HAPPEX Parity Violation Experiments at
Jefferson Lab
D.S. Armstrong † and the HAPPEX and Hall A Collaborations †
†
Dept. of Physics, College of William & Mary, Williamsburg VA 23187, USA
Thomas Jefferson National Accelerator Facility. Newport News, VA 23606, USA
Abstract. The HAPPEX program of measurements of parity-violation in elastic electron scattering,
in Hall A of Jefferson Lab, is presented. The results of the recently completed measurement on
the proton at Q 2 = 0.48 GeV2 are briefly reviewed. The plans are presented for the upcoming
HAPPEX II measurement on the proton at Q 2 = 0.1 GeV2 , as well as the companion measurement
with a 4 He target at the same momentum transfer. These experiments are sensitive to strange
quark contributions to the vector structure of the nucleon. The two new experiments will provide
a precision measurement of the strangeness radius parameter, and the combination of the two
experiments will also determine the strange contribution to the proton’s magnetic moment.
INTRODUCTION
Our knowledge of the electromagnetic form factors of the nucleon has become increasingly refined in recent years [1]. However, a basic question remains open: how does
the quark-antiquark sea contribute to the distributions of charge and magnetization represented by these form factors? As the nucleon possesses no net strangeness, strange
quarks contributions represent a promising window onto the effects arising from the sea.
A significant role for strange quarks in certain nucleon properties is suggested by results on a number of observables. Deep inelastic neutrino scattering experiments [2, 3]
indicate that the quark structure functions sx and sx are significant at low x. Analyses of spin-dependent deep inelastic scattering [4, 5, 6, 7] suggest that strange quarks
may contribute a sizable fraction to the nucleon’s spin. The pion-nucleon sigma term,
accessible through low-energy π scattering, gives information on scalar strange matrix
elements. The comparison of the π N scattering results with hyperon mass systematics
indicates sizable scalar strange matrix elements for the nucleon [8, 9]. Low-energy antiproton annihilation measurements [10] find large enhancements of φ meson production
over predictions based on the OZI rule, which have been interpreted as evidence of large
polarized ss components in the nucleon [11, 12].
Parity violation in electron scattering provides another avenue by which to investigate
strange quarks in the nucleon. In contrast to the techniques discussed above, the primary
sensitivity here is to the vector matrix elements, N sγ µ sN , about which we have
little information.
In electron scattering, the interference of the γ exchange and Z 0 exchange diagrams
leads to parity violation in the scattering rate, which can be used as a probe of hadron
structure. In essence, while the photon’s coupling to the quarks is proportional to their
CP675, Spin 2002: 15th Int'l. Spin Physics Symposium and Workshop on Polarized Electron
Sources and Polarimeters, edited by Y. I. Makdisi, A. U. Luccio, and W. W. MacKay
© 2003 American Institute of Physics 0-7354-0136-5/03/$20.00
267
electric charges, the Z 0 coupling is proportional to their weak charges. Thus a measurement of parity violation allows a quark flavor decomposition of the vector structure of
the nucleon.
The vector matrix elements of the nucleon can be represented in terms of the Sachs
form factors, GE Q2 and GM Q2 , (Q2 is the momentum transfer) which parameterize
the electric and magnetic response, respectively. Similarly, one defines form factors for
each flavor i, GiE and GiM ; of particular interest here are the strange quark versions, GsE
and GsM .
σ σ
The measured asymmetry is A σR σL where σR and σL are the cross sections
R
L
for scattering with right- and left-handed helicity electrons respectively. The dominant
parity-conserving term from the square of the γ -exchange diagram drops out from the
difference, and one is left with a term proportional to the γ -Z interference.
For elastic scattering from the proton, the asymmetry is given by
ε G0 τ G0M
A A0 τ 2 4 sin2 θW γEp
ε GE τ GγMp
AA
3 are the SU(3) flavor singlet form facG M
tors, ε and τ are kinematic factors, θ is the lab scattering angle, and A 0 AA is
2πα
where G0E M τ GuE M GdE M GsE M
F
2
p
the axial vector form factor, which is small at the forward scattering angles used here.
The asymmetry involves the term ε G 0E τ G0M ε GγEp τ GγMp which allows one to
extract G0E M if the electromagnetic form factors are known. Assuming charge symmetry
one can extract the strange quark form factors from the G0 ’s. Combining with the
nucleon electromagnetic form factors allows the extraction of some combination of G sE
and GsM ; the exact combination depends on the kinematics of the measurement.
