STAR Forward Spin Physics at STAR RHIC BNL

STAR Forward Spin Physics at STAR RHIC, BNL Forward jet-like event spin-dependent properties in polarized p+p collisions at = 200 Ge. V Nikola Poljak University of Zagreb For the STAR Collaboration 30. 09. 2010.

2 Single spin asymmetry • Definition: • dσ↑(↓) – differential cross section of p 0 when incoming proton has spin up(down) Left positive AN: p 0, x. F<0 p 0, x. F>0 more p 0 going p left to polarized beam p Right

Published measurements - AN PRL 97, 152302 (2006) nucl-ex/0602011 RUN 6 : PRL 101 (222001) • Large transverse single-spin asymmetries at large x. F • x. F dependence matches Sivers effect expectations qualitatively • p. T dependence at fixed x. F not consistent with 1/p. T expectation of p. QCD-based calculations • Obtained with the FPD / At. FPD++ this energy the cross-section is consistent with NLO p. QCD (run 2 + run 3) and included in global fits on fragmentation functions Phys. Rev. D 75: 114010, 2007 3

Published measurements - AN RUN 8 : AIP conf. proceedings 1149, ar. Xi. V: 0901. 2828 (N. Poljak for the STAR collaboration) • Confirmed x. F dependence of the asymmetry with a new detector (Forward Meson Spectrometer) • Exctracted postive x. F asymmetry dependence on the angle with respect to the spin axis 4

5 Separating Sivers and Collins effects Sivers mechanism: asymmetry in the forward jet or γ production Phys Rev D 41 (1990) 83; 43 (1991) 261 SP k. T, q p Collins mechanism: asymmetry in the forward jet fragmentation Nucl Phys B 396 (1993) 161 SP p p Sensitive to proton spin – parton transverse motion correlations p Sensitive to transversity Sq To discriminate between the two effects we need to go beyond π0 detection to jet-like events k. T, π

6 “jet-like” event measurements Obtained with the FMS; event selection done with: • • >15 cells with energy > 0. 4 Ge. V in the event (no single pions in the event sample) cone radius = 0. 5 (eta-phi space) “Jet-like” p. T > 1 Ge. V/c ; x. F > 0. 2 2 perimeter fiducial volume cut (small/large cells) ar. Xi. V: 0901. 2828 N. Poljak for the STAR collaboration Reasonable agreement of data and simulations

Idea: use FPD++ (trigger/geometry) 7 RUN-8; FMS RUN-6; FPD++ Hightower trigger Summed module energy trigger Monolithic detector Modular detector with explicit azimuthal symmetry

8 Analysis overview - calibration • Small cells calibrated for the RUN-6 published results • Large cells proved more difficult due to the trigger which is sensitive only to the small cell deposited energy, different from the initial plan • 2 step calibration was employed; first is the energy deposition slope matching, used to get the calibration factors within 10 to 15 %, second is the di-photon invariant mass calibration, with the following event selection: WN 114 simulations E(Ge. V) OVERLAY OF SIMULATIONS AND DATA BEFORE AFTER CALIBRATION

9 Goals • Extract a “jet-like” event sample and compare reconstructed PYTHIA+GSTAR events against reconstructed data events • Verify consistency with SPIN 08 results • Reproduce transverse single spin asymmetries from FPD++ from data for p↑+ p -> p 0 + X (verify consistency with RUN 6 published results) • Extract the Collins angle and calculate the single spin asymmetries integrated over it and differential in it

“Jet-like” event selection • • At least 4 towers with E ≥ 0. 4 Ge. V weighted sum of towers ≥ 10 (w(small)=1, w(large)=1. 52) “jet-like” p. T ≥ 1. 5 Ge. V/c “jet-like” E ≥ 20 Ge. V Max. cone radius of 0. 5 (eta-phi space) 2 perimeter fiducial volume cut The event selection different than for FMS due to geometry (weighted sum of towers cut) 10

First “jet-like” events results • Simulations set up to mimic the data trigger • The agreement between data and simulations is very convincing and repeats itself over a variety of results and throughout the x. F range 11

First results compared with SPIN 08 12 • Trigger differences can give rise to differences in results from the FMS and from the FPD++ • The weighted sum of cells cut is effectively a larger multiplicity cut in the FMS than in the FPD++ • Differences dealt with by mimicking the FMS trigger in the FPD++ and by use of the weighted multiplicity cut to account for geometry differences The results obtained with the FPD++ and the FMS show agreement.

Event selection • RUN 6 published results used only data from the inner cells • looking at the π0 + X “jet-like” final event state, the requirement that the large cell readout is alive was imposed • the analyzing power for this subset was calculated to compare it with previously published results; the average asymmetry for x. F>0. 3 is equal to 0. 0313± 0. 006. RUN 6 published result Result with the “large cell live readout” condition The new event selection doesn’t introduce any asymmetries. 13

Association analysis 14 The neutral pion is well reconstructed and carries most of the energy of the event. What about the “jet-like” event ? "jet-like" event clusters reconstructed from a PYTHIA/GSTAR simulation are found to be associated with a hard-scattered or a radiated parton. The “jetevent” axis is given well with the direction of the parton.

Collins angle – definition and results • The angle which the plane defined by the leading pion and the fragmenting quark closes with the reaction plane • Defined mirror symmetrically around the axis of symmetry of the detector system (here parallel to Sproton), with 0 towards the beam and ±π away from the beam 15 =γ • Collins angle well reconstructed as confirmed by association analysis • The pion carries most of the transverse momentum of an event, so it is expected that the distribution will be peaked near small angles The Collins angle distributions show agreement in data/simulations and are behaving as predicted.

Results - asymmetry 16 • The asymmetry for the events was calculated in bins in the cosine of the Collins angle • The negative x. F asymmetry is consistent with zero • The x. F>0 asymmetry is greater than zero in all bins (av. 0. 031± 0. 014), but doesn’t show a dependence on cos(γ) The “jet-like” events x. F>0 asymmetry is positive, but doesn’t show any Collins effect contributions.

Conclusions & Outlook • Reconstruction and calibration procedures used in FPD and FMS succesfully used in the FPD++ • Data shows very good agreement with the simulated sample of events for the “jet-like” sample • Verified consistency with published RUN 6 results and the RUN 8 results, obtained with the FMS • The Collins angle was found and is behaving as predicted • The calculated positive x. F asymmetry is greater than zero (av. of 0. 031) and doesn’t show any Collins contributions 17

Thank you