This progress report describes the research activities of the W.~K.~Kellogg Radiation Laboratory at the California Institute of Technology, supported by NSF grant PHY94-20470, during the period December 1997 to August 1998. The period from December 1997 to November 1998 represents the fourth year of a five year continuing grant. The research program involves a combination of experimental and theoretical work; the lab thus provides a unique and productive environment for the study of nuclear physics and related subjects. This broad range of activities is especially advantageous and attractive to the young scientists and students who participate in the program. Weekly seminars and journal club presentations foster substantial interaction and collaboration among faculty, research staff, and students in both the experimental and theoretical programs. Common access to the laboratory's resources such as computers, instrumentation, and technical expertise are also important aspects of the laboratory structure. \goodbreak
The last year's activities involve a number of significant accomplishments and new initiatives. We have advanced our program of parity violating electron scattering by implementing many improvements to the SAMPLE experiment (which is just beginning a production run at the Bates Laboratory), beginning construction of the liquid hydrogen target for the G0 experiment at Jefferson Lab, and obtaining first stage approval for a new experiment in parity-violating M\o ller scattering at SLAC. The HERMES experiment at DESY is now producing many new results in spin-dependent deep inelastic scattering and installed a new ring-imaging Cerenkov detector (RICH), which was a joint effort between our laboratory (with funding provided by Caltech) and Argonne National Laboratory. We played a major role in the design, construction, and assembly of a new high-density polarized 3He target for Hall A at Jefferson Laboratory, and obtained approval for two proposals to study spin-dependent electron scattering from polarized 3He. We have also participated in initiating two new programs at the Los Alamos Neutron Science Center: one to study the beta asymmetry in polarized neutron decay and one to search for the electric dipole moment of the neutron with greater sensitivity.
The theoretical program in the laboratory continues to attack a broad array of important problems in nuclear structure physics, computational studies of nuclear matter, lattice QCD, neutrino mass effects in supernovae, and chiral perturbation theory. This latter effort has been greatly strengthened by the addition of Senior Research Fellow Bira van Kolck (jointly funded by Caltech). Van Kolck (along with Visiting Associate Ryoichi Seki and Martin Savage from the University of Washington) organized a workshop in the application of effective field theory to nuclear physics. This very successful workshop, hosted by the Kellogg Laboratory and jointly supported by the Institute for Nuclear Theory, was attended by 28 theorists from around the world. In addition, we have continued our program of theoretical visitors which included H. Bethe, G. Brown, D. Riska, R. Schiavilla, K.-F. Liu, M. Chu, and S. Schramm.
There have been many personnel changes over the last year. In addition to Bira van Kolck, we have added postdoctoral scholars W. Fink (theorist funded by S. Koonin's grant) and T. Ito (experimentalist). J. Gao will join us as a postdoctoral scholar in experimental physics in January 1999. With these additions we will have the manpower strength we require to continue our vigorous program and meet our existing commitments. Research Fellow Todd Averett is leaving to accept an Assistant Professor position at the College of William and Mary. We graduated four Ph.D. students during the last year: J. Arrington, B. Bray, J. White, and P. Wrean. Finally, we proudly note that a recent Ph.D. graduate, Bryon Mueller, was awarded the Peter Demos Prize for oustanding Ph.D. work at the Bates Laboratory in conjunction with his work on the SAMPLE experiment and former Research Fellow E. J. Beise (now a professor at Univ. Maryland) was awarded the Maria Goeppert-Mayer Award of the American Physical Society.
The laboratory infrastructure was greatly enhanced during the last year by the addition of a farm of 20 dual 300MHz Pentium processors running LINUX. This system, funded entirely by funds provided by Caltech, represents a major increase in our ability to analyze data from our ongoing experiments at DESY, JLAB, and Bates and allows us to function as an important center for data analysis for these projects. In addition, our laboratory network and connection to the campus network is being upgraded as part of the CITNET-2000 effort at Caltech.
Progress on the many research projects in the laboratory is detailed in the following sections, followed by a budget proposal for the next fiscal year and our list of publications.
II.A.1 SAMPLE
[T. Averett, R.W. Carr, B.W. Filippone, T. Ito, C.E. Jones, and
R.D. McKeown]
The magnetic moment of the proton is a basic property related to the nucleon quark substructure. The analogous coupling to the Z boson is equally significant, and can provide new information related to the strange quark-antiquark contribution to the proton's magnetic moment.
The magnetic moment of the proton can be decomposed into the individual quark components:
\mup = 2/3 \muu - 1/3 \mud - 1/3 \mus
The neutral weak magnetic coupling of the nucleon can then be related to the electroweak couplings and a remaining contribution from the strange quarks:
\mupZ = (\mup-\mun ) - 4\sin2\thetaW \> \mup - \mus\;,
where \mup and \mun are the (electromagnetic) nucleon magnetic moments, and \thetaW is the weak mixing angle. The value of \mus has been of considerable interest, and most theoretical predictions give \mus<0 with a typical magnitude of about -1/3 [M. J. Musolf et al., Phys. Rep. 239 1 (1994)].
The SAMPLE collaboration includes the University of Illinois, University of Maryland, RPI, Virginia Tech, Louisiana Tech, University of Kentucky, and MIT-Bates. The experiment measures parity violating electron-proton scattering at low Q2. At backward angles, the parity violating asymmetry (i.e., the asymmetry between cross sections for right- and left-handed incident electrons) is approximately proportional to the weak form factor of interest. [R. D. McKeown, Phys. Lett. B219, 140 (1989)]
The experiment is performed using a 200 MeV polarized electron beam incident on a liquid hydrogen target. The scattered electrons are detected in a large solid angle (~1.5 sr) Cerenkov detector at backward angles (130o< \theta < 170o), giving an average Q2 \approx 0.1 (GeV/c)2.
The SAMPLE collaboration has reported the first measurement of the neutral weak magnetic moment of the proton [B. A. Mueller, et al, Phys. Rev. Lett. 78 3824 (1997)] based on data acquired during 1995-96 at Bates. Since that time, significant efforts have been made to improve polarized beam reliability and to reduce systematic errors. This includes work in Kellogg to develop a feedback system to monitor and compensate helicity correlated laser beam motion (see Section II.C.3).
In February 1998 the Bates laboratory held a readiness review of the SAMPLE experiment. The review committee recommended a high priority effort to complete the hydrogen data set during 1998. As a result, we were scheduled for a test/demonstration run in May and production running in July-September 1998. We have completed the May run and the experiment is completely operational again with many demonstrated improvements. These include a new intensity feedback system that reduces helicity correlations in the beam, an energy feedback system to stabilize the energy of the beam separately for the ten time slots, new compensated solenoids in the injector for better spin control, new toroids with reduced noise, symmetric array of halo and luminosity monitors, more complete set of scintillation counters for pulse counting studies of the Cerenkov detectors, and a new modern data analysis system based on the new CERN package ROOT. The experience in the May run indicates that the experiment will take higher quality data with much improved reliability this summer. We expect to report new higher precision data for \mupZ during the next year, and we hope to be scheduled for the deuterium running in summer 1999.
[T. Averett, E.W. Hughes, J.S. Jensen, Y.G. Kolomensky]
a1. SLAC Experiment E154
SLAC Experiment E154 (spokesperson: Hughes) was the first 50 GeV fixed target experiment performed at SLAC. The experiment used a 50 GeV 80% polarized electron beam scattering off a polarized 3He gas target to study the neutron spin structure function. The experiment collected data in October and November of 1995. The E154 experiment represented a substantial improvement in statistical and systematic precision compared to the previous E142 experiment.
Two Physical Review Letters and one Physics Letter from the E154 experiment were published in the July 1997 issues. Work on this experiment is now complete.
a2. SLAC Experiment E155
In March and April of 1997, SLAC experiment E155 collected a large data sample, effectively reaching the proposal goals. The experiment involved scattering a 50 GeV polarized electron beam off a polarized ammonia target to extract the proton spin structure function g1p and a 6 LiD target to extract the deuteron spin structure function g1d. In addition to the two existing spectrometers from experiment E154, a third spectrometer arm at 10.5o was constructed to measure the nucleon spin structure functions at large Q2. The data is expected to be important for studying the Q2 variation of the nucleon spin structure functions and for extracting information on the gluon contribution to the nucleon's spin.
A short data collection period of approximately one week using the E155 targets polarized in the transverse direction was also performed. This data will provide information on the spin structure functions g1p and g2d over the new extended kinematic range. \medskip
Figure II.A.2a shows preliminary results on the proton spin structure function g1p from experiment E155 compared to the SLAC E143 results and the CERN SMC results. Figure II.A.2b shows preliminary results on the deuteron spin structure function g1d from experiment E155 compared to the SLAC E143 results and the CERN SMC results.
Although no final published results exist for the E155 experiment, progress in the analysis continues. With approximately a factor of two improvement in the statistical precision of the result compared to experiment E143, many of the individual asymmetry measurements in Bjorken-x bins are now limited by systematic uncertainties, both a blessing and a curse. The ultimate goal will be to extract the Q2 dependence of the proton and deuteron spin structure function g1. Preliminary results on the proton and deuteron spin structure functions have been released in the last year. Figures II.A.2a and II.A.2b show the new high precision data compared to the previous E143 results.
Over the past year, the Caltech group participated in a December 1987 test run to check the calibration of the E155 lead glass calorimeter, assembled at Caltech. The analysis and results of this test run are important for untangling the energy measurement of the electrons in the 10.5 degree spectrometer used for the measurements of the spin structure functions for the highest Q2 data. Publication of the E155 data will include the larger kinematic reach provided by this new spectrometer arm.
Last September the E155 Collaboration submitted and received full approval for a measurement of the transverse spin structure function, g2 for the proton and deuteron. The new experiment, called E155x, is scheduled to collect data in January 1999. The E155x experiment will perform the first precision measurement of the g2 structure function in a dedicated two month run. The experiment will search for non-zero values of the g2 structure function approaching the precision of the predicted g2WW prediction. The measurement should also determine the d2 matrix element over the range in Bjorken-x from 0.014 to 0.7.