At low Q2 , it is convenient to consider the leading moments of the form factors, given
dGs τ by the strangeness magnetic moment µ s GsM 0 and radius parameter ρs dEτ τ 0 .
Many theoretical predictions for the strange quark form factors are available, using a
variety of models of non-perturbative QCD, and they range for µs from -0.6 to +0.4, and
from -3.0 to 3.2. for ρs. Recent reviews of the field are available elsewhere [13, 14, 15].
HAPPEX - I
The Hall A Proton Parity Experiment (HAPPEX) [16] measured the asymmetry in
scattering of longitudinally-polarized 3.3 GeV electrons from a liquid hydrogen target at
Q2 048 GeV2 . Two identical high-resolution magnetic spectrometers, both located at
123Æ to the beam, were used to detect the scattered electrons. Elastic scattering events
were detected in the focal plane of each spectrometer using total absorption counters
made from a lead-lucite sandwich.
The data-taking took place in two runs. In the 1998 run, the electron beam current was
typically 100 µ A, and the polarization was about 39%. The polarized electrons were
produced using a circularly-polarized laser shining on a bulk GaAs photocathode. In the
268
1999 run, the photocathode was replaced with a “strained crystal" GaAs photocathode,
which afforded higher polarization ( 73%) with reduced intensity (35 µ A) 1 . The beam
polarization was monitored periodically using a Mott polarimeter at the accelerator
injector, a Møller polarimeter, and (for the 1999 run) a Compton polarimeter, which
enabled continuous non-invasive polarization measurements [17].
The helicity of the beam was reversed at 30 Hz (to average over 60 Hz noise), in a
pseudorandom manner. The detector signals were integrated in customized 16-bit ADCs
during each 33 ms helicity window; by integrating the signal, data acquisition dead-time
was eliminated.
Helicity-correlated changes in the beam energy were of the order of 108 , and typical
helicity-correlated beam position differences, integrated over the runs, were a few nm.
The total effect of these helicity-correlated variations in beam properties on the experimental asymmetry was small ( 1 ppm).
The focal-plane detector signals are normalized by the beam current, as measured
by two independent rf cavity monitors. Non-linearity in the detector response or the
electronics can produce false asymmetries if the intensity exhibits significant helicitycorrelated fluctuations. These were kept at the 1 ppm level using a slow-feedback system
at the electron source, based on the voltage applied to the helicity-reversal Pockels cell.
Backgrounds from inelastic scattering events and scattering from the target windows
were measured with runs at reduced beam intensity, taken with the full instrumentation
package of the spectrometers. The background processes contributed 0.2% and 1.5%,
respectively, to the measured signal.
The two data-taking runs yielded consistent results, and the combined asymmetry
is A 150 11 106 [16], leading to the combination of strange quark form
factors GsE 0392GsM 0025 0020 0014, at Q2 048 GeV2 . The first error is
the combined statistical and experimental systematic errors, and the second arises from
uncertainties in the electromagnetic form factors.
The result is consistent with the absence of strange quark effects, however, note that
in several models GsE and GsM are predicted to have opposite sign. Thus it may be that
the results reflect an ‘accidental’ cancellation of the form factors. Also, some models
with substantial strange quark contributions predict form factors that cross zero at this
Q2 , but that are large at other momentum transfers. Several model predictions are ruled
out, however the data are consistent with other models which include substantial strange
quark contributions. Measurements at different kinematic points are therefore of interest.
HAPPEX-II AND 4 HE
In 2003, the HAPPEX experiment will be extended down to Q 2 01 GeV2 [18]. To
achieve this lower Q2 , a pair of septum magnets will be installed, which will allow the
spectrometers to view a smaller scattering angle (6Æ ). The expected asymmetry at this
kinematics is small (A 16 ppm), however the scattering rate will be 65 MHz per
1
More recently currents in excess of 100 µ A have been achieved with the high-polarization source.
269
spectrometer, so a statistical precision of 4.6% will be obtained in the approved 700 hr
run. New radiation-hard focal plane detectors, consisting of 5-layer sandwiches of brass
and fused silica, have been constructed and tested for this experiment.
The experiment demands increased attention to helicity-correlated fluctuations in
beam parameters, since the statistical precision corresponds to an ambitious 75 ppb.
These fluctuations will be reduced by using simultaneous feedback at the electron source
on the beam position and beam current asymmetries.
