P1252109249ahLJY

1
Forward Transverse Spin
at STAR
Andrew Gordon RHIC AGS Annual Users
Meeting Workshop 5 May 27, 2008
2
Outline of report on transverse data at STAR
Part I Large AN can be gainfully employed to
track beam polarization
Part II Recent measurements at STAR
Part III Status of Run 8 data and measurements
3
Part I Sizeable asymmetries can be used to
measure beam polarization
STAR BBC
Inner tiles cover 3.5lthlt5
Can use large asymmetries to measure (relative)
bunch-by-bunch polarization
4
Inner tiles of BBC accumulated every clock cycle

Scaler Boards

Discriminated phototube outputs
24-bit word is histogrammed every clock cycle

East-West Coincidence

Bunch Crossing (7-bits)
5
Bunch-by-bunch polarization from colliding beams
Width1.22?0.15
Yellow (east BBC)
Expected spin up
Expected spin down
sig
Width1.05?0.12
Blue (west BBC)
sig
3.5 s (statistical) measurement of polarization
per bunch per hour
(Pbx-ltPbxgt)
Sigbx
sbx
Statistical uncertainties only
6
Part II Recent asymmetry measurements at STAR
Polarization here valence quark spin effects
p0
p
p
xgp
xqp
Polarization here low x-gluons and other partons
High rapidity ps (hp4) from asymmetric partonic
collisions
pp p0, hp3.8, vs200GeV
Mostly high-x valence quark on low-x gluon
(0.3 lt xqlt 0.7, 0.001lt xg lt 0.1)
Fragmentation z nearly constant and high 0.7
0.8
NLO pQCD Jaeger, Stratmann, Vogelsang, Kretzer
7
Two examples of sources of transverse spin effects
Sivers mechanism correlation between proton spin
and quark KT (implies orbital angular momentum).
Collins mechanism correlation between quark
polarization (transversity) spin and asymmetry in
jet fragmentation
Initial state transversely polarized quark
Final state transversely polarized quark
Asymmetry in fragmentation
Depends on transversity distribution. Potential
asymmetry in fragmentation provides ability to
see transversity, (and vice versa).
One important goal To separate these
experimentally.
8
Cross sections at ?s200 GeV are consistent with
pQCD
but well described at 200 GeV
Cross sections significantly under-predicted at
lower vs
STAR
vs52.8GeV
vs200 GeV
vs23.3GeV

2 NLO collinear calculations with different
scale pT and pT/2
ISR data
(STAR), PRL 92 (2004) 171801

• Bourrely and Soffer hep-ph/0311110 (Eur. Phys.
  J C36 (2004) 371) data references therein

Suggests that asymmetry data can be described
within the context of pQCD.
9
Large asymmetries persist at high ?s
Examples
?s 20 GeV
?s 62 GeV
?s 200 GeV
p? p ? p X, ?s 20 GeV
p? p ? p0 X, ?s 200 GeV