It is worth noting that the final complete analysis of the E143 proton and deuteron measurements have been submitted for publication to Physical Review D in February 1998. This is essentially the same collaboration as E155. The differences are a new deuteron target and a higher beam energy for E155.
[P. Carter, A. Dvoredsky, B. W. Filippone, R. D. McKeown]
The main goal of the HERMES experiment is the isolation of the quark contribution to the proton spin. Inclusive measurements provide information on the total quark contribution while semi-inclusive measurements can yield information on the separate quark flavor contributions and distinguish between valence and sea as well as quark and anti-quark distributions. Additional measurements of charmonium, charmed meson production and multi-hadron distributions may provide the first data on the gluon contribution to the nucleon spin.
During 1997, high precision measurements of the inclusive proton spin structure function have been obtained (see Fig. II.A.3a). These high precision data provide important information on the momentum transfer (Q2) dependence of the structure function. These results have recently been submitted for publication.
Fig. II.A.3a HERMES data on inclusive measurements of the spin structure function of the proton g1/F1 compared with SLAC experiment E143 and CERN experiment SMC.
New information on the unpolarized distributions of the anti-quark sea have also been obtained from unpolarized semi-inclusive measurements (see Fig. II.A.3b). Comparison of \pi+ and \pi- yields from unpolarized hydrogen and deuterium targets allows access to the anti-quark distributions. These results help characterize the momentum transfer dependence of quark distribution functions when compared with the high Q2 measurments from Fermilab experiment E866. Detailed knowledge of the quark distribution functions is essential in interpreting forthcoming data from the Relativistic Heavy-Ion Collider (RHIC). Promising analysis of other physics from unpolarized targets is also underway, with studies of the nuclear transparency of 14N and \Lambda polarization. Caltech is playing a significant role in the analysis of the latter.
Fig. II.A.3b a) Comparison of HERMES data for {(\overline{d} - \overline{u})/(u - d)} vs x with several published parameterizations of the quark distribution functions. b) Comparison of HERMES data for \overline{d} - \overline{u} with Fermilab experiment E866.
For 1998 and 1999, installation of a deuterium atomic beam source will allow precise measurements of the deuteron spin structure function and provide the best data (when combined with the semi-inclusive data from the proton target) on the flavor decomposition of the quark contribution to the nucleon spin. In addition a newly installed dual-radiator Ring Imaging Cherenkov (RICH) detector will provide access (for the first time) to semi-inclusive kaon asymmetries. These data may permit direct access to the strange quark contribution to the nucleon spin.
The Caltech group has played a major role in the construction of the RICH detector, being responsible for the aerogel radiator installation (see section II.C.1). Using the newly developed highly transparent aerogels combined with a gas radiator (C4 F10), imaging of the Cherenkov rings will provide pion, kaon and proton indentification over the full range of the HERMES momentum acceptance (2 - 20 GeV/c). Caltech has also played leading roles in the design, construction and operation of other target and detector systems including the Target Optical Monitor (TOM) for monitoring of the 3He target polarization and the trigger hodoscope and Pb-scintillator pre-shower detector which are major components of the first-level HERMES trigger.
The data taken in 1995 formed the Ph.D. thesis of Bruce Bray (Ph.D. 1997), while the data from 1996 and 1997 will form the Ph.D. theses of Andrea Dvoredsky and Paul Carter.
[R. Carr, B. Filippone, E.W. Hughes, J.S. Jensen, C. Jones, Yu. G. Kolomensky, R. McKeown, D. Pripstein]
The Moller experiment is a precision measurement of the electroweak theory at low Q2. The experiment uses the new 50 GeV polarized electron beam facility at SLAC to scatter polarized electrons off unpolarized electons in a liquid hydrogen target. The Moller scattered electrons are then detected in an open forward angle spectrometer and the asymmetry in polarized electron-electron scattering is measured. The relationship between the Moller electroweak asymmetry, AMoller and the electroweak mixing parameter, \sin2 \thetaw, is directly calculable and given below:
AMoller = 8 x 10-8 \cdot Pbeam \cdot Ebeam (1 - 4 \sin2 \thetaw)
where Pbeam and Ebeam are the beam polarization and energy, respectively.
The asymmetry turns out to be very small at 50 GeV, on the order of 10-7, but the Moller scattering cross section is large (20 \mu barns). In a three month experiment, the asymmetry is expected to be determined to a statistical precision of approximately 10-8, yielding an uncertainty on \sin2\thetaw of 0.0008. Systematic uncertainties, in particular, related to false asymmetries coming from the control of the polarized beam need to be kept down to a level of better than 10-8. Previous parity violation experiments performed at BATES have achieved systematic uncertainties somewhat larger (2 x 10-8), but comparable to the goals of the Moller experiment. If the experiment achieves the stated precision, it would represent the most precise test of the electroweak theory at low Q2 (20 MeV2), and would be substantially better than what is expected to come from present and future atomic parity violation experiments.
In September 1998, the SLAC Program Advisory Committee granted first stage approval to SLAC Experiment E158 (Spokespersons: Kumar, Hughes, Souder). The meaning of the approval is that the case for the physics motivation has been successfully defended, but full funding and scheduling of the experiment depends on passing a series of technical milestones. The primary issues for achieving the milestones are the demonstration of adequate control of the electron beam and a sufficiently high resolution in the beam position and current measurement so as not to degrade the statistical uncertainty.
The improvements required to achieve the precision involve installing a 16 bit ADC developed by the Princeton group for a recent Hall A CEBAF experiment, and upgrading the signal processing and cable plant so as to reduce the rf noise in the present beam position monitors (BPM) and toroids (Section II.C.7). The beam test will be performed at 1 GeV beginning in January 1999 at SLAC.
In addition, a number of paper studies have been requested by SLAC in order to evaluate the feasibility of the Moller spectrometer and the Moller target, in particular. The Caltech group is responsible for the construction of the 1.5 meter long liquid hydrogen target. A report from the Caltech group is due at the beginning of 1999, including a preliminary design and feasibility study.
Over the past year, Caltech has been actively involved in running the local meetings with Hughes as the local coordinator from September 1997 until August 1998.
The goal of the upcoming year is to satisfy the SLAC milestones by winter, which would effectively open up the full funding for the experiment. At that time, the Caltech group would focus primarily on the full design and construction of the liquid hydrogen target. If all stays on track, the goal is to have the experiment ready to run in the autumn of 2000.
[B. W. Filippone, T. Ito, C. E. Jones and R. D. McKeown]
Analogous to the magnetic form factor discussed in II.A.1 above, the charge form factor of the proton may also have a contribution from strange quark-antiquark pairs [D.H. Beck, Phys. Rev. D 39, 3248 (1989)]. In this case, the effect vanishes at Q2=0 (since there is no net strangeness) but it may be sizeable at finite Q2. The ``G0'' experiment is a major approved proposal [91-017; D. Beck of Illinois is spokesperson] to study the strange vector (electric and magnetic) form factors of the nucleon at TJNAF. These measurements will be in the range 0.2 < Q2 < 1.0 (GeV/c)2, higher than the SAMPLE experiment at Bates. By performing measurements of the parity violating asymmetry in elastic electron scattering at both forward and backward angles, the electric and magnetic neutral weak form factors will be separated. One can then solve for both the strange electric and magnetic form factors in this Q2 range.
The experimental design consists of a superconducting toroidal magnetic spectrometer, a liquid hydrogen target, scintillation counters, and high rate electronics. The Caltech group is responsible for the liquid hydrogen target, which will be based on the design of the SAMPLE target and LH targets that have been constructed at JLAB for other experiments. Engineering, drafting, machining, assembly, and test of the target is the responsibility of the Kellogg Laboratory in collaboration with E. Beise at the University of Maryland. (Construction funds for the G0 target will be part of the supplemental equipment proposal for this experiment.) Construction and design are presently underway and progress is described in section III.C.4.
The backward angle electron scattering measurements will be made by detecting the backward scattered electrons directly. The forward angle elastic electron scattering will be studied by detecting the recoil protons. The combination of magnetic analysis and time-of-flight measurement will allow separation of elastic events.
The magnetic spectrometer is scheduled for delivery and test at the University of Illinois in early 1999. At present, installation of the experiment at JLAB is anticipated during 2000 followed by beam commissioning studies in 2001.
[B. W. Filippone and R. D. McKeown]
A spinless nucleus with zero isospin can have only two vector form factors: one from the light quarks (up and down) and one from the strange quarks. These form factors can be separated by measuring both the electromagnetic and neutral weak form factors. The strange form factor may arise from either the nucleon strange electric form factor or through many-body nuclear effects.
Our group is involved in an approved proposal at TJNAF [91-004; E. Beise, the spokesperson, was a Senior Research Fellow in our group when the proposal was written] to measure the neutral weak elastic form factor in 4He via parity violating electron scattering. The measurement will be performed at Q2=0.6 (GeV/c)2 corresponding to the second maximum in the elastic form factor. Predicted effects from strange quarks are large enough to even reverse the sign of the parity violating asymmetry at these kinematics. The proposed experiment will use the high resolution spectrometers in Hall A with high current polarized beam at 3.6 GeV incident on a high pressure circulating 4He gas target.
[T. Averett, B. Filippone, S. Jensen, C. Jones, E. Hughes, R. McKeown, and D. Pripstein]
We are collaborating with Princeton (G. Cates), Temple (Z.E. Meziani) and others on an approved proposal at TJNAF [94-010] to study the spin dependent structure function of the neutron in deep inelastic scattering as a function of Q2 using polarized 3He. The physics goal is to connect the high-Q2 deep inelastic structure function to the low-Q2 inelastic structure governed by the resonance region. There are several theoretical predictions for the connection of the Ellis-Jaffe sum rules to the Drell-Hearn-Gerasimov sum rules that can be tested in such experiments.
