CLAS CLAS 12 Measurements with Polarized Targets H

CLAS & CLAS 12 Measurements with Polarized Targets H. Avakian, N. Kalantarians Workshop on High Luminosity Polarized Targets for the 12 Ge. V Era JLab, June 18 Jlab, June 18, 2010 1

Outline Transverse structure of the nucleon and partonic correlations • Physics motivation • k. T-effects with unpolarized and longitudinally polarized target data –Double spin asymmetries –Single Spin Asymmetries • Physics with transversely polarized hadrons and quarks –k. T-effects and SSA in pion production –correlations between transverse degrees of freedom • Studies of 3 D PDFs at CLAS at 6 Ge. V and beyond • Summary Jlab, June 18, 2010 2

Structure of the Nucleon Wpu(k, r. T) “Mother” Wigner distributions d 2 r. T TMD PDFs f 1 u(x, k. T), . . h 1 u(x, k. T) d 2 k. T GPDs/IPDs H, E, …HT d 2 k. T d 2 r. T PDFs f 1 u(x), . . h 1 u(x) Analysis of SIDIS and DVMP are complementary Jlab, June 18, 2010 3

k. T-dependent PDFs and FFs: “new testament” Baccetta, Diehl, Goeke, Metz, Mulders, Schlegel EPJ-2007 No analog in twist-2 appear in sinf moment of ALU and AUL quark-gluon-quark correlations responsible for azimuthal moments in the cross section Jlab, June 18, 2010 4

SIDIS: partonic cross sections k. T p┴ PT = p┴ +z k. T Ji, Ma, Yuan Phys. Rev. D 71: 034005, 2005 Is the info on x-k. T correlations accessible in k. T integrated observables? Jlab, June 18, 2010 5

Collins Effect: azimuthal modulation of the fragmentation function FUT∞h 1 H 1┴ y STIntial quark polarization f. S hadronizing quark s. T(q×PT)↔ H 1┴ f. C PT f C s. T D(z, PT)=D 1(z, PT)+H 1┴(z, PT) sin(fh- f. S’) fh f. S’ = p-f. S spin of quark flips wrt y-axis sin(fh+f. S) x PT y f. S = p/2+fh f. S fh x PT PT y f. S=fh fh fh x y f. S=p x HT function related to force on Jlab, June 18, the 2010 quark. M. Burkardt (2008) 6

Sivers mechanisms for SSA PT FS f. S Proton polarization - Correlation between quark transverse momentum and the proton spin fk. T x SIDIS Drell-Yan M. Burkardt (2000) HT asymmetries (T-odd) No leading twist, provide access to quark-gluon correlations Jlab, June 18, 2010 7

Quark distributions at large k. T: lattice B. Musch ar. Xiv: 0907. 2381 Analysis of same distributions using gaussian distributions in terms of f 1&g 1 and q+/q- yield different PDFs. H. H. Avakian, Jlab, Avakian, June JLab, 18, 2010 April May 26 3 8

Quark distributions at large k. T: lattice H. Mkrtchyan et al. Phys. Lett. B 665: 20 -25, 2008. B. Musch ar. Xiv: 0907. 2381 Higher probability to find a d -quark at large k. T Jlab, June 18, 2010 9

Transverse momentum distributions of hadrons Gauss CLAS slightly lower, but may have bigger Anselmino et al from EMC data → = 0. 25 Wider H. Jlab, Avakian, June. JLab, 18, 2010 May 3 at smaller beam energies? 10

CLAS 12 LTCC Track resolution: FTOF dp (Ge. V/c) 0. 003 p + 0. 001 p 2 dq (mr) < 1 df (mr) < 3 DC R 1, R 2, R 3 PCAL Central Detector Wide detector and physics acceptance (current/target fragmentation) EC Solenoid 5 T Forward Detector CTOF SVT HTCC High beam polarization 85% High target polarization 85% Lumi > 1035 cm-1 s-1 Forward carriage TORUS Jlab, June 18, 2010 11

CLAS Longitudinally Polarized Solid Target LHe transfer line He-4 Pump Out He-4 Pumping tube Needle Valves Heat shield Beam 10 cm LHe fill tube Outer vacuum jacket 1 K heat exchanger • • Beam tube separator wave guide Horizontal DNP Polarized Target. Field provided by warm bore solenoid. Center of the target is at 2 m from the solenoid entrance, surrounded by SVT. Beam along axis of refrigerator. 4 He Evaporation Refrigerator. One target cell (no ladder). UVa, ODU, CNU, JLab Jlab, June 18, 2010 12

CLAS 12: Kinematical coverage ep. X SIDIS kinematics Q 2>1 Ge. V 2 W 2>4 Ge. V 2(10) y<0. 85 MX>2 Ge. V Large Q 2 accessible with CLAS 12 are important for separation of HT contributions Jlab, June 18, 2010 13

Longitudinal Target SSA measurements at CLAS (E 05 -113) • Complete azimuthal coverage crucial for separation of sinf, sin 2 f moments W 2>4 Ge. V 2 Q 2>1. 1 Ge. V 2 y<0. 85 0. 4<z<0. 7 MX>1. 4 Ge. V PT<1 Ge. V 0. 12<x<0. 48 Large sinf for all pions and no indication of sin 2 f for p 0 s suggests Sivers dominance Jlab, June 18, 2010 14

