Transversity Signals in Two-Hadron Production at COMPASS

1. Christian Schill Universität Freiburg Transversity Signals in Two-Hadron Production at COMPASS On behalf of the COMPASS Collaboration • Transverse spin physics at COMPASS • Azimuthal asymmetries in two-hadron production • Results for identified ππ, πK and KK pairs on the deuteron • Discussion and Outlook DIS 2007, Munich, April 18

2. Transverse Spin Physics q(x) f1(x) Dq(x) g1(x) DTq(x) = q↑↑(x) - q↑↓(x) h1(x) 3 distribution functions are necessary to describe the spin structure of the nucleon at LO: momentum distribution well known - unpolarized DIS helicity distribution known - polarized DIS transversity distribution still unknown DTq(x) decouples from inclusive DIS: helicity flip of quark  SIDIS experiment

3. Transversity: How to measure it in SIDIS? Transversity Tq (x) chiral odd: observable effect only in combination with chiral odd fragmentation function • Suggested quark polarimeters in SIDIS: • Azimuthal distribution of single hadrons • Collins fragmentation function • Azimuthal dependance of the plane • containing a hadron pair • 2-hadron interference fragmentation • function • Measure transverse polarization of Λ • fragmentation function q Λ • COMPASS results Talk by A. Bressan • This talk • COMPASS results

4. The COMPASS Spectrometer at CERN • Beam: 160 GeV µ+ (2 · 108 µ+ / 5s spill) • Polarized target • Particle identification: • RICH MuonWall SM2 E/ H CAL E/ H CAL 50 m SM1 Polarised Target MuonWall SciFi Straws RICH Silicon Drift Chambers Micromegas MWPC GEMs µ+ beam

5. The polarised 6LiD-Target 3He – 4He dilution refrigerator Superconducting Dipole (0.5 T) Two 60 cm long target cells with opposite polarisation Vertex distribution: T=50mK Polarization: 50 % Dilution factor: 0.38 Transverse target polarization: Reversed one a week

6. The polarised 6LiD-Target 3He – 4He dilution refrigerator Superconducting Dipole (0.5 T) Two 60 cm long target cells with opposite polarisation Vertex distribution: T=50mK Polarization: 50 % Dilution factor: 0.38 Transverse target polarization: Reversed one a week

7. Ring Imaging Cherenkov Detector 3 m 6 m Mirror wall 5 m CsI MWPC Radiator: C4F10 Identification of π, K and protons Cherenkov thresholds:π ≈ 3 GeV/c K ≈ 9 GeV/c p ≈ 17 GeV/c 2σπ/K separation at 43 GeV/c

8. Transversity Data Sample transversely polarised deuteron target ~ 20% of the running time 2002 2002 11 days 11 days of data taking (19), of data taking 2 periods trigger (large x, Q2) + PID (ECAL, RICH) 2003 2003 9 days 9 days of data taking (14), of data taking 1 period 2004 2004 14 days 14 days of data taking (24), of data taking 2 periods reconstructed reconstructed DIS events DIS events (10 (10 ) ) 6 6

9. The Coordinate System z-axis = virtual photon direction x-z plane = lepton scattering plane R= angle between lepton scattering plane and two-hadron plane S= azimuthal angle of initial quark versus lepton scattering plane S´= π - S (fragmenting quark) RS= R - S´ = R + S - π R (A. Bacchetta, M. Radici, hep-ph/0407345) (X. Artru, hep-ph/0207309)

10. Azimuthal asymmetry for two-hadron production Target single spin asymmetry ARS(x,z,Mh2): z=z1+z2 and N±(RS): Number of events for target spin up (+) and down (-) f: Dilution factor ≈ 0.38 D: Depolarisation factor D=(1-y)/(1-y+y2/2) PT: Target polarisation ≈ 0.5 presently unknown can be measured in e+e- (BELLE) expected to depend on the hadron pair invariant mass (X. Artru, hep-ph/0207309)

11. Event selection • Hadron selection: • z1,2 > 0.1 (current fragmentation) • xF1,2 > 0.1 • z1+z2 < 0.9 (exclusive rho) • RICH identification of π, K • DIS cuts: • Q2 > 1 GeV2/c2 • 0.1 < y < 0.9 • W > 5 GeV/c2

12. RICH hadron identification all hadron pairs: 5.3 M 0.1 M (1.6%) 3.7 M (70%) 0.25 M (4.5%) 0.3 M (5.8%)

13. Results for π+π- pairs

14. Results for π+ K- pairs

15. Results for K+π- pairs

16. Results for K+ K- pairs

17. sin Dependance Cross section σUT for two-π fragmentation depends on sin: (Interference of s- and p-wave of the 2π-state) (A. Bacchetta and M. Radici, hep-ph/0212300)  : Angle of h1 in the two-hadron CMS to the direction of Ph=Ph1+Ph2 <sin> = 0.95 <sin> = 0.90  small contribution in the kinematic region of COMPASS

18. Discussion • First results for two identified hadron pair asymmetries • Asymmetries are small for all hadron pairs ππ, πK, and KK • Systematic effects are considerably smaller • than the statistical error Comparison with theoretical models 

19. Comparison with Theory Model calculations for COMPASS kinematics (M. Radici, QCDN 06, hep-ph/0608037): Model for transversity: • Soffer, Stratmann, Vogelsang, • P.R. D65 (02) 114024 • Korotkov, Nowak, Oganessian, • E.P.J. C18 (01) 639 • Schweitzer et al., • P.R. D64 (01) 034013 • Wakamatsu • P.L. B509 (01) 59 ΔTq(x) xBj Prediction for COMPASS: • small asymmetries on the deuteron combination of isospin- symmetric target and properties of dihadron- fragmentation function Hqh(z) ARS Mh

20. Outlook • Complementary measurements with proton target planned this year • Data (of comparable statistics) will be collected on a transversely • polarized proton target (NH3) and will allow flavour separation Predictions for two-hadron asymmetries on the proton at COMPASS: • Soffer, Stratmann, Vogelsang, • P.R. D65 (02) 114024 • Korotkov, Nowak, Oganessian, • E.P.J. C18 (01) 639 • Wakamatsu • P.L. B509 (01) 59 - ARS (M. Radici, hep-ex/0608037)

21. Thank you!

22. Results for two unidentified hadrons all combinations of +ve (h1) and –ve (h2) hadrons 2002-2004 COMPASS data 2002-2004 COMPASS data

23. Partial wave expansion of Hqh(z) A. Baccetta, hep-ph/0708037 cos-Distributions of our data:

24. Frame independant definition of φR • A. Baccetta, • hep-ph/0608037 where P1T and P2T are the transverse components of the hadron momenta with: we define:

25. Double Ratio Method double ratio: Nup/down: upstream / downstream target cell N: target polarization vector pointing up / down fit function: final asymmetry:

26. COMPASS kinematic region

27. Transverse Λ Polarization

28. Transverse Λ Polarization