Tremendous progress has been made over the past year in the preparation for this first polarized helium-3 target experiments to be conducted in Hall A at CEBAF. The experiment is presently scheduled to collect data in the fall of 1998. The Caltech group has been an active member of this new helium-3 collaboration with one research associate and one graduate student working full time on the project. The graduate student, Steffen Jensen, will earn his PhD thesis on the Drell-Hearn experiment.
By August of this year, the full helium-3 target construction will be complete and assembled at the lab. Already the target without the full laser system has recorded a polarized helium-3 signal and a water signal. First data collection with this target in beam is expected to begin in September.
The Caltech group specifically has been in charge of the laser and optics setup and the construction of the inner target ladder and target support assembly. The laser setup is a true state-of-the-art system, employing seven 30 Watt diode systems in a compact and flexible optical line. Four of the diode lasers will be used to pump the target rubidium in the longitudinal spin configuration and the other three will be used for transverse pumping. The optics and lasers will be located approximately 3 meters from the polarized target in a shielded laser hut.
The target ladder and support which holds the polarized target, dummy target cells and a beam position measurement device was designed by CEBAF and Caltech in collaboration, and constructed in the Caltech shops.
The future plans of the effort are to participate actively in the running of the first helium-3 target experiments and in the subsequent analysis of the data to begin in the winter following the first experiment. Although no final decision has been made yet on where the analysis will be performed, the Kellogg group has a large new pentium-based PC computer farm as well as four 266 MHz Alphas that were used in the analysis of the SLAC E154 experiment in 1996. It is likely that the Caltech group will choose to use some of this enormous computer power for the CEBAF polarized helium-3 analyses.
[J. Arrington, T. Averett, B. Bray, J. S. Jensen, B. W. Filippone, E. Hughes, and R. D. McKeown]
High energy electron scattering from nuclei can provide important information on the wave function of nucleons in the nucleus. In particular, with simple assumptions about the reaction mechanism, scaling functions can be deduced that, if shown to scale (i.e. are independent of length scale or momentum transfer), can provide information on the momentum and energy distribution of nucleons in the nucleus. Theoretical studies have also suggested that such measurements may provide direct access to short-range nucleon-nucleon correlations.
Figure II.A.8a Scaling function F(y) vs. the scaling variable y for an Fe target at different scattering angles (corresponding to different four-momentum transfers).
Initial analysis of experiment 89-008 at Jefferson Lab [B. W. Filippone - cospokesperson] has been completed. This measurement significantly extends the kinematic coverage of the earlier SLAC experiment NE3 (in which we were major participants). The scaling function F(y) from the new experiment is shown in Fig. II.A.8a for a range of four-momentum transfers.
In a simple impulse approximation picture for the reaction, the scaling function, defined as the inclusive cross section scaled by the e-N cross section, should be independent of momentum transfer and reflect an integral over the momentum distribution of nucleons in the nucleus. In this approximation the scaling variable y is simply the minimum momentum of the struck nucleon along the direction of the virtual photon. The data for y < 0 appear to show the predicted scaling behavior. For y > 0 other inelastic processes (beyond quasielastic scattering) begin to dominate the reaction. A more careful look at the scaling function vs. Q2 (see Fig. II.A.8b) shows a violation of the scaling for smaller values of Q2 as is expected if final-state interactions of the struck nucleon play a significant role. In addition a small scaling violation is suggested at high Q2 for the lowest value of y suggesting the increasing importance of other inelastic processes beyond quasielastic scattering (eg. pion production and deep inelastic scattering).
Figure II.A.8b Scaling function at fixed value of y vs. four-momentum transfer. Also shown are the earlier data from SLAC experiment NE3.
Our group was responsible for the design, assembly and testing of the trigger electronics for the two spectrometers in Hall C. We have also led the analysis effort for the experiment with John Arrington receiving his Ph.D. thesis from this work. This analysis has been a critical component in providing a detailed understanding of the two spectrometers in Hall C.
[C. E. Jones and R. D. McKeown]
We have joined a collaboration to measure the electric dipole moment of the neutron using ultracold neutrons (UCN) at LANSCE. A nonzero edm of the neutron is a signature of time reversal violation and of parity violation. The observation of T violation in the neutral kaon system has led to a standard model prediction of a neutron edm of order < 10-31 e-cm, a value well beyond the sensitivity of any existing experimental technique. However, extensions to the standard model which are needed to explain the observed baryon-antibaryon asymmetry in the universe increase the predicted neutron edm to the range 6 x 10-28 < d_n < 2 x 10-25. A high precision measurement of the neutron edm combined with the observed baryon-antibaryon asymmetry would severely constrain the acceptable theories. The goal of the experiment proposed for LANSCE is to ultimately achieve a sensitivity of 4 x 10-28 e-cm, an improvement of two orders of magnitude over present experimental limits and well within the range for dn predicted by current theories that explain the baryon-antibaryon asymmetry in the universe. The EDM collaboration submitted a letter of intent to the LANSCE management at a review of fundamental neutron experiments organized by the laboratory in November 1997. The experiment received a favorable reception at that time and has been allocated beamtime for tests this coming fall.
The experimental technique proposed for the experiment involves trapping UCNs from the LANSCE spallation source in a superfluid helium bath that contains a small fraction (~10-9) of polarized 3He. The bath will be in a volume of diameter ~10 cm covered by a uniform magnetic field of magnitude ~5 mG and a uniform electric field of magnitude 5 kV/mm. The polarized 3He serves three purposes; to polarize the ultracold neutrons, as a magnetometer to monitor the holding field, and as the UCN spin precession analyzer to distinguish the component of neutron spin rotation dependent upon the electric field, which can only arise as a consequence of a nonzero edm.
At Caltech, we will initially be responsible for developing and testing ways to introduce polarized 3He into the superfluid helium and maintain the polarization of the 3He for sufficiently long times to make a significant measurement of the neutron edm. C. Jones has been working with an undergraduate student to put together and test the laser used for the metastability exchange optically pumped polarized 3He target developed earlier at Caltech. We have also been involved in discussions with physicists at Los Alamos regarding the design of the optical pumping volume and transfer tube to feed the polarized 3He into the superfluid. We will come up with a set of measurements to determine the relaxation rate of the polarized 3He in the superfluid volume. We plan to take part in the test runs this coming fall and to ship the LNA laser system to LANSCE this winter for integration and testing in the prototype setup next spring and summer.
[B. W. Filippone, T. Ito, C. Jones, R. D. McKeown, J. Yuan]
We are major collaborators on a new experiment to measure the angular correlation between the neutron spin direction and the electron momentum following neutron beta-decay using ultra-cold neutrons (UCN). This measurement is usually characterized, via angular integration, by a rate asymmetry A for detected electrons along and opposite the neutron spin direction. Assuming time-reversal invariance, A is directly related to the ratio of the vector to axial-vector weak coupling constants GA. When combined with precision measurements of the neutron lifetime, GA and GV can be separately determined. Precise measurements of these quantities are key inputs in the Cabbibo-Kobyashi-Maskawa (CKM) matrix of standard model couplings.
Previous determinations of GV from nuclear beta-decay suggests a disagreement with the unitarity of the CKM matrix. Also previous measurements of A (which has a magnitude of about 0.1) with fractional uncertainties of ~1% disagree at the 3-4% level. This new measurement could achieve a total uncertainty of ~0.2% with significantly different systematics compared to the previous experiments. A comparison of the proposed measurment with the earlier experiments is shown in Fig. II.A.10a. While the earlier measurements used cold neutrons produced at reactor facilities, we would use the pulsed 0.8 GeV proton beam at LAMPF to produce pulses of ultra-cold neutrons. The pulsed beam considerably reduces background by allowing beta-decay detection to occur during beam-off periods, something not possible at reactor experiments. Using ultra-cold neutrons allows the production of essentially 100% neutron polarization via a solenoidal magnetic field filter.
Figure II.A.10a Expected results from the proposed measurement compared with previous measurements.
A schematic overview of the experiment is shown in Fig. II.A.10b. Production of ultra-cold neutrons has been tested with a rotor source (neutrons backscattering from a moving ``wall'') potentially providing a density of 10 - 20 n/cc. However the availability of the pulsed proton beam for neutron production may allow a significant increase in UCN density. Simulation studies of a solid deuterium moderator suggest dramatic increases in density may be possible - up to 100 - 1000 n/cc. The best UCN density presently available is at the ILL reactor where ~100 n/cc have been obtained. These higher densities could potentially push the fractional statistical error in the determination of A below 0.1%.
Figure II.A.10b Schematic diagram of apparatus used to measure the correlation of the neutron spin with the electron momentum following neutron beta-decay.
Caltech is responsible for developing, constructing and installing the beta-decay detectors for the measurement. We are investigating the use of tracking detectors to measure the electron angle and to reject background from backscattered electrons. The angle measurement will allow an important systematic check of the angular correlation (most previous experiments averaged over the distribution). This development will be funded through Caltech; no equipment money for the project is requested in this update.
While at Argonne National Laboratory, C. E. Jones submitted proposal 93-016 to TJNAF to measure the spin-dependent asymmetry in quasielastic scattering of polarized electrons from polarized tritium. The proposal was conditionally approved at the time, pending demonstration of the target using hydrogen. The experiment would run in Hall C at TJNAF using two spectrometers to make simultaneous measurements of the transverse and transverse-longitudinal asymmetries in the range Q2 = 0.2 - 0.8 (GeV/c)2 with sufficient accuracy to serve as a benchmark for theoretical calculations of the spin observables in the three-body system. The target would contain polarized tritium with a small amount of polarized hydrogen, so that the measured quasielastic scattering asymmetries from tritium could be directly compared with the elastic scattering asymmetries from hydrogen to look for possible medium modification of nucleon form factors. Since the proton form factors are well known, this kind of information about the ability to extract nucleon form factors from quasielastic scattering from three-body targets is important to experiments which propose to use polarized 3He to study the properties of the neutron.
