Preliminary Result of AN Measurement in pp Elastic Scattering at RHIC, at s 200 GeV Wlodek Guryn for

1
Preliminary Result of AN Measurement in p?p?
Elastic Scattering at RHIC, at ? s 200 GeV
Wlodek Gurynfor pp2pp collaboration
Brookhaven National Laboratory, Upton, NY, USA

• OUTLINE of the TALK
• Description of the experiment
• Description of analysis
• Results and interpretation
• Where do we go from here?

2
Total and Differential Cross Sections, and
Polarization Effects in pp Elastic Scattering at
RHIC

• S. Bültmann, I. H. Chiang, R.E. Chrien, A. Drees,
  R. Gill, W. Guryn, J. Landgraf, T.A. Ljubicic,
  D. Lynn, C. Pearson, P. Pile, A. Rusek, M.
  Sakitt, S. Tepikian, K. Yip
• Brookhaven National Laboratory, USA
• J. Chwastowski, B. Pawlik
• Institute of Nuclear Physics, Cracow, Poland
• M. Haguenauer
• Ecole Polytechnique/IN2P3-CNRS, Palaiseau, France
• A. A. Bogdanov, S.B. Nurushev, M.F Runtzo, M. N.
  Strikhanov
• Moscow Engineering Physics Institute (MEPHI),
  Moscow, Russia
• I. G. Alekseev, V. P. Kanavets, L. I. Koroleva,
  B. V. Morozov, D. N. Svirida
• ITEP, Moscow, Russia
• S. Khodinov, M. Rijssenbeek, L. Whitehead
• SUNY Stony Brook, USA
• K. De, N. Ozturk

3
Spin Dependence in Elastic Scattering
Five helicity amplitudes describe proton-proton
elastic scattering
Some of the measured quantities are
4
Source of single spin analyzing power AN
Single spin asymmetry AN arises in the CNI region
is due to the interference of hadronic non-flip
amplitude with electromagnetic spin-flip
amplitude (originally called Schwinger
asymmetry) Any difference from the above is an
indication if other contributions, hadronic spin
flip caused by resonance (Reggeon) or vacuum
exchange (Pomeron) contributions.
5
The world data on AN (pp)
6
RHIC/AGS Measurements
There will also be results from polarized jet at
RHIC
7
Experimental Determination of AN
Use Square-Root-Formula to calculate spin ( ??,
?? ) and false asymmetries (??, ?? .) This
formula cancels luminosity dependence and
apparatus asymmetries.
Since AN is a relative measurement the
efficiencies e(t, f) cancel
8
Principle of the Measurement

• Elastically scattered protons have very small
  scattering angle ?, hence beam transport magnets
  determine trajectory scattered protons
• The optimal position for the detectors is where
  scattered protons are well separated from beam
  protons
• Need Roman Pot to measure scattered protons close
  to the beam without breaking accelerator vacuum

Beam transport equations relate measured position
at the detector to scattering angle.
x0,y0 Position at Interaction Point Tx Ty
Scattering Angle at IP xD, yD Position at
Detector TxD, TyD Angle at Detector
9
The Setup
10
Elastic Event Identification
11
Trigger Active
area
Only inner pots used for trigger and analysis,
biggest acceptance Analyze the data for the
closest position (¾ of all data)
12
Angle (hit) Correlations Before the CutsEvents
with only eight hits
Note the background appears enhanced because of
the saturation of the main band
13
Hit selection

• Pedestal value, pedestal width (s) and dead
  channels (only six) were determined
• Valid hit, single strip, has dE/dx gt 5s above the
  pedestal
• Cluster size is ? 5 consecutive strips above
  pedestal cut
• Valid hit in the Si plane for event
  reconstruction
• is a cluster whose dE/dx gt 20 ADC counts above
  pedestal and
• is within fiducial area of the detector (slide)
• has for a y-plane y gt 0.2mm from the edge of the
  detector.
• Coordinate for x and y formed from adjacent hits
  in Sin for each Roman Pot

14
Elastic Events Display Dave Morse
15
Elastic Event Selection III

• After finding matching hits in x and y
• Choose events with one track in x and one track
  in y and ? 6hits.
• Veto on the Sc signal in the opposite arm, TDC
  cut.
• Choose collinear tracks within 3s in angles.
• Plot dN/dt and calculate asymmetries.

16
CollinearityDQx before and after z-correction,
and DQy
17
Event Yields after Cuts
18
dN/dt
19
Determination of AN

• Use Square-Root-Formula to calculate raw
  asymmetries.
• It cancels cancel luminosity dependence and
  effects of apparatus asymmetries.
• It uses ??, ?? bunch combinations.

Since AN is a relative measurement the
efficiencies e(t, f) cancel
20
Preliminary Results Full bin 0.011 lt -t lt 0.029
(GeV/c)2
Fit AN cos(j) dependence to obtain AN
Statistical errors
Statistical errors
21
Results PYellow PBlue 0.67 (central value)
and CNI curve (stot , r from world data, B from
pp2pp result)
Preliminary
Statistical errors
Statistical errors
Note a) PYellow PBlue 0.67 can vary by 15
( a working number) b) Systematic
effects due to the beam transport uncertainty are
small compared to the beam polarization
uncertainty.
22
Results comparison with CNIdifferent
polarizations and CNI parameters
Statistical errors
Statistical errors
Curves show dependence of CNI on stot, r and B
23
Prediction using E704 and E950 dataLarry Trueman
LT The predictions are very uncertain because
E704 makes a very large contribution with large
errors to the Regge parameters.
The upper curve has Regge contribution and the
Pomeron coupling set to zero. Here t 0.02
0.01i. Not much change  in the curve. Note that
the f/a2 is the dominant Regge contribution and
this leads to a very large energy dependence,
especially to Re(t), in going from 704 energy to
RHIC energy. At s 400 GeV2, t 0.1281
0.0285 i
24
Theoretical Interpretation Boris Kopeliovich
Assume the non-flip Re f/ Im f p/2e 0.15,
where e 0.1, is the power of s in
energy dependence of stot. 1) The dashed curve
is calculation with central values for r5 as was
published by E950. Need for energy dependence of
r5 2) The dash-dotted uses Im r5 ?0.12, from pp
and pp data at low energies. (BK SPIN2002 and
hep-ph/0211061.) The phase is assumed to be the
same as for the non-flip amplitude, i.e. Re r5/Im
r50.15. 4) Solid curve assumes r5 rising as se,
Regge Model of stot i.e. the spin-flip amplitude
rises as s2e and results in a) Im r5 ? 0.24 (
twice that at low energy.) b) Re r5p/22eIm
r5 ? 0.075.
25
Summary

• We have measured the single spin analyzing power
  AN in polarized pp elastic scattering at ?s 200
  GeV, highest to date, in t-range 0.011,0.29
  (GeV/c)2.
• The AN is ? 4-5s from zero (statistical error.)
• The AN is ? 2s away from a CNI curve, which does
  not have hadronic spin flip amplitude.
• We discussed interpretation of the result by
  Larry Trueman and Boris Kopeliovich. (there is
  clear ?s dependence in r5, a hadronic spin flip
  is present.)
• In order to understand better underlying dynamics
  one needs to map ?s and t-dependence of AN and
  also measure other spin related variables ANN.
• RHIC is a great and unique place to do this
  physics and we have a plan how to do it and do it
  better!

26
Future Possibility Big Improvement