ALL PT-dependence in SIDIS E 05 -113 M. Anselmino et al hep-ph/0608048 02=0. 25 Ge. V 2 D 2=0. 2 Ge. V 2 0. 4<z<0. 7 • New experiment with 10 times more data will study the PT-dependence for different quark helicities and flavors for bins in x to check if m 0< m 2 Jlab, June 18, 2010 15

Extracting widths from A 1 Assuming the widths of f 1/g 1 x, z and flavor independent Fits to unpolarized data Anselmino et al Collins et al Jlab, June 18, 2010 16

CLAS transversely polarized HD-Ice target vs std nuclear targets Heat extraction is accomplished with thin aluminum wires running through the target (can operate at T~500 -750 m. K) Pros 1. Small field (∫Bdl~0. 005 -0. 05 Tm) Multiple scattering and attenuation 2. Small dilution (fraction of events from polarized material) in nuclear environment introduces 3. Less radiation length additional PT-dependence for hadrons 4. Less nuclear background (no nuclear attenuation) HD-Ice target at ~2 n. A ~ NH 3 5. Wider acceptance at 5 Na (L~5. 1033 cm-2 s-1) much better FOM, especially for deuteron Cons 1. HD target is highly complex and there is a need for redundancy due to the very long polarizing times (months). p+ yield to deuteron 2. Need to demonstrate that the target can remain polarized forratios long periods with an electron beam with currents of order of 1 -2 n. A 3. Additional shielding of Moller electrons necessary Jlab, June 18, 2010 17

Collins SSAs CLAS E 08 -015 (2011) H. A. , A. Efremov, P. Schweitzer, F. Yuan Anselmino et al helicity-transversity=pretzelosity CLAS with a transversely polarized target will allow measurements of transverse spin distributions and constrain Collins fragmentation function Jlab, June 18, 2010 18

Measurement of Sivers function and GPD-E (DVCS) (SIDIS) CLAS E 08 -015 GPD-E=0 DVCS Transverse asymmetry (function of momentum transfer to proton) is large and has strong sensitivity to GPD-E CLAS will provide a measurements of Sivers asymmetry at large x, where the effect is large and models unconstrained by previous measurements. Meissner, Metz & Goeke (2007) Jlab, June 18, 2010 19

Summary Measurements of azimuthal dependences of double and single spin asymmetries in SIDIS (TMDs) indicate that there are significant correlations between spin and transverse momentum distribution of quarks. Sizable higher twist asymmetries measured in SIDIS indicate the quark-gluon correlations may be significant at moderate Q 2. Upcoming JLab SIDIS experiments at 6 and 12 Ge. V will significantly improve the statistical precision of polarized target data, allowing studies of correlations between longitudinal and transverse degrees of freedom Measurements of TMDs at JLab in the valence region provide important input into the global analysis of Transverse Momentum Distributions (involving HERMES, COMPASS, RHIC, BELLE, BABAR+JPARC, GSI, EIC) Jlab, June 18, 2010 20

Support slides…. Jlab, JLab, June. Nov 18, 25 2010 21

SIDIS (g*p->p. X) x-section at leading twist TMD PDFs • Measure Boer-Mulders distribution functions and probe the polarized fragmentation function • Measurements from different experiments consistent Jlab, June 18, 2010 22

Boer-Mulders Asymmetry with CLAS 12 & EIC 5 -7 Ge. V - e p 50 -150 Ge. V Transversely polarized quarks in the unpolarized nucleon CLAS 12 sin(f. C) =cos(2 fh) EIC Nonperturbative TMD Perturbative region CLAS 12 and ELIC studies of transition from non-perturbative to perturbative regime will provide complementary info on spin-orbit correlations and test unified theory Jlab, June 18, 2010 23

PT-dependence of beam SSA ssinf. LU(UL) ~FLU(UL)~ 1/Q (Twist-3) 1/PT Nonperturbative TMD 1/Q Perturbative region Check of the higher twist nature of observed SSA critical SSA test transition from non-perturbative to perturbative region Jlab, June 18, 2010 24

A 1 PT-dependence Anselmino Collins CLAS data suggests that width of g 1 is less than the width of f 1 Jlab, June 18, 2010 25

Kotzinian-Mulders Asymmetries HERMES CLAS (5 days) Worm gear TMDs are unique (no analog in GPDs) Jlab, June 18, 2010 B. Musch ar. Xiv: 0907. 2381 B. Pasquini et al, ar. Xiv: 0910. 1677 26

Collins fragmentation: Longitudinally polarized target Kotzinian-Mulders Asymmetry Transversely polarized quarks in the long. polarized nucleon KM s. UL ~ curves, c. QSM from Efremov et al • KM sin 2 f moment, sensitive to spin-orbit correlations: the only leading twist azimuthal moment for longitudinally polarized target • More info will be available from SIDIS (COMPASS, EIC) and DY (RHIC, GSI) Jlab, June 18, 2010 27

cosf moment in ALL-PT-dependence hep-ph/0608048 02=0. 25 Ge. V 2 D 2=0. 2 Ge. V 2 CLAS PRELIMINARY PT-dependence of cosf moment of double spin asymmetry is most sensitive to k. Tdistributions of quarks with spin orientations along and opposite to the proton spin. Jlab, June 18, 2010 28