We intend to continue development of the target for this experiment at Caltech (see section II.C.6).
[C. E. Jones]
We are participating in experiments CE66 and CE68 at IUCF. CE66 is a test of the laser-driven polarized hydrogen and deuterium internal targets developed at Argonne National Laboratory and the University of Illinois Urbana-Champaign, and CE68 is an experiment which will use the polarized deuterium internal target to measure the deuteron wavefunction through spin-dependent quasielastic scattering of polarized protons from polarized deuterium. The experimental collaboration consists of UIUC, ANL, IUCF, Caltech, Univ. of Erlangen, MIT and Univ. of Colorado.
The internal target was installed in the ring at IUCF in late 1996. After short runs in March and June 1997 to test the detectors and the target control and monitor systems, we had a longer run in September 1997 to measure the target polarization. During that run the target was quite unstable, and following the run substantial progress was made towards improving the stability of the polarization and dissociation fraction of the polarized gas in the target region, primarily through improved coatings and through the introduction of a small amount of O2 into the dissociator which acted to clean the surfaces of the target. Dry runs without beam were made in December and January during which the target performance was quite robust. In March 1998 we had a long run to measure the figure of merits of the polarized hydrogen and deuterium targets. Analysis of the analyzing power data from that run is ongoing and is anticipated to be finished by October 1998. Preliminary analysis of the hydrogen data shows that the nuclear polarization measured in the scattering reaction agrees with the nuclear polarization one infers from measurements of the atomic polarization and the molecular fraction, assuming the system is in spin temperature equilibrium. The results of the measurements on polarized deuterium confirm that the target is nuclear polarized, a result obtained using a low energy fusion reaction by the group at Argonne National Laboratory. We await the final analysis to make statements about the magnitude of the polarization.
These runs were the first tests of the polarization of a laser-driven source in an accelerator beamline. This kind of optically-pumped source for the hydrogen isotopes has been under development for a decade and may offer higher polarization density than the more conventional atomic beam source. If these tests indicate that the polarization and density are sufficiently high and robust under experimental conditions, then this type of source for a polarized proton or deuterium internal target will be considered for the Hermes experiment. We plan to have a long run at IUCF late this year to collect the data for experiment CE68.
[L. Diaconescu, R. Seki, and U. van Kolck]
Currently, the only way to solve QCD in the non-perturbative regime is on the computer, with lattice regularization. At sufficiently low energies, this solution can be reproduced by the most general effective field theory (EFT) consistent with the known symmetries of QCD. Chiral Perturbation Theory (\chi PT) has emerged in the last few years as a promising tool to systematically understand long-known nuclear features in a framework which is consistent with QCD and other low-energy hadronic processes, for which facilities such as IUCF, Uppsala, TUNL, TRIUMF, Mainz, Saskatoon, and Bates have produced very precise data.
The application of EFT ideas to nuclear physics involves a novel interplay of perturbative and non-perturbative physics, mixing subtle issues of regularization, renormalization, fine-tuning, power counting, and role of potentials. A joint Caltech/INT workshop was organized by R. Seki, U. van Kolck, and M. Savage (University of Washington), and held at Kellogg on February 26 and 27, 1998, to discuss these issues ({\it Nuclear Physics with Effective Field Theory}, R. Seki, U. van Kolck, and M. Savage (editors), World Scientific, Singapore, to appear). Previous successes in describing nucleon-nucleon (NN) scattering and reactions on the deuteron were reviewed (U. van Kolck, ``Overview of the Workshop on Nuclear Physics with Effective Field Theory'', to appear in the Proceedings), unresolved issues were settled, and new research directions were identified.
Our research in this period has focused on investigating chiral symmetry breaking by instantons, further understanding renormalization in a non-perturbative context, applying EFT methods to the solution of the three-nucleon (3N) system, studying 3N forces, considering other reactions involving the deuteron, and extending these ideas to kaonic systems.
1.1 Role of instantons in finite-temperature QCD with dynamical quarks
[S.M. Ouellette (Lauritsen Laboratory, Caltech), R. Seki, M. Chu (Chinese University of Hong Kong), and S. Schramm (GSI)]
We are investigating the role of instantons in chiral symmetry restoration by carrying out finite-temperature lattice QCD calculations with T3E's at NERSC (National Energy Research Supercomputer Center), SDSC (San Diego Supercomputer Center), and TACC (Texas Advanced Computing Center). We use the lattice size 163 x Nt with Nt running from 4 to 16 by steps of 2, with \beta=5.54 and two flavors of light staggered quarks of mass mqa= 0.0125. In each case, we use a molecular dynamics time step of dt=0.02 and apply ample thermalization steps, 300 steps or more. The cooling is made for 200 steps. The temperature of a gauge configuration, T = 1/ Nt a, is varied by changing the number of time slices Nt, instead of changing a. The latter is often used, but it involves an uncertainty because a itself is a function of the lattice inverse coupling \beta. In the asymptotic scaling regime, a(\beta) can be calculated by perturbation theory, but one is restricted to a narrow range of \beta, and hence temperature. We choose to keep \beta fixed, and hence a. We find that the instanton size parameter deduced from the correlation function decreases from 0.44fm below the phase-transition temperature Tc (\approx 150MeV) to 0.33fm at 1.3 Tc. The topological susceptibility decreases rapidly below Tc, showing the apparent restoration of the U(1)A symmetry already at T \approx Tc. A short manuscript is in preparation.
1.2 Effective field theory of short-range forces
[U. van Kolck]
For typical momenta much smaller than the pion mass, the EFT for nucleons contains short-range interactions only. In this case, an explicit solution of the low-energy two-heavy-particle system can be given, whether there is fine-tuning (as for the NN system) or not. I have shown that a controlled expansion in momentum does exist for the full two-body amplitude in any regularization scheme. When applied consistently, the renormalized EFT is equivalent to the effective range expansion, to a Schr\"odinger equation with a pseudopotential, and to an energy expansion of a generic boundary condition at the origin. I have also discussed when the amplitude can alternatively be obtained from an expansion of the potential. (U. van Kolck, ``The Aleph'', to appear in the Proceedings of the Joint Caltech/INT Workshop on ``Nuclear Physics with Effective Field Theory''; and U. van Kolck, ``Effective Field Theory of Short--Range Forces'', in preparation.)
1.3 Effective theory solution of the three-nucleon system
[U. van Kolck, P.F. Bedaque (Institute for Nuclear Theory), and H.-W. Hammer (TRIUMF)]
The systematic momentum expansion of the two-body amplitude can be used to solve the three-body system. We have applied this method to neutron-deuteron (Nd) scattering below the deuteron break-up threshold. In the J=3/2 channel, contact three-body forces are suppressed by the Pauli principle, and to third order in the expansion we can calculate the Nd scattering amplitude using low-energy constants entirely determined from low-energy NN scattering. We have found good agreement with an Nd phase-shift analysis at non-zero energies, and obtained a theoretical Nd scattering length, ath=6.33\pm 0.10 fm, which compares very well with the experimental value, aexp=6.35\pm 0.02 fm (P.F. Bedaque, H.-W. Hammer, and U.van Kolck, ``Effective Theory for Neutron-Deuteron Scattering: Energy Dependence'', Phys. Rev. C, to appear). We are now studying the J=1/2 channel, where puzzling renormalization effects seem to demand three-body forces in leading order in the momentum expansion (P.F. Bedaque, H.-W. Hammer, and U.van Kolck, in progress).
1.4 Chiral symmetry and three-nucleon forces
[U. van Kolck, J.L. Friar (Los Alamos National Laboratory), and D. H\"uber (Los Alamos National Laboratory)]
For typical momenta comparable to the pion mass, the EFT contains pions as explicitly degrees of freedom as well. This EFT can be used to derive the most general 3N potential consistent with chiral symmetry in a momentum expansion (U. van Kolck, Phys. Rev. C 49 (1994) 2932). We have reviewed the role 3N forces play when few-nucleon systems are solved with ``realistic'' NN potentials, and discussed the \chi PT approach to these forces. We have compared 3N forces that are used today regarding their two-pion-exchange components stemming from the (nominal) leading- and subleading-order Born terms and pion-rescattering graphs. We have demonstrated that the short-range c-term of the Tucson-Melbourne force violates chiral symmetry; numerically, this term is unnaturally large and should be dropped. The class of two-pion-exchange 3N forces then becomes rather uniform. (J.L. Friar, D. H\"uber, and U. van Kolck, ``Chiral Symmetry and Three-Nucleon Forces'', submitted to Phys. Rev. C.)
1.5 The N N \rightarrow d \pi reaction near threshold
[U. van Kolck, C.A. da Rocha (Universidade Estadual Paulista), and G.A. Miller (University of Washington)]
The reaction pp \rightarrow pp \pi0 near threshold is of great interest because it is sensitive to short-range nuclear dynamics. It can be described using power-counting arguments as an organizing principle for interactions from the most general chiral Lagrangian (including an explicit isobar field); however, uncertainties arising from short-range one-pion-two-nucleon operators and from the choice of ``realistic'' NN potential are large (U. van Kolck, G.A. Miller, and D.O. Riska, Phys. Lett. B 388 (1996) 679). We are extending this approach to the N N \rightarrow d \pi reaction near threshold, expected to be under better theoretical control. We are examining not only the three formally leading mechanisms ---Weinberg-Tomozawa (WT) term, impulse term, and isobar excitation--- but also sub-leading yet potentially large contributions ---including S-wave pion-rescattering, the Galilean correction to the WT term, and short-ranged contributions. (C.A. da Rocha, G.A. Miller, and U. van Kolck, ``The N N \rightarrow d \pi Reaction near Threshold in a Chiral Power Counting Approach'', in preparation.)
