arXiv:nucl-ex/0612020v1 18 Dec 2006
Absolute Polarization Measurements at RHIC in
the Coulomb Nuclear Interference Region
K.O. Eyser∗, I. Alekseev†, A. Bravar∗∗, G. Bunce‡,§, S. Dhawan¶, R. Gill‡,
W. Haeberli?, H. Huang‡, O. Jinnouchi††, Y. Makdisi‡, I. Nakagawa‡‡,§,
A. Nass§§, H. Okada‡‡,¶¶, E. Stephenson∗∗∗, D. Svirida†, T. Wise?, J. Wood‡
and A. Zelenski‡
∗University of California, Riverside, CA 92521, USA
†Institute for Theoretical and Experimental Physics (ITEP), 117259 Moscow, Russia
∗∗University of Geneva, 1205 Geneva, Switzerland
‡Brookhaven National Laboratory, Upton, NY 11973, USA
§RIKEN BNL Research Center, Upton, NY 11973, USA
¶Yale University, New Haven, CT 06520, USA
?University of Wisconsin, Madison, WI 53706, USA
††KEK, Tsukuba, Ibaraki 305-0801, Japan
‡‡RIKEN, Wako, Saitama 351-0198, Japan
§§University of Erlangen, 91058 Erlangen, Germany
¶¶Kyoto University, Kyoto 606-8502, Japan
∗∗∗Indiana University, Bloomington, IN 47408, USA
Abstract. The Relativistic Heavy Ion Collider at Brookhaven National Laboratory provides polar-
ized protonbeams forthe investigationof the nucleonspin structure.Forpolarimetry,carbon-proton
and proton-proton scattering is used in the Coulomb nuclear interference region at small momen-
tum transfer (−t). Fast polarization measurements of each beam are carried out with carbon fiber
targets at several times during an accelerator store. A polarized hydrogengas jet target is needed for
absolute normalization over multiple stores, while the target polarization is constantly monitored
in a Breit-Rabi polarimeter. In 2005, the jet polarimeter has been used with both RHIC beams. We
present results from the jet polarimeter including a detailed analysis of backgroundcontributions to
asymmetries and to the beam polarization.
Keywords: Elastic Scattering, Polarimeter, Polarized Protons
PACS: 24.70.+s, 29.25.Lj, 29.27.Hj
tivistic Heavy Ion Collider (RHIC). Measurements of beam polarizations utilize scatter-
ing processes. Fast measurements are based on proton-carbon scattering off carbon fiber
targets , for which the necessary normalization is provided from elastic proton-proton
scattering with a polarized hydrogen gas jet target. At high energies, the asymmetries
are driven by the electromagnetic spin-flip amplitude at small four-momentum trans-
fer (10−3< |−t| < 10−2). Asymmetries are small (few percent) and contain unknown
contributions from hadronic amplitudes in this so-called Coulomb nuclear interference
region. The existing data, therefore, not only serve as a necessary input to other RHIC
experiments, but on the other hand can also be used to further confine the helicity am-
~ 80 cm
to the interaction point. In routine operation only one RHIC beam hits the target, while the other beam is
Detector setup of the jet polarimeter at RHIC. The six detectors are centered with respect
plitudes φ1−φ5and the hadronic spin-flip contribution in particular .
SETUP AND ANALYSIS
The jet polarimeter is located at one of the collision points in the RHIC accelerator and
reflects the kinematics of elastic proton-proton scattering at small |−t|. It consists of
a polarized hydrogen atomic jet target with a Breit-Rabi polarimeter  and a set of
six separate detectors each of 16 silicon strips, see figure 1. The target polarization is
prepared in an inhomogeneous magnetic field in combination with a radio-frequency
transition unit and constantly monitored (not shown in the figure). The hydrogen molec-
ularcontenthas been determinedand is corrected for.Theeffectivepolarizationamounts
to Ptarget= (92.4±1.8)%.
The detectors are centered with respect to the interaction point of the proton beam
with the target. In this geometry, eight downstream strips on each detector can detect
elastically scattered recoil protons when one RHIC beam hits the target. In routine
operation, one of the two RHIC beams is displaced, so the respective eight non-signal
strips can be used to estimate the background fraction below the elastic signal peak.