Another challenging requirement is the liquid hydrogen target. Beam-induced density
fluctuations on the 30-Hz helicity-flip time scale, such as due to local boiling, will spoil
the statistical precision. We require such fluctuations to be below 100 ppm in order not
to degrade the statistics. Two different designs of thin-walled high-power target cells are
being developed.
Systematic errors will include the polarization and Q 2 measurements (2% and 0.6%),
background contributions (0.75%), the electromagnetic form factors (1.4%) and radiative corrections (1%). The measurement will yield ρ s µ p µs to a precision of 03
which is compared to model predictions in Fig. 1 (see [13] for theory references).
An approved companion measurement to HAPPEX-II will be a measurement of the
elastic scattering from 4 He, at the same Q2 [19]. Essentially the same instrumentation
will be adopted. Scattering from a spin-zero target has the advantage that there are no
magnetic or axial-vector contributions - only charge scattering contributes. Thus the
asymmetry can be interpreted directly in terms of the strangeness radius ρ s .
The detected elastic scattering rate will be 12 MHz in each spectrometer, and the
predicted asymmetry is A 84 ppm. The 700 hr run will provide a statistical error of
2.2% on A, yielding a measurement of ρ s to a precision (including systematics) of 0.5.
This will considerably reduce the allowed model space for the strange form factors (see
Figure 1), and, in combination with HAPPEX-II, will allow us to disentangle the electric
from the magnetic contributions. The combined experiments will provide a measurement
of µs to 022.
CONCLUSIONS
The strange quark contributions to the vector structure of the nucleon remain elusive,
and are, to date, consistent with zero, despite a variety of models which suggest substantial effects. This could, however, be due to accidental cancellations between the magnetic and electric form factors, or it may be that they happen to have Q 2 dependences
which make the effects small at the measured kinematics. The next-generation HAPPEX
measurements on hydrogen and helium will provide a precise separation of the strange
electric and magnetic form factors at Q2 01. If, after these measurements, the strange
contributions remain consistent with zero, this will beg the interesting question: what
mechanism suppresses the effect of the qq sea on the electromagnetic structure of the
nucleon?
270
FIGURE 1. Various model predictions (points) for the leading moments of the strange form factors, ρ s
vs. µs , compared with the expected precision of the HAPPEX-II and 4 He parity experiments (bands). The
location of the bands shown is arbitrary, only the slopes and widths are relevant.
ACKNOWLEDGMENTS
This work was supported by the National Science Foundation (grants PHY-9602901,
PHY-0099557) and by DOE contract DE-AC05-84ER40150 under which the Southeastern Universities Research Association (SURA) operates The Thomas Jefferson National
Accelerator Facility.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
See contributions to these proceedings from J. Kelly, M. Seimetz, B. Plaster, N. Savvinov, B. Reitz,
and C.F. Perdrisat.
T. Adams, et al., Phys. Rev. D 64, 112006 (2001).
A. O. Bazarko, et al., Z. Phys. C 65, 189 (1995).
K. Abe, et al., Phys. Rev. Lett. 79, 26 (1997).
B. Adeva, et al., Phys. Rev. D 58, 112002 (1998).
H. Lipkin and M. Karliner, Phys. Lett. B 461, 280 (1999).
A. Miller, these proceedings.
M. Pavan, π N Newsletter 19, 118 (1999).
J. Gasser, H. Leutwyler, and M.E. Saino, Phys. Lett. B 253, 260 (1991).
C. Amsler, et al., Phys. Lett. B 346, 363 (1995).
J. Ellis, et al., Phys. Lett. B 353, 319 (1995).
M.G. Sapozhnikov, Nucl. Phys. A692, 63 (2001).
K.S. Kumar and P.A. Souder, Prog. Part. Nucl. Phys. 45, S333 (2000).
D.H. Beck and B.R. Holstein, Int. J. Mod. Phys. E 10, 1 (2001).
D.H. Beck and R.D. McKeown, Ann. Rev. Nucl. Part. Sci. 51, 189 (2001).
K.A. Aniol, et al., Phys. Rev. Lett. 82, 1096 (1999); K.A. Aniol, et al., Phys. Lett. B 509, 211 (2001).
M. Baylac, et al., Phys. Lett. B 539, 8 (2002).
Jefferson Lab experiment 99-115, K.S. Kumar and D. Lhuillier, spokespersons.
Jefferson Lab experiment 00-114, D.S. Armstrong and R. Michaels, spokespersons.
271