1.6 Meson exchange currents in parity violating electron scattering on the deuteron
[L. Diaconescu, U. van Kolck, and R. Schiavilla (Old Dominion University)]
Extraction of the nucleon strange magnetic form factor from parity violating electron scattering on the proton at backward angles is limited by the uncertainty in radiative corrections to the axial vector coupling form factor. These contributions can be disentangled with scattering on a deuteron target to be carried out by the SAMPLE collaboration. We are examining the effects of meson exchange currents in the observable asymmetry. We started with pionic currents, whose effects are expected to be larger due to the long-range nature of the pion and the large size of the deuteron. These currents are being calculated in \chi PT, and their numerical size will be estimated using a realistic deuteron wave-function.
1.7 Heavy kaons in effective field theory
[S.M. Ouellette (Lauritsen Laboratory, Caltech), R.Seki, and U. van Kolck]
In standard chiral perturbation theory, kaons are treated in the same way as other pseudoscalar bosons (pions and eta). These bosons are treated fundamentally as Goldstone bosons which pick up (small) finite masses due to the explicit chiral symmetry breaking by the quark masses. Phenomenologically, the kaons and the eta are much heavier than the pions, and there is evidence that their dynamics is also richer in the sense that it might give rise to shallow bound states in \bar{K} K and \bar{K} N systems. In the region of momenta much smaller than their masses, the kaons and the eta can alternatively be treated as heavy particles. We are investigating consequences of applying the technique of heavy-quark effective theory to these bosons. We expect that this approach will simplify the description of their dynamics, allowing for example a systematic implementation of unitarization procedures. We are examining the simpler case of \pi K interactions, and plan to extend this approach to \bar{K}K, K N, and \bar{K} N interactions involving the \Lambda (1405).
[H.-M. M\"uller, J.A. White, G. Mart\'{\i}nez-Pinedo, S.E. Koonin, R. Seki, U. van Kolck, and P. Vogel]
Methods for the exact solution of the nuclear shell model based on Monte Carlo techniques have been developed and implemented at Kellogg for a number of years. Shell Model Monte Carlo (SMMC) methods can overcome some limitations of conventional diagonalization techniques, and calculations based on these methods continue to be pursued for finite nuclei, including processes of astrophysics interest, and nuclear matter.
2.1 Studies of nuclear matter in coordinate space
[H.-M. M\"uller, S.E. Koonin, R. Seki, and U. van Kolck]
We are studying thermal properties of infinite nuclear matter by applying SMMC methods on a lattice. The nucleons interact by Skyrme-like Hamiltonian entailing on-site and nearest-neighbor interactions; this makes the system similar to a multi-species Hubbard model, so that a variety of numerical methods developed in that context can be employed. Our studies will concentrate on the behavior of nuclear matter at various temperatures and densities: questions about fragmentation, pair correlations, and sound modes will be addressed. We have implemented a full three-dimensional model on the computer that includes spin-spin and isospin-isospin exchange terms in a Skyrme-like fashion. The computer code has been rigorously tested. We have been pursuing two strategies. One is a phenomenological approach in which potential parameters are fitted to binding energies and saturation densities. Even though the potential parameters are constrained by the Monte Carlo sign problem, we are able to reproduce energies and densities quite well. Furthermore, we are beginning to see phase transitions from a Fermi gas to a clustered system at reasonable temperatures. So far, cases of symmetric nuclear matter and pure neutron matter have been investigated. The other strategy involves the use of an NN potential determined from scattering data. We are solving the Schr\"odinger equation on a lattice to express potential parameters in terms of scattering length and effective range. The context of our Hamiltonian and a consistent power counting scheme, as it is used in effective field theories, are being investigated. The incorporation of more terms like spin-orbit and tensor interactions is under consideration. This work will form the Ph.D. thesis of H.-M. M\"uller. (Preliminary results were reported in H.-M. M\"uller and R. Seki, ``Lattice Regularization and Nuclear Matter Calculation,'' to appear in the Proceedings of the Joint Caltech/INT Workshop on ``Nuclear Physics with Effective Field Theory''.)
2.2 SMMC investigations of nuclear structure in rare earth nuclei
[J.A. White, S.E. Koonin, and D.J. Dean (Oak Ridge National Laboratory and University of Tennessee)]
This work demonstrated the first full oscillator basis shell model calculations in rare earth nuclei using the Monte Carlo shell model technique. Exact results with a pairing plus quadrupole Hamiltonian were compared with mean field and SPA solutions in several dysprosium isotopes from A=152-162, including the odd mass 153Dy. These nuclei were much heavier than can be handled in the conventional shell model. Properties of these nuclei at various temperatures and spin were explored. These included energy, deformation, moments of inertia, band crossing, pairing channel strengths, and evolution of shell model occupation numbers. Level densities were also calculated and, in the case of 162Dy, compared with experimental data. Exact level densities are important for improving the accuracy of neutron capture cross sections required in nucleosynthesis calculations. This project was the dissertation for J. White. What remains is to calculate level densities for several other masses to detect shell effects.
2.3 Shell model calculation of the \beta- and \beta+ partial halflives of 54Mn and other unique second forbidden \beta decays
[G. Mart\'{\i}nez-Pinedo and P. Vogel]
The nucleus 54Mn has been observed in cosmic rays. In astrophysical environments it is fully stripped of its atomic electrons and its decay is dominated by the \beta- branch to the 54Fe ground state. Application of 54Mn based chronometer to study the confinement of the iron group cosmic rays requires knowledge of the corresponding halflife, but its measurement is impossible at the present time. However, the branching ratio for the related \beta+ decay of 54Mn was determined recently. We use the shell model with only a minimal truncation and calculate both \beta+ and \beta- decay rates of 54Mn. Good agreement for the \beta+ branch suggests that the calculated partial halflife of the \beta- decay, (4.94\pm 0.06) x 105 years, should be reliable. However, this halflife is noticeably shorter than the range 1-2 x 106 y indicated by the fit based on the 54Mn abundance in cosmic rays. We also evaluate other known unique second forbidden \beta decays from the nuclear p and sd shells (10Be, 22Na, and two decay branches of 26Al) and show that the shell model can describe them with reasonable accuracy as well. (G. Mart\'{\i}nez-Pinedo and P. Vogel, Phys. Rev. C 81 (1998) 281.)
[J.F. Beacom, Y.-Z. Qian, and P. Vogel]
Astronomy with photons of eV energies is sensitive to the atomic processes in the object being studied. The nuclear processes in the object can be probed (for example) with neutrinos of MeV energies. Many new detectors for neutrino astronomy are being built, and interpretation of the data will require the tools of nuclear physics. Besides the astrophysical measurements, the same data can be used to determine or limit neutrino properties, in particular masses and mixing.
We also study physics of core-collapse supernovae including neutrino emission and heavy element nucleosynthesis. We are interested in how strong magnetic fields affect interactions of neutrinos inside and their emission from the compact neutron star. We also look for observable signatures of heavy element nucleosynthesis via rapid neutron capture (the r-process) in supernovae. These studies apply nuclear physics to understand astrophysical phenomena and make important connections between these two subject areas.
3.1 Solar neutrinos and matter-enhanced oscillations
[J.F. Beacom, A.B. Balantekin (University of Wisconsin), and J.M. Fetter (University of Wisconsin)]
The basis for the solar neutrino problem is that the measured flux of neutrinos from the nuclear fusion reactions in the solar core is lower than expected. In fact, current data seem to indicate that the suppression is energy-dependent and that the most tenable explanation is matter-enhanced neutrino oscillations. The mathematics of this problem can be cast in the form of a coupled-channels problem. There is a supersymmetry which relates the two channels, and this facilitates analytic studies, e.g., with a uniform semiclassical approximation. The general aspects of this problem were developed, and may be useful for mathematically similar problems in nuclear and atomic physics. The resulting solution is the electron neutrino suppression factor as a function of energy and the density profile. Since the form of the density is left unspecified, measurement of the energy dependence would allow inversion for the density as a function of radius. This would provide a probe of the solar interior complementary to helioseismology. While current detectors do not have the statistical power to make the inversion, the analytic solution is also useful in the forward problem, since numerical solution of these equations is tedious when it must be done for a wide range of mixing parameters and a non-point source of neutrinos. (J.F. Beacom and A.B. Balantekin, ``A semiclassical approach to level crossing in supersymmetric quantum mechanics,'' in {\it Springer Lecture Notes in Physics No. 502: Supersymmetry and Integrable Models}, H. Aratyn et al. (editors), Springer-Verlag, Berlin, 1998; and A.B. Balantekin, J.F. Beacom, and J.M. Fetter, Phys. Lett. B 427 (1998) 317.)
3.2 Supernova neutrinos and the \nu\tau mass
[J.F. Beacom and P. Vogel]
A type-II supernova produces of order 1058 neutrinos and antineutrinos of all flavors over several seconds, with a hierarchy of average energies. Typical predicted average energies are: \langle E \rangle \simeq 11 MeV for \nue, \langle E \rangle \simeq 16 MeV for \bar{\nu}e, and \langle E \rangle \simeq 25 MeV for \nu\mu, \nu\tau, \bar{\nu}\mu, and \bar{\nu}\tau. This hierarchy is caused by the different opacities in the proto neutron star for the different flavors. The SuperKamiokande detector (online now), and the Sudbury Neutrino Observatory (coming online this year) have an unprecedented sensitivity for detecting a Galactic supernova. Such a detection would allow precise measurement of the time and energy dependence of the burst. It would also allow a time-of-flight determination of the \nu\tau neutrino mass. The charged-current rate for the (almost massless) \nue and \bar{\nu}e neutrinos determines the supernova luminosity. Because of the hierarchy of average energies, the neutral-current scattering rate is dominated by the \nu\mu and \nu\tau neutrinos (and their antiparticles). We have shown how to compare the charged-current and neutral-current scattering rates to determine the \nu\tau mass. In the presence of statistical fluctuations, SuperKamiokande is sensitive to a mass down to about 50 eV, and Sudbury down to about 30 eV. We have shown in detail how to separately study the supernova and neutrino physics aspects using the same data, and how different assumed input parameters would affect the mass sensitivity. (J.F. Beacom and P. Vogel, ``Mass signature of supernova \nu\mu and \nu\tau neutrinos in SuperKamiokande,'' Phys. Rev. D, to appear; and J.F. Beacom and P. Vogel, ``Mass signature of supernova \nu\mu and \nu\tau neutrinos in the Sudbury Neutrino Observatory,'' submitted to Phys. Rev. D.)