The silicon detectors are read out with waveform digitizers (running at 420 MHz
and synchronized with the RHIC clock) that send the pre-processed ADC spectra to
the DAQ-PC. Two different α-sources (Am and Gd) are used for energy calibration
and estimation of the entrance window thickness of the detectors. Additional time of
flight offsets are individually adjusted for each strip using the pronounced proton signal.
Particle identification is based on time of flight and energy. Elastic scattering further
correlates the detector geometry, i.e. the scattering angle, and the small recoil energy of
theproton. A time-of-flightcut ofafew ns removesthemajorpart ofpromptbackground
events below 5 MeV.
In 2005, one of the two RHIC beams was centered on the jet target for several days to
are expected asymmetries from formal descriptions in terms of helicity amplitudes  in the Coulomb
nuclear interference region scaled with the polarization values.
Target and beam asymmetries as functions of recoil energy TR. The solid and dashed lines
accumulate enough statistics for a precise measurement of the beam polarization. Both
beams have been measured repeatedly over the course of a few weeks.
For the determination of the RHIC beam polarization, a set of four different vertical
polarization combinations of target and beam is used. The yields are then combined to
separate asymmetries resulting from the beam and target polarizations. While a certain
polarization direction would result in a specific left-right asymmetry in the detectors,
respective yields can be coupled with their opposite polarization directions in order
to remove detector acceptances and efficiencies and luminosity effects, . The beam
polarization Pbeamis then derived from the ratio of the measured asymmetries εbeamand
εtargetand the known target polarization Ptarget:
The analyzing power AN= ε/P is not flat in the considered t-region and, therefore,
the beam polarization was determined in separate steps of the recoil energy TRbefore
calculating the weighted mean. Also, the kinematic correlation of the detector strip
position and the recoil energy suppresses the background outside of the considered
energy range, where no elastic signal is seen.
The ratio of asymmetries is very robust with regards to background contributions
to the yields, as long as the background has no pronounced polarization dependence.
Background fractions can be as large as 7% in certain energy ranges, which affect the
beam polarization only in second order.
Target and beam asymmetries have been measured as functions of energy of the recoil
proton. Figure 2 shows the typical peak of the asymmetries (and the analyzing power) at
TR≈1.5 MeV in the CNI region from a subset of the data. The asymmetry ratio between
1.0 and 4.0 MeV is used to determine the beam polarization. Lower and higher recoil
energies are discarded due to increased acceptance asymmetries and highly increased
background. Also, the asymmetries in figure 2 are compared to the expected shape of the
analyzing power ANfrom a formal description in terms of helicity amplitudes  that
has been scaled with the target and beam polarizations. The shown curves are neither
fitted to the analyzed data, nor do they include contributions from hadronic spin-flip
In 2005, beam polarizations of nearly 50% have been measured in RHIC with good
statistical accuracy. Systematic errors were estimated from the background yields from
empty target measurements and subdivision with respect to separate proton bunches in
to the polarization directions of bunches and calculating the resulting asymmetries.
Background contributions have been identified from molecular content in the jet tar-
get, from beam gas, and from the displaced beam that is threaded around the target. No
significantbeam or target polarizationdependence has been observedin thebackground,
which is uniformly distributed over the detector halves within the statistical uncertain-
ties. The effect on the determination of the beam polarization has been estimated by
purposefully widening the signal region on the detectors, this way increasing the back-
ground by factors of up to four and consecutivelylowering the beam and target asymme-
tries. The ratio of asymmetries, on the other hand, is largely unaffected by the growing
background and an upper limit of ∆P/P = 1.1% has been assigned to the beam related
background below the signal. The hydrogen molecular content of the target causes an
additional uncertainty that is already taken into account in the target polarization error.
Fastbeam polarizationmeasurementsat RHIC are carried outwithtwo carbon polarime-
ters . Absolute polarization normalization is done alternatingly with a polarized hy-
drogen jet target. Beam polarizationsofnearly 50% havebeen measured in thelatter half
of the 2005 run. Total uncertainties have reached the accuracy that experiments in the
RHIC spin programm have called for. The existing data, also, have been used to further
constrain the knowledge of hadronic spin-flip amplitudes at√s =13.7 and 6.9 GeV .
1. I. Nakagawa, these proceedings (2006).
2. H. Okada et al., Phys. Lett. B638, 450 (2006) and H. Okada, these proceedings (2006).
3. H.G. Gaul and E. Steffens, NIM A316, 297 (1992).
4. G.G. Ohlsen, P.W. Keaton, Nucl. Instrum. Meth. 109, 41 (1973).