p? p ? p X, ?s 62 GeV
pT0.5-2.0 GeV/c
Arsene et al. (BRAHMS), submitted to Phys. Rev.
Lett. arXivnucl-ex/0801.1078
RHIC, Brahms, 2007
(STAR) Phys. Rev. Lett. 92 (2004) 171801
??0 E704, Phys.Lett. B261 (1991) 201.
??/- E704, Phys.Lett. B264 (1991) 462.
RHIC, STAR, 2004
Fermilab, Fixed target, E704, 1991
Perturbative cross section
Non-Perturbative cross section
10
Runs 3, 5, and 6 forward p0 asymmetry
PT vs XF coverage of data
Steeply falling with PT and XF.
FPD set at different distances from beam for the
different lthgt ranges.
arXiv0801.2990v1, submitted to PRL
11
Run 3, 5, and 6 asymmetry data (cont) Theory can
predict XF dependence based on Sivers function
fits to p/p- asymmetries
Data B.I. Abelev et al. (STAR), submitted to
PRL arXivhep-ex/0801.2990v1, submitted to
PRL Theory (red) M. Boglione, U. DAlesio, F.
Murgia arXivhep-ph/0712.4240 Theory (blue)
C. Kouvaris, J. Qiu, W. Vogelsang, F. Yuan, PRD
74 (2006) 114013
12
but rising PT dependence is not predicted by the
same fits
Admixture of Collins and Sivers?
XFgt0.4
Current data can extend PT reach of measurements
B.I. Abelev et al. (STAR) arXivhep-ex/0801.2990v
1, ), submitted to PRL
Data broken out in XF bins
13
STAR Results vs. Di-Jet Pseudorapidity Sum Run-6
Result
14
Part III Run 8 data
Because of lower polarization than expected,
figure of Merit (P2L) fell short by roughly a
factor of two.
Run 8 Integrated FMS transverse figure-of-merit.
Fast detectors only.
However, the FMS provides roughly 20 times the
coverage of previous runs in the forward region.
15
New forward detector for Run 8 FMS
FMS provides nearly 20x the coverage of previous
forward detectors
Run 5 FPD
16
FMS Physical Location
Far West side of Hall, at the opening to RHIC
tunnel.
7.5 meters from interaction point.
FMS
17
Run8 expectations for forward p0 asymmetry
Projections for 9 pb-1 recorded in FMS with 70
polarization. With actual figure of merit, error
bars will increase by roughly factor of 2.
Run 8 data should be able to extend kinematic
reach for inclusive p0 and heavy mesons.
18
Run 8 data current status
1) p0 reconstructions are readily available
19
Run 6 resolution of d(Mp)/Mp10 should be
possible.
Calorimeter stable at level of 1.
20
More run 8 opportunities with FMS
Near term
Increase reach of inclusive p0 asymmetry (above)
Spin-dependent jet-like events and spin-dependent
p0p0 correlations
Far term
Measurements of spin-dependent inclusive direct g
and gjet
Benchmarking of abilities to measure
spin-dependent Drell-Yan
21
Spin-dependent p0p0 correlations
Look for 4-photon events in FMS consistent with
two p0 decays
Two azimuthal angles
Angle of p0p0 system relative to spin direction
Angle of leading p0 relative to p0p0 system
Can provide information about Collins/Sivers
separation
22
Strategy to find p0p0 events in FMS
Look for all events with gt4 photons candidates in
FMS.
For all photon pairs, calculate a vertex location
such that the two photons combine to the pion
mass.
Candidate p0p0 events are events where two photon
pairs have consistent vertices.
Pythiafast FMS simulation
Caveat 1 FMS at ideal resolution in simulation
DZgt175 cm
All found pairs
Combinatoric background
Caveat 2 FMS at ideal hadronic response in
simulation (ie, hadrons ignored)
M(p0p0) (GeV)
KS decays to p0p0
Caveat 3 Photon reconstruction efficiency at
ideal level in simulation (100). No cluster
merging, etc.
DZZFIT-ZBBC (cm)
KS p0p0 show up at high displaced vertex
ZFITerror-weighted average of two pion condidates
23
Future measurement of spin-dependent direct g
Transversely polarized proton
q
photon
Quark jet
g
Unpolarized proton
Lack of photon fragmentation provides clear
access to Sivers function
Theory (Kouvaris, Qiu, Vogelsang, Yuan, PRD 74
(2006) 114013) predicts a sign flip in left/right
asymmetry for inclusive direct photon.
(Background from fragmenting quarks predicted to
have opposite sign.)
24
Large FMS acceptance allows rejection of p0
background
Search for photons in yellow band and reject
events with nearby photon to reduce background
from p0 and h decays. Large acceptance of FMS
makes this possible.
For Eggt25, 95 of second photon from p0 decays
occur within radius of 4 large cells (23 cm).
Example p0 rejection region.
Large acceptance also allows calorimeter
isolation criterion to help reduce background
from photons that result from fragmentation.
25
Pythia fast-FMS simulation
Restricting the measurement of the forward photon
to Egt35 GeV at lthgt3.2 produces a
signalbackground ratio of 2.1.
26
Benchmarking Drell-Yan
As a test of our ability to do future
spin-dependent Drell-Yan measurements, we can
look for lepton pairs in the Run 8 data.
Mass distribution of isolated clusters using
subset of Run 8 data and initial FMS calibrations.
Widths and signal to background ratios will be
strongly dependent on ongoing calibration efforts.
p0 mass region
Can be used to calibrate simulation of ability to
identify ee- pairs with FMS, an important first
step towards a Drell-Yan measurement.
27
Conclusions
Forward pion asymmetries from previous runs have
been completed and submitted for publication.
Run 8 FMS calibrations underway. All of run 8
should be available soon thereafter.
20x increase in detector size with FMS should
allow significant increase in physics reach.