3.3 Asymmetric neutrino emission and the origin of pulsar kicks
[Y.-Z. Qian and D. Lai (Cornell University)]
Rotating magnetized neutron stars, or pulsars, are observed to have large space velocities, which are most likely caused by kicks received at their birth. We studied one kind of kick mechanism which is based on asymmetric neutrino emission induced by strong magnetic fields of new-born neutron stars. We first investigated the asymmetry due to parity violation in neutrino interactions and its possible enhancement due to multiple neutrino scatterings (D. Lai and Y.-Z. Qian, Astrophys. J. 495 (1998) L103). Later we concluded that enhancement of this asymmetry is negligible in the bulk interior of the neutron star where local thermodynamic equilibrium holds to a good approximation (D. Lai and Y.-Z. Qian, Astrophys. J. 501 (1998) L155 (E)). We then studied the effects of strong magnetic fields on cross sections for neutrino absorption on free nucleons. We found that only extremely strong magnetic fields can change these cross sections enough to give significant asymmetry in neutrino emission through asymmetric distribution of the magnetic fields (D. Lai and Y.-Z. Qian, ``Neutrino Transport in Strongly Magnetized Proto-Neutron Stars and the Origin of Pulsar Kicks: the Effect of Asymmetric Magnetic Field Topology,'' Astrophys. J., to appear).
3.4 Gamma-ray signatures of supernova r-process
[Y.-Z. Qian, P. Vogel, and G.J. Wasserburg (Geological and Planetary Sciences, Caltech)]
We studied gamma-ray emission from the decay of progenitor nuclei made through the r-process. We found that if the r-process occurs in a supernova, a number of progenitor nuclei live long enough to produce gamma rays after the supernova becomes transparent to them. The gamma rays from a Galactic supernova would be detectable with the proposed Advanced Telescope for High Energy Nuclear Astrophysics (ATHENA). Detection of these gamma rays from a supernova would establish that supernovae are a site of the r-process. We also found a particularly interesting nucleus, 126Sn, whose long lifetime facilitates possible detection by ATHENA of its gamma rays from the nearby young Vela supernova remnant and from past supernovae in the Galaxy (Y.-Z. Qian, P. Vogel, and G. J. Wasserburg, ``Supernovae as the Site of the r-Process: Implications for Gamma-Ray Astronomy,'' Astrophys. J., to appear).
II.B.4 Quantum computation and information theory
[C. Adami, N.J. Cerf, R.M. Gingrich, and S.E. Koonin]
During the last year, we have analyzed several issues related to the foundations of quantum mechanics using an information-theoretic formulation of quantum entanglement. This work is a direct continuation of our previous research during the year 1996-97. More specifically, our effort has been centered on four research projects: (1) analysis of quantum coding and quantum inseparability by use of an information-theoretic approach previously developed by us; (2) study of quantum copying machines and derivation of no-cloning uncertainty relations; (3) investigation of the optical simulation of quantum networks; and (4) development of an improved quantum search algorithm based on nesting.
4.1 Information-theoretic approach to quantum coding and inseparability
[C. Adami, N.J. Cerf, and R.M. Gingrich]
We have investigated several applications of the unified framework for quantum information that we had developed in earlier work. First, we have considered the calculation of entropic bounds on the achievable rate of reliable transmission of quantum information through a noisy quantum channel (N.J. Cerf, Phys. Rev. A 57 (1998) 3330). In analogy with its classical counterpart, a noisy quantum channel can be characterized by its loss, a quantity that depends on the channel input and the quantum operation performed by the channel. The loss reflects the transmission quality: if the loss is zero, quantum information can be perfectly transmitted at a rate measured by the quantum source entropy. By using block coding based on sequences of n entangled symbols, the average loss (defined as the overall loss of the joint n-symbol channel divided by n, when n tends to infinity) can be made lower than the loss for a single use of the channel. In this context, we have examined several upper bounds on the rate at which quantum information can be transmitted reliably via a noisy channel, that is, with an asymptotically vanishing average loss while the one-symbol loss of the channel is non-zero. These bounds on the channel capacity rely on the entropic Singleton bound on quantum error-correcting codes that we had derived in earlier work. For the quantum erasure channel, our bound coincides with the exact capacity that was recently found by Bennett and coworkers. We have also analyzed the entropic Singleton bound in the case when the noisy quantum channel is supplemented with a classical auxiliary channel, and we concluded that a classical channel cannot be used to increase the Singleton bound.
Second, we have analyzed quantum inseparability using this information-theoretic approach. We have investigated the properties of the conditional amplitude operator, the quantum analog of the conditional probability (N.J. Cerf and C. Adami, submitted to Phys. Rev. A). We showed that the spectrum of the conditional operator characterizing a quantum bipartite system is invariant under local unitary transformations and reflects its inseparability. Furthermore, we proved that the conditional amplitude operator of a separable state cannot have an eigenvalue exceeding 1, which results in a necessary condition for separability. We then derived a related ``reductive'' separability criterion, based on the positive map \rho \to (Tr \rho) - \rho. Any separable state is mapped by the tensor product of this map and the identity into a non-negative operator, which provides a simple necessary condition for separability (N.J. Cerf, C. Adami, and R.M. Gingrich, submitted to Phys. Rev. A). This map reduces to time-reversal in the special case of two quantum bits, so that the reduction separability condition is then equivalent to partial transposition. A simple connection between this map and complex conjugation in the so-called ``magic'' basis was also displayed.
Finally, we have investigated the duality between the concepts of ``physical reality'' and quantum reality in terms of quantum information theory. We have shown that this formalism suggests a solution to quantum paradoxes such as the Einstein-Podolsky-Rosen and the Schroedinger-cat paradoxes (C. Adami and N.J. Cerf, in Lecture Notes in Computer Science, Proceedings of the 1st NASA Workshop on Quantum Computation and Quantum Communication, C.P. Williams (editor), Springer-Verlag, to appear).
4.2 Quantum copying machines and no-cloning uncertainty relations
[N.J. Cerf]
First, we have discussed an information-theoretic approach to quantum copying, relying on the notion of quantum loss, a quantity that reflects the transmission quality in a noisy quantum channel.
More specifically, an entropic no-cloning inequality has been derived for a Hilbert space of arbitrary dimension, which describes the tradeoff between the losses of the channels leading to the two copies: LA+Lb \ge 2S, where LA and LB are the losses characterizing outputs A and B of the cloner, respectively, while S is the source entropy. (N.J. Cerf, in {\it Lecture Notes in Computer Science, Proceedings of the 1st NASA Workshop on Quantum Computation and Quantum Communication}, C.P. Williams (editor), Springer-Verlag, to appear).
Then, we have introduced a family of asymmetric Pauli cloning machines (PCM), which produces two distinct (approximate) copies of a single quantum bit, each emerging from a Pauli channel. This is in contrast with the cloning machines considered in the literature, which are symmetric (both outputs being characterized by the same density operator). The family of PCMs relies on a parametrization of 4-qubit wave functions for which all qubit pairs are in a mixture of Bell states, and generalizes the universal quantum copying machine introduced by Buzek and Hillery. Using a particular class of asymmetric PCMs whose outputs emerge from (distinct) depolarizing channels, we have derived a no-cloning uncertainty relation governing the tradeoff between the quality of the copies of a quantum bit: a2+ab+b2\ge 1, where a2 and b2 are the depolarizing fractions of the channels associated with outputs A and B, respectively. It is, by construction, a tight inequality which is saturated using our PCM. Finally, the subclass of symmetric PCMs has been used in order to express an upper bound on the quantum capacity of the Pauli channel, extending the bound that was previously known for the depolarizing channel. In particular, the capacity of the Pauli channel of probabilities px=x2, py=y2 and pz=z2, was shown to vanish if (x,y,z) lies outside an ellipsoid whose pole coincides with the depolarizing channel underlying the UCM. This implies an upper bound on the capacity C of any Pauli channel associated with a point (x,y,z) located inside the ellipsoid, namely C \le 1-2(x^2+y^2+z^2+xy+xz+yz). (N.J. Cerf, submitted to Phys. Rev. Lett.).
We have considered the cloning of quantum states in a Hilbert space of arbitrary dimension (N.J. Cerf, Acta Phys. Slovaca {\bf 48} (1998), in press). We have generalized the Pauli cloning machines and introduced an (asymmetric) N-dimensional cloning machine. We then showed that the complementarity between the two copies produced by a Pauli cloning machine and its N-dimensional extension results from a genuine uncertainty principle, much like that associated with Fourier transforms. This uncertainty principle simply relates the probability distributions underlying the channels leading to the two outputs of the cloner. Finally, we have analyzed the semi-classical limit of an infinite-dimensional cloning machine.
4.3 Optical simulation of quantum networks
[C. Adami, N.J. Cerf, and P.G. Kwiat (Los Alamos National Laboratory)]
We have suggested a constructive method for simulating small-scale quantum circuits using optical setups composed of linear optical devices. It relies on the representation of several quantum bits by a single photon, and on the implementation of universal quantum gates using simple optical components such as beam splitters, phase shifters, etc. This avoids the recourse to non-linear Kerr media to effect quantum conditional dynamics, a severe constraint in the usual optical realization of quantum circuits. A drawback of this technique is clearly the exponential increase of the resources (optical devices) with the size of the circuit. Nevertheless, as optical components that simulate 1- and 2-bit universal quantum gates can be cascaded straightforwardly, a non-trivial quantum computing optical device can be constructed if the number of component qubits is not too large. As an illustration, we have presented the optical analogue of the Brassard et al.'s 3-bit circuit for quantum teleportation. This suggests that the optical realization of small quantum networks with present-day quantum optics technology is a reasonable goal, and that such a technique could be useful for demonstrating basic concepts of simple quantum algorithms or error-correction schemes. (N.J. Cerf, C. Adami, and P.G. Kwiat, Phys. Rev. A {\bf 57} (1998) R1477, and C. Adami and N.J. Cerf, in {\it Lecture Notes in Computer Science, Proceedings of the 1st NASA Workshop on Quantum Computation and Quantum Communication}, C.P. Williams (editor), Springer-Verlag, to appear).