CLAS configurations ep→e’p. X Inner Calorimeter e pp+ Transversely polarized target Longitudinaly polarized target 1) Polarized NH 3/ND 3 (no IC, ~5 days) 2) Polarized NH 3/ND 3 with IC 60 days 3) Polarized HD-Ice (no IC, 25 days) ØPolarizations: ØBeam: ~80% ØNH 3 proton 80%, ND 3 ~30% ØHD (H-75%, D-25%) Jlab, June 18, 2010 29

CLAS Longitudinally polarized target run (eg 1 -dvcs) Polarized target Helium tube Inner Calo Jlab, June 18, 2010 30

eg 1 -dvcs Run Conditions Beam polarization: >80% Target polarization: >75% Rastering: R=~7. mm eg 1 -dvcs 7. 5 mm Polarized target Inner Calo Helium tube Jlab, June 18, 2010 31

eg 1 -dvcs: Monitoring polarizations HWP→IN HWP→OUT Monitoring the time dependence of the beam polarization using the single spin asymmetry in ep→e’p. X Monitoring the time dependence of the product of target and beam polarizations using the elastic asymmetry Jlab, June 18, 2010 32

SIDIS with JLab at 6 Ge. V Scattering of 5. 7 Ge. V electrons off polarized proton and deuteron targets Ø DIS kinematics, Q 2>1 Ge. V 2, W 2>4 Ge. V 2, y<0. 85 Ø 0. 4>z>0. 7, MX 2>2 Ge. V 2 ep. X 2 Large PT range and full coverage in azimuthal angle f crucial for studies Jlab, June 18, 2010 33

e 1&eg 1 -dvcs: Monitoring and calibration IC operating in high background was stable. IC time and energy resolutions after calibration using 2 photon events. Jlab, June 18, 2010 34

SSA with long. polarized target quark polarization Jlab, June 18, 2010 35

SSA with long. polarized target quark polarization Jlab, June 18, 2010 36

SSA with unpolarized target quark polarization Jlab, June 18, 2010 37

SSA with unpolarized target quark polarization Jlab, June 18, 2010 38

Beam SSA: ALU from CLAS @ JLab Photon Sivers Effect Afanasev & Carlson, Metz & Schlegel 0. 5<z<0. 8 Beam SSA from initial distribution (Boer-Mulders TMD) F. Yuan using h 1┴ from MIT bag model Beam SSA from hadronization (Collins effect) by Schweitzer et al. Jlab, June 18, 2010 39

Jlab, JLab, June. Nov 18, 25 2010 40

CLAS vs CLAS 12? • Existing inconsistencies between HERMES and COMPASS require new independent input. CLAS data will be crucial in developing global analysis of 3 D parton distributions, TMDs GPDs and Wigner distributions. • Study the Q 2 dependence of different SSA at fixed Bjorken-x will require measurements with different beam energies, including 6 Ge. V. • Virtual photon has some angle with beam direction, so measurements with longitudinal and transverse target are important for each other. • Enables early understanding of higher order corrections and higher twist contributions. • CLAS data at 6 Ge. V will be important to analyze future CLAS 12 data (different systematics). • Reaction asymmetries from events ray traced back to vertex gives tomographic decomposition of target polarization vs e running time. -> essential to optimize plans for 12 Ge. V experiments Jlab, JLab, June. Nov 18, 25 2010 41

Transverse force on the polarized quarks Quark polarized in the x-direction with k. T in the y-direction Force on the active quark right after scattering (t=0) Interpreting HT (quark-gluon-quark correlations) as force on the quarks (Burkardt hep-ph: 0810. 3589) Jlab, JLab, June. Nov 18, 25 2010 42

SSA with unpolarized target quark polarization Jlab, JLab, June. Nov 18, 25 2010 43

SSA with unpolarized target quark polarization Jlab, JLab, June. Nov 18, 25 2010 44

SSA with unpolarized target quark polarization Jlab, JLab, June. Nov 18, 25 2010 45

SSA with unpolarized target quark polarization Jlab, JLab, June. Nov 18, 25 2010 46

Beam SSA: ALU from CLAS @ JLab Photon Sivers Effect Afanasev & Carlson, Metz & Schlegel 5. 7 Ge. V p+ 4. 3 Ge. V p+ 5. 7 Ge. V p 0 0. 5<z<0. 8 Beam SSA from initial distribution (Boer-Mulders TMD) F. Yuan using h 1┴ from MIT bag model Beam SSA from hadronization (Collins effect) by Schweitzer et al. Beam SSA for p 0 and p+ are comparable indicating small Collins type contributions Jlab, JLab, June. Nov 18, 25 2010 47

Inbendin/outbending configurations Different polarities increase the acceptance of positive and negative hadrons. Jlab, JLab, June. Nov 18, 25 2010 48