.560.02 0.02
(GeV/c)(GeV/c)-1-1
22/NDF=6.2/3/NDF=6.2/3Assume that Assume that
exponent describes exponent describes soft physics soft physics contribution also at contribution also at higher phigher pTT soft soft physics contribution physics contribution at pat pTT>2 GeV/c is >2 GeV/c is <10%<10%