4.4 Nested quantum search algorithm
[N.J. Cerf, C.P. Williams (Jet Propulsion Laboratory), and L.K. Grover (Lucent)]
We have initiated a collaboration which concerns the development of improved quantum algorithms for NP-complete problems. We have shown that a quantum algorithm can be devised that exploits the structure of a tree search problem by nesting the standard (unstructured) quantum search algorithm. The latter is used to construct a search operator which, itself, is nested within another search algorithm at an upper nesting level. The expected number of iterations required to find the solution of an average instance of a constraint satisfaction problem is estimated to scale as \sqrt{d^\alpha}, where d is the dimension of the search space and \alpha<1 is a scaling coefficient that depends on the nesting depth and the problem considered. When applying a single nesting level to a problem with constraints of size 2 such as the graph coloring problem, this constant \alpha is estimated to be around~0.62 for average instances of maximum difficulty. This corresponds to a square-root speedup over a classical nested search algorithm, of which our algorithm is the quantum counterpart. Nevertheless, it is an exponential speedup with respect to the time needed to solve the problem by use of the standard unstructured quantum search algorithm, which scales as \sqrt{d}. We believe that this technique is applicable in many structured search problems. (N.J. Cerf, L.K. Grover, and C.P. Williams, to appear in Applicable Algebra in Engineering, Communication and Computing).
[C. Adami, W. Fink, S.E. Koonin, R. Seki, and J.A. White]
The aim is to successfully apply methods of Theoretical Physics to various problems in other scientific disciplines, e.g., Biology, Chemistry, and Medical sciences (Ophthalmology in particular).
5.1 De novo protein design: effective sequence selection algorithms
[W. Fink, R. Seki, J.A. White, S.E. Koonin, S.L. Mayo (Braun Laboratories, Caltech), and D.B. Gordon (Braun Laboratories, Caltech)]
De novo protein design consists of two steps: selecting a desired tertiary structure and finding a sequence of rotamers (amino acid side-chain positions) that would stabilize this fold. Therefore, the task is to find the global energy minimum of a given protein by choosing the appropriate rotamers at each residue (protein site) with respect to their self-energy and interaction energy that are given by biological/chemical models. In general there is a vast number of possible rotamer combinations. Therefore, exhaustive search and other deterministic algorithms (e.g., branch-and-bound and dead-end elimination of Desmet et al.) are computationally limited to smaller proteins. To go beyond those limits one has to rely on non-deterministic methods such as self-consistent ensemble optimization (mean-field approach), stochastic methods, and genetic algorithms. We have developed a genetic algorithm and a Monte Carlo algorithm, the latter being an extension of the stochastic method that the group has fruitfully used in nuclear structure studies. These algorithms have already been successfully applied to smaller proteins, and application to larger proteins is in progress. The future goal is to develop algorithms that expand the range of computational protein design significantly beyond today's limits. Success would have a strong beneficial impact on other disciplines, biochemical, biomedical, and pharmaceutical research/industry, etc.
5.2 Chemical reaction networks -- origin of life
[W. Fink, C. Adami, and S.E. Koonin]
Tracing the origin of life backwards in Earth's history one has to answer the question: ``Which have been the ancestral processes of early life?" A scrutiny on recent literature suggests that the ``prebiotic broth theory" of Miller et al. may not be the only answer to that question. To examine alternative theories, we have developed a simulation platform for chemical reaction networks that allows one to calculate the concentrations of all participating chemicals as a function of time according to a set of reaction equations. Furthermore, we have successfully implemented chemical networks that simulate simple and compound boolean functions. An extension of the simulation platform for training a chemical reaction network to perform certain tasks is currently under development. Future goals are to understand how processes like autocatalysis and self-reproduction can evolve at random. Furthermore, we plan to explore how a chemical network can establish and maintain itself. This would improve the understanding of how a biological system like a cell can arise.
5.3 Realistic human computer eye model in Ophthalmology
[W. Fink, with University Eye Clinic Tuebingen, Eye Clinic Buch (Berlin), and University Eye Clinic CGC (TU Dresden)]
In 1909 Gullstrand introduced his famous eye model, which he called the ``exact schematic eye". It has six refractive surfaces: two for the cornea and four for a two-shell crystalline lens. Based upon this eye model we have developed a computer-based ray tracer which allows to simulate the visual mapping under various kinds of eye defects both, with and without refractive corrections (glasses, contact lenses, and intra-ocular lenses). Since the Gullstrand eye model suffers from the fact that it consists of spherical surfaces only, a more general model with arbitrarily shaped surfaces (B-splines, etc.) would be required in order to be able to visualize eye defects such as regular and irregular astigmatism. Furthermore, since PRK (photorefractive keratectomy) has become the state-of-the-art method to correct for myopia and hyperopia, it is of interest, to calculate the visual mapping of a cornea before, during, and after laser treatment. One of the future goals is to have a realistic and adjustable human computer eye model that allows to simulate the visual mapping of any given human eye under examination. Furthermore, it would be an interesting perspective, to computer-assist the surgeon while performing PRK, e.g., to calculate the ``final" cornea profile that is needed to remove high ametropia and to give the surgeon a feedback control of the progress that has been achieved so far during laser treatment. (A. Frohn and W. Fink, Invest. Ophthalmol. Vis. Sci. 39 (1998) S314; and W. Fink, H.J. Huebscher, and T. Seiler, Invest. Ophthalmol. Vis. Sci. 39 (1998) S1033.)
[R. Carr, P. Carter, B. W. Filippone]
A significant upgrade for the HERMES experiment has recently been completed with the installation of a Ring Imaging CHerenkov (RICH) detector for identification of pions, kaons and protons from 2 - 20 GeV. This will greatly improve the capabilities of the experiment to measure semi-inclusive asymmetries from identified hadronic final states. Aside from providing the first semi-inclusive asymmetries with kaons, this detector will greatly improve the ability to identify decaying particles, eg. \phi's, \Lambda's, and charmed mesons and baryons.
The detector takes advantage of new developments in the production of highly transparent aerogel, where a significant fraction of the Cherenkov photons can traverse the material without scattering. By imaging rings both from the aerogel and from a heavy gas (C4F10), \pi, K, p identification can be performed over the full energy acceptance of HERMES. This detector will be the first RICH detector to use aerogel as a radiator.
Caltech was responsible for design, testing and procurement of the aerogel radiator. This construction was funded through Caltech (about $100K) in order to allow completion of the project in time to install during the regular shutdown of HERA at DESY. An extensive series of tests have been conducted at Caltech to characterize the optical properties of this new material. Furthermore, we are involved in an ongoing study of optical characteristics of aerogel including measurements of the yield of Cherenkov light, refractive index, transmission, and scattering properties.
Figure II.C.1 Angular distribution of laser light scattered by aerogel
Among the properties measured to understand light propagation through aerogel is the length scale of structural inhomogeneities which scatter light through small angles. The angular distribution of radiation scattered by aerogel was measured with a laser and silicon photodiode. Figure II.C.1 shows the amount of radiation scattered within a maximum angle \Theta. The error bars are not statistical, but represent the range of length scales found in different aerogel samples. The scattering is well described by modeling aerogel as a random two phase network with the distance scale of fluctuations in the dielectric constant on the order of a few hundred microns. Such a large length scale indicates excellent optical quality which will translate to well resolved Cherenkov rings.
The full design, construction, and installation of the aerogel radiator, was completed on schedule in the spring of 1998.
[R. Carr, E. Goldberg, E.W. Hughes, J.S. Jensen, R. McKeown, D. Pripstein]
The new polarized helium-3 target laboratory that was built by Caltech in room 100 Kellogg is now fully operational. Construction of the laboratory was completed in June of 1997 and the first polarized helium-3 and water signals were obtained in July 1997.
Since last year the main focus of the laboratory has been geared towards preparing the new optical setup for the CEBAF helium-3 experiments. A prototype laser setup was built in the lab in winter of 1998 and recently a full optical assembly was completed in the laboratory and shipped to CEBAF in July 1998. A photograph of the complete assembly is shown below.
Over the past year, polarizations on the order of 20\% have been achieved with some of the old target cells from the SLAC experiments. These target cells had typically produced 35\% polarizations during the SLAC experiments. However, the laser power that exists in Kellogg is limited. The Caltech group has only approximately half the laser power compared to the SLAC experiments.
Recently, we have found a third Argon ion pump laser and intend to fix a second one that we have. This year we expect to have three operational Ti:sapphire lasers pumped by 20 Watt Argon ion lasers. These systems will give us more laser power and more importantly allow us to tune the laser light to different wavelengths. The explosion in diode laser technology has replaced Ti:sapphire lasers in most recent applications. However, for flexibility in wavelength and narrowband work for studying how optical quality may affect the polarization process, tunable Ti:sapphire lasers are still an excellent source. We have recently purchased a modern laser profile monitor and it is our intention to use this to study the optical quality and how it relates to the efficiency of polarizing our targets. Shown below is a measurement of the light coming from the diode lasers to be used at CEBAF using this new profile monitor.
Of course, the laboratory will its full target setup provides a wonderful environment for training new graduate students at Caltech for experiments elsewhere using polarized helium-3 targets. A new first year graduate student is planning to begin working in this laboratory starting in the autumn of 1998.
[T. Averett, C.E. Jones, P. Lee, K. McCarty R.D. McKeown, V. Savu]
The production of longitudinally polarized electron beams for parity violation experiments is generally accomplished via photoemission from GaAs with a polarized laser beam. The helicity of the electron beam is reversed by reversal of the helicity of the laser beam. In order to minimize systematic errors in the measurement of small (< parts per million) parity-violating asymmetries, one strives to insure that all other properties of the electron beam remain constant under helicity reversal. This includes, but is not necessarily limited to, the beam intensity and position. A necessary prerequisite to appropriate stability of the electron beam under helicity reversal is the stability of the laser beam used to produce the electron beam.
The laser beam helicity is reversed by a Pockels cell used as a \pm \lambda/4 plate. Unfortunately, the Pockels cell causes small but non-trivial distortions of the laser beam that result in helicity-correlated variations in various parameters of the laser beam. Our recent studies using the polarized source laser at Bates show that it is typical to observe helicity-correlated motion of the laser beam at the few hundred nanometer level. It appears that this effect is associated with distortion of the laser beam by the Pockels cell, coupled with interactions with the additional optics between the Pockels cell and the GaAs crystal. Although careful alignment of the Pockels cell and other measures have been shown to reduce these effects, one typically finds a residual helicity correlated beam motion at the GaAs location of order 100-200 nm.
We have developed a feedback system to steer the beam in a helicity-correlated fashion in order to remove these residual systematic laser beam motions. The laser beam position is continually monitored (separately for left- and right-helicities) using a quad-split diode that is capable of measuring the helicity correlated beam motion (in both x- and y-directions) to less than 100nm in a few minutes. An AR-coated glass plate is then tilted by a piezoelectric transducer (in sync with the helicity flip) by an amount computed from the measured beam motion in order to cancel the observed helicity correlated beam motion (on the average).
Figure II.C.3a Beam position asymmetries in x and y as a function of time. Arrows indicate where feedback was turned on and off.
The system has been tested and demonstrated to perform well using a HeNe laser system with a Pockels cell and mirror transport system to simulate the laser system used with polarized electron sources. Figure II.C.3a shows data for the helicity correlated position differences in \Delta x and \Delta y versus time. Arrows indicate where feedback was turned on and off. With the feedback off, one can clearly see helicity correlated beam motion of order several hundred nanometers due to long-term instabilities in the laser and Pockels cell. With the feedback on, the beam is consistently kept near zero in both x and y.
We are adapting some aspects of this system (particularly the piezo-driven steerer) at Bates for the SAMPLE run in summer 1998. Clearly, these techniques will be applicable at JLAB and SLAC in future parity-violation experiments and we intend to continue this development for potential future use at these laboratories.
[T. Averett, R. Carr, T. Ito, C. E. Jones and R. D. McKeown]
The Caltech group has the responsibility of designing and building the major part of the cryogenic liquid hydrogen target for the G0 experiment at JLab, consisting of the target cell, the cryogenic target loop, the gas handling system, and the transverse motion mechanism. The G0 experiment will measure parity violating electron scattering from hydrogen at forward and backward scattering angles to separate the strange electric and magnetic form factors of the nucleon (see section II.A.5 above), and will need a high density target capable of handling 40 \mu A of minimum ionizing electron current in order to obtain the high statistics needed. Because of Caltech's earlier experience designing, building and operating the LH2 target for the SAMPLE experiment, which has operated successfully in that experiment, the design of the G0 target is quite similar to the SAMPLE target.
Figure II.C.4a Schematic diagram of the cryogenic loop, hydrogen cell and helium cell of the liquid hydrogen target for the G0 experiment.
We have completed the design of the target cell, manifold, cryogenic loop and gas handling system. Figure II.C.4a shows the design of the G0 cryogenic target loop. One significant change to the target design that has happened in the last year is the choice of a more compact fin-tube style heat exchanger, which will reduce the total volume of liquid hydrogen in the target system. This heat exchanger has been built for us by the same company who supplied the heat exchangers for the JLab cryotargets. In the past year, we have also shown that it will be possible to operate the liquid hydrogen pump motor inside the cryoloop, in the field of the superconducting G0 magnet. Detailed design drawings using Autocad of the target and cryogenic loop are being made at Caltech.
Figure II.C.4b Closeup view of the target region of the liquid hydrogen cryotarget.
The target cell for the G0 experiment needs to be thinner than that used for SAMPLE because of the need to minimize multiple scattering and energy loss of the low energy protons which are detected in the experiment. A thickness of 0.007'' was chosen for the hydrogen cell wall and endcap thickness, with a 0.003'' thick truncated cone flow separator inside the liquid hydrogen region (see Figure II.C.4b). The construction technique used to make the thicker SAMPLE target cells was not able to be applied successfully to make uniformly thin G0 cells, so we have come up with a new technique and the tooling necessary to make the G0 hydrogen target cell. We have constructed and tested the flow separator cone and constructed prototype cells with the desired wall thickness which were burst tested to 70 psi differential pressure, well above the target operating pressure of 25 psi.
In the upcoming year we will complete construction of the liquid hydrogen cryotarget and design and construct the mechanism for positioning the target and moving it in and out of the beam path. The control and monitoring system for the target is being designed and built in parallel by the group at the University of Maryland.
[R. Carr, E. Hughes, C. E. Jones, and R. McKeown]
We plan to build a 1.5 meter long liquid hydrogen target for use in the parity-violating Moller scattering experiment at SLAC (see section II.A.4). The target will consist of liquid hydrogen flowing through a closed loop system, and its design will be based upon those the cryogenic targets for the SAMPLE and G0 experiments, both of which were done by the Caltech group. A schematic of this target is shown in Figure II.C.5a.
Figure II.C.5a Schematic diagram showing the layout of the target cell and recirculation loop for the Moller PV experiment.
The target cell will be cylindrical, 3 inches in diameter and 150 cm long. The entrance and exit windows will have the same radius of curvature to ensure that the liquid hydrogen target thickness is independent of the position of the incident electron beam. A separate helium cell will be used to maintain a constant pressure on the concave entrance window.
The cryogenic loop will consist of a heat exchanger, liquid hydrogen pump, hydrogen target cell and manifold. A heater within the target loop will be in a feedback loop with the electron beam current to compensate for a loss of beam on target, maintaining a constant hydrogen target temperature. Target control will be done using a PC which monitors the state of the gas handling system and the temperature and pressure of the target hydrogen and helium at several locations around the cryoloop. Stepping motors will be used to position the target both horizontally and vertically.
Density fluctuations in the Moller target must be maintained to \le 10-4 on a pulse- to -pulse basis or the statistical error on the PV asymmetry will be substantial increased. At present, the SAMPLE experiment has determined that with a 40 \mu Amp beam current, the target experiences less than 0.1% density fluctuations. For the Moller target, we plan to monitor and study the luminosity on the target using a low scattering angle (1 mrad) detector to search for density fluctuations. At these small angles the asymmetry in the elastic scattering cross-section is negligible; thus the low angle detector can be used to measure a zero of the asymmetry (a powerful calibration tool) and to monitor and measure the size of density fluctuations.
[C. E. Jones]
We intend to begin development at Caltech of a polarized tritium target for use in an experiment at TJNAF to study the wavefunction of the three body system and to look for possible medium modification of the nucleon form factors (see section II.A.11). We proposed to develop a sealed target based upon optical pumping of the hydrogen isotopes, similar to the optically pumped polarized hydrogen and deuterium internal targets developed at Argonne National Laboratory and the University of Illinois. During the past year, measurements of nuclear polarization of that type of target at IUCF confirm that it is possible to obtain nuclear polarization in hydrogen, which is very similar to tritium, through optical pumping of the atomic substates (see section II.A.12). In order to continue the development work at Caltech which C. E. Jones began at Argonne National Laboratory, the medium energy group at ANL has loaned Caltech an argon ion laser and a Ti-sapphire laser to use for this project. We are currently in the process of setting up a laser lab to begin work on developing optically pumped polarized targets of the hydrogen isotopes. All development work will be done with hydrogen initially before working with tritium because of the similar atomic and nuclear properties of the two isotopes.
[E. Hughes and Y. Kolomensky]
As mentioned in the section on the SLAC Moller scattering experiment (II.A.4), funding for the experiment requires a successful hardware test with a 1 GeV electron beam to be performed at SLAC in the winter of 1999. A primary condition of the test is to demonstrate adequate resolution for the beam position and current. The requirements on the beam position are a one micron resolution on a pulse to pulse basis and a beam current measurement at the level of 3 x 10-5. Both these conditions are approximately an order of magnitude better than what has been achieved in the past.
The investment in such an upgrade is expected to involve improving the electronics and readout. Using a 16 bit ADC developped for a parity violation measurement at CEBAF and with a redesign of the SLAC preamplifier and cable plant, the belief is that these new requirements can be met. The Caltech group has taken on responsibility for the measurements to be performed in the January 1999 test run. Precision monitoring and control of the electron beam parameters is probably the single greatest challenge to the experiment. The goal is to maintain control over false asymmetries at the level of a few times 10-9, approximately an order of magnitude better than what has been done in any previous electron scattering parity violation experiment. Adequate resolution in the beam monitoring devices is necessary so as to keep the statistical error limited by counting statistics.
With funding from Caltech, we have recently installed a CPU "farm" which consists of twenty dual 300 MHz Pentium II computers. Each node of the ``farm'' contains 256 Mbytes of memory and 8.4 Gbytes of disk space. An additional dual 300 MHz computer acts as a server node to manage the farm. The server has one 9 Gbyte and six 18 Gbyte disks. Communication between the nodes of the farm is achieved through a 24-port, 100 Mbit ethernet switch.
This system gives the lab a computing system that is significantly superior to the PC farm that is operated by the HERMES collaboration at DESY. We are presently running the HERMES Monte Carlo program on this farm and expect to continue this analysis as well as begin analysis of on-going experiments on this